DEVELOPMENT OF A NEWS SUBSCRIPTION MOTIVATION SCALE By Weiyue Chen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Information and Media - Doctor of Philosophy 2022 ABSTRACT DEVELOPMENT OF A NEWS SUBSCRIPTION MOTIVATION SCALE By Weiyue Chen As news organizations face accelerated loss in advertising revenue, increasing importance is placed on strategies to increase subscription sales. Although previous studies have found several predictors of paywall, willingness to pay, and paying for news research, these factors did not fit into one clear conceptual framework that links them together. In this dissertation, I aim to introduce a new construct, News Subscription Motivation, that provides theoretical linkages between different predictors of paying for news. Mixed method research was employed to conceptualize and operationalize this new construct. In Chapter 1, I discuss my thought process developing this study, the purpose of the study, and why this topic matters in the context of digital economy. Chapter 2 includes a review of previous research on what drives people to pay for news, and the literature on consumer decision-making processes, consumer decision-making styles, and consumer motivation in general. The need to develop a new construct and measurement tools that are specially designed for news consumption was also addressed. In Chapter 3, I conducted 22 in-depth interviews to generate possible dimensions of the construct, analyzed the qualitative data to propose a conceptual framework and definition. Study 1 results suggested nine possible dimensions: content utility, journalism quality, price, convenience, hitting the paywall, surveillance, being a good citizen, brand reputation, and journalism. Conceptual definitions of each dimension were also elaborated. Chapter 4 focuses on the operationalization of News Subscription Motivation. An initial items pool was generated based on Study 1. After the pilot test, I recruited two independent samples, and they were respectively subjected to Exploratory Factor Analysis and Confirmatory Factor Analysis. The final scale included six dimensions with 19 items, and this scale demonstrated robust model fit and adequate convergent and discriminant validity. Six dimensions of News Subscription Motivation were identified: supporting journalism, journalism quality, triggered by the paywall, community attachment, price, and content utility. In Chapter 5, I aim to establish the nomological validity of News Subscription Motivation. Factors extracted from Chapter 4 demonstrated statistically significant relationships with numbers of news subscription people report paying for, types of subscriptions people get, and individuals’ intention to maintain their primary subscriptions in the next 3 months, 6 months, and 12 months. Finally, I discuss the theoretical and practical implications of the scale of News Subscription Motivation in Chapter 6. This dissertation is dedicated to my grandmothers, Li Yuzhen and Ren Xuemei. Yuzhen was illiterate, and Xuemei never finished elementary school. Thank you for making higher education possible for me. iv ACKNOWLEDGEMENTS While this degree will be granted to me, I have not done this alone. I am in great debt to many people, which probably makes this section the easiest one to write. I have been blessed with a wonderful committee who demonstrate exemplary research and mentorship. Not only they care about my growth as a researcher, they also guided me in many ways to develop as a person. I am privileged to learn from them, be supported and encouraged by them, and hope to pass their intellectual legacy and love to my students. My advisor, Dr. Esther Thorson, is far beyond what one can possibly ask for an advisor. From offering personal financial support for me to attend ICPSR to babysitting my dissertation writing process, she has been actively creating opportunities for me to grow and succeed. I have no doubt she would legally adopt me if that’s what it takes for me to finish this program. Seeing her doing interdisciplinary research at the front row, I am also very proud to be part of her legacy and so fortunate to have her as a role model in life. I would like to further extend my appreciation to my committee members: Drs. Serena Miller, Stephen Lacy, and Johannes Bauer. I could thank them individually here, but all of them have supported me in every possible way: timely and helpful feedback that advance my work to the next level, flexible deadlines at the expense of their own convenience, individual coaching that tailored to my own research interests, and personal insights on well-being and managing stress. Every time I encounter a struggling student, I can always find an example in how they have helped me. Having a chorionic health issue during doctoral studies has not been easy. I could not make it in this far if my health providers are not incredibly good at their job. I am grateful for v Caroline Silvia, Dr. Leigh White, Dr. Debra Duxbury, and Anne Buffington for lending me their assistance throughout my time at MSU. This degree reflects their dedication as well. My RA and TA supervisors have also shaped me into who I am today. Dr. Rachel Mourão has taught me the structures of writing and organization of multiple projects. I have benefited course designs and teaching styles from Drs. Kjerstin Thorson, Perry Parks, Howard Bossen, and Professor Amy Haimerl. Dr. Lucinda Davenport also advised me on summer research plans and helped me develop my first course. I am forever grateful for their mentorship and guidance. I have also received countless support from my friends and communities. Yingqian and Jing had significant impact on me for navigating challenges in life; LC, Ming, Na Rae, and Mia constantly sent me care packages and offered me shoulders to lean on; Jessica, Alyssa, Huei, prayed for and encouraged me to confront my fears; Lin and Mel offered me insights surviving grad school; and CK, Mengyan, Ran, and Sevgi shared their tips that helped me to better plan my steps in this program. I am also grateful for my friends from #OneInJC, MSU Bridges International, #Bethesda, Lansing Chinese Christian Church, Riverview Grad Life Group, College Park Church Indy, and Faculty Commons at Butler. I could not imagine bringing my unfinished dissertation to an institution other than Butler University. My colleagues, Drs. Brooke Barnett, Margaretha Geertsema-Sligh, Lee Farquhar, and Rose Campbell have offered me immense support in the past three semesters. When I was lost in the meaning of journalism business, my students also inspired me to carry on this project. Their intellectual curiosity and faith in quality journalism motivated me to move forward. vi I am also grateful for my parents. This is not the path they have hoped and planned for me, but they have showered me with unwavering encouragement and support in the past five years. I cannot wait for my turn to support them. Finally, thanks be to God, who is the ultimate source of meaning and grace. This work, in my opinion, is not there yet to “glorify Him.” But I would like to see it as an offering to God, acknowledging His love and faithfulness. vii TABLE OF CONTENTS LIST OF TABLES .......................................................................................................................... x LIST OF FIGURES ....................................................................................................................... xi CHAPTER 1: INTRODUCTION ................................................................................................... 1 Study Purpose ............................................................................................................................. 1 Why Subscription Motivation Matters in the Digital Economy ................................................. 3 REFERENCES ............................................................................................................................... 7 CHAPTER 2: THEORETICAL BACKGROUND ...................................................................... 10 Previous Research on Paying for News .................................................................................... 10 Consumer decision-making process ......................................................................................... 11 Consumer Decision-making Style ............................................................................................ 12 High quality-consciousness/perfectionism. .......................................................................... 14 Price/value consciousness. .................................................................................................... 14 Brand consciousness. ............................................................................................................ 