THREE ECONOMIC ESSAYS ON THE U.S. MEAT AND POULTRY INDUSTRIES 

By 

Kelsey A. Hopkins 

A DISSERTATION 

Submitted to 
Michigan State University 
in partial fulfillment of the requirements   
for the degree of 

Agricultural, Food, and Resource Economics – Doctor of Philosophy 

2024 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ABSTRACT 

This dissertation focuses broadly on the U.S. domestic meat and poultry industries, more 

specifically  on  issues  related  to  developing  and  adopting  policies  and  programs  for  two  niche 

markets  –  halal  and  farm  animal  welfare  friendly.  Halal  meat  and  poultry  products  meet  the 

religious  dietary  restrictions  for  Muslim  consumers,  specifically  those  related  to  the  slaughter 

process for animals. The first two essays focus on the halal market, while the third focuses on farm 

animal welfare policy adoption.  

The first chapter is concerned with meat and poultry processors’ and retailers’ decision-

making patterns related to supplying halal meat and poultry in the U.S. domestic market. There 

has been strong ongoing demand growth for halal meat and poultry products in the U.S., but a 

relative dearth of processors and retailers entering the market to supply these goods. This essay 

seeks to understand if there are differences in preferences and business decision-making behavior 

between agents in the halal market and agents outside to suggest methods in which to increase 

market  participation.  To  do  this,  I  utilize  a  mixed  methods  design  consisting  of  qualitative 

interviews with halal processors and retailers and survey data from halal and non-halal processors 

and retailers. My methodology consists of analyzing Likert scale data using descriptive statistics, 

principal components analysis, and k-means clustering to reveal patterns and group respondents 

for comparison. My results show that businesses that may expand into the halal market have been 

established longer and more likely to be retailers or further processors. 

The  second  chapter  focuses  on  market  participants’  preferences  for  designing  a  U.S. 

national halal meat and poultry certification program. The development of such a program serves 

as a potential solution to food fraud stemming from an overabundance of confusing and commonly 

contradictory certifications already in the market, similar to the issues that lead to the creation of 

 
the USDA Organic standard. I again use a mixed methods approach of halal consumer, retailer, 

and processor qualitative interviews paired with national stacked surveys containing best-worst 

scaling questions to investigate preferences for the design of a U.S. national halal meat and poultry 

certification program. Results show that the market overall prefers that program designers consider 

most  carefully  Who/What  is  Certified,  Halal  Standards,  and  Costs.  Additional  results  show 

preferences for which organizations should be involved in setting and/or enforcing this program, 

namely government (enforcement only), non-government, religious, and certifier organizations. 

Finally, the data indicated that multiple transparency and traceability measures should be included 

to  ensure  a  robust  and  trustworthy  program.  In  all,  this  chapter  aids  in  bolstering  halal  meat 

consumer confidence in product authenticity and improves the equity of the U.S. food system. 

The final chapter explores modeling of farm animal welfare regulation adoption across the 

U.S. In the U.S., 19 state-level bills and ballot initiatives concerning farm animal welfare (FAW) 

have been adopted across 12 states. In this chapter, I and my co-authors seek to model the evolution 

of  the  state-level  FAW  regulatory  landscape  as  a  function  of  legislature  characteristics  and 

constituent demographics. More specifically, we utilize a two-stage model known as a multinomial 

endogenous  switching  regression  to  assess  whether  and  when  a  given  state  considers  FAW 

measures, and if so, the likelihood the measures are passed. Using this model, we estimate the 

likelihood of FAW adoption for all 50 states. Additionally, we find that the cost to the egg and 

pork industries to upgrade to cage- and crate-free production methods in the states most likely to 

pass a FAW regulation in the future is small relative to the size of the industry. Our findings will 

assist producers and industry stakeholders in gauging the future of the regulatory landscape and 

provide  guidance  on  whether  to  upgrade  existing  enclosures  to  comply  with  mandates  on  the 

horizon or to continue operating with “conventional” enclosures. 

To my mom for her unwavering support and belief in me – not to mention her stellar 
proofreading skills. 

To my husband, Alex, for his love and encouragement in everything I do. 

To my twin daughters, Violet and Vivienne, who serve as both heralds of my next chapter and 
strong motivators in the completion of this degree. 

iv 

 
 
 
 
ACKNOWLEDGEMENTS 

Completing  this  dissertation  would  not  have  been  possible  without  support  and 

encouragement from many channels. First and foremost, I thank my spectacular superwoman of 

an advisor, Dr. Melissa G.S. McKendree, for her faith in me and the many hours we spent together 

working towards the finish line of this degree. I thank my committee – Professors Kimberly Chung, 

David  Ortega,  and  Felicia  Wu  –  for  their  expertise  and  experience  that  greatly  improved  this 

dissertation.  

I am also thankful for the friendships I have made while at Michigan State, and how these 

wonderful humans shaped my time there. I am especially grateful for my friendships with Sarah 

Klammer,  Angelos  Lagoudakis,  and  Caitlin  and  Matt  Herrington  –  for  all  the  coffees  drank, 

dinners eaten, problems solved, laughs had, and memories made. 

Finally,  I  am  thankful  for  my  family.  I  give  thanks  for  my  mom,  Sharon  McCoy,  for 

everything  she  has  done  for  me  and  the  sacrifices  she  has  made,  and  my  husband,  Alexander 

Hopkins, for his love and belief in me. 

v 

 
 
 
 
Islamic Dietary Laws 

PREFACE 

In order for a meat product to be halal – “permissible” for consumption – there are many 

qualifications that must be met; if they are not, the product is haram, or “forbidden.” According to 

Bonne and Verbeke (2007, 2008), most practicing Muslims require halal certifications that ensure 

that Islamic dietary laws are followed at all stages of the supply chain. Islamic dietary restrictions 

are nuanced, and as with many religious texts, open to interpretation. Therefore, I detail only the 

most widely accepted views and dietary laws here. 

   Muslim teachings prohibit consumption of certain species or types of meat, namely pork 

and  dead  meat.  Dead  meat  refers  to  an  animal  whose  spinal  cord  is  severed  in  the  process  of 

slaughter, rendering the heart unable to pump and thus the animal does not die of exsanguination. 

Additionally, animals must be raised on a natural, vegetarian diet that excludes filth or any animal 

proteins.  Animals  must  be  treated  humanely;  they  must  be  well-nourished,  not  stressed  before 

slaughter, the knife should not be sharpened in front of the animal, and no animal should witness 

the slaughter of another animal. Animals must be alive – both heart and brain still fully functioning 

– at the time of slaughter and must die of blood loss. As such, there is much debate about the use 

of electrical stunning prior to slaying an animal; many Muslims are opposed to the practice because 

of  the  risk  of  premature  death,  but  it  is  common  and  considered  humane  in  conventional  U.S. 

slaughter.

Preferably, animals should be slaughtered by hand; machine slaughter is not fully accepted.  

The knife used should be sharp enough to kill the animal with one cut. The slaughter person must 

be  a  sane  adult  Muslim  who  invokes  the  name  of  Allah  prior  to  each  individual  slaughter. 

Recordings  of  blessings  that  are  played  on  loop  are  not  recognized  as  halal.  Finally,  cross-

vi 

 
 
 
contamination  with  haram  products  renders  halal  products  haram.  As  such,  all  steps  of  the 

slaughter process and the following supply chain members must be certified halal.

vii 

 
 
 
TABLE OF CONTENTS 

CHAPTER 1. PROCESSOR AND RETAILER MOTIVATIONS FOR HALAL MEAT AND 
POULTRY MARKET PARTICIPATION: PROFIT, PIETY, AND PURPOSE .......................... 1 
1. Introduction and Motivation ................................................................................................... 1 
2. Research Methods ................................................................................................................... 4 
3. Quantitative Data Collection Process ................................................................................... 10 
4. Quantitative Analysis Methods ............................................................................................. 14 
5. Summary Statistics................................................................................................................ 21 
6. Results & Discussion ............................................................................................................ 31 
BIBLIOGRAPHY ..................................................................................................................... 47 

CHAPTER 2. MARKET PARTICIPANTS’ PREFERENCES FOR A NATIONAL HALAL 
MEAT CERTIFICATION PROGRAM ....................................................................................... 52 
1. Introduction and Motivation ................................................................................................. 52 
2. Current Landscape of U.S. Domestic Halal Certifications ................................................... 54 
3. Halal Certification Around the World .................................................................................. 57 
3. Mixed Methods and Survey Design ...................................................................................... 59 
4. Methodology & Statistical Framework for Analysis ............................................................ 74 
5. Results and Discussion ......................................................................................................... 78 
6. Implications for Implementation ......................................................................................... 102 
7. Conclusions, Limitations, and Future Work ....................................................................... 105 
BIBLIOGRAPHY ................................................................................................................... 109 

CHAPTER 3. RESOLVING THE REALITY GAP IN FARM REGULATION VOTING 
MODELS .................................................................................................................................... 114 
1. Introduction ......................................................................................................................... 114 
2. Background ......................................................................................................................... 117 
3. Methodology ....................................................................................................................... 121 
4. Results ................................................................................................................................. 132 
5. Implications ......................................................................................................................... 142 
6. Conclusion .......................................................................................................................... 146 
BIBLIOGRAPHY ................................................................................................................... 148 

APPENDIX A.1 INTERVIEW PROCESS AND ANALYSIS OVERVIEW ............................ 153 

APPENDIX A.2 PROCESSOR, RETAILER, AND CONSUMER SURVEY QUESTIONS .. 155 

APPENDIX A.3 BEST-WORST SCALING RESULTS AND ADDITIONAL TABLES ....... 221 

APPENDIX A.4 FARM ANIMAL WELFARE REGULATIONS ADDITIONAL TABLES .. 236 

APPENDIX A.5 FARM ANIMAL WELFARE DATA COLLECTION METHODS .............. 242 

APPENDIX A.6 CALCULATION OF FARM ANIMAL WELFARE INDUSTRY COSTS .. 245 

viii 

 
 
CHAPTER 1. PROCESSOR AND RETAILER MOTIVATIONS 
FOR HALAL MEAT AND POULTRY MARKET 
PARTICIPATION: PROFIT, PIETY, AND PURPOSE 

1. Introduction and Motivation 

Demand for halal meat and poultry (hereafter referred to as halal meat for brevity) products 

in the U.S. is growing quickly. The population of U.S. Muslim consumers is increasing rapidly 

and is projected to double by 2050, from about 3-5 million to 6-10 million people (Pew Research 

2018). This growth in demand is in part due to immigration patterns that grow the consumer base, 

but also the vertical mobility by second and third generation Muslims who have begun to consume 

more meat products (Bereaud-Blackler 2004, Bonne & Verbeke 2007). Additionally, non-Muslim 

consumers have demonstrated demand for halal meat products (Campbell et al. 2011).

This rapid growth in demand for halal meat products represents an opportunity for U.S. 

meat processors and retailers to enter the market and expand halal meat availability. Despite this 

strong  demand  and  the  corresponding  opportunities,  domestic  supply  of  halal  meat  products  is 

relatively low. Further, halal meat and poultry products are not readily available outside of major 

metropolitan  areas  or  areas  with  relatively  concentrated  Muslim  populations.  Considering  this 

largely untapped market opportunity, this chapter addresses several research questions. First, how 

did current halal meat retailers and processors decide to offer halal meat products in their stores? 

Are there incentives and barriers that exist when entering the halal meat market, and how do these 

retailers and processors perceive them? Further, what motivations do current halal meat retailers 

and processors have for supplying [certified] halal meat? Taking this research one step further, 

how do the answers to these questions change for non-halal meat processors and retailers, and what 

comparisons can be made? 

1 

 
 
 
As with any market, participants face incentives and barriers to entry; thus, understanding 

how  participants  view  these  incentives  and  barriers  is  essential  to  devising  policies  to  increase 

halal market participation. Compounding these aspects, the U.S. domestic halal meat market has 

multiple religious community considerations in addition to the usual pecuniary motivations that 

influence processor and retailer decision making in specialty market participation. 

The U.S. domestic halal meat and poultry market is understudied. To my knowledge, there 

is  no  agricultural  economic  academic  research  published  to  date  that  focuses  on  the  U.S.  halal 

market. However, there is a recent study discussing religiosity and minority community behavioral 

patterns and their relation to Muslims’ desires to consume halal meat (Mumuni et al 2018). The 

majority of agricultural economic research into the supply side of halal meat markets has been 

conducted in Europe (e.g., Ahmed 2008, Fuseini et al. 2017, Fuseini et al. 2021, Lever et al. 2010, 

Lever & Miele 2012, Masudin et al. 2020, Tieman et al. 2012), Southeast Asia (e.g., Ab Rashid & 

Bojei 2019, Salindal 2019, Shahijan et al. 2014, Tieman et al. 2012,), Australia (e.g., Zulfakar et 

al. 2013, Zulfakar et al. 2018, Zulfakar et al. 2019), and Brazil (e.g., de Araújo 2019). In general, 

these papers find that there are hurdles to halal meat market participation – including obtaining 

certifications, implementing halal slaughter and processing methods, and ensuring supply chain 

integrity. However, they also note that the benefits to supplying halal meat products are numerous, 

including entry into niche markets, the ability to charge a price premium, stronger transparency in 

the meat supply chain, and access to export markets. The U.S. meat market faces many of the same 

obstacles and has the potential to reap many of these benefits.  

Research  into  the  U.S.  meat  sector  in  general  has  uncovered  several  barriers  to  and 

incentives  for  market  entry.  Barriers  include  strong  regulatory  standards  and  food  safety 

regulations  that  can  be  daunting  to  potential  entrants  (Worosz  et  al.  2008);  for  halal  and  other 

2 

 
 
specialty or niche meat sectors, these standards include additional regulations and requirements to 

achieve  certification,  likely  compounding  these  issues.  Additionally,  there  are  issues  related  to 

labor shortages and the seasonality of the livestock slaughter and processing industry (Lewis & 

Peters 2011, Choe 2023), which can lead to logistical challenges for processors. However, research 

in  New  England  found  that  livestock  producers  commonly  demanded  more  slaughter  and 

processing  capacity  than  regional  slaughterhouses  were  able  to  supply  (Lewis  &  Peters  2011), 

which suggests there is room for more processors in the market. Indeed, during the height of the 

COVID-19 pandemic, processing capacity was a concern, largely due to continued labor shortage 

concerns  related  to  virus  transmission  and  policies  and  regulations  impacting  supply  chain 

resiliency (Hobbs 2021, Ijaz et al. 2021, Larue 2021, Taylor et al. 2020, Weersink et al. 2021). 

Thus,  there  are  multiple  incentives  to  market  entry,  including  major  market  opportunities  and 

government  grant  programs  –  such  as  the  2022  Meat  and  Poultry  Inspection  Readiness  Grant 

(MPRIG) – designed to offset the cost of inspected meat processing for small and very small meat 

processors (USDA-AMS 2022).  

This essay contributes to the literature in several ways. First, this study serves to define the 

baseline for the current state of the industry for future work, given the dearth of research on the 

U.S. domestic halal meat supply chain. Second, this research contributes to a better understanding 

of  the  motivations  and  attitudes  of  current  halal  meat  processors  and  retailers  towards  market 

participation, as well as the barriers to and incentives for entry they face(d). Additionally, it is the 

first-ever  evaluation  of  non-halal  processors’  and  retailers’  pecuniary  and  non-pecuniary 

motivations for their decision to not enter the halal meat market, including their perceptions of 

barriers  to  and  incentives  for  halal  meat  market  entry.  This  assessment  provides  necessary 

identification of common concerns, misconceptions, and other factors dissuading processors and 

3 

 
 
retailers  from  entering  the  U.S.  domestic  halal  meat  market  that  can  potentially  be  addressed 

through increased outreach and education. Further, by utilizing a mixed methods approach with 

active stakeholder engagement to ensure supply-side market participants’ concerns and opinions 

are heard and incorporated into study design, this chapter expands the current body of literature on 

addressing systemic inequities in the U.S. food system. In all, this work contributes to efforts to 

promote equitable access to food for Muslim consumers through investigating methods to increase 

the supply of authentic halal meat products nationwide. 

2. Research Methods 

The unique religious and niche nature of the halal market, coupled with the lack of research 

to  understand  the  behavior  of  U.S.  halal  meat  suppliers  requires  a  methodological  approach 

allowing  for  exploration  and  learning  feedback.  As  such,  I  use  a  mixed  methods  approach 

composed of both qualitative and quantitative components. The qualitative component consisted 

of in-depth semi-structured interviews with halal meat and poultry market participants – halal meat 

and poultry processors and retailers, as well as Muslim halal meat and poultry consumers. The 

quantitative component consisted of nationwide online surveys to collect further data; the structure 

and content of these surveys was informed by the data obtained in the qualitative interviews.   

The synergy between methods is exhibited in two ways. In the first stage of this project, I 

formally interviewed multiple halal meat processors and retailers; interviews were qualitative and 

open-ended. These interviews were necessary to learn about the current state of the industry and 

understand  the  nuances  of  market  participation,  halal  production  methods,  and  product 

certification.  I  used  these  conversations  to  inform  the  development  of  the  second  stage  of  this 

project – further narrowing my research questions and the design of the quantitative data collection 

tools.  From  the  qualitative  interviews,  I  determined  the  most  appropriate  research  questions  to 

4 

 
 
explore in relation to halal meat processing and retailing revolve around barriers and incentives to 

market participation, as well as perceptions of halal meat. This focus led me to decide that my 

quantitative  data  collection  should  utilize  Likert  scale  type  survey  questions  to  measure 

perceptions of barriers and incentives to entry, as well as survey questions to benchmark current 

levels of knowledge about halal meat. The data from the quantitative portion of this chapter is 

analyzed  and  interesting,  unexpected,  or  particularly  important  results  are  highlighted  and 

discussed in the context of information collected from the qualitative interviews. In this manner, 

the research comes full circle and incorporates findings from both methodologies throughout the 

project. 

The  two  components  of  this  chapter’s  methods  are  described  in  more  detail  in  the  next 

subsections. 

 2.1 Phase 1: Qualitative Exploration of U.S. Domestic Halal Meat and Poultry Industry 

The qualitative portion of this study was designed using suggested methods from Patton 

(2014),  Rubin  and  Rubin  (2011),  and  Maxwell  (2012).  Interview  questions  were  grouped  by 

topic/area of interest and were open-ended to allow interviewees room for robust answers (Rubin 

and  Rubin  2011).  These  qualitative  interviews  were  crucial  components  of  this  study;  they 

provided the opportunity to obtain information about the U.S. domestic halal meat and poultry 

industry  previously  unknown  to  me,  as  this  market  is  understudied.  Information  collected  via 

qualitative interviews included retailers’ and processors’ motivations for supplying halal meat and 

their preferences between different certifications to use in their operations. Collecting information 

directly from market participants in this way allowed me to design a more robust and relevant 

research  program  for  the  quantitative  portion  of  this  project.  The  qualitative  interviews  also 

provided vital context and explanatory power for the quantitative survey findings. The interview 

5 

 
 
guide for these qualitative interviews, information for how interviews were conducted, and the 

outline of the interview analysis process are found in Appendix A.1.   

Interviewees were recruited from lists of retailers and processors registered as certified to 

supply halal meat by a reputable halal meat certification organization active in the industry for two 

decades.  These  lists  are  available  on  the  certification  organization’s  website.  The  interview 

candidates were narrowed to those in Midwestern states for ease of access and improved likelihood 

of  name  recognition  for  Michigan  State  University.  A  series  of  eight  in-depth,  semi-structured 

interviews with Midwestern halal meat retailers and 10 in-depth, semi-structured interviews with 

Midwestern halal meat processors were conducted December 2021 through April 2022.   

 2.2 Phase 2: Online Survey Methods for Further Exploration of Themes 

The second stage of this chapter involves conducting two national online surveys, one with 

a sample of meat retailers and another with meat processors. The goal of these surveys is to elicit 

nationally representative opinions related to the information collected and to dig deeper into the 

common  themes  discovered  via  the  qualitative  interviews.  The  sample  for  both  retailers  and 

processors included those who currently offer halal products and those that do not. Including both 

current  halal  meat  market  participants  and  those  outside  the  market  is  important  to  be  able  to 

compare differences in motivations and perceived barriers and incentives. This comparison will 

allow us to understand what could increase market participation by meat processors and retailers. 

Potentially,  the  results  will  inform  policies  to  increase  market  participation,  thereby  increasing 

supply of halal meat products and fulfilling demand. Each survey includes Likert scale questions 

to elicit participants’ attitudes towards different incentives and barriers to supplying halal meat 

and  poultry  products  and  using  halal  certifications.  The  surveys  also  included  questions  about 

6 

 
 
participant  and  business  demographics,  with  emphasis  on  religious  and  cultural  demographic 

information. 

In the qualitative interviews, the majority of halal meat processors and retailers described 

strong support for increased certification utilization across the halal meat supply chain. Muslim 

processors and retailers discussed the importance of supplying halal meat to support their cultural 

and religious community members, while non-Muslim owners cited access to niche markets as a 

motivation  for  supplying  halal  meat  products.  Along  these  lines,  most  of  the  processors 

interviewed expressed interest in adopting new traceability strategies – including certifications, 

blockchain, and revised (electronic) paperwork for inventory – in their operations to strengthen 

the integrity of their operation and the halal meat market in general. Additionally, processors and 

retailers reported a variety of barriers to and incentives for halal meat market participation that 

influenced their decision making. Barriers or challenges related to halal meat market participation 

included racial, ethnic, cultural, or religious biases and discrimination, limited access to financial 

assistance such as Islam-compliant business loans, input shortages, and fraudulent competition. 

Incentives included a desire to supply a necessary niche product to their community, the conviction 

that financially supporting themselves/their families through halal methods was morally correct, 

the belief that they are supplying a higher quality product, and the opportunity to access a niche 

market.  

Thus,  the  areas  of  interest  for  Likert  scale  questions  were  selected  based  on  common 

themes from the Phase 1 qualitative interviews. These qualitative interview findings suggested that 

the Likert scale questions in the quantitative survey should be designed to include questions to 

evaluate  participants'  attitudes  towards  potential  barriers  and  incentives  to  halal  meat  market 

participation  and  their  knowledge  of  halal  meat  religious  requirements.  An  example  of  these 

7 

 
 
questions for retailers and processors are included in Figure 1, Figure 2, Figure 3, and Figure 4. 

Participants  selected  how  likely  each  option  was  or  would  be  an  incentive  (motivation)  or 

disincentive (barrier) for adding a halal program to their operations. 

Finally, members of the Islamic community were asked to look over the survey for clarity 

and to ensure there were no misrepresentations prior to distribution. Both the processor and retailer 

surveys in entirety can be found in Appendix A.2. 

Figure 1: Example Likert Scale Questions for Halal Retailing Incentives 

8 

 
 
 
 
 
 
 
 
Figure 2: Example Likert Scale Questions for Halal Processing Incentives 

Figure 3: Example Likert Scale Questions for Barriers to Halal Retailing 

9 

 
 
 
 
 
 
 
Figure 4: Example Likert Scale Questions for Barriers to Halal Processing 

3. Quantitative Data Collection Process  

3.1 Data Collection Process & Summary Statistics 

Surveying  supply-side  agents  is  notoriously  difficult  relative  to  surveying  consumers. 

Typically, supply-side studies receive a very low number of responses, due to a variety of reasons 

– a smaller population to sample from, a lack of relevant pre-established survey panels, and the 

opportunity cost of a business owner’s time being just a few. Nevertheless, I conducted multiple 

efforts  over  many  months  to  contact  and  obtain  survey  responses  from  both  the  processor  and 

retailer populations for this study; details on these efforts are given below. 

10 

 
 
 
 
 
 
 
3.1.1 Processor Sample and Data Collection  

I recruited processor participants from three sources: 1) USDA Food Safety and Inspection 

Service  (FSIS)  list  of  registered  meat  processors,  2)  registered  processors  on  Halal  Monitoring 

Services’ (HMS) website, and 3) the American Association of Meat Processors (AAMP). Poultry, 

lamb, beef, and goat processors were included in the sample – I excluded processors that only 

processed pork since the species is not acceptable for consumption in Islam. While the processors 

listed on HMS’ website are known to be halal, it is also likely that some processors on the USDA 

FSIS database and the AAMP membership list also process halal meat or poultry products, though 

the exact number is unknown. 

The  USDA  FSIS  database  lists  all  6,788  USDA-inspected  processing  establishments. 

Removing pork-only, siluriform-only, and egg-only facilities left 5,859 establishments. Of these 

establishments, 2,736 are classified as “very small” by the USDA with less than 10 employees or 

less than $2.5 million in annual sales, 2,656 are classified as “small” by the USDA with 10-499 

employees and 440 are “large” establishments with 500 or more employees. I conducted stratified 

random sampling of the three groups of establishments in the USDA FSIS data file, using Excel 

to  generate  random  number  lists  to  select  establishments  from  the  populations.  As  very  small, 

small, and large processors make up 46.7%, 45.3%, and 7.5% of the total population, respectively, 

these percentages were used to determine how many establishments to sample from each group. 

There  were  1,049  processors  contacted  from  the  USDA  FSIS  database  including  20  large 

processors, 451 small processors, and 578 very small processors. Establishments were called by a 

team of undergraduate research assistants beginning in early November 2022 to determine who at 

the establishment should respond to the survey and obtain email addresses. Email addresses for 

11 

 
 
 
 
establishments without a phone number listed or those that did not answer or return calls were 

retrieved from business websites when available.  

Individual Qualtrics survey URLs were sent via email to the USDA FSIS sample. The first 

round  of  emails  was  sent  using  MS  Word  mail  merge  on  December  9,  2022,  with  follow-up 

reminder emails on December 13 and 16, 2022. The next reminder email was sent using Constant 

Contact on January 11, 2023. From the first round of emails, approximately 50 bounced back as 

undeliverable, and 12 businesses responded saying they did not qualify or would not be taking the 

survey. Thus, a total of 987 processors from the USDA FSIS list received the survey.  

There  were  an  additional  58  registered  processors  on  the  HMS  website  which  were  all 

included  in  the  sample.  Emails  were  obtained  from  the  certifier,  and  individualized  Qualtrics 

survey URLs were sent using MS Word mail merge in January 2023, with follow-up reminder 

emails in January and February 2023. Additionally, these processors were given the option to take 

the survey in either Arabic or Urdu if they preferred. 

The  AAMP  membership  list  was  contacted  via  an  association  representative,  who 

distributed an anonymous Qualtrics survey URL to the membership email listserv in March 2023, 

with reminder emails in March and April 2023. It is likely that many AAMP members received 

the survey who also were included in my USDA FSIS recruitment efforts. However, the response 

rate  for  both  samples  was  very  low,  so  I  do  not  anticipate  there  were  any  duplicate  survey 

responses. 

Despite  contacting  businesses  in  three  different  samples,  I  received  only  195  total 

responses,  with  only  95  remaining  after  data  cleaning.  I  received  responses  mainly  from  very 

small,  small,  and  medium  processing  plants,  both  because  these  make  up  over  90%  of  meat 

12 

 
 
 
 
 
processors in the nation and, anecdotally, because larger processors typically do not respond to 

surveys. Summary statistics and a discussion of valid responses is presented later in this chapter. 

3.1.2 Retailer Sample and Data Collection  

Non-halal  and  halal  meat  retailers  were  recruited  between  February  2023  and  October 

2023. Non-halal retailers were recruited via multiple state-level grocers associations and from a 

membership  list  from  the  National  Grocers  Association  (NGA).  First,  I  attempted  to  recruit 

retailers via the state-level grocers associations and the NGA email listservs with the assistance of 

association representatives. However, only 18 responses were received via these efforts, so another 

recruitment approach was needed. In May of 2023, a team of undergraduate research assistants 

called retailers from the 2019 winter NGA membership list (National Grocers Association 2019) 

– the most recent available online – between May 2023 and September 2023 to collect point of 

contact email addresses. After removing closed businesses, a total of 946 retail stores were called, 

and 236 email addresses were obtained. As in the processor case, it is possible that some of these 

retailers actually did have a halal program at the time of the survey, though the exact number of 

these stores is unknown. 

Known  halal  retailers  were  recruited  from  halal  certifiers’  online  lists  of  registered 

businesses and through a nationwide web scraping of Yellow Pages using the following key terms 

and phrases: “halal meat grocery store,” “halal meat,” “Indo-Pak grocery,” “African grocery,” and 

“international  grocery  store.”  The  results  of  the  web  scraping  were  compiled,  then  duplicates, 

unrelated businesses, and closed businesses were removed from the list. A team of undergraduate 

research assistants called the remaining 919 stores between July 2023 and October 2023 to collect 

email contact information; 96 email addresses were obtained. 

13 

 
 
 
 
Emails with survey links were sent three times to each category of retailers between August 

25  and  October  10,  2023.  Incentive  payments  of  $25  were  offered  for  complete  and  quality 

responses, though not all respondents claimed their incentive. In total, 50 responses were collected 

from the retailer samples, and after data cleaning, 39 viable survey responses remained.  

4. Quantitative Analysis Methods 

The Likert scale data collected from the surveys was analyzed using three methods: Count 

data and descriptive statistics, principal component analysis (PCA), and k-means clustering. The 

data  in  this  study  consists  of  two  small  samples  from  related  populations  –  meat  and  poultry 

processors and retailers. While there are certainly similarities between businesses within each of 

these  two  populations  (e.g.,  retailer-to-retailer  or  processor-to-processor),  I  am  interested  in 

similarities  and  differences  between  businesses  across  these  two  populations  (e.g.,  retailer-to-

processor). Specifically, I am interested in ascertaining classifying businesses by their patterns of 

behavior so that generalizations based on these classifications can be made and used to prescribe 

potential  methods  to  increase  halal  meat  market  participation.  Thus,  I  first  present  an  overall 

summary of the data collected by establishment type (processor or retailer) and halal status using 

descriptive analysis. Then, I undertake a more complex statistical analysis by pooling the data and 

conducting PCA and k-mean clustering on the combined data to generate variables (PCs) and sort 

businesses based on these common characteristics (k-means). 

4.1 Principal Components Analysis Statistical Framework 

There are multiple methods, including econometric regressions, available to analyze Likert 

scale data; however, the data in this study requires a method that can handle both multicollinearity 

between large numbers of variables and a relatively small sample size. The analysis method that 

best fits this description is PCA, which is commonly used to condense high-dimensional Likert 

14 

 
 
 
 
scale responses into a lower-dimensional form to identify overlapping variability in the sample. 

PCA  and  k-means  clustering  are  commonly  used  together  for  statistical  analysis  of  multi-

dimensional  data  sets.  K-means  clustering  has  been  used  in  conjunction  with  PCA  to  identify 

market segments for agricultural products including seafood (Hanson et al. 1994), apples (Bejaei 

et al. 2020), and beer (Malone & Lusk 2018, and has been popular in the marketing and data-

mining literature (Arabie & Hubert 1996).  It is used to group similar participants without requiring 

large  sample  sizes  nor  some  of  the  more  rigorous  assumptions  required  by  discrete  choice 

experiments and conjoint analysis (Arabie & Hubert 1996).  

PCA was first developed in the early 1900s, when Pearson (1901) and Hotelling (1933) 

endeavored to mathematically define patterns in data to describe large numbers of variables using 

a smaller subset of independent variables. These independent variables – the principal components 

– are chosen to maximize their explanatory power for the total variance of the original variables, 

and  the  components  that  are  derived  in  this  way  were  termed  ‘principal  components.’  In  the 

successive century since PCA was developed, thousands of studies and papers across all fields of 

science have used this method to condense complex data to discover important patterns (e.g., Hsu 

et al. 2009, Sinha et al. 1969, Calder et al. 2001).  

PCA  is  an  orthogonal  linear  transformation;  it  translates  higher-dimensional  data  into  a 

lower-dimensional  coordinate  system  so  that  the  first  coordinate  –  that  is,  the  first  principal 

component  (PC1)  –  represents  some  linear  projection  of  the  greatest  variance  of  the  data. 

Additional principal components (e.g., PC2 and PC3) would lie on the second, third, and so forth 

coordinates with their corresponding second, third, and so forth greatest variances in the data. The 

underlying concepts and procedures are illustrated mathematically below. 

15 

 
 
 
 
 
Suppose that I have a random vector: 

𝑋 =

⎞
⎟

𝑋!
𝑋"
⋮
𝑋#⎠

⎛
⎜

⎝

with population variance-covariance matrix  

𝑣𝑎𝑟(𝑋) = 	Σ = 	 1

" ⋯ 𝜎!#
𝜎!
⋮
⋮
⋱
"
𝜎#! ⋯ 𝜎#

5 

Equation 1 

Equation 2 

Consider the linear combinations, each of which can be thought of as a linear regression 

predicting 𝑌$ from 𝑋!, 𝑋", … , 𝑋#: 

𝑌! = 	 𝑒!!𝑋! + 𝑒!"𝑋" + ⋯ + 𝑒!#𝑋# 

𝑌" = 	 𝑒"!𝑋! + 𝑒""𝑋" + ⋯ + 𝑒"#𝑋# 

⋮ 

𝑌# = 	 𝑒#!𝑋! + 𝑒#"𝑋" + ⋯ + 𝑒##𝑋# 

Since the 𝑌$ are functions of random data, I have: 

#

#

𝑣𝑎𝑟(𝑌$) = 	 ; 	
%&!

; 𝑒$%𝑒$’𝜎%’
’&!

= 	 𝑒$

( ; 𝑒$ 

#

#

𝑐𝑜𝑣>𝑌$, 𝑌)? = 	 ; 	
%&!

; 𝑒$%𝑒)’𝜎%’
’&!

= 	 𝑒$

( ; 𝑒) 

Equation 3 

Equation 4 

Now, the first principal component (PC1) accounts for as much variation in the data as 

possible. It is expressed as the linear combination of x-variables that has maximum variance among 

16 

 
 
 
all the linear combinations. I must maximize the variance subject to the constraint that the sum of 

the squared coefficients 𝒆𝟏 equals one. Expressed mathematically this is: 

#

#

𝑣𝑎𝑟(𝑌!) = 	 ; 	
%&!

; 𝑒!%𝑒!’𝜎%’
’&!

= 	 𝑒!

( ; 𝑒! 

such that: 

#
"
(𝑒! = ; 𝑒!%
𝑒!
%&!

= 1 

Equation 5 

Equation 6 

Additional  principal  components  are  computed  in  much  the  same  way,  with  the  added 

constraint  that  the  components  are  uncorrelated.  That  is,  for  the  ith  PC,  I  find  the  vector  𝒆𝒊  of 

coefficients to solve: 

such that  

and  

Equation 7 

Equation 8 

#

#

𝑣𝑎𝑟(𝑌$) = 	 ; 	
%&!

; 𝑒$%𝑒$’𝜎%’
’&!

= 	 𝑒$

( ; 𝑒$ 

#
"
(𝑒! = ; 𝑒!%
𝑒!
%&!

= 1 

#

#

𝑐𝑜𝑣(𝑌!, 𝑌$) = 	 ; 	
%&!

; 𝑒!%𝑒)’𝜎%’
’&!

= 	 𝑒!

( ; 𝑒$ = 0 

#

#

𝑐𝑜𝑣(𝑌", 𝑌$) = 	 ; 	
%&!

; 𝑒"%𝑒)’𝜎%’
’&!

= 	 𝑒"

( ; 𝑒$ = 0 

17 

 
 
⋮ 

#

#

𝑐𝑜𝑣(𝑌$,!, 𝑌$) = 	 ; 	
%&!

; 𝑒$,!%𝑒)’𝜎%’
’&!

= 	 𝑒$,!

( ; 𝑒$ = 0 

Equation 9 

To  solve  for  the  coefficients  𝑒$),  I  must  use  the  eigenvalues  and  eigenvectors  of  the 

variance-covariance matrix Σ. Letting 𝜆! ≥ 𝜆" … ≥ 	 𝜆# be the eigenvalues of Σ and 𝒆𝟏, 𝒆𝟐, … , 𝒆𝒑 

be  the  corresponding  eigenvectors,  I  have  that  the  elements  for  the  eigenvectors  are  the  PC 

coefficients. Additionally, the variance of the ith PC is equivalent to the ith eigenvalue: 

𝑣𝑎𝑟(𝑌$) = 𝑣𝑎𝑟>𝑒$!𝑋! + 𝑒$"𝑋" + ⋯ + 𝑒$#𝑋#? = 	 𝜆$ 

Now,  to  obtain  the  principal  components  of  a  sample  of  our  data,  I  must  compute  the 

corresponding  sample  eigenvalues  𝜆E

$  and  eigenvectors  𝒆F 𝒊  of  the  sample  variance-covariance 

matrix S. I can then define each estimated PC as a linear combination using the eigenvectors as 

Equation 10 

coefficients: 

𝑌G! = 	 𝑒̂!!𝑋! + 	 𝑒̂!"𝑋" + ⋯ + 	 𝑒̂!#𝑋#	 

𝑌G" = 	 𝑒̂"!𝑋! + 	 𝑒̂""𝑋" + ⋯ + 	 𝑒̂"#𝑋#	 

⋮ 

𝑌G# = 	 𝑒̂#!𝑋! + 	 𝑒̂#"𝑋" + ⋯ + 	 𝑒̂##𝑋#	 

In PCA, I seek to retain minimal PCs so that the proportion of the variance is described 

by the first g PCs is large and ideally close to one. That is: 

Equation 11 

18 

 
 
! + 	 𝜆E
𝜆E
! + 	 𝜆E
𝜆E

" + ⋯ + 	 𝜆E
/
" + ⋯ + 	 𝜆E
#

	 ≅ 1 

 This allows us to obtain the simplest possible relationship between the original data and 

the new, condensed variables (the PCs). If the first few PCs explain a small amount of variation, I 

need  more  of  them  to  explain  a  desired  percentage  of  total  variance  resulting  in  a  large  g.  To 

determine the number of PCs for this study, I retained any PC that accounts for at least 10% of the 

Equation 12 

common variance (Malone and Lusk 2017).1 

4.2 K-means Clustering Motivation and Statistical Framework  

K-means  clustering  is  advantageous  for  analyzing  the  data  in  this  study  because  it  is 

straightforward to implement and interpret – preferable attributes for performing an exploratory 

study. Additionally, k-means clustering is ideal as it segments the market into groups that exhibit 

common views.  

Given the exploratory nature of this study, I conduct simple nonhierarchical cluster analysis 

where individuals are grouped using the least squares method to minimize the Euclidean distance 

within a specified k number of groups or clusters. This algorithm partitions the data space in a way 

so that data points within the same cluster are as similar as possible (intra-class similarity) with 

respect to their PC scores, while data points from different clusters are as dissimilar as possible 

(inter-class  similarity).  In  k-means,  each  cluster  is  characterized  by  its  centroid,  which  is  the 

arithmetic mean of the data points assigned to the cluster, but it might not be a member of the 

dataset.  K-means  randomly  selects  data  points  as  possible  centroids  of  the  clusters  and  then 

1 I used the principal components analysis commands in Stata to conduct this analysis, using the nine common 
Likert scale questions between the processor are retailer surveys as the independent variables for which to develop 
the PCs from. I did not use an orthogonal rotation method, as I am not using PCA to compare across different scales 
or groups of questions in the data. Specifically, the code used was: pca PricePremium - LackKnowledge 

19 

 
 
 
 
iteratively recalculates new centroids to converge to a final clustering of the data points. K-means 

assigns every data point in the dataset to the nearest centroid, meaning that a data point is in a 

given cluster if it is closer to that cluster’s centroid than any other centroid. The algorithm repeats 

the selection of centroids and sorting of data points until the sum of distances between data points 

and their given centroid is minimized, the maximum number of iterations is reached, or there are 

no changes in centroid values.2 

The Likert scale data resulted in a 9-dimensional space. I indexed this data using the three 

PCs, and then partition the observations into three groups (clusters) based on their PC values. As 

such, the specific objective of this cluster analysis is to minimize the within-group variation of the 

PC values. Specifically, the k-means method in our context minimizes the squared distance from 

each observation (𝑥$) to the center of the observation’s associated cluster (𝑋K$0): 

2

min(𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒1) = 𝑚𝑖𝑛U;(𝑥$ − 𝑋K$0)"

$&!

Equation 13 

Once  I  select  the  k  PCs  that  best  describe  most  of  the  variance,  I  use  these  PCs  as 

independent  variables  on  which  to  cluster  or  “sort”  individuals  into  groups  based  on  their 

individual predicted PC values. After condensing the data in this manner, k-means clustering is 

applied to group the observations into clusters that exhibit similar response patterns. That is, rather 

than grouping variables as is done in PCA, cluster analysis allows us to group people – or in our 

case, businesses. In this study, cluster analysis will combine all participants’ responses and then 

2 I conducted k-means cluster analysis in Stata using the cluster kmeans command on the three PCs selected from 
the PCA to sort the data into three groups with a set starting seed value and a maximum of 1,000,000 iterations. In 
particular, the code used is: cluster kmeans pc1 pc2 pc3, k(3) name(threegroups) start(random(12281996)) 
iterate(1000000) 

20 

 
 
 
 
 
sort them into new groups; I anticipate this segmentation will group some non-halal processors 

and  retailers  with  their  halal  counterparts,  revealing  what  common  characteristics  of  non-halal 

businesses are most suitable to future expansion into the halal market.  

To  determine  the  unique  characteristics  of  the  different  market  segments  created  via  k-

mean  clustering,  F-tests  on  the  means  between  groups  for  each  demographic  category  are 

conducted. Values with 10% or stronger statistical significance are then evaluated with pairwise t-

tests  to  discover  more  specific  differences  between  the  groups.  Ultimately,  this  will  allow  for 

policy prescriptions tailored to their unique positions in the market that are not easily seen when 

analyzing participants within their original (halal versus non-halal) categories. 

5. Summary Statistics 

The  summary  statistics  of  respondents  are  given  by  business  type  in  the  following 

subsections. 

5.1 Processor Summary Statistics 

The processor survey received 195 responses. After dropping ineligible and responses that 

were less than 50% complete, 95 responses remained.  Establishments represented in this data are 

in 39 states.3  

5.1.1 Processor Respondent Characteristics 

Summary  statistics  of  processor  respondent  characteristics  are  given  in  Table  1. 

Respondents were mainly male and white. The majority had at least a 4-year college degree. When 

asked their political affiliation, most said they were Republicans or Independents, although 31% 

3 The states represented in the data are Alabama, Arkansas, California, Colorado, Connecticut, Florida, Georgia, 
Illinois, Indiana, Kansas, Kentucky, Louisiana, Maine, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, 
Montana, Nebraska, Nevada, New Jersey, New York, North Carolina, North Dakota, Ohio, Oklahoma, Oregon, 
Pennsylvania, Rhode Island, South Dakota, Tennessee, Texas, Utah, Virginia, Washington, West Virginia, 
Wisconsin, and Wyoming. 

21 

 
 
 
 
 
opted not to disclose their political affiliation. Twenty-one percent indicated they were second-

generation immigrants (i.e., their parent(s) were born outside of the U.S., but they were born in the 

U.S.). Immigration information is relevant to this study, as more recent immigrants or younger 

generations  of  immigrants  are  more  likely  to  identify  with  and  participate  in  the  cultural  and 

religious heritage from their home or ancestral country (Lopez 2017). The most recent waves in 

immigration of Muslims to the U.S. occurred in the 1980s and in the 2000s onward, as the U.S. 

relaxed  immigration  laws  and  civil  unrest  and  climate  catastrophes  in  the  Arab  world  and 

Southeast Asia contributed to a refugee crisis (Carmichael 2020); as such, it is possible that more 

younger generations of immigrants are active in the halal meat and poultry supply chain. Further, 

21% of the sample indicated they are religious. As halal is a religious product, this information is 

potentially important for understanding processors’ motivations - or lack thereof - for providing 

halal products. 

22 

 
 
 
 
Table 1: Processor Respondent Summary Statistics 

Category 
Gender (n = 94) 
Male 
Female 
Prefer not to disclose 
Education Level (n = 94) 
High School 
Some College 
2-Year Degree (Associates) 
4-Year Degree (Bachelor's) 
Master's Degree 
Professional Degree 
Race (n = 94) 
White 
Black 
Native American or Alaskan Native 
Native Hawaiian or Pacific Islander 
Asian 
Other 
Prefer not to disclose 
Hispanic (N = 93) 
No 
Yes 
Prefer not to disclose 
Political Party (N = 94) 
Democrat 
Republican 
Independent 
Other 
Prefer not to disclose 
2nd Generation Immigrant (N = 94) 
No 
Yes 
Prefer not to disclose 
Currently Religious (N = 94) 
Yes 
No 
Prefer not to disclose 
Previously Religious (N = 20) 
Yes 
No 
Prefer not to disclose 

23 

Percent 

76% 
19% 
5% 

6% 
13% 
10% 
44% 
22% 
5% 

76% 
1% 
0% 
0% 
3% 
6% 
14% 

89% 
3% 
8% 

5% 
35% 
26% 
3% 
31% 

66% 
21% 
13% 

21% 
56% 
22% 

45% 
40% 
15% 

 
 
 
 
 
 
 
 
 
 
 
5.1.2 Processor Establishment Characteristics 

Processing  establishment  summary  statistics  are  given  in  Table  2.  Establishments 

represented in the sample consisted of facilities that slaughtered, processed, both slaughtered and 

processed,  or  conducted  other  meat  and  poultry  product  handling  (“other”).  Seven  percent  of 

respondents only slaughtered meat or poultry, 44% of respondents only processed meat or poultry, 

45% both slaughtered and processed meat or poultry products, and 3% of respondents conducted 

other  meat  and  poultry  product  handling.  Most  respondents  represented  very  small  or  small 

processors;  the  mean  and  median  business  size  of  our  sample  was  62  and  20  employees, 

respectively. This is not surprising, as over 95% of meat and poultry processing establishments in 

the U.S. are small or very small per USDA standards. On average, these establishments had been 

open  for  31  years.  Unsurprisingly,  given  the  nature  of  meat  and  poultry  processing,  48%  of 

respondents indicated their establishment is in a rural area. 

Many  slaughter  and  processing  establishments  handle  multiple  species.  Of  the  95 

establishments in our sample, 83% indicated they handled beef, 34% veal, 55% lamb, 46% pork, 

35% turkey, 38% chicken, 55% goat, and 35% handled exotics (game, specialty poultry, etc.). 

Respondents were asked if they currently or had ever slaughtered, processed, or handled 

halal meat or poultry products. Thirty-four percent responded they were currently operating a halal 

program, 4% had previously had a halal program but did not currently, and 62% indicated they 

had never operated a halal program at their establishment.  

24 

 
 
 
 
 
 
 
 
 
Table 2: Processing Establishment Summary Statistics, N = 95 

Category 
Establishment Type 
Slaughter without processing 
Processing without slaughter 
Slaughter and processing 
Other 
Location 
Rural 
Suburban 
Urban 
Prefer not to disclose 
Type of Animal 
Beef 
Veal 
Lamb 
Pork 
Turkey 
Chicken 
Goat 
Exotics 
Halal Status 
Current Halal 
Past Halal 
Never Halal 
Number of Employees  Mean  Median 

Year Established 

62 

20 

Mean  Median 
1992 

2001 

Percent 

7% 
44% 
45% 
3% 

48% 
23% 
23% 
6% 

83% 
34% 
55% 
46% 
35% 
38% 
55% 
35% 

34% 
4% 
62% 

Min 
2 
Min 
1902 

Max 
850 
Max 
2022 

Looking more specifically at the 32 establishments that reported currently operating a halal 

program, over three-quarters indicated they have a zabiha (hand-slaughtered) halal program (Table 

3). Further, over one-half of these establishments reported operating a halal program for more than 

7 years. Most of these establishments are certified by a third-party halal certification group, which 

is  not  surprising  given  our  sampling  strategy.  For  operations  that  conduct  halal  slaughter  or 

processing, most reported that halal slaughter or processing was over 50% of their operation.  

25 

 
 
 
 
 
 
 
 
 
 
Table 3: Current Halal Establishment Summary Statistics, N = 32 

Category 
Halal slaughter method 
Zabiha 
Machine 
Unsure/Don’t know 
Years operating a halal program 
< 1 year 
1-3 years 
4-6 years 
7-10 years 
11-20 years 
> 20 years 
Certified by 3rd party 
Yes 
No 
Unsure 
Percent of slaughter that is halal 
< 10% 
11-25% 
26-50% 
50-75% 
> 75% 
Percent of processing that is halal 
< 10% 
11-25% 
26-50% 
50-75% 
> 75% 

Percent 

77% 
6% 
16% 

6% 
29% 
13% 
16% 
13% 
23% 

78% 
16% 
6% 

9% 
4% 
4% 
74% 
9% 

24% 
12% 
4% 
4% 
56% 

5.2 Retailer Summary Statistics 

The  retailer  survey  received  50  responses.  After  dropping  ineligible  responses  and 

responses that were less than 50% complete, 41 responses remained. The survey was distributed 

nationwide; stores represented in this data are located in 21 states.4 

4 The states represented in the data are Alaska, California, Hawaii, Illinois, Indiana, Maine, Maryland, 
Massachusetts, Michigan, Montana, Nebraska, New Jersey, North Carolina, North Dakota, Ohio, Rhode Island, 
South Dakota, Texas, Utah, Virginia, and Wisconsin. 

26 

 
 
 
 
 
 
 
 
5.2.1 Retailer Respondent Characteristics 

Retailer  respondent  demographic  summary  statistics  are  given  in  Table  4.  Like  in  the 

processor  survey,  respondents  were  mainly  male  and  white,  and  the  majority  had  at  least  a 

bachelor’s degree. When asked their political affiliation, most indicted they were Republicans or 

Independents,  although  24%  opted  not  to  disclose  their  political  affiliation.  Fifteen  percent 

indicated they were second generation immigrants (i.e., their parent(s) were born outside of the 

U.S.,  but  they  were  born  in  the  U.S.).  Further,  36%  of  the  sample  indicated  they  are  currently 

religious.  

27 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table 4: Retailer Respondent Demographic Summary Statistics 

Category 
Gender (n = 33) 
Male 
Female 
Education Level (n = 32) 
High School 
Some College 
2-Year Degree (Associates) 
4-Year Degree (Bachelor’s) 
Master’s Degree 
Professional Degree 
Race (n = 36) 
White 
Black 
Native American or Alaskan Native 
Native Hawaiian or Pacific Islander 
Asian 
Other 
Prefer not to disclose 
Hispanic (n = 33) 
No 
Yes 
Political Party (n = 33) 
Democrat 
Republican 
Independent 
Other 
Prefer not to disclose 
1st or 2nd Generation Immigrant (n = 33) 
No 
Yes 
Prefer not to disclose 
Currently Religious (n = 33) 
Yes 
No 
Prefer not to disclose 
Previously Religious (n = 12) 
Yes 
No 

Percent 

79% 
21% 

3% 
13% 
28% 
25% 
22% 
9% 

66% 
0% 
5% 
2% 
10% 
2% 
2% 

97% 
3% 

6% 
27% 
36% 
6% 
24% 

82% 
15% 
3% 

36% 
48% 
15% 

58% 
42% 

28 

 
 
 
 
 
 
 
 
 
 
 
5.2.2 Retail Store Characteristics  

Summary  statistics  of  the  retail  stores  in  our  sample  are  given  in  Table  5.  Respondents 

consisted of grocery stores (78%), butcher shops or delis (14%), and “other” (8%) retail stores – 

likely  ethnic  grocery  store-restaurant  hybrids,  which  are  common  in  areas  with  high  ethnic 

populations.  Most  retailers  represented  in  the  sample  were  very  small  or  small  independent 

retailers, with an average of 45 employees. This is not surprising, as we sampled mainly from the 

National Grocers Association membership list. Further, approximately 33% of retail grocery stores 

in the U.S. are small or very small independent retailers, per the National Grocers Association. On 

average, these stores have been open for 44 years. Over half of respondents indicated their store is 

in a rural area. 

Respondents were asked if they currently or had ever sold halal meat or poultry products: 

22% responded they were currently selling halal meat or poultry products, 10% had previously 

sold halal meat or poultry products, 51% had never sold halal meat or poultry products at their 

store, and 17% were unsure if they had ever sold halal. For the purposes of the survey and this 

analysis, those who were unsure if they had ever sold halal were classified as “never sold halal” 

for further questions about their operations.  

29 

 
 
 
 
 
 
 
 
 
 
 
Table 5: Retail Store Summary Statistics (n = 41) 
Participant Type 
Grocery Store 
Butcher Shop/Deli 
Other 
Rural 
Suburban 
Urban 
Currently selling halal 
Previously sold halal 
Never sold halal 
Unsure if ever sold halal 
Number of Employees 

Frequency 
33 
5 
3 
20 
11 
2 
9 
4 
21 
7 
Mean 
45 
Mean 
1979 

Year Established 

Percentage 
80% 
12% 
7% 
61% 
33% 
6% 
22% 
10% 
51% 
17% 
Min 
600 
Max 
2021 

Min 
1 
Min 
1867 

Retail food stores sell multiple species of meat or poultry; this information is presented in 

Table 6. Of the 41 stores in our sample, 100% indicated they sold beef, 41% veal, 56% lamb, 85% 

pork, 95% turkey, 100% chicken, 17% goat, and 34% exotics (game, specialty poultry, etc.). For 

the 13 stores that currently sell or previously sold halal meat or poultry products, 62% sold halal 

beef, 38% halal veal, 46% halal lamb, 38% halal turkey, 92% halal chicken, 54% halal goat, and 

8% halal exotics. 

Table 6: Percent of Retail Stores that Carry Different Species of Meat and Poultry 
Type of Animal 
Beef 
Veal 
Lamb 
Pork 
Turkey 
Chicken 
Goat 
Exotics 

Current and Past Halal Stores Only, n = 13 
62% 
38% 
46% 
0% 
38% 
92% 
54% 
8% 

All Stores, n = 41 
100% 
41% 
56% 
85% 
95% 
100% 
17% 
34% 

The summary statistics for the nine current halal retailers are given in Table 7. Most of 

these stores sell zabiha (hand-slaughtered) halal products. Further, all stores in the sample had sold 

30 

 
 
 
 
 
 
 
halal meat and poultry products for less than a decade at the time of the survey. Most of these 

stores are certified by a third-party halal certification group. However, halal meat and poultry sales 

make up less than 50% of each store’s overall meat and poultry sales.   

Table 7: Current Halal Retailer Summary Statistics (n=9) 

Category 
Zabiha 
Machine 
Unsure/don’t know 
Years Selling Halal Meat & Poultry 
< 1 year 
1-3 years 
4-6 years 
6-10 years 
11-20 years 
> 20 years 
Store is Halal Certified 
Yes 
No 
Unsure 
Percent of Meat & Poultry Sales that 
are Halal 
< 10% 
11-25% 
26-50% 
50-75% 
> 75% 

Frequency  Percent 

7 
1 
2 

2 
1 
3 
3 
0 
0 

5 
2 
2 

1 
2 
6 
0 
0 

70% 
10% 
20% 

22% 
11% 
33% 
33% 
0% 
0% 

56% 
22% 
22% 

11% 
22% 
67% 
0% 
0% 

6. Results & Discussion 

The  results  of  the  analyses  presented  in  this  paper  shed  light  on  potential  avenues  to 

increase the availability of halal meat and poultry products. In the following subsections, results 

are given by analysis method – Likert scales, PCA, and k-means clustering - and related discussion 

is included. 