14 Novelty/fashion consciousness. ............................................................................................ 15 Recreation consciousness...................................................................................................... 15 Impulsive/careless consumer. ............................................................................................... 15 Habitual, brand-loyal consumer. ........................................................................................... 16 Confused by over-choice. ..................................................................................................... 16 Consumer Motivation ............................................................................................................... 16 Proposing a New Construct: News Subscription Motivation ................................................... 18 REFERENCES ............................................................................................................................. 19 CHAPTER 3: CONCEPTUALIZING NEWS SUBSCRIPTION MOTIVATION...................... 23 Method ...................................................................................................................................... 23 Participants and Sampling......................................................................................................... 23 Summary of interview participants. ...................................................................................... 25 Procedure .................................................................................................................................. 26 Preparation of Interview Data ................................................................................................... 28 Thematic Coding....................................................................................................................... 28 Results ....................................................................................................................................... 31 Content Utility. ..................................................................................................................... 31 Journalism Quality ................................................................................................................ 32 Affordability ......................................................................................................................... 33 Convenience.......................................................................................................................... 34 Hitting the Paywall. .............................................................................................................. 34 Surveillance........................................................................................................................... 35 Being A Good Citizen........................................................................................................... 35 Brand Reputation. ................................................................................................................. 36 Supporting Journalism. ......................................................................................................... 37 Construct Development/Definitions ......................................................................................... 38 viii APPENDIX ................................................................................................................................... 41 REFERENCES ............................................................................................................................. 46 CHAPTER 4: OPERATIONALIZING NEWS SUBSCRIPTION MOTIVATION .................... 49 The Initial Process of Scale Development ................................................................................ 49 Existing scales....................................................................................................................... 49 Item generation. .................................................................................................................... 50 Expert feedback and pre-test. ................................................................................................ 50 Scale response categories and lead-in question .................................................................... 50 Pilot Study................................................................................................................................. 53 Discussion of Pilot Study Results. ............................................................................................ 62 Exploratory Factor Analysis (Study 2) ..................................................................................... 64 Sample................................................................................................................................... 64 Verifying data quality. .......................................................................................................... 66 Determining optimal number of factors. ............................................................................... 66 Item reduction. ...................................................................................................................... 67 Final items............................................................................................................................. 68 Discussion of Exploratory Factor Analysis Results.................................................................. 72 Scale Validation: Confirmatory Factor Analysis (Study 3) ...................................................... 74 Sample and procedure. .......................................................................................................... 74 Results of CFA...................................................................................................................... 75 Convergent and discriminant validity ................................................................................... 78 Conclusion ................................................................................................................................ 80 APPENDICES .............................................................................................................................. 81 Appendix A: Survey - Exploratory Factor Analysis (Study 2) ................................................. 82 Appendix B: Survey - Confirmatory Factor Analysis (Study 3) .............................................. 88 REFERENCES ............................................................................................................................. 95 CHAPTER 5: TESTING NEWS SUBSCRIPTION MOTIVATIONS AS PREDICTORS ...... 100 Research Questions ................................................................................................................. 100 Dependent Measures ............................................................................................................... 101 Results ..................................................................................................................................... 103 Discussion ............................................................................................................................... 108 REFERENCES ........................................................................................................................... 111 CHAPTER 6: GENERAL DISCUSSION .................................................................................. 113 Discussion of the Scale Development Results ........................................................................ 113 Discussion of NSM as Predictors of News Paying Behavior ................................................. 117 Explained Variance. ............................................................................................................ 118 Effect size............................................................................................................................ 119 Predictor Evaluation............................................................................................................ 119 Theoretical Implications ......................................................................................................... 122 Practical Implications.............................................................................................................. 124 Limitations .............................................................................................................................. 126 REFERENCES ........................................................................................................................... 127 ix LIST OF TABLES Table 1. Overview of demographics of interview participants ..................................................... 27 Table 2. Frequency of news subscription motivations among interview participants .................. 30 Table 3. Prior subscales developed in scholarly research ............................................................. 49 Table 4. Proposed items for developing news subscription motivation scale .............................. 52 Table 5. Summary of demographics of participants in the pilot study ......................................... 55 Table 6. EFA results for pilot study with factor loadings above .50 ............................................ 57 Table 7. EFA results for pilot study with factor loadings above .70 ............................................ 60 Table 8. Revised items for content utility dimension ................................................................... 63 Table 9. Respondents’ demographics from the API (2018) study (N=4,113) .............................. 65 Table 10. Summary of participants in Study 2 (N=417)............................................................... 65 Table 11. Exploratory factor analysis results (final scale) ............................................................ 69 Table 12. Correlations between factors extracted from EFA ....................................................... 71 Table 13. Summary of participants in study 3 (N=506) ............................................................... 75 Table 14. Similar subscales for factors of news subscription motivation..................................... 78 Table 15. Descriptive statistics of dependent variables .............................................................. 102 Table 16. Predicting the monetary amount subscribers are paying for news subscriptions ....... 103 Table 17. Predicting the quantity of publications subscribers are paying for ............................. 104 Table 18. Predicting subscribers' retention in the future ............................................................ 107 Table 19. Predicting paying for national or local news subscription .......................................... 108 x LIST OF FIGURES Figure 1. Plot from parallel analysis ............................................................................................. 56 Figure 2. Q-Q plot from multivariate normality test..................................................................... 67 Figure 3. Confirmatory factor analysis results .............................................................................. 77 xi CHAPTER 1: INTRODUCTION Study Purpose During doctoral studies at MSU, my work has mostly explored the reasons that lead to people paying for news subscriptions (e.g., Chen & Thorson, 2021). This research surveyed the a representative sample of U.S. population, compared the effects of several variables on amount of paying for news, and surprisingly, found that the perceived content quality of news and perceived societal value of news do not significantly affect what people pay for news. This finding suggests that, when it comes to paying for news subscriptions, people are not motivated by either concern for “news content quality” dimensions as they are defined by journalists, or for appreciation of the role that news plays in a democracy, such as representing the public and serving as the watchdog for the society. The evaluation of the news product, either its informational function or social impact, is surprisingly irrelevant to people's consumption decision of subscribing to news sources. Thus it is important for scholars to examine the decision-making process for paying for news subscriptions and explore other mechanisms motivating people’s paying for news subscriptions. For this reason, I intend to conduct a marketing-type study that aims to test whether a typology of “consumer decision styles,” which was developed as a prediction model and can be used to predict who will or will not pay for news. The importance of news organizations figuring out what/who will pay for news is three-fold. First, the market of news, especially in a digitally based environment, is largely affected by network effects (Knieps & Bauer, 2016). The increasing number of subscribers will attract more advertisers, and it is found that circulation/subscription revenue is positively related to advertising sales (Chen, Thorson, & Lacy, 2004). Second, the price of digital products/services will tend toward zero in the digital 1 economy, and that makes it harder for paywalled news organizations to compete with their free alternatives. Search costs, reproduction costs, transportation costs, tracking costs, and verification costs will drop significantly and probably approach zero (Goldfarb & Tucker, 2019). The trivial cost further makes it possible for companies to set their price lower or even provide products/service for free. Hence it is important to understand what factors motivate people to pay for news subscriptions. Third, given news organizations also serve as “the fourth estate” in the society, increases in subscription revenue are not only crucial for news organizations’ financial performance, but survival of these organizations may be critical from a societal and democratic functioning point of view (Schultz, 1998, p. 51). Previous studies have found various predictors of paying behavior or willingness to pay for news, but there are no linkages between these factors from theoretical perspective. Predictors of paying behavior or willingness to pay include: individual characteristics such as demographics (Chyi, 2012; Chyi & Lee, 2013); news use related factors such as frequency (Chyi & Lee, 2013), habit strength (Chyi & Lee, 2013; Chen & Thorson, 2021), and format preference (Berger et al., 2015); price/cost/value of the news publication (Reuters Institute, 2020; Fletcher & Nielsen, 2017); content value, comprehensiveness, and quality (American Press Institute, 2018; Reuters Institute, 2020; Li & Thorson, 2015); attachment to the local community (Olsen, 2020; Goyanes, 2020); convenience of accessing news online or in print (American Press Institute, 2018; Reuters Institute, 2020); altruism motivations such as a belief in the value of funding good journalism and helping news organizations’ get through the current financial crises (Goyanes, 2020; Reuters Institute, 2020); getting the latest update on what’s going on in the world (News Media Alliance, 2019); continuance of reading when hitting a paywall (American Press Institute, 2018); and desire to follow a particular journalist (American Press Institute, 2018). 2 Additionally, none of these studies have thoroughly examined the decision-making processes involved in deciding to consume news and whether and how much to pay for it. In their widely-cited model, Engel, Kollat, and Blackwell (1978) argue that consumers’ decision- making processes generally include five different stages: problem recognition, search, evaluation of alternatives to chosen brands, the purchase act, and evaluation of the purchase act after it occurs. However, this basic framework does not include some external factors that might influence the process. For example, a habitual user might automatically purchase a product without going through the decision-making process. Another crucial factor is individual characteristics, as they determine the way people approach purchasing a certain product. The discrepancies in individual characteristics influence people's unmet wants and needs, and further affect activities within all five stages of decision- making processing (Engel et al., 1978). Hence, it is essential to understand individual consumer characteristics in order to understand them as they play a role in Engel et al’s (1978) five-stage model. Why Subscription Motivation Matters in the Digital Economy The need to focus on subscriptions became an urgent matter as news organizations experienced drastic loss of revenues, first due to the loss of classifieds to Craigslist, and more recently to losses due to losing the competition with digital advertising. In 2021, Internet giants (Google, Amazon, Yahoo, Microsoft) and social networking companies (Facebook, YouTube, LinkedIn) consumed more than 70% of the digital ad sales in the U.S (eMarketer, 2021). The rise of specialization websites also negatively affected newspapers’ revenue from classification ads, and further, the overall financial status of the organizations. In Western Europe, the loss of classified ads accounted for 44% of news organizations’ revenue decline. It has become clear 3 that advertising and classifieds can no longer be relied on to support newspaper companies, and that at least so far, the best alternative appears to be requiring users to pay the true cost of news content. Therefore, understanding what drives people to purchase news subscription and further boosting the