31 

 
 
 
 
 
 
 
 
 
 
 
6.1 Processors’ and Retailers’ Perceptions of Motivations and Barriers 

Likert  scales  were  used  to  assess  processors’  and  retailers’  perceptions  of  motivations 

(incentives)  and  barriers  (disincentives)  in  the  decision  to  establish  a  halal  program  at  their 

operation.  I  am  interested  in  the  differences  between  businesses  who  currently  or  have  never 

offered halal meat or poultry products, as these differences may indicate avenues for supporting 

increased  market  entry  into  the  halal  processing  and  retailing  industry.  Note,  I  do  not  report 

specific statistics for past halal processors and retailers out of this analysis, as this group includes 

only eight businesses.  

6.1.1 Likert Scales: Motivations and Incentives 

Retailers  and  processors  were  asked  to  indicate  which  motivations  or  incentives  they 

considered for currently or potentially selling or processing halal meat or poultry products. An 

example of these questions for retailers and processors are included above in Figure 1 and Figure 

2, respectively. In answering these questions, participants selected how likely each option was or 

would be an incentive or motivation for adding a halal program to their operations. 

The  results  of  these  groups  of  Likert  questions  are  in  Table  8,  reported  as  percentages. 

Results are presented by type of establishment – retailers and processors, halal and non-halal – and 

by level of incentive in decision-making. 

32 

 
 
 
 
  
Table 8: Results of Likert Scale Questions for Halal Incentives   

Incentives 
Considered for 
Halal 

Higher Price 

Access New 
Markets 

Supply Minority 
Communities 

Financial Aid to 
Establish 

Assistance from 
Organizations 

Compete with 
Similar 
Businesses 

Likert 
Scale 
Response 
Likely 
Neutral 
Unlikely 
Likely 
Neutral 
Unlikely 
Likely 
Neutral 
Unlikely 
Likely 
Neutral 
Unlikely 
Likely 
Neutral 
Unlikely 
Likely 
Neutral 
Unlikely 

Halal 
Retailers 
n = 9 
0% 
33% 
67% 
67% 
11% 
22% 
0% 
89% 
11% 
N/A 
N/A 
N/A 
33% 
22% 
44% 
56% 
11% 
33% 

Non-Halal 
Retailers 
n = 26 
31% 
38% 
31% 
42% 
31% 
27% 
19% 
54% 
27% 
N/A 
N/A 
N/A 
23% 
31% 
46% 
8% 
50% 
42% 

All Retailers 
n = 39 

20% 
37% 
39% 
49% 
22% 
24% 
15% 
59% 
22% 
N/A 
N/A 
N/A 
27% 
27% 
41% 
17% 
41% 
37% 

Halal 
Processors 
n = 18 
26% 
37% 
37% 
78% 
17% 
6% 
56% 
28% 
17% 
11% 
11% 
78% 
17% 
22% 
61% 
N/A 
N/A 
N/A 

Non-Halal 
Processors 
n = 57 
19% 
32% 
49% 
28% 
25% 
47% 
16% 
37% 
47% 
30% 
25% 
46% 
30% 
28% 
42% 
N/A 
N/A 
N/A 

All 
Processors 
n = 79 
21% 
34% 
45% 
40% 
24% 
35% 
25% 
36% 
38% 
25% 
23% 
51% 
26% 
28% 
45% 
N/A 
N/A 
N/A 

33 

 
 
 
Nine retailers and 18 processors reported that they are currently supplying halal meat or 

poultry products and answered the Likert scale questions in the survey. When making the decision 

to offer halal meat or poultry products at their business, the ability to sell halal products at a higher 

price than conventional products was not an incentive for current halal retailers and processors. 

However, the ability to access new markets was a strong incentive for both retailers and processors. 

Surprisingly, the ability to supply minority communities was not an incentive for halal retailers, 

with most current halal retailers indicating it was a neutral factor in their decision. This may be 

due to current halal retailers operating in areas of the country in which there are large Muslim 

populations,  in  which  these  retailers  may  not  see  Muslims  as  a  minority  in  their  area.  Indeed, 

according  to  the  Public  Religion  Research  Institute  (2023),  Muslim-Americans  are  mainly 

concentrated in major metropolitan areas. However, the ability to supply products for minority 

communities was generally a strong incentive for current halal processors.  Obtaining assistance 

from outside organizations to set up a halal program at their stores was in general a slight incentive 

for current halal retailers, and obtaining financial aid and assistance setting up a halal program 

were not incentives that were valued by current halal processors. Finally, the ability to compete 

with  similar  businesses  was  generally  a  strong  incentive  for  current  halal  retailers  and  a  slight 

incentive for halal processors. 

For the 26 retailers and 57 processors that have never had a halal program, I asked which 

of these six factors would incentivize them to add a halal program. The majority indicated that all 

six factors would not be an incentive or would be a neutral factor in their decision. Access to new 

markets (42%) was the factor most likely to incentivize retailers to add a halal program to their 

store – determined by the overall percentage of the time retailers selected the factor as likely was.  

This is unsurprising, as it is likely that retailers that have never sold halal products are not members 

34 

 
 
 
 
of the Muslim community; most retailer respondents in our survey  who said they are religious 

indicated  they  were  not  Muslim.  Therefore,  they  are  unlikely  to  have  considered  adding  halal 

products to their stores and may not have known what halal was prior to receiving this survey. 

Indeed, several survey respondents indicated in the comments box they did not know what halal 

was prior to taking the survey. The factors most likely to incentivize processors to add a halal 

program to their establishment were financial aid to establish a halal program (30%), assistance 

from organizations to set up a halal program (30%), and the ability to access new markets (28%), 

though these are not practically significant figures.  

6.1.2 Likert Scales: Barriers and Disincentives 

Participants were also asked to indicate which potential barriers or disincentives they faced 

or considered for currently or potentially offering halal meat or poultry products at their operations. 

An example of this question for retailers is given above in Figure 3 and an example of this question 

for processors is given in Figure 4. When answering these questions, participants could select how 

likely each option was or would be a barrier in their decision to supply or not supply halal meat or 

poultry products.  

The descriptive statistics of these responses are given in Table 9, reported as percentages. 

Results are presented by type of establishment – retailers and processors, halal and non-halal – and 

by level of disincentive in decision-making. 

35 

 
 
 
 
Table 9: Results of Likert Scale Questions for Halal Disincentives 

Barriers Considered 
for Halal 

Likert Scale 
Response 

Costs to Participate 

Discrimination from 
Regulators 

Backlash from Non-
Muslims 

Lack of Muslim Labor 

Limited Local Market 

Lack of Halal 
Knowledge 

Unlikely 
Neutral 
Likely 
Unlikely 
Neutral 
Likely 
Unlikely 
Neutral 
Likely 
Unlikely 
Neutral 
Likely 
Unlikely 
Neutral 
Likely 
Unlikely 
Neutral 
Likely 

Current 
Halal 
Retailers 
n = 9 
44% 
33% 
22% 
67% 
33% 
0% 
100% 
0% 
0% 
N/A 
N/A 
N/A 
44% 
11% 
44% 
56% 
44% 
0% 

Current 
Halal 
Processors 
n = 19 
56% 
31% 
13% 
72% 
22% 
6% 
72% 
22% 
6% 
11% 
50% 
39% 
68% 
5% 
26% 
72% 
6% 
22% 

Non-Halal 
Processors 
n = 57 

All 
Processors 
n = 77 

19% 
33% 
47% 
53% 
44% 
4% 
56% 
39% 
5% 
25% 
35% 
40% 
23% 
30% 
47% 
21% 
39% 
40% 

26% 
31% 
43% 
56% 
41% 
4% 
59% 
35% 
5% 
20% 
39% 
41% 
34% 
25% 
41% 
32% 
33% 
35% 

Non-Halal 
Retailers 
n = 26 

All 
Retailers 
n = 39 

26% 
26% 
38% 
49% 
36% 
5% 
62% 
23% 
5% 
N/A 
N/A 
N/A 
15% 
8% 
67% 
28% 
36% 
26% 

21% 
25% 
46% 
46% 
39% 
7% 
54% 
32% 
7% 
N/A 
N/A 
N/A 
7% 
7% 
79% 
21% 
36% 
36% 

36 

 
 
 
 
 
For the nine halal retailers and 19 halal processors in our sample, costs to participate in a 

halal certification program to sell products at their businesses were not a major barrier for retailers 

nor processors. Evidence from interviews conducted in the first phase of this project indicated that 

multiple halal processors have experienced discriminatory or hostile behavior from USDA meat 

and poultry plant inspectors. However, the results of the national online survey show that many 

current halal retailers, processors, and regulators did not consider discrimination from regulators 

or inspectors to be a barrier to halal production. 

Halal retailers and processors were not concerned about backlash from their non-Muslim 

customers. However, while some halal processors slaughter exclusively halal, some of the meat 

and poultry processing plants interviewed for this project that have a halal program said that they 

run halal on their processing lines relatively infrequently – from once or twice a month to once 

every few months – while the rest of their processing is for non-halal customers. Lack of Muslim 

labor  was  a  concern  for  some  halal  processors.  In  interviews,  some  processors  said  they  must 

contract with their certifier to bring in a Muslim slaughter person for the days they run halal on 

their line and cited reasons related to their local area (e.g., rural non-Muslim communities). For 

current halal retailers, a limited local market was equally indicated to be a barrier and not a barrier. 

Contrastingly, a limited local market was not a barrier for processors; this is not surprising, as meat 

processors typically distribute their products outside of their local market, regardless of whether 

they are halal or not. Finally, a lack of halal knowledge was not a challenge faced by current halal 

retailers when establishing the program at their store – again, likely because many of these retailers 

are themselves Muslim. However, a lack of halal knowledge was a challenge faced by 22% of 

halal processors when establishing the program at their establishment – this may be because of the 

37 

 
 
 
numerous and rigorous standards that must be upheld for halal slaughter. It may also be because 

many of these processors run halal infrequently and are likely not themselves Muslim. 

The 26 retailers and 57 processors who have never offered halal meat or poultry products 

at their businesses in general differ from halal retailers and processors in what challenges they 

perceive they would face if they were to offer halal meat or poultry products at their establishments 

in the future. These retailers and processors perceive that the costs to participate in a halal certified 

program  would  be  prohibitive.  Discrimination  from  regulators  and  backlash  from  non-Muslim 

consumers are not seen as major challenges for retailers nor processors. Retailers’ and processors’ 

lack  of  concern  about  discrimination  and  backlash  is  likely  because  these  businesses  are,  in 

general,  not  themselves  ethnic  or  religious  minorities,  and  therefore  are  unlikely  to  be  worried 

about this behavior. Non-halal retailers are strongly concerned about the lack of a local market for 

halal meat and poultry products, as 79% of them indicated this would be a barrier for their store to 

offer halal, while processors are not as concerned about the lack of a local market. This makes 

sense, as meat and poultry products are commonly shipped across state lines from processors to 

retail  stores,  while  retailers  rely  on  customers  near  their  stores.  Finally,  non-halal  retailers  and 

processors indicated that a lack of knowledge about halal meat and poultry would be a disincentive 

or a neutral factor in their decision to offer halal meat or poultry products in the future.  

6.2 Principal Components Analysis and K-Means Clustering 

PCA was conducted on the combined processors’ and retailers’ data sets, again excluding 

past halal retailers. This allows us to analyze patterns across all halal and non-halal businesses 

together and potentially reveal commonalities that point to outreach methods or policy design to 

facilitate  halal  market  expansion.  Additionally,  PCA  facilitates  analysis  of  how  participants 

38 

 
 
 
responded to all questions, instead of looking at each question’s response individually. This allows 

me to better pinpoint participants’ overall attitudes and opinions towards halal business decisions. 

In  the  data,  an  increasing  Likert  scale  value  for  a  motivation  (incentive)  indicates  an 

increase in the attractiveness of that motivation; that is, higher values correspond to a business 

feeling  more  incentivized.  Contrastingly,  an  increasing  Likert  scale  value  for  a  barrier 

(disincentive) indicates an increase in how prohibitive it is; that is, higher values correspond to a 

business feeling more disincentivized. Only the Likert scale questions that were included on both 

the retailer and the processor surveys were used for the PCA. 

The results of PCA for this analysis are given in Table 10 and Table 11. A PC was retained 

if it accounted for at least 10% of the variation in the data. The PCA resulted in three latent factors 

representing a total of 71.84% of the variance in the original dataset; PC1 represented 37.29% of 

the variation, PC2 represented 22.35%, and PC3 represented 12.20%. Thus, there are now three 

new  variables  (PC1,  PC2,  and  PC3)  that  contain  information  and  patterns  between  the  nine 

variables  in  the  original  dataset  (the  motivations  and  barriers).  These  three  new  variables  now 

facilitate more straightforward analysis of the responses, as they represent the three major trends 

or themes in the data in which we can classify behavior. I call these new variables “Concerns” 

(PC1),  “Motivators”  (PC2),  and  “Niche  Minority  Considerations”  (PC3)  and  explain  their 

elements below.  

PC1, “Concerns,” has positive correlations of approximately equal magnitude with all the 

features of a business’ decision of whether to offer halal products; thus, businesses with high values 

of PC1 are equally motivated by the incentives and dissuaded by the barriers. PC2, “Motivators,” 

has a positive correlation with all motivations and a negative correlation with all barriers; thus, 

businesses with high values for PC2 consider the motivations to be attractive and the barriers to 

39 

 
 
not be prohibitive in their decision-making process. PC3, “Niche Minority Considerations,” has a 

positive correlation with the first three motivations: Price Premium, Access to New Markets, and 

Providing  a  Religious  Minority  Product,  as  well  as  two  of  the  barriers:  Discrimination  from 

Regulating  Bodies  and  Backlash  from  Non-Muslim  Customers.  It  follows  that  businesses  with 

high values of PC3 have mixed opinions on the motivations and barriers, specifically those related 

to supplying religious minority products when deciding to offer halal at their business.   

Table  10:  Eigenvalues  and  Proportion  of  Variance  for  Principal  Components  Analysis  of 
Retailer and Processor Data, n = 109 
Principal Components Metrics 
Eigenvalues 
Proportion of Variance 

PC1 
3.36 
37.29% 

PC2 
2.01 
22.35% 

PC3 
1.10 
12.20% 

PC1 

Price Premium 

Motivation or Barrier 

Table 11: Loadings for Principal Components Analysis of Retailer and Processor Data, n = 
109 
Motivation 
or Barrier 
Motivation 
Motivation  Access to New Markets 
Motivation 
Motivation  Technical Assistance Setting Up 
Barrier 
Barrier 
Barrier 
Barrier 
Barrier 

Halal Certification Program Costs 
Discrimination from Regulating Bodies 
Backlash from Non-Muslim Customers 
Limited Local Market Opportunities 
Lack of Halal Knowledge  

0.26 
0.47 
0.47 
0.29 
-0.17 
-0.36 
-0.40 
-0.22 
-0.18 

0.04 
0.13 
0.32 
-0.12 
-0.40 
0.49 
0.50 
-0.40 
-0.21 

0.38 
0.28 
0.21 
0.36 
0.37 
0.31 
0.28 
0.34 
0.41 

Providing a Religious Minority Product 

PC2 

PC3 

K-means  clustering  was  then  conducted  using  the  three  PCs  from  the  PCA  as  the 

underlying factors on which to sort the respondents. Three groups (k = 3) were chosen due to the 

small number of responses, as conducting k-means with more groups in each data set resulted in 

at least one group with very small membership relative to the other groups.  

Table 12 gives the group sizes and their mean PC values. Group 1 had 35 members, with 

the most negative average values for PC1 “Concerns” (-1.82), PC2 “Motivators” (-1.03), and PC3 

40 

 
 
 
 
“Niche Minority Considerations” (-0.23) of the three groups, though the magnitude of the PC3 

averages is small. The negative values of all three PCs combined suggest that the motivations are 

strongly negatively weighted for Group 1, while the weight placed on the barriers is close to zero 

in magnitude. Group 1 is therefore least likely to find the motivations for offering halal meat and 

poultry  products  to  be  attractive  and  are  neutral  towards  the  barriers  to  offering  halal.  When 

comparing  their  opinions  to  the  operational  and  demographic  data  (Table  13),  the  findings  are 

logical;  these  are  the  smallest  businesses  on  average  and  contain  a  large  percentage  of  rural 

businesses, as well as a large percentage of non-halal businesses and a small percentage of halal 

businesses. Taken together, it is possible that these smaller, more rural businesses do not have the 

local market to sell halal products to or may not have the capacity to add another program to their 

establishment. As such, I would not expect these businesses to be strong candidates for adding a 

halal program to their operation in the future.   

Group  2  had  28  members,  with  the  most  positive  average  value  (1.56)  for  PC2 

“Motivators” of the three groups. The average values for Group 2’s PC1 “Concerns” (-0.42) and 

PC3 “Niche Minority Considerations” (0.07) are the mid-range value relative to Groups 2 and 3, 

though  PC3  is  relatively  small  in  magnitude  (Table  12).  Thus,  Group  2  was  most  strongly 

positively  influenced  by  the  motivations  and  negatively  influenced  by  the  barriers  in  PC2. 

However, Group 2’s PC1 magnitude indicates a slight counteracting negative influence to PC2. 

Therefore, Group 2 contains businesses that are most likely to find motivations for offering halal 

to be attractive and the barriers to be prohibitive. The conflicting negative and positive influence 

of PC2 and PC1 indicate the barriers are overall more prohibitive in Group 2’s decision-making 

than the motivations are attractive. Group 2 contains the highest percentage of retailers, the highest 

percentage of halal businesses, the lowest percentage of non-halal businesses, are the largest on 

41 

 
 
 
average, the most recently established, and are more likely to be in urban or suburban areas than 

the other two groups (Table 13). Since Group 2 has the highest percentage of halal businesses, it 

is likely that businesses in this group that do not currently offer halal products are good candidates 

to do so in the future, as they must have other qualities in common that align with halal market 

participation. Indeed, these businesses are more likely to be in areas with larger Muslim consumer 

populations  and  may  have  the  capacity  to  add  an  additional  program  to  their  operation. 

Additionally, retailing of halal products requires much less capital investment and knowledge of 

the halal slaughter process than what is required for slaughterhouses and processors, and therefore 

may allow these businesses to adopt halal more easily. However, as these businesses are slightly 

younger on average, they may not have the capital to support expanding into a new market, which 

may explain their larger concerns with the barriers to market entry. Additionally, it is worth noting 

that Group 2 is the smallest group, with only 13 non-halal businesses, meaning the proportion of 

businesses that are good candidates to adopt a halal program in the future is small relative to the 

overall sample of this study. 

Group 3 had 46 members, with the most positive values for PC1 “Concerns” (1.64), and 

mid-range values for PC2 “Motivators” (-0.17) and PC3 “Niche Minority Considerations” (0.13) 

that are relatively small in magnitude. Thus, Group 3 was most strongly influenced by PC1, with 

slight  counteracting  effects  from  PC2  and  PC3.  Group  3  therefore  contains  businesses  that  are 

likely to find the motivations for halal to be attractive and the barriers to be prohibitive, with the 

influence of the motivations and barriers fairly equal and opposite each other relative to Group 2. 

Group 3’s businesses are on average the oldest of the three groups, have the highest percentage of 

processors, and a relatively high percentage of retailers. Retailing and processing halal products is 

logistically  simpler  than  slaughtering,  which  would  make  adoption  of  a  halal  program  more 

42 

 
 
straightforward for these businesses than groups with more slaughterhouses. Additionally, older 

businesses are likely well-established and could have more access to capital resources to expand 

their operations. Thus, these businesses may be good candidates for expanding into the halal meat 

and  poultry  business  in  the  future  if  given  enough  information  on  the  motivations  and  barriers 

specific to their operations (e.g., a feasibility study).  

Table 13 gives operation and respondent demographic percentages across the three groups 

of businesses. Note that not all percentages sum to 100% within a group’s category (e.g., race), as 

some respondents declined to answer all survey questions. From the F-tests and resulting t-tests, I 

find that there exist significant differences for non-halal, current halal, only processors, number of 

employees,  at  most  high  school  diploma  holders,  and  Asians.  There  are  statistically  significant 

differences at the 1% level between Groups 1 and 2 for the percentages of non-halal and current 

halal businesses, and a difference at the 5% level for the number of employees. There also exists 

a  statistically  significant  difference  between  Groups  1  and  3  at  the  5%  level  for  high  school 

diploma holders. Finally, there exists a statistically significant difference between Groups 2 and 3 

at the 1% for halal and non-halal businesses, and at the 5% level for number of employees, Asian 

respondents,  and  high  school  diploma  holders.  Altogether,  I  see  the  most  distinct  operational 

demographic patterns between Groups 1 and 3 versus Group 2.  

Table 12: K-Means Groups and Means of PC1, PC2, and PC3, Retailers and Processors, n 
= 109 

Group 
1 
2 
3 

Number of Members  Mean of PC1  Mean of PC2  Mean of PC3 

35 
28 
46 

-1.82 
-0.42 
1.64 

-1.03 
1.56 
-0.17 

-0.23 
0.07 
0.13 

43 

 
 
 
 
 
 
 
 
Table 13: Demographic Composition of Groups, n = 109 

Demographics 

Group 1 
(n = 35) 

Group 2 
(n = 28) 

Group 3 
(n = 46) 

85.71%a*** 
14.29% a*** 
34.29% 
5.71% 
28.57% 
25.71% 
1991 
28a** 
51.43% 
20.00% 
8.57% 

86.96% c*** 
13.04% c*** 
21.74% 
2.17% 
36.96% 
34.78% 
1986 
32 c** 
52.17% 
23.91% 
8.70% 

46.43%a***, c*** 
53.57% a***, c*** 
32.14% 
7.14% 
21.43% 
35.71% 
1993 
63a**, c** 
39.29% 
32.14% 
17.86% 

Operation Demographics 
Non-Halal Businesses 
Current Halal Businesses 
Slaughter and Process 
Only Slaughter 
Only Process 
Retailer 
Year Established 
Number of Employees 
Rural 
Suburban 
Urban 
Respondent Demographics 
White 
78.26% 
Black 
0.00% 
0.00% c** 
Asian 
Native American or Alaskan Native 
2.17% 
Hispanic 
4.35% 
Democrat 
4.35% 
Republican 
32.61% 
Independent 
28.26% 
First or Second-Generation Immigrant 
4.35% 
Female 
23.91% 
0.00%b**, c** 
High School 
Bachelor’s 
32.61% 
Graduate Degree 
21.74% 
Religious 
50.00% 
Notes: F-tests conducted on all three groups and statistically significant findings evaluated with pairwise t-
tests. Statistical significance between groups from the pairwise t-tests is indicated as follows: a indicates 
significance between Groups 1 and 2, b indicates significance between Groups 1 and 3, and c indicates 
significance between Groups 2 and 3. Additionally, *** indicates p < 0.01, ** indicates p < 0.05, and * 
indicates p < 0.1. 

75.00% 
3.57% 
10.71% c** 
0.00% 
3.57% 
10.71% 
28.57% 
32.14% 
10.71% 
10.71% 
10.71% c** 
35.71% 
14.29% 
42.86% 

62.86% 
0.00% 
5.71% 
2.86% 
0.00% 
5.71% 
25.71% 
22.86% 
5.71% 
20.00% 
11.43%b** 
25.71% 
17.14% 
54.29% 

7. Conclusions, Limitations, and Future Work 

This  study  utilized  a  mixed  methods  approach  consisting  of  qualitative  interviews  and 

nationwide online surveys with U.S. meat and poultry processors and retailers. The purpose of this 

research  was  to  investigate  the  decision-making  process  for  offering  halal  meat  and  poultry 

44 

 
 
 
 
 
 
 
 
 
products; specifically, the incentives and barriers that exist when entering the halal meat market 

and  how  retailers  and  processors  perceive  them.  The  goal  of  this  research  was  to  determine 

characteristics of businesses that are mostly likely to adopt a halal program in the future and how 

to facilitate their market participation. Both halal and non-halal businesses were sorted into three 

groups via principal components analysis (PCA) and non-hierarchical k-means clustering using 

the  data  collected  on  the  motivations  and  barriers  related  to  offering  [certified]  halal  meat  and 

poultry products.  

 Altogether, my analysis sheds light on potential avenues for supply-side expansion of the 

U.S. domestic halal meat and poultry industry. By studying patterns in businesses’ perceptions of 

the  motivations  and  barriers  and  operational  demographics,  I  can  determine  common 

characteristics that lend themselves well to the adoption of a halal program. Indeed, the analysis 

shows that urban, suburban, retail, younger, and larger businesses are more likely to currently have 

a halal program (Group 2), while businesses that may expand into the halal market in the future 

are older and more likely to be retailers or processors instead of slaughterhouses. As retailing and 

processing of halal meat and poultry products is generally less capital intensive and requires less 

knowledge and skill to implement, these businesses are likely the strongest candidates to expand 

into  halal  in  the  future.  Furthermore,  if  this  expansion  in  the  downstream  supply  chain  can  be 

achieved, it may make adding a halal program upstream to existing slaughtering facilities more 

attractive and provide demand-driven pressure for halal slaughterhouse market entry. 

There are a few notable limitations of this study. First, collecting data from supply-side 

agents  in  any  market  is  notoriously  difficult,  which  leads  to  lower  response  rates  and  lower 

numbers of quality observations in our dataset. Thus, more complex statistical analyses were not 

possible. Second, there are relatively few halal meat and poultry businesses in the U.S. compared 

45 

 
 
to traditional meat and poultry businesses, so halal businesses are not equally represented in our 

survey  sampling.  Finally,  even  though  efforts  were  made  to  survey  participants  in  their  native 

language, it is likely that some halal businesses did not take the survey due to English language 

barriers.  

There is still need for more research focused on the U.S. domestic halal meat supply chain, 

and  future  studies  can  expand  on  this  work  in  a  few  ways.  As  my  analysis  was  exploratory  in 

nature, additional research can and should be done into the noteworthy motivations and barriers 

for  the  different  types  of  businesses  in  this  study  –  halal  retailers,  halal  processors,  non-halal 

retailers,  and  non-halal  processors.  A  deeper  understanding  of  these  businesses’  differences  in 

perceptions of adopting a halal program would help design more effective policies and incentive 

structures for a more robust halal meat supply chain. To achieve this, future work would benefit 

from larger sample sizes to facilitate more advanced statistical techniques for data analysis. There 

are two methods that I believe could be effective for increasing sample size when working with 

these populations. First, utilizing a team of researchers to conduct in-person or virtual (e.g., Zoom 

or phone) surveys in real-time may increase response rates and quality, as processors and retailers 

are typically less likely to complete surveys. Secondly, some of the U.S. Muslim community are 

nonnative  English  speakers,  especially  older  individuals  and  recent  immigrants.  As  such, 

researchers  may  benefit  from  close  partnerships  with  native  Arabic  and  Urdu  speakers  when 

collecting data from halal businesses to increase participation rates. Altogether, these suggestions 

will  help  future  work  make  meaningful  contributions  to  our  understanding  and  support  of  this 

unique market, as well as add to the literature on meat and poultry businesses’ decision making 

and design of policy to support supply chain development. 

46 

 
 
 
 
BIBLIOGRAPHY 

Ab Rashid, N., & Bojei, J. (2019). The relationship between halal traceability system adoption and 
environmental factors on halal food supply chain integrity in Malaysia. Journal of Islamic 
Marketing, 11(1), 117–142. https://doi.org/10.1108/JIMA-01-2018-0016 

Ahmed, A. (2008). Marketing of halal meat in the United Kingdom: Supermarkets versus local shops. 

British Food Journal, 110(7), 655–670. https://doi.org/10.1108/00070700810887149 

Arabie, P., & Hubert, L. (1996). Advances in Cluster Analysis Relevant to Marketing Research. In W. 
Gaul & D. Pfeifer (Eds.), From Data to Knowledge (pp. 3–19). Springer Berlin Heidelberg. 
https://doi.org/10.1007/978-3-642-79999-0_1 

Bejaei, M., Cliff, M. A., & Singh, A. (2020). Multiple Correspondence and Hierarchical Cluster 
Analyses for the Profiling of Fresh Apple Customers Using Data from Two Marketplaces. 
Foods, 9(7), 873. https://doi.org/10.3390/foods9070873 

Bergeaud-Blackler, F. (2004). Social definitions of halal quality: the case of Maghrebi Muslims in 

France. 

Bonne, K., & Verbeke, W. (2007). Religious values informing halal meat production and the control 

and delivery of halal credence quality. Agriculture and Human Values, 25(1), 35–47. 
https://doi.org/10.1007/s10460-007-9076-y. 

Bonne, K., & Verbeke, W. (2008). Muslim consumer trust in halal meat status and control in 
Belgium. Meat Science, 79(1), 113–123. https://doi.org/10.1016/j.meatsci.2007.08.007. 

Bonne, K., Vermeir, I., & Verbeke, W. (2008). Impact of Religion on Halal Meat Consumption 

Decision Making in Belgium. Journal of International Food & Agribusiness Marketing, 21(1), 
5–26. https://doi.org/10.1080/08974430802480628. 

Calder, A. J., Burton, A. M., Miller, P., Young, A. W., & Akamatsu, S. (2001). A principal 

component analysis of facial expressions. Vision Research, 41(9), 1179–1208. 
https://doi.org/10.1016/S0042-6989(01)00002-5 

Campbell, H., Murcott, A., & MacKenzie, A. (2011). Kosher in New York City, halal in Aquitaine: 
Challenging the relationship between neoliberalism and food auditing. Agriculture and Human 
Values, 28(1), 67–79. https://doi.org/10.1007/s10460-010-9260-3 

Carmichael, L. (2020). Muslims in America. Zwemer Center for Muslim Studies. 

https://www.zwemercenter.com/muslims-in-america/  

Choe, J. (2023). Outlook for Livestock and Poultry in 2022. USDA’s 99th Agricultural Outlook 

Forum. United States Department of Agriculture. 
https://www.usda.gov/sites/default/files/documents/2023AOF-livestock-poultry-outlook.pdf  

47 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Fuseini, A., Hadley, P., & Knowles, T. (2021). Halal food marketing: An evaluation of UK halal 

standards. Journal of Islamic Marketing, 12(5), 977–991. https://doi.org/10.1108/JIMA-02-2020-
0037 

Fuseini, A., Wotton, S. B., Knowles, T. G., & Hadley, P. J. (2017). Halal Meat Fraud and Safety 

Issues in the UK: A Review in the Context of the European Union. Food Ethics, 1(2), 127–142. 
https://doi.org/10.1007/s41055-017-0009-1 

de Araújo, S.H. (2019). Assembling halal meat and poultry production in Brazil: Agents, practices, 

power, and sites. Geoforum, 100, 220–228. https://doi.org/10.1016/j.geoforum.2019.01.014 

Durán, K., & Pipes, D. (2002). Muslim Immigrants in the United States. Center for Immigration 

Studies. CIS.org. https://cis.org/Report/Muslim-Immigrants-United-States 

Hanson, G. D., Rauniyar, G. P., & Herrmann, R. O. (1994). Using consumer profiles to increase the 

U.S. market for seafood: Implications for aquaculture. Aquaculture, 127(4), 303–316. 
https://doi.org/10.1016/0044-8486(94)90233-X 

Hobbs, J. E. (2021). The Covid-19 pandemic and meat supply chains. Meat Science, 181, 108459. 

https://doi.org/10.1016/j.meatsci.2021.108459 

Hotelling, H. (1933). Analysis of a complex of statistical variables into principal components. Journal 

of Educational Psychology, 24(6), 417–441. https://doi.org/10.1037/h0071325 

Hsu, F.-C., Kritchevsky, S. B., Liu, Y., Kanaya, A., Newman, A. B., Perry, S. E., Visser, M., Pahor, 
M., Harris, T. B., Nicklas, B. J., & for the Health ABC Study. (2009). Association Between 
Inflammatory Components and Physical Function in the Health, Aging, and Body Composition 
Study: A Principal Component Analysis Approach. The Journals of Gerontology Series A: 
Biological Sciences and Medical Sciences, 64A (5), 581–589. 
https://doi.org/10.1093/gerona/glp005 

Ijaz, M., Yar, M. K., Badar, I. H., Ali, S., Islam, Md. S., Jaspal, M. H., Hayat, Z., Sardar, A., Ullah, 

S., & Guevara-Ruiz, D. (2021). Meat Production and Supply Chain Under COVID-19 Scenario: 
Current Trends and Future Prospects. Frontiers in Veterinary Science, 8, 660736. 
https://doi.org/10.3389/fvets.2021.660736 

Larue, B. (2021). COVID‐19 and labor issues: An assessment. Canadian Journal of Agricultural 

Economics/Revue Canadienne d’agroeconomie, 69(2), 269–279. 
https://doi.org/10.1111/cjag.12288 

Lever, J., & Miele, M. (2012). The growth of halal meat markets in Europe: An exploration of the 

supply side theory of religion. Journal of Rural Studies, 28(4), 528–537. 
https://doi.org/10.1016/j.jrurstud.2012.06.004 

Lever, J., Puig De La Bellacasa, M., Miele, M., & Higgin, M. (2010). From the Slaughterhouse to the 
Consumer: Transparency and information in the distribution of Halal and Kosher meat. Dialrel. 

48 

 
 
 
 
 
 
 
 
 
 
 
 
 
Malone, T., & Lusk, J. L. (2018). If you brew it, who will come? Market segments in the U.S. beer 

market. Agribusiness, 34(2), 204–221. https://doi.org/10.1002/agr.21511 

Lopez, M. H. (2017, December 20). Hispanic identity fades across generations as immigrant 

connections fall away. Pew Research Center’s Hispanic Trends Project. 
https://www.pewresearch.org/hispanic/2017/12/20/hispanic-identity-fades-across-generations-as-
immigrant-connections-fall-away/  

Masudin, I., Jie, F., & Widayat, W. (2020). Impact of halal supplier service quality and staff readiness 
to adopt halal technology on halal logistics performance: A study of Indonesian halal meat 
supply chain. International Journal of Agile Systems and Management, 13(3), 315-338. 

Maxwell, J. A. (2012). Qualitative research design: An interactive approach. Sage Publications. 

Miles, M. B., Huberman, A. M., & Saldaña, J. (2020). Qualitative data analysis: A methods 

sourcebook (Fourth edition). SAGE. 

Mumuni, A. G., Veeck, A., Luqmani, M., Quraeshi, Z. A., & Kamarulzaman, Y. (2018). Religious 

identity, community, and religious minorities’ search efforts for religiously sanctioned food: The 
case of halal food in non-Muslim majority markets. International Journal of Consumer Studies, 
42(6), 586–598. https://doi.org/10.1111/ijcs.12423 

National Grocers Association. (2019). NGA Retail Membership List by State – Winter 2019. National 
Grocers  Association.  https://www.nationalgrocers.org/wp-content/uploads/2018/12/NGA-Retail-
Membership-List-Winter-2019-by-State.pdf  

Patton,  M.  Q.  (2014). Qualitative  research  &  evaluation  methods:  Integrating  theory  and  practice. 

Sage Publications. 

Pearson, K. (1901). LIII. On lines and planes of closest fit to systems of points in space. The London, 
Edinburgh,  and  Dublin  Philosophical  Magazine  and  Journal  of  Science,  2(11),  559–572. 
https://doi.org/10.1080/14786440109462720 

Public Religion Research Institute. (2023, November 3). The 2020 PRRI Census of American religion. 
https://www.prri.org/research/2020-census-of-american-religion/#:~:text=Muslim%20 

PRRI. 
Americans%20are%20primarily%20concentrated,County%2C%20New%20York%20(4%25) 

Rubin, H. J., & Rubin, I. S. (2011). Qualitative interviewing: The art of hearing data. Sage Publications. 

Saldaña, J. (2013). The coding manual for qualitative researchers (2. ed). SAGE Publ. 

Salindal, N. A. (2019). Halal certification compliance and its effects on companies’ innovative and 

market performance. Journal of Islamic Marketing, 10(2), 589–605. 
https://doi.org/10.1108/JIMA-04-2018-0080 

49 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Shahijan, M.K., Rezaei, S., Nigel Preece, C., & Khairuzzaman Wan Ismail, W. (2014). Examining 

retailers’ behaviour in managing critical points in Halal meat handling: A PLS analysis. Journal 
of Islamic Marketing, 5(3), 446–472. https://doi.org/10.1108/JIMA-10-2013-0070 

Sinha, R. N., Wallace, H. A. H., & Chebib, F. S. (1969). Principal-Component Analysis of 

Interrelations Among Fungi, Mites, and Insects in Grain Bulk Ecosystems. Ecology, 50(4), 536–
547. https://doi.org/10.2307/1936244 

Stata. (2013, February 11). Statistical Software for data science | Stata. Stata.com. Retrieved August 

8, 2022, from https://www.stata.com/manuals13/mvfactor.pdf  

Taylor, C. A., Boulos, C., & Almond, D. (2020). Livestock plants and COVID-19 transmission. 

Proceedings of the National Academy of Sciences, 117(50), 31706–31715. 
https://doi.org/10.1073/pnas.2010115117 

Tieman, M., van der Vorst, J. G. A. J., & Che Ghazali, M. (2012). Principles in halal supply chain 

management. Journal of Islamic Marketing, 3(3), 217–243. 
https://doi.org/10.1108/17590831211259727 

United States Department of Agriculture (USDA) Agricultural Marketing Service (AMS). 

(2022). Meat and poultry inspection readiness grant. Meat and Poultry Inspection Readiness 
Grant | Agricultural Marketing Service. Retrieved August 5, 2022, from 
https://www.ams.usda.gov/services/grants/mpirg  

Verbeke, W., Rutsaert, P., Bonne, K., & Vermeir, I. (2013). Credence quality coordination and 

consumers’ willingness-to-pay for certified halal labelled meat. Meat Science, 95(4), 790–797. 

Weersink, A., Von Massow, M., Bannon, N., Ifft, J., Maples, J., McEwan, K., McKendree, M. G. S., 

Nicholson, C., Novakovic, A., Rangarajan, A., Richards, T., Rickard, B., Rude, J., Schipanski, 
M., Schnitkey, G., Schulz, L., Schuurman, D., Schwartzkopf-Genswein, K., Stephenson, M., … 
Wood, K. (2021). COVID-19 and the agri-food system in the United States and Canada. 
Agricultural Systems, 188, 103039. https://doi.org/10.1016/j.agsy.2020.103039 

Winship, C., & Mare, R. D. (1984). Regression Models with Ordinal Variables. American 

Sociological Review, 49(4), 512. https://doi.org/10.2307/2095465 

Zulfakar, M. H., Jie, F., & Chan, C. (2013, December). Critical success factors for a successful 

implementation of halal red meat supply chain in Australia: meat processor’s perspective. In 11th 
ANZAM Operations, Supply Chain and Services Management Symposium, Melbourne (pp. 20-
21). 

Zulfakar, M. H., Chan, C., & Jie, F. (2018). Institutional forces on Australian halal meat supply chain 
(AHMSC) operations. Journal of Islamic Marketing, 9(1), 80–98. https://doi.org/10.1108/JIMA-
01-2016-0005 

50 

 
 
 
 
 
 
 
 
 
 
 
 
 
Zulfakar, M. H., Chan, C., Jie, F., & Sundram, V. P. K. (2019). Halal accreditation and certification in 
a non-Muslim country setting: Insights from Australia halal meat supply chain. International 
Journal of Supply Chain Management, 8(1), 10-17. 

51 

 
 
 
 
 
CHAPTER 2. MARKET PARTICIPANTS’ PREFERENCES FOR 
A NATIONAL HALAL MEAT CERTIFICATION PROGRAM 

1. Introduction and Motivation 

There  are  over  200  different  halal  food  certification  organizations  around  the  globe 

(Zabihah  2021,  Verify  Halal  2022),  and  in  the  U.S.  alone,  there  are  over  20  verified  halal 

certification organizations. Given that the global Muslim population does not have a universally 

accepted  and  verified  halal  meat  standard  and  the  resulting  astoundingly  large  number  of 

certifications  and  certifying  organizations  operating  in  the  market,  it  is  unsurprising  that  halal 

foods are the fourth most likely food in the U.S. to be fraudulent (FSNS 2020). In most cases, halal 

meat and poultry certification organizations have their own internal halal standards that they use 

to evaluate and then grant certification; these depend on how strictly they interpret the religious 

laws from The Qur’an and the teachings of the Prophet Muhammad (peace be upon him) written 

in  The  Hadith  that  govern  halal  meat.  The  plethora  of  distinct  and  sometimes  contradictory 

certifications  can  make  selecting  a  product  confusing  for  consumers,  accurately  sharing 

information  with  customers  difficult  for  retailers,  and  determining  requirements  for  halal  meat 

processing challenging for processors.  

There is an overabundance and a lack of clarity and standardization in certifications in the 

U.S. halal meat and poultry market. The current halal meat and poultry market situation is similar 

to the U.S. organic market prior to the development of the USDA organic certification, when there 

were many different and competing organic certifications available for producers, and consumers 

necessarily  incurred  search  and  information  collection  costs  to  determine  which  labels  and 

standards  were  on  products  (Lohr  1998).  As  in  the  case  of  the  pre-USDA  certification  organic 

market,  this  likely  makes  it  challenging  for  participants  to  engage  fully  in  the  halal  meat  and 

poultry market and reduces the domestic market’s competitiveness relative to imported products 

52 

 
 
 
(Lohr  1998).  Further,  an  overabundance  of  certification  and  regulations  serves  as  a  barrier  to 

market entry for specialty meat products (Worosz et. al 2008) – especially in the case of smaller 

or  new  operations,  as  previously  seen  in  the  organic  farming  industry  (Guthman  1998).  These 

barriers result in a loss of market efficiency, and in the long term, a harm to consumers who are 

thereby  underserved.  Thus,  this  research  aims  to  undertake  the  first  step  in  the  process  of 

developing a U.S. NHMC program to help streamline market participation. 

These shared concerns suggest there is room for a verification strategy in the form of a 

unified U.S. national halal meat and poultry certification (NHMC) program. Correspondingly, the 

purpose of this study is to understand what the U.S. domestic halal market participants want out 

of a NHMC program, and how these desires compare to what is feasible to implement. Therefore, 

to determine what such a program would look like and how it would operate, this chapter addresses 

three major research questions. First, what characteristics of a hypothetical NHMC are important 

to halal meat processors, halal meat retailers, and Muslim halal meat consumers? Second, are there 

differences in preferences for NHMC program characteristics between groups, and if so, what is 

the nature of these differences? And finally, how could differences in preferences between groups 

impact  the  design  and  implementation  of  a  NHMC  program?  This  research  will  answer  these 

questions  by  investigating  the  preferences  for  a  NHMC  program  at  three  levels  of  the  market 

(processors, retailers, and Muslim halal meat consumers). The differences in preferences across 

groups will be compared to each other and will serve as a baseline understanding of the market’s 

desires, which future research can expand upon. 

In all, this research contributes to the literature on labeling and certification for improving 

trust  in  products  with  credence  attributes.  This  chapter  provides  an  overview  of  the  variety  of 

different halal meat certifications, certification programs, and certifying bodies around the world, 

53 

 
 
with  a  direct  comparison  to  the  current  U.S.  halal  meat  certification  landscape,  which  to  the 

author’s knowledge has not been explicitly explored before. Altogether, the findings of this chapter 

will aid in bolstering halal meat consumer confidence in the authenticity of their products, as well 

as improve the equity of the U.S. food system. 

2. Current Landscape of U.S. Domestic Halal Certifications 

As  early  as  the  1970s  and  1980s,  U.S.  Muslim  consumers  –  concerned  about  the 

authenticity of the halal meat and poultry products they were buying – began to demand more 

verification  for  their  products  in  the  form  of  third-party  labels  and  certifications.  From  this 

movement, certifying organizations were established; these organizations typically charge a fee – 

per unit, an annual lump sum, or some formulaic combination – to inspect slaughter, processing, 

and retail facilities and provide certification that their products and/or establishment conforms to 

a given set of halal standards.  

As the demand for halal meat and poultry products has grown and evolved over the past 50 

years,  so  too  has  the  number  of  certifiers,  the  variety  of  certifications,  and  the  ways  in  which 

certifiers  operate.  Today,  there  are  several  major  halal  meat  and  poultry  certifiers  and  another 

dozen or so smaller local or regional certification agencies operating in the U.S.5 These certifiers 

are broken into two organizational categories: non-profit foundations that rely mainly on volunteer 

work and donations from the public to operate, and for-profit businesses which charge fees for 

their  services  and  retain  employees  to  conduct  the  certification  process.6  The  non-profit 

foundations include Halal Monitoring Services (HMS) of Shariah Board of America and Halal 

Food Standards Alliance of America (HFSAA), among others. The for-profit businesses include 

5 Not all of these certifiers work exclusively with meat and poultry products. 
6 These classifications are based upon what the author learned in qualitative interviews with halal meat and poultry 
certifiers during this project and are not necessarily mutually exclusive nor immutable. 

54 

 
 
 
 
Islamic  Services  of  America  (ISA),  Islamic  Society  of  the  Washington  Area  (ISWA),  Halal 

Transactions of Omaha (HTO), Islamic Food and Nutrition Council of America (IFANCA), and 

American Halal Foundation (AHF), among others. 

Many  of  the  for-profit  certifiers  have  certification  via  the  United  States  Department  of 

Agriculture’s  Food  Safety  and  Inspection  Service  (USDA-FSIS)  to  operate  in  the  import  and 

export markets, while the non-profit foundations generally do not.7 For-profit certifiers offer both 

zabiha and machine-slaughtered halal certifications, while the non-profit certifiers focus mainly 

on the zabiha market.8 Correspondingly, for-profit certifiers typically contract with larger domestic 

and multinational corporations and establishments looking for international business opportunities, 

while non-profits operate mainly in the domestic small and local business arena. Thus, for-profit 

certifiers currently make up most of the certification market by both quantity of products certified 

and revenues received, therefore holding the majority of power in the halal certification market. 

However, the non-profit certifiers work with a larger number of domestic businesses, though the 

quantity of products and overall revenues is lower. 

As with any market, power dynamics and market structure play a pivotal role in market 

growth, policy development, and the implementation of new methods of production.9 Indeed, the 

U.S. domestic halal meat and poultry certification market is no exception. According to multiple 

certifiers and Muslim consumer advocacy groups, efforts have been underway for over a decade 

to  create  a  set  of  industry-wide  U.S.  national  halal  standards,  though  progress  in  reaching 

agreements on what specifications and requirements to include has been slow. For-profit and non-

7 These certifications are a service provided by the Food Safety and Inspection Service (FSIS) of the United States 
Department of Agriculture (USDA) and requires businesses to pay a fee to obtain export certification and/or import 
products from foreign countries. 
8 The term “zabiha” refers to hand-slaughtered animal products. More information is given in the preface to this 
dissertation. 
9 Historically, markets with an imbalance of power (e.g., those with monopolies or oligopolies) see declines in 
growth, equity, innovation, and efficiency losses (Washington Center for Equitable Growth 2018). 

55 

 
 
 
profit certifiers commonly disagree on which standards should be included in a U.S. national halal 

certification, as well as how such a certification should be implemented and verified. What’s more, 

Muslim  halal  consumers  have  varying  opinions  due  to  the  extremely  diverse  nature  of  the 

population  in  terms  of  ethnicity,  religious  sect,  immigration  and  citizenship  status,  and  other 

cultural factors. The dissenting views on halal certification from both the supply and demand side 

of the market contribute to the ongoing struggle to develop a single cohesive certification program.  

Further complicating the matter are the views of the general public and federal and state 

government representatives. Islam is a minority religion in the U.S., and multiple domestic and 

international events over the past several decades have reinforced prejudices that negatively impact 

U.S. Muslims; indeed, a Pew Research poll revealed that a significant percentage of Americans 

believe Muslims face “a lot of” discrimination and are viewed more negatively than other religions 

(Mohammed 2021). As such, it is unlikely that many government representatives will move to 

make changes to state or federal laws to improve halal verification and transparency, nor support 

the development of a national-level halal meat and poultry certification. Instead, general efforts to 

increase  authenticity  in  the  food  system,  strengthen  traceability,  and  improve  labeling 

requirements are more realistic avenues for addressing halal certification issues. 

2.1 Legal Enforcement of Certifications in the U.S. 

In the U.S., there are a lack of labeling regulations in place to ensure that consumers know 

which standards are involved in the production of food products. Outside of a handful of states 

that have instituted forms of consumer-right-to-know legislation (California, Illinois, Maryland, 

Michigan, Minnesota, New  Jersey, New York,  Texas,  and Virginia), there is no federal law or 

regulation that requires a business that claims to be certified to supply a given specialty product to 

prove that they are certified by displaying the certificate or registering with the state department 

56 

 
 
of  agriculture.  Furthermore,  these  state  laws  that  require  proof  of  certification  to  be  readily 

available  often  are  not  enforced.  Altogether,  there  is  ample  room  for  fraudulent  behavior  by 

uncertified actors throughout the U.S.; this is likely a large factor contributing to halal food being 

the fourth most fraudulent industry in the U.S.  

When  looking  to  certify  an  attribute,  one  of  the  main  questions  is  who  should  be  the 

certifying body. Many studies have focused on preferences for certification entities and have found 

that  government  certification  is  in  general  more  trusted  than  third-party  certification  in  a  wide 

variety of contexts; see for example McKendree et al. (2013), Johnston et al. (2001), Ortega et al. 

(2011, 2012), and Sønderskov & Daugbjerg (2011). However, for the halal market in the U.S., the 

First Amendment prohibits the government from setting any religious product standards, as this 

would  be  an  infringement  of  religious  freedoms.  California,  Illinois,  Maryland,  Michigan, 

Minnesota, New Jersey, New York, Texas, and Virginia have all circumvented these constitutional 

rights  concerns  by  enacting  laws  in  the  spirit  of  consumer-right-to-know  considerations.  These 

include allowing the state to enforce the proper use of halal labels, including outlawing mislabeling 

and false representation, requiring businesses to display their halal meat Department of Agriculture 

registration,  and  requiring  separate  facilities/machinery  for  halal  products  to  ensure  no  cross 

contamination can occur (Illinois Halal Food Act 2002). A similar law exists in many states for 

kosher meat. A federal law and enforcement akin to these states’ laws would provide another layer 

of authentication to prevent halal food fraud. 

3. Halal Certification Around the World 

Of the plethora of certification programs operating around the world, there are several that 

oversee  halal  production  standards  on  a  national  or  even  multi-national  scope  –  such  as  the 

Department of Islamic Development Malaysia (JAKIM) program or the Australian national halal 

57 

 
 
 
meat program (Rhaman et al. 2018, Australian Halal 2022). Though the U.S. does not have the 

same  institutional  framework  as  these  countries,  lessons  can  be  learned  from  national-level 

programs already in place.  

In countries like Australia, Singapore, the United Arab Emirates (UAE), and Malaysia, the 

national halal meat certification (NHMC) standard is set by, and administered through, the federal 

government. This allows for a generalization of the halal meat standards within these countries; 

there is only one halal meat standard or a set of predetermined levels of halal meat standards. This 

makes adoption by supply-side agents in the halal meat market straightforward, as they can either 

adhere to the universal standard or not – there is no room for interpretation or for confusion as to 

which certification a supplier needs to enter the market. This is also advantageous for consumers, 

as  the  halal  meat  standards  will  be  identical  across  the  country,  so  there  is  no  room  for 

miscommunication or confusion with competing labels or standards. On the flip side, this does 

remove diversity from the halal meat market, and so consumers who desire different standards for 

halal meat products may still need to look elsewhere for their products. While a program in which 

local,  state,  or  federal  government  sets  halal  standards  may  not  be  possible  in  the  U.S.  due  to 

concerns about First Amendment violations, the structure and development of a national program 

is  still  useful  to  understand.  Furthermore,  a  U.S.  national  program  could  potentially  have 

government involvement in the enforcing of standards set by non-governmental organizations. 

In contrast, in other countries, there is no national halal program, leaving room for multiple 

third-party  agents,  such  as  Islamic  community  groups,  imams  (faith  leaders  in  Islam),  and 

certifying corporations to design their own halal meat standards and certification implementation 

processes. This system is akin to what is currently in place in the U.S., Canada, and most other 

countries across the globe. This system can be very confusing to consumers and supply-side agents 

58 

 
 
alike, especially when there is little to no transparency in the standards required to obtain a given 

certification.  Consumers  are  faced  with  an  overload  of  information  and  sorting  out  which 

certifications align with their beliefs is mentally taxing. Supply-side agents have other issues – 

selecting  one  certification  to  use  in  their  operation  may  preclude  them  from  some  market 

opportunities or result in backlash from consumer groups who do not agree with the standards they 

follow. These concerns are especially relevant in the U.S., as, to the author’s knowledge, the U.S. 

has the largest number of halal meat certifiers and certifications currently available in the market 

of any country in the world. However, this assortment of certifications can also be a boon – for 

consumers with very niche requirements, they may be able to find a certification that aligns with 

their beliefs, whereas a universal standard may not align. Altogether, the future of U.S. domestic 

halal certification is complex, and understanding and addressing the market’s needs will require 

significant effort on behalf of researchers and policy makers alike. 

3. Mixed Methods and Survey Design 

As  discussed,  the  U.S.  domestic  halal  meat  and  poultry  market  has  distinctive  niche 

religious requirements that must be upheld. The nature of the halal market, coupled with the lack 

of  research  to  understand  the  certification  landscape  of  the  U.S.  halal  meat  market  poses  an 

exciting opportunity to focus this study on multiple groups throughout the domestic market. This 

ensures that the main actors in the halal meat market that would be impacted by a certification – 

consumers, retailers, and processors – are included in the study. This understudied market also 

requires a methodological approach designed for exploration and learning feedback. Therefore, I 

employ a mixed methods research design using both qualitative interviews and quantitative survey 

methods in this project. The synergy between methods is exhibited in two ways. In the first stage 

of this project, I conducted qualitative interviews with halal meat processors, retailers, and Muslim 

59 

 
 
consumers. These qualitative interviews were necessary to develop an understanding of and elicit 

opinions on halal meat market certification practices in the U.S. domestic market. I used the data 

from these conversations to narrow my research questions and design quantitative data collection 

tools. From these interviews, I determined the most appropriate questions to explore in relation to 

halal  meat  and  poultry  certification  revolve  around  an  overabundance  of  certifications,  unclear 

standards,  and  a  lack  of  transparency  in  certification  use  and  enforcement.  These  qualitative 

interview  findings  in  turn  inform  the  quantitative  components  of  the  study.  In  particular,  the 

patterns uncovered in the qualitative research indicated that the quantitative data collection should 

utilize best-worst scaling (BWS) to measure processors’, retailers’, and consumers’ preferences 

for possible attributes of such a program. 

 3.1 Phase 1: Qualitative Interview Methods 

The  qualitative  portion  of  this  project  was  designed  with  methods  from  Patton  (2014), 

Rubin and Rubin (2011), and Maxwell (2012). The purpose of these qualitative interviews was to 

obtain  information  about  the  U.S.  domestic  halal  meat  and  poultry  industry’s  certification 

standards  and  process  previously  unknown  to  me,  as  this  market  is  understudied.  Information 

collected via qualitative interviews included Muslim halal consumers’, halal retailers’, and halal 

processors’  opinions  on  the  current  certification  landscape  and  their  preferences  for  future 

certification  development.  Collecting  information  directly  from  market  participants  in  this  way 

allowed me to design a more robust and relevant research program for the quantitative portion of 

this project. The qualitative interviews also provided vital context and explanatory power for the 

quantitative survey findings. Interview questions were grouped by topic and were open-ended to 

allow  for  robust  answers  (Patton  2014,  Rubin  and  Rubin  2011,  Maxwell  2012).  The  interview 

60 

 
 
guide for these interviews, information for how interviews were conducted, and the outline of the 

interview analysis process are found in Appendix A.1. 

Supply-side interviewees were recruited from lists of retailers and processors registered as 

certified to supply halal meat by a reputable halal meat certifier. The lists of supply-side interview 

candidates were narrowed to those in Midwestern states for ease of access and improved likelihood 

of  name  recognition  for  Michigan  State  University.  A  series  of  eight  in-depth,  semi-structured 

interviews with Midwestern halal meat retailers and 10 in-depth, semi-structured interviews with 

Midwestern halal meat processors were conducted December 2021 through April 2022. Supply-

side  interviews  were  conducted  via  Zoom  or  over  the  phone.  Consumer  interviewees  were 

recruited  in  person  from  five  different  certified  halal  meat  retailers  in  Illinois,  Wisconsin,  and 

Michigan; the list of interested participants was then randomly sampled from. Selected consumers 

were interviewed over Zoom or via phone and were compensated for their participation with a $50 

gift card. In total, 12 consumer interviews were conducted in April 2022 through June 2022. 

3.2 Phase 2: Quantitative Survey Methods 

The second stage of this chapter involves conducting three stacked national online surveys, 

one  with  a  sample  of  meat  processors,  one  with  a  sample  of  meat  retailers,  and  another  with 

Muslim halal meat consumers.10 These parallel stacked surveys are designed to assess preferences 

related to the development of a hypothetical U.S. NHMC program. The surveys had three major 

purposes: first, to further explore the patterns uncovered in the qualitative interviews; second, to 

be able to gauge these groups’ preferences and opinions at a national level; and third, to directly 

compare preferences and response patterns across the three market groups – a key advantage of 

stacked surveys. This method is particularly attractive for application in this study, as consumers, 

10 Surveys were available in English, Arabic, and Urdu for all study participants. 

61 

 
 
 
retailers, and processors are the main actors in the halal meat market that would be impacted by a 

certification  and  should  therefore  be  included.  Additionally,  the  shared  desire  for  a  NHMC 

program across the three groups interviewed suggests that comparing preferences directly in this 

manner will be useful in determining consensus or differences in opinions. 

To compare preferences for attributes of a U.S. NHMC program directly across the three 

groups using a consistent and unitless ranking system, I use best-worst scaling (BWS), also known 

as maximum difference (max-diff) scaling or most-least scaling. BWS is a choice analysis method 

that  asks  participants  to  repeatedly  –  over  a  series  of  different  choice  sets  –  select  the  most 

preferred  (best)  and  least  preferred  (worst)  options  out  of  a  given  set  of  items  and  allows  the 

researcher to thereby understand preferences between items. BWS was introduced by Finn and 

Louviere (1992) with theoretical properties of probabilistic, best-worst choice models being more 

recently explained by Marley and Louviere (2005).  BWS is advantageous relative to Likert scale 

questions, as Likert scales make it challenging to distinguish the actual importance of attributes. 

For  example,  multiple  Likert  scale  questions  can  all  have  the  same  mean  level  of  importance. 

Likert scale questions are additionally problematic due to scale subjectivity – what is considered 

a “4” on one individual’s scale may be a “5” on another (Lusk and Briggeman 2009, Lusk and 

Parker 2009, Wolf and Tonsor 2013). Additionally, as BWS is a tradeoff method, I achieve shares 

of  importance  that  can  be  directly  interpreted  from  a  ratio  scale.  The  sum  of  shares  among  all 

attributes analyzed must equal one. If attributes j and k have importance shares of 0.2 and 0.1, 

respectively, attribute j is two times as important as k. BWS also provides five to ten times more 

differentiation than most scaling methods, such as the aforementioned Likert scales (Horne 2012). 

Directly  interpretable  shares  and  more  differentiation  provides  further  insight  into  the  exact 

importance of each attribute.   

62 

 
 
In each survey, participants were asked a series of seven BWS questions with different 

combinations of seven attributes of a potential future U.S. NHMC program. In these questions, 

participants were asked to select which of three attributes they thought was the most important and 

which was the least important to consider when designing a U.S. NHMC program. An example of 

one of the questions that participants saw is shown in Figure 5.  

Figure 5: Example Best-Worst Question for Potential National Halal Certification 

3.3 Attribute Selection for the Best-Worst Scaling Survey Instrument 

The BWS for this survey includes seven attributes for consideration. The attributes for the 

BWS questions were selected based on the characteristics of current certification programs around 

the  world  (both  within  and  outside  of  halal  meat),  discussion  in  the  literature,  and  information 

collected from interviews. Justification for these choices is broken down by attribute below. These 

attributes differ slightly in their precise language across the three surveyed groups to better fit the 

audience; attributes included here are cost, administrative body, traceability, information collected 

or  available,  certification  type,  audit  characteristics,  and  halal  type.  The  description  of  these 

attribute categories and the relevant questions for each surveyed group are presented in Table 14.   

63 

 
 
 
 
Table 14: Best-Worst Scaling Attributes and Question Language by Group 

Attribute wording by group 

Referred to in 
paper as 

Consumers 

Retailers 

Processors 

Costs 

Additional costs I would need to pay 
for certified products 

Costs associated with certification program involvement 
(certification fees, infrastructure, labor) 

Enforcement 

What organization enforces the program standards 

Benefits 

Confidence gained by purchasing a 
certified product 

Benefits associated with certification program involvement 
(access to new markets, price premiums, etc.) 

Information  What information is available to me 

What information is passed on to my customers 

e
t
u
b
i
r
t
t

A

Who/What 
Certified 

What or who should be required to be certified (products, retailers, slaughterhouses, etc.) 

Inspections 

How retailers and processors are 
audited/inspected 

How my suppliers’ & my operation 
are inspected 

How my operation is 
inspected 

Halal 
Standards 

What halal standards are required (hand- versus machine-slaughtered, stunning or no stunning, etc.) 

64 

 
 
 
 
3.3.1 Costs 

Cost is a major component of a certification’s success and is therefore vital to include in 

the BWS questionnaire. Supply-side agents are unlikely to adopt a new certification, technology, 

or  production  method  if  they  perceive  the  costs  of  adoption  outweigh  the  benefits  (Pearson  & 

Henryks 2008, El-Osta & Morehart 2000, Ugochukwu & Phillips 2018). Likewise, consumers are 

unlikely to pay the price premium for a certified product if the premium elevated the cost of the 

product outside of their budget (Pearson & Henryks 2008).  The qualitative data indicated that 

supply-side agents were concerned about bringing new or additional certification programs into 

their operations, as these are likely to raise their cost of production, which would be mostly passed 

onto  their  consumers.  Similarly,  consumers  interviewed  were  also  hesitant  about  potential 

increased prices on their halal meat products, which are already more expensive than non-halal 

products.  

3.3.2 Enforcement 

It is also essential to consider preferences that market participants may have for a NHMC 

program’s administering body, as opinions on government versus non-government oversight are 

varied. McKendree et al. (2013), Johnston et al. (2001), Ortega et al. (2011, 2012), and Sønderskov 

and Daugbjerg (2011) have all found that certifications administered by a government agency are 

in general more trusted than those run by a third-party organization. However, in the case of halal 

meat, which is a religious standard, a government-run certification program would likely violate 

the First Amendment, and so an alternate organizer is needed. Further, halal meat processors and 

retailers interviewed communicated trepidation with additional government involvement in their 

operations.  In  contrast,  most  consumers  interviewed  were  in  favor  of  increased  government 

involvement in regulating halal certified meat products. 

65 

 
 
3.3.3 Benefits 

Given the rampant food fraud found in the halal market (FSNS 2020) and market agents’ 

corresponding  concerns  about  transparency,  the  benefits  to  consumers  and  supply  chain  agents 

alike  of  a  NHMC  program  are  necessary  to  consider.  On  the  supply  side,  access  to  additional 

markets is possible, and consumer trust and loyalty can be improved using certifications. Wary 

consumers are likely to be willing to pay more for certified products to have enhanced peace of 

mind. Processors and retailers interviewed described interest in using a national certification to 

access new markets and share product attributes easily; consumers likewise would like the ability 

to verify product information more easily.  

3.3.4 Information 

Market agents have preferences for what types of information is collected throughout the 

supply chain. There is a fine line for some individuals between collecting relevant and necessary 

information for ensuring a certification is met and being overly intrusive in an operation. Some 

processors  interviewed  were  concerned  that  collecting  additional  data  on  their  day-to-day 

operations would slow down their processing. Others expressed displeasure with the idea of yet 

another agency “sticking their noses” into business.  

Further,  it  is  important  to  consider  what,  if  any,  of  the  information  collected  should  be 

available  for  downstream  agents  to  access.  In  interviews  with  processors  and  retailers,  some 

discussed experiences with adding a halal meat program to their organization, only to be met with 

backlash from non-Muslim consumers. As such, some processors and retailers, especially larger 

non-Muslim  owned  operations  may  be  hesitant  to  have  their  certification  information  publicly 

available.  However,  Muslim-owned  processing  plants  and  retailers,  as  well  as  the  Muslim 

consumers interviewed, want this information to be readily available to the public. This increased 

66 

 
 
 
availability of information can aid in halal meat consumers’ search efforts and help supply chain 

agents compete.  

3.3.5 Who or What is Certified 

Certification type refers to what the certification program applies to – in this case, whether 

individual halal meat products should be certified (i.e., a brand of chicken breasts), whether an 

entire operation should be certified (i.e., a slaughterhouse or a retailer), the entire supply chain 

(i.e.,  processors,  wholesalers/distributors,  and  retailers),  or  some  combination  of  these  options. 

Current  halal  meat  certification  agencies  in  the  U.S.  certify  at  all  these  levels;  some  certify 

individual products or supply chain agents (e.g., Islamic Services of America 2022), and some 

certify the entire supply chain (e.g., Halal Monitoring Services 2022). In interviews, members of 

all  three  groups  –  consumers,  retailers,  and  processors  –  had  preferences  between  the  types  of 

certifications  they  preferred,  with  some  insisting  on  full  supply  chain  certification  and  others 

accepting piecemeal product certificates. 

3.3.6 Inspections 

Audit characteristics refers to the type of audit (i.e., scheduled or surprise) and how often 

a certification must be renewed or confirmed (i.e., yearly, monthly, daily). Some of the certifiers 

in the U.S. have a scheduled annual audit for renewal for their certifications (e.g., Islamic Services 

of  America),  while  others  have  both  a  scheduled  annual  audit  and  surprise  audits  (e.g.,  Halal 

Monitoring  Services).  Further,  in  the  case  of  meat,  it  is  important  to  consider  the  value  to  the 

consumer in confirming that the certification has been upheld for each day a processor conducts 

halal slaughter. This distinction in audit type and frequency may be important to consumers who 

do not trust current certifications currently in the U.S.  

67 

 
 
 
3.3.7 Halal Standards 

Last, but likely most important to Muslim halal meat consumers, is preferences for halal 

type, which are likely to have a strong impact on preferences for a NHMC program’s attributes. 

There is a very long list of attributes that consumers could prefer to be certified as part of a halal 

standard. These include whether the animal was hand- or machine-harvested, pre-stunned before 

harvest, facing Mecca at the time of slaughter, harvested by a Muslim or a Person of the Book, and 

many other considerations. These attributes vary in their ease of implementation in industrialized 

food  systems;  hand  slaughter  results  in  a  slower  line  speed  than  machine  slaughter  but  facing 

Mecca  may  be  relatively  simple  to  accommodate.  Regardless  of  the  ease  of  adopting  these 

methods,  these  attributes  are  vitally  important  to  Muslims  and  non-negotiable  for  many.  For 

instance, in interviews, consumers said they have never and will never eat machine-slaughtered 

halal  meat  products;  that  is,  they  require  a  halal  meat  certification  that  uses  hand-slaughter 

methods. On the other hand, processors interviewed are concerned about the competitiveness of 

their operations when their line speed is significantly slower under hand-slaughter methods.  

3.4 Additional Survey Questions & Analysis Methods 

In addition to the BWS questionnaire, the survey included follow-up questions for each 

attribute category to clarify what preferences the three groups have. For example, for the attribute 

“Enforcement and Regulation,” the follow-up question shown to participants is given in Figure 6. 

The full list of follow-up questions for the seven attributes are given in Figure 7. These follow-up 

questions  elicit  more  specific  preferences  that  the  surveyed  groups  have  and  allow  for  a  more 

detailed comparison of choices across the market groups. 

68 

 
 
 
 
Figure 6: Example Best-Worst Scaling Follow-Up Question 

69 

 
 
 
Figure 7: List of Follow-Up Best-Worst Scaling Questions 

Costs

• C: How much would they be willing to pay in addition for this 

certification on a product (%)

• P&R: List of costs they may consider important in their decision-

making process

Enforcement

• A list of possible organizations that could run the program (e.g., NGO, 

religious group, current certifiers, etc.)

Benefits

• P&R: List of benefits they could consider important

Information

• C: Types of info that should be included on packages/in stores
• P&R: Types of info that should be passed on to customers

Who/What Certified

• What should be certified/carry a certification stamp (products, retailers, 

suppliers, etc.)

Inspection

• List of inspection types/techniques (scheduled, random, etc.) for 

certification renewal

Halal Standards

• List of standards that could be included in a certification

There were also demographics questions included in the surveys. These consisted of the 

standard socioeconomic questions for individuals typically seen in survey data collection, as well 

as  questions  for  consumers  about  their  ethnic  and  cultural  background,  immigration  status  or 

70 

 
 
 
generation  of  citizenship, 

religious  history 

(i.e.,  born  and 

raised  Muslim  versus 

converting/reverting to the faith), subsect of Islam that they practice, and other cultural or religious 

characteristics that could influence perceptions and preferences related to the U.S. domestic halal 

meat  market.  For  processors  and  retailers,  the  survey  also  included  operation  demographics 

questions.  The  survey  instruments  were  reviewed  with  members  of  the  Islamic  community  for 

clarity and to ensure there were no misrepresentations. 

3.5 Participant Recruitment 

The recruitment processes for each of the three groups surveyed – processors, retailers, and 

consumers – are given below.  

3.5.1 Processor Recruitment 

I recruited processor participants from three sources: 1) USDA Food Safety and Inspection 

Service  (FSIS)  list  of  registered  meat  processors,  2)  registered  processors  on  Halal  Monitoring 

Services’ (HMS) website, and 3) the American Association of Meat Processors (AAMP). Poultry, 

lamb, beef, and goat processors were included in the sample. While the processors listed on HMS’ 

website are known to be halal, it is also likely that some processors on the USDA FSIS database 

and  the  AAMP  membership  list  also  process  halal  meat  or  poultry  products,  though  the  exact 

number is unknown. 

The  USDA  FSIS  database  lists  5,859  USDA-inspected  poultry,  lamb,  beef,  and  goat 

processing establishments. Of these establishments, the USDA classifies 2,736 as “very small”, 

2,656 as “small”, and 440 as “large” establishments.11 I conducted stratified random sampling of 

the three groups of establishments in the USDA FSIS data file, using Excel to generate random 

11 The USDA classifies processors as “very small” if they have less than 10 employees or less than $2.5 million in 
annual sales, “small” if they have 10-499 employees, “large” if they have 500 or more employees. 

71 

 
 
 
 
 
number lists to select establishments from the populations.12 Establishments were called by a team 

of undergraduate research assistants beginning in early November 2022 to determine who at the 

establishment  should  respond  to  the  survey  and  obtain  email  addresses.  Email  addresses  for 

establishments without a phone number listed or those that did not answer or return calls were 

retrieved from business websites when available.  

Individual Qualtrics survey links were sent via email to the USDA FSIS sample. The first 

round  of  emails  was  sent  using  MS  Word  mail  merge  on  December  9,  2022,  with  follow-up 

reminder emails on December 13 and 16, 2022. The next reminder email was sent using Constant 

Contact on January 11, 2023. From the first round of emails, approximately 50 bounced back as 

undeliverable, and 12 businesses responded saying they did not qualify or would not be taking the 

survey. Thus, a total of 987 processors from the USDA FSIS list received the survey.  

There  were  an  additional  58  registered  processors  on  the  Halal  Monitoring  Services’ 

(HMS) website which were all included in the sample. Emails were obtained from the certifier, 

and individualized Qualtrics survey links were sent using MS Word mail merge in January 2023, 

with follow-up reminder emails in January and February 2023. Additionally, these processors were 

given the option to take the survey in either Arabic or Urdu if they preferred. 

The  AAMP  membership  list  was  contacted  via  an  association  representative,  who 

distributed an anonymous Qualtrics survey link to the membership email listserv in March 2023, 

with reminder emails in March and April 2023. It is likely that many AAMP members received 

the survey who also were included in my USDA FSIS recruitment efforts. However, the response 

12 As very small, small, and large processors make up 46.7%, 45.3%, and 7.5% of the total population, respectively, 
these percentages were used to determine how many establishments to sample from each group. There were 1,049 
processors contacted from the USDA FSIS database including 20 large processors, 451 small processors, and 578 
very small processors. 

72 

 
 
 
 
 
rate  for  both  samples  was  very  low,  so  I  do  not  anticipate  there  were  any  duplicate  survey 

responses. 

Despite  contacting  businesses  in  three  different  samples,  I  received  only  195  total 

responses, with only 95 complete responses remaining after data cleaning. I received responses 

mainly from very small, small, and medium processing plants, both because these make up over 

90% of meat processors in the nation and, anecdotally, because larger processors typically do not 

respond to surveys.  

3.5.2 Retailer Recruitment 

General and halal meat retailers were recruited between February 2023 and October 2023. 

Non-halal  retailers  were  recruited  via  multiple  state-level  grocers  associations  and  from  a 

membership  list  from  the  National  Grocers  Association  (NGA).  First,  I  attempted  to  recruit 

retailers via the state-level grocers associations and the NGA email listservs with the assistance of 

association representatives. However, only 18 responses were received via these efforts, so another 

recruitment approach was needed. In May of 2023, a team of undergraduate research assistants 

called retailers from the 2019 winter NGA membership list (National Grocers Association 2019) 

– the most recent available online – between May 2023 and September 2023 to collect point of 

contact email addresses. After removing closed businesses, a total of 946 retail stores were called, 

and 236 email addresses were obtained. As in the processor case, it is possible that some of these 

retailers actually did have a halal program at the time of the survey, though the exact number of 

these stores is unknown. 

Known  halal  retailers  were  recruited  from  halal  certifiers’  online  lists  of  registered 

businesses and through a nationwide web scraping of Yellow Pages using the following key terms 

and phrases: “halal meat grocery store,” “halal meat,” “Indo-Pak grocery,” “African grocery,” and 

73 

 
 
 
 
 
“international  grocery  store.”  The  results  of  the  web  scraping  were  compiled,  and  a  team  of 

undergraduate research assistants called the 919 stores between July 2023 and October 2023 to 

collect email contact information; 96 email addresses were obtained. 

Emails with survey links were sent three times to each category of retailers between August 

25  and  October  10,  2023.  Incentive  payments  of  $25  were  offered  for  complete  and  quality 

responses, though not all respondents claimed their incentive. In total, 50 responses were collected 

from the retailer samples, and after data cleaning, 39 viable survey responses remained. 

3.5.3 Consumer Recruitment 

For Muslim halal meat and poultry consumer recruitment, I partnered with Qualtrics ™ to 

collect responses to ensure a nationally representative population to sample from. Multiple rounds 

of recruitment were necessary due to the relatively small number of these consumers nationwide. 

To qualify for the survey, consumers needed to be practicing Muslims, over the age of 18, the 

primary grocery shopper for their household, and have bought a halal meat or poultry product in 

the last 12 months. In total, 507 complete and clean responses to the online survey were collected 

between May 2023 and December 2023.  

4. Methodology & Statistical Framework for Analysis 

Multiple methods for data analysis are used in this chapter. First, descriptive statistics are 

used  to  detail  the  characteristics  of  survey  participants  and  general  patterns  in  their  responses. 

Then, BWS scores are used to determine which attribute to use as the base case for further analysis 

with random parameter logits (RPLs) models. In particular, common practice is to set the base case 

for a logit model as the largest or smallest BWS score. From the RPLs, BWS shares are calculated 

and compared across groups using Poe tests to determine what statistically significant differences 

in preferences exist between market groups. 

74 

 
 
 
  
 
4.1 Count Data and Descriptive Best-Worst Scaling Statistics 

A descriptive analysis of the BWS data was conducted using BWS scores. BWS scores are 

computed as the number of times an attribute was selected as most preferred, minus the number of 

times that attribute was selected as least preferred, divided by the number of times the attribute 

appears in the design. In this study, each attribute appeared three times in the design. These BWS 

scores provide another descriptive method for analyzing the data, as well as inform the selection 

of the base-case in logit models. In the RPL models discussed in the next section, I selected the 

base case as the smallest BWS score (Inspections) for two out of the three survey groups. 

4.2 Best-Worst Scaling Theoretical Foundation & Analytical Methodology 

BWS  is  rooted  in  random  utility  theory  (RUT),  which  assumes  that  agents  seek  to 

maximize their expected utility subject to the choices they are presented (McFadden 1974).  In 

RUT,  it  is  assumed  that  the  relative  preference  for  object  A  over  object  B  is  a  function  of  the 

relative frequency with which A is chosen as better than B for an individual (Louviere et al. 2013).  

Individuals make choices randomly, with some error involved, to maximize their utility. 

The best-worst scaling method presents each individual multiple answer options (in our analysis, 

attributes of a NHMC program) and asks them to select one as “best” (or most important) and one 

as  “worst”  (or  least  important).  In  practice,  the  BWS  method  consists  of  a  series  of  several 

questions, each comprised of different combinations of attributes per question.  According to RUT, 

the utility for respondent n in selecting alternative i in choice set t is:   

𝑈!"# = 	 𝑉!"# + 𝜀!"# 

where 𝑉!"# is the deterministic portion of utility dependent upon the attributes of the alternative 

and 𝜀!"# is the stochastic component of utility, which is independently and identically distributed 

over all alternatives and choice scenarios.   

Equation 14 

75 

 
 
Generally, when respondents are presented with a choice set, they make choices based on 

maximizing the utility they can receive from each alternative in the choice set. For example, in 

making a choice between alternative j and alternative k, respondent n will pick alternative j over 

alternative k when:   

𝑈!$# > 𝑈!%#	𝑓𝑜𝑟	𝑎𝑙𝑙	𝑗 ≠ 𝑘 

Equation 15 

Given that each choice set has J attributes, the pair of attributes chosen represents a choice 

from all J(J-1) possible pairs, which maximizes the difference in importance. Following Lusk and 

Briggeman  (2009)  and  McKendree  et  al.  (2018),  let  the  true  or  latent  unobservable  level  of 

importance for individual n be represented by 𝐼3) = 𝜆) + 𝜖3) where 𝜆) represents j’s location on 

the scale of importance and 𝜖3) is the random error term.  

The  probability  that  pair  (j,  k)  is  chosen  out  of  a  choice  set  with  J  attributes,  where  j 

represents  the  most  important  attribute  and  k  represents  the  least  important  attribute,  is  the 

probability that the difference between j and k is larger than all the J(J – 1) – 1 other possible 

differences in the choice set. When the error terms 𝜖3) are independent and identically distributed 

type I extreme value, the multinomial logit (MNL) form of this probability is:  

𝑃𝑟𝑜𝑏!(𝑗	𝑐ℎ𝑜𝑠𝑒𝑛	𝑎𝑠	𝑚𝑜𝑠𝑡, 𝑘	𝑎𝑠	𝑙𝑒𝑎𝑠𝑡) = 	

𝑒 &!’&"
(
)*+

𝑒 &!’&" − 𝐽

(
∑ ∑
,*+

From this probability statement, by maximizing the log-likelihood function, parameters 𝜆$ can be 

estimated. When doing this, the dependent variable is 1 for the chosen most-least attribute pair and 

Equation 16 

0 for the remaining J(J – 1) – 1 pairs.   

An  MNL  assumes  respondents  have  homogenous  views  of  the  attributes  analyzed. 

However,  past  studies  have  found  heterogenous  preferences  among  agricultural  supply  chain 

76 

 
 
 
agents (e.g., McKendree et al. 2018, Schulz and Tonsor 2010, Ortega et. al 2019) and consumers 

in relation to agricultural products (e.g., Ortega et. al 2011, Bazzani et. al 2017, Ubilava & Foster 

2009, McKendree et. al 2013). Not only do individuals behave differently, it is likely they also 

have  different  motivations  behind  these  decisions.  Accordingly,  to  account  for  response 

heterogeneity, both an uncorrelated and correlated random parameters logit (RPL) were estimated 

(Boxall and Adamowicz 2002) and compared to the MNL for each of the three groups surveyed.  

The MNL and RPL models were conducted using NLogit 6 on the BWS data for all three 

groups in the study. In each group, the correlated and uncorrelated RPLs resulted in statistically 

significant standard deviations for all attributes and statistically significant coefficients for at least 

some  of  the  attributes;  in  the  case  of  the  correlated  RPLs,  the  covariances  were  also  strongly 

statistically significant, confirming heterogeneity of opinions within each group. Thus, I chose to 

use the results of the correlated RPLs for further analysis and comparisons across the groups. 

RPL coefficient estimates cannot be directly interpreted. However, a “share of importance” 

estimate based on a ratio scale can be calculated for each of the seven attributes of a hypothetical 

U.S. national halal program included in the BWS portion of the surveys:  

Share of importance for attribute j = 
∑

4!"#

%
$&’

#
4!$

Equation 17 

These shares provide a more intuitive approach to analyzing the data. The shares and p-

values  were  calculated  in  MATLAB  with  the  coefficient  matrices  from  NLogit  6.  The  sum  of 

shares  among  all  seven  attributes  analyzed  must  equal  one  (or  100%).  If  attribute  j  has  an 

importance share of 0.3 (30%) and attribute k has an importance share of 0.1 (10%), then j is three 

times  as  important  as  k.  Krinsky-Robb  confidence  intervals  (Krinsky  &  Robb  1986)  were 

calculated as a conservative way to compare statistical differences across importance shares both 

77 

 
 
 
within and across survey groups. Additionally, to test for differences in preference shares across 

the samples, I use the full combinatorial method from Poe et. al (2005). This allows me to draw 

conclusions for which attributes are most important to different groups, and how these preferences 

are displayed across the market. The Krinsky-Robb and Poe tests were conducted in MATLAB 

using the coefficient and variance matrices obtained from modeling the correlated RPLs in NLogit 

6. 

5. Results and Discussion 

The results of the analysis methods used in this chapter and corresponding discussion are 

presented by method type in the sections below. 

5.1 Descriptive Statistics 

Participant  demographics  by  survey  group  are  presented  in  Table  15  and  Table  16.  For 

retailers and processors, the demographic statistics are representative of the survey respondent; 

relevant operation demographics are included in Table 17.  

78 

 
 
 
 
Table 15: General Participant Demographics by Survey Group 
Category 

Processors 
(%) 

Retailers 
(%) 

Consumers 
(%) 

Gender (n = 94, n = 33, n = 507) 
Male 
Female 
Prefer not to disclose 
Education Level (n = 94, n = 32, n = 507) 
Less than High School  
High School 
Some College 
2-Year Degree (Associates) 
4-Year Degree (Bachelor's) 
Master's Degree 
Professional Degree 
Prefer not to Disclose 
Race (n = 94, n = 36) 
White 
Black 
Native American or Alaskan Native 
Native Hawaiian or Pacific Islander 
Asian 
Other 
Prefer not to disclose 
Political Party (n = 94, n = 33, n = 507) 
Democrat 
Republican 
Independent 
Other 
Prefer not to disclose 
1st Generation Immigrant (n = 94, n = 33, n = 507) 
No 
Yes 
Prefer not to disclose 
2nd Generation Immigrant (n = 94, n = 33, n = 507) 
No 
Yes 
Prefer not to disclose 
Currently Religious (n = 94, n = 33) 
Yes 
No 
Prefer not to disclose 
Previously Religious (n = 20, n = 12) 
Yes 
No 
Prefer not to disclose 

76% 
19% 
5% 

0% 
6% 
13% 
10% 
44% 
22% 
5% 
0% 

76% 
1% 
0% 
0% 
3% 
6% 
14% 

5% 
35% 
26% 
3% 
31% 

82% 
10% 
9% 

66% 
21% 
13% 

21% 
56% 
22% 

45% 
40% 
15% 

79 

49% 
50% 
<1% 

2% 
20% 
18% 
11% 
27% 
17% 
5% 
<1% 

40% 
21% 
28% 
1% 
9% 

70% 
27% 
4% 

52% 
41% 
8% 

79% 
21% 
0% 

0% 
3% 
13% 
28% 
25% 
22% 
9% 
0% 

66% 
0% 
5% 
2% 
10% 
2% 
2% 

6% 
27% 
36% 
6% 
24% 

91% 
6% 
3% 

82% 
15% 
3% 

36% 
48% 
15% 

58% 
42% 
0% 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Increased specificity for consumer demographics is important to include, as preferences 

and opinions of Muslim consumers will be influenced more strongly and in different ways by their 

race, ethnicity, sect of Islam, and status as a convert or revert to Islam (Table 16). 

Percentage 

45% 
27% 
23% 
2% 
8% 
16% 
<1% 
3% 
1% 

Table 16: U.S. Muslim Halal Consumer Demographics: Race, Ethnicity, and Religion (n = 
507) 
Category 
Race and Ethnicity 
White 
Black or African American 
Middle Eastern or North African 
American Indian or Alaskan Native 
Asian 
South or Southeast Asian 
Native Hawaiian or Other Pacific Islander 
Other 
Prefer not to Disclose 
Sect of Islam 
Sunni 
Shia or Shiite   
Ibadi 
Non-denominational 
Other 
Prefer not to disclose 
Convert or Revert to Islam 
Years Since Converting or Reverting to Islam (n = 165) 
0-5 years 
6-10 years 
11-15 years 
16-20 years 
21-25 years 
26-30 years 
31-35 years 
Over 35 years 
Prefer not to Disclose 
Previous Religion (n = 165) 
Christianity 
Hinduism 
Buddhism 
Judaism 
Sikhism 
Other 
None 
Prefer not to Disclose 

23% 
17% 
17% 
10% 
7% 
7% 
5% 
8% 
6% 

50% 
9% 
2% 
5% 
2% 
5% 
19% 
7% 

63% 
11% 
6% 
13% 
2% 
5% 
33% 

80 

 
 
 
 
 
 
Unsurprisingly, the demographic patterns of the supply-side agents differ from consumers. 

While  the  majority  of  supply-side  agents  were  male,  white,  and  had  at  least  a  4-year  degree, 

consumers  were  evenly  split  amongst  genders,  were  a  much  wider  variety  of  races,  and  were 

generally  less  educated  overall.  Supply-side  agents  were  more  commonly  Republicans  or 

Independents, while consumers were more likely to be Democrats. Consumers were more likely 

to  be  first-  or  second-generation  immigrants  and  100%  were  currently  religious  (non-religious 

individuals  were  screened  out  of  the  survey),  compared  to  21%  and  36%  of  processors  and 

retailers, respectively. These general differences in survey groups may help to explain differences 

in preferences and opinions across groups for a potential U.S. NHMC program. 

Muslim consumers’ demographics provide even more detailed information that can explain 

variability  in  preferences.  While  the  largest  group  of  consumers  is  white,  there  are  also  large 

percentages  of  other  races  and  ethnicities  represented  in  the  survey.  Likewise,  the  majority  of 

consumers represented are Sunni Muslims (the most common across the world and the most likely 

to  follow  stricter  halal  dietary  standards),  though  other  denominations  are  well  represented. 

Further, one-third of the sample were Islamic converts (also commonly referred to as reverts), with 

the majority converting within the last 15 years. Recent converts may display different preferences 

than life-long Muslims. Finally, one-half of converts were previously Christians, who may display 

different  preferences  or  opinions  than  converts  from  other  religions  or  those  who  had  not 

previously followed a religion. 

81 

 
 
 
 
Retailers (n = 41) 

Processors (n = 95) 

48% 
23% 
23% 
6% 

7% 
44% 
45% 
3% 
N/A 
N/A 
N/A 

Table 17: Processing and Retailing Establishment Summary Statistics 
Category 
Establishment Type 
Slaughter without processing 
Processing without slaughter 
Slaughter and processing 
Other 
Grocery Store 
Butcher Shop/Deli 
Other 
Location 
Rural 
Suburban 
Urban 
Prefer not to disclose 
Type of Animals  
Processed/Sold 
Beef 
Veal 
Lamb 
Pork 
Turkey 
Chicken 
Goat 
Exotics 
Halal Status 
Current Halal 
Past Halal 
Never Halal 
Unsure 
No. of Employees 

83% 
34% 
55% 
46% 
35% 
38% 
55% 
35% 

34% 
4% 
62% 
N/A 

Mean 
62 
Mean 
1992 

Min 
2 
Min 
1902 

Max 
850 
Max 
2022 

Mean 
45 
Mean 
1979 

Year Established 

N/A 
N/A 
N/A 
N/A 
80% 
12% 
7% 

61% 
33% 
6% 
N/A 

100% 
41% 
56% 
85% 
95% 
100% 
17% 
34% 

22% 
10% 
51% 
17% 
Min 
1 
Min 
1867 

Min 
600 
Max 
2021 

The processors and retailer operation demographics in Table 17 show that supply-side 

responses came from a variety of different business types, ages, and sizes. For processors, many 

establishments slaughtered multiple species of animals; likewise, retailers sold products from 

82 

 
 
 
 
 
 
 
 
 
 
 
 
 
multiple species. Finally, halal businesses were represented in each sample. The variety of 

businesses and heterogeneity in their characteristics helps to ensure the data collected is fairly 

representative of each group’s national population. 

5.2 Attribute Preferences 

The  count  data  and  corresponding  descriptive  discussion  of  the  best-worst  question 

responses is given in Appendix A.3. Figure 8 presents the descriptive results based on the BWS 

scores. The scores reveal the most and least important attributes to consider for each of the three 

groups. The BWS scores provide information on the ideal base case for the RPL model; in this 

instance, I selected the least important attribute for two of the three groups (Inspections) as the 

base case.  

83 

 
 
 
 
 
Figure 8: Best-Worst Scores for Processors, Retailers, and Consumers 

Processor Best-Worst Scores

Costs

Information

Inspections

Enforcement

Who/What 
Certified

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

Halal Standards

Benefits

Retailer Best-Worst Scores

Costs

Enforcement

Information

Inspections

Halal Standards

Who/What 
Certified

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

Benefits

Consumer Best-Worst Scores

Costs

Information

Who/What Certified

Enforcement

Inspections

Halal Standards

Benefits

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

The  coefficients  for  the  correlated  RPLs  are  given  in  Table  18,  while  the  additional 

variance and correlation matrices for the correlated RPLs and the results for the uncorrelated RPLs 

and MNLs are given in Appendix A.3. In all models, the Inspection attribute of a hypothetical U.S.  

NHMC program – has been dropped to avoid multicollinearity. Therefore, the coefficients reflect 

the importance of each of the six attributes relative to Inspection, which was normalized to zero 

for identification purposes as it was the least important attribute to processors and retailers, and 

the second least important attribute to consumers (based on the BWS scores). 

84 

 
 
 
 
 
 
 
The correlated RPL results are generally consistent with the BWS scores. For processors, 

all attributes of the best-worst scales are statistically significant at least at the 5% level, excluding 

Benefits, which was not significant. For retailers, all attributes are statistically significant at least 

at the 10% level, excluding Enforcement and Halal Standards; for consumers, all attributes are 

statistically significant at the 10% level, excluding Benefits. The results indicate that Who/What 

is Certified is the most preferred attribute for retailers and processors, closely followed by Halal 

Standards and Costs for processors, as well as Costs for retailers. For consumers, Halal Standards 

and Who/What is Certified are the top two preferred attributes. On the other hand, Information 

was the least preferred attribute (least important to consider) for processors and retailers, while 

consumers indicated that the Costs were least important.  

The importance shares from the correlated RPLs are given in Table 19. Importance shares 

results for the uncorrelated RPLs and MNLs are given in Appendix A.3. Results show that 23% 

of processors view the attributes Who/What is Certified and Halal Standards as the most important 

attribute for program designers to consider when developing a hypothetical U.S. NHMC program. 

Who/What  is  Certified  was  the  most  important  attribute  for  program  designers  to  consider  for 

retailers (27%) and the second most important for program designers to consider for consumers 

(18%). Halal Standards was the fourth most important attribute for program designers to consider 

for  retailers  (13%)  and  the  most  important  attribute  for  program  designers  to  consider  for 

consumers (22%). I see the most consensus in rankings of attribute importance between the two 

supply  chain  members  (processors  and  retailers),  while  the  preferences  of  consumers  are  more 

divergent. These differences between the supply and demand sides of the market are unsurprising 

and illustrate the general challenges of food policy making and, more specifically, the diversity of 

85 

 
 
preferences for what is most important to affected market participants when designing a NHMC 

program. 

Finally, I conducted pair-wise Poe tests on the preference shares associated with each of 

the seven attributes considered in the BWS for a hypothetical NHMC program. Table 20 presents 

the  p-values  from  the  Poe  tests,  with  values  under  0.05  indicating  the  two  groups’  shares  are 

statistically different at the 5% level. Poe test results for the uncorrelated RPLs and MNLs are 

given in Appendix A.3. Overall, the results reveal patterns of preferences between the three groups 

in the study. Unsurprisingly, the Poe test shows that consumers’ preferences across the attributes 

of a hypothetical national U.S. halal certification program are significantly different from those of 

the  processors  and  retailers.  Consumers  place  stronger,  more  positive  emphasis  on  Inspection 

relative to retailers and Halal Standards. On the other hand, consumers place stronger and more 

negative emphasis on the Costs and Who/What is Certified than both processors and retailers, and 

Benefits  and  Inspection  when  compared  to  processors.  As  for  processors  and  retailers,  their 

responses are overall very similar; there are only two statistically significant differences on the 

attributes  Enforcement  and  Benefits.  The  BWS  shares  indicated  that  processors  valued 

Enforcement more than retailers, while retailers valued Benefits more than processors.  

86 

 
 
 
   
Table 18: Correlated Random Parameters Logit Results of Best-Worst Scaling for Hypothetical U.S. National Halal Meat and 
Poultry Certification Program: Processors (n = 82), Retailers (n = 33), and Consumers (n = 507) 

Best-Worst Scaling Attributes 

Cost of Certification 

Enforcement and Regulation 

Information Collected/Available 

Who/What Must be Certified 

Which Halal Standards are Included 

Benefits of Certification 

Processors 
Coefficient 
(st. error) 
0.92*** 
(0.16) 

0.29** 
(0.14) 

-0.81*** 
(0.16) 

0.99*** 
(0.15) 

0.97*** 
(0.15) 

-0.047 
(0.15) 

Retailers 
Coefficient 
(st. error) 
0.91*** 
(0.25) 

-0.18 
(0.23) 

-0.79*** 
(0.24) 

1.05*** 
(0.23) 

0.35 
(0.24) 

0.40* 
(0.24) 

Consumers 
Coefficient 
(st. error) 
-0.35*** 
(0.05) 

-0.09* 
(0.05) 

-0.20*** 
(0.05) 

0.29*** 
(0.05) 

0.52*** 
(0.05) 

0.07 
(0.05) 

McFadden Pseudo R2 
N 
Log likelihood 
AIC 
AIC/N 
Note: *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1. Base attribute normalized to zero is Inspection Type/Frequency. 

0.23 
574 
-860.64 
1775.30 
3.09 

0.24 
231 
-342.56 
739.10 
3.20 

0.12 
3549 
-6053.56 
12161.1 
3.43 

87 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table 19: Best-Worst Scaling Shares of Preferences for Hypothetical U.S. National Halal Meat and Poultry Certification 
Program: Consumers, Retailers, & Processors: Correlated Random Parameters Logit 

National Halal Meat & Poultry 
Certification Program Attribute 
Cost of Certification 
Enforcement and Regulation 
Information Collected/Available 
Who/What Must be Certified 
Inspection Frequency and Type 
Which Halal Standards are Included 
Benefits of Certification 

Processors (n = 82) 
Share 
0.22*** 
0.12*** 
0.04*** 
0.23*** 
0.09*** 
0.23*** 
0.08*** 

95% CI 
[0.18, 0.26] 
[0.10, 0.14] 
[0.03, 0.05] 
[0.20, 0.27] 
[0.07, 0.10] 
[0.19, 0.27] 
[0.07, 0.10] 

Retailers (n = 33) 

Share 
0.24*** 
0.08*** 
0.04*** 
0.27*** 
0.09*** 
0.13*** 
0.14*** 

95% CI 
[0.17, 0.30] 
[0.06, 0.11] 
[0.03, 0.06] 
[0.21, 0.34] 
[0.07, 0.13] 
[0.09, 0.19] 
[0.10, 0.19] 

Consumers (n = 507) 
95% CI 
Share 
[0.09, 0.10] 
0.09*** 
[0.11, 0.13] 
0.12*** 
[0.10, 0.12] 
0.11*** 
[0.17, 0.19] 
0.18*** 
[0.12, 0.14] 
0.13*** 
[0.21, 0.24] 
0.22*** 
[0.13, 0.15] 
0.14*** 

Note: *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1. 

Table 20: P-values from Poe tests for Hypothetical U.S. National Halal Meat and Poultry Certification Program: Pair-wise 
Comparisons Between Processors (n = 82), Retailers (n = 33), and Consumers (507) 

National Halal Meat & Poultry 
Certification Program Attribute 
Cost of Certification 
Enforcement and Regulation 
Information Collected/Available 
Who/What Must be Certified 
Inspection Frequency and Type 
Which Halal Standards are Included 
Benefits of Certification 

Processors vs. Retailers 

Processors vs. Consumers  Retailers vs. Consumers 

0.68 
0.03 
0.34 
0.16 
0.33 
1.00 
0.00 

0.00 
0.68 
0.01 
0.00 
0.00 
0.58 
0.00 

0.00 
0.99 
0.01 
0.00 
0.02 
0.00 
0.44 

Note: Values that are statistically significant at the 5% level or better are bolded.

88 

 
 
 
5.3 Follow-Up Questions: Certification Specifics 

Follow-up questions for each of the BWS attributes allow me to dig deeper into the nature 

of participants’ preferences for a hypothetical U.S. NHMC program. Descriptive statistics for these 

follow-up  questions  are  given  by  survey  group  in  Table  21-Table  30.  Note  that  the  number  of 

respondents  for  processors  and  retailers  is  slightly  higher  for  these  questions  than  the  BWS 

questions, as not all processors and retailers completed all BWS questions, thus excluding these 

participants from the MNL and RPL panel analyses. 

The  BWS  scores,  RPL  coefficients,  and  BWS  shares  for  the  three  groups  provided 

consistent preference patterns for different attributes of a hypothetical U.S. NHMC program. To 

summarize, processors’ results showed Who/What is Certified, Halal Standards, and Costs as the 

most important attributes for program designers to consider when developing a hypothetical U.S. 

NHMC  program.  Retailers’  results  showed  Who/What  is  Certified  and  Costs  as  the  two  most 

important  attributes  for  program  designers  to  consider  when  developing  a  hypothetical  U.S.  

NHMC program. Muslim halal consumers’ results identified Halal Standards and Who/What is 

Certified as the two most important attributes for program designers to consider when developing 

a hypothetical U.S. NHMC program. I now discuss these results in more detail by group – and by 

halal  versus  non-halal  supply  chain  members  where  applicable  –  using  additional  information 

gathered from the follow-up questions and qualitative interview quotes for further context.  

Table 21 provides more detail on the Halal Standards preferred by group. Between non-

halal processors and halal processors only, the results show an increased preference for each of the 

given possible standards by halal processors. This increase in strength of preferences makes sense, 

as halal processors are both more knowledgeable of the halal process and have more at stake when 

defining standards than those currently outside of the market. These responses also support the 

89 

 
 
 
interview findings that halal processors generally favor standards that are more closely associated 

with hand slaughter (e.g., zabiha slaughter method, slaughterers of Muslim faith, individual spoken 

blessings, and no stunning).  

For  halal  consumers,  the  results  show  more  variability  in  preferences,  with  consumers 

indicating stronger preferences for additional quality attributes not typically included in current 

halal certifications used in the U.S. (e.g., animals face Mecca and non-GMO) (Table 21). These 

variations  in  consumers’  preferences  and  the  generally  smaller  percentages  of  consumers 

indicating that any one standard was preferred are unsurprising given the diversity in Islamic sect, 

racial, and ethnic backgrounds of the consumer sample. Furthermore, consumers interviewed in 

the earlier phase of this project who are less strict in their halal dietary requirements indicated little 

to no preference between zabiha and machine slaughter or the faith of the slaughterer, for example. 

Table 22 provides more information on opinions on the attribute Who/What is Certified. 

For all retailers and processors, the majority want a hypothetical U.S. national halal certification 

program to require certification of individual products (>63%) and supply chain members (>50%). 

For those that want supply chain members to have certification, the supply chain agents indicated 

slaughter (>91%) and processing (>83%) establishments were the most preferred to be certified. 

A higher percentage of halal processors and retailers indicated that supply chain agents should be 

certified, which matches the opinions of nearly all halal processors and retailers interviewed. The 

interviewees  valued  a  halal  supply  chain  with  strict  and  well-defined  certifications  for  each 

member  to  ensure  their  good  name  and  product  quality  were  maintained.  Additionally,  halal 

retailers interviewed indicated they rely heavily on their reputation within their local communities 

and word-of-mouth of happy customers; thus, strong certification utilization is important. 

90 

 
 
I  asked  processors  and  retailers  to  indicate  which  of  the  given  costs  of  a  U.S.  NHMC 

program would be most important to their businesses (Table 23). The cost categories indicated as 

important  were  consistent  between  non-halal  processors  and  halal  processors,  with  the  most 

notable  difference  being  the  costs  of  potential  establishment  modifications;  67%  of  non-halal 

processors  and  34%  of  halal  processors  indicated  this  was  an  important  factor.  The  higher 

percentage of non-halal processors who considered the costs of establishment modifications to be 

important is logical, as these processors do not currently have a halal program and therefore their 

establishments  may  not  currently  have  the  infrastructure  to  add  a  halal  program.  There  were 

notable differences in which cost categories were most important for non-halal retailers and halal 

retailers. Forty-one percent of non-halal retailers and 67% of halal retailers indicated certification 

fees were an important factor, while 50% of non-halal retailers and 33% of halal retailers indicated 

the costs of potential establishment modifications were important. 

Table 21: Market Participants’ Preferred Halal Standards for a U.S. National Halal Meat 
and Poultry Certification Program 

Preferred Halal 
Standards 

Zabiha (hand-slaughter) 
Machine slaughter 
Slaughterers of Muslim 
faith 
Slaughterers of Jewish or 
Christian faith 
Individual spoken 
blessings 
Animals not stunned 
Animals face Mecca 
Non-GMO 
Other 

Non-Halal 
Processors 
(n = 66) 
76% 
9% 

Halal 
Processors 
(n = 32) 
88% 
22% 

Non-Halal 
Retailers 
(n = 32) 
22% 
16% 

Halal 
Retailer
s (n = 9) 
78% 
0% 

Halal 
Consumers 
(n = 507) 
53% 
19% 

6% 

9% 

16% 

22% 
3% 
22% 
13% 

67% 

33% 

33% 

44% 
33% 
11% 
0% 

58% 

14% 

36% 

35% 
33% 
32% 
0% 

12% 

8% 

8% 

3% 
3% 
3% 
8% 

75% 

16% 

47% 

38% 
22% 
6% 
6% 

91 

 
 
 
 
 
Table 22: Market Participants’ Opinions on Who or What Should be Certified Under a 
U.S. National Halal Meat and Poultry Certification Program 
Non-
Halal 
Retailers 
(n = 32) 
63% 
50% 

Non-Halal  
Processors 
(n = 66) 

Halal 
Processors 
(n = 32) 

Halal 
Retailers 
(n = 9) 

Halal 
Consumers 
(n = 507) 

Who or What Certified 

75% 
72% 

79% 
77% 

78% 
67% 

80% 
58% 

Individual Products 
Supply Chain Members 
Which Supply Chain 
Members? (n = 61, 38, 
22, 16, 392) 

Slaughter 
Establishments 
Processing 
Establishments 
Transportation/ 
Distributors 
Retailers/Wholesalers 
Restaurants and Food 
Service 
Other 

92% 

84% 

24% 

32% 

34% 

5% 

96% 

91% 

57% 

65% 

61% 

0% 

94% 

83% 

88% 

100% 

31% 

44% 

38% 

0% 

100% 

100% 

100% 

0% 

76% 

66% 

52% 

66% 

57% 

1% 

Table 23: Costs of U.S. National Halal Meat and Poultry Certification Considered Most 
Important  

Program Cost Category 

Certification Fees 
Establishment Modifications 
Increased Labor Hours Needed 
Cost of Traceability Equipment 
Other 

Non-Halal 
Processors 
(n = 66) 
35% 
67% 
42% 
38% 
2% 

Halal 
Processors 
(n = 32) 
47% 
34% 
28% 
31% 
0% 

Non-Halal 
Retailers 
(n = 32) 
41% 
50% 
28% 
44% 
0% 

Halal 
Retailers 
(n = 9) 
67% 
33% 
33% 
44% 
0% 

5.4 Follow-Up Questions: Program Implementation and Transparency 

The development of a U.S. NHMC program will be complex and subject to a wide array 

of  market  participants’  preferences,  as  seen  in  previous  U.S.  national  certification  program 

development processes (e.g., USDA Organic). Nonetheless, the results of my surveys and analysis 

provide initial context and guidance for the potential future development of such a program. There 

are no glaring differences in opinions and preferences that would preclude the development of a 

92 

 
 
 
 
 
 
 
 
 
national program across the three study groups. However, other factors must be considered when 

designing a national-level certification program; specifically, who would set the standards for and 

enforce such a program, how certifications would be administered and audited, and the amount 

and nature of program transparency.  

5.4.1 Program Standards and Enforcement 

Halal is a religious attribute of a food product and is defined by interpreting religious texts; 

thus, the First Amendment to the U.S. Constitution prohibits federal, state, and local governments 

from setting or specifying halal standards. Therefore, the choice of which halal standards to include 

or  exclude  from  a  national  program  would  fall  to  non-government  organizations  (NGOs)  – 

potentially including consumer advocacy groups, religious organizations, certifier and producer 

organizations, and supply chain members. Table 24 shows market participants’ opinions of which 

of  these  groups  should  have  a  say  in  setting  these  standards.  Reaching  a  consensus  on  halal 

standards amongst so many different voices is likely to be difficult; therefore, identifying which 

types of organizations should lead this effort is critical. In interviews, multiple halal processors, 

halal retailers, and Muslim halal consumers all indicated that religious organizations were their 

most preferred option for setting halal standards, followed to a lesser extent by certifiers and non-

government groups. Indeed, based on the results of market participants’ opinions presented here, 

the ideal candidates to work together to set halal standards should be religious (44-89%), certifier 

(31-48%) and non-government (11-28%) organizations.  

Enforcement  of  a  predetermined  religious  standard  is  not  prohibited  by  the  First 

Amendment;  therefore,  options  for  program  enforcement  include  federal  and  state  government 

organizations such as the USDA or state-level departments of agriculture. With this expansion of 

options, the organizations most preferred to be in charge of program management and enforcement 

93 

 
 
 
are religious (22-67%), U.S. government (11-53%), and certifier organizations (19-35%). These 

findings match interview findings; multiple halal processors, retailers, and consumers all expressed 

their  desire  for  a  U.S.  government-backed  certification  program,  with  input  from  religious  and 

certifier organizations.  

94 

 
 
Table 24: Market Participants’ Opinions on Who Should Set and Enforce Standards for a U.S. National Halal Meat and 
Poultry Certification Program, Aggregated 

Organizations 

Non-Halal 
Processors 
(n = 66) 

Halal 
Processors 
(n = 32) 

Non-Halal 
Retailers 
(n = 32) 

Halal 
Retailers 
(n = 9) 

Halal 
Consumers 
(n = 507) 

Set 
U.S. Government Organizations  N/A 
State Government Organizations  N/A 
23% 
Non-Government Organizations 
49% 
Religious Organizations 
32% 
Certifier Organizations 
20% 
Producer Organizations 
28% 
Slaughterers & Processors 
5% 
Wholesalers & Distributors 
3% 
Retailers & Restaurants 
6% 
Other 
Note: Due to the religious freedom protections of the First Amendment, governments are prohibited from setting or defining religious practices; in this case, halal 
standards. 

Enforce 
44% 
18% 
14% 
30% 
33% 
11% 
11% 
3% 
0% 
3% 

Enforce 
51% 
35% 
21% 
49% 
35% 
23% 
31% 
23% 
18% 
0% 

Enforce 
53% 
22% 
19% 
44% 
19% 
0% 
19% 
0% 
0% 
3% 

Enforce 
11% 
22% 
44% 
67% 
33% 
0% 
22% 
0% 
0% 
0% 

Enforce 
28% 
28% 
16% 
22% 
34% 
6% 
13% 
3% 
3% 
0% 

Set 
N/A 
N/A 
28% 
63% 
31% 
9% 
28% 
3% 
3% 
6% 

Set 
N/A 
N/A 
25% 
66% 
48% 
30% 
29% 
42% 
22% 
0% 

Set 
N/A 
N/A 
28% 
44% 
34% 
19% 
6% 
25% 
6% 
0% 

Set 
N/A 
N/A 
11% 
89% 
33% 
22% 
0% 
33% 
0% 
0% 

95 

 
 
5.4.2 Certification Administration and Auditing 

Proper  enforcement  of  a  U.S.  NHMC  program  will  require  defined  standards  for 

administering certifications and auditing certified agents and products. However, administration 

and auditing of certifications is more than just setting standards – a defined method and timeline 

for  inspections  is  ideal  to  ensure  initial  and  ongoing  compliance.  Table  25  shows  survey 

participants’ preferences for the types of inspections or audits that should be conducted to ensure 

supply chain members’ compliance and retain certification. The results show that the majority of 

the market prefers random or a mixture of scheduled and random inspections or audits. This is 

unsurprising, as multiple interview participants across all three groups expressed their concern that 

exclusively using scheduled audits could allow for certifications to be granted to dishonest supply 

chain  members.  Furthermore,  these  preferences  are  in  line  with  the  nature  in  which  many 

certification  programs  are  managed  in  a  variety  of  contexts,  and  therefore  should  not  pose  a 

challenge to implement. 