subscription revenue is paramount for news organizations to survive their financial crisis. To understand the importance of subscription revenue, one must also understand how media organizations operate in multi-sided markets and the impact of network effects. News organizations, like many other media organizations, essentially operate in multi-sided markets. That is, they serve at least two distinct user groups: 1) audiences of news content, and 2) advertisers who are targeting the news publication’s audience. While previous research has established the positive correlations between subscription and advertising revenue (Chen, Thorson, & Lacy, 2005), the case is even more so in the context of digital economy. According to Reuters Institute, about seven out of ten leading newspapers in the United States and the European Union are creating paywalls online (Simon & Graves, 2019). Thus a growing number of news organizations are operating and competing in digital multi-sided markets, in which the direct and indirect network effects are of great importance (Knieps and Bauer, 2016). The growing significance of network effects also motivates companies to be more responsive to advertisers rather than users (audiences). Consumer sovereignty is still important, although it is indirect in multi-sided markets (Anderson, 2012). Yet, media companies still need to refine their products/services to first attract larger audiences, and then sell those audiences to advertisers. For news organizations, it means they need to focus on what value they are providing to their audiences, getting them to subscribe, and then eventually serving more and/or loftier advertisers. In other words, subscriptions revenues do not hurt, but rather help advertising 4 sales and the overall financial performance of the news organization. Thus, the exploration of what drives individual motivation to subscribe is important. Getting more people to subscribe also taps into strategies of competition in the digital economy: 1) price competition would not work in the long run; and 2) quality competition will be vital. Price competition is problematic for news organizations because the price of digital products/services is significantly lower, as advertisers can provide cross-subsidies and digital technology has significantly reduced costs. Anderson (2009) argues that as it takes nothing for digital technology to reproduce and distribute ads (marginal costs equals zero). Goldfarb and Tucker (2019) also suggest that, in the digital economy, search costs, reproduction costs, transportation costs, tracking costs, and verification costs will drop significantly and eventually approach zero. The trivial cost further makes it possible for companies to set their price lower or even provide products/service for free. Chances to survive and thrive through price competition are slim. Then, moving to an alternative direction, competition among news organizations will be heavily focused on quality of the product/service. Owen and Wildman (1992, pp.93) note that in the advertiser-supported model, television networks attract audiences through product competition, not price competition. The goal is to use quality news products to attract audiences, and then sell those audiences to advertisers. This conclusion is also suitable for digital multi- sided market of news, while news organizations have been relying on advertising revenue for decades. Then the strategic focus of companies would be how to create and deliver unique value for their consumers by improving the quality of their products and services (Kranz and Picot, 2016). From the consumers’ perspective, the quality of news products is defined as the ability to 5 meet consumers needs and wants (Lacy, 1989), the present study begins the exploration of what needs can be fulfilled through news subscription. 6 REFERENCES 7 REFERENCES Anderson, C. (2009). Free: The future of a radical price. Random House. Anderson, S. (2012). Advertising on the Internet. The Oxford handbook of the digital economy, 355-396. Chen, R., Thorson, E., & Lacy, S. (2005). The impact of newsroom investment on newspaper revenues and profits: Small and medium newspapers, 1998–2002. Journalism & Mass Communication Quarterly, 82(3), 516-532. Chen, W., & Thorson, E. (2021). Perceived individual and societal values of news and paying for subscriptions. Journalism, 22(6), 1296-1316. eMarketer. (2021). Digital Ad Revenue Share, by Company, U.S. eMarketer, Retrieved from: https://forecasts- na1.emarketer.com/584b26021403070290f93a4a/5851918a0626310a2c1869c2 Engel, J. F., Kollat, D. T., & Blackwell, R. D. (1978). Consumer behavior, 3rd ed. Hinsdale, IL: Dryden. Goldfarb, A., & Tucker, C. (2019). Digital economics. Journal of Economic Literature, 57(1), 3- 43. Knieps, G. & Bauer, J. M. (2016). The industrial organization of the Internet. In Bauer, J. M., & Latzer, M. (eds). Handbooks on the economics of the Internet. Edward Elgar Publishing, UK. Krans, J. J., & Picot, A. (2016). Internet Business Strategies. In Bauer, J. M., & Latzer, M. (eds). Handbooks on the economics of the Internet. Edward Elgar Publishing, UK. Lacy, S. (1989). A model of demand for news: Impact of competition on newspaper content. Journalism Quarterly, 66(1), 40-48. Li, Y., & Thorson, E. (2015). Increasing news content and diversity improve revenue. Newspaper Research Journal, 36(4), 382-398. Owen, B. M., & Wildman, S. S. (1992). Video economics. La Editorial, UPR. Schultz, J. (1998). Reviving the fourth estate: Democracy, accountability and the media. Cambridge University Press. Simon, F. & Graves, L. (2019, May 23). “Newspaper Paywalls Slowly Increasing, but Online News is Still Mostly Free.” Reuters Institute, Retrieved from: 8 https://reutersinstitute.politics.ox.ac.uk/risj-review/newspaper-paywalls-slowly- increasing-online-news-still-mostly-free 9 CHAPTER 2: THEORETICAL BACKGROUND Previous Research on Paying for News At the national level, various factors were found to be related to paying behavior or willingness to pay for news. Age is the first significant factor – younger people show higher paying intent for online news. Men tend to have higher paying intent than women, and people who are more interested in news and who use more online news are more likely to pay (Chyi, 2012; Chyi & Lee, 2013). Preference for online formats also leads to higher news paying intent (Chyi & Lee, 2013) and customers’ willingness to pay (Berger et al., 2015). Higher prices charged for news creates a negative impact on willingness to pay (Berger et al., 2015), and people are more likely to pay for online news when they perceive the price to be relatively “inexpensive” (Fletcher & Nielsen, 2017). Fletcher and Nielsen (2017) also found that higher usage of public service media is positively associated with paying for online news. It is important to note that given subscriptions to the print newspaper are mostly bundled with access to digital content, the distinction of paying for what content format has considerably weakened. Other than these predictors, Chen and Thorson (2021) find that habit strength of news use, social identity motivations, and entertainment spending are also associated with how much people pay for news. Research about what prompts people to pay for news have also been examined in the local context. Olsen (2020) expanded on the concept of perceived worthiness of news and readers’ perception of proximity. She suggests that the perceptions of news and willingness to pay for news vary among local communities. For example, those who see local news as a vital part of community life are more likely to pay for news subscriptions than those who think local news is “over-exaggerating insignificant events.” Additionally, Goyanes (2020) finds that 10 audiences’ perceptions of the financial state of local news organizations, journalists’ engagement with local communities, and readers’ attachment to local community, users' local content creation, and social media use are highly predictive of paying for local news subscriptions. From the perspective of current news subscribers, several studies have investigated why people pay for news (American Press Institute, 2018; New Media Alliance, 2019; Reuters Institute, 2020). Although these analyses are descriptive in nature, they do offer possible explanations of why individuals decide to purchase news subscriptions. News subscribers reported numerous reasons of paying: 1) Community - the community connection of accessing local news and supporting local news organizations; 2) Content - obtaining interesting, useful, and valuable content; 3) Convenience - convenience of accessing news and digital payments; 4) Cost – getting a cheaper price or a discount; 5) Response to paywalls – Being able to continue to read when the content is locked behind a paywall; 6) Altruism – funding and supporting good journalism; and 7) Getting the latest updates. These factors, however, have been reported in an organic and exploratory way. How these factors could be linked together from a theoretical view is still unknown. In other words, which of these factors are most predictive? Little research has investigated the overall research question, “what makes people pay for news,” utilizing theories in the marketing field, let alone considering consumers’ decision-making process. This dissertation aims to begin that exploration. Consumer decision-making process The most cited model for consumer decision-making process is the five-stage model proposed by Engel, Kollat, and Blackwell (1978). They suggest that when a consumer decides to make a purchase, he/she typically go through five different stages: problem recognition, search, 11 evaluation of alternatives to chosen brands, the purchase act, and evaluation of the purchase act after it occurs. At the problem recognition stage, people realize their need for a certain product that can mitigate or solve a focal problem. Next, hoping to satisfy the focal need, people search for information about available and suitable products. The evaluation stage is when consumers compare different options they have, and come up with what they think is the best (s). At the purchase stage, consumers go ahead and purchase the selected choice. During the after-purchase phase, consumers generate new perceptions and attitudes of the product based on their personal experience with it, creating perceptions and evaluations that will influence their repurchase intentions and behaviors. To focus more on the stage of satisfaction of consumer needs, Darley, Blankson, and Luethge (2010) extended the EKB model by adding external factors that might influence consumers’ decision-making process. They argue that the external factors can be divided into four categories: 1) individual differences such as personality, motives, and lifestyle; 2) social influences such as culture, class, and subgroups; 3) situational and economic factors; and 4) relationships to the online shopping environment. They argue that these possible external factors are important and therefore they are included here as we develop a model for news consumer’s decision-making processes. Consumer Decision-making Style Consumer Decision-making Style is one of the marketing concepts that taps into the external factors of the decision-making process. To better understand and profile individual consumer characteristics, Sproles and Kendall (1986) developed a Consumer Styles Inventory, which identified eight types of consumer styles. The scale was adapted and tested in multiple countries, among different age generations, and across several product categories (e.g., Lysonski, 12 Durvasula, & Zotos, 1996; Bakewell & Mitchell, 2003; Zhou, Arnold, Pereira, & Yu, 2010; Kang, Johnson, & Wu, 2014). In general, it does a good job of predicting purchase intention (Prakash, Singh, & Yadav, 2018). Through the conceptual lens, Sproles and Kendall (1986) define decision-making style as the mental orientation when individuals face a consumption decision. This mental orientation motivates individuals to look for affective and cognitive characteristics in the product before and during shopping events. Therefore, consumer style can be a determinant of the consumption decision. For example, while shopping for cars, a quality-driven consumer would look for cars that function well, while a fashion-driven consumer would prefer cars with innovative design and exterior. In sum, different consumer styles reflect people’s preferences and attributes they value, which in turn influence their purchase decision. Assumed in this approach is that consumers perceive shopping with different mental orientations that preexist the consumption process. Sproles and Kendall (1986) argue that consumer style is similar to one’s “personality” in shopping. The notion of personality suggests that consumers could have some fundamental decision-making modes that make them approach shopping differently, thus creating variations in consumer preferences and tastes, which further lead to differences in purchase decisions. Essentially, determinants of the consumption decision are relevant to individuals’ discrepancies in their inclinations towards purchasing products. Although Sproles and Kendall (1986) claim that these consumer styles are basic characteristics of individuals, predispositions can vary with the nature of the product. Attributes that are important for one product may not be identical for the another. Moreover, the cost of the product could also affect consumers’ inclinations towards different attributes of the product. For instance, 13 consumers might value quality, function, and sustainability while buying a car that cost $25,000, but might prioritize novelty and aesthetic attributes over quality while buying a $20 dress. As well as defining the concept of consumer style, Sproles and Kendall (1986) also development a measurement scale for various kinds of consumer styles. The instrument, Consumer Style Inventory, contains 40 original items and was first tested with a sample of the U.S. general population. Factor analysis of the items identified eight basic characteristics of consumer decision-making styles: High quality-consciousness/perfectionism. This decision style emphasizes product quality. The present study follows Lacy’s definition (1989) and refers to quality as the ability to fulfill the needs and wants of consumers. “Needs” refers to individuals’ physical or psychological necessity and “wants” indicates consumers’ conscious desire for something (Lacy, 1989). Consumers with this trait tend to pursue what they think is the best possible product and are reluctant to settle for something that does not meet their expectations (Kamaruddin & Mokhlis, 2003). Price/value consciousness. This consumer style indicates consumers’ desire to get low prices whenever it is possible. Price/value-consciousness consumers are sensitive to price changes and are more likely to respond to pitches for lower price and to accept discount prices when they are offered. They also tend to be comparison shoppers who aim to get a relatively low price for the choice they make. Brand consciousness. Brand consciousness refers to individuals’ emphasis on getting a well-known brand while making purchase decisions. Consumers with this trait prefer to shop at department stores where expensive and high-end brands are prevalent (Sproles & Kendall, 1986). This tendency of getting well-known brands is often associated with one’s desire to use brands to 14 demonstrate his or her social status (Nelissen & Meijers, 2011). Brand-conscious consumers also regard a higher price as indicative of better quality. Consequently, they tend to buy the best- selling and well-advertised products (Kang et al., 2014). As this definition indicates brand consciousness is both a heuristic for quality and desire to impress others for identity purposes, this factor might correlate with other factors. Novelty/fashion consciousness. This decision style represents consumers’ preference for discovering and purchasing innovative and novel products. Consumers with this trait usually take pleasure in keeping up with new styles and trends (Bakewell and Mitchell, 2003). Sproles and Kendall (1986) argue that novelty/fashion consciousness is fundamentally motivated by people’s motivation for variety seeking. Recreation consciousness. Recreation-conscious describes consumers who take pleasure from shopping and enjoy the stimulation of browsing and picking up products (Sproles & Kendall, 1986; Bakewell & Mitchell, 2003). This decision-making style suggests that shopping, in its very nature, is pleasant and entertaining. Marketing scholars also suggest that people who have higher hedonic tendencies enjoy browsing new products and obtaining new information while they are shopping (Kim & Eastin, 2012). Since recreation conscious consumers enjoy the process of shopping, they are more likely to take pleasure in single use purchase rather than habitual buying. Impulsive/careless consumer. This decision style describes consumers who do not plan their shopping and are not concerned with how much money they spend (Sproles & Kendall, 1986). Bakewell and Mitchell (2003) also argue that consumers with this trait are more likely to show buyers’ remorse. 