Table 25: Market Participants’ Opinions for how Certified Suppliers Under a U.S. 
National Halal Meat and Poultry Certification Program Should be Audited 

Inspection Type 

Non-Halal 
Processors 
(n = 38) 
47% 
5% 

Halal 
Processors 
(n = 23) 
22% 
22% 

Non-Halal 
Retailers 
(n = 16) 
44% 
6% 

Halal 
Retailers 
(n = 6) 
0% 
50% 

Halal 
Consumers 
(n = 392) 
19% 
29% 

Scheduled 
Random 
Mixed Scheduled 
& Random 
Other 
Note:  The  number  of  participants  is  lower  for  this  question,  as  only  survey  participants  who  indicated  that  they 
preferred supply chain members be required to hold certification saw this question in the survey. 

50% 

52% 

42% 

57% 

50% 

0% 

0% 

5% 

0% 

0% 

5.4.3 Program Transparency and Traceability 

Finally, it is vital to consider the transparency and traceability of a certification granted 

under a hypothetical U.S. NHMC program. Consumers are increasingly interested in having access 

to information about where and how their food is produced, as well as having the ability to self-

96 

 
 
 
 
 
 
 
authenticate labels and other quality indicators in real-time. Likewise, supply chain members are 

in favor of employing transparency and traceability efforts to ensure their products are viewed as 

authentic and trustworthy to their customers. I asked survey participants about who should have 

access to four main types of information that relate to a hypothetical U.S. NHMC program; the 

results are shown in Table 26 - Table 29. 

First,  survey  participants  were  asked  which  groups  should  have  access  to  traceability 

information for individual products (Table 26) and who should have access to a list of all certified 

establishments (Table 27). The majority of survey participants indicated that the general public, 

slaughter and processing establishments, wholesalers and distributors, and retailers and restaurants 

should have access to both of these types of information. Halal supply chain members were more 

in favor of these groups having access to this information than non-halal supply chain members. 

Second,  survey  participants  were  asked  who  should  have  access  to  a  list  of  all  enforcement 

agencies  that  would  be  responsible  for  issuing  and  auditing  certifications  (Table  28)  and  who 

should have access to the list of halal standards included in the certification (Table 29). Again, 

most survey participants indicated that the general public, slaughter and processing establishments, 

wholesalers and distributors, and retailers and restaurants should have access to this information; 

additionally, the majority of halal processors again indicated that government organizations should 

be able to access this information. Again, halal supply chain members were more in favor of these 

groups having access to this information than non-halal supply chain members. Finally, I asked 

survey  participants  how  the  general  public  should  have  access  to  any  of  the  traceability  or 

transparency information asked about in these four questions; these results are shown in Table 30. 

. All three groups were strongly in favor of online access and halal processors, non-halal 

and halal retailers, and consumers also favored the ability to use a QR code or cell phone app. 

97 

 
 
These responses show that generally, there is strong interest in ensuring a U.S. NHMC program 

has robust transparency and traceability attributes. Overall, the results of these survey questions 

align with opinions expressed by interviewees; the majority of interviewees were in favor of a U.S. 

NHMC with robust transparency and traceability attributes. 

98 

 
 
Table 26: Market Participants’ Opinions on Which Groups Should Have Access to Traceability Information for Individual 
Products Under a U.S. National Halal Meat and Poultry Certification Program 

Groups That Should Have Access to 
Traceability Information 

General Public 
Slaughter & Processing Establishments 
Wholesalers & Distributors 
Retailers & Restaurants 
Government Organizations 
None 

Non-Halal 
Processors 
(n = 66) 
59% 
59% 
59% 
52% 
41% 
8% 

Halal 
Processors 
(n = 32) 
84% 
75% 
81% 
84% 
66% 
0% 

Non-Halal 
Retailers 
(n = 32) 
53% 
53% 
59% 
53% 
31% 
3% 

Halal 
Retailers 
(n = 9) 
56% 
78% 
78% 
56% 
33% 
0% 

Halal 
Consumers 
(n = 507) 
62% 
59% 
55% 
55% 
32% 
2% 

Table 27: Market Participants’ Opinions on Which Groups Should Have Access to a List of all Certified Establishments Under 
a U.S. National Halal Meat and Poultry Certification Program 

Groups That Should Have Access to 
List of all Certified Establishments 

General Public 
Slaughter & Processing Establishments 
Wholesalers & Distributors 
Retailers & Restaurants 
Government Organizations 
None 

Non-Halal 
Processors 
(n = 66) 
64% 
59% 
64% 
58% 
47% 
9% 

Halal 
Processors 
(n = 32) 
88% 
81% 
88% 
88% 
66% 
0% 

Non-Halal 
Retailers 
(n = 32) 
59% 
53% 
56% 
56% 
38% 
3% 

Halal 
Retailers 
(n = 9) 
67% 
67% 
67% 
56% 
33% 
0% 

Halal 
Consumers 
(n = 507) 
60% 
59% 
55% 
55% 
33% 
2% 

99 

 
 
 
 
 
 
 
 
 
 
Table 28: Market Participants’ Opinions on Which Groups Should Have Access to a List of all Enforcement Agencies Under a 
U.S. National Halal Meat and Poultry Certification Program 

Groups That Should Have Access to 
List of all Enforcement Agencies 

General Public 
Slaughter & Processing Establishments 
Wholesalers & Distributors 
Retailers & Restaurants 
Government Organizations 
None 

Non-Halal 
Processors 
(n = 66) 
55% 
56% 
53% 
45% 
49% 
11% 

Halal 
Processors 
(n = 32) 
81% 
75% 
75% 
75% 
66% 
0% 

Non-Halal 
Retailers 
(n = 32) 
53% 
53% 
59% 
53% 
34% 
3% 

Halal 
Retailers 
(n = 9) 
78% 
56% 
67% 
56% 
33% 
0% 

Halal 
Consumers 
(n = 507) 
60% 
53% 
53% 
50% 
35% 
2% 

Table 29: Market Participants’ Opinions on Which Groups Should Have Access to a List of Halal Standards Included in a U.S. 
National Halal Meat and Poultry Certification Program 

Groups That Should Have Access to 
List of Halal Standards Used 

General Public 
Slaughter & Processing Establishments 
Wholesalers & Distributors 
Retailers & Restaurants 
Government Organizations 
None 

Non-Halal 
Processors 
(n = 66) 
64% 
62% 
59% 
52% 
48% 
8% 

Halal 
Processors 
(n = 32) 
81% 
81% 
78% 
72% 
69% 
0% 

Non-Halal 
Retailers 
(n = 32) 
53% 
53% 
53% 
50% 
44% 
3% 

Halal 
Retailers 
(n = 9) 
78% 
56% 
67% 
56% 
33% 
0% 

Halal 
Consumers 
(n = 507) 
57% 
58% 
57% 
55% 
33% 
2% 

100 

 
 
 
 
 
 
 
 
 
 
Table 30: Market Participants’ Opinions on how the General Public Should Have Access to Information Related to a U.S. 
National Halal Meat and Poultry Certification Program 

Information Access Method 

Halal 
Processors 
(n = 29) 
100% 
Online website 
52% 
Using a QR code or cell phone app 
31% 
Freedom of Information Act (FOIA) 
0% 
Other 
Note: The number of participants is lower for this question, as only survey participants who indicated that the general public should have access to information 
related to a U.S. NHMC program saw this question in the survey. 

Halal 
Consumers 
(n = 411) 
74% 
53% 
55% 
<1% 

Non-Halal  
Processors 
(n = 46) 
100% 
39% 
35% 
0% 

Non-Halal  
Retailers 
(n = 20) 
95% 
65% 
35% 
0% 

Halal 
Retailers 
(n = 7) 
100% 
86% 
0% 
0% 

101 

 
 
 
6. Implications for Implementation 

 As  previously  discussed,  the  modern-day  U.S.  halal  meat  and  poultry  certification 

landscape is complex. There are many competing players with differing standards, approaches to 

certification, and market segments. This complicates efforts to develop a uniform standard for the 

U.S. market. However, halal meat and poultry market participants’ interest in a U.S. government-

backed approach to increased market regulation is coming to fruition, though not specifically for 

halal  products.  The  USDA  under  the  Biden  administration  has  made  efforts  to  expand 

transparency, diversity, and accessibility in local meat and poultry processing, via the Executive 

Order on Promoting Competition in the American Economy, the 2021 American Rescue Plan, and 

the USDA’s Meat and Poultry Supply Chain initiatives, among other methods (The United States 

Government 2022, United States Department of Agriculture 2023).  

Additional  concerns  of  halal  market  participants  can  be  addressed  by  the  federal 

government.  Supply-side  survey  respondents  indicated  that  the  potential  costs–  such  as 

certification fees and establishment modifications – of participating in a NHMC program are of 

concern; however, it is possible that federal grant programs could alleviate these financial burdens. 

The  Biden  administration  has  provided  the  USDA  with  increased  funding  for  grant  programs 

including  the  Meat  and  Poultry  Inspection  Readiness  Grant  Program  (MPIRG),  the  Meat  and 

Poultry Processing Capacity Technical Assistance Program (MPPTA), the Local Meat Capacity 

Grant (LocalMCap), and the Indigenous Animals Grant (IAG), which together aim to support the 

growth  of  the  meat  and  poultry  supply  chain,  increase  access  to  inspection  and  certification 

programs, expand domestic processing capacity, and improve the ability of independent facilities 

to serve more customers in more markets (United States Department of Agriculture 2023). These 

grants  are  available  to  very  small  and  small  meat  processors  to  “Increas[e]  access  to 

102 

 
 
slaughter/processing  facilities  for  smaller  farms  and  ranches,  new  and  beginning  farmers  and 

ranchers, socially disadvantaged producers, veteran producers, and/or underserved communities” 

and  promote  efforts  for  “developing  new  and  expanding  existing  markets”  (United  States 

Department  of  Agriculture  Agricultural  Marketing  Service  2024). These  efforts  are  important 

steps for the overall U.S. meat and poultry supply chain that also benefit the halal market.  

Furthermore,  additional  rulemaking  endeavors  and  investigations  are  underway  by  the 

USDA’s  Packers  and  Stockyards  Division  to  expand  and  strengthen  the  USDA’s  ability  and 

authority to foster and regulate a more diverse, equitable, and transparent meat and poultry supply 

chain (The United States Government 2022). Results of these investigations may provide new legal 

precedent for enforcing and regulating transparency and labeling in the meat and poultry supply 

chain,  which  will  address  some  of  halal  market  participants’  concerns  about  the  appropriate 

utilization of halal certifications and information accessibility. Likewise, the new rules will enable 

the USDA to prosecute unfair, deceptive, and anticompetitive behavior in the meat and poultry 

industry, which will strengthen the authenticity of labeled products. Altogether, the financial and 

legal efforts at the federal level point to increased interest in revitalizing the U.S. domestic meat 

and poultry industry. This revitalization will include diversification of the products available to 

consumers – such as certified halal products – and improvements in enforcement of traceability, 

labeling, and other authenticity verification strategies that would positively impact the certification 

landscape of the halal meat and poultry market. While these efforts are not explicitly directed at 

the U.S. domestic halal meat and poultry certification landscape, they have and will continue to 

improve upon certification-related issues such as transparency, information access, certification 

costs and benefits, enforcement, and regular inspections – all of which are areas of concern for the 

halal market. 

103 

 
 
Taken together, public opinion and current presidential leadership do not provide a clear 

answer as to whether a NHMC program will be developed in the near future, nor how it would 

likely be implemented. However, a more transparent, well-regulated U.S. domestic halal meat and 

poultry  market  is  possible  without  direct  federal  intervention  in  the  halal  market,  as  discussed 

above.  Further,  consumer  right-to-know  legislation  provides  legal  protections  for  enforcing 

appropriate  certification  utilization  without  violating  the  religious  protections  in  the  First 

Amendment. Thus, we have the necessary legal framework to facilitate the proper use of a U.S. 

domestic NHMC program, were one to be developed. The only remaining piece of the certification 

implementation process is the certification program itself. The results of this study help identify 

market’s  preferences  for  the  attributes  of  a  NHMC  program.  However,  a  significant  hurdle  in 

developing a NHMC will be reaching a consensus on standards amongst the many different groups 

in  the  market.  If  this  can  be  achieved,  U.S.  domestic  halal  meat  certifiers  develop  the  NHMC 

program, thereby simplifying and strengthening U.S. domestic halal meat and poultry certification. 

Finally, the NHMC program can be implemented effectively by coordinating with federal and state 

governments as needed for proper labeling enforcement and management.  

However, the development of a national certification program in the U.S. is complex, as 

seen in the years-long refinement and adoption of the USDA organic certification. Nonetheless, 

the logistical and institutional experience of developing the USDA organic certification can inform 

and aid in the process of designing a U.S. national halal meat certification program. Most notably, 

perhaps, is the way in which program development can incorporate the preferences of stakeholders 

and consumers. The variation in opinions and preferences for a NHMC program’s standards and 

structure from market agents and how they may impact how a NHMC program would operate are 

important  to  consider  for  the  program  to  be  successful.  Additionally,  when  developing  such  a 

104 

 
 
program, the U.S. can take cues from current halal meat and poultry certification programs in place 

throughout the world, in terms of organizational structure, transparency standards, implementation 

process, and many other attributes.  

7. Conclusions, Limitations, and Future Work  

There is a lack of clarity and standardization in certifications in the U.S. halal meat and 

poultry  market  that  makes  it  challenging  for  participant  –  including  processors,  retailers,  and 

consumers – to engage fully and confidently in the market. Indeed, in interviews conducted for 

this  research,  Muslim  halal  meat  consumers  and  current  halal  meat  retailers  and  processors 

expressed  concern  over  the  lack  of  transparent  and  standardized  certification  requirements  and 

shared desires for the development of such a system. These issues leave much to be desired for 

Muslim Americans. Thus, the purpose of this study was to understand what the U.S. domestic halal 

market  participants  want  out  of  a  NHMC  program,  and  how  these  desires  compare  to  what  is 

feasible  to  implement.  I  achieved  this  objective  by  investigating  the  preferences  for  a  NHMC 

program  for  processors,  retailers,  and  Muslim  halal  meat  consumers.  I  used  a  mixed  methods 

design employing qualitative interviews and stacked surveys with BWS. In all, the findings of this 

chapter aid in bolstering halal meat consumer confidence in the authenticity of their products, as 

well as improve the equity of the U.S. food system. 

My  analysis  reveals  the  preferences  of  market  participants  for  the  design  and 

implementation  of  a  U.S.  NHMC  program  and  the  potential  challenges  that  may  be  faced  in 

developing  such  a  program.  By  studying  patterns  in  processors’,  retailers’,  and  consumers’ 

preferences for different attributes of a U.S. NHMC program, I determine common characteristics 

that should be carefully considered in program design to meet the market’s needs. Results show 

that  the  market  overall  prefers  that  program  designers  consider  most  carefully  Who/What  is 

105 

 
 
 
Certified, Halal Standards, and Costs when developing a U.S. NHMC program. The results show 

that  the  implementation  of  a  NHMC  designed  by  non-governmental  agencies  and  backed  by 

federal consumer-right-to-know legislation may be ideal. In this manner, the First Amendment is 

not  violated,  but  Muslim  Americans  are  granted  further  religious  security.  Finally,  the  data 

indicated  that  multiple  transparency  and  traceability  measures  should  be  included  to  ensure  a 

robust and trustworthy program. 

The hurdles to designing a U.S. NHMC program described in this research are notable, but 

not insurmountable. The design of a U.S. NHMC program will involve multiple groups’ opinions 

and the need to consider a variety of religious and non-religious preferences for program attributes.  

The most difficult part of the process will be reaching a consensus on standards amongst many 

different groups without government involvement, and then coordinating with federal and state 

governments  as  needed  for  proper  enforcement  and  management  without  violating  the  First 

Amendment.  Despite  this  challenge,  the  results  of  the  analyses  described  in  this  work  provide 

detailed information on which attributes are important to consider while developing a U.S. NHMC 

program,  and  also  shed  light  on  how  to  best  suit  the  needs  and  wants  of  the  market. With  this 

information,  program  designers  will  be  well  equipped  to  develop  a  U.S.  NHMC  program. 

Furthermore, steps to strengthen the U.S. meat and poultry industry’s transparency, equity, and 

authenticity in general have already begun to be implemented by the Biden administration and the 

USDA,  and  consumer  right-to-know  legislation  is  in  place  to  enforce  proper  certification 

utilization in the food system overall. Thus, the U.S. domestic halal meat and poultry market is in 

106 

 
 
a prime position to implement a national certification; the only missing piece is the certification 

program itself. 

Despite these novel findings, there are some limitations of this study. First, collecting data 

from supply-side agents in any market is notoriously difficult, which led to lower response rates 

and lower numbers of quality observations in my BWS datasets. As such, the sample sizes for 

processors  and  retailers  were  much  smaller  than  the  sample  of  consumers  and  may  not  be 

representative  of  the  industry  in  this  analysis.  However,  the  findings  from  the  interviews  are 

consistent with those from the surveys, suggesting that the results and conclusions are reasonably 

sound and representative of the market despite the small sample sizes.   

Moving forward there is need for additional research focused on the potential structure and 

design  of  a  U.S.  NHMC  program.  As  my  analysis  is  exploratory  in  nature,  additional  research 

should be conducted to better describe the attribute preferences expressed in the BWS. A deeper 

understanding of these preferences would aid in the design of a more effective program. To achieve 

this, future work would benefit from larger sample sizes for processors and retailers to ensure more 

representative  data.  As  in  the  case  of  Essay  1,  there  are  two  methods  that  I  believe  could  be 

effective for increasing sample size when working with these populations. First, utilizing a team 

of  researchers  to  conduct  in-person  or  virtual  (e.g.,  Zoom  or  phone)  surveys  in  real-time  may 

increase response rates and quality, as processors and retailers are typically less likely to complete 

surveys.  Secondly,  some  of  the  U.S.  halal  processor  and  retailer  communities  are  nonnative 

English speakers, especially older individuals and recent immigrants. As such, researchers may 

benefit from close partnerships with native Arabic and Urdu speakers when collecting data from 

halal businesses to increase participation rates.  

107 

 
 
 
Additionally,  future  investigation  into  the  design  and  implementation  of  a  U.S.  NHMC 

program  would  benefit  from  more  information  on  the  process  for  making  complex  food  and 

agricultural policies, regulations and laws, especially in the case of religious standards. Potential 

avenues  to  acquire  this  knowledge  would  be  interviews  with  representatives  of  the  USDA, 

Congresspersons, individuals who helped to design the USDA Organic standards, and government 

or certifier representatives from other countries around the world that have national level halal 

meat and poultry certification programs. 

Overall,  my  findings  and  additional  suggestions  for  additional  research  will  help  future 

work  make  meaningful  contributions  to  our  understanding  and  support  of  regulation  and 

certification within this unique market, as well as add to the literature on the design of policies to 

support both supply-side agents and consumers. 

108 

 
 
 
BIBLIOGRAPHY 

An Update on Food Fraud. (2020, January 31). Food Safety Net Services (FSNS). Retrieved 
November 15, 2021, from https://fsns.com/2020/01/31/an-update-on-food-fraud/#. 

Australian Halal. (2022). Australian Halal. https://australiahalal.com/en/index.html  

Bandoim, Lana. 2019. “Can Blockchain and Chip Technology Improve Beef Sourcing 

Transparency?” Forbes. Accessed Aug 8, 2022. 
https://www.forbes.com/sites/lanabandoim/2019/04/30/can-blockchain-and-chip- technology-
improve-beef-sourcing-transparency/#441ecf473284  

Batz, F. ‐J., Peters, K. J., & Janssen, W. (1999). The influence of technology characteristics on the 

rate and speed of adoption. Agricultural Economics, 21(2), 121–130. 
https://doi.org/10.1111/j.1574-0862.1999.tb00588.x 

Bazzani, C., Caputo, V., Nayga, R. M., & Canavari, M. (2017). TESTING COMMITMENT COST 

THEORY IN CHOICE EXPERIMENTS: COMMITMENT COST IN CHOICE 
EXPERIMENTS. Economic Inquiry, 55(1), 383–396. https://doi.org/10.1111/ecin.12377 

Boxall, P.C., Adamowicz, W.L. Understanding Heterogeneous Preferences in Random Utility 

Models: A Latent Class Approach. Environmental and Resource Economics 23, 421–446 (2002). 
https://doi.org/10.1023/A:1021351721619  

Caswell, J. A. (1998). How Labeling of Safety and Process Attributes Affects Markets for Food. 

Agricultural and Resource Economics Review, 27(2), 151–158. 
https://doi.org/10.1017/S106828050000647X. 

Collart, A. J., & Canales, E. (2022). How might broad adoption of blockchain‐based traceability 

impact the U.S. fresh produce supply chain? Applied Economic Perspectives and Policy, 44(1), 
219–236. https://doi.org/10.1002/aepp.13134 

Economically Motivated Adulteration (Food Fraud). (2021, November 4). U.S. Food & Drug 

Administration (FDA). Retrieved November 15, 2021 from 
https://www.fda.gov/food/compliance-enforcement-food/economically-motivated-adulteration-
food-fraud. 

El-Osta, H. S., & Morehart, M. J. (2000). Technology Adoption and Its Impact on Production 

Performance of Dairy Operations. Review of Agricultural Economics, 22(2), 477–498. 
https://doi.org/10.1111/1058-7195.00034 

Finn, A., & Louviere, J. J. (1992). Determining the Appropriate Response to Evidence of Public 

Concern: The Case of Food Safety. Journal of Public Policy & Marketing, 11(2), 12–25. 
https://doi.org/10.1177/074391569201100202 

109 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
Guthman, J. (1998). Regulating Meaning, Appropriating Nature: The Codification of California 
Organic Agriculture. Antipode, 30(2), 135–154. https://doi.org/10.1111/1467-8330.00071 

Halal Food Act, 410 Ill. Compiled Statute 637 (2002). 

https://www.ilga.gov/legislation/ilcs/ilcs3.asp?ActID=1581&ChapterID=35. 

Halal Food Market: Global Industry Trends, Share, Size, Growth, Opportunity and Forecast 2021-

2026. (2021, April). ResearchandMarkets.com. Retrieved November 15, 2021 from 
https://www.researchandmarkets.com/reports/5311860/halal-food-market-global-industry-trends-
share?utm_source=BW&utm_medium=PressRelease&utm_code=5w4lr3&utm_campaign=1527
350+-+%241.9+Trillion+Halal+Food+Market+-
+Global+Industry+Trends%2c+Share%2c+Size%2c+Growth%2c+Opportunity+and+Forecast+t
o+2026&utm_exec=chdo54prd. 

Halal Food Market in U.S. 2020-2024. (2020, June). ResearchandMarkets.com. Retrieved November 

15, 2021 from https://www.researchandmarkets.com/reports/5125564/halal-food-market-in-us-
2020-
2024?utm_source=BW&utm_medium=PressRelease&utm_code=nvjb5t&utm_campaign=14495
38+-+Halal+Food+Market+in+U.S.+2020-
2024%3a+Market+Size%2c+Forecasts%2c+Trends%2c+Drivers%2c+and+Challenges&utm_ex
ec=chdo54prd. 

Halal Monitoring Services. (2022). About HMS. Halal Monitoring Services – About Us. Retrieved 

August 8, 2022, from https://hmsusa.org/about.html  

Hassan, F., & Hanif, A. (2017). Halal issues in processed food: Misuse of the Halal logo. Journal of 

Emerging Economies and Islamic Research, 5(3), 1. https://doi.org/10.24191/jeeir.v5i3.8826 

Horne, J. (2012). How anchored tradeoffs reveal customer preferences. Market Strategies 

International. 

Islamic Services of America. (2022). Halal Meat & Halal Food Certification: ISA. Islamic Services 
of America. Retrieved August 8, 2022, from https://www.isahalal.com/about-isa/what-I-do.  

Johnston, R. J., Roheim, C. A., Donath, H., Asche, F., Johnston, R. J., Roheim, C. A., Donath, H., & 

Asche, F. (2001). MEASURING CONSUMER PREFERENCES FOR ECOLABELED 
SEAFOOD: AN INTERNATIONAL COMPARISON. https://doi.org/10.22004/AG.ECON.31157 

Lin, W., Ortega, D. L., Ufer, D., Caputo, V., & Awokuse, T. (2020). Blockchain‐based traceability 
and demand for U.S. beef in China. Applied Economic Perspectives and Policy, aepp.13135. 
https://doi.org/10.1002/aepp.13135 

Lohr, L. (1998). Implications of Organic Certification for Market Structure and Trade. American 
Journal of Agricultural Economics, 80(5), 1125–1129. https://doi.org/10.2307/1244216 

110 

 
 
 
 
 
 
 
 
 
 
 
 
 
Louviere, J., Lings, I., Islam, T., Gudergan, S., & Flynn, T. (2013). An introduction to the application 

of (case 1) best–worst scaling in marketing research. International Journal of Research in 
Marketing, 30(3), 292–303. https://doi.org/10.1016/j.ijresmar.2012.10.002  

Lusk, J. L., & Briggeman, B. C. (2009). Food Values. American Journal of Agricultural Economics, 

91(1), 184–196. https://doi.org/10.1111/j.1467-8276.2008.01175.x 

Marley, A. A. J., & Louviere, J. J. (2005). Some probabilistic models of best, worst, and best–worst 

choices. Journal of Mathematical Psychology, 49(6), 464–480. 
https://doi.org/10.1016/j.jmp.2005.05.003 

McFadden, D. (1974). Conditional logit analysis of qualitative choice behavior. Zarembka (Ed.), 

Frontiers of Econometrics, Academic, New York. 

McKendree, M.G.S., Widmar, N. O., Ortega, D. L., & Foster, K. A. (2013). “Consumer 

preferences for verified pork-rearing practices in the production of ham products.” Journal of 
Agricultural and Resource Economics, 38(3), 397-417. 

McKendree, M. G. S., Tonsor, G. T., & Wolf, C. A. (2018). Animal Welfare Perceptions of the U.S. 

Public and Cow-Calf Producers. Journal of Agricultural and Applied Economics, 50(4), 544–
578. https://doi.org/10.1017/aae.2018.14 

Mohamed, Besheer. (2021, September 1). Muslims are a growing presence in U.S., but still face 

negative views from the public. Pew Research Center. https://www.pewresearch.org/short-
reads/2021/09/01/muslims-are-a-growing-presence-in-u-s-but-still-face-negative-views-from-
the-public/  

Mohamed, Besheer. (2018, January 3). New estimates show U.S. Muslim population continues to 

grow. Pew Research. https://www.pewresearch.org/fact-tank/2018/01/03/new-estimates-show-u-
s-muslim-population-continues-to-grow/. 

Ortega, David L., and Christopher A. Wolf. “Demand for Farm Animal Welfare and Producer 

Implications: Results from a Field Experiment in Michigan.” Food Policy, vol. 74, Jan. 2018, pp. 
74–81. DOI.org (Crossref), doi: 10.1016/j.foodpol.2017.11.006  

Ortega, D. L., Wang, H. H., Wu, L., & Olynk, N. J. (2011). Modeling heterogeneity in consumer 

preferences for select food safety attributes in China. Food Policy, 36(2), 318–324. 
https://doi.org/10.1016/j.foodpol.2010.11.030 

Ortega, D. L., Ward, P. S., & Caputo, V. (2019). Evaluating producer preferences and information 
processing strategies for drought risk management tools in Bangladesh. World Development 
Perspectives, 15, 100132. https://doi.org/10.1016/j.wdp.2019.100132 

Pearson, D., & Henryks, J. (2008). Marketing Organic Products: Exploring Some of the Pervasive 

Issues. Journal of Food Products Marketing, 14(4), 95–108. 
https://doi.org/10.1080/10454440801986421 

111 

 
 
 
 
 
 
 
 
 
 
 
 
 
Poe, G. L., Giraud, K. L., & Loomis, J. B. (2005). Computational Methods for Measuring the 

Difference of Empirical Distributions. American Journal of Agricultural Economics, 87(2), 353–
365. http://www.jstor.org/stable/3697850 

Rahman, A.A, Md. Ismail, C.T. & Abdullah, N.A. (2018). Regulating Halal Food Consumption: 
Malaysian Scenario. International Journal of Law, Government and Communication, 3 (13), 
313-321.  

Sønderskov, K. M., & Daugbjerg, C. (2011). The state and consumer confidence in eco-labeling: 

Organic labeling in Denmark, Sweden, The United Kingdom, and The United States. Agriculture 
and Human Values, 28(4), 507–517. https://doi.org/10.1007/s10460-010-9295-5. 

Stata. (2013, February 11). Statistical Software for data science | Stata. Stata.com. Retrieved August 

8, 2022, from https://www.stata.com/manuals13/mvfactor.pdf.  

Statista Research Department. (2016, Oct. 3). Major growth opportunities of halal food market 

worldwide as of 2016. Statista. https://www.statista.com/statistics/784921/halal-food-global-
market-opportunities/. 

The United States Government. (2022, January 3). Fact Sheet: The Biden-Harris Action Plan for a 
Fairer, More Competitive, and More Resilient Meat and Poultry Supply Chain. The White 
House. https://www.whitehouse.gov/briefing-room/statements-releases/2022/01/03/fact-sheet-
the-biden-harris-action-plan-for-a-fairer-more-competitive-and-more-resilient-meat-and-poultry-
supply-chain/  

Tian, Feng. (2016). An agri-food supply chain traceability system for China based on RFID & 

blockchain technology. 2016 13th International Conference on Service Systems and Service 
Management (ICSSSM), 1–6. https://doi.org/10.1109/ICSSSM.2016.7538424 

Ubilava, D., & Foster, K. (2009). Quality certification vs. product traceability: Consumer preferences 

for informational attributes of pork in Georgia. Food Policy, 34(3), 305–310. 
https://doi.org/10.1016/j.foodpol.2009.02.002 

United States Department of Agriculture. (2023). USDA Announces Funding Availability to Expand 
Meat and Poultry Processing Options for Underserved Producers and tribal Communities. 
United States Department of Agriculture. https://www.usda.gov/media/press-
releases/2023/04/19/usda-announces-funding-availability-expand-meat-and-
poultry#:~:text=These%20new%20grant%20programs%2C%20the,midsized%20meat%20and%
20poultry%20processors 

United States Department of Agriculture, Agricultural Marketing Service. (2024). Meat and Poultry 
Inspection Readiness Grant. Meat and Poultry Inspection Readiness Grant | Agricultural 
Marketing Service. https://www.ams.usda.gov/services/grants/mpirg  

112 

 
 
 
 
 
 
 
 
 
 
 
 
 
Ugochukwu, A. I., & Phillips, P. W. B. (2018). Technology Adoption by Agricultural Producers: A 
Review of the Literature. In N. Kalaitzandonakes, E. G. Carayannis, E. Grigoroudis, & S. 
Rozakis (Eds.), From Agriscience to Agribusiness (pp. 361–377). Springer International 
Publishing. https://doi.org/10.1007/978-3-319-67958-7_17 

Verify Halal. (2022). Verify halal. Verify Halal. Retrieved August 6, 2022, from 

https://www.verifyhalal.com/  

Washington Center for Equitable Growth. (2018, May 23). Market Power in the U.S. Economy Today. 

Washington Center for Equitable Growth. https://equitablegrowth.org/market-power-in-the-u-s-
economy-today/  

Winship, C., & Mare, R. D. (1984). Regression Models with Ordinal Variables. American 

Sociological Review, 49(4), 512. https://doi.org/10.2307/2095465 

Wolf, C. A., & Tonsor, G. T. (2013). Dairy Farmer Policy Preferences. 

https://doi.org/10.22004/AG.ECON.158286 

Zabihah. (2021). Halal certifying authorities. https://www.zabihah.com/aut. 

113 

 
 
 
 
 
 
 
 
 
CHAPTER 3. RESOLVING THE REALITY GAP IN FARM 
REGULATION VOTING MODELS 

A version of this Chapter was previously published in Food Policy and is reproduced with the 

permission of the coauthors. DOI: https://doi.org/10.1016/j.foodpol.2022.102357  

Hopkins, K.A., McKendree, M.G.S., & Schaefer, K. Aleks. (2022). Resolving the reality gap in 

farm regulation voting models. Food Policy, 112, 102357. 

© 2022 Elsevier Ltd. All rights reserved. 

1. Introduction 

Throughout the world, many national, provincial, state, and local jurisdictions engage in 

legislative efforts to regulate food production systems beyond federal or overarching requirements. 

This  behavior  has  been  seen  recently  with  members  of  the  European  Union  (EU),  including 

Germany,  Italy,  Spain,  France,  and  the  United  Kingdom  (UK)13,  who  have  passed  additional 

agricultural production regulations beyond the EU's standards (Vogeler, 2019a, 2019b). Similarly, 

over the past two decades in the United States (US), some states have made concerted efforts to 

wrest from Congress the regulatory control of food production systems within their borders (Neill 

et  al.,  2020).  These  state  regulatory  efforts  tend  to  focus  on  socially  controversial  agricultural 

practices,  such  as  genetically  modified  varieties  or  use  of  production  enclosures  or  farming 

practices deemed not to promote farm animal welfare (FAW).  

The  resulting  laws  can  have  substantial  negative  economic  effects  for  agricultural 

producers,  including  unfunded  mandates  and  vote-buy  gaps  (Sumner  et  al.,  2008).  Unfunded 

mandates in agriculture arise when a law passes that requires changes to agricultural production 

13 These regulations were passed prior to the UK's withdrawal from the EU in 2020. 

114 

 
 
 
 
 
practices but provides no monetary assistance to the producer to implement these changes (Paul et 

al.,  2019).  A  vote-buy  gap  occurs  when  citizens  vote  or  express  support  for  a  law  to  regulate 

products, but then demonstrate little demand for these specialized products (Norwood et al., 2019). 

Further,  these  regulations  effect  producers  and  consumers  both  in-state  (Sumner  et  al.,  2008; 

Malone & Lusk, 2016; Mullally & Lusk, 2018; Ortega & Wolf, 2018) and out-of-state (Carter et 

al., 2021; Carter & Schaefer, 2019; Sumner, 2017).  

The rise of piecemeal state legislation surrounding labeling of genetically modified foods 

led  Congress  to  implement  the  National  Bioengineered  Food  Disclosure  Standard  (NBFDS), 

which  preempts  state  efforts  to  regulate  the  labeling  of  genetically  modified  foods  (Bovay  & 

Alston,  2018).  However,  states  have  almost-unfettered  power  to  develop  new  mandates  or 

restrictions in the area of FAW. As shown in Figure 9, this authority has culminated in 19 state-

level  bills  and  ballot  initiatives  concerning  FAW  across  13  states  through  2020  (Animal 

Agricultural  Alliance,  2021).  In  light  of  the  substantial  negative  economic  effects  of  FAW 

regulations for agricultural producers and other stakeholders, a natural question emerges – can one 

empirically assess how and why these measures occur in US states' regulatory landscapes?  

Figure 9: Timeline of All Enacted FAW Regulations 2000-2021 

Note: Bills are above the horizontal axis, have square markers, and are blue in color; ballot initiatives are below the 
axis, have round markers, and are orange in color. The bills in Kentucky and Ohio were administrative regulations or 
revised  statutes.  Regulations  occurring  after  2020  were  not  included  in  this  article’s  analysis  due  to  limited  data 
availability. 

115 

 
 
 
 
We seek to model the evolution of the state-level FAW regulatory landscape as a function 

of legislature characteristics and constituent demographics. More specifically, we utilize a two-

stage model to assess (i) whether and when a given state considers FAW measures, and (ii) if so, 

the  likelihood  the  measures  are  passed.  Using  this  model,  we  estimate  the  likelihood  of  FAW 

adoption outcomes for all 50 states. Using these predictions, we then estimate the cost to the egg 

and pork industries to upgrade to cage- and crate-free production methods in the states most likely 

to pass a FAW regulation in the future. We believe this exercise will assist producers and industry 

stakeholders in gauging the future of the regulatory landscape and provide guidance on whether to 

upgrade existing enclosures to comply with mandates on the horizon or to continue operating with 

“conventional” enclosures.  

Of  course,  we  are  not  the  first  to  attempt  to  understand  the  uptake  of  state-level  farm 

regulation. Videras (2006) first analyzed whether religious demographic variables could be used 

to predict voting outcomes in the context of the 2002 Florida Animal Cruelty Amendment. Results 

showed that Catholicism and Evangelism had strong, conflicting effects on support for the FAW 

ballot initiative. Smithson et al. (2014) expanded on (Videras, 2006) by analyzing the demographic 

drivers  of  voting  under  the  2008  California  Proposition  2  ballot  initiative.  The  authors  further 

created predictions for all 50 states to determine which states and animal agriculture industries 

have a high probability of future FAW regulations. Bovay and Alston (2016) develop a similar 

approach to model the probability of genetically modified organism (GMO) labeling restrictions 

across all 50 states based on California Proposition 37. Similarly, Bovay and Sumner (2019) used 

voting results from both California Proposition 2 and the Prevention of Farm Animal Cruelty Act 

in Massachusetts in 2016 to draw connections between political party affiliation and support for 

FAW initiatives.  

116 

 
 
However, one persistent puzzle in this line of research is that studies tend to over-predict 

state-level FAW regulation. Smithson et al. (2014), for example, predicted between 46% and 100% 

of  all  50  US  states  would  pass  a  FAW  law  through  a  ballot  initiative,  if  such  initiatives  were 

considered. Similarly, Bovay and Sumner (2019) predicted that nearly all 50 states would have 

passed FAW regulations in the 2008 and 2016 election years. If FAW measures are so universally 

popular, why has this widespread adoption not already occurred?  

We posit that this gap between the literature and reality is a function of inherent selection 

in whether and when states consider this type of regulation. In contrast to previous papers,14 our 

model utilizes a two-stage, three-part process to account for potential selection bias between bills 

and  ballot  initiatives  by  state  legislators.  A  large  body  of  research  in  political  science  has 

established how the use of ballots versus bills functions in state legislature behavior (Matsusaka, 

1992; Boehmke & Patty, 2007). Consider, for example, the different costs to legislators and citizen 

groups of voting on a proposed law. It is much cheaper for an interest group to lobby a handful of 

legislators than the population of a state, and so this is usually the first step (Matsusaka, 1992). For 

the legislators, costs are more complex: they must weigh their personal views, party views, and 

the views of their constituents before deciding to vote. Our specification allows us to consider the 

effect that legislative decisions have on the success of a proposed bill or ballot initiative, especially 

when using these models to forecast voting outcomes in other states.  

2. Background 

In the US, individual states can enact their own FAW laws and regulations, so long as these 

laws do not contradict laws passed by the federal government.  This type of legislative process is 

similar to that exercised by members of the EU, in which some countries have chosen to enact 

14 Previous papers include an earlier version of this work, Hopkins et al. (2020). 

117 

 
 
 
their own laws to regulate agricultural production that go beyond what is required by the EU. In 

the US, FAW legislation is passed at the state level through two major avenues: through a bill 

voted on in the legislature or through a ballot initiative voted on directly by citizens.  

Since 2007, 11 FAW laws have been enacted through the legislative process. All but one 

US state has a bicameral legislative body, which is a two-body legislature made up of the State 

Senate  and  State  House  of  Representatives.15  The  process  to  pass  a  legislative  bill  in  the  US 

involves  several  stages.  Typically,  one  or  more  representatives  drafts  a  bill  to  present  to  the 

legislative body that they reside in.16 The bill will then be considered by smaller, more focused 

committees within the respective body, and if it passes in committee(s), it can be voted on by the 

entire body.17 Once a bill passes by majority vote in either the House or Senate, it is sent to the 

other legislative body and goes through the entire process again. If it passes a vote in the second 

legislative body, it is then referred to the Governor of the state, who can either sign it into law or 

veto it.18  

Ballot initiatives are an option to create laws in 24 of the 50 US states.19 A ballot initiative 

is typically proposed by citizens of the state, an interest group, or some other non-governmental 

organization (NGO). Ballot initiatives in the US fall into two categories: direct and indirect ballot 

initiatives. Direct ballots bypass the legislature at every step of the ballot initiative process; that is, 

they do no not require approval or action by legislatures to be placed on a ballot, so long as a state's 

15 Nebraska is the only state with a unicameral legislature, meaning it only has one legislative body. 
16 Representatives may draft legislation to support their own beliefs, the beliefs of their constituents, or in response 
to lobbying from interest groups. 
17 It is common that bills do not make it through committee or do not receive a vote on the floor of the Senate or 
House once through a committee. 
18 If there is two-thirds support of the vetoed law in both legislative bodies, vetoed laws can be passed into law 
without the Governor's signature, though this is rare.  
19 Each state has its own requirements and processes that must be followed before the initiative can be considered, 
and these requirements vary greatly across states; therefore, our empirical analysis cannot differentiate between 
direct versus indirect ballot initiatives. 

118 

 
 
 
signature and legal filing requirements have been met.  On the other hand, indirect ballot initiatives 

must be approved by the state legislature before they can appear on a ballot. In either case, typically 

a subset of the state legislature, such as a Budget Committee, is involved in the ballot initiative 

process to the extent that they conduct a financial and legal analysis of the proposed law to present 

to the petitioners. Some states also require the legislature to hold formal hearings or an open forum 

about the initiative proposed. At any point in the direct or indirect ballot initiative process, the 

legislature can decide to pass a bill to enact the regulations proposed in the ballot initiative. Once 

an initiative is on a ballot, it is voted on by citizens and will become law if it passes with a majority 

of the vote.20,21 Six FAW laws have been passed through ballot initiatives since 2002.  

A list of the FAW bills and ballot initiatives analyzed in this article is given in Table 31. 

Of the laws enacted, 10 of them involve confinement standards for egg-laying hens or the sale of 

eggs from hens raised in battery cages, 11 involve confinement standards for gestating sows, and 

11  involve  confinement  of  veal  calves.  There  were  an  additional  two  bills  regulating  the 

confinement of veal calves and gestating sows that were vetoed by the states' governors (MI 2019 

and NJ 2013).  

20 Some states have specific requirements for majority vote, such as a 60% super-majority or a majority of all ballots 
cast, even if a person declined to vote on the initiative. 
21 Once passed by majority vote, the outcome may need to be confirmed by the legislature before being passed into 
law and may still require the Governor's signature. However, this additional legislative confirmation process is 
normally more of a formality; once the citizens have expressed majority support for a ballot initiative, the law will 
be enacted. 

119 

 
 
 
 
 
Table 31: Farm Animal Welfare Legislative Bills and Ballot Initiatives Analyzed 

Type 
Ballot 

Title of Legislation 

Amendment 10 – HSUS Ballot Initiative: 
Gestating Sows 
Proposition 204 
Or. Rev. Stat. §600.150 
SB 201 
Proposition 2 
Mich. Comp. Laws. Ann. §287.746 

Ballot 
Bill 
Bill 
BallotD 
Bill 
BallotD  Amendment 2 – Livestock Care Standards 

Year  State 
FL 
2002 

Industry Affected 
Pork 

Pork, Veal 
Pork 
Pork, Veal 

2006  AZ 
2007  OR 
2008  CO 
2008  CA  Eggs, Pork, Veal 
2009  MI 
Eggs, Pork, Veal 
2009  OH  Veal 

Bill 
Bill 
Bill 
Bill 
Bill* 
Ballot 
Bill* 
Bill 

BallotD 
Bill 
Bill 
Bill 

Amendment 
2010  CA  Eggs 
AB 1437 
SB 805 – Relating to egg-laying hens 
2011  OR  Eggs 
Wash. Rev. Code §69.25.065 and §69.25.107  2011  WA  Eggs 
SB 2191 
SB 1921 
Prevention of Farm Animal Cruelty Act 
SB 660 
HB 7456 – Unlawful Confinement of a 
Covered Animal 
Proposition 12 
SB 174 
S.B. 1019 
HB 2049 – Concerning commercial egg layer 
operations 

Pork, Veal 
Pork 

RI 
2012 
2013  NJ 
2016  MA  Eggs, Pork, Veal 
2018  MI 
RI 
2018 

Eggs, Pork 
Eggs, Pork, Veal 

2018  CA  Eggs, Pork, Veal 
2019  MI 
Eggs 
2019  OR  Eggs 
2019  WA  Eggs 

* Represents a bill that passed but was vetoed by the state’s governor. All other bills or ballots have been enacted.  D 
Indicates a direct ballot initiative. All other ballot initiatives are indirect. 

Relationships  between  demographic  characteristics  and  support  for  FAW  are  well 

documented in agricultural economics and political science literature and are therefore important 

to consider when studying FAW regulation adoption. Smithson et al. (2014) found that an increase 

in median household income and an increase in poverty rate both correlated with a decrease in 

support for FAW regulations. Educational achievement has been shown to correlate with lower 

support for regulations to increase FAW, as more educated individuals are more likely to view 

animal and human similarities and differences more scientifically and this may change their views 

on FAW (Jerolmack, 2003).  

Previous research suggests that religion plays a large role in an individual's view of the 

natural world and thus impacts views on animals and animal welfare (Videras, 2006). For instance, 

120 

 
 
 
 
Catholics tend to be more supportive of animal welfare issues than Protestants and Evangelicals 

(Smithson et al., 2014; Oldmixon, 2017). Overall, non-religious, non-Christians, and Catholics are 

most  in  favor  of  FAW  over  Christians  (Cornish  et  al.,  2016;  Jerolmack,  2003;  Flynn,  2001). 

Jerolmack (2003) found that Jewish and other religions were more likely to support animal rights. 

In  Islam,  concern  for  animal  welfare  and  animal  rights  are  key  moral  and  religious  values 

(Gharebaghi et al., 2007). Several studies have also shown that non-white Americans tend to view 

regulation to increase FAW more positively (Jerolmack, 2003; Franklin et al., 2001; Nibert, 1994; 

Peek  et  al.,  1996;  Uyeki  &  Holland,  2000).  Over  the  past  two  decades,  FAW  has  become  an 

increasingly  politicized  issue  in  the  US  (Feindt  et  al.,  2020;  Lai  et  al.,  2021;  Vogeler,  2020). 

Membership in the Democratic party has been linked to higher concern for animal welfare in a 

wide variety of past studies and contexts, including McKendree et al. (2014), Deemer and Lobao 

(2011), Czech and Borkhataria (2001), Miele et al. (1993), and Heleski et al. (2006). Furthermore, 

from past studies using voter data, we know that liberals are more supportive of animal welfare 

measures, in general, than conservatives (Smithson et al., 2014).  

3. Methodology 

We utilize a two-stage, three-part multinomial endogenous switching regression (MESR) 

to model the implementation of FAW regulations and account for decision selection bias.22,23 In 

22 Selection bias is a common challenge in studies using nonrandomized data to model decisions and outcomes. 
Methodologically, most studies in this area have used propensity score matching (PSM); however, the PSM 
approach does not correct selection bias from unobserved factors (Abdulai, 2016; Jaleta et al., 2016). Unlike PSM, 
MESR models employ a selection correction method by calculating an Inverse Mills Ratio (IMR) using the theory of 
truncated normal distribution to correct selection bias (Bourguignon et al., 2007). The IMR is the ratio of the 
probability density function to the complementary cumulative distribution function of a distribution. This technique 
is commonly used in development economics to account for unobserved heterogeneity and selection bias in farmers' 
cropping decisions (Kassie et al., 2015; Di Falco, 2014). 
23 One method for adjusting for this bias may be Heckman's (Heckman, 1976) two-stage model; however, we must 
consider the different levels of data aggregation, inclusion of population weights, use of both panel and cross-
sectional data sets, and likelihood of unobserved heterogeneity in our analysis. Thus, the Heckman approach would 
give inconsistent estimates if selection bias originating from observed and unobserved heterogeneity is not 
addressed. 

121 

 
 
 
the MESR framework, the two stages are modeled simultaneously.  A simple schematic of our 

MESR model is depicted in Figure 10. We hypothesize the likely source of selection bias occurs 

at the state legislative level – legislatures may endogenously self-select different FAW legislative 

actions, and decisions are likely to be influenced by unobserved factors that may be correlated 

with  outcome  variables.  Additionally,  accounting  for  institutional  characteristics  is  crucial  to 

understanding the relationship between policy changes and politics, particularly in FAW where 

passing regulations using democratic instruments such as ballot initiatives is common (Vogeler, 

2020).  The  MESR  allows  us  to  take  into  account  the  characteristics  of  individual  states' 

governments and their impact on the likelihood of a FAW bill or ballot initiative occurring and 

passing. These relationships are modeled in Stage 1 (Legislature Action Decisions) of our model. 

We then account for decision selection bias in our second stage FAW regulation voting outcome 

models  –  Stage  2.1  (Ballot  Initiative  Voting  Outcomes)  and  Stage  2.2  (Legislative  Bill  Voting 

Outcomes) by including the first-stage Inverse Mills Ratio (IMR) as an explanatory variable and 

bootstrapping to compute coefficient and standard error estimates.  

3.1. Econometric Model 

We discuss each of the components in turn. 

122 

 
 
 
 
Figure 10: Farm Animal Welfare Voting Two-Stage Selection Model Schematic 

3.1.1 Stage 1: Legislative Action Decisions  

Here, we model state legislatures' choice of alternative legislative actions for addressing 

FAW concerns – no action, bill proposed, or ballot initiative allowed – using a multinomial logit 

selection  (MNLS)  model  accounting  for  unobserved  heterogeneity  (Equation  18).  Stage  1  is 

aggregated at the state level and consists of 20 years of annual panel data for 49 of the 50 states (n 

= 980).24 We cluster the standard errors by state. The dependent variable can take three values: 

y = 0 if no action was taken, 𝑦 = 1 if a ballot initiative was placed on a ballot, or 𝑦 = 2 if a bill 

was proposed to the state legislature.  

We estimate the following first-stage model: 

𝑌$ = {0,1,2} = (𝑋$

( × β) + (𝐷$

( × δ) + (𝐶$

( × κ) + ε$

where 𝑌$ is the predicted action outcome of a state legislature, 𝑋$ is the matrix of state legislature 

political variables – dummy variables to indicate whether the state house, senate and governor are 

all  of  the  same  political  party  (denoted  TRIFECTA_D  and  TRIFECTA_R),  and  continuous 

Equation 18 

24 Nebraska is excluded from this data set due to its unique unicameral state government system and because the 
state's legislators are not required to affiliate with a political party. 

123 

 
 
 
 
 
variables  indicating  the  percent  of  house  and  senate  seats  occupied  by  Democrats  (denoted 

HOUSE%D  and  SENATE%D)  –  with  β  the  corresponding  coefficients.  𝐷$  is  the  matrix  of 

legislative  characteristic  variables  –  counts  of  previous  animal  welfare  legislation  (denoted 

COUNT_PASSED_PREV), a dummy variable indicating whether the state had previously passed 

FAW  regulation  (denoted  PREV_LAW),25  and  a  dummy  variable  indicating  whether  the  state 

allows ballot initiatives (denoted ALLOW_BALLOT) – with δ the corresponding coefficients. 𝐶$ is 

the  matrix  of  state  agricultural  industry  density  variables  (denoted  HENS_PER_1000  and 

HOGS_PER_1000) which are the number of egg-laying hens and gestating sows per 1,000 people 

in each state, respectively – with κ the corresponding coefficients. ε$ is the error term.  

The IMRs for each of the second stage regressions are then calculated (Equation 19 and 

Equation 20) from their respective estimated outcome probabilities in the MNLS model: 

𝐼𝑀𝑅$ =

𝑓(𝑌$)
𝐹K(𝑌$)

𝐹K(𝑌$) = 𝑃𝑟(𝑌$ > 𝑦$) = k 𝑓(𝑠)𝑑𝑠

6

7(

Equation 19 

Equation 20 

where 𝑖 = 0,1,2 for the no action, ballot, and bill outcomes, respectively, 𝑓(𝑌$) is the standard 

normal probability density function (PDF), 𝐹K(𝑌$) is the standard normal cumulative distribution 

function (CDF), and 𝑠 is the integration argument. 

25 COUNT_PASSED_PREV and PREV_LAW account for the "snowball effect" of previous laws. A "snowball 
effect" is a situation in which one event influences the likelihood that a similar event occurs (Matsusaka, 2005). 
These two variables also proxy for the diffusion of media coverage related to FAW within and between states, 
increased public awareness of FAW, and other time effects. 

124 

 
 
 
 
 
 
 
3.1.2. Stage 2: Ballot and Bill Voting Outcomes 

In the second stage, we evaluate the voting outcomes for a ballot initiative (Stage 2.1) or a 

legislative bill (Stage 2.2) using ordinary least squares (OLS) with IMRs as additional covariates 

to account for selection bias from time-varying unobserved heterogeneity. 

Stage  2.1  Ballot  Initiative  Voting  Outcomes:  In  Stage  2.1,  we  model  county-level  voting 

outcomes for six ballot initiatives in Arizona (AZ; 2006), California (CA; 2008 & 2018), Florida 

(FL; 2002), Massachusetts (MA; 2016), and Ohio (OH; 2009) (n = 299).  To model these voting 

outcomes, we estimate a log-linear model: 

𝑙𝑛 m

𝑉0
1 − 𝑉0

o = (𝐼𝑀𝑅0

( × ϕ) + (𝐵0

( × ζ) + 𝑢0

Equation 21 

where the dependent variable is the log-odds of the predicted “yes” portion of the ballot initiative 

vote  𝑉0.  𝐼𝑀𝑅0  is  the  IMR  value  from  the  Stage  1  ballot  outcome  with  ϕ  the  corresponding 

coefficient.26  𝐵0 

is 

the  matrix  of  county  demographics 

including  people  per  farm 

(PEOPLE_PER_FARM),27  percent  of  voters  in  the  county  who  voted  for  the  Democratic 

presidential candidate in the year closest to when the initiative was on the ballot (%DEMOCRAT), 

median household income in thousands of dollars (HOUSEHOLD_INCOME_1000), the percent 

of  people  in  poverty  (POVERTY_RATE),  the  percent  of  persons  of  25+  years  of  age  with  a 

bachelor's  degree  (EDUCATION),  the  percent  of  white  (%WHITE),  Black  (%BLACK),  and 

Hispanic 

citizens 

(%HISPANIC), 

and 

the 

percent 

of  Mainline 

Protestants 

(%MAINLINE_PROTESTANT),  Evangelical  Protestants    (%EVANGELICAL_PROTESTANT), 

26 A table showing the distribution of the Stage 1 ballot IMRs is given in Appendix A.4. 
27 This variable (PEOPLE_PER_FARM) is a proxy for citizens' familiarity with agriculture. 