15 Habitual, brand-loyal consumer. According to the original definition (Sproles & Kendall, 1986), this consumer style is a characteristic that involves repetitively choosing the same favorite brands and stores. Consumers with this trait tend to buy the same brand and go to the same product provider again and again, and they are very less likely to withdraw their financial commitments. Confused by over-choice. Confused by over-choice describes individuals who have difficulty making choices when multiple options are available to them. This decision style stems from consumers’ lack of confidence or inability choose and distinguish the numerous choices are available to them. Consumers with this style tend to experience information overload (Bakewell and Mitchell, 2003) and believe it is impossible to select products since there are too many choices (Kang et al., 2014). One caveat of the Consumer Style Inventory, however, is that the scale development process never included qualitative prior analysis. While the concept of consumer style does reflect external factors of the consumer decision-making process, the construct validity is relatively weak without justifying why these eight dimensions are included and none others are. Moreover, the scale was designed for the general context of shopping, and was developed before digitalization. Thus, direct use of this scale would likely be problematic in today’s research on news consumption. Consumer Motivation Another important concept to take into consideration is consumer motivation. Ryan and Deci (2000) argue that motivation is fundamental in psychology research as it is “at the core of biological, cognitive, and social regulation.” In early consumer behavior research, motivation was the explanatory concept for why people buy (Britt, 1950), and was defined as the drives and 16 desires that lead to behavioral change (Bayton, 1958). In recent studies, Pincus (2004) emphasized how motivations stem from consumers’ unmet needs. Mallalieu and Nakamoto (2008) contend that motivations are driven by consumers’ desire to achieve specific goal(s). Other scholars have also highlighted the influence of fundamental motives and evolutionary needs on consumer behavior (Griskevicius & Kenrick, 2013; Schaller et al., 2017). While a conceptual definition was not provided in the publication, Barbopoulos and Johansson (2016) note that consumer motivation is driven by the gain goal, the hedonic goal, and the normative goal. Three quantitative studies were conducted in Sweden to develop a scale. Participants were asked to answer the question “When you [insert consumption context and product here], how important it was for you too...” and rate each provided statements on a 5- point Likert scale. The results demonstrated five dimensions of consumer motivation. The first one is labeled as thrift, which reflects consumers desire to get the cheapest price possible. Safety is the second dimension that describes consumers’ goal to guarantee financial security and psychological well-being. Instant gratification reflects consumers’ urge to have immediate satisfaction and comfort. The last two dimensions moral and social norms, represent consumers’ aim to act within moral convictions and social conventions. Barbopoulos and Johansson’s (2016) scale of consumer motivation is limited in two major ways. First, although a literature review was conducted, the initial dimensions for developing the scale were not justified by qualitative research. Second, this study was executed in largely different contexts from news consumption in the U.S. The participants were directed to think about their shopping experiences when getting groceries, deciding on times used for leisure, and making financial savings and investments. The survey instruments were originally 17 presented and tested in Swedish, and were later translated into English at the publication stage. Hence, the consumer motivation is likely not suitable to be directly used in news consumption surveys among Americans. Proposing a New Construct: News Subscription Motivation Although the overviewed scales of consumer motivation and consumer style provide instruments to survey news consumers, these scales have significant limitations. First, neither scale was initiated by the guidance of subjected qualitative research, thus their construct validity, and how these dimensions emerged and connected are problematic. Second, neither scale has been tested in the context of news consumption. Unlike other commodities, news products also have features of public goods, and offer societal value (Doyle, 2013). Thus, a new scale is needed, and development of this scale should include qualitative research and be specifically designed for news consumers. The present study is intended to apply to a narrower focus with two more distinctions. First, antecedents of purchasing (buying) a product and donation (charitable) behaviors can be distinct (e.g., Park & Kim, 2003; Zhang, Cai, & Shi, 2021). In this project, I focus only on consumers’ purchase of news subscription, rather than people’s donation behaviors when it comes to news. Second, a distinction is important to be made for existing consumers and non- consumers, since significant attitudinal and behavioral differences can be found among these two groups (e.g., Diamantopoulos, Schlegelmilch, & Allpress, 1990; Sohail & Al-Jabri, 2014). The present study focuses on people who are already buying or have bought news subscriptions in the past. Built on this, I thus propose a new construct: News Subscription Motivation, which examines news subscribers’ motivation to purchase news subscription(s). The next chapter will discuss the conceptualization of this new construct. 18 REFERENCES 19 REFERENCES American Press Institute. (2018). Paths to Subscription: Why recent subscribers chose to pay for news. American Press Institute. Retrieved from: https://www.americanpressinstitute.org/publications/reports/survey-research/subscribers- appendix-1/ Bakewell, C., & Mitchell, V. W. (2003). Generation Y female consumer decision‐making styles. International journal of retail & distribution management. Barbopoulos, I., & Johansson, L. O. (2016). A multi-dimensional approach to consumer motivation: exploring economic, hedonic, and normative consumption goals. Journal of Consumer Marketing. Bayton, J. A. (1958). Motivation, cognition, learning—Basic factors in consumer behavior. Journal of Marketing, 22(3), 282-289. Berger, B., Matt, C., Steininger, D. M., & Hess, T. (2015). It is not just about competition with “free”: Differences between content formats in consumer preferences and willingness to pay. Journal of Management Information Systems, 32(3), 105-128. Britt, S. H. (1950). The strategy of consumer motivation. Journal of Marketing, 14(5), 666-674. Chen, W., & Thorson, E. (2021). Perceived individual and societal values of news and paying for subscriptions. Journalism, 22(6), 1296-1316. Chyi, H. I. (2012). Paying for what? How much? And why (not)? Predictors of paying intent for multiplatform newspapers. International Journal on Media Management, 14(3), 227-250. Chyi, H. I., & Lee, A. M. (2013). Online news consumption: A structural model linking preference, use, and paying intent. Digital journalism, 1(2), 194-211. Darley, W. K., Blankson, C., & Luethge, D. J. (2010). Toward an integrated framework for online consumer behavior and decision making process: A review. Psychology & marketing, 27(2), 94-116. Doyle, G. (2013). Understanding media economics. Sage. Diamantopoulos, A., Schlegelmilch, B. B., & Allpress, C. (1990). Export marketing research in practice: a comparison of users and non‐users. Journal of Marketing Management, 6(3), 257-273. 20 Engel, J. F., Kollat, D. T., & Blackwell, R. D. (1978). Consumer behavior, 3rd ed. Hinsdale, IL: Dryden. Fletcher, R., & Nielsen, R. K. (2017). Are news audiences increasingly fragmented? A cross- national comparative analysis of cross-platform news audience fragmentation and duplication. Journal of Communication, 67(4), 476-498. Goyanes, M. (2020). Why do citizens pay for online political news and public affairs? socio- psychological antecedents of local news paying behaviour. Journalism Studies, 21(4), 547-563. Griskevicius, Vladas, and Douglas T. Kenrick. "Fundamental motives: How evolutionary needs influence consumer behavior." Journal of Consumer Psychology 23.3 (2013): 372-386. Kamaruddin, A. R., & Mokhlis, S. (2003). Consumer socialization, social structural factors and decision‐making styles: a case study of adolescents in Malaysia. International Journal of Consumer Studies, 27(2), 145-156. Kang, J. Y. M., Johnson, K. K., & Wu, J. (2014). Consumer style inventory and intent to social shop online for apparel using social networking sites. Journal of fashion marketing and management. Kim, S., & Eastin, M. S. (2011). Hedonic tendencies and the online consumer: An investigation of the online shopping process. Journal of Internet Commerce, 10(1), 68-90. Lacy, S. (1989). A model of demand for news: Impact of competition on newspaper content. Journalism Quarterly, 66(1), 40-48. Lysonski, S., Durvasula, S., & Zotos, Y. (1996). Consumer decision‐making styles: a multi‐ country investigation. European journal of Marketing. Nelissen, R. M., & Meijers, M. H. (2011). Social benefits of luxury brands as costly signals of wealth and status. Evolution and human behavior, 32(5), 343-355. News Media Alliance. (2019). Media Subscribers’ Motivations and Preferences: A Research Report. News Media Alliance, Retrieved from: http://www.newsmediaalliance.org/wp- content/uploads/2019/10/For-WEB_FINAL_NewsMediaAlliance_ConnectingtheDots- DIGSUBReport_10-29-19.pdf Mallalieu, L., & Nakamoto, K. (2008). Understanding the role of consumer motivation and salesperson behavior in inducing positive cognitive and emotional responses during a sales encounter. Journal of Marketing Theory and Practice, 16(3), 183-198. Olsen, R. K. (2020). Understanding the relationship people in their early adulthood have to small-town news and paywalls. Journalism, 21(4), 507-523. 21 Park, C. H., & Kim, Y. G. (2003). Identifying key factors affecting consumer purchase behavior in an online shopping context. International journal of retail & distribution management. Pincus, J. (2004). The consequences of unmet needs: The evolving role of motivation in consumer research. Journal of Consumer Behaviour: An International Research Review, 3(4), 375-387. Prakash, G., Singh, P. K., & Yadav, R. (2018). Application of consumer style inventory (CSI) to predict young Indian consumer’s intention to purchase organic food products. Food quality and preference, 68, 90-97. Reuters Institute. (2020). Digital News Report 2020. Retrieved from: Reuters Institute for the Study of Journalism, Retrieved from: https://www.digitalnewsreport.org/survey/2020/how-and-why-people-are-paying-for- online-news/ Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American psychologist, 55(1), 68. Schaller, M., Kenrick, D. T., Neel, R., & Neuberg, S. L. (2017). Evolution and human motivation: A fundamental motives framework. Social and Personality Psychology Compass, 11(6), e12319. Sohail, M. S., & Al-Jabri, I. M. (2014). Attitudes towards mobile banking: are there any differences between users and non-users?. Behaviour & information technology, 33(4), 335-344. Sprotles, G. B., & Kendall, E. L. (1986). A methodology for profiling consumers' decision‐ making styles. Journal of Consumer Affairs, 20(2), 267-279. Zhang, K., Cai, F., & Shi, Z. (2021). Do Promotions Make Consumers More Generous? The Impact of Price Promotions on Consumers’ Donation Behavior. Journal of Marketing, 85(3), 240-255. Zhou, J. X., Arnold, M. J., Pereira, A., & Yu, J. (2010). Chinese consumer decision-making styles: A comparison between the coastal and inland regions. Journal of Business Research, 63(1), 45-51. 22 CHAPTER 3: CONCEPTUALIZING NEWS SUBSCRIPTION MOTIVATION Method Qualitative methods are powerful tools for refining existing or building new theories (Shah & Corley, 2006). Carpenter (2018) points out that qualitative research is an essential part of scale development. Since News Subscription Motivation is a new construct, its conceptualization and operationalization require qualitative research to build the foundation. The qualitative methods of focus groups or interviews are particularly useful in getting different perspectives on a topic. In the process of scale development, qualitative studies help researchers discover possible dimensions for the construct and generate potential scale items from participants' responses (Boateng et al., 2018). Participants and Sampling The present qualitative study was conducted in January 2021, when COVID-19 restrictions were still in place for in-person activities in the United States. It was necessary to conduct the study through Zoom. The effectiveness of using Zoom to conduct interviews and focus groups has also been attested by previous research (e.g., Archibald et al., 2019; Marques et al., 2020). I decided to choose individual interviews over focus groups for two reasons. First, individual in-depth interviews allow me to investigate consumers' underlying issues more closely, and thus gather a more sophisticated understanding of people's motivation of paying for news subscriptions (Stokes & Bergin, 2006). Second, given possible technical issues during the video conferencing, individual in-depth interviews allow me to have more control over the data collection process and ensure the data quality of interview conversations. 23 The goal of the qualitative study is to gather information from news subscribers in the United States. Participants were recruited through an online research platform called Prolific. Prolific allows researchers to set parameters for the samples such as nationality, the primary language used, and whether the participant agrees to sign up for a video interview. I utilized Prolific to recruit participants through the following steps: Step #1: I sent out a screening survey to get news subscribers. I set up an initial pool for participants who reside in the United States, are proficient in English, and agree to join a video interview study. Participants in this pool were invited to fill out a screening survey, which asked if they are paying for a news subscription, what publication and how much they are paying for it, and some basic demographic questions. Step #2: I then looked through the data from the screening survey, gathered Prolific IDs for those who indicated paying for news, and then invited them to sign up for the interview study. 193 people completed the screening survey from step #1, and only 39 people indicated that they are subscribing to news. Among these 39 respondents, 35 were Caucasian, three were Asian, and one indicated being in an ethnic group other than Caucasian, African American, Hispanic, or Asian. Step #3: Although a previous study by the American Press Institute (2018) noted that 88% of news subscribers are Caucasian, it is also crucial for the present study to include respondents from other races to understand why people pay for news subscriptions. Therefore, I launched another screening survey on Prolific. The participants' pool was set to recruit Hispanic and African American participants who reside in the U.S. and are proficient in English. 131 subjects participated in the screening survey, and 34 participants indicated they were paying for news subscriptions. 24 Step #4: I further sent out invitations for zoom interviews to 73 news subscribers from Step#2 and Step#3. 24 people signed up for interviews, but only 18 showed up for the interview. Among these 18 interviewees, only one of them was over 50 years old. Given previous studies indicated news subscribers tend to be older than non-subscribers and the general population (Thorson, Chen, & Lacy, 2020; American Press Institute, 2018), I decided to recruit more participants who are over 50 years old. Step #5: To recruit participants who are over 50 years old, I went back to the list of news subscribers from Step #3 and Step #4, and shortlisted 9 participants who are over 50 years old. I also started another screening survey on Prolific targeting U.S. residents over 50 years old. 30 participants responded to the screening survey and I got six more news subscribers signed up for interviews. Eventually, four participants showed up for zoom interviews. Step #6: During the interviews with the last four participants, it became clear to me that the qualitative interviews had reached saturation. Therefore, I stopped recruiting more participants for the qualitative study. In total, I recruited 354 participants from the screening survey, and 79 participants were news subscribers. 22 news subscribers completed the interview, which leaves a 28% final response rate. Summary of interview participants. The average age of interviewees was 35.8. The sample was mostly male (59.1%). White participants made up 54.2% of the sample. African American, Asian, and Hispanic participants each counted 13.6% of the sample. 18.2% of the participants completed high school and 4.5% went to some college. 9.1% of the interviewees have an Associate's degree, 36.4% have a Bachelor's degree, and 31.8% have a Master's degree or higher. In terms of income, 9.1% have less than $20,000, 22.7% earn $20,000 - $40,000, 25 27.4% have $40,000 - $60,000, 22.7% earn $60,000 - $80,000, and 18.2% have more than $80,000 for annual household income. 