125 

 
 
 
 
and Catholics (%CATHOLIC) – with ζ the corresponding coefficients.28 There is little accurate 

data available for Jewish and Muslim populations at the county level, and so we unfortunately do 

not include these religions in our model.29 𝑢0 is the error term. Standard errors in the model are 

clustered by state, as we are aggregating data from five different states. 

Stage 2.2 Legislative Bill Voting Outcomes: In Stage 2.2, we model 13 individual bill voting 

outcomes in California (CA; 2010), Colorado (CO; 2008), Michigan (MI; 2009, 2018 & 2019), 

New Jersey (NJ; 2013), Oregon (OR; 2007, 2011 & 2019), Rhode Island (RI; 2012 & 2018), and 

Washington (WA; 2011 & 2019) (n = 1,583).  To estimate bill-voting outcomes, we use a linear 

probability  model  (LPM)  with  continuous  state  legislative-district  level  demographic  data  to 

predict the vote of individual legislators.30 The dependent variable can take two values: 𝑍 = 0 if 

the legislator voted “no”, was absent, or declined to vote on an FAW bill, or 𝑍 = 1 if the legislator 

voted “yes”. We estimate the following model: 

𝑍) = {0,1} = >𝐼𝑀𝑅)

( × τ? + >𝐿)

( × γ? + 𝑢)

where 𝑍) is the actual vote of a given legislator. 𝐼𝑀𝑅) is the vector of IMR values from the Stage 

1 bill outcome with τ the corresponding coefficient. 𝐿) is the matrix of state legislative district 

Equation 22 

28 These variables were selected to be consistent with previous literature on the relationships between demographic 
characteristics and FAW ballot initiative outcomes (Videras, 2006; Smithson et al., 2014; Bovay & Sumner, 2019), 
as well as documented correlations between demographics and support for FAW in general (McKendree et al., 2014; 
Jerolmack, 2003; Deemer & Lobao, 2011; Czech and Borkhataria, 2001; Miele et al., 1993; Heleski et al., 2006; 
Oldmixon, 2017). 
29 We collected county-level data from ARDA the Association of Religious Data Archives 
(http://www.thearda.com/QL2010/) for the percentage of Jewish and Muslim citizens for the bill and ballot states 
involved in our Stage 2 models. This data was very sparse, as these populations are small relative to the overall 
population of the US. Roughly 3/4 of the counties collected did not report a percentage estimate for Muslim citizens 
and about 2/3 did not report a percentage for Jewish citizens.  We attempted to interpolate the data using fractional 
probit and OLS models; however, the results were poor. For example, the model predicted negative Jewish and 
Muslim population estimates for some counties, which is nonsensical. 
30 In order to use the MESR model with IMRs, the second-stage models must be linear, that is, a probit or logit 
function is not compatible. Further, an LPM is as appropriate as a logit or probit for this data so long as our standard 
errors are robust (Bellemare, 2015). 

126 

 
 
 
 
demographic  variables.  These  variables  are  equivalent  to  those  in  Stage  2.1,  except  at  the 

legislative  district,  rather  than  county,  level.31  Additionally,  we  include  a  dummy  variable  to 

indicate the political party of the legislator (denoted DEMOCRAT). Corresponding coefficients are 

represented by γ. 𝑢) is the error term.  

Stage 2 Bootstrapping and Standard-Error Clustering: To help with bias and inconsistency 

and improve inference in the IMR-adjusted second stage models, we bootstrap coefficients and 

standard  errors  using  a  stationary  cluster  block  bootstrap  method.  This  allows  us  to  estimate 

coefficients simultaneously across the two stages of our model (Politis & White, 2004; Politis & 

Romano, 1994; Hall et al., 1995). Our block bootstrapping procedure randomly selects 45 states 

to use in our model and tests out-of-sample prediction accuracy using the five states randomly 

excluded in each repetition.32 

3.2 Data by Stage 

In this section, we describe our data selection and collection process, aggregation levels, 

and summary statistics by stage.  More explicit information on how data were accessed and data 

sources are given in Appendix A.5.  

The variable names, descriptions, and summary statistics for Stage 1 (Legislative Action 

Decisions) are included in Table 32. The majority of our data were collected from the respective 

31 As above, these include people per farm (PEOPLE_PER_FARM), median household income 
(HOUSEHOLD_INCOME_1000), the percent of people in poverty (POVERTY_RATE), the percent of persons 25+ 
years of age with a bachelor's degree (EDUCATION), the percent of white (%WHITE), Black (%BLACK), and 
Hispanic citizens (%HISPANIC), and the percent of Mainline Protestants (%MAINLINE_PROTESTANT), 
Evangelical Protestants (%EVANGELICAL_PROTESTANT), and Catholics (%CATHOLIC). We selected the 
variables based on documented correlations between demographics and support for FAW (McKendree et al., 2014; 
Jerolmack, 2003; Deemer & Lobao, 2011; Czech & Borkhataria, 2001; Miele et al., 1993; Heleski et al., 2006; 
Oldmixon, 2017) and for comparison to the ballot model. 
32 We chose to bootstrap using blocks rather than only clustering by state, as only clustering standard errors by state 
in the Stage 1 MNLS model would likely lead to correlation between observations from the same state. We 
conducted 1,000 repetitions with 35, 40, and 45 states per block in Stage 1.  Models were robust to block size and 
thus we chose the 45 block model. 

127 

 
 
 
state  government  websites  and  Ballotpedia.org.  Agricultural  industry  data  was  collected  from 

USDA QuickStats and the Livestock Marketing Information Center (LMIC).  

Variable names, descriptions, and summary statistics for Stage 2.1 (Ballot Initiative Voting 

Outcomes) are given in Table 33. Counties with less than 2,000 people were dropped, then each 

county within a state was given a weight corresponding to the fraction of its population relative to 

the overall state population, minus any dropped counties.33  

Variable names, descriptions, and summary statistics for Stage 2.2 (Legislative Bill Voting 

Outcomes) are given in Table 34. There is no need to weight this data, as legislative districts are 

drawn to be proportional based on the most recent census data.  

The majority of the demographic data for Stage 2.1 and 2.2 were downloaded from the 

United States Census website (census.gov). Data were used from the census year closest to the 

year the ballot or bill was considered. Stage 2.1 data on vote outcomes were collected from each 

state's records, available online. Stage 2.2 data on legislators' individual votes was collected from 

each state legislature's records, available online. Farm data was collected from USDA QuickStats. 

All religion data were retrieved from the Association of Religious Data Archives.34 

33 Weighting is important when evaluating demographic makeup, as weighting the data by county population 
ensures more realistic and accurate predictions. In the U.S., state-level elections are determined by a majority vote; 
therefore, relative populations should be considered when collecting the data. Our weighting procedure is similar to 
that used by Smithson et al. (2014). This weighting method is intuitive: if County A has two times the population of 
County B, then the demographic percentages and voting outcomes from County A will be given two times the 
importance of those from County B in the state-wide calculations. 
34 As religious data is not reported for all states by legislative district, religious affiliation data for legislative districts 
was obtained by aggregating and averaging the data from counties in the district. 

128 

 
 
 
Table 32: Variable Names and Descriptions for Stage 1 – Legislative Action Decisions 

Variable Name 

Description 

𝑌$, 𝑖 = 0,1,2 

HENS_PER_1000 

HOGS_PER_1000 

TRIFECTA_D 

TRIFECTA_R 

HOUSE%D 

SENATE%D 

COUNT_PASSED_PREV 

PREV_LAW 

ALLOW_BALLOT 

Categorical variable = 0 if the state legislature took no action, 
= 1 if the state legislature allowed a ballot initiative, and = 2 if 
the state legislature proposed a legislative bill 
Continuous variable equal to the number of egg-laying hens 
per 1,000 people in a state in a given year 
Continuous variable equal to the number of gestating sows per 
1,000 people in a state in a given year 
Dummy variable = 1 if all three bodies of state legislative 
branch are controlled by Democrats, = 0 otherwise 
Dummy variable = 1 if all three bodies of state legislative 
branch are controlled by Republicans, = 0 otherwise 
Continuous variable equal to the percent of the state’s House of 
Representatives that belong to the Democratic Party 
Continuous variable equal to the percent of the state’s Senate 
that belong to the Democratic Party 
Discrete variable equal to the total number of farm animal 
welfare regulations in place throughout the US 
Dummy variable = 1 if the state has a previous farm animal 
welfare law in place, = 0 otherwise 
Dummy variable = 1 if the state allows ballot initiatives, = 0 
otherwise 

Mean 

Standard 
Deviation 

Frequency 
(%) 
𝑖 = 0, 95.61% 
𝑖 = 1, 0.92% 
𝑖 = 2, 3.47% 

1430.57 

2474.54 

61.78 

123.08 

0.21 

0.33 

0.41 

0.47 

48.76 

16.89 

47.04 

19.11 

8.60 

0.12 

0.48 

7.55 

0.32 

0.50 

129 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table 33: Variable Names and Descriptions for Stage 2.1 – Ballot Voting Outcomes 

Variable Name 

𝑙𝑛 m

𝑉0
1 − 𝑉0

o 

PEOPLE_PER_FARM 

HOUSEHOLD_INCOME_1000 

POVERTY_RATE 

EDUCATION 

%WHITE 

%BLACK 

%HISPANIC 

%DEMOCRAT 

%CATHOLIC 

%EVANGELICAL_PROTESTANT 

%MAINLINE_PROTESTANT 

Description 

Log of the odds of success of the ballot initiative 

Continuous variable equal to the number of people per farm in the 
county 
Continuous variable equal to the median household income in the 
county in thousands of dollars 
Continuous variable equal to the percent of people in the county who 
are below the poverty line 
Continuous variable equal to the percent of people in the county 25+ 
years old with a 4-year college degree 
Continuous variable equal to the percent of people in the county who 
are white 
Continuous variable equal to the percent of people in the county who 
are Black 
Continuous variable equal to the percent of people in the county who 
are Hispanic 
Continuous variable equal to the percent of voters in the county who 
voted for the Democratic presidential candidate in the presidential 
election year closest to when the initiative was on the ballot 
Continuous variable equal to the percent of people in the county who 
are Catholic 
Continuous variable equal to the percent of people in the county who 
are Evangelical Protestants 
Continuous variable equal to the percent of people in the county who 
are Mainline Protestants 

Mean 

0.31 

Standard 
Deviation 
0.65 

1392.42 

9326.10 

49.56 

16.26 

5.19 

5.23 

20.69 

9.98 

71.99 

21.01 

5.92 

7.44 

15.42 

17.32 

47.42 

12.06 

17.67 

14.04 

12.87 

8.89 

6.52 

4.99 

130 

 
 
 
 
 
 
 
 
 
Table 34: Variable Names and Descriptions for Stage 2.2 – Legislative Bill Voting Outcomes 

Variable Name 

𝑍) 

PEOPLE_PER_FARM 

HOUSEHOLD_INCOME_1000 

POVERTY_RATE 

EDUCATION 

%WHITE 

%BLACK 

%HISPANIC 

DEMOCRAT 

%CATHOLIC 

%EVANGELICAL_PROTESTANT 

%MAINLINE_PROTESTANT 

Description 

Discrete variable = 1 if the legislator voted yes on a farm animal welfare 
bill, = 0 otherwise 
Continuous variable equal to the number of people per farm in the 
legislative district 
Continuous variable equal to the median household income in the 
legislative district in thousands of dollars 
Continuous variable equal to the percent of constituents in the legislative 
district who are below the poverty line 
Continuous variable equal to the percent of constituents in the legislative 
district 25+ years old with a 4-year college degree 
Continuous variable equal to the percent of constituents in the legislative 
district who are white 
Continuous variable equal to the percent of constituents in the legislative 
district who are Black 
Continuous variable equal to the percent of constituents in the legislative 
district who are Hispanic 
Dummy variable = 1 if legislator belongs to the Democratic Party 
Continuous variable equal to the percent of constituents in the legislative 
district who are Catholic 
Continuous variable equal to the percent of constituents in the legislative 
district who are Evangelical Protestants 
Continuous variable equal to the percent of constituents in the legislative 
district who are Mainline Protestants 

Mean 

Standard 
Deviation 

0.80 

0.40 

395.71 

2358.62 

60.57 

18.43 

13.76 

6.92 

30.98 

14.01 

80.81 

16.94 

9.04 

13.90 

13.32 

14.42 

0.59 

0.49 

21.20 

13.47 

10.05 

4.77 

5.75 

22.28 

131 

 
 
 
4. Results 

We present the results of each stage of our model in turn.  

4.1. Stage 1: Legislature Action Decisions 

Results of the multinomial logit model for Stage 1 are in Table 35. with results for the 

ballot outcome on the left and bill outcome on the right. The base case is a legislature not taking 

any action on FAW regulations; that is, neither a bill nor ballot initiative is proposed. For ease of 

interpretation, Stage 1 model coefficients are given in relative risk ratio (RRR) format.35 A value 

greater than one means the risk of the outcome falling in the comparison group rather than the base 

group  increases  as  the  variable  increases.  A  RRR  less  than  one  means  the  risk  of  the  outcome 

falling  in  the  comparison  group  over  the  base  group  decreases  as  the  variable  increases.  For 

example, the coefficient of 1.695 on PREV_LAW for the bill outcome means that a bill is 69.5% 

more  likely  to  occur  than  no  action,  all  else  equal.  Similarly,  the  coefficient  of  0.281  on 

TRIFECTA_R in the ballot column means that a ballot is 71.9% [1.00 - 0.281 = 0.719] less likely 

to occur than no action, all else equal.  

A state's ties to the pork and egg industries resulted in interesting impacts on the likelihood 

that a bill or ballot would be proposed over no action taken. The ratio of gestating sows and egg-

laying  hens  to  1,000  people  within  the  state,  HOGS_PER_1000  and  HENS_PER_1000,  have 

notable predicted influences. While the number of egg-laying hens per 1,000 people does not have 

a correlated effect on the likelihood a ballot or bill is proposed, the number of gestating sows per 

1,000  people  does.  As  the  number  of  gestating  sows  per  1,000  people  increases,  we  see  a 

corresponding 5.2% decrease in the likelihood that a ballot is proposed and a 0.8% decrease in the 

likelihood that a bill is proposed. This implies that states with higher ratios of gestating sows to 

35 A table with non-transformed coefficients is in Appendix A.4. 

132 

 
 
 
people are less likely to propose FAW regulations. This is sensible. States in which this ratio is 

higher likely have larger populations of people familiar with the pork industry, and this familiarity 

will influence the peoples' choice to regulate FAW in pork production. Further, state legislators in 

states  with  prominent  agricultural  industries  likely  have  closer  connections  with  local  Farm 

Bureaus and other agricultural lobbyists. Introducing legislation that is counter to these groups' 

interests can have high political costs from the legislators' perspective.  

The history of FAW regulations is also important. As the number of previous FAW laws 

in the nation overall rises (COUNT_PASSED_PREV), we see a correlation with an increase in the 

likelihood  that  either  a  ballot  or  a  bill  is  proposed  in  any  state,  though  this  correlation  is  only 

significant in the bill case. These results support (Matsusaka, 2005) who asserts that there can be 

a bandwagon effect where legislation in one state leads to proposals of similar legislation in other 

states. When there is already a FAW law in place in a state (PREV_LAW), a bill is more likely to 

be proposed in a given state during a given year than no action taken. This association makes sense, 

as citizens of a state that already had a FAW regulation in place have likely been exposed to more 

media coverage of FAW issues. On the other hand, in states with a previous law, it is 51.5\% less 

likely that a ballot is proposed than no action taken. This makes sense as citizens are unlikely to 

feel the need to petition for a ballot initiative for another FAW law if FAW concerns have already 

been  addressed.  The  option  within  a  state  to  take  action  through  a  ballot  initiative 

(ALLOW_BALLOT) correlates with an over 18-fold (19.391) increase in the likelihood that a ballot 

will be proposed in a given year than no action taken.  

Interestingly,  the  influence  of  political  party  and  legislative  composition  was  more 

pronounced  in  the  bill  model.  Our  results  show  that  the  presence  of  a  Democratic  trifecta 

(TRIFECTA_D) is associated with a significant increase in the likelihood that a bill is proposed 

133 

 
 
and a decrease in the likelihood that a ballot is proposed. On the other hand, a Republican trifecta 

(TRIFECTA_R) correlates with a decrease in the chances of either a FAW bill or ballot, though 

the correlation is only significant in the case of a bill. As the percent of Democrats in the House 

(HOUSE%D) increases, we see an increase in the likelihood of proposing a bill or a ballot, though 

these effects are not statistically significant. These findings are in line with previous studies, which 

found  that  Democrats  tend  to  support  FAW  more  than  Republicans  (McKendree  et  al.,  2014; 

Deemer and Lobao, 2011; Czech and Borkhataria, 2001; Miele et al., 1993; Heleski et al., 2006). 

On the other hand, as the percent of Democrats in the Senate increases (SENATE%D), we see a 

small associated decrease in the likelihood of a proposed bill or ballot. Finally, the constant terms 

in both the bill and ballot outcomes are very close to zero, meaning the probability of no action is 

almost 100% more likely than a bill or ballot. Our out-of-sample model predictions (the five states 

randomly excluded from the clustering in each of our 1,000 bootstrapping repetitions) are over 

92% accurate.36  

In Table 36 we present the predicted mean probability of the three outcomes in a given 

state for a given year from 2000 to 2019.  We note that the mean predicted probability for no action 

in a state in each year is about 96%, going as low as about 49% and as high as nearly 100%. These 

predictions suggest that the FAW bills and ballots that are currently in place were unlikely to occur. 

4.2. Stage 2.1: Ballot Initiative Voting Outcomes 

The results of the OLS regression for the ballot initiative model in Stage 2.1 are given in 

Table  37  with  the  ballot  model  on  the  left  in  column  (1).  The  dependent  variable  in  the  ballot 

36 See Appendix A.4 for descriptive statistics on the accuracy of out-of-sample predictions. 

134 

 
 
 
regression is the log-odds of success of the initiative. The results of the model are contingent on a 

ballot being allowed and put to a vote.37  

In  our  ballot  model,  all  variables  are  statistically  significant.  We  note  first  that 

PEOPLE_PER_FARM, %DEMOCRAT, %WHITE, %BLACK, and %HISPANIC are all positive, 

indicating that an increase in these variables is associated with an increase in the likelihood of a 

ballot initiative's success. These effects are in line with those in Smithson et al. (2014). On the 

other hand, an increase in the percent of adults with at least a 4-year degree (EDUCATION) and 

the  percent  of  citizens  living  below  the  poverty  line  (POVERTY_RATE)  both  correlate  to  a 

statistically significant decrease on the success of a ballot initiative. The sign on POVERTY_RATE 

matches  the  models  presented  in  Smithson  et  al.  (2014).  The  negative  association  of  increased 

education on support for FAW regulations is supported by Jerolmack (2003), who suggest that 

educated individuals are more likely to view animal and human similarities and differences more 

scientifically. Additionally, the positive sign on the coefficient for HOUSEHOLD_INCOME_1000 

indicates  a  predicted  increase  in  support  for  FAW  regulations  as  median  household  income 

increases  –  this  makes  sense,  as  households  with  higher  income  levels  have  more  disposable 

income to spend on specialty products.  

An increase in the percent of Catholic citizens (%CATHOLIC) suggests a slight decrease 

in  the  likelihood  that  a  FAW  ballot  initiative  succeeds,  which  is  in  line  with  previous  findings 

(Smithson 

et 

al., 

2014;  Videras, 

2006;  Oldmixon, 

2017).  The 

signs 

of 

%EVANGELICAL_PROTESTANT  and  %MAINLINE_PROTESTANT  are  negative  and  positive, 

respectively.  Our  results  for  Evangelical  Protestants  align  with  results  in  previous  papers 

(Smithson et al., 2014; Videras, 2006; Oldmixon, 2017). While our results for Mainline Protestants 

37 K-density plots showing the distribution of the R-squared values, IMRs, and residuals for the ballot model are 
given in Appendix A.4. Additionally, descriptive statistics for the IMRs are given by year in Appendix A.4. 

135 

 
 
 
conflict with (Videras, 2006) and Smithson et al. (2014), we believe the inclusion of more data in 

our  model,  particularly  from  states  with  higher  percentages  of  Mainline  Protestants,  allows  for 

more  robust  estimates  of  the  effects  on  support  for  FAW  regulations  of  numbers  of  Mainline 

Protestants. The IMR values from Stage 1 (BALLOT_IMR) are statistically significant and positive. 

Thus, there exists a selection process that would bias results if not taken into account.38 Finally, 

the constant term has a positive and significant value, suggesting that a ballot is likely to pass once 

considered. 

Table 35: Stage 1 Legislature Action Decisions Output 

VARIABLES 

HENS_PER_1000 

HOGS_PER_1000 

COUNT_PASSED_PREV 

PREV_LAW 

ALLOW_BALLOT 

TRIFECTA_D 

TRIFECTA_R 

HOUSE%D 

SENATE%D 

CONSTANT 

(1) 
Ballot Outcome 
1.000 
(<0.001) 
0.948** 
(0.024) 
1.050 
(0.045) 
0.485 
(0.467) 
19.391*** 
(21.053) 
0.537 
(0.312) 
0.281 
(0.328) 
1.019 
(0.042) 
0.995 
(0.038) 
0.001*** 
(0.002) 

(2) 
Bill Outcome 
1.000 
(<0.001) 
0.992** 
(0.003) 
1.097*** 
(0.038) 
1.695 
(0.793) 
1.443 
(0.617) 
2.442* 
(1.233) 
0.179* 
(0.159) 
1.045 
(0.034) 
0.987 
(0.026) 
0.002*** 
(0.002) 

Observations 
Note: Robust standard error form in parentheses, *** p<0.01, ** p<0.05, * p<0.1. Standard errors are given in relative 
risk ratio format. 

980 

38 An unadjusted ballot model, similar in structure to those estimated by Smithson et al. (2014) and Videras (2006) is 
included in Appendix A.4. 

136 

 
 
 
 
 
 
 
 
 
Table 36: Mean Predicted Probability of Stage 1 Legislative Action Decisions Outcomes 
Max. Prob. 
0.99940 
0.51945 
0.09639 

Predicted Outcome  Mean Prob. 
No Action 
Bill Proposed 
Ballot Proposed 
Note: 36 FAW bills were proposed between 2000 and 2019, but only 13 made it to a vote in a state legislature; the 
“Bill Proposed” outcome does reflect this full consideration rate. 

Min. Prob. 
0.47826 
0.00021 
0.00014 

Std. Dev. 
0.06766 
0.06342 
0.01403 

0.95714 
0.03367 
0.00918 

Table 37: Stage 2.1 Ballot Initiative and Stage 2.2 Legislative Bill Voting Outcomes 

Variables 

PEOPLE_PER_FARM 

%DEMOCRAT 

DEMOCRAT 

%WHITE 

%BLACK 

%HISPANIC 

EDUCATION 

POVERTY_RATE 

HOUSEHOLD_INCOME_1000 

%CATHOLIC 

%EVANGELICAL_PROTESTANT 

%MAINLINE_PROTESTANT 

BALLOT_IMR 

BILL_IMR 

CONSTANT 

(1) 
Ballot Model 
<0.001*** 
(<0.001) 
0.014*** 
(<0.001) 

0.008*** 
(<0.001) 
0.001*** 
(<0.001) 
0.008*** 
(<0.001) 
-0.009*** 
(<0.001) 
-0.023*** 
(<0.001) 
0.002*** 
(<0.001) 
-0.003*** 
(<0.001) 
-0.034*** 
(<0.001) 
0.030*** 
(<0.001) 
0.110*** 
(<0.001) 

0.650*** 
(0.001) 

(2) 
Bill Model 
<0.001*** 
(<0.001) 

0.208*** 
(0.001) 
<0.001*** 
(<0.001) 
0.002*** 
(<0.001) 
0.002*** 
(<0.001) 
<0.001*** 
(<0.001) 
-0.006*** 
(<0.001) 
-0.001*** 
(<0.001) 
<0.001*** 
(<0.001) 
0.003*** 
(<0.001) 
0.013*** 
(<0.001) 

0.009*** 
(<0.001) 
0.678*** 
(0.004) 

Observations 
Note:  Robust  standard  errors  in  parentheses.  Coefficients  and  errors  are  obtained  from  cluster  bootstrapping.  *** 
p<0.01, ** p<0.05, * p<0.1. All variables in (1) are weighted by county population as a proportion of the total state 
population. 

1,583 

299 

137 

 
 
 
 
 
 
 
 
 
 
4.3. Stage 2.2: Legislative Bill Voting Outcomes 

The results of the LPM regression for the legislative bill model in Stage 2.2 are given in 

Table 37 on the right in column (2). The dependent variable here is the vote of the legislator, with 

a value of 1 equating to a “yes” vote and a value of 0 otherwise. The results of this model are 

contingent on a bill being proposed and put to a vote. 

An increase in the number of people per farm (PEOPLE_PER_FARM) correlates with a 

significant  increase  in  the  likelihood  of  a  legislator  voting  "yes"  on  a  FAW  bill,  though  the 

magnitude of this coefficient is small. This implies that legislators from districts with relatively 

less  representation  and  familiarity  with  agriculture  are  more  likely  to  vote  in  support  of  FAW 

regulations. Legislators who identify as Democrats are significantly more likely to vote “yes” on 

an FAW regulation than to vote “no”, as evidenced by the positive coefficient on DEMOCRAT. 

Again, this aligns with previous literature that Democrats are more likely to support FAW than 

Republicans (McKendree et al., 2014; Deemer & Lobao, 2011; Czech & Borkhataria, 2001; Miele 

et al., 1993; Heleski et al., 2006). Increases in the percent of white (%WHITE), Black (%BLACK), 

Hispanic  (%HISPANIC),  adults  with  at  least  a  4-year  college  degree  (EDUCATION),  Catholic 

(%CATHOLIC),  Evangelical  Protestant  (%EVANGELICAL_PROTESTANT),  and  Mainline 

Protestant (%MAINLINE_PROTESTANT) constituents all correlate with an increased likelihood 

that a legislator votes “yes” on a FAW bill.  However, all these effects are small. The positive signs 

on EDUCATION, %CATHOLIC, and %EVANGELICAL_PROTESTANT are opposite to the signs 

in  the  ballot  model.  Increases  in  the  percent  of  constituents  living  below  the  poverty  line 

(POVERTY_RATE)  and  the  median  household  income  (HOUSEHOLD_INCOME_1000)  in  the 

legislative district are associated with a decrease in the likelihood that a legislator votes “yes” on 

a  FAW  bill,  but  again,  these  effects  are  small.  The  negative  correlation  between  increases  in 

138 

 
 
HOUSEHOLD_INCOME_1000 and the likelihood that a legislator supports a FAW bill differs 

from  the  correlation  found  in  the  ballot  model.  The  IMR  values  from  Stage  1  are  statistically 

significant, confirming the need to include the selection process from Stage 1. Additionally, the 

constant coefficient, which is statistically significant and positive, suggests that once a bill is put 

to a vote, it is more likely to pass than fail.  

4.4. Ballot (Stage 2.1) and Bill (Stage 2.2) Model Comparisons 

There are general differences in the importance and influence of demographics between 

the  bill  and  ballot  initiative  models.  The  only  demographic  variable  that  has  a  meaningful 

magnitude associated with success of a FAW regulation in the bill model is the legislator's political 

party (DEMOCRAT). In the ballot initiative model, the demographic variables are all relatively 

similar in their overall magnitude and correlation to voting outcomes. When comparing these two 

models,  we  conclude  that  legislators  tend  to  vote  along  party  lines  when  it  comes  to  FAW 

regulations. For ballot initiatives, constituents' preferences and demographic characteristics more 

directly impact the outcome of FAW regulation, and as such, we see that the effect of demographic 

variables  are  more  influential  on  voting  outcomes.  Indeed,  Tolbert  and  Smith  (2006)  find  that 

policies that result from a popular vote are more likely to be representative of voter preferences 

than policies that result from legislative votes.  

4.5. Ballot Model Predictions 

To  understand  the  implications  of  using  a  multi-stage  model  and  accounting  for  the 

decision of whether or not a ballot goes before the people, we generate predictions for all 50 states 

for our novel IMR adjusted ballot model using data from 2019 ( 

139 

 
 
 
Figure 11). Accurate predictions are shaded in dark grey in  

140 

 
 
 
Figure  11.  We  use  the  ballot  model  to  predict  outcomes,  but  count  either  a  bill, 

administrative action, or a ballot as a pass in “reality”. Since only 24 of the 50 states have a ballot 

process, we use the outcomes from our model as a prediction of the opinions of a state's population. 

Our  predictions  for  the  non-ballot  states  should  be  considered  as  indicators  for  public  opinion 

towards FAW, which influence the likelihood of a legislature proposing a bill but do not indicate 

that  a  legislature  will  necessarily  pass  a  law.  These  results  are  also  presented  in  Table  51  in 

Appendix A.4. In reality, 24% of all 50 states and 41.7% of the 24 ballot initiative states have 

passed FAW regulations. With our ballot model we predict that 26% of all 50 states and 30.4% of 

the 24 ballot states would pass FAW regulations. The predicted pass rates given by our novel IMR 

adjusted  model  are  close  to  the  actual  percentage  of  states  with  FAW  regulations  in  place.  

Furthermore, our model is 74% and 75% accurate in predicting the presence of a FAW law in all 

50 states and the 24 ballot states, respectively. These results show that the inclusion of legislature-

level behavior in our multi-stage model is important to consider and yields predictions that are 

more consistent with actual FAW regulation outcomes.  

141 

 
 
 
 
Figure 11: Map of Ballot Model Predictions 

WA*

✓L

OR*

✓L

ID*

MT*

WY*

NV*

UT*

CA*

✓B, L

ND*

SD*

NE*

MN

IA

CO*

✓L

KS

MO*

AZ*

✓B

NM

OK*

TX

AR*

LA

AK*

HI

NH

VT

NY

PA

VA

NC

ME*
✓A

MA*

✓B

✓L

RI
CT

NJ

V
✓L

DE

MD

WI

MI*
✓L

IL*

IN

OH*
✓B

WV

✓A

KY

TN

AL

GA

MS*

SC

FL*

✓B

Powered by Bing
© GeoNames, Microsoft, TomTom

Note: States shaded in grey represent where the model prediction matches reality. States with a (P) have passed a 
FAW  regulation  as  of  2019.  The  subscripts  “A”,  “B”,  and  “L”  correspond  to  regulations  passed  through  an 
administrative regulation, a ballot initiative, and a legislative bill, respectively. The superscript “V” denotes that the 
regulation was vetoed by the governor of the state and therefore is not in effect. States with a (*) allow ballot initiatives.  

5. Implications 

Predicting  the  outcome  of  future  FAW  regulations  is  important  for  several  reasons. 

Producers, animal agriculture supply chain stakeholders, and consumers are all impacted by the 

outcomes of FAW regulations. In some cases, such as in California (CA) and Michigan (MI), FAW 

regulations also prohibit the import of products not produced in a manner that adheres to their state 

regulation(s). As such, consumers within a state with a FAW regulation can be precluded from 

purchasing these products, which negatively effects some consumers' welfare. These stakeholder 

effects  could  be  seen  in  other  instances  of  national,  provincial,  state,  and  local  government 

regulations (such as in the EU) that go beyond what is required by the overarching government or 

organization. 

142 

 
 
 
 
The state-level regulatory process of the US is likely to remain the same in the future, with 

all  states  offering  legislative  bills  and  only  24  states  allowing  ballot  initiatives  as  a  means  for 

passing  new  laws.  As  such,  our  predictions  can  assist  producers  and  industry  stakeholders  in 

gauging  the  future  of  the  regulatory  landscape  and  provide  guidance  on  whether  to  upgrade 

existing production methods to comply with anticipated mandates. Our model predicts where new 

FAW regulations are most likely to be passed; namely in the seven states that were predicted by 

our  model  to  have  a  FAW  law  in  place  in  2019  but  did  not.  These  states  are  Alaska  (AK), 

Connecticut  (CT),  Delaware  (DE),  Maryland  (MD),  Nevada  (NV),  New  Hampshire  (NH),  and 

Vermont (VT). In Table 38 and Table 39, we present the predicted percentage of the population 

within these states that would be in favor of FAW regulations, the number of egg-laying hens and 

gestating sows in the state, the relative size of the state's industry to the national total, and the 

estimated costs to the industry in each of these states to update to cage-free egg and crate-free pork 

production methods. We assume that 18.32% of egg-laying hens and 18.67% of gestating sows 

are already in cage-free and crate-free housing systems.39  

Table 38: Annual Cost to Update Cage-Free Egg Production in Seven States Predicted to 
Pass FAW Regulation 

State 

Number of 
Egg-Laying 
Hens 

Percent of Egg-
Laying Hens in 
the Nation 

Alaska* 
Connecticut 
Delaware 
Maryland 
Nevada 
New Hampshire 
Vermont 
Total 
Note: States with a (*) allow ballot initiatives. We assume a $6.95 increase in production cost per hen per year. Collar 
values have been inflated to 2022 dollars. 

8,360 
3,249,703 
3,249,703 
2,971,918 
15,964 
246,099 
173,241 
9,914,988 

0.002 
0.882 
0.882 
0.807 
0.004 
0.067 
0.047 
2.69 

Annual Cost to 
Update to 
Cage-Free 
System 
$47,477.80 
$18,455,593.01 
$18,455,593.01 
$16,878,005.68 
$90,662.15 
$1,397,636.25 
$983,863.82 
$56,308,831.72 

Percent of 
Population in 
Favor of FAW 
Regulation 
57.224 
57.659 
61.573 
59.082 
55.794 
62.032 
63.689 
N/A 

39 For a detailed explanation of how we calculated these numbers and the estimated costs to the other states without 
a FAW regulation in place, please see Appendix A.6. 

143 

 
 
 
Table 39: Annual Cost to Update to Crate-Free Pork Production in Seven States Predicted to Pass FAW Regulation 

State 

Alaska* 
Connecticut 
Delaware 
Maryland 
Nevada 
New Hampshire 
Vermont 
Total 
Note: States with a (*) allow ballot initiatives. We assume an increase in production costs of $51.53 (lower bound) and $87.77 (upper bound) per sow per year. 
Dollar values have been inflated to 2022 dollars.  

Annual Cost to 
Update to Crate-Free 
System Lower Bound 
$16,754.39 
$25,131.59 
$167,543.93 
$154,978.13 
$16,754.39 
$25,131.59 
$33,508.79 
$439,802.81 

Annual Cost to 
Update to Crate-Free 
System Upper Bound 
$28,534.83 
$42,802.24 
$285,348.25 
$263,947.13 
$28,534.83 
$42,802.24 
$57,069.65 
$749,039.16 

Percent of 
Population in Favor 
of FAW Regulation 
57.224 
57.659 
61.573 
59.082 
55.794 
62.032 
63.689 
N/A 

Percent of 
Gestating Sows in 
the Nation 
0.003 
0.005 
0.031 
0.029 
0.003 
0.005 
0.006 
0.080 

Number of 
Gestating 
Sows 
400 
600 
4,000 
3,700 
400 
600 
800 
10,500 

144 

 
 
 
As seen in Table 38, these seven states make up almost 3% of national egg production. 

Matthews  and  Sumner  (2015)  estimate  an  annual  increase  of  $6.95  per  hen  in  2022  dollars  to 

update to a cage-free production method. The total estimated annual cost to update all seven states' 

egg industries to cage-free production methods is over $56.3 million annually. Likewise, in Table 

39, these seven states make up less than 1% of national pork production. Ortega and Wolf (2018) 

estimate an annual increase of between $51.53 and $87.77 per sow in 2022 dollars to update to 

crate-free  production  methods.  The  estimated  cost  to  update  all  seven  states'  pork  industries  to 

crate-free production is between $439.8 thousand and $749.1 thousand annually. These estimated 

values do not take into account any effects on interstate commerce. In the case that future FAW 

regulations impose restrictions on what products can be sold within a state, the predicted costs to 

consumers would be added to these estimates.  

Outside of the US, the correlations we find between demographic variables and predicted 

public support for FAW regulations are likely transmutable. These associations can help inform 

policy  makers  and  industry  stakeholders  of  potential  future  FAW  regulations  in  countries  or 

organizations with similar legislative and regulatory processes to the US. Furthermore, we have 

provided a novel two-stage, three-part MESR analysis method to incorporate multiple stages of 

regulatory processes into predictions that can be applied to other legislative processes throughout 

the world. Additionally, an MESR model like the one presented here can be used to model the 

regulatory  process  for  other  types  of  agricultural  policies  and  laws,  such  as  laws  related  to 

regulating agricultural pollution.  

Further, as affluence increases in developing countries worldwide, it is likely that there will 

be an increase in these consumers' activity in food system regulations in the future. Our results 

shed light on likely regulatory outcomes in these countries. Countries in which ballot initiatives or 

145 

 
 
a  similar  process  are  allowed  can  anticipate  citizens'  voting  behavior  through  analyzing 

demographic characteristics. On the other hand, countries in which laws are only passed through 

a legislature of elected representatives can anticipate that political party-line voting will occur. 

6. Conclusion 

Nineteen state-level bills and ballot initiatives concerning farm animal welfare (FAW) have 

been adopted across 12 states. In this research, we seek to model the evolution of the state-level 

FAW  regulatory  landscape  as  a  function  of  legislature  characteristics  and  constituent 

demographics. More specifically, we utilize a two-stage model to assess (i) whether and when a 

given state considers FAW measures, and (ii) if so, the likelihood the measures are passed. We 

find that state legislature characteristics influence the likelihood of taking FAW regulatory action 

differently between ballot initiatives and legislative bills. Moreover, political party has a stronger 

influence on the outcome of votes on legislative bills, while demographics have a stronger effect 

on the outcome of votes on ballot initiatives.  Finally, we find that new FAW regulations are most 

likely to be passed in Alaska, Connecticut, Delaware, Maryland, Nevada, New Hampshire, and 

Vermont. We estimate the costs to the egg and pork industries to update to cage- and crate-free 

production systems in these states to be small. 

Of course, our analysis is not without limitations. For example, our analysis necessarily 

reduces nuanced regulation into a binary outcome. However, not all FAW regulations are alike. 

Underlying “yes” or “no” outcomes we have modeled are distinct rules that may affect markets 

differently. We are unable to account for these differences in our specifications.  Further, there are 

likely multiple factors leading to over-prediction in FAW regulation voting models, including the 

presence of social desirability bias (SDB) in public voting (Lai et al., 2021). Moreover, and perhaps 

most  importantly,  our  analysis  relies  on  an  assumption  that  the  future  –  both  in  terms  of  what 

146 

 
 
policies  are  considered  and  how  those  considerations  play  out  –  is  like  the  past.  The  US  is 

constantly experiencing changes in discourse, policy environments, and business strategies at the 

firm level. These changes may impact the probabilities that FAW legislation is proposed and how 

it is voted on, so out of sample predictions in the future using our model may not be possible. To 

the  extent  that  the  COVID-19  pandemic,  ongoing  climate  crisis,  and  other  current  issues  take 

center stage in the policy arena, these considerations could “crowd out” agricultural policies, such 

as the farm animal welfare regulations considered in this analysis. This could reduce the ongoing 

external validity of our results. Despite these concerns, our model does provide a good indication 

of FAW support within each state and can serve as a tool for policy makers, industry members, 

and other associated groups to understand the FAW regulatory landscape and provide guidance on 

whether to upgrade existing enclosures to comply with mandates on the horizon or to continue 

operating with “conventional” enclosures. 

147 

 
 
 
BIBLIOGRAPHY 

Abdulai,  Abdul  Nafeo.  2016.  “Impact  of  conservation  agriculture  technology  on  household 

welfare in Zambia.” Agricultural economics, 47(6): 729-741. 

Animal 

Agricultural 

Alliance. 

2021. 

“Issues: 

Legislation.” 

https://animalagalliance.org/issues/legislation/?fbclid=IwAR08JAohHlmE3zOpGobAOy
VBE_9OlB4jEtFAwDXau5L2iUPvjSWh64FLhEs  

Bellemare, Marc F. 2015. “A rant on estimation with binary dependent variables (technical)Marc.” 

Boehmke, Frederick J, and John W Patty. 2007. “The selection of policies for ballot initiatives: 

What voters can learn from legislative inaction.” Economics & Politics, 19(1): 97-121. 

Bourguignon,  François,  Martin  Fournier,  and  Marc  Gurgand.  2007.  “Selection  bias  corrections 
based on the multinomial logit model: Monte Carlo comparisons.” Journal of Economic 
Surveys, 21(1): 174-205. 

Bovay,  John,  and  Daniel  A  Sumner.  2019.  “Animal  welfare,  ideology,  and  political  labels: 
evidence  from  California’s  proposition  2  and  Massachusetts’s  question  3.”  Journal  of 
Agricultural and Resource Economics, 44(1835-2019-1542): 246-266. 

Bovay, John, and Julian M Alston. 2016. “GM labeling regulations by plebiscite: analysis of voting 
on Proposition 37 in California.” Journal of Agricultural and Resource Economics, 161-
188. 

Bovay,  John,  and  Julian  M  Alston.  2018.  “GMO  food  labels  in  the  United  States:  Economic 

implications of the new law.” Food Policy, 78: 14-25. 

Carter, Colin A, and K Aleks Schaefer. 2019. “Impacts of Mandatory GE Food Labeling: A Quasi-
Natural Experiment.” American Journal of Agricultural Economics, 101(1): 58-73. 

Carter, Colin A, K Aleks Schaefer, and Daniel Scheitrum. 2021. “Piecemeal Farm Regulation and 
the US Commerce Clause.” American Journal of Agricultural Economics, 103(3): 1141-
1163. 

Cornish,  Amelia,  David  Raubenheimer,  and  Paul  McGreevy.  2016.  “What  we  know  about  the 
public level of concern for farm animal welfare in food production in developed countries.” 
Animals, 6(11): 74. 

Czech,  Brian,  and  Rena  Borkhataria.  2001.  “The  relationship  of  political  party  affiliation  to 

wildlife conservation attitudes.” Politics and the Life Sciences, 3-12. 

Deemer,  Danielle  R,  and  Linda  M  Lobao.  2011.  “Public  concern  with  farm-animal  welfare: 
Religion,  politics,  and  human  disadvantage  in  the  food  sector.”  Rural  Sociology,  76(2): 
167-196. 

148 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Di Falco, Salvatore. 2014. “Adaptation to climate change in Sub-Saharan agriculture: assessing 
the  evidence  and  rethinking  the  drivers.”  European  Review  of  Agricultural  Economics, 
41(3): 405-430. 

Feindt, Peter H, Sandra Schwindenhammer, and Jale Tosun. 2020. “Politicization, depoliticization 
and policy change: A comparative theoretical perspective an agri-food policy.” Journal of 
Comparative Policy Analysis: Research and Practice, 1-17. 

Flynn,  Clifton.  2001.  “Acknowledging  the  “Zoological  connection”:  A  sociological  analysis  of 

animal cruelty.” Society & Animals, 9(1): 71-87. 

Franklin, Adrian, Bruce Tranter, and Robert White. 2001. “Explaining support for animal rights: 
to  humans,  nonhuman  animals,  and 

two  recent  approaches 

A  comparison  of 
postmodernity.” Society & Animals, 9(2): 127-144. 

Gharebaghi, Reza, MR Vaez Mahdavi, Hasan Ghasemi, Amir Dibaei, and Fatemeh Heidary. 2007. 

“Animal rights in Islam.” AATEX, 14: 63. 

Hall, Peter, Joel L Horowitz, and Bing-Yi Jing. 1995. “On blocking rules for the bootstrap with 

dependent data.” Biometrika, 82(3): 561-574. 

Heckman,  James  J.  1976.  “The  common  structure  of  statistical  models  of  truncation,  sample 
selection  and  limited  dependent  variables  and  a  simple  estimator  for  such  models.”  In 
Annals of economic and social measurement, volume 5, number 4. 475-492. NBER. 

Heleski, Camie R, Angela G Mertig, and Adroaldo J Zanella. 2006. “Stakeholder attitudes toward 

farm animal welfare.” Anthrozoös, 19(4): 290-307. 

Hopkins,  Kelsey  A,  Melissa  GS  McKendree,  K.  Aleks  Schaefer,  and  Emma  D  Rice.  2020. 
“Understanding  the  US  Publics’  Voting  on  Animal  Welfare  and  Genetically  Modified 
Organism Labeling Ballot Initiatives.” Agricultural and Applied Economics Association 
Annual Meeting, [Paper Presentation]. Virtual. 

Jaleta,  Daniel,  Boniface  Mbilinyi,  Henry  Mahoo,  and  Mulugeta  Lemenih.  2016.  “Eucalyptus 
expansion  as  relieving  and  provocative  tree  in  Eithiopia.”  Journal  of  Agriculture  and 
Ecology Research International, 1-12. 

Jerolmack,  Colin.  2003.  “Tracing  the  profile  of  animal  rights  supporters:  A  preliminary 

investigation.” Society & Animals, 11(3): 245-263. 

Kassie, Menale, Hailemariam Teklewold, Paswel Marenya, Moti Jaleta, and Olaf Erenstein. 2015. 
“Production  risks  and  food  security  under  alternative  technology  choices  in  Malawi: 
Application of a multinomial endogenous switching regression.” Journal of Agricultural 
Economics, 66(3): 640-659. 

149 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
Lai, Yufeng, Albert Boaitey, and Kota Minegishi. 2021. “Behind the veil: Social desirability bias 

and animal welfare ballot initiatives.” Food Policy, 102184. 

Langemeier, Michael. 2019. “Long-Term Trends In Pigs Per Litter.” 

Malone, Trey, and Jayson L Lusk. 2016. “Putting the Chicken Before the Egg Price: An “Ex Post” 
Analysis  of  California’s  Battery  Cage  Ban.”  Journal  of  Agricultural  and  Resource 
Economics, 518-532. 

Matsusaka, John G. 1992. “Economics of direct legislation.” The Quarterly Journal of Economics, 

107(2): 541-571. 

Matsusaka, John G. 2005. “Direct democracy works.” Journal of Economic Perspectives, 19(2): 

185-206. 

Matthews, William A, and Daniel A Sumner. 2015. “Effects of housing system on the costs of 

commercial egg production.” Poultry Science, 94(3): 552-557. 

McKendree,  Melissa  GS,  Candace  C  Croney,  and  NJ  Olynk  Widmar.  2014.  “Effects  of 
demographic factors and information sources on United States consumer perceptions of 
animal welfare.” Journal of Animal Science, 92(7): 3161-3173. 

Miele, Joseph, Leanne Tingley, Robert Kimball, and John Broida. 1993. “Personality differences 

between pro- and antivivisectionists.” Society & Animals, 1(2): 129-144. 

Mullally, Conner, and Jayson L Lusk. 2018. “The impact of farm animal housing restrictions on 
egg  prices,  consumer  welfare,  and  production  in  California.”  American  Journal  of 
Agricultural Economics, 100(3): 649-669. 

Neill, Clinton L, Rodney B Holcomb, and Jayson L Lusk. 2020. “Estimating potential beggar-thy-

neighbor effects of state labeling programs.” Agribusiness, 36(1): 3-19. 

Nibert, David A. 1994. “Animal rights and human social issues.” Society & Animals, 2(2): 115-

124. 

Norwood, Franklin Bailey, Glynn Tonsor, and Jayson L Lusk. 2019. “I will give you my vote but 
not my money: Preferences for public versus private action in addressing social issues.” 
Applied Economic Perspectives and Policy, 41(1): 96-132. 

Oldmixon, Elizabeth A. 2017. “Religious representation and animal welfare in the US senate.” 

Journal for the Scientific Study of Religion, 56(1): 162-178. 

Ortega, David L, and Christopher A Wolf. 2018. “Demand for farm animal welfare and producer 
implications: Results from a field experiment in Michigan.” Food Policy, 74: 74-81. 

150 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Paul, Andrew S, Jayson L Lusk, F Bailey Norwood, and Glynn T Tonsor. 2019. “An experiment 
on  the  vote-buy  gap  with  application  to  cage-free  eggs.”  Journal  of  Behavioral  and 
Experimental Economics, 79: 102-109. 

Peek, Charles W, Nancy J Bell, and Charlotte C Dunham. 1996. “Gender, gender ideology, and 

animal rights advocacy.” Gender & Society, 10(4): 464-478. 

Politis, Dimitris N, and Halbert White. 2004. “Automatic block-length selection for the dependent 

bootstrap.” Econometric Reviews, 23(1): 53-70. 

Politis,  Dimitris  N,  and  Joseph  P  Romano.  1994.  “The  stationary  bootstrap.”  Journal  of  the 

American Statistical Association, 89(428): 1303-1313. 

Smithson, Katie, Max Corbin, Jayson L Lusk, and F Bailey Norwood. 2014. “Predicting state-
wide  votes  on  ballot  initiatives  to  ban  battery  cages  and  gestation  crates.”  Journal  of 
Agricultural and Applied Economics, 46(1379-2016-113917): 107-124. 

Sumner,  Daniel  A.  2017.  “Economics  of  US  state  and  local  regulation  of  farm  practices,  with 
emphasis on restrictions of interstate trade.” Annual Review of Resource Economics, 9: 13-
31. 

Sumner,  Daniel  A,  J  Thomas  Rosen-Molina,  William  A  Matthews,  Joy  A  Mench,  and  Kurt  R 
Richter.  2008.  Economic  effects  of  proposed  restrictions  on  egg-laying  hen  housing  in 
California. University of California Agricultural Issues Center Davis, CA. 

Tolbert, Caroline J, and Daniel A Smith. 2006. “Representation and direct democracy in the United 

States.” Representation, 42(1): 25-44. 

United Egg Producers. 2022. “Facts and Stats.” 

Uyeki, Eugene S, and Lani J Holland. 2000. “Diffusion of pro-environment attitudes?” American 

Behavioral Scientist, 43(4): 646-662. 

Videras, Julio. 2006. “Religion and animal welfare: Evidence from voting data.” The Journal of 

Socio-Economics, 35(4): 652-659. 

Vogeler,  Colette  S.  2019a.  “Market-Based  governance  in  farm  animal  welfare:  A  comparative 
analysis of public and private policies in Germany and France.” Animals, 9(5): 267. 

Vogeler, Colette S. 2019b. “Why do farm animal welfare regulations vary between EU member 
states?  A  comparative  analysis  of  societal  and  party  political  determinants  in  France, 
Germany, Italy, Spain, and the UK.” JCMS: Journal of Common Market Studies, 57(2): 
317-335. 

151 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Vogeler, Colette S. 2020. “Politicizing farm animal welfare: A comparative study of policy change 
in the United States of America.” Journal of Comparative Policy Analysis: Research and 
Practice, 1-18. 

World Animal Protection. 2021. “Quit Stalling: Are companies making good on promises to end 

sow confinement?” 

152 

 
 
 
 
APPENDIX A.1 INTERVIEW PROCESS AND ANALYSIS 
OVERVIEW 

The  interviews  were  audio  or  video  recorded  using  Zoom  or  a  recorder  app,  then 
transcribed within 24 hours of the interview – using an automated transcription service – and then 
I further cleaned and edited the transcripts for grammar and transcription errors. The analysis of 
the interview data took part in eight main stages, an expanded version of what is suggested as the 
typical structure of qualitative analysis by Miles, Huberman, and Saldaña (2020). Retailers and 
processors were analyzed separately, though the steps were identical. These stages are illustrated 
in Figure 1. 

I  identified  recurring  themes,  concepts,  and  motivations  in  the  interviews  and  defined 
explicit major themes and concepts. Using these concepts and themes, I developed a concept map 
and the first round of subject codes following techniques and suggestions from Saldaña (2013). 
Then I reviewed the transcripts again, and refined the concept map and codes, to allow for the 
addition of new codes/themes or to condense codes/themes together. I applied each cycle of codes 
to the interviews until I was satisfied that the codes accurately captured the information needed for 
the next phase of data collection. 

153 

 
 
Read through 
of transcripts

Identification 
of major 
themes and 
concepts

Concept map 
and code 
revising

Development 
of concept 
map and 
codes

Application 
of codes to 
transcripts

Summary 
statement 
writing

Answering 
research 
questions

Compilation 
of findings 
and drawing 
conclusions

Figure 1: Stages of Qualitative Data Analysis 

For each interview transcript, a summary statement was composed for each of the codes. 
Then,  these  individual  summaries  were  condensed  into  a  single  summary  statement  for  each 
subject code to convey general interview findings. This technique is used in qualitative research 
to display data in an organized manner and see trends or gaps in information (Miles, Huberman, 
and Saldaña 2014). These summaries were used to draw conclusions and provide answers to the 
research questions and inform the design of the online survey. I intend to review all transcripts at 
least one additional time after the quantitative data is collected and analyzed; I think it is prudent 
to  also  review  the  transcripts  through  the  lens  of  quantitative  findings  and  potential 
interconnections and effects between the groups of interviewees rather than assuming each group 
operates in a vacuum. 

154 

 
 
 
APPENDIX A.2 PROCESSOR, RETAILER, AND CONSUMER 
SURVEY QUESTIONS 

Meat Processor Survey 

Q1 You are being asked to participate in a research study of meat and poultry slaughter and 
processing establishment preferences. Processing includes packing, freezing, canning, salting, 
smoking, and eviscerating meat products. You should feel free to ask the researchers any 
questions you may have. Your participation in this study will take about fifteen to twenty 
minutes. You will be asked to respond to a series of questions about how you make decisions at 
your establishment. There are 46 questions asking about your preferences for business practices 
in addition to questions asking about your establishment’s operations. I also ask some basic 
demographic questions. This project will assist researchers to benchmark awareness of food-
related issues and study events that could affect demand. You can choose to not complete the 
survey without penalty. 

 Study Title: U.S. Meat Industry Overview: Consumer Preferences, Retailer Motivations, and 
Processor Practices Researcher Title and Contact Information: Melissa G.S. McKendree, PhD, 
mckend14@msu.edu and Kelsey Hopkins, PhD Candidate, hopki190@msu.edu, 847-513-1708 
Department and Institution: Dept. of Agricultural, Food, and Resource Economics, Michigan 
State University Sponsor: USDA-Agriculture and Food Research Initiative 

The researchers will not have access to your name or your establishment's name.  At no point 
will a data file be constructed in which your personal information is linked with your responses.  
The data will only be released in summaries in which no individual’s answers can be identified.  
You have the right to say no to participating in this research. You can stop at any time after you 
have started. There will be no consequences if you stop and you will not be criticized.  You will 
not lose any benefits that you normally receive. 

 If you have concerns or questions about this study, such as scientific issues, how to do any part 
of it, or to report an injury, please contact the researcher Melissa G.S. McKendree, 202 Morrill 
Hall of Agriculture, East Lansing, MI, 48824, mckend14@msu.edu. If you have questions about 
your role and rights as a research participant, would like to obtain information or offer input, or 
would like to register a complaint about this study, you may contact, anonymously if you wish, 
the Michigan State University’s Human Research Protection Program at 517-355-2180, Fax 517-
432-4503, or e-mail irb@msu.edu or regular mail at 4000 Collins Rd, Suite 136, Lansing, MI 
48910. 

 Continuing with the survey means that you voluntarily agree to participate in this research study. 

Q2 What type of establishment do you have?   

155 

 
 
 
 
  
  
 
 
 
 
  
 Note: By definition, "processing" includes packing, freezing, canning, salting, smoking, and 
eviscerating/cutting meat products after slaughter. 

o  Slaughter establishment (without further processing)  
o  Processing establishment (without slaughter)  
o  Slaughter and processing establishment  
o  Distributor/storage facility  
o  Other (e.g., custom butcher shop, on-farm custom exempt slaughter)  
o  None of the above  

Q3 Does your establishment slaughter or process ONLY pork, seafood, fish, and/or egg 
products? 

o  Yes  
o  No  
o  Not applicable  

Q4 To the best of your knowledge, which of these are requirements for a meat or poultry product 
to be halal? Select all that apply. 

▢  A blessing was spoken over the animal at the time of slaughter  
▢  It is free of alcohol or alcohol derivatives  
▢  The animal is slaughtered by a Muslim or Person of the Book (Jewish or Christian)  
▢  The animal or meat product is imported from a Muslim majority country  
▢  It is free of pork/porcine products or derivatives  
▢  ⊗I am not sure/don’t know  

Q5 Has your establishment ever supplied halal meat or poultry products? 

o  Yes, currently supplying halal meat/poultry products  
o  Yes, supplied halal meat/poultry products in the past but not currently  
o  No, never supplied halal meat/poultry products  

Q6 Is your establishment certified for any other niche/specialty or value-added products (e.g., 
organic, kosher, antibiotic free, no added hormones, grass-fed, humanely raised, branded, etc.)? 

o  No  
o  Yes; please specify: __________________________________________________ 

Q7 In what year did your establishment begin operations? _______________________ 
Q8 In what U.S. state is your establishment?  

▼ Alabama ... I do not reside in the United States 

Q9 Approximately how many people are employed at your establishment?_______________ 
Q10 Approximately what percentage of your products are exported for international sale? 

o  0%  
o  1-25%  
o  26-50%  
o  51-75%  
o  75% or more  
o  Not applicable/unsure  

Q11 What type(s) of animal(s) do you slaughter/process at your establishment? 

156 

 
 
Slaughter 

Process 

Beef  
Veal  
Lamb or sheep  
Pork  
Turkey  
Chicken  
Goat  
Other (e.g., deer, 
bison, or exotic fowl)  
Q15 Please select the top three (3) motivations for your decision to supply halal meat or poultry 
products: 

Neither Slaughter nor 
Process 
▢   
▢   
▢   
▢   
▢   
▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   
▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   
▢   
▢   
▢   

▢  My establishment has a halal program so I can sell our products at a premium price  
▢  My establishment has a halal program so I can sell to more customers  
▢  My establishment has a halal program so I can compete with other similar businesses 
with halal programs  
▢  My establishment has a halal program so I can provide halal products for Muslim 
communities  
▢  My establishment has a halal program so I can provide halal products for people with 
diverse cultural backgrounds  

Q12 What type of halal meat or poultry does your establishment provide? (Select all that apply) 

▢  Zabiha (hand-slaughtered) halal  
▢  Machine-slaughtered halal  

Display This Question: 

If Q12 = Zabiha (hand-slaughtered) halal 

Q13 For your method of zabiha (hand-slaughtered) halal, is the cut to the animal's throat vertical 
or horizontal? 

o  Vertical  
o  Horizontal  
o  I don't know/ I am not sure  

Q14 How many years have you had a halal program at your establishment? 

o  Less than 1 year  
o  1-3 years  
o  4-5 years  
o  6-10 years  
o  11-20 years  
o  More than 20 years  
o  Q94 Please indicate to what extent each of the following INCENTIVIZED you to start 
a halal program to your establishment: 

157 

 
 
 
Neutral 

Not an 
incentive 
o 
o 
o 

An 
incentive 
o 
o 
o 

o 
o 
o 

Higher price for my products  
Access to new markets to sell my products  
Supplying niche religious products to minority 
communities  
The ability to apply for grants or financial aid to help set 
up the program  
Assistance from organizations (certifiers, universities, 
etc.) to coordinate setting up the halal program  
Competing with similar businesses that had a halal 
program  
Q93 Please indicate to what extent each of the following served as a BARRIER when you started 
a halal program to your establishment: 

o 

o 

o 

o 

o 

o 

o 

o 

o 

Costs associated with certification program involvement 
(certification fees, infrastructure, traceability technology 
materials, labor, etc.)  
Religious or racial discrimination from regulating bodies  
Backlash or displeasure from my non-Muslim customers  
Lack of Muslim laborers available near me  
Limited local market opportunities to sell my product  
Lack of knowledge for how to implement a halal program at 
my operation  
Display This Question: 

Not a 
barrier 
o 

o 
o 
o 
o 
o 

Neutral  A 

barrier 
o 

o 
o 
o 
o 
o 

o 

o 
o 
o 
o 
o 

If Q2 = Slaughter establishment (without further processing) 
Or Q2 = Slaughter and processing establishment 
Or Q2 = Other (e.g., custom butcher shop, on-farm custom exempt slaughter) 
Q16 Approximately what percent of your establishment’s total slaughter is halal? 

o  Less than 10%  
o  11-25%  
o  26 - 50%  
o  51-75%  
o  More than 75%  
o  Not applicable  
Display This Question: 

If Q2 = Processing establishment (without slaughter) 
Or Q2 = Slaughter and processing establishment 
Or Q2 = Other (e.g., custom butcher shop, on-farm custom exempt slaughter) 
Q17 Approximately what percent of your establishment’s total processing is halal? 

o  Less than 10%  
o  11-25%  
o  26-50%  
o  51-75%  
o  More than 75%  

158 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
o  Not applicable  

Q18 Approximately what percentage of your halal products are exported for international sale? 

o  0%  
o  1-25%  
o  26-50%  
o  51-75%  
o  75% or more  
o  Not applicable/unsure  

Display This Question: 
If Q18 = 0% 

Q19 Why does your establishment not export halal meat or poultry products for international 
sale? Select all that apply. 

▢  Our halal certification program does not support exports  
▢  I do not know how to coordinate exporting our halal products  
▢  I do not have excess halal product available for export  
▢  I do not want to export our halal products  
▢  Other; please specify: __________________________________________________ 

Q20 Is your operation certified by a third party to provide halal meat or poultry products? 

o  Yes  
o  No  
o  I am not sure  
Display This Question: 
If Q20 = Yes 

Q21 Which organization(s) provides your halal certification? Select all that apply. 

Islamic Services of America (ISA)  
Halal Food Standards Alliance of America (HFSAA)  
Halal Monitoring Services (HMS)  
Halal Transactions of Omaha (HTO)  
Other; please specify:  
Display This Question: 

Current 
certifier 
▢   
▢   
▢   
▢   
▢   

Past 
certifier 
▢   
▢   
▢   
▢   
▢   

Never been 
my certifier 
▢   
▢   
▢   
▢   
▢   

If Q2 = Processing establishment (without slaughter) 
Or Q2 = Slaughter and processing establishment 
Or Q2 = Other (e.g., custom butcher shop, on-farm custom exempt slaughter) 
Q22 Approximately what percent of your establishment’s total processing was halal?  

o  Less than 10%  
o  11-25%  
o  26-50%  
o  51-75%  
o  More than 75%  
o  Not applicable  
Display This Question: 

If Q2 = Slaughter establishment (without further processing) 
Or Q2 = Slaughter and processing establishment 

159 

 
 
 
Or Q2 = Other (e.g., custom butcher shop, on-farm custom exempt slaughter) 
Q23 Approximately what percent of your establishment’s total slaughter was halal?  

o  Less than 10%  
o  11-25%  
o  26-50%  
o  51-75%  
o  More than 75%  
o  Not applicable  

Q24 Approximately what percentage of your halal meat or poultry products were exported for 
international sale? 

o  0%  
o  1-25%  
o  26-50%  
o  51-75%  
o  75% or more  
o  Not applicable/unsure  

Q25 What type of halal meat or poultry did your establishment provide? (Select all that apply) 

▢  Zabiha (hand-slaughtered) halal  
▢  Machine-slaughtered halal  

Display This Question: 

If Q25 = Zabiha (hand-slaughtered) halal 

Q26 For your method of zabiha (hand-slaughtered) halal, was the cut to the animal's throat 
vertical or horizontal? 
o  Vertical  
o  Horizontal  
o  I don't know/ I am not sure  

Q27 Please select the top three (3) motivations for your decision to supply halal meat or poultry 
products in the past: 

▢  My establishment had a halal program so I could sell our products at a premium price  
▢  My establishment had a halal program so I could sell to more customers  
▢  My establishment had a halal program so I could compete with other similar businesses 
with halal programs  
▢  My establishment had a halal program so I could provide halal products for Muslim 
communities  
▢  My establishment had a halal program so I could provide halal products for people with 
diverse cultural backgrounds  

Q28 How many years did you have a halal program at your establishment? 

o  Less than 1 year  
o  1-3 years  
o  4-5 years  
o  6-10 years  
o  11-20 years  
o  More than 20 years  

Q29 Was your operation certified by a third party to provide halal meat or poultry products? 

o  Yes  
o  No  

160 

 
 
o  I am not sure  
Display This Question: 
If Q29 = Yes 

Q30 Why did you end your halal certification? (Check all that apply) 

▢  Costs became prohibitive  
▢  Wanted to offer different products  
▢  Standards were too strict/hard to meet  
▢  Poor working relationship with certifier  
▢  Insufficient demand for halal products  
▢  Other; please specify: __________________________________________________ 

Display This Question: 
If Q29 = Yes 

Q31 Which organization(s) was your halal certification from? Select all that apply. 

▢  Islamic Services of America (ISA)  
▢  Halal Food Standards Alliance of America (HFSAA)  
▢  Halal Monitoring Services (HMS)  
▢  Halal Transactions of Omaha (HTO)  
▢  Other; please specify: __________________________________________________ 

Q32 Have you ever considered adding a halal program to your establishment again? 

o  Yes  
o  No  

Q33 Please indicate how likely each of the following are to INCENTIVIZE you to add 
a halal program back to your establishment: 

Unlikely  Neutral 

o 
o 
o 

Higher price for my products  
Access to new markets to sell my products  
Supplying niche religious products to minority 
communities  
The ability to apply for grants or financial aid to help set up 
the program  
Assistance from organizations (certifiers, universities, etc.) 
to coordinate setting up the halal program  
Competing with similar businesses that had a halal program  
Q34 Please indicate how likely each of the following serve as a BARRIER in your decision to 
add a halal program back to your establishment: 

o 
o 
o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

Likely 
o 
o 
o 

161 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Unlikely  Neutral 

o 

Costs associated with certification program involvement 
(certification fees, infrastructure, traceability technology 
materials, labor, etc.)  
Religious or racial discrimination from regulating bodies  
Backlash or displeasure from my non-Muslim customers  
Lack of Muslim laborers available near me  
Limited local market opportunities to sell my product  
Lack of knowledge for how to implement a halal program 
at my operation  
Q35 Have you ever considered adding a halal program to your establishment? 

o 
o 
o 
o 
o 

o 

o 
o 
o 
o 
o 

Likely 
o 

o 
o 
o 
o 
o 

o  Yes  
o  No  
o  I do/did not know what halal is  

Q36 Are you interested in learning more about halal meat production opportunities for your 
establishment? 
o  Yes  
o  No  

Q37 Please indicate how likely each of the following are to INCENTIVIZE you to add 
a halal program to your establishment: 

Unlikely  Neutral 

Likely 
o 
o 
o 
o 

Higher price for my products  
Access to new markets to sell my products  
Supplying niche religious products to minority communities  
The ability to apply for grants or financial aid to help set up 
the program  
Assistance from organizations (certifiers, universities, etc.) 
to coordinate setting up the halal program  
Competing with similar businesses that had a halal program  
Q38 Please indicate how likely each of the following serve as a BARRIER in your decision to 
add a halal program to your establishment: 

o 
o 
o 
o 

o 
o 
o 
o 

o 

o 

o 

o 

o 

o 

Unlikely  Neutral 

Likely 
o 

o 

o 

Costs associated with certification program involvement 
(certification fees, infrastructure, traceability technology 
materials, labor, etc.)  
Religious or racial discrimination from regulating bodies  
Backlash or displeasure from my non-Muslim customers  
Lack of Muslim laborers available near me  
Limited local market opportunities to sell my product  
Lack of knowledge for how to implement a halal program at 
my operation  
Q39 There is not a NHMC program in the U.S. I am interested in your opinions to help design a 
future national U.S. meat and poultry halal certification program. If you do not currently have 
a halal program, imagine that you are considering adding a halal program to your establishment. 

o 
o 
o 
o 
o 

o 
o 
o 
o 
o 

o 
o 
o 
o 
o 

 In the following section of the survey, you will be presented seven scenarios. Please consider the 

162 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
three factors presented, and indicate which one factor is the least important and which one is the 
most important to you when designing a national U.S. halal meat and poultry certification 
program. Please select one factor as least important AND one factor as most important in each 
question. 

 The questions look similar but contain different comparisons of factors. Please treat each 
question individually. 

 To help, I have given an example below with ice cream, where flavor is the most important 
factor and price is the least important factor in your decision to buy ice cream.  
EXAMPLE  
 Of the following three factors, which one is the least important and which one is the most 
important in your decision to buy ice cream?  

Q40 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 

Most Important 
 (Check only one) 

o 

o 

o 

⊗Which group(s) will enforce the program (e.g., 
government, religious organization, private non-
religious organization) 
⊗What will be required to be certified (e.g., 
products, retailers, slaughter and processing 
establishments) 
⊗What halal standards will be required (e.g., 
hand versus machine slaughter, stunned or not 
stunned) 

o 

o 

o 

Q41 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 
o 

o 

o 

⊗Which group(s) will enforce the program (e.g., 
government, religious organization, private non-
religious organization) 
⊗Inspection process (e.g., frequency, random or 
scheduled) 
⊗Benefits associated with certification program 
involvement (e.g., access to new markets, price 
premiums) 

Most Important 
 (Check only one) 
o 

o 

o 

163 

 
 
  
  
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q42 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 

Most Important 
 (Check only one) 

o 

o 

o 

⊗Costs associated with certification program 
involvement (e.g., certification fees, infrastructure, 
traceability technology materials, labor) 
⊗Which group(s) will enforce the program (e.g., 
government, religious organization, private non-
religious organization) 
⊗What information will be passed on to my 
customers and how they access it (e.g., only 
available through the Freedom of Information Act 
(FOIA) versus accessible online) 

o 

o 

o 

Q43 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 

Most Important 
 (Check only one) 

o 

o 

o 

⊗What information will be passed on to my 
customers and how they access it (e.g., only 
available through the Freedom of Information Act 
(FOIA) versus accessible online) 
⊗Inspection process (e.g., frequency, random or 
scheduled) 
⊗What halal standards will be required (e.g., 
hand versus machine slaughter, stunned or not 
stunned) 

o 

o 

o 

Q44 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 
o 

o 

o 

⊗Benefits associated with certification program 
involvement (e.g., access to new markets, price 
premiums) 
⊗What halal standards will be required (e.g., 
hand versus machine slaughter, stunned or not 
stunned) 
⊗Costs associated with certification program 
involvement (e.g., certification fees, infrastructure, 
traceability technology materials, labor) 

Most Important 
 (Check only one) 
o 

o 

o 

164 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q45 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 

Most Important 
 (Check only one) 

o 

o 

o 

⊗Costs associated with certification program 
(e.g., facility modifications, traceability equipment, 
certification fees, labor) 
⊗What will be required to be certified (e.g., 
products, retailers, slaughter and processing 
establishments) 
⊗Inspection process (e.g., frequency, random or 
scheduled) 

o 

o 

o 

Q46 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 

Most Important 
 (Check only one) 

⊗What information will be passed on to my 
customers and how they access it (e.g., only 
available through the Freedom of Information Act 
(FOIA) versus accessible online) 
⊗Benefits associated with certification program 
involvement (e.g., access to new markets, price 
premiums) 
⊗What will be required to be certified (e.g., 
products, retailers, slaughter and processing 
establishments) 

165 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q47 If you were to use a national halal certification at your establishment, would you want it to 
be internationally accredited or recognized for export? 

o  Yes  
o  No  
o  Maybe  
o  I would not want to use a national halal certification at my establishment  

Q48 Should individual products or supply chain members be required to have a national halal 
certification to ensure authentic halal meat and poultry products? 

o  Individual products  
o  Supply chain members (e.g., processor, wholesaler, retailer)  
o  Both individual products and supply chain members  

Display This Question: 

If Q48 = Supply chain members (e.g., processor, wholesaler, retailer) 
Or Q48 = Both individual products and supply chain members 

Q49 Which members of the supply chain should be required to have a national halal 
certification? Select all that apply. 
▢  Slaughter establishments  
▢  Processing establishments  
▢  Distributors or transportation services  
▢  Retailers and wholesalers  
▢  Restaurants and food service  
▢  Other; please specify: __________________________________________________ 

Q50 Under a national halal certification, how should a certified business be inspected? 

o  Pre-scheduled inspections  
o  Random/surprise inspections  
o  A mixture of pre-scheduled and random inspections  
o  Other; please specify: __________________________________________________ 
Q51 Which benefits of a national halal certification would be most important to your business? 
Select up to three. 

▢  Access to new domestic (U.S.) markets  
▢  Access to new export (international) markets  
▢  Ability to charge a higher price for my products  
▢  Increased consumer trust  
▢  Ease of identifying a certified product  
▢  Ease of communicating product attributes  
▢  Other; please specify: __________________________________________________ 
▢  ⊗None of the above  

Q52 Which costs of a national halal certification would be most important to your business? 
Select up to three. 

▢  Reoccurring certification fees  
▢  Cost of establishment modifications  
▢  Increased labor hours needed  
▢  Cost of traceability and/or verification equipment  
▢  Other; please specify: __________________________________________________ 

166 

 
 
▢  ⊗None of the above  

Q53 In your opinion, what standards should be included in a national halal certification for 
meat and poultry? Select all that apply. 

▢  Zabiha (hand-slaughter)  
▢  Machine-slaughter  
▢  Slaughterers of Muslim faith  
▢  Slaughterers of Christian or Jewish faith  
▢  Individual spoken blessings  
▢  Animal(s) not stunned  
▢  Animal(s) face Mecca at time of slaughter  
▢  GMO-free  
▢  Other; please specify: __________________________________________________ 
▢  ⊗I don't know  

Q54 Please indicate which parties should have access to the names of establishment(s) at which 
a halal meat or poultry product was slaughtered and/or processed using a national halal 
certification. Select all that apply. 

▢  General public  
▢  Processors/slaughterers  
▢  Wholesaler/distributor  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Q55 Please indicate which parties should have access to a national list of halal certified 
meat/poultry establishments certified with a national halal certification. Select all that apply. 

▢  General public  
▢  Processors/slaughterers  
▢  Wholesalers/distributors  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above 

Q56 Please indicate which parties should have access to the name(s) of enforcement agencies 
that certify halal establishments with a national halal certification. Select all that apply. 

▢  General public  
▢  Processors/slaughterers  
▢  Wholesalers/distributors  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Q57 Please indicate which parties should have access to information regarding the halal 
standards used in slaughter and/or processing under a national halal certification. Select all that 
apply. 

▢  General public  
▢  Processors/slaughterers  

167 

 
 
▢  Wholesalers/distributors  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Display This Question: 

If Q54 = General public 
Or Q55 = General public 
Or Q56 = General public 
Or Q57 = General public 

Q58 How should the general public be able to access information related to a national halal 
certification program for meat/poultry? Select all that apply. 
▢  Online (e.g., company website, online database)  
▢  Using a QR code/cell phone app  
▢  Freedom of Information Act (FOIA) request  
▢  Other; please specify: __________________________________________________ 

Q59 Who should set standards for a new national halal certification for meat and poultry? 
Select all that apply.  
 Note: The U.S. and state governments are not legally allowed to define standards related to 
religious products. 

▢  Non-government organizations  
▢  Religious organizations  
▢  Certifier-led organizations  
▢  Producer-led organizations  
▢  Wholesalers/distributors  
▢  Slaughterers/processors  
▢  Retailers/restaurants  
▢  Other; please specify: __________________________________________________ 
Q60 Who should enforce a new national halal certification for meat and poultry? Select all that 
apply. 

▢  U.S. government organization (e.g., the USDA)  
▢  State government organization (e.g., state department of agriculture)  
▢  Non-government organizations  
▢  Religious organizations  
▢  Certifier-led organizations  
▢  Producer-led organizations  
▢  Slaughterers/processors  
▢  Wholesalers/distributors  
▢  Retailers/restaurants  
▢  Other; please specify: __________________________________________________ 

Q61 Please select the degree to which you agree or disagree with the following statements: 

168 

 
 
Agree  Neutral  Disagree 
o 

o 

o 

o 

o 

o 

o 

o 

o 

Halal meat or poultry tastes better than non-halal meat or 
poultry  
All halal meat or poultry slaughter or processing 
establishments must be halal certified  
Halal establishments that are certified have a stronger 
reputation than halal establishments that are not certified  
Halal meat/poultry is more sanitary than non-halal meat/ 
poultry  
The halal slaughter process is more humane for the animal  
Establishments that are not halal certified cannot be trusted to 
supply authentic halal products  
Halal meat and poultry is higher quality than non-halal meat 
and poultry  
Halal meat and poultry is healthier than non-halal meat and 
poultry  
If an establishment has a good reputation for supplying halal 
meat and poultry products, it does not need to be certified as 
halal  
Q62 What is your role or job at your establishment? __________________________________ 
Q63 What is your current age? 

o 
o 

o 
o 

o 
o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o  18-24 years old  
o  25-34 years old  
o  35-44 years old  
o  45-54 years old  
o  55-64 years old  
o  65-74 years old  
o  75 years or older  
o  Prefer not to disclose  
Q64 Were you born in the U.S.? 

o  Yes  
o  No, I was born in this country: __________________________________________ 
o  Prefer not to disclose  

Q65 Were your parents and grandparents born in the U.S.? 

o  Yes  
o  No; they were born in this/these countries:____________________________________ 
o  Prefer not to disclose  

Display This Question: 

If Q64 = No, I was born in this country: 
Or Were you born in the U.S.? Text Response Is Not Empty 

Q66 How long have you lived in the U.S.? 

o  0-5 years  
o  6-10 years  
o  11-15 years  
o  16-20 years  
o  21-25 years  

169 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
o  Over 25 years  
o  Prefer not to disclose  

Q67 What is your gender? 

o  Male  
o  Female  
o  Prefer to self-describe: __________________________________________________ 
o  Prefer not to disclose  

Q68 What is the highest level of education you have completed? 

o  Less than High School  
o  High school graduate or GED  
o  Some college  
o  2-year degree (Associates)  
o  4-year degree (BA, BS)  
o  Professional degree (JD, MD, PhD, etc.)  
o  Prefer not to disclose  

Q69 Are you of Hispanic, Latino, or Spanish origin? 

o  No  
o  Yes; please specify: __________________________________________________ 
o  Prefer not to disclose  

Q70 What is your race? Select all that apply. 

▢  White  
▢  Black or African American  
▢  Native American or Alaska Native  
▢  Native Hawaiian or Pacific Islander  
▢  Asian  
▢  Other; please specify: __________________________________________________ 
▢  ⊗Prefer not to disclose  

Q71 Which political party do you most identify with? 

o  Democrat  
o  Republican  
o  I am an independent  
o  Other; please specify: __________________________________________________ 
o  Prefer to not disclose  

Q72 Which best describes the area in which you live? 

o  Rural  
o  Suburban  
o  Urban  
o  Prefer not to disclose  

Q73 Do you consider yourself to be religious? 

o  No  
o  Yes, I am: __________________________________________________ 
o  Prefer not to disclose  

Display This Question: 

If Q73 = No 

Q74 Have you followed a religion in the past even if you do not do so now? 

o  No  

170 

 
 
o  Yes, I was: __________________________________________________ 
o  Prefer not to disclose  

Q75 Do you have any final thoughts or comments you wish to share about using/not using a 
halal meat program at your establishment? 

________________________________________________________________ 
________________________________________________________________ 

171 

 
 
 
 
 
 
 
Meat Retailer Survey 

Q1 You are being asked to participate in a research study of meat and poultry retailers. You 
should feel free to ask the researchers any questions you may have. Your participation in this 
study will take about fifteen to twenty minutes. You will be asked to respond to a series of 
questions about how you make decisions at your retail establishment. There are questions asking 
about your preferences for business practices in addition to questions asking about your store’s 
operations. I also ask some basic demographic questions. This project will assist researchers to 
benchmark awareness of food-related issues and study events that could affect demand. You can 
choose to not complete the survey without penalty. 

Study Title: U.S. Meat Industry Overview: Consumer Preferences, Retailer Motivations, and 
Processor Practices  
Researcher Title and Contact Information: Melissa G.S. McKendree, PhD, mckend14@msu.edu 
and Kelsey Hopkins, PhD Candidate, hopki190@msu.edu, 847-513-1708  
Department and Institution: Dept. of Agricultural, Food, and Resource Economics, Michigan 
State University  
Sponsor: USDA-Agriculture and Food Research Initiative 

At no point will a data file be constructed in which your personal information is linked with your 
responses.  The data will only be released in summaries in which no individual’s answers can be 
identified.  You have the right to say no to participating in this research. You can stop at any 
time after you have started. There will be no consequences if you stop, and you will not be 
criticized.  You will not lose any benefits that you normally receive. 

 If you have concerns or questions about this study, such as scientific issues, how to do any part 
of it, or to report an injury, please contact the researcher Melissa G.S. McKendree, 202 Morrill 
Hall of Agriculture, East Lansing, MI, 48824, mckend14@msu.edu. If you have questions about 
your role and rights as a research participant, would like to obtain information or offer input, or 
would like to register a complaint about this study, you may contact, anonymously if you wish, 
the Michigan State University’s Human Research Protection Program at 517-355-2180, Fax 517-
432-4503, or e-mail irb@msu.edu or regular mail at 4000 Collins Rd, Suite 136, Lansing, MI 
48910. 

 Continuing with the survey means that you voluntarily agree to participate in this research 
study.  

Please certify that you are over 18 years of age and agree to voluntarily participate in this survey. 

o  I am over 18 and agree to participate  
o  I am not over 18 or do not agree to participate  

172 

 
 
 
 
  
 
 
 
 
 
Q144 Please ensure you are carefully reading through the survey questions and making 
thoughtful selections in order to qualify for the $25 survey incentive. Any nonsense answers, 
quality issues, or speeding will be disqualified without incentive. 

o  Agree  
o  Disagree  

Skip To: End of Survey If Q144 = Disagree 
Q2 Do you represent (work for or own) a retail store that sells meat or poultry products? 

o  No  
o  Yes 

Q3 What is your role or job at your store?____________________________________________ 
Q4 Which of the following best describes your store? 
o  Supermarket (e.g., Walmart, Target, Meijer)  
o  Club membership store (e.g., Costco, Sam's Club)  
o  Grocery store (e.g., independent small local or regional store)  
o  Convenience store  
o  Butcher shop/deli  
o  Other, please specify: __________________________________________________ 
o  None of the above  

Q5 To the best of your knowledge, which of these are requirements for a meat or poultry product 
to be halal? Select all that apply. 

▢  A blessing was spoken over the animal at the time of slaughter  
▢  It is free of alcohol or alcohol derivatives  
▢  The animal is slaughtered by a Muslim or Person of the Book (Jewish or Christian)  
▢  The animal or meat product is imported from a Muslim majority country  
▢  It is free of pork/porcine products or derivatives  
▢  ⊗I am not sure/don’t know  

Q6 Has your retail store ever sold halal meat or poultry products? 

o  Yes, currently selling halal meat/poultry products  
o  Yes, sold halal meat/poultry products in the past but not currently  
o  No, never sold halal meat/poultry products  
o  I am unsure/don't know  

Q7 Does your retail store sell any other certified niche/specialty or value-added meat or poultry 
products besides halal (e.g., organic, kosher, antibiotic free, no added hormones, grass-fed, 
humanely raised, branded)? 

o  No, no other specialty meat or poultry products  
o  Yes, I sell other specialty meat or poultry products. Please specify: __________________ 

Q8 The following questions ask for basic information about your retail store. If you are 
responsible for multiple retail stores, please answer the questions based on your primary or 
flagship retail location. 
Q9 In what year did your store begin operations? Please enter a full year, such as 1990. _______ 
Q10 In what U.S. state is your store?  

▼ Alabama ... My store is not in the United States 

173 

 
 
 
Q11 In what type of area is your store located? 

o  Rural or countryside  
o  Suburban or small-mid size city  
o  Urban or large city  
o  Prefer not to disclose  

Q12 Approximately how many people are employed at your store full time or full time 
equivalent? Note: For example, if an employee works 20 hours per week, they are considered 0.5 
full time equivalent. 
_________________________________________________ 
Q13 The following questions ask you about the meat and poultry products you sell in your 
store. 
Q14 Where or how do you sell meat or poultry products at your store? Select all that apply. 

o  Products are sold prepackaged on shelves (shelf stable products)  
o  Products are sold prepackaged in a refrigerated or frozen display case (case ready)  
o  Products are sold prepackaged in a refrigerated or frozen display case (not case ready/cut 

in store) Products are sold at a service deli counter  

o  Products are sold at a service butcher counter  
o  Other location or method. Please specify: ______________________________________ 

174 

 
 
I do not 
sell 
products 
from this 
species 
at my 
store 

▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   

Q15 What species and types of meat or poultry products do you sell at your store? 
Frozen 
fully 
cooked 

Deli 
meats/
poultry 

Canned, 
smoked, or 
cured 
meat/poultry 
products 

Ready to 
eat 
products 
(e.g., 
snack 
sticks, 
jerky) 

Frozen 
whole 
muscle cuts 
or ground 
(e.g., 
chicken 
breasts, 
roasts, 
ground 
turkey) 

Fresh 
whole 
muscle 
cuts or 
ground 
(e.g., 
chicken 
breasts, 
roasts, 
ground 
turkey) 
▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   

Beef  
Veal  
Lamb or 
sheep  
Pork  
Turkey  
Chicken  
Goat  
Other 
(e.g., 
deer, 
bison, 
camel, 
or exotic 
fowl)  
Q16 How many years have halal meat/poultry products been sold at your store? 

▢   
▢   
▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   

▢   
▢   
▢   
▢   
▢   

o  Less than 1 year  
o  1-3 years  
o  4-5 years  
o  6-10 years  
o  11-20 years  
o  More than 20 years  

Q17 Approximately what percent of your store’s total meat/poultry sales are halal? 

o  Less than 10%  
o  11-25%  
o  26 - 50%  
o  51-75%  
o  More than 75%  

Q18 What standard of halal meat or poultry does your store sell? Select all that apply. 

▢  Zabiha (hand-slaughtered) halal  
▢  Machine-slaughtered halal  
▢  ⊗I am not sure/don't know  

Display This Question: 

175 

 
 
 
If Q18 = Zabiha (hand-slaughtered) halal 

Q19 For your zabiha (hand-slaughtered) halal products, is the cut to the animal's throat vertical 
or horizontal? 

o  Vertical (up and down)  
o  Horizontal (ear to ear)  
o  I don't know/ I am not sure  

Q20 What halal meat or poultry products do you sell at your store? Select all that apply. 
▢  Fresh whole muscle cuts or ground (e.g., chicken breasts, roasts, ground turkey)  
▢  Frozen whole muscle cuts or ground (e.g., chicken breasts, roasts, ground turkey)  
▢  Frozen fully cooked products  
▢  Ready to eat shelf stable products (e.g., snack sticks, jerky)  
▢  Deli meats/poultry  
▢  Canned, smoked, or cured meats/poultry  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q20 = <strong>Fresh</strong> whole muscle cuts or ground (e.g., chicken breasts, 

roasts, ground turkey) 
Q21 What types of halal fresh whole muscle cuts do you sell at your store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q20 = <strong>Frozen</strong> whole muscle cuts or ground (e.g., chicken breasts, 

roasts, ground turkey) 
Q22 What types of halal frozen whole muscle cuts do you sell at your store? Select all that 
apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q20 = Frozen fully cooked products 

Q23 What types of halal frozen fully cooked products do you sell at your store? Select all that 
apply. 

▢  Beef  
▢  Veal  

176 

 
 
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q20 = Ready to eat shelf stable products (e.g., snack sticks, jerky) 

Q24 What types of halal ready to eat (shelf stable) products do you sell at your store? Select all 
that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q20 = Deli meats/poultry 

Q25 What types of halal deli products do you sell at your store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q20 = Canned, smoked, or cured meats/poultry 

Q26 What types of halal canned, smoked, or cured products do you sell at your store? Select all 
that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If What&nbsp; halal meat or poultry products do you sell at your store?&nbsp;Select all 

that apply.  Text Response Is Not Empty 
Q27 What types of halal ${Q20/ChoiceTextEntryValue/7} products do you sell at your 
store? Select all that apply. 

▢  Beef  

177 

 
 
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Q28 Would you like to sell other halal meat or poultry products at your store that you do not 
currently offer? 

o  Yes. Please specify other halal meat or poultry products you would like to sell: ________ 
o  No  

Q29 Please select the top three (3) motivations for your decision to sell halal meat/poultry 
products: 

▢  My store offers halal meat/poultry products so I can sell these products at a higher retail 
margin  
▢  My store offers halal meat/poultry products so I can attract more customers  
▢  My store offers halal meat/poultry products so I can compete with other similar 
businesses that offer halal meat/poultry products  
▢  My store offers halal meat/poultry products so I can provide halal meat/poultry products 
for Muslim communities  
▢  My store offers halal meat/poultry products so I can provide halal meat/poultry products 
for people with diverse cultural backgrounds  

Q30 Are any of the halal products your establishment provides certified by a third party? (e.g., 
individual products have a stamp or label that says "halal") 

o  Yes  
o  No  
o  I am not sure  

Q31 Is your store certified by a third party to provide halal meat or poultry products? 

o  Yes  
o  No  
o  I am not sure  
Display This Question: 
If Q31 = Yes 

Q32 Which organization(s) provides your store's halal certification? Select all that apply. 

Islamic Services of America (ISA)  
Halal Food Standards Alliance of America (HFSAA)  
Halal Monitoring Services (HMS)  
Halal Transactions of Omaha (HTO)  
The Islamic Society of the Washington Area (ISWA)  
Other. Please specify:  
Display This Question: 
If Q30 = Yes 

Current 
certifier 
▢   
▢   
▢   
▢   
▢   
▢   

Past 
certifier 
▢   
▢   
▢   
▢   
▢   
▢   

Never been 
my certifier 
▢   
▢   
▢   
▢   
▢   
▢   

Q33 Which organization(s) provides your products' halal certification? Select all that apply. 

178 

 
 
 
Current 
certifier 

Past 
certifier 

Never 
been my 
certifier 

Unsure if 
they have 
ever been 
my certifier 
▢   
▢   

▢   
▢   

▢   
▢   

Islamic Services of America (ISA)  
Halal Food Standards Alliance of America 
(HFSAA)  
Halal Monitoring Services (HMS)  
Halal Transactions of Omaha (HTO)  
The Islamic Society of the Washington Area 
(ISWA)  
Other. Please specify:  
▢   
Q34 Please indicate to what extent each of the following MOTIVATED your store to start a 
halal program: 

▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   
▢   

▢   
▢   

▢   

▢   

▢   

Neutral 

Not a 
motivation 
o 
o 
o 

A 
motivation 
o 
o 
o 

Higher retail margin for halal products  
Access to new customers to sell halal products  
Supplying niche religious products to minority 
communities  
The ability to compete with other similar businesses 
that offer halal meat/poultry products  
Assistance from organizations (certifiers, producer 
groups, etc.) to coordinate finding a supplier  
Q35 Please indicate to what extent each of the following was a CHALLENGE when you started 
a halal program to your store: 

o 
o 
o 

o 

o 

o 

o 

o 

o 

Costs associated with certification program involvement 
(certification fees, infrastructure, traceability technology 
materials, labor, etc.)  
Religious or racial discrimination from regulating bodies  
Backlash or displeasure from my non-Muslim customers  
Limited local customer base to sell halal products  
Lack of knowledge for how to sell halal products at my 
store  
Display This Question: 

Not a 
challenge 
o 

o 
o 
o 
o 

Neutral 

o 

o 
o 
o 
o 

A 
challenge 
o 

o 
o 
o 
o 

If Q35 = Costs associated with certification program involvement (certification fees, 

infrastructure, traceability technology materials, labor, etc.) [ Neutral ] 

Or Q35 = Costs associated with certification program involvement (certification fees, 

infrastructure, traceability technology materials, labor, etc.) [ A challenge ] 
Q36 In the previous question, you indicated that costs were a challenge or barrier when 
beginning your halal program. Which of the following costs were the most challenging? Select 
all that apply. 

▢  Reoccurring certification fees  

179 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
▢  One-time infrastructure costs (e.g., additional refrigerators or shelves)  
▢  Increased labor  
▢  Traceability technology costs  
▢  Other. Please specify: __________________________________________________ 

Q37 How many years did you sell halal meat/poultry products at your store? 

o  Less than 1 year  
o  1-3 years  
o  4-5 years  
o  6-10 years  
o  11-20 years  
o  More than 20 years  

Q38 Approximately what percent of your store’s total meat/poultry sales were halal?  

o  Less than 10%  
o  11-25%  
o  26-50%  
o  51-75%  
o  More than 75%  

Q39 What standard of halal meat or poultry products did your store sell? Select all that apply. 

▢  Zabiha (hand-slaughtered) halal  
▢  Machine-slaughtered halal  
▢  ⊗I don't know/am not sure  

Display This Question: 

If Q39 = Zabiha (hand-slaughtered) halal 

Q40 For your zabiha (hand-slaughtered) halal products, was the cut to the animal's throat 
vertical or horizontal? 

o  Vertical (up and down)  
o  Horizontal (ear to ear)  
o  I don't know/ I am not sure  

Q41 What halal meat or poultry products did you sell at your store? Select all that apply. 
▢  Fresh whole muscle cuts or ground (e.g., chicken breasts, roasts, ground turkey)  
▢  Frozen whole muscle cuts or ground (e.g., chicken breasts, roasts, ground turkey)  
▢  Frozen fully cooked products  
▢  Ready to eat shelf stable products (e.g., snack sticks, jerky)  
▢  Deli meats/poultry  
▢  Canned, smoked, or cured meats/poultry  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q41 = <strong>Fresh</strong> whole muscle cuts or ground (e.g., chicken breasts, 

roasts, ground turkey) 
Q42 What types of halal fresh whole muscle cuts did you sell at your store? Select all that 
apply. 

▢  Beef  
▢  Veal  
▢  Lamb  

180 

 
 
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q41 = <strong>Frozen</strong> whole muscle cuts or ground (e.g., chicken breasts, 

roasts, ground turkey) 
Q43 What types of halal frozen whole muscle cuts did you sell at your store? Select all that 
apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q41 = Frozen fully cooked products 

Q44 What types of halal frozen fully cooked products did you sell at your store? Select all that 
apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q41 = Ready to eat shelf stable products (e.g., snack sticks, jerky) 

Q45 What types of halal ready to eat (shelf stable) products did you sell at your store? Select all 
that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q41 = Deli meats/poultry 

Q46 What types of halal deli products did you sell at your store? Select all that apply. 

▢  Beef  
▢  Veal  

181 

 
 
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q41 = Canned, smoked, or cured meats/poultry 

Q47 What species of halal canned, smoked, or cured products did you sell at your store? Select 
all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If What&nbsp;halal meat or poultry products did you sell at your store?&nbsp;Select all 

that apply. Text Response Is Not Empty 
Q48 What species of halal ${Q41/ChoiceTextEntryValue/7} products did you sell at your 
store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Q49 Please select the top three (3) motivations for your decision to sell halal meat/poultry 
products: 

▢  My store offered halal meat/poultry products so I could sell those products at a higher 
retail margin  
▢  My store offered halal meat/poultry products so I could attract more customers  
▢  My store offered halal meat/poultry products so I could compete with other similar 
businesses that offered halal meat/poultry products  
▢  My store offered halal meat/poultry products so I could provide halal meat/poultry 
products for Muslim communities  
▢  My store offered halal meat/poultry products so I could provide halal meat/poultry 
products for people with diverse cultural backgrounds  
Q50 Why did you stop offering halal? (Check all that apply) 

▢  Costs became prohibitive  
▢  Wanted to offer different products  
▢  Standards were too strict/hard to meet  

182 

 
 
▢  Poor working relationship with certifier  
▢  Insufficient demand for halal meat/poultry products  
▢  Insufficient supply of halal meat/poultry products  
▢  Other. Please specify: __________________________________________________ 

Q51 Were any of your halal meat/poultry products certified by a third party? 

o  Yes  
o  No  
o  I am not sure  

Q52 Was your store certified by a third party to provide halal meat or poultry products? 

o  Yes  
o  No  
o  I am not sure  
Display This Question: 
If Q51 = Yes 

Q53 Which organization(s) was your store's halal certification from? Select all that apply. 

▢  Islamic Services of America (ISA)  
▢  Halal Food Standards Alliance of America (HFSAA)  
▢  Halal Monitoring Services (HMS)  
▢  Halal Transactions of Omaha (HTO)  
▢  The Islamic Society of the Washington Area (ISWA)  
▢  Other. Please specify: __________________________________________________ 
▢  ⊗Unsure/don't know  

Display This Question: 
If Q51 = Yes 

Q54 Which organization(s) were your products' halal certification from? Select all that apply. 

▢  Islamic Services of America (ISA)  
▢  Halal Food Standards Alliance of America (HFSAA)  
▢  Halal Monitoring Services (HMS)  
▢  Halal Transactions of Omaha (HTO)  
▢  The Islamic Society of the Washington Area (ISWA)  
▢  Other. Please specify: __________________________________________________ 
▢  ⊗Unsure/don't know  

Q55 Have you ever considered selling halal meat/poultry products at your store again? 

o  Yes  
o  No  

Q56 Please indicate how likely each of the following are to MOTIVATE you to 
sell halal meat/poultry products at your store again: 

183 

 
 
Unlikely  Neutral 

Higher retail margin for halal products  
Access to new customers to sell halal products  
Supplying niche religious products to minority communities  
The ability to compete with other similar businesses that 
offer halal meat/poultry products  
Assistance from organizations (certifiers, producer groups, 
etc.) to coordinate finding a supplier  
Q57 Please indicate how likely each of the following would be a CHALLENGE if you were to 
sell halal meat/poultry products at your store again: 

o 
o 
o 
o 

o 
o 
o 
o 

o 

o 

o 

Likely 
o 
o 
o 
o 

Costs associated with certification program involvement 
(certification fees, infrastructure, traceability technology 
materials, labor, etc.)  
Religious or racial discrimination from regulating bodies  
Backlash or displeasure from my non-Muslim customers  
Limited local customer base to sell halal products  
Lack of knowledge for how to sell halal products at my 
store  
Display This Question: 

Unlikely  Neutral 

o 

o 
o 
o 
o 

o 

o 
o 
o 
o 

Likely 
o 

o 
o 
o 
o 

If Q57 = Costs associated with certification program involvement (certification fees, 

infrastructure, traceability technology materials, labor, etc.) [ Neutral ] 

Or Q57 = Costs associated with certification program involvement (certification fees, 

infrastructure, traceability technology materials, labor, etc.) [ Likely ] 
Q58 In the previous question, you indicated that costs would be a challenge or barrier when 
starting a halal program at your store again. Which of the following costs would be the most 
challenging? Select all that apply. 

▢  Reoccurring certification fees  
▢  One-time infrastructure costs (e.g., additional refrigerators or shelves)  
▢  Increased labor  
▢  Traceability technology costs  
▢  Other. Please specify: __________________________________________________ 

Q59 Have you ever considered selling halal meat/poultry products at your store? 

o  Yes  
o  No  
o  I do/did not know what halal is  

Q60 Are you interested in learning more about halal meat/poultry retail opportunities for your 
store? 

o  Yes  
o  No  

Display This Question: 
If Q59 = Yes 
Or Q60 = Yes 

Q61 What halal meat or poultry products would you like to sell at your store? Select all that 
apply. 

184 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
▢  Fresh whole muscle cuts or ground (e.g., chicken breasts, roasts, ground turkey)  
▢  Frozen whole muscle cuts or ground (e.g., chicken breasts, roasts, ground turkey)  
▢  Frozen fully cooked products  
▢  Ready to eat shelf stable products (e.g., snack sticks, jerky)  
▢  Deli meats/poultry  
▢  Canned, smoked, or cured meats/poultry  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q61 = <strong>Fresh</strong> whole muscle cuts or ground (e.g., chicken breasts, 

roasts, ground turkey) 
Q62 What types of halal fresh whole muscle cuts would you like to sell at your store? Select all 
that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q61 = <strong>Frozen</strong> whole muscle cuts or ground (e.g., chicken breasts, 

roasts, ground turkey) 
Q63 What types of halal frozen whole muscle cuts would you like to sell at your store? Select 
all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q61 = Frozen fully cooked products 

Q64 What types of halal frozen fully cooked products would you like to sell at your 
store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

185 

 
 
If Q61 = Ready to eat shelf stable products (e.g., snack sticks, jerky) 

Q65 What types of halal ready to eat (shelf stable) products would you like to sell at your 
store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q61 = Deli meats/poultry 

Q66 What types of halal deli products would you like to sell at your store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q61 = Canned, smoked, or cured meats/poultry 

Q67 What species of halal canned, smoked, or cured products would you like to sell at your 
store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If What&nbsp;halal meat or poultry products would you like to sell at your 

store?&nbsp;Select all that apply. Text Response Is Not Empty 
Q68 What species of halal ${Q61/ChoiceTextEntryValue/7} products would you like to sell at 
your store? Select all that apply. 

▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 

186 

 
 
Q69 Please indicate how likely each of the following are to MOTIVATE you to sell halal meat 
or poultry products at your store: 

Unlikely  Neutral 

Likely 
o 
o 
o 
o 

Higher retail margin for halal products  
Access to new customers to sell halal products  
Supplying niche religious products to minority communities  
The ability to compete with other similar businesses that 
offer halal meat/poultry products  
Assistance from organizations (certifiers, producer groups, 
etc.) to coordinate finding a supplier  
Q70 Please indicate how likely each of the following serve as a CHALLENGE in your decision 
to sell halal meat or poultry products at your store: 

o 
o 
o 
o 

o 
o 
o 
o 

o 

o 

o 

Costs associated with certification program involvement 
(certification fees, infrastructure, traceability technology 
materials, labor, etc.)  
Religious or racial discrimination from regulating bodies  
Backlash or displeasure from my non-Muslim customers  
Limited local customer bases to sell halal products  
Lack of knowledge for how to sell halal products at my 
store  
Display This Question: 

Unlikely  Neutral 

o 

o 
o 
o 
o 

o 

o 
o 
o 
o 

Likely 
o 

o 
o 
o 
o 

If Q70 = Costs associated with certification program involvement (certification fees, 

infrastructure, traceability technology materials, labor, etc.) [ Neutral ] 

Or Q70 = Costs associated with certification program involvement (certification fees, 

infrastructure, traceability technology materials, labor, etc.) [ Likely ] 
Q71 In the previous question, you indicated that costs would be a challenge or barrier when 
starting a halal program at your store. Which of the following costs would be the most 
challenging? Select all that apply. 

▢  Reoccurring certification fees  
▢  One-time infrastructure costs (e.g., additional refrigerators or shelves)  
▢  Increased labor  
▢  Traceability technology costs  
▢  Other. Please specify: __________________________________________________ 
Q146 The following question is to verify that you are paying attention. Please select the animal 
that has hooves from the options below. 

o  Cat  
o  Horse  
o  Dog  
o  Hamster  

Skip To: End of Survey If Q146 != Horse 
Q72 There is not a  NHMC program in the U.S. I am interested in your opinions to help design a 
future national U.S. meat and poultry halal certification program. If you do not currently 
sell halal meat or poultry products, imagine that you are considering selling halal meat or 

187 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
poultry products at your store. 

 In the following section of the survey, you will be presented seven scenarios. Please consider the 
three factors presented, and indicate which one factor is the least important and which one is the 
most important to you when designing a national U.S. halal meat and poultry certification 
program. Please select one factor as least important AND one factor as most important in each 
question. 

 The questions look similar but contain different comparisons of factors. Please treat each 
question individually. 

 To help, I have given an example below with ice cream, where flavor is the most important 
factor and price is the least important factor in your decision to buy ice cream.  
EXAMPLE  
 Of the following three factors, which one is the least important and which one is the most 
important in your decision to buy ice cream?  

Please click the arrow to continue. 

Q73 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 

Least Important 
 (Check only one) 

Most Important 
 (Check only one) 

o 

o 

o 

⊗Which group(s) will enforce the program (e.g., 
government, religious organization, private non-
religious organization) 
⊗What will be required to be certified (e.g., 
products, retailers, slaughter and processing 
establishments) 
⊗What halal standards will be required (e.g., 
hand versus machine slaughter, stunned or not 
stunned) 

o 

o 

o 

188 

 
 
  
  
  
  
 
 
 
 
 
 
 
 
 
 
 
Q74 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 
Least Important 
 (Check only one)  

Most Important 
 (Check only one)  

o 

o 

o 

⊗Which group(s) will enforce the program (e.g., 
government, religious organization, private non-
religious organization)   
⊗Inspection process (e.g., frequency, random or 
scheduled)   
⊗Benefits associated with certification program 
involvement (e.g., access to new markets, higher 
retail margin, increased consumer trust)  

o 

o 

o 

Q75 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 
Least Important 
 (Check only one)  

Most Important 
 (Check only one)  

o 

o 

o 

⊗Costs associated with certification program 
involvement (e.g., certification fees, infrastructure, 
traceability technology materials, labor)  
⊗Which group(s) will enforce the program (e.g., 
government, religious organization, private non-
religious organization)  
⊗What information will be passed on to my 
customers and how they access it (e.g., only 
available through the Freedom of Information Act 
(FOIA) versus accessible online)   

o 

o 

o 

Q76 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 
Least Important 
 (Check only one)  

Most Important 
 (Check only one)  

o 

o 

o 

⊗What information will be passed on to my 
customers and how they access it (e.g., only 
available through the Freedom of Information Act 
(FOIA) versus accessible online)   
⊗Inspection process (e.g., frequency, random or 
scheduled)   
⊗What halal standards will be required (e.g., 
hand versus machine slaughter, stunned or not 
stunned)  

o 

o 

o 

189 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q77 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 
Least Important 
 (Check only one)  

Most Important 
 (Check only one)  

o 

o 

o 

⊗Benefits associated with certification program 
involvement (e.g., access to new markets, higher 
retail margin, increased consumer trust)  
⊗What halal standards will be required (e.g., 
hand versus machine slaughter, stunned or not 
stunned)   
⊗Costs associated with certification program 
involvement (e.g., certification fees, infrastructure, 
traceability technology materials, labor)  

o 

o 

o 

Q78 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 
Least Important 
 (Check only one)  

Most Important 
 (Check only one)  

o 

o 

o 

⊗Costs associated with certification program 
(e.g., facility modifications, traceability equipment, 
certification fees, labor)   
⊗What will be required to be certified (e.g., 
products, retailers, slaughter and processing 
establishments)   
⊗Inspection process (e.g., frequency, random or 
scheduled)   

o 

o 

o 

Q79 Of the following three factors, which one is the least important and which one is the most 
important to consider when designing a national U.S. halal meat and poultry certification 
program? 
Least Important 
 (Check only one)  

Most Important 
 (Check only one)  

⊗What information will be passed on to my 
customers and how they access it (e.g., only 
available through the Freedom of Information Act 
(FOIA) versus accessible online)   
⊗Benefits associated with certification program 
involvement (e.g., access to new markets, higher 
retail margin, increased consumer trust)  
⊗What will be required to be certified (e.g., 
products, retailers, slaughter and processing 
establishments)   

190 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q80 If there were a national halal certification, should individual products or supply chain 
members be required to have a certification to ensure authentic halal meat and poultry products? 

o  Individual products  
o  Supply chain members (e.g., processor, wholesaler, retailer)  
o  Both individual products and supply chain members  

Display This Question: 

If Q80 = Supply chain members (e.g., processor, wholesaler, retailer) 
Or Q80 = Both individual products and supply chain members 

Q81 Which members of the supply chain should be required to have a national halal 
certification for their establishment? Select all that apply. 

▢  Slaughter establishments  
▢  Processing establishments  
▢  Distributors or transportation services  
▢  Retailers and wholesalers  
▢  Restaurants and food service  
▢  Other. Please specify: __________________________________________________ 

Display This Question: 

If Q80 = Supply chain members (e.g., processor, wholesaler, retailer) 
Or Q80 = Both individual products and supply chain members 

Q82 Under a national halal certification, how should a certified business be inspected? 

o  Pre-scheduled inspections  
o  Random/surprise inspections  
o  A mixture of pre-scheduled and random inspections  
o  Other. Please specify: __________________________________________________ 
Q83 Under a national halal certification, how should a certified meat or poultry product be 
verified or traced? Select all that apply. 

▢  Paper trail/certificates  
▢  Online universal internet or cloud-based system (e.g., blockchain, RFID)  
▢  Online store-specific system (e.g., store records)  
▢  Laboratory tests to ensure no pork DNA  
▢  Government audits  
▢  Third party certifier audits (non-government)  
▢  Other. Please specify: __________________________________________________ 
Q84 Which benefits of a national halal certification would be most important to your business? 
Select all that apply. 

▢  Access to new markets  
▢  Ability to receive a higher retail margin for my products  
▢  Increased consumer trust  
▢  Ease of identifying a certified product  
▢  Ease of communicating product attributes  
▢  Other. Please specify: __________________________________________________ 
Q85 Which costs of a national halal certification would be most important to your business? 
Select all that apply. 

▢  Reoccurring certification fees  

191 

 
 
▢  Cost of establishment modifications (e.g., more shelves/coolers)  
▢  Increased labor hours needed  
▢  Cost of traceability and/or verification equipment  
▢  Other. Please specify: __________________________________________________ 
Q86 In your opinion, what standards should be included in a national halal certification for 
meat and poultry? Select all that apply. 

▢  Zabiha (hand-slaughter)  
▢  Machine-slaughter  
▢  Slaughterers of Muslim faith  
▢  Slaughterers of Christian or Jewish faith  
▢  Individual spoken blessings  
▢  Animal(s) not stunned  
▢  Animal(s) face Mecca at time of slaughter  
▢  Non-GMO  
▢  Other. Please specify: __________________________________________________ 
Q87 For a national halal certification, please indicate which parties should have access to the 
names of slaughter and/or processing establishment(s) at which an individual halal meat or 
poultry product was produced. Select all that apply. 

▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Q88 Please indicate which parties should have access to a national list of halal certified 
meat/poultry establishments certified with a national halal certification. Select all that apply. 

▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Q89 Please indicate which parties should have access to the name(s) of enforcement agencies 
that certify halal establishments with a national halal certification. Select all that apply. 

▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Q90 Please indicate which parties should have access to information regarding the halal 
standards used in slaughter and/or processing under a national halal certification. Select all that 
apply. 

192 

 
 
▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Display This Question: 

If Q87 = General public 
Or Q88 = General public 
Or Q89 = General public 
Or Q90 = General public 

Q91 How should the general public be able to access information related to a national halal 
certification program for meat/poultry (e.g., where it is was produced, what halal standards were 
used)? Select all that apply. 

▢  Online (e.g., company website, online database)  
▢  Using a QR code/cell phone app  
▢  Freedom of Information Act (FOIA) request  
▢  Other. Please specify: __________________________________________________ 

Q92 Who should set standards for a new national halal certification for meat and poultry? 
Select all that apply. 
Note: U.S. and state governments are not legally allowed to define standards for religious 
products. 

▢  Non-government organizations  
▢  Religious organizations  
▢  Certifier-led organizations  
▢  Producer-led organizations  
▢  Slaughterers/processors  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Other. Please specify: __________________________________________________ 
Q93 Who should enforce a new national halal certification for meat and poultry? Select all that 
apply. 

▢  U.S. government organization (e.g., the USDA)  
▢  State government organization (e.g., state department of agriculture)  
▢  Non-government organizations  
▢  Religious organizations  
▢  Certifier-led organizations  
▢  Producer-led organizations  
▢  Slaughterers/processors  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Other. Please specify: __________________________________________________ 

Q94 Please select the degree to which you agree or disagree with the following statements: 

193 

 
 
Agree  Neutral  Disagree 
o 

o 

o 

o 
o 

o 
o 

o 
o 

Halal meat or poultry tastes better than non-halal meat or 
poultry  
All halal meat or poultry retail stores must be halal certified  
Halal retail stores that are certified have a stronger reputation 
than halal retail stores that are not certified  
Halal meat/poultry is more sanitary than non-halal meat/ 
poultry  
The halal slaughter process is more humane for the animal  
Retail stores that are not halal certified cannot be trusted to 
supply authentic halal products  
Halal meat and poultry is higher quality than non-halal meat 
and poultry  
Halal meat and poultry is healthier than non-halal meat and 
poultry  
If a retail store has a good reputation for supplying halal meat 
and poultry products, it does not need to be certified as halal  
Q145 The following question is to verify that you are a real person.   Which of the following is 
equal to 10 plus 21? 

o 
o 

o 
o 

o 
o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o  11  
o  17  
o  31  

Skip To: End of Survey If Q145 != 31                                           
Q95 What is your current age? 

o  18-24 years old  
o  25-34 years old  
o  35-44 years old  
o  45-54 years old  
o  55-64 years old  
o  65-74 years old  
o  75 years or older  
o  Prefer not to disclose  
Q96 Were you born in the U.S.? 

o  Yes  
o  No, I was born in this country: ______________________________________________ 
o  Prefer not to disclose  

Q97 Were your parents and grandparents born in the U.S.? 

o  Yes  
o  No, they were born in this/these countries: ____________________________________ 
o  Prefer not to disclose  

Display This Question: 

If Q96 = No, I was born in this country: 
Or Were you born in the U.S.? Text Response Is Not Empty 

Q98 How long have you lived in the U.S.? 

o  0-5 years  
o  6-10 years  

194 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
       
o  11-15 years  
o  16-20 years  
o  21-25 years  
o  Over 25 years  
o  Prefer not to disclose  

Q99 What is your gender? 

o  Male  
o  Female  
o  Prefer to self-describe: __________________________________________________ 
o  Prefer not to disclose  

Q100 What is the highest level of education you have completed? 

o  Less than High School  
o  High school graduate or GED  
o  Some college  
o  2-year degree (Associates)  
o  4-year degree (BA, BS)  
o  Professional degree (JD, MD, PhD, etc.)  
o  Prefer not to disclose  

Q101 Are you of Hispanic, Latino, or Spanish origin? 

o  No  
o  Yes; please specify: __________________________________________________ 
o  Prefer not to disclose  

Q102 What is your race? Select all that apply. 

▢  White  
▢  Black or African American  
▢  Native American or Alaska Native  
▢  Native Hawaiian or Pacific Islander  
▢  Asian  
▢  Other. Please specify: __________________________________________________ 
▢  ⊗Prefer not to disclose  

Q103 Which U.S. political party do you most identify with? 

o  Democrat  
o  Republican  
o  I am an independent  
o  Other. Please specify: __________________________________________________ 
o  Prefer to not disclose  

Q104 Which best describes the area in which you live? 

o  Rural  
o  Suburban or mid-size city  
o  Urban or large city  
o  Prefer not to disclose  

Q105 Do you consider yourself to be religious? 

o  No  
o  Yes, I am: __________________________________________________ 
o  Prefer not to disclose  

Display This Question: 

195 

 
 
If Q105 = No 

Q106 Have you followed a religion in the past even if you do not do so now? 

o  No  
o  Yes, I was: __________________________________________________ 
o  Prefer not to disclose  

Q107 Do you have any final thoughts or comments you wish to share about selling/not selling 
halal meat or poultry at your store? 

________________________________________________________________ 
________________________________________________________________ 

Q143 Thank you for completing this survey! If you would like to retrieve your $25 reward for 
participating, please enter your email address in the box below, and I will reach out to you with 
your gift card. If you do not wish to receive your gift card, you can click the "Next" button at the 
bottom of the screen to skip this question. Please note that in order to qualify for this gift card, 
your survey response will need to meet data quality standards and will be reviewed by the 
research team. Gift cards will be sent out on a weekly or bi-weekly basis, so it may take up to 
14 days to receive your gift card. Thank you for your participation in this research project! 

 Email address for gift card:__________________________________________________ 

196 

 
 
 
  
 
 
 
Halal Meat Consumer Survey 
Q1 You are being asked to participate in a research study of U.S. Muslim consumer preferences 
for halal meat and poultry retail purchases.      Your participation in this study will take about 
twenty (20) minutes. You will be asked to respond to a series of questions about how you 
purchase halal meat, your preferences for retail locations, and halal meat certifications. We also 
ask some basic demographic questions. This survey will assist researchers to anticipate the 
demand for various halal meat products and improve awareness of halal-related issues or events 
that could affect demand. Researchers are required to provide a consent form to inform you 
about the research study, to convey that participation is voluntary, to explain risks and benefits of 
participation, and to empower you to make an informed decision. You should feel free to ask the 
researchers any questions you may have.   
 Study Title: U.S. Meat Industry Overview: Consumer Preferences, Retailer Motivations, and 
Processor Practices 

 Researcher Title and Contact Information: Melissa G.S. McKendree, PhD, 
mckend14@msu.edu and Kelsey Hopkins, PhD candidate, hopki190@msu.edu, 847-513-1708 

 Department and Institution: Department of Agricultural, Food, and Resource Economics, 
Michigan State University 

 Sponsor: USDA-Agriculture and Food Research Initiative    
 The risks associated with this study are minimal. The risks are not greater than those ordinarily 
encountered in daily life. Moreover, you may stop the survey at any time. The data will only be 
released in summaries in which no individual’s answers can be identified.  You have the right to 
say no to participate in the research. You can stop at any time after it has already started. There 
will be no consequences if you stop, and you will not be criticized.  You will not lose any 
benefits that you normally receive.    
 If you have concerns or questions about this study, such as scientific issues, how to do any part 
of it, or to report an injury, please contact the researcher Melissa G.S. McKendree, 202 Morrill 
Hall of Agriculture, East Lansing, MI, 48824, mckend14@msu.edu.  If you have questions or 
concerns about your role and rights as a research participant, would like to obtain information or 
offer input, or would like to register a complaint about this study, you may contact, anonymously 
if you wish, the Michigan State University’s Human Research Protection Program at 517-355-
2180, Fax 517-432-4503, or e-mail irb@msu.edu or regular mail at 4000 Collins Rd, Suite 136, 
Lansing, MI 48910.   
 Continuing with the survey means that you voluntarily agree to participate in this research 
study. Please click the arrow at the bottom of the screen to continue. 

Q2 Please ensure you are carefully reading through the statements and making thoughtful 
selections in order to qualify for the incentive. Any nonsense answers, keyboard slamming, 
quality issues, or speeding will be disqualified without incentive. 
o  Agree and continue  

o  Disagree and exit  

Skip To: End of Block If Please ensure you are carefully reading through the statements and 
making thoughtful selections i... = Disagree and exit 

197 

 
 
  
  
  
 
Q3 Are you 18 years or older and live in the U.S.? 

o  Yes  
o  No  

Skip To: End of Block If Are you 18 years or older and live in the U.S.? = No 
Q4 Have you purchased a halal meat or poultry product in the last twelve (12) months? 

o  Yes  
o  No  

Skip To: End of Block If Have you purchased a halal meat or poultry product in the last twelve 
(12) months? = No 
Q5 Do you identify as Muslim? 

o  Yes  
o  No  

Skip To: End of Block If Do you identify as Muslim? = No 
Q6 Are you one of the primary grocery shoppers for your household?  

o  Yes  
o  No  

Skip To: End of Block If Are you one of the primary grocery shoppers for your household?  = No 
Q7 Do you follow a vegetarian, vegan, or pescatarian diet? (That is, you do not eat meat.) 
o  Yes  
o  No  
Q8 How often do you eat halal meat products? 
o  Always. I never eat non-halal meat products.  
o  Very often. It is rare that I eat meat that is non-halal.  
o  Often. I frequently eat halal meat products.  
o  Somewhat often. I eat halal meat products, but I eat non-halal meat products just as 

frequently.  

o  Almost never. I typically eat non-halal meat products.  
o  Never.  
Q9 If halal meat products are not available, will you purchase kosher meat products to eat 
instead? 
o  Always. I always purchase kosher meat if halal meat is not available.  
o  Very often. It is very common that I purchase kosher meat if halal meat is not available.  
o  Often. I frequently purchase kosher meat if halal meat is not available.  
o  Somewhat often. I purchase kosher meat about half of the time if halal meat is not available.  
o  Almost never. It is rare that I purchase kosher meat if halal meat is not available.  
o  Never. I do not purchase kosher meat if halal meat is not available.  

198 

 
 
 
Q10 If halal meat products are not available, will you purchase vegetarian or pescatarian 
(fish/seafood) options instead? 
o  Always. I always purchase vegetarian or pescatarian options if halal meat is not available.  
o  Very often. It is very common that I purchase vegetarian or pescatarian options if halal meat 

is not available.  

o  Often. I frequently purchase vegetarian or pescatarian options if halal meat is not available.  
o  Somewhat often. I purchase vegetarian or pescatarian options about half of the time if halal 

meat is not available.  

o  Almost never. It is rare that I purchase vegetarian or pescatarian options if halal meat is not 

available.  

o  Never. I do not purchase vegetarian or pescatarian options if halal meat is not available.  
Q11 Please indicate whether you agree or disagree with the following statements: 

▼ Strongly agree ... Not applicable/ Unsure 

▼ Strongly agree ... Not applicable/ Unsure 

▼ Strongly agree ... Not applicable/ Unsure 

⊗I prefer that my halal meat retailer is 
reliable  
⊗I prefer that my halal meat retailer is 
considerate and nice  
⊗I prefer that my halal meat retailer is well-
stocked  
⊗I prefer that my halal meat retailer sells 
quality halal meat products  
⊗I prefer that my halal meat retailer is 
transparent and honest  
Q12 What is the name of the store you purchase the majority of your everyday halal meat and 
poultry products from?  
 Note: "Everyday" halal meat and poultry products are those you purchase for daily 
consumption outside of religious holidays such as Eid. 
o  Name of store: __________________________________________________ 
Q13 This section of the survey asks you about your buying relationship with 
${e://Field/Store%20Name}.  
Q14 Please indicate whether you agree or disagree with the following statements: 

▼ Strongly agree ... Not applicable/ Unsure 

▼ Strongly agree ... Not applicable/ Unsure 

${e://Field/Store%20Name} demonstrates empathy and kindness 
toward me and treats everyone fairly  
${e://Field/Store%20Name} openly shares information, motives, 
and choices in straightforward and plain language  
${e://Field/Store%20Name} consistently and dependably delivers 
on their promises  
${e://Field/Store%20Name} communicates product characteristics 
in plain and easy to understand language  
${e://Field/Store%20Name} resolves issues in an adequate and 
timely manner  

▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 

199 

 
 
 
 
 
 
 
Q15 Please indicate whether you agree or disagree with the following statements: 

${e://Field/Store%20Name} quickly resolves issues with safety, 
security and satisfaction in mind  
${e://Field/Store%20Name} values and respects everyone, 
regardless of background, identity or beliefs  
${e://Field/Store%20Name} values the good of society and the 
environment, not just profit  
${e://Field/Store%20Name} is upfront about how they make and 
spend money from our interactions  
${e://Field/Store%20Name} is clear and upfront about fees and 
costs of products, services and experiences  
Q16 Please indicate whether you agree or disagree with the following statements: 

▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 

${e://Field/Store%20Name}'s products are good quality, accessible 
and safe to use  
${e://Field/Store%20Name}'s prices are good value for the money   ▼ Strongly agree ... 

▼ Strongly agree ... 
Not applicable/ Unsure 

${e://Field/Store%20Name}'s employees and leadership are 
competent and understand how to respond to my needs  
${e://Field/Store%20Name} can be counted on to improve the 
quality of their products and services  
${e://Field/Store%20Name} consistently delivers products, services, 
and experiences with quality  
Q17 Please indicate whether you agree or disagree with the following statements: 

Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 

⊗${e://Field/Store%20Name} sells quality halal meat products  

⊗${e://Field/Store%20Name} takes care of their employees  

▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 
▼ Strongly agree ... 
Not applicable/ Unsure 

⊗${e://Field/Store%20Name}'s marketing and communications are 
accurate and honest  
⊗${e://Field/Store%20Name} creates long term solutions and 
improvements that work well for me  
⊗${e://Field/Store%20Name} facilitates digital interactions that 
run smoothly and work when needed (e.g., placing online or phone 
orders)  
Q18 Please rate your overall relationship (in terms of trustworthiness, friendliness, reliability, 
and transparency) with ${e://Field/Store%20Name}. A score of "100" indicates your relationship 
with ${e://Field/Store%20Name} is ideal or perfect, a score of "0" indicates your relationship is 
nonexistent. 

0  10  20  30  40  50  60  70  80  90  100 

200 

 
 
 
 
 
 
 
 
 
 
Relationship 

Q19 In this section of the survey, we will be asking you questions about 
${e://Field/Store%20Name} and the halal meat or poultry products you purchase there. 
Q20 What type of store is ${e://Field/Store%20Name}? 
o  Small ethnic grocery store (for example, an Asian or Indo-Pakistani grocery store)  
o  Small local or regional non-ethnic grocery store  
o  Box/chain store (for example, Walmart, Whole Foods, Meijer)  
o  Membership store (for example, Costco, Restaurant Depot)  
o  Butcher shop (that is, they sell only meat or poultry products)  
o  Online retailer (for example, One Stop Halal, Crescent Foods)  
o  Other. Please specify: __________________________________________________ 
Q21 How often do you purchase halal meat or poultry products from 
${e://Field/Store%20Name}? 
o  Daily  
o  4-6 times a week  
o  2-3 times a week  
o  Once a week  
o  Every other week  
o  Once a month  
o  Every other month  
o  3-5 times per year  
o  1-2 times per year  
Q22 When you purchase halal meat or poultry, do you purchase only for your household, or do 
you purchase on behalf of a group or multiple households? 
o  Only my household  
o  Multiple households or a group  
Q23 When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what kinds 
of products do you purchase? (Select all that apply) 
▢  Fresh whole cuts or ground products (for example, chicken breasts or ground beef purchased 

at a butcher counter)  

▢  Pre-packaged fresh whole cuts or ground products (for example, steak or ground turkey 

available in coolers/on refrigerated shelves)  

▢  Frozen packaged whole cuts (for example, frozen chicken breasts)  
▢  Refrigerated processed products (for example, lunch/deli meat, hot dogs)  
▢  Frozen processed products (for example, frozen dinners, frozen hamburger patties)  
▢  Ready to eat processed products (for example, jerky, snack sticks)  
▢  Other. Please specify: __________________________________________________ 
Display This Question: 

If When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what 
kinds of product... = Fresh whole cuts or ground products (for example, chicken breasts or 
ground beef purchased at a butcher counter) 
Q24 When shopping at ${e://Field/Store%20Name}, which types of halal meat or poultry do 
you purchase as fresh whole cuts (for example, chicken breasts or ground beef purchased from a 

201 

 
 
 
butcher counter)? 
 Select all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Display This Question: 

If When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what 
kinds of product... = Pre-packaged fresh whole cuts or ground products (for example, steak or 
ground turkey available in coolers/on refrigerated shelves) 
Q25 When shopping at ${e://Field/Store%20Name}, which types of halal meat or poultry do 
you purchase as pre-packaged refrigerated whole cuts (for example, steaks or ground turkey 
available in coolers/on refrigerated shelves)? Select all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Display This Question: 

If When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what 

kinds of product... = Frozen packaged whole cuts (for example, frozen chicken breasts) 
Q26 When shopping at ${e://Field/Store%20Name}, which types of halal meat or poultry do 
you purchase as frozen packaged whole cuts (for example, chicken breasts available in 
freezers)? Select all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Display This Question: 

If When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what 
kinds of product... = Refrigerated processed products (for example, lunch/deli meat, hot dogs) 
Q27 When shopping at ${e://Field/Store%20Name}, which types of halal meat or poultry do 
you purchase as refrigerated processed products (for example, lunch/deli meat, hotdogs)? 
 Select all that apply. 
▢  Beef  
▢  Veal  

202 

 
 
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Display This Question: 

If When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what 
kinds of product... = Frozen processed products (for example, frozen dinners, frozen hamburger 
patties) 
Q28 When shopping at ${e://Field/Store%20Name}, which types of halal meat or poultry do 
you purchase as frozen processed products (for example, frozen dinners, frozen hamburger 
patties)? Select all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Display This Question: 

If When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what 

kinds of product... = Ready to eat processed products (for example, jerky, snack sticks) 
Q29 When shopping at ${e://Field/Store%20Name}, which types of halal meat or poultry do 
you purchase as ready to eat processed products (for example, jerky, snack sticks)? 
 Select all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Display This Question: 

If If When you buy halal meat or poultry products from ${e://Field/Store%20Name}, what 

kinds of product... Text Response Is Not Empty 
Q30 When shopping at ${e://Field/Store%20Name}, which types of halal meat or poultry do 
you purchase as ${Q23/ChoiceTextEntryValue/6}? 
 Select all that apply. 
▢  Beef  
▢  Veal  
▢  Lamb  
▢  Goat  
▢  Turkey  

203 

 
 
▢  Chicken  
▢  Other. Please specify: __________________________________________________ 
Q31 Is ${e://Field/Store%20Name} certified by a halal certification organization (for example, 
the store has a sign/certificate displayed from a certifier)? 
o  Yes. Please specify certifier if known: ____________________________________ 
o  No  
o  I don't know  
Q32 Does ${e://Field/Store%20Name} sell individually halal certified meat products? (for 
example, a package has a label or stamp on it that says "halal") 
o  Yes. Please specify certifier if known: _____________________________________ 
o  No  
o  I don't know  
Q33 Please select all factors that contributed to your decision to purchase from 
${e://Field/Store%20Name}. (Select all that apply) 
▢  Recommendations from friends or family  
▢  Good customer service  
▢  Online reviews  
▢  Recommendations from mosque, Islamic community center, or other religious leader/group  
▢  Reputation or history as a trustworthy retailer  
▢  Store carries a halal certification for all meat or poultry products  
▢  Certified halal meat products available  
▢  Owner / staff is Muslim  
▢  Retailer serves my country's community (for example, Indo-Pakistani, Somali, Turkish)  
▢  Low/no sales tax  
▢  Near other businesses I shop at  
▢  Low/fair prices  
▢  Near where I live  
▢  Near where I work  
▢  Cleanliness / sanitary environment  
▢  Quality of products available  
▢  Other. Please specify: __________________________________________________ 
Q34 What standard of halal meat or poultry products do you purchase? 
o  Zabiha (hand-slaughtered) halal  
o  Machine-slaughtered halal  
o  Whichever is available  
o  I do not have a preference  
o  I do not know the difference between Zabiha and machine-slaughtered halal  
Q35 What standards do you prefer when you purchase your halal meat and poultry products? 
Select all that apply. 
▢  Zabiha (hand-slaughter)  
▢  Machine-slaughter  
▢  Slaughterers of Muslim faith  
▢  Slaughterers of Christian or Jewish faith  

204 

 
 
▢  Individual spoken blessings  
▢  Animal(s) not stunned  
▢  Animal(s) face Mecca at time of slaughter  
▢  Does not contain genetically modified organisms (non-GMO)  
▢  Other. Please specify: __________________________________________________ 
Q36 Where or from whom do you get information about what standards are required for meat 
or poultry to be halal? Select all that apply. 
▢  Friends / family  
▢  Religious leader / group (for example, from your imam, your mosque, or Islamic community 

center)  

▢  Social media page(s) (for example, a Facebook group)  
▢  Store websites / advertisements  
▢  Halal certifiers  
▢  Other. Please specify: __________________________________________________ 
Q37 Where or from whom do you get information about where to find authentic halal meat or 
poultry? Select all that apply. 
▢  Friends / family  
▢  Religious leader / group (for example, from your imam, your mosque, or Islamic community 

center)  

▢  Social media page(s) (for example, a Facebook group)  
▢  Store websites / advertisements  
▢  Halal certifiers  
▢  Consumer apps / websites (for example, Scan Halal)  
▢  Other. Please specify: __________________________________________________ 
Q56 The following question is to verify that you are a real person.   Which of the following is 
equal to 10 + 21? 
o  9  
o  31  
o  53  
Skip To: End of Block If The following question is to verify that you are a real person. Which of 
the following is equal t... != 31 
Q38 Please read these instructions and descriptions carefully, as they are important for the 
next questions.  

In this section of the survey, you will be asked eight (8) questions where you will choose 
between two (2) different stores to purchase halal meat and poultry products, or have the option 
to not shop at either store. 

When comparing the stores, you will see different characteristics:   The use of a comprehensive 
store-wide halal certification  A halal certification only on individual products   Your one-way 
travel time to the store   
the stores are the same (for example, prices and variety of products available) - that is, 
your choice should depend on the information shown.  
 In the following questions, you will be shown pictures of the stores you are choosing between to 

Your relationship with the retailer.    All other characteristics of 

205 

 
 
  
  
help you to visualize your choice.  
 ____________________________________________________________________________ 
 Halal Certifications 
 For the purpose of this survey, the halal standards for these certifications are those that you 
prefer (for example, hand or machine cut, stunned or not stunned). 

 On the front of these stores, you will see different halal certification logos, which are also 
shown here with a description:   
      The store above with the green 8-point star is certified halal, meaning that all the meat and 
poultry products sold inside are certified as halal.     The store above with the blue 18-point star 
carries halal certified meat and poultry products, but not all meat and poultry products sold 
inside are necessarily certified as halal. Halal certified meat and poultry products would be 
labeled individually with the halal certification.      The store above with the green 8-point star 
and blue 18-point star is both halal certified and carries halal certified meat and poultry 
products, meaning that all the meat and poultry products sold inside are certified as halal and all 
of these products are additionally individually labeled as halal certified.  
 ____________________________________________________________________________ 
 Travel Time 
 Travel time is the one-way time shown is the minutes it takes you to get to the retailer from 
your home. Travel times will be either 15, 30, 45, or 60 minutes one-way. 
 ____________________________________________________________________________ 
 Relationship 
 You will see four (4) different possible levels of your relationship with the retailer; these are 
given below with definitions. 

 No relationship with this retailer: You have never purchased from this store before and do not 
know anything about their trustworthiness, friendliness, reliability, or transparency. 

 Relationship I have with my current retailer: This is the current relationship you have with 
the retailer you purchase from most frequently, in terms of trustworthiness, friendliness, 
reliability, and transparency. 

 Best retailer relationship I have experienced: This is the best relationship you have had with a 
retailer in terms of trustworthiness, friendliness, reliability, and transparency. 

 Ideal or perfect retailer relationship: This is the most trustworthy, friendly, reliable, and 
transparent retailer you can imagine purchasing halal meat and poultry products from. 

Q39 IMPORTANT: Previous similar surveys have found that people often state they are willing 
to shop at different stores when they are not actually willing to do so. Accordingly, it is 
important that you make each of your upcoming selections like you would if you were actually 
facing these exact choices; that is, noting that choosing to shop at one store means that you 
would not shop at the other location for your halal meat or poultry products. The accuracy of 
your responses is very important, as the information collected here will be used to help design 
future policies and regulations for halal meat and poultry products. 

206 

 
 
  
  
  
  
  
 
  
 Please read carefully and be aware that every question has different information even 
though they may look very similar.  
Q40 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 No certifications present 60 minutes total time spent traveling to the store  Best retailer 
relationship I have experienced  
 Store and product certifications present   45 minutes total time spent traveling to the store  
Ideal or perfect retailer relationship  

o 

o  I would not shop for halal meat products at either of these two stores   
Q41 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 No certifications present 15 minutes total time spent traveling to the store  Ideal or perfect 
retailer relationship  
 Product and store certifications present   60 minutes total time spent traveling to the store  
No relationship with this retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q42 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 
 Product certification present 30 minutes total time spent traveling to the store  No 
relationship with this retailer  
 Store certification present   15 minutes total time spent traveling to the store  Relationship I 
have with my current retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q43 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Product certification present 45 minutes total time spent traveling to the store  Relationship I 
have with my current retailer  
 Store certification present  30 minutes total time spent traveling to the store  Best retailer 
relationship I have experienced  

o 

o  I would not shop for halal meat products at either of these two stores   
Q44 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account certifications, 
one-way travel time to the store, and your relationship with the retailer? 
o 

 Store certification present 45 minutes total time spent traveling to the store  No relationship 
with this retailer  
 Product certification present  30 minutes total time spent traveling to the store  Relationship 
I have with my current retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q45 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 

207 

 
 
o 

o 

 Store certification present  30 minutes total time spent traveling to the store  Relationship I 
have with my current retailer  
 Product certification present   15 minutes total time spent traveling to the store  Best retailer 
relationship I have experienced  

o  I would not shop for halal meat products at either of these two stores  
Q46 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Store and product certifications present  15 minutes total time spent traveling to the store  
Best retailer relationship I have experienced  
 No certifications present   60 minutes total time spent traveling to the store  Ideal or perfect 
retailer relationship  

o 

o  I would not shop for halal meat products at either of these two stores   
Q47 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Store and product certifications present  60 minutes total time spent traveling to the store  
Ideal or perfect retailer relationship  
 No certifications present   45 minutes total time spent traveling to the store  No relationship 
with this retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q48 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 No certifications present  15 minutes total time spent traveling to the store  No relationship 
with this retailer  
 Store and product certifications present   60 minutes total time spent traveling to the store  
Relationship I have with my current retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q49 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 No certifications present  60 minutes total time spent traveling to the store  Relationship I 
have with my current retailer  
 Store and product certifications present   45 minutes total time spent traveling to the store  
Best retailer relationship I have experienced  

o 

o  I would not shop for halal meat products at either of these two stores   
Q50 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Product certification present  45 minutes total time spent traveling to the store  Best retailer 
relationship I have experienced  
 Store certification present   30 minutes total time spent traveling to the store  Ideal or perfect 
retailer relationship  

o 

o  I would not shop for halal meat products at either of these two stores   

208 

 
 
Q51 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Product certification present  30 minutes total time spent traveling to the store  Ideal or 
perfect retailer relationship  
 Store certification present   15 minutes total time spent traveling to the store  No relationship 
with this retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q52 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Store certification present  30 minutes total time spent traveling to the store  Best retailer 
relationship I have experienced  
 Product certification present   15 minutes total time spent traveling to the store  Ideal or 
perfect retailer relationship  

o 

o  I would not shop for halal meat products at either of these two stores   
Q53 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Store certification present  45 minutes total time spent traveling to the store  Ideal or perfect 
retailer relationship  
 Product certification present   30 minutes total time spent traveling to the store  No 
relationship with this retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q54 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Store and product certifications present  60 minutes total time spent traveling to the store  
No relationship with this retailer  
 No certification present   45 minutes total time spent traveling to the store  Relationship I 
have with my current retailer  

o 

o  I would not shop for halal meat products at either of these two stores   
Q55 Imagine you are grocery shopping and want to purchase halal meat or poultry products. Of 
the stores below, which would you prefer to purchase from, taking into account halal 
certifications, one-way travel time to the store, and your relationship with the retailer? 
o 

 Store and product certifications present  15 minutes total time spent traveling to the store  
Relationship I have with my current retailer  
 No certification present   60 minutes total time spent traveling to the store  Best retailer 
relationship I have experienced  

o 

o  I would not shop for halal meat products at either of these two stores   
Q57 Please select the degree to which you agree or disagree with the following statements: 

209 

 
 
Halal meat and poultry is healthier to eat than non-halal meat and 
poultry  
I will never eat meat or poultry that is not halal certified  

▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 

Halal meat and poultry is cleaner/more hygienic than non-halal 
meat or poultry  
I have access to good information about halal certified meat and 
poultry in the U.S.  
Halal meat and poultry tastes better than non-halal meat and poultry   ▼ Strongly agree ... 

Q58 Please select the degree to which you agree or disagree with the following statements: 

Not applicable / Unsure 

I will not buy a meat or poultry product if my peers or family have 
doubts about whether it is truly halal  
I always check labels to see if all ingredients are halal before 
purchasing  
Halal meat and poultry slaughtering is more humane than non-halal 
meat and poultry slaughtering  
It is easy for me to tell if a meat or poultry product is halal  

I am willing to travel extra miles to get authentic halal meat or 
poultry products  
I check to see if a restaurant serves halal food before I eat there  

▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 

Q59 Please select the degree to which you agree or disagree with the following statements: 

I am willing to pay more for meat or poultry that has been certified 
as halal  
If halal meat or poultry is not available, I will chose a seafood or 
vegetarian option instead  
I always check to see if a meat or poultry product is certified halal 
before eating it  
I have enough knowledge about U.S. halal meat and poultry to tell 
the difference between halal certified and non-halal meat and 
poultry  
Q60 Now we will move onto the next portion of the survey, which will ask you questions about 
your opinions about a national halal certification program. 
Q61 There is not a NHMC program in the U.S. We are interested in your opinions to help design 
a future national U.S. meat and poultry halal certification program.  

▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 
▼ Strongly agree ... 
Not applicable / Unsure 

In the following section of the survey, you will be presented seven (7) scenarios. Please consider 
the three (3) factors presented, and indicate which one (1) factor is the least important and which 
one (1) is the most important to you when designing a national U.S. halal meat and poultry 
certification program. Please select one factor as least important AND one factor as most 
important in each question. 

210 

 
 
 
 
 
 
 
 
  
 The questions look similar but contain different comparisons of factors. Please treat each 
question individually. 

 To help, we have given an example below with ice cream, where flavor is the most important 
factor and price is the least important factor in your decision to buy ice cream.  
EXAMPLE  
 Of the following three (3) factors, which one (1) is the least important and which one (1) is the 
most important in your decision to buy ice cream?  

 Q62 Of the following three (3) factors, which one (1) is the least important and which one (1) is 
the most important to consider when designing a national U.S. halal meat and poultry 
certification program? 
Least Important 
 (Check only one)  
o 

Most Important 
 (Check only one)  
o 

⊗Which group(s) will enforce the program (for 
example, government, religious organization, private 
non-religious organization)  
⊗What will be required to be certified (for 
example, products, retailers, slaughter and processing 
establishments)  
⊗What halal standards will be required (for 
example, hand versus machine slaughter, stunned or 
not stunned)  

Q63 Of the following three (3) factors, which one (1) is the least important and which one (1) is 
the most important to consider when designing a national U.S. halal meat and poultry 
certification program? 
Least Important 
 (Check only one)  
o 

Most Important 
 (Check only one)  
o 

⊗Which group(s) will enforce the program (for 
example, government, religious organization, private 
non-religious organization)  
⊗Inspection process (for example, frequency, 
random or scheduled)  
⊗Benefits associated with certified products (for 
example, transparency, reliability, quality)  

o 

o 

o 

o 

o 

o 

o 

o 

211 

 
 
  
  
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q64 Of the following three (3) factors, which one (1) is the least important and which one (1) is 
the most important to consider when designing a national U.S. halal meat and poultry 
certification program? 
Least Important 
 (Check only one)  
o 

Most Important 
 (Check only one)  
o 

Q65 Of the following three (3) factors, which one (1) is the least important and which one (1) is 
the most important to consider when designing a national U.S. halal meat and poultry 
certification program? 
Least Important 
 (Check only one)  
o 

Most Important 
 (Check only one)  
o 

⊗Costs associated with certified products (for 
example, higher prices)  
⊗Which group(s) will enforce the program (for 
example, government, religious organization, private 
non-religious organization)  
⊗What product information I have access to and 
how I can access it (for example, only available 
through the Freedom of Information Act (FOIA) 
versus accessible online)  

⊗What product information I have access to and 
how I can access it (for example, only available 
through the Freedom of Information Act (FOIA) 
versus accessible online)  
⊗Inspection process (for example, frequency, 
random or scheduled)  
⊗What halal standards will be required (for 
example, hand versus machine slaughter, stunned or 
not stunned)  

Q66 Of the following three (3) factors, which one (1) is the least important and which one (1) is 
the most important to consider when designing a national U.S. halal meat and poultry 
certification program? 
Least Important 
 (Check only one)  
o 

Most Important 
 (Check only one)  
o 

⊗Benefits associated with certified products (for 
example, transparency, reliability, quality)  
⊗What halal standards will be required (for 
example, hand versus machine slaughter, stunned or 
not stunned)  
⊗Costs associated with certified products (for 
example, higher prices)  

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

o 

212 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q67 Of the following three (3) factors, which one (1) is the least important and which one (1) is 
the most important to consider when designing a national U.S. halal meat and poultry 
certification program? 
Least Important 
 (Check only one)  
o 

Most Important 
 (Check only one)  
o 

o 

o 

⊗Costs associated with certified products (for 
example, higher prices)  
⊗What will be required to be certified (for 
example, products, retailers, slaughter and 
processing establishments)  
⊗Inspection process (for example, frequency, 
random or scheduled)  

o 

o 

Q68 Of the following three (3) factors, which one (1) is the least important and which one (1) is 
the most important to consider when designing a national U.S. halal meat and poultry 
certification program? 
Least Important 
 (Check only one)  

Most Important 
 (Check only one)  

⊗What product information I have access to and 
how I can access it (for example, only available 
through the Freedom of Information Act (FOIA) 
versus accessible online)  
⊗Benefits associated with certified products (for 
example, transparency, reliability, quality)  
⊗What will be required to be certified (for 
example, products, retailers, slaughter and processing 
establishments)  

213 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Q69 Should individual products or supply chain members be required to have a national halal 
certification to ensure authentic halal meat and poultry products? 
o  Individual products  
o  Supply chain members (for example, processor, wholesaler, retailer)  
o  Both individual products and supply chain members  
Display This Question: 

If Should individual products or supply chain members be required to have a national halal 

certifica... = Supply chain members (for example, processor, wholesaler, retailer) 

Or Should individual products or supply chain members be required to have a national halal 

certifica... = Both individual products and supply chain members 
Q70 Which members of the supply chain should be required to have a national halal 
certification? Select all that apply. 
▢  Slaughter establishments  
▢  Processing establishments  
▢  Distributors or transportation services  
▢  Retailers and wholesalers  
▢  Restaurants and food service  
▢  Other. Please specify: __________________________________________________ 
Q71 Under a national halal certification, how should a certified business be inspected? 
o  Pre-scheduled inspections  
o  Random/surprise inspections  
o  A mixture of pre-scheduled and random inspections  
o  Other. Please specify: __________________________________________________ 
Q72 If there were a national halal certification for meat and poultry, what should be the 
minimum requirements for the certification? Select all that apply. 
▢  Zabiha (hand-slaughter)  
▢  Machine-slaughter  
▢  Slaughterers of Muslim faith  
▢  Slaughterers of Christian or Jewish faith  
▢  Individual spoken blessings  
▢  Animal(s) not stunned  
▢  Animal(s) face Mecca at time of slaughter  
▢  Not genetically modified (Non-GMO)  
▢  Other. Please specify: __________________________________________________ 
Q73 Please indicate which parties should have access to the names of establishment(s) at which 
a halal meat or poultry product was slaughtered and/or processed using a national halal 
certification. Select all that apply. 
▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

214 

 
 
Q74 Please indicate which parties should have access to a national list of halal certified 
meat/poultry establishments certified with a national halal certification. Select all that apply. 
▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  
Q75 Please indicate which parties should have access to the name(s) of enforcement agencies 
that certify halal establishments with a national halal certification. Select all that apply. 
▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  
Q76 Please indicate which parties should have access to information regarding the halal 
standards used in slaughter and/or processing under a national halal certification. Select all that 
apply. 
▢  General public  
▢  Processors/slaughterers  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Government organizations  
▢  ⊗None of the above  

Display This Question: 

If Please indicate which parties should have access to the names of establishment(s) at 

which a hala... = General public 

Or Please indicate which parties should have access to a national list of halal certified 

meat/poult... = General public 

Or Please indicate which parties should have access to the name(s) of enforcement agencies 

that cert... = General public 

Or Please indicate which parties should have access to information regarding the halal 

standards use... = General public 
Q77 How should the general public be able to access information related to a national halal 
certification program for meat/poultry? Select all that apply. 
▢  Online (for example, company website, online database)  
▢  Using a QR code/cell phone app  
▢  Freedom of Information Act (FOIA) request  
▢  Other. Please specify: __________________________________________________ 
Q78 Who should set standards for a new national halal certification for meat and poultry? Select 
all that apply. 
Note: U.S. and state governments are not legally allowed to define standards for religious 
products. 

215 

 
 
▢  Non-government organizations  
▢  Religious organizations  
▢  Certifier-led organizations  
▢  Producer-led organizations  
▢  Slaughterers/processors  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Other. Please specify: __________________________________________________ 
Q79 Who should enforce a new national halal certification for meat and poultry? Select all that 
apply. 
▢  U.S. government organization (for example, the USDA)  
▢  State government organization (for example, state department of agriculture)  
▢  Non-government organizations  
▢  Religious organizations  
▢  Certifier-led organizations  
▢  Producer-led organizations  
▢  Slaughterers/processors  
▢  Distributors/wholesalers  
▢  Retailers/restaurants  
▢  Other. Please specify: __________________________________________________ 
Q80 Please select the degree to which you agree or disagree with the following statements: 

Agree  Neutral 
o 
o 

Disagree 
o 

o 

o 

o 

o 

o 

o 

o 

o 

All halal meat or poultry slaughter or processing 
establishments must be halal certified  
Halal establishments that are certified have a stronger 
reputation than halal establishments that are not certified  
Establishments that are not halal certified cannot be trusted to 
supply authentic halal products  
If an establishment has a good reputation for supplying halal 
meat and poultry products, it does not need to be certified as 
halal  
Q81 Now we are going to ask you questions about yourself. These responses will only be used 
by the research team and will not be shared with any identifying information attached.  
Q82 Which branch of Islam do you practice? 
o  Sunni  
o  Shia or Shiite  
o  Ibadi  
o  Non-denominational  
o  Other. Please specify: __________________________________________________ 
o  Prefer not to disclose  
Q83 Have you converted or reverted to Islam? 
o  Yes  
o  No  
o  Prefer not to disclose  

o 

216 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Display This Question: 

If Have you converted or reverted to Islam? = Yes 

Q84 Which religion did you follow prior to converting or reverting to Islam? 
o  Christianity  
o  Hinduism  
o  Buddhism  
o  Judaism  
o  Sikhism  
o  Other. Please specify: __________________________________________________ 
o  None/ not applicable  
o  Prefer not to disclose  
Display This Question: 

If Which religion did you follow prior to converting or reverting to Islam? = Christianity 

Q85 Which branch of Christianity did you follow prior to converting or reverting to Islam? 
o  Catholicism  
o  Protestant (for example, Baptist, Lutheran, Methodist, Presbyterian, Pentecostal, 

Episcopalian)  

o  Orthodox (for example, Greek, Eastern)  
o  Prefer not to disclose  
Display This Question: 

If Have you converted or reverted to Islam? = Yes 

Q86 When did you convert to Islam? 
o  0-5 years ago  
o  6-10 years ago  
o  11-15 years ago  
o  16-20 years ago  
o  21-25 years ago  
o  26-30 years ago  
o  31-35 years ago  
o  Over 35 years ago  
o  Prefer not to disclose  
Q87 I consider myself: 
o  A vegetarian  
o  A vegan  
o  None of the above  
o  Prefer not to disclose  
Q88 What is your current age? 
o  18 - 24 years old  
o  25 - 34 years old  
o  35 - 44 years old  
o  45 - 54 years old  
o  55 - 64 years old  
o  65 - 74 years old  
o  75 years or older  
o  Prefer not to disclose  
Q89 What is your gender? 

217 

 
 
o  Male  
o  Female  
o  Non-binary / third gender  
o  Prefer to self-describe: __________________________________________________ 
o  Prefer not to disclose  
Q90 What is your current marital status? 
o  Single, Never Married  
o  Married  
o  Separated  
o  Divorced  
o  Widowed  
o  Prefer not to disclose  
Q91 How many people (including yourself) live in your household? 
o  1  
o  2  
o  3  
o  4  
o  5 or more  
o  Prefer not to disclose  
Q92 Are there children under the age of 12 living in your household? 
o  Yes  
o  No  
o  Prefer not to disclose  
Q93 Have you ever received food stamps? 
o  Yes  
o  No  
o  Prefer not to disclose  
Q94 Are you currently on food stamps? 
o  Yes  
o  No  
o  Prefer not to disclose  
Q95 In what U.S. state do you live? 

▼ Alabama ... Wyoming 

Q96 What is your ZIP code? _____________________________________________________ 
Q97 Please select your U.S. citizenship status. 
o  U.S. Citizen  
o  Lawful Permanent Resident  
o  Temporary Resident (e.g., visitor, student)  
o  Prefer not to disclose  
Q98 Were you born in the U.S.? 
o  Yes  
o  No, I was born in this country: ____________________________________________ 
o  Prefer not to disclose  
Q99 Were your parents born in the U.S.? 
o  Yes  
o  No, they were born in this country: __________________________________________ 

218 

 
 
o  Prefer not to disclose  
Display This Question: 

If Were you born in the U.S.? = No, I was born in this country: 
Or Or Were you born in the U.S.? Text Response Is Not Empty 

Q100 How long have you lived in the U.S.? 
o  0-5 years  
o  6-10 years  
o  11-15 years  
o  16-20 years  
o  21-25 years  
o  26-30 years  
o  31-35 years  
o  More than 35 years  
o  Prefer not to disclose  
Q101 What is the highest level of education you have completed? 
o  Less than High School  
o  High School/GED  
o  Some College  
o  2-Year College Degree (Associates)  
o  4-Year College Degree (BA, BS)  
o  Master's Degree  
o  Professional Degree (Ph.D., J.D., M.D., etc.)  
o  Prefer not to disclose  
Q102  What is your approximate annual household income before taxes? 
o  Less than $20,000  
o  $20,000 - $39,999  
o  $40,000 - $59,999  
o  $60,000 - $79,999  
o  $80,000 - $99,999  
o  $100,000 - $119,999  
o  $120,000 - $139,999  
o  $140,000 - $159,999  
o  $160,000 or greater  
o  Prefer not to disclose  
Q103 Which category best describes you? Select all that apply. 
▢  White (for example, German, Irish, English, Italian, Polish, French, etc.)  
▢  Hispanic, Latino or Spanish origin (for example, Mexican or Mexican American, Puerto 

Rican, Cuban, Salvadoran, Dominican, Colombian, etc.)  

▢  Black or African American (for example, African American, Jamaican, Haitian, Nigerian, 

Ethiopian, Somalian, etc.)  

▢  Middle Eastern or North African (for example, Lebanese, Iranian, Egyptian, Syrian, 

Moroccan, Algerian, etc.)  

▢  American Indian or Alaskan Native (for example, Navajo nation, Blackfeet tribe, Mayan, 

Aztec, Native Village or Barrow Inupiat Traditional Government, Nome Eskimo 
Community, etc.)  

▢  Asian (for example, Chinese, Korean, Japanese, etc.)  

219 

 
 
▢  South or Southeast Asian (for example, Indian, Pakistani, Filipino, Vietnamese, Malaysian, 

etc.)  

▢  Native Hawaiian or Other Pacific Islander (for example, Native Hawaiian, Samoan, 

Chamorro, Tongan, Fijian, etc.)  

▢  Other. Please specify: __________________________________________________ 
▢  Prefer not to disclose  
Q104 Which U.S. political party do you most identify with? 
o  Democratic  
o  Republican  
o  I am an independent  
o  Other. Please specify: __________________________________________________ 
o  Prefer not to disclose  
Q105 Which best describes the area in which you live? 
o  Rural  
o  Suburban  
o  Urban  
o  Prefer not to disclose  
Q106 Do you have any final comments to share about your responses or this study? 

________________________________________________________________ 
________________________________________________________________ 

220 

 
 
 
 
 
 
APPENDIX A.3 BEST-WORST SCALING RESULTS AND 
ADDITIONAL TABLES 

The count data of the best-worst question responses is given in Figure 12, Figure 13, and 
Figure 14.  Figure 12 aggregates these responses from halal and non-halal processors, Figure 13 
gives the responses from halal processors only, and Figure 14 gives the responses from non-halal 
processors only. We exclude past halal processors from this analysis. Looking at the data in Figure 
12,  we  may  conclude  that  the  meat  and  poultry  processing  industry  overall  is  not  strongly 
concerned about what requirements for information and transparency across the supply chain may 
be imposed by a national halal certification. Likewise, the benefits of certification and inspection 
type and frequency appear to have low importance to the industry overall. Costs of certification, 
which standards are included in the certification, and what establishments are required to carry a 
certification are overall more important to the industry. 

However, when disaggregated into halal and non-halal processors, the count data provides 
a more nuanced insight into the market’s preferences. For halal processors, which standards are 
included in a certification, what establishments are required to be certified, and who is in charge 
of enforcing the certification are highly important, while the remaining four factors are relatively 
unimportant to halal processors. For non-halal processors, the most important characteristics to 
consider when designing a national halal certification program are costs, what is included in the 
certification, and which establishments are required to be certified. The type of information and 
transparency passed along the supply chain appears very unimportant, while the remaining three 
factors  are  of  relatively  mild  importance.  When  comparing  halal  and  non-halal  processors,  I 
conclude  that  halal  processors  are  overall  more  interested  in  the  rigor  of  the  standards  and 
enforcement  integrity  of  a  potential  national  halal  certification,  while  non-halal  processors  are 
most concerned with the costs and how they may need to adjust their business to meet standards. 

Figure 12: Best-Worst Count Data by Attribute of Potential National Halal Certification 
Program, Current Halal and Non-Halal Processors Combined, n = 98 

All Processors Best-Worst Count Data

Which standards are included

45%

21%

34%

What is required to be certified

Costs of certification

41%

38%

16%

22%

Who is in charge of enforcement

28%

32%

Benefits of certification

22%

Inspection frequency and type

22%

34%

33%

Type of information passed down the supply chain

10%

49%

43%

39%

40%

43%

46%

41%

% of time chosen as most important

% of time chosen as least important

% of time not chosen

0%

10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

221 

 
 
 
 
 
Figure 13: Best-Worst Count Data by Attribute of Potential National Halal Certification 
Program, Current Halal Processors, n = 32 

Halal Processors Best-Worst Counts

Which standards are included

66%

8%

27%

What is required to be certified

48%

18%

Costs of certification

16%

50%

34%

34%

Who is in charge of enforcement

41%

27%

32%

Benefits of certification

17%

Inspection frequency and type

16%

38%

40%

Type of information passed down the supply chain

14%

37%

46%

44%

49%

% of time chosen as most important

% of time chosen as least important

% of time not chosen

0%

10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Figure 14: Best-Worst Count Data by Attribute of Potential National Halal Certification 
Program, Non-Halal Processors, n = 66 

Non-Halal Processors Best-Worst Counts

Which standards are included

What is required to be certified

39%

41%

28%

34%

16%

43%

Costs of certification

49%

12%

39%

Who is in charge of enforcement

24%

34%

Benefits of certification

28%

34%

Inspection frequency and type

25%

32%

Type of information passed down the supply chain

7%

56%

43%

38%

43%

37%

% of time chosen as most important

% of time chosen as least important

% of time not chosen

0%

10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

222 

 
 
 
 
 
 
 
Table 40: Multinomial Logit Results of Best-Worst Scaling for Hypothetical U.S. National 
Halal Meat and Poultry Certification Program: Processors (n = 82), Retailers (n = 33), and 
Consumers (n = 507) 

Best-Worst Scaling Attributes 

Cost of Certification 

Enforcement and Regulation 

Information Collected/Available 

Who/What Must be Certified 

Which Halal Standards are Included 

Benefits of Certification 

Processors 
Coefficient 
(st. error) 
0.46*** 
(0.11) 
0.17 
(0.11) 
-0.41*** 
(0.11) 
0.60*** 
(0.11) 
0.55*** 
(0.11) 
0.01 
(0.11) 

Retailers 
Coefficient 
(st. error) 
0.39** 
(0.17) 
-0.14 
(0.17) 
-0.46*** 
(0.17) 
0.62*** 
(0.17) 
0.22 
(0.17) 
0.14 
(0.17) 

Consumers 
Coefficient 
(st. error) 
-0.25*** 
(0.04) 
-0.07 
(0.04) 
-0.16*** 
(0.04) 
0.22*** 
(0.04) 
0.37*** 
(0.04) 
0.06 
(0.04) 

N 
Log likelihood 
AIC 
AIC/N 
Note: *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1. Base attribute normalized to zero is 
Inspection Type/Frequency. 

3549 
-6202.82 
12417.6 
3.499 

231 
-388.49 
789.0 
3.415 

574 
-960.86 
1933.7 
3.369 

223 

 
 
 
 
 
 
 
Table 41: Best-Worst Scaling Shares of Preferences for Hypothetical U.S. National Halal Meat and Poultry Certification 
Program: Consumers, Retailers, & Processors: Multinomial Logit 
Processors (n = 82) 
Share 
0.18*** 
0.13*** 
0.07*** 
0.20*** 
0.11*** 
0.19*** 
0.11*** 

National Halal Meat & Poultry 
Certification Program Attribute 
Cost of Certification 
Enforcement and Regulation 
Information Collected/Available 
Who/What Must be Certified 
Inspection Frequency and Type 
Which Halal Standards are Included 
Benefits of Certification 

Consumers (n = 507) 
95% CI 
Share 
[0.10, 0.11] 
0.11*** 
[0.12, 0.13] 
0.13*** 
[0.11, 0.12] 
0.12*** 
[0.16, 0.18] 
0.17*** 
[0.13, 0.14] 
0.14*** 
[0.19, 0.21] 
0.20*** 
[0.14, 0.15] 
0.14*** 

95% CI 
[0.15, 0.20] 
[0.11, 0.15] 
[0.06, 0.09] 
[0.18, 0.23] 
[0.10, 0.13] 
[0.17, 0.22] 
[0.10, 0.13] 

95% CI 
[0.14, 0.22] 
[0.08, 0.13] 
[0.06, 0.10] 
[0.18, 0.27] 
[0.10, 0.15] 
[0.12, 0.19] 
[0.11, 0.17] 

Share 
0.18*** 
0.11*** 
0.08*** 
0.22*** 
0.12*** 
0.15*** 
0.14*** 

Retailers (n = 33) 

Note: *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1. 

Table 42: P-values from Poe tests for Hypothetical U.S. National Halal Meat and Poultry Certification Program: Pair-wise 
Comparisons Between Processors (n = 82), Retailers (n = 33), and Consumers (507), Multinomial Logit 

National Halal Meat & Poultry 
Certification Program Attribute 
Cost of Certification 
Enforcement and Regulation 
Information Collected/Available 
Who/What Must be Certified 
Inspection Frequency and Type 
Which Halal Standards are Included 
Benefits of Certification 

Processors vs. Retailers 

Processors vs. Consumers  Retailers vs. Consumers 

0.54 
0.05 
0.40 
0.19 
0.26 
0.97 
0.05 

0.00 
0.35 
1.00 
0.01 
0.00 
0.32 
0.00 

0.00 
0.94 
1.00 
0.00 
0.15 
0.00 
0.36 

Note: Values that are statistically significant at the 5% level or better are bolded.

224 

 
 
 
 
Table 43: Uncorrelated Random Parameters Logit Results of Best-Worst Scaling for 
Hypothetical U.S. National Halal Meat and Poultry Certification Program: Processors (n = 
82), Retailers (n = 33), and Consumers (n = 507) 

Best-Worst Scaling Attributes 

Cost of Certification 

Enforcement and Regulation 

Information Collected/Available 

Who/What Must be Certified 

Which Halal Standards are Included 

Benefits of Certification 

Processors 
Coefficient 
(st. error) 
0.80*** 
(0.15) 
0.25* 
(0.14) 
-0.74*** 
(0.14) 
0.92*** 
(0.14) 
0.95*** 
(0.14) 
-0.04 
(0.14) 

Retailers 
Coefficient 
(st. error) 
0.67*** 
(0.23) 
-0.41* 
(0.23) 
-0.78*** 
(0.23) 
0.92*** 
(0.22) 
0.43* 
(0.23) 
0.16 
(0.22) 

Consumers 
Coefficient 
(st. error) 
-0.34*** 
(0.05) 
-0.10** 
(0.05) 
-0.20*** 
(0.05) 
0.27*** 
(0.05) 
0.49*** 
(0.05) 
0.07 
(0.04) 

McFadden Pseudo R2 
N 
Log likelihood 
AIC 
AIC/N 
Note: *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1. Base attribute normalized to zero is 
Inspection Type/Frequency. 

0.12 
3549 
-6095.59 
12215.2 
3.442 

0.21 
231 
-353.08 
730.2 
3.161 

0.21 
574 
-879.30 
1782.6 
3.106 

225 

 
 
 
 
 
 
 
 
Table 44: Best-Worst Scaling Shares of Preferences for Hypothetical U.S. National Halal Meat and Poultry Certification 
Program: Consumers, Retailers, & Processors: Uncorrelated Random Parameters Logit 

National Halal Meat & Poultry 
Certification Program Attribute 
Cost of Certification 
Enforcement and Regulation 
Information Collected/Available 
Who/What Must be Certified 
Inspection Frequency and Type 
Which Halal Standards are Included 
Benefits of Certification 

Processors (n = 82) 
Share 
0.20 
0.12 
0.04 
0.23 
0.09 
0.23 
0.09 

95% CI 
[0.16, 0.24] 
[0.09, 0.14] 
[0.03, 0.05] 
[0.19, 0.27] 
[0.07, 0.11] 
[0.19, 0.27] 
[0.07, 0.11] 

Retailers (n = 33) 

Share 
0.21 
0.07 
0.05 
0.27 
0.11 
0.17 
0.13 

95% CI 
[0.16, 0.27] 
[0.05, 0.10] 
[0.03, 0.07] 
[0.20, 0.34] 
[0.08, 0.14] 
[0.12, 0.22] 
[0.09, 0.17] 

Consumers (n = 507) 
95% CI 
Share 
[0.09, 0.10] 
0.10 
[0.11, 0.13] 
0.12 
[0.10, 0.12] 
0.11 
[0.16, 0.18] 
0.17 
[0.13, 0.14] 
0.13 
[0.21, 0.23] 
0.22 
[0.13, 0.15] 
0.14 

Note: *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1. 

Table 45: P-values from Poe tests for Hypothetical U.S. National Halal Meat and Poultry Certification Program: Pair-wise 
Comparisons Between Processors (n = 82), Retailers (n = 33), and Consumers (507), Uncorrelated Random Parameters Logit 

National Halal Meat & Poultry 
Certification Program Attribute 
Cost of Certification 
Enforcement and Regulation 
Information Collected/Available 
Who/What Must be Certified 
Inspection Frequency and Type 
Which Halal Standards are Included 
Benefits of Certification 

Processors vs. Retailers 

Processors vs. Consumers  Retailers vs. Consumers 

0.57 
0.01 
0.30 
0.16 
0.16 
0.97 
0.03 

0.00 
0.68 
1.00 
0.00 
0.00 
0.72 
0.00 

0.00 
1.00 
1.00 
0.00 
0.06 
0.03 
0.19 

Note: Values that are statistically significant at the 5% level or better are bolded.

226 

 
 
 
Figure 15: All Output of Correlated Random Parameters Logit Results of Best-Worst 
Scaling for Hypothetical U.S. National Halal Meat and Poultry Certification Program, 
Consumers (n = 507) 

Iterative procedure has converged 

Normal exit:  38 iterations. Status=0, F=    .6053561D+04 

----------------------------------------------------------------------------- 

Random Parameters Multinom. Logit Model 

Dependent variable               CHOICE 

Log likelihood function     -6053.56104 

Restricted log likelihood   -6906.03512 

Chi squared [ 27](P= .000)   1704.94815 

Significance level               .00000 

McFadden Pseudo R-squared      .1234390 

Estimation based on N =   3549, K =  27 

Inf.Cr.AIC  =  12161.1 AIC/N =    3.427 

--------------------------------------- 

            Log likelihood R-sqrd R2Adj 

No coefficients -6906.0351  .1234 .1221 

Constants only can be computed directly 

               Use NLOGIT ;...;RHS=ONE$ 

At start values -6202.8157  .0241 .0226 

Note: R-sqrd = 1 - logL/Logl(constants) 

Root Likelihood:Geom. Mean of P^  .1816 

Warning:  Model does not contain a full 

set of ASCs. R-sqrd is problematic. Use 

model setup with ;RHS=one to get LogL0. 

--------------------------------------- 

Response data are given as ind. choices 

Replications for simulated probs. =1000 

Used Halton sequences in simulations. 

RPL model with panel has     507 groups 

Fixed number of obsrvs./group=        7 

Number of obs.=  3549, skipped    0 obs 

227 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 15 (cont’d)  
--------+------------------------------------------------------------------
-- 

        |                  Standard            Prob.      95% Confidence 

  CHOICE|  Coefficient       Error       z    |z|>Z*         Interval 
--------+------------------------------------------------------------------
-- 

        |Random parameters in utility functions.......................... 

      A1|    -.34640***      .04825    -7.18  .0000     -.44098   -.25183 

      A2|    -.09280*        .04750    -1.95  .0507     -.18589    .00030 

      A3|    -.20008***      .04678    -4.28  .0000     -.29177   -.10839 

      A4|     .28766***      .04682     6.14  .0000      .19591    .37942 

      A6|     .51522***      .04916    10.48  .0000      .41887    .61157 

      A7|     .06852         .04628     1.48  .1387     -.02219    .15924 

        |Diagonal values in Cholesky matrix, L........................... 

    NsA1|     .88157***      .05371    16.41  .0000      .77630    .98683 

    NsA2|     .88719***      .04802    18.47  .0000      .79307    .98131 

    NsA3|     .59115***      .04163    14.20  .0000      .50955    .67274 

    NsA4|     .60445***      .04064    14.87  .0000      .52480    .68410 

    NsA6|     .58399***      .04228    13.81  .0000      .50112    .66685 

    NsA7|     .36926***      .03704     9.97  .0000      .29665    .44186 

        |Below diagonal values in L matrix. V = L*Lt..................... 

   A2:A1|    -.21233***      .04916    -4.32  .0000     -.30868   -.11598 

   A3:A1|    -.33569***      .04866    -6.90  .0000     -.43106   -.24031 

   A3:A2|    -.29119***      .04222    -6.90  .0000     -.37394   -.20844 

   A4:A1|    -.08463*        .04841    -1.75  .0804     -.17952    .01026 

   A4:A2|    -.28091***      .04249    -6.61  .0000     -.36418   -.19764 

   A4:A3|     .10671***      .03916     2.73  .0064      .02996    .18346 

   A6:A1|    -.13966***      .04991    -2.80  .0051     -.23749   -.04183 

   A6:A2|    -.32461***      .04452    -7.29  .0000     -.41186   -.23735 

   A6:A3|     .21022***      .04121     5.10  .0000      .12946    .29098 

   A6:A4|     .65385***      .04287    15.25  .0000      .56983    .73788 

   A7:A1|    -.38867***      .04841    -8.03  .0000     -.48356   -.29378 

   A7:A2|    -.11276***      .04168    -2.71  .0068     -.19445   -.03106 

   A7:A3|     .31559***      .03918     8.06  .0000      .23880    .39238 

   A7:A4|     .26119***      .03784     6.90  .0000      .18703    .33535 

   A7:A6|     .08272**       .03635     2.28  .0229      .01146    .15397 

228 

 
 
 
 
 
 
 
Figure 15 (cont’d) 

        |Standard deviations of parameter distributions.................. 

    sdA1|     .88157***      .05371    16.41  .0000      .77630    .98683 

    sdA2|     .91224***      .04805    18.99  .0000      .81807   1.00641 

    sdA3|     .73955***      .04251    17.40  .0000      .65623    .82287 

    sdA4|     .68031***      .04087    16.65  .0000      .60020    .76041 

    sdA6|     .96831***      .04050    23.91  .0000      .88894   1.04769 

    sdA7|     .68905***      .04121    16.72  .0000      .60827    .76982 

        |Covariances of Random Parameters................................ 

   A2:A1|    -.18718***      .03935    -4.76  .0000     -.26430   -.11007 

   A3:A1|    -.29593***      .03810    -7.77  .0000     -.37061   -.22125 

   A3:A2|    -.18706***      .04227    -4.43  .0000     -.26991   -.10421 

   A4:A1|    -.07461*        .04097    -1.82  .0686     -.15491    .00569 

   A4:A2|    -.23125***      .03825    -6.05  .0000     -.30623   -.15627 

   A4:A3|     .17329***      .03888     4.46  .0000      .09708    .24950 

   A6:A1|    -.12312***      .04109    -3.00  .0027     -.20367   -.04258 

   A6:A2|    -.25833***      .04053    -6.37  .0000     -.33777   -.17889 

   A6:A3|     .26567***      .04215     6.30  .0000      .18307    .34828 

   A6:A4|     .52066***      .05060    10.29  .0000      .42148    .61983 

   A7:A1|    -.34264***      .03733    -9.18  .0000     -.41580   -.26948 

   A7:A2|    -.01751         .04338     -.40  .6865     -.10253    .06751 

   A7:A3|     .34986***      .04672     7.49  .0000      .25830    .44143 

   A7:A4|     .25612***      .04050     6.32  .0000      .17674    .33551 

   A7:A6|     .37631***      .04928     7.64  .0000      .27972    .47290 
--------+------------------------------------------------------------------
-- 

***, **, * ==>  Significance at 1%, 5%, 10% level. 

Model was estimated on Jan 01, 2024 at 09:24:34 PM 
---------------------------------------------------------------------------
-- 

Correlation Matrix for Random Parameters 

--------+----------------------------------------------------- 

Cor.Mat.|      A1       A2       A3       A4       A6       A7 

--------+----------------------------------------------------- 

      A1| 1.00000  -.23276  -.45391  -.12440  -.14423  -.56407 

      A2| -.23276  1.00000  -.27727  -.37262  -.29245  -.02786 

      A3| -.45391  -.27727  1.00000   .34443   .37099   .68657 

      A4| -.12440  -.37262   .34443  1.00000   .79037   .54638 

      A6| -.14423  -.29245   .37099   .79037  1.00000   .56400 

      A7| -.56407  -.02786   .68657   .54638   .56400  1.00000 

229 

 
 
 
Figure 16: All Output of Correlated Random Parameters Logit Results of Best-Worst 
Scaling for Hypothetical U.S. National Halal Meat and Poultry Certification Program, 
Retailers (n = 33) 

Iterative procedure has converged 

Normal exit:  21 iterations. Status=0, F=    .3425588D+03 

---------------------------------------------------------------------------
-- 

Random Parameters Multinom. Logit Model 

Dependent variable               CHOICE 

Log likelihood function      -342.55881 

Restricted log likelihood    -449.50524 

Chi squared [ 27](P= .000)    213.89288 

Significance level               .00000 

McFadden Pseudo R-squared      .2379203 

Estimation based on N =    231, K =  27 

Inf.Cr.AIC  =    739.1 AIC/N =    3.200 

--------------------------------------- 

            Log likelihood R-sqrd R2Adj 

No coefficients  -449.5052  .2379 .2197 

Constants only can be computed directly 

               Use NLOGIT ;...;RHS=ONE$ 

At start values  -388.4877  .1182 .0971 

Note: R-sqrd = 1 - logL/Logl(constants) 

Root Likelihood:Geom. Mean of P^  .2270 

Warning:  Model does not contain a full 

set of ASCs. R-sqrd is problematic. Use 

model setup with ;RHS=one to get LogL0. 

--------------------------------------- 

Response data are given as ind. choices 

Replications for simulated probs. =1000 

Used Halton sequences in simulations. 

RPL model with panel has      33 groups 

Fixed number of obsrvs./group=        7 

Number of obs.=   231, skipped    0 obs 

230 

 
 
 
 
 
 
 
 
 
 
Figure 16 (cont’d) 
--------+------------------------------------------------------------------
-- 

        |                  Standard            Prob.      95% Confidence 

  CHOICE|  Coefficient       Error       z    |z|>Z*         Interval 
--------+------------------------------------------------------------------
-- 

        |Random parameters in utility functions.......................... 

      A1|     .91392***      .24643     3.71  .0002      .43093   1.39691 

      A2|    -.17735         .23345     -.76  .4474     -.63491    .28021 

      A3|    -.79397***      .24061    -3.30  .0010    -1.26556   -.32237 

      A4|    1.05049***      .23288     4.51  .0000      .59405   1.50692 

      A6|     .35458         .23941     1.48  .1386     -.11467    .82382 

      A7|     .39813*        .23971     1.66  .0967     -.07168    .86795 

        |Diagonal values in Cholesky matrix, L........................... 

    NsA1|    1.85616***      .29287     6.34  .0000     1.28215   2.43017 

    NsA2|    1.73116***      .26483     6.54  .0000     1.21210   2.25022 

    NsA3|    1.08359***      .22861     4.74  .0000      .63552   1.53166 

    NsA4|     .23369         .19074     1.23  .2205     -.14015    .60752 

    NsA6|     .01543         .19983      .08  .9385     -.37623    .40709 

    NsA7|    1.18455***      .22404     5.29  .0000      .74543   1.62367 

        |Below diagonal values in L matrix. V = L*Lt..................... 

   A2:A1|     .56945**       .23733     2.40  .0164      .10430   1.03460 

   A3:A1|     .31574         .23456     1.35  .1783     -.14399    .77547 

   A3:A2|     .35253         .22054     1.60  .1099     -.07973    .78479 

   A4:A1|     .71177***      .23914     2.98  .0029      .24307   1.18047 

   A4:A2|     .28354         .20243     1.40  .1613     -.11321    .68029 

   A4:A3|    -.09126         .18869     -.48  .6286     -.46109    .27858 

   A6:A1|    -.30952         .25500    -1.21  .2248     -.80930    .19026 

   A6:A2|     .95030***      .23396     4.06  .0000      .49174   1.40886 

   A6:A3|     .63784***      .23022     2.77  .0056      .18662   1.08905 

   A6:A4|   -1.66497***      .28356    -5.87  .0000    -2.22074  -1.10919 

   A7:A1|    1.32502***      .26996     4.91  .0000      .79592   1.85413 

   A7:A2|     .33871         .21834     1.55  .1208     -.08923    .76665 

   A7:A3|    -.66067***      .20993    -3.15  .0016    -1.07213   -.24922 

   A7:A4|     .92612***      .23406     3.96  .0001      .46738   1.38486 

   A7:A6|   -1.00791***      .21907    -4.60  .0000    -1.43727   -.57855 

231 

 
 
 
 
 
 
 
Figure 16 (cont’d) 

        |Standard deviations of parameter distributions.................. 

    sdA1|    1.85616***      .29287     6.34  .0000     1.28215   2.43017 

    sdA2|    1.82241***      .26765     6.81  .0000     1.29784   2.34699 

    sdA3|    1.18243***      .22299     5.30  .0000      .74539   1.61948 

    sdA4|     .80619***      .23139     3.48  .0005      .35268   1.25971 

    sdA6|    2.04403***      .23082     8.86  .0000     1.59164   2.49643 

    sdA7|    2.36297***      .20718    11.41  .0000     1.95690   2.76905 

        |Covariances of Random Parameters................................ 

   A2:A1|    1.05699*        .54743     1.93  .0535     -.01595   2.12993 

   A3:A1|     .58607         .47567     1.23  .2179     -.34623   1.51837 

   A3:A2|     .79009*        .46112     1.71  .0866     -.11369   1.69387 

   A4:A1|    1.32116**       .55949     2.36  .0182      .22458   2.41773 

   A4:A2|     .89617*        .47633     1.88  .0599     -.03742   1.82976 

   A4:A3|     .22581         .30547      .74  .4598     -.37291    .82452 

   A6:A1|    -.57452         .46464    -1.24  .2163    -1.48518    .33615 

   A6:A2|    1.46887***      .55642     2.64  .0083      .37831   2.55944 

   A6:A3|     .92844**       .40721     2.28  .0226      .13033   1.72656 

   A6:A4|    -.39814         .46307     -.86  .3899    -1.30575    .50946 

   A7:A1|    2.45946***      .75575     3.25  .0011      .97822   3.94069 

   A7:A2|    1.34090**       .60810     2.21  .0274      .14904   2.53275 

   A7:A3|    -.17813         .44834     -.40  .6911    -1.05685    .70060 

   A7:A4|    1.31586**       .54121     2.43  .0150      .25511   2.37661 

   A7:A6|   -2.06715***      .78743    -2.63  .0087    -3.61048   -.52382 
--------+------------------------------------------------------------------
-- 

***, **, * ==>  Significance at 1%, 5%, 10% level. 

Model was estimated on Jan 01, 2024 at 07:44:43 PM 
---------------------------------------------------------------------------
-- 

Correlation Matrix for Random Parameters 

--------+----------------------------------------------------- 

Cor.Mat.|      A1       A2       A3       A4       A6       A7 

--------+----------------------------------------------------- 

      A1| 1.00000   .31247   .26703   .88288  -.15143   .56074 

      A2|  .31247  1.00000   .36665   .60996   .39432   .31138 

      A3|  .26703   .36665  1.00000   .23688   .38414  -.06375 

      A4|  .88288   .60996   .23688  1.00000  -.24161   .69074 

      A6| -.15143   .39432   .38414  -.24161  1.00000  -.42798 

      A7|  .56074   .31138  -.06375   .69074  -.42798  1.00000 

232 

 
 
 
Figure 17: All Output of Correlated Random Parameters Logit Results of Best-Worst 
Scaling for Hypothetical U.S. National Halal Meat and Poultry Certification Program, 
Processors (n = 96) 

Iterative procedure has converged 

Normal exit:  32 iterations. Status=0, F=    .8606388D+03 

---------------------------------------------------------------------------
-- 

Random Parameters Multinom. Logit Model 

Dependent variable               CHOICE 

Log likelihood function      -860.63878 

Restricted log likelihood   -1116.95243 

Chi squared [ 27](P= .000)    512.62728 

Significance level               .00000 

McFadden Pseudo R-squared      .2294759 

Estimation based on N =    574, K =  27 

Inf.Cr.AIC  =   1775.3 AIC/N =    3.093 

--------------------------------------- 

            Log likelihood R-sqrd R2Adj 

No coefficients -1116.9524  .2295 .2222 

Constants only can be computed directly 

               Use NLOGIT ;...;RHS=ONE$ 

At start values  -960.8607  .1043 .0958 

Note: R-sqrd = 1 - logL/Logl(constants) 

Root Likelihood:Geom. Mean of P^  .2233 

Warning:  Model does not contain a full 

set of ASCs. R-sqrd is problematic. Use 

model setup with ;RHS=one to get LogL0. 

--------------------------------------- 

Response data are given as ind. choices 

Replications for simulated probs. =1000 

Used Halton sequences in simulations. 

RPL model with panel has      82 groups 

Fixed number of obsrvs./group=        7 

Number of obs.=   574, skipped    0 obs 

233 

 
 
 
 
 
 
 
 
 
 
Figure 17 (cont’d) 
--------+------------------------------------------------------------------
-- 

        |                  Standard            Prob.      95% Confidence 

  CHOICE|  Coefficient       Error       z    |z|>Z*         Interval 
--------+------------------------------------------------------------------
-- 

        |Random parameters in utility functions.......................... 

      A1|     .91671***      .15806     5.80  .0000      .60691   1.22651 

      A2|     .28945**       .14281     2.03  .0427      .00955    .56935 

      A3|    -.81308***      .15676    -5.19  .0000    -1.12032   -.50584 

      A4|     .98937***      .14577     6.79  .0000      .70366   1.27509 

      A6|     .97008***      .15223     6.37  .0000      .67172   1.26845 

      A7|    -.04691         .14636     -.32  .7486     -.33378    .23995 

        |Diagonal values in Cholesky matrix, L........................... 

    NsA1|    2.09648***      .21355     9.82  .0000     1.67792   2.51503 

    NsA2|    1.31529***      .15257     8.62  .0000     1.01627   1.61432 

    NsA3|    1.46573***      .15789     9.28  .0000     1.15628   1.77518 

    NsA4|     .82066***      .12589     6.52  .0000      .57392   1.06740 

    NsA6|    1.31810***      .15276     8.63  .0000     1.01871   1.61750 

    NsA7|    1.39940***      .15315     9.14  .0000     1.09923   1.69958 

        |Below diagonal values in L matrix. V = L*Lt..................... 

   A2:A1|     .53761***      .15864     3.39  .0007      .22669    .84854 

   A3:A1|    -.18392         .15577    -1.18  .2377     -.48922    .12138 

   A3:A2|     .34598**       .13530     2.56  .0106      .08081    .61115 

   A4:A1|    -.28950*        .15286    -1.89  .0582     -.58911    .01011 

   A4:A2|     .04337         .12925      .34  .7372     -.20996    .29670 

   A4:A3|     .31768***      .12000     2.65  .0081      .08249    .55287 

   A6:A1|     .45964***      .16344     2.81  .0049      .13931    .77997 

   A6:A2|     .32291**       .13953     2.31  .0206      .04945    .59638 

   A6:A3|     .37361***      .12727     2.94  .0033      .12415    .62306 

   A6:A4|    -.70761***      .13096    -5.40  .0000     -.96430   -.45093 

   A7:A1|    -.74733***      .16135    -4.63  .0000    -1.06356   -.43110 

   A7:A2|     .59629***      .13425     4.44  .0000      .33316    .85941 

   A7:A3|     .83063***      .13309     6.24  .0000      .56977   1.09149 

   A7:A4|    -.16944         .12007    -1.41  .1582     -.40477    .06589 

   A7:A6|     .24544**       .12172     2.02  .0438      .00688    .48401 

234 

 
 
 
 
 
 
 
Figure 17 (cont’d) 

        |Standard deviations of parameter distributions.................. 

    sdA1|    2.09648***      .21355     9.82  .0000     1.67792   2.51503 

    sdA2|    1.42092***      .16430     8.65  .0000     1.09890   1.74295 

    sdA3|    1.51720***      .15719     9.65  .0000     1.20912   1.82528 

    sdA4|     .92742***      .12721     7.29  .0000      .67810   1.17674 

    sdA6|    1.64111***      .14855    11.05  .0000     1.34996   1.93225 

    sdA7|    1.91084***      .13770    13.88  .0000     1.64094   2.18073 

        |Covariances of Random Parameters................................ 

   A2:A1|    1.12709***      .38691     2.91  .0036      .36875   1.88543 

   A3:A1|    -.38558         .32129    -1.20  .2301    -1.01529    .24414 

   A3:A2|     .35619*        .21533     1.65  .0981     -.06585    .77823 

   A4:A1|    -.60693*        .31371    -1.93  .0530    -1.22178    .00793 

   A4:A2|    -.09859         .19115     -.52  .6060     -.47324    .27605 

   A4:A3|     .53389**       .21289     2.51  .0121      .11663    .95114 

   A6:A1|     .96363**       .38080     2.53  .0114      .21729   1.70998 

   A6:A2|     .67184**       .26875     2.50  .0124      .14509   1.19858 

   A6:A3|     .57479**       .23677     2.43  .0152      .11073   1.03885 

   A6:A4|    -.58109***      .17591    -3.30  .0010     -.92586   -.23631 

   A7:A1|   -1.56676***      .31778    -4.93  .0000    -2.18960   -.94393 

   A7:A2|     .38252         .24389     1.57  .1168     -.09549    .86053 

   A7:A3|    1.56123***      .33940     4.60  .0000      .89603   2.22643 

   A7:A4|     .36703         .22372     1.64  .1009     -.07145    .80552 

   A7:A6|     .60279**       .27302     2.21  .0273      .06768   1.13789 
--------+------------------------------------------------------------------
-- 

***, **, * ==>  Significance at 1%, 5%, 10% level. 

Model was estimated on Jan 01, 2024 at 08:01:49 PM 
---------------------------------------------------------------------------
-- 

Correlation Matrix for Random Parameters 

--------+----------------------------------------------------- 

Cor.Mat.|      A1       A2       A3       A4       A6       A7 

--------+----------------------------------------------------- 

      A1| 1.00000   .37835  -.12122  -.31216   .28008  -.39110 

      A2|  .37835  1.00000   .16522  -.07482   .28811   .14088 

      A3| -.12122   .16522  1.00000   .37943   .23085   .53852 

      A4| -.31216  -.07482   .37943  1.00000  -.38179   .20711 

      A6|  .28008   .28811   .23085  -.38179  1.00000   .19222 

      A7| -.39110   .14088   .53852   .20711   .19222  1.00000 

235 

 
 
 
 
 
APPENDIX A.4 FARM ANIMAL WELFARE REGULATIONS 
ADDITIONAL TABLES 

Table 46: Stage 1 Legislative Action Decision Outcomes Coefficients 

VARIABLES 

HENS_PER_1000 

HOGS_PER_1000 

COUNT_PASSED_PREV 

PREV_LAW 

ALLOW_BALLOT 

TRIFECTA_D 

TRIFECTA_R 

HOUSE%D 

SENATE%D 

CONSTANT 

(1) 
Ballot Outcome MNL 
Coefficients 
<0.001 
(<0.001) 
-0.054** 
(0.025) 
0.049 
(0.043) 
-0.724 
(0.963) 
2.965*** 
(1.086) 
-0.622 
(0.580) 
-1.268 
(1.165) 
0.019 
(0.041) 
-0.005 
(0.039) 
-6.740*** 
(1.645) 

980 
Observations 
Note: Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1. 

Table 47: Out of Sample Stage 1 Accurate Prediction Percentage 

(2) 
Bill Outcome MNL 
Coefficients 
<-0.001 
(<0.001) 
-0.008** 
(0.003) 
0.093*** 
(0.034) 
0.528 
(0.468) 
0.367 
(0.428) 
0.893* 
(0.505) 
-1.719* 
(0.888) 
0.044 
(0.032) 
-0.013 
(0.026) 
-6.106*** 
(1.077) 

Predicted Outcome Accuracy 

Mean 
0.931 
Note:  We  used  clusters  of  45  states  in  the  Stage  1  bootstrap;  the  remaining  5  states  are  used  for  out  of  sample 
predictions for each repetition. An accurate prediction is when the real-world outcome matches the overall prediction 
of the model. As there are three possible outcomes (no action, ballot proposed, bill proposed), the predicted probability 
for any given outcome must be larger than the remaining two outcomes for it to be considered the model’s overall 
prediction. 

Standard Deviation 
0.218 

236 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 18: K-Density Plots: Ballot Model Residuals (top), Ballot Model R-Squared 
(middle), and Ballot Model Inverse Mills Ratios (bottom) 

K-Density Ballot Model Residuals

y
t
i
s
n
e
D

5
.
1

1

5
.

0

0
4

0
3

y
t
i
s
n
e
D

0
2

0
1

0

8
0
.

6
0
.

y
t
i
s
n
e
D

4
0
.

2
0
.

0

-4

-2

0

Residuals

kernel = epanechnikov, bandwidth = 0.0262

2

4

K-Density Ballot Model R-squared

.3

.4

.5
R-Squared

.6

.7

kernel = epanechnikov, bandwidth = 0.0007

K-Density Ballot Model Inverse Mills Ratio

-200

-150

-100
Ballot Inverse Mills Ratio

-50

0

kernel = epanechnikov, bandwidth = 0.2838

237 

 
 
 
 
 
 
 
 
 
Table 48: Ballot IMR by Year 
Year 
2000 
2001 
2002 
2003 
2004 
2005 
2006 
2007 
2008 
2009 
2010 
2011 
2012 
2013 
2014 
2015 
2016 
2017 
2018 
2019 

Mean 
-12.726 
-12.693 
-12.679 
-12.646 
-12.598 
-12.570 
-12.297 
-12.261 
-12.006 
-11.876 
-12.081 
-12.503 
-12.163 
-12.237 
-12.424 
-12.163 
-12.053 
-12.042 
-11.931 
-11.751 

Standard Deviation 
2.826 
2.835 
2.695 
2.720 
2.788 
2.788 
2.766 
2.570 
2.566 
2.620 
2.630 
2.756 
2.767 
2.789 
2.719 
2.546 
2.804 
2.886 
2.843 
2.525 

Table 49: Coefficients with of Percentage of “Yes” Votes as Independent Variable 

Variables 

PEOPLE_PER_FARM 
%DEMOCRAT 
%WHITE 
%BLACK 
%HISPANIC 
EDUCATION 
POVERTY_RATE 
HOUSEHOLD_INCOME_1000 
%CATHOLIC 
%EVANGELICAL PROTESTANT 
%MAINLINE PROTESTANT 
BALLOT_IMR 
CONSTANT 
Observations 
Note:  Coefficients  in  (1)  are  obtained  from  cluster  bootstrapping.  ***  p<0.01,  **  p<0.05,  *  p<0.1.  To  obtain  the 
coefficients in terms of the percentage of “yes” votes from our adjusted model, we must exponentiate to remove the 
natural log and rearrange terms. The rearranged equation is: 

(2) 
Transformed Coefficients 
0.500*** 
0.504*** 
0.502*** 
0.500*** 
0.502*** 
-0.498*** 
-0.494*** 
0.501*** 
-0.499*** 
-0.491*** 
0.507*** 
0.527*** 
0.657*** 

(1) 
OLS Coefficients 
0.000*** 
0.014*** 
0.008*** 
0.001*** 
0.008*** 
-0.009*** 
-0.023*** 
0.002*** 
-0.003*** 
-0.034*** 
0.030*** 
0.110*** 
0.650*** 

299 

𝑉) =

𝑒𝑥𝑝[(𝐼𝑀𝑅 × ϕ) + (𝐵* × ζ))]
1 + 𝑒𝑥𝑝[(𝐼𝑀𝑅 × ϕ) + (𝐵* × ζ))]

Equation 23 

238 

 
 
 
 
 
Table 50: Ballot Model Without MESR and IMRs 

VARIABLES 

PEOPLE_PER_FARM 

%DEMOCRAT 

%WHITE 

%BLACK 

%HISPANIC 

EDUCATION 

POVERTY_RATE 

HOUSEHOLD_INCOME_1000 

%CATHOLIC 

%EVANGELICAL PROTESTANT 

%MAINLINE PROTESTANT 

CONSTANT 

Unadjusted (no IMR) 
Ballot Model Coefficients 
<0.001 
(<0.001) 
0.014** 
(0.006) 
0.009** 
(0.004) 
-0.006 
(0.009) 
0.008 
(0.005) 
-0.006 
(0.007) 
-0.008 
(0.014) 
0.005 
(0.003) 
0.004 
(0.004) 
-0.037** 
(0.016) 
0.029*** 
(0.010) 
-0.881 
(0.622) 

Observations 
R-squared 
Note: Robust standard errors are in parentheses, *** p<0.01, ** p<0.05, * p<0.1. All variables are weighted by county 
population as a proportion of the total state population.  

299 
0.311 

239 

 
 
 
 
 
 
 
 
Table 51: Comparing Real World Outcomes to Ballot Model Predictions 

State 

Alabama 
Alaska* 
Arizona* 
Arkansas* 
California* 
Colorado* 
Connecticut 
Delaware 
Florida* 
Georgia 
Hawaii 
Idaho* 
Illinois* 
Indiana 
Iowa 
Kansas 
KentuckyR 
Louisiana 
Maine*R 
Maryland 
Massachusetts* 
Michigan* 
Minnesota 
Mississippi* 
Missouri* 
Montana* 
Nebraska* 
Nevada* 
New Hampshire 
New JerseyV 
New Mexico 
New York 
North Carolina 
North Dakota* 
Ohio* 
Oklahoma* 
Oregon* 

Reality 
No regulation in place 
No regulation in place 
Regulation in place 
No regulation in place 
Regulation in place 
Regulation in place 
No regulation in place 
No regulation in place 
Regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
Regulation in place 
No regulation in place 
Regulation in place 
No regulation in place 
Regulation in place 
Regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
Regulation in place 
No regulation in place 
Regulation in place 

Ballot Model 
Fail 
Pass 
Fail 
Fail 
Pass 
Fail 
Pass 
Pass 
Fail 
Fail 
Fail 
Fail 
Fail 
Fail 
Fail 
Fail 
Fail 
Fail 
Pass 
Pass 
Pass 
Pass 
Fail 
Fail 
Fail 
Fail 
Fail 
Pass 
Pass 
Fail 
Fail 
Fail 
Fail 
Fail 
Fail 
Fail 
Pass 

240 

 
 
 
Table 46 (cont’d) 
State 

Ballot Model 
Fail 
Pennsylvania 
Fail 
Rhode Island 
Fail 
South Carolina 
Fail 
South Dakota* 
Fail 
Tennessee 
Fail 
Texas 
Fail 
Utah* 
Pass 
Vermont 
Fail 
Virginia 
Pass 
Washington* 
Fail 
West Virginia 
Fail 
Wisconsin 
Fail 
Wyoming* 
26% 
All 50 States 
30.4% 
Ballot States 
74% 
Accuracy All 50 States 
Accuracy Ballot States 
75% 
Note: States with a (*) are states that allow ballot initiatives. (V) indicates that a FAW law passes by the legislature 
was vetoed by the governor and is therefore not in effect in reality. Shaded cells correspond to predictions that match 
what  exist  in  reality.  The  superscript  (R)  indicates  a  state  that  has  a  FAW  regulation  in  place  that  was  not  passed 
through a bill or ballot. 

Reality 
No regulation in place 
Regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
Regulation in place 
No regulation in place 
No regulation in place 
No regulation in place 
24% 
41.7% 
N/A 
N/A 

241 

 
 
 
 
 
APPENDIX A.5 FARM ANIMAL WELFARE DATA 
COLLECTION METHODS 

Stage 1: 

State legislature political party information was collected from state legislature websites. 
In some instances, these websites were incomplete, and these gaps were filled in by the lead author 
using a range of sources, including Ballotpedia.org, state and local newspapers, and political party 
websites. Industry size information for egg-laying hens was downloaded from USDA QuickStats 
and industry size information for gestating sows was downloaded from the Livestock Marketing 
Information Center (LMIC). 

Stage 2: 

for 

the 

file 

The 

2006 

Records 

Votes for State Initiative 
Arizona 2006, Proposition 204: Votes on Proposition 204 came from the Arizona’s Secretary of 
State 
Election 
General 
(https://apps.azsos.gov/election/2006/General/ElectionInformation.htm).  The 
“Official 
Election Results (PDF)” was downloaded on February 8, 2019. Data regarding the voting outcome 
of Proposition 204, by county, is found on page 15 of the document. 
California 2008, Proposition 2: Votes on Proposition 2 came from California’s Statement of Vote 
from  the  2008  General  Election  (  https://www.sos.ca.gov/elections/prior-elections/statewide-
file 
election-results/presidential-general-election-november-4-2008/statement-vote/). 
“Complete Statement of Vote (PDF)” was downloaded on January 11, 2019. Data regarding the 
voting outcome of Proposition 2, by county, is found on page 57 of the document. 
California 2018, Proposition 12: Votes on Proposition 12 came from California’s Statement of 
Vote  from  the  November  6,  2018  General  Election  (https://www.sos.ca.gov/elections/prior-
elections/statewide-election-results/general-election-november-6-2018/statement-vote/).  The  file 
“Complete Statement of Vote (PDF)” was downloaded on February 10, 2019. Data regarding the 
voting outcome of Proposition 12, by county, is found on page 98 of the document. 
Florida 2002, Amendment 10: Votes on Amendment 10 came from Florida’s Department of State 
2002 
results 
(https://results.elections.myflorida.com/downloadresults.asp?ElectionDate=11/5/2002&DATAM
ODE=). 
Massachusetts 2016, Question 3: Votes on Question 3 came from Massachusetts’ Secretary of 
the 
results 
General 
(http://electionstats.state.ma.us/ballot_questions/view/2741/). 
Ohio 2009, Issue 2: Votes on Issue 2 came from the Ohio’s Statement of Vote from the November 
3,  2009  General  Election  (https://www.sos.state.oh.us/elections/election-results-and-data/2009-
election-results/state-issue-2-november-3-2009/).  The  file  “State  Issue  2  Official  Results: 
November 3, 2009” was downloaded on February 10, 2019. 

Commonwealth 

Elections 

Election 

General 

2016 

Votes for State Legislation 
California  2010:  The 
https://leginfo.legislature.ca.gov/faces/billVotesClient.xhtml?bill_id=200920100AB1437%20  
Colorado  2008:  The 
http://www.leg.state.co.us/clics/cslFrontPages.nsf/PrevSessionInfo?OpenForm  

legislators’  votes  on 

legislators’  votes  on 

record  of 

record  of 

this  bill 

this  bill 

came 

came 

from 

from 

242 

 
 
 
 
 
 
came 

came 

came 

this  bill 

this  bill 

this  bill 

record  of 

record  of 

record  of 

record  of 

2007:  The 

legislators’  votes  on 

legislators’  votes  on 

legislators’  votes  on 

legislators’  votes  on 

from 
Michigan  2009:  The 
http://www.legislature.mi.gov/(S(45zoivtdskmlve2f2f03jh3a))/mileg.aspx?page=getObject&obje
ctName=2009-HB-5127  
Michigan  2018:  The 
from 
http://www.legislature.mi.gov/(S(rvjeyosf413g1xjdeb3o32sh))/mileg.aspx?page=GetObject&obj
ectname=2017-SB-0660  
Michigan  2019:  The 
from 
http://www.legislature.mi.gov/(S(trsczgqyfjcjfrm3ho5b0gni))/mileg.aspx?page=getObject&objec
tName=2019-SB-0174  
New  Jersey  2013:  The 
https://www.njleg.state.nj.us/bills/BillsByNumber.asp%20  
this 
of 
record 
Oregon 
https://olis.oregonlegislature.gov/liz/2007R1/Measures/Overview/SB600  
Oregon 
this 
of 
https://olis.oregonlegislature.gov/liz/2011R1/Measures/Overview/SB805  
Oregon 
this 
of 
https://olis.oregonlegislature.gov/liz/2019R1/Measures/Overview/SB1019  
Rhode  Island  2012:  The 
from 
http://webserver.rilin.state.ri.us/search/search.idq?CiRestriction=SB+2191&CiMaxRecordsPerPa
ge=25&CiScope=%2FJournals12%2F&CiSort=DocTitle%5Ba%5D&HTMLQueryForm=%2Fse
arch%2Fsearch%2Easp&Abstractt=1  
Rhode  Island  2018:  The 
http://webserver.rilin.state.ri.us/journals18/senatejournals18/senatejournals18.html  
Washington  2011:  The 
https://app.leg.wa.gov/billsummary?BillNumber=5487&Year=2011&Initiative=false  
Washington  2019:  The 
https://app.leg.wa.gov/billsummary?BillNumber=2049&Year=2019&Initiative=false 

legislators’  votes  on 

legislators’  votes  on 

legislators’  votes  on 

legislators’  votes  on 

this  bill  came 

this  bill  came 

this  bill  came 

this  bill  came 

this  bill  came 

2011:  The 

2018:  The 

record  of 

record  of 

legislators’ 

legislators’ 

legislators’ 

record  of 

record  of 

record 

record 

votes 

votes 

votes 

came 

came 

came 

from 

from 

from 

from 

from 

from 

from 

bill 

bill 

bill 

on 

on 

on 

all 

data 

Demographic Information 
File  Download  website 
For 
(https://www.census.gov/support/USACdataDownloads.html), all reference codes are found in the 
file, Mastdata.xls, under “Reference Information Files”. The data available closest to the vote in 
question was used for both counties and legislative districts. 

the  Counties  Data 

retrieved 

from 

Vote for Democratic Presidential Candidate 
Arizona  2006,  Proposition  204:  Votes  for  the  Democratic  ticket,  Kerry-Edwards,  in  the  2004 
Presidential election came from Arizona’s Secretary of State Record for the 2004 General Election 
(https://apps.azsos.gov/election/2004/Info/ElectionInformation.htm). 
California  2008,  Proposition  2:  Votes  for  the  Democratic  ticket,  Obama-Biden,  in  the  2008 
Presidential Election came from the California’s Statement of Vote Records for the 2008 General 
(https://www.sos.ca.gov/elections/prior-elections/statewide-%20election-
Election 
results/presidential-general-election-november-4-2008/statement-vote/). 
California  2018,  Proposition  12:  Votes  for  the  Democratic  ticket,  Clinton-Kaine,  in  the  2016 
Presidential election came from the California’s Statement of Vote Records for the 2016 General 
Election (https://elections.cdn.sos.ca.gov/sov/2016-general/sov/2016-%20complete-sov.pdf). 

243 

 
 
 
 
6, 

of 

on 

the 

2012 

Votes 

General 

November 

Florida  2002,  Amendment  3:  Votes  for  the  Democratic  ticket,  Clinton-Gore,  in  the  1996 
Presidential election came from Florida’s Department of State records for the November 5, 1996 
Abstract 
Election 
(https://results.elections.myflorida.com/Index.aspElectionDate=11/5/1996&DATAMODE=). 
*Note: The 1996 Presidential election was used opposed to the 2000 Presidential election due to 
the controversial results in Florida in 2000. 
Massachusetts 2016, Question 3: Votes for the Democratic ticket, Clinton-Kaine, in the 2016 
Presidential election came from the Massachusetts’ Secretary of the Commonwealth results for the 
2016  General  Election 
(https://elections.cdn.sos.ca.gov/sov/2016-%20general/sov/2016-
complete-sov.pdf). 
Ohio  2009,  Issue  2:  Votes  for  the  Democratic  ticket,  Obama-Biden,  in  the  2008  Presidential 
election  came  from  the  Ohio’s  Secretary  of  State  Results  for  the  November  4,  2008  General 
Election 
(https://www.sos.state.oh.us/elections/election-resuls-and-%20data/2008-election-
results/). 

Median Household Income 
Data  regarding  median  household  income  is  found  on  the  “Small  Area  Income  and  Poverty 
Estimates 
the  US  Census  Bureau  website 
on 
(https://www.census.gov/programs-surveys/saipe.html). 

Program” 

(SAIPE) 

page 

Percent of People all Ages in Poverty 
Data  regarding  poverty  is  found  on  the  “Small  Area  Income  and  Poverty  Estimates  (SAIPE) 
the  US  Census  Bureau  website  (https://www.census.gov/programs-
Program”  page  on 
surveys/saipe.html). 

Persons 25+ years of age with a Bachelor’s degree or higher 
Education  data 
(https://www.census.gov/support/USACdataDownloads.html). 

found  on 

is 

the  USA  Counties  Data  File  Download  website 

Race – White, Black, Hispanic 
Race 
on 
is 
(https://www.census.gov/support/USACdataDownloads.html). 

found 

data 

the  USA  Counties  Data  File  Download  website 

Religious Data – Catholic, Mainline Protestant, Evangelical Protestant 
Religious 
of  Religious  Data  Archives 
the  Association 
(http://www.thearda.com/QL2010/). Collect data from the “Percent” column for the 2010 religious 
census. 

came 

from 

data 

People per Farm 
Information on farm numbers per county was downloaded from USDA QuickStats. 

244 

 
 
 
 
 
 
 
 
 
 
APPENDIX A.6 CALCULATION OF FARM ANIMAL WELFARE 
INDUSTRY COSTS 

Eggs 

The  calculation  of  the  estimated  cost  to  the  egg  industry  of  updating  to  cage-free  egg 
production in states that do not currently have a FAW regulation in place utilized data from the 
United  Egg  Producers  (United  Egg  Producers,  2022),  USDA  Quick  Stats,  and  Matthews  and 
Sumner (2015). Data from 2017 (the most recent available) on the inventory of egg-laying hens in 
each  state  was  downloaded  from  USDA  Quick  Stats.  In  five  states  –  Arizona,  Connecticut, 
Delaware, Kansas, and Maine – the number of egg-laying hens was not provided. For these five 
states, we took the number of egg-laying hens in the nation overall, subtracted the total known 
from the 45 states that reported inventory numbers, and then divided the unaccounted for inventory 
evenly between these five states. According to the United Egg Producers, at the end of 2020, 28% 
of all egg-laying hens were in cage-free systems. We removed the percentage accounted for by the 
seven  states  that  have  a  cage-free  regulation  –  California,  Colorado,  Massachusetts,  Michigan, 
Oregon, Rhode Island, and Washington. We assumed that 100% of the egg-laying hens in these 
states were in cage-free systems. We then calculated the number of egg-laying hens in cage-free 
systems in the remaining 43 states without cage-free regulations and computed the new percentage 
of  egg-laying  hens  in  cage-free  systems.  This  updated  estimate  is  18.32%  of  egg-laying  hens, 
meaning 81.68% of all egg-laying hens are in conventional housing in the 43 states without a cage-
free  regulation.  This  percentage  of  conventionally  housed  hens  was  multiplied  by  the  total 
inventory in a given state to calculate the number of egg-laying hens in conventional housing.  

To estimate the cost of upgrading to cage-free production for these 81.68% of hens, we 
used estimated changes in cost of producing a dozen eggs under conventional versus cage-free 
systems from Matthews and Sumner (2015), scaled to the average number of eggs produced per 
hen  in  2020  and  inflated  to  2022  dollars.  According  to  the  United  Egg  Producers  (United  Egg 
Producers, 2022), on average a hen laid 296 eggs in 2020. Converting this number to dozens of 
eggs, we have the estimated cost of upgrading from conventional to cage-free production per egg-
laying hen, which is $6.95 per hen per year in 2022 dollars. We present the predicted percentage 
of  the  population  within  the  31  states  not  predicted  by  our  model  to  be  likely  to  pass  a  FAW 
regulation, the number of egg-laying hens in those states, the relative size of the states’ industry to 
the national total, and the estimated costs to the industry in each of these states to update to cage-
free egg production methods Table 52.  

Pork  

The calculation of the estimated cost to the pork industry of updating to crate-free pork 
production in states that do not currently have a FAW regulation in place utilized data from World 
Animal Protection (World Animal Protection, 2021), the Livestock Marketing Information Center 
(LMIC), Purdue University’s Center for Commercial Agriculture (Langemeier, 2019), and Ortega 
and Wolf (2018). Data from 2019 (corresponding to the year for the data used in our predictions) 
on the inventory of gestating sows in each state was downloaded from LMIC. 

According  to  World  Animal  Protection’s  “Quit  Stalling”  report  on  crate-free  pork 
production (World Animal Protection, 2021), at the end of 2020, about 25% of all gestating sows 
were in crate-free systems. We removed the percentage accounted for by the eleven states that 
have  a  cage-free  regulation  –  Arizona,  California,  Colorado,  Florida,  Maine,  Massachusetts, 
Michigan, Ohio, Oregon, Rhode Island, and Washington. We assumed that 100% of the gestating 

245 

 
 
 
 
sows in these states were in crate-free systems. We then calculated the number of gestating sows 
in crate-free systems in the remaining 39 states without crate-free regulations and computed the 
new  percentage  of  gestating  sows  in  crate-free  systems.  This  updated  estimate  is  18.67%  of 
gestating sows, meaning 81.34% of all gestating sows are in conventional housing in the 39 states 
without  a  crate-free  regulation.  This  percentage  of  conventionally  housed  gestating  sows  was 
multiplied  by  the  total  inventory  in  a  given  state  to  calculate  the  number  of  gestating  sows  in 
conventional housing.  

To estimate the cost of upgrading to crate-free production for these 81.34% of sows, we 
used estimated changes in cost per weaned pig under conventional versus two types of crate-free 
systems from Ortega and Wolf (2018), scaled to the average number of pigs produced per sow in 
2019  and  inflated  to  2022  dollars.  According  to  the  Purdue  University  Center  for  Commercial 
Agriculture (Langemeier, 2019), in 2019, on average a sow had two litters of piglets per year, with 
an average litter size of 11, for an estimated total of 22 piglets per sow per year. Converting to cost 
per sow, we have the estimated cost of upgrading from conventional to crate-free production per 
sow,  which  is  between  $51.53  and  $87.77  per  sow  per  year  in  2022  dollars.  We  present  the 
predicted percentage of the population within the 31 states not predicted by our model to be likely 
to pass a FAW regulation, the number of gestating sows in those states, the relative size of the 
states’ industry to the national total, and the estimated costs to the industry in each of these states 
to update to crate-free pork production methods in Table 53. 

246 

 
 
 
 
 
Table 52: Cost to Update to Cage-Free Egg Production in 31 States not Predicted to Pass a 
FAW Regulation 

State 

Alabama 
Arkansas* 
Georgia 
Hawaii 
Idaho* 
Illinois* 
Indiana 
Iowa 
Kansas 
Louisiana 
Minnesota 
Mississippi* 
Missouri* 
Montana* 
Nebraska* 
New Jersey 
New Mexico 
New York 
North Carolina 
North Dakota* 
Oklahoma* 
Pennsylvania 
South Carolina 
South Dakota* 
Tennessee 
Texas 
Utah* 
Virginia 
West Virginia 
Wisconsin 
Wyoming* 

Percent of 
Population in 
Favor of FAW 
Regulation 
21.451 
34.367 
35.904 
38.745 
42.278 
46.797 
23.841 
0.171 
5.630 
34.514 
11.274 
20.817 
11.958 
39.778 
N/A 
55.172 
38.939 
54.224 
10.165 
12.272 
0.345 
61.928 
25.678 
0.011 
20.968 
34.773 
52.484 
51.360 
39.490 
48.259 
5.369 

Number of 
Egg-Laying 
Hens 

Percent of Egg-
Laying Hens in 
the Nation 

Cost to Update to 
Cage-Free 
System 

7,867,738 
12,285,533 
17,966,521 
192,185 
472,192 
5,470,158 
26,354,377 
56,554,774 
3,249,703 
1,970,896 
10,849,607 
5,828,262 
11,306,386 
931,006 
7,353,761 
1,631,775 
102,020 
6,058,141 
14,160,452 
81,364 
3,354,460 
26,317,523 
4,002,121 
2,708,331 
1,986,321 
21,006,254 
4,480,850 
2,447,718 
1,215,655 
7,639,627 
29,550 

2.137 
3.336 
4.879 
0.052 
0.128 
1.485 
7.157 
15.358 
0.882 
0.535 
2.946 
1.583 
3.070 
0.253 
1.997 
0.443 
0.028 
1.645 
3.845 
0.022 
0.911 
7.147 
1.087 
0.735 
0.539 
5.704 
1.217 
0.665 
0.330 
2.075 
0.008 

$44,682,163.73 
$69,771,540.05 
$102,034,794.87 
$1,091,449.87 
$2,681,655.17 
$31,065,916.96 
$149,670,793.32 
$321,183,759.74 
$18,455,593.01 
$11,193,038.93 
$61,616,682.76 
$33,099,647.82 
$64,210,804.99 
$5,287,334.49 
$41,763,204.75 
$9,267,115.62 
$579,388.17 
$34,405,168.05 
$80,419,510.00 
$462,079.39 
$19,050,523.92 
$149,461,493.46 
$22,728,696.07 
$15,381,052.24 
$11,280,640.02 
$119,297,932.97 
$25,447,475.93 
$13,900,988.62 
$6,903,902.46 
$43,386,684.26 
$167,819.26 

$1,509,948,850.89 
265,875,261 
Total 
Note: States with a (*) are states that allow ballot initiatives. The predicted percent of the population in favor of a 
FAW  regulation  in  Nebraska  could  not  be  estimated  since  Nebraska  was  left  out  of  the  Stage  1  model  due  to  its 
unicameral state legislature. 

72.20 

N/A 

247 

 
 
 
 
 
 
 
 
 
 
Table 53: Cost to Update to Crate-Free Pork Production in 31 States not Predicted to Pass 
a FAW Regulation 

Percent of 
Population 
in Favor of 
FAW 
Regulation 

Number 
of 
Gestating 
Sows 

21.451 
34.367 
35.904 
38.745 
42.278 
46.797 
23.841 
0.171 
5.630 
34.514 
11.274 
20.817 
11.958 
39.778 
N/A 
55.172 
38.939 
54.224 
10.165 
12.272 
0.345 
61.928 
25.678 
0.011 
20.968 
34.773 
52.484 
51.360 
39.490 
48.259 
5.369 

15,312 
57,816 
21,114 
2,225 
8,238 
464,442 
280,559 
917,567 
174,810 
2,018 
572,545 
47,797 
334,240 
20,933 
391,551 
685 
318 
10,923 
896,231 
35,147 
425,387 
103,064 
9,195 
167,015 
15,466 
83,017 
16,842 
8,460 
1,362 
43,716 
16,842 

Percent 
of 
Gestating 
Sows in 
the 
Nation 
0.271 
1.025 
0.374 
0.039 
0.146 
8.232 
4.973 
16.263 
3.098 
0.036 
10.148 
0.847 
5.924 
0.371 
6.940 
0.012 
0.006 
0.194 
15.885 
0.623 
7.540 
1.827 
0.163 
2.960 
0.274 
1.471 
0.299 
0.150 
0.024 
0.775 
0.299 

Cost to Update 
to Crate-Free 
System Lower 
Bound 

Cost to Update 
to Crate-Free 
System Upper 
Bound 

$1,141,393.00 
$670,175.71 
$6,634,346.82 
$3,895,396.30 
$2,568,134.25 
$1,507,895.34 
$142,674.13 
$83,771.96 
$706,236.92 
$414,671.22 
$79,540,824.82 
$46,702,869.62 
$34,598,475.37 
$20,314,701.14 
$88,798,281.25  $151,234,572.76 
$25,253,320.17 
$14,827,637.53 
$28,534.83 
$16,754.39 
$85,247,789.83 
$50,053,748.16 
$6,348,998.57 
$3,727,852.37 
$72,193,107.37 
$42,388,613.50 
$4,565,572.01 
$2,680,702.83 
$54,858,201.16 
$32,210,319.95 
$57,069.65 
$33,508.79 
$28,534.83 
$16,754.39 
$370,952.73 
$217,807.10 

$78,955,075.55  $134,470,363.04 
$5,207,605.57 
$3,057,676.67 
$62,063,244.48 
$36,440,804.10 
$15,480,142.59 
$9,089,258.03 
$1,319,735.66 
$774,890.66 
$36,310,564.88 
$21,319,964.70 
$3,887,869.91 
$2,282,786.00 
$19,261,006.91 
$11,309,215.07 
$11,984,626.52 
$7,036,844.93 
$477,958.32 
$280,636.08 
$49,935.94 
$29,320.19 
$7,062,369.20 
$4,146,712.19 
$5,171,937.04 
$3,036,733.67 

State 

Alabama 
Arkansas* 
Georgia 
Hawaii 
Idaho* 
Illinois* 
Indiana 
Iowa 
Kansas 
Louisiana 
Minnesota 
Mississippi* 
Missouri* 
Montana* 
Nebraska* 
New Jersey 
New Mexico 
New York 
North Carolina 
North Dakota* 
Oklahoma* 
Pennsylvania 
South Carolina 
South Dakota* 
Tennessee 
Texas 
Utah* 
Virginia 
West Virginia 
Wisconsin 
Wyoming* 

Total 
Note: States with a (*) are states that allow ballot initiatives. The predicted percent of the population in favor of a 
FAW regulation in Nebraska could not be estimated since Nebraska was left out of the Stage 1 model due to its 
unicameral state legislature. 

$486,321,379.40  $828,266,099.29 

5,144,837 

91.19 

N/A 

248