59.1% of participants indicated being strong Democrat, Democrat, and leaning Democrat. Independents take up 18.2% of the sample. Strong Republican, Republican, and people leaning Republican consist of 22.7% of the interviewees. The detailed demographics of each interview participant are illustrated in Table 1 on the next page. Procedure I conducted in-depth interviews with participants with semi-structured questions. Each participant scheduled their individual session with the author through Calendly.com, an online scheduling tool, and entered the zoom meeting through links automatically generated by Calendly. These interviews were conducted from January 13 to January 24, with sessions ranging from 20.1 minutes to 70.5 minutes. At the beginning of each zoom meeting, I read consent forms to each participant and informed them about the purpose of the study, the compensation for completing the study, their rights to withdraw, and confidentiality. I also asked each participant whether they consented to be recorded by a voice recorder during the interview. Interviewees were first asked about what news publications they are currently subscribing to, whether the subscription is digital or print, how much they pay for it/them each month, and how long they have been paying for it/them. Then they were asked why did they decided to pay for each news subscription. Follow-up questions such as "What do you mean by [KEYWORD OF THEIR REASONS]?" and "Can you tell me more about [KEYWORD OF THEIR REASONS]?" were asked to elicit more information. After interviewees responded, they were asked about subsequent factors from the literature review (American Press Institute, 2018; New Media Alliance, 2019; Reuters Institute, 26 Table 1. Overview of demographics of interview participants Subject Age Gender Race Education Income Partisanship No. 1 31 Female White Bachelor's degree $40,000 - $49,999 Independent leaning Republican 2 65 Female White HS diploma/GED $10,000 - $19,999 Independent leaning Democrat 3 30 Male White Bachelor's degree $100,000 - $149,999 Democrat 4 24 Female Asian Bachelor's degree $30,000 - 39,999 Democrat 5 31 Male White Master's or higher $100,000 - $149,999 Strong Democrat 6 20 Male Hispanic Associates degree $60,000 - $69,999 Strong Democrat 7 64 Male White Associates degree $40,000 - $49,999 Independent leaning Republican 8 31 Female Hispanic Bachelor's degree $80,000 - $89,999 Democrat 9 25 Female White HS diploma/GED $70,000 - $79,999 Republican 10 42 Male Black HS diploma/GED $50,000 - $59,999 Strong Democrat 11 20 Male White Bachelor's degree Less than $10,000 Strong Democrat 12 32 Male White Bachelor's degree $20,000 - $29,999 Independent 13 53 Male White Master's or higher $60,000 - $69,999 Strong Democrat 14 25 Female Black Master's or higher More than $150,000 Democrat 15 65 Male White Master's or higher $60,000 - $69,999 Republican 16 22 Male Hispanic HS diploma/GED $40,000 - $49,999 Independent 17 21 Male Black Some college $50,000 - $59,999 Democrat 18 40 Female White Master's or higher $30,000 - 39,999 Independent 19 25 Male Asian Bachelor's degree $20,000 - $29,999 Independent leaning Democrat 20 24 Female White Bachelor's degree $30,000-39,999 Democrat 21 67 Male White Master's or higher $40,000 - $49,999 Independent 22 31 Female Asian Master's or higher $70,000 - $79,999 Independent leaning Republican 27 2020) or previous interview responses. These subsequent factors include 1) content, 2) affordability, 3) convenience, 4) hitting the paywall, 5) surveillance, 6) supporting good journalism or local news organization, 7) following a particular journalist, and 8) being a good citizen. Participants were asked, "How important was [SUBSEQUENT FACTOR] in your decision to pay for [NEWS SOURCE]?" If the participant indicated the subsequent factor is important, I asked them to explain why it is crucial. Finally, I asked participants if they have anything to add and have any questions regarding this study. The interview protocol can be found in the Appendix. Preparation of Interview Data All interviews were recorded by a digital voice recorder. Recordings were submitted to an AI transcription website, Temi.com. The original transcriptions produced by Temi resulted in around a 90% accuracy rate. The author first cleaned the transcripts by deleting personal information, the interviewer's audio line, and conversations that were irrelevant to news subscription motivation. Then recordings were played again to compare with the remaining transcripts. Mistakes and missing information were corrected in this step. Then, the interview data was submitted for qualitative analysis. Thematic Coding To identify possible dimensions for the construct, I referred to the theme development process suggested by Vaismoradi et al. (2016) and thematic coding steps by Carpenter et al. (2016). At the initiation stage, an initial codebook was developed by the author. The preliminary dimensions were determined by the literature review: content, affordability, convenience, hitting the meter (paywall), surveillance, altruism, and following a particular journalist. After reading 28 the transcriptions multiple times, I deleted the dimension for following a particular journalist and added two more dimensions: brand reputation and being a good citizen. Only one participant stated he paid for a news subscription to follow a particular journalist, while others did not think this is an important motivation. Instead, several participants offered a counterargument and indicated they decided to pay for news subscriptions because of the news brand, not the news writers. Meanwhile, one participant suggested that she paid for a news subscription because she wanted to be an informed citizen and thus be more involved in society. This motivation is also confirmed by some participants in later interviews in which they expressed desires to be informed and be active in the democratic process. Then I also consulted two other researchers of the initial text clusters of why participants paid for news subscriptions. In this process, we identified two subthemes in the content category: content utility and journalism quality. Some participants paid for news subscriptions because they think the contents are useful in some way: either they found news useful to their work or investment, or the content is interesting, original, or unique and thus worth paying for. Meanwhile, some participants also indicated the superior journalism quality motivates them to pay. This division of how news subscribers want different things also reflects the difference between consumers needs and professional standard (Lacy & Rosenstiel, 2015). Thus the revised codebook included the following dimensions: content utility, journalism quality, affordability, convenience, hitting the paywall, surveillance, altruism, brand reputation, and being a good citizen. To code the qualitative data, two coders were recruited to enhance the coding process and reduce possible issues of validity, reliability, and generalizability of the results (Carpenter et al., 29 2016; Vaismoradi et al., 2016). At the time of the study, both coders were upper-level journalism students from a small primarily undergraduate institution in the Midwest. The coding process proceeds as follows. First, two coders read through interview transcriptions multiple times, highlighted noteworthy contents, identified preliminary categories on their own, and wrote reflective notes. Second, I conducted a training session with the coders to review coding schemes and sample transcripts with applied codes. Then the coding team completed seven transcripts together during the training session. This step followed the train- practice-code approach by Giesen and Roeser (2020, Figure 1). Third, coders coded the remaining 15 transcripts independently and, in a subsequent meeting, compared the coding results and resolved the differences through discussions. Fourth, I further used the qualitative analysis software to identify any other additional themes that were missed by the coders. This step is completed through the cluster retrieval function by QDA Miner. Next, I changed the label "altruism" into "supporting journalism" in consultant with another expert, who suggested that some participants indicated their motivations to support news organizations are not entirely selfless since their investment to benefit the community could also benefit themselves. Finally, I concluded the coding results and drafted participants' narratives based on the analysis. A frequency of coding results is shown in Table 2 below. Table 2. Frequency of news subscription motivations among interview participants Respondents News Subscription Motivations n % Content Utility 18 81.80% Journalism Quality 15 68.20% Affordability 19 86.40% Convenience 10 45.50% Hitting the Paywall 9 40.90% Surveillance 11 50% Being A Good Citizen 7 31.80% 30 Table 2 (cont’d) Brand Reputation 15 68.20% Supporting Journalism 12 54.50% Results In this section I aim to illustrate emerging themes from the qualitative study. Based on these results, I hope to provide conceptual understanding for each suggested dimensions of News Subscription Motivation. Content Utility. Several participants said the functional aspect of news content drove their decision to pay for news subscriptions. They see some utility of subscription content and believe they can directly benefit from the content they have purchased. Some noted that the content they get from news subscriptions is useful to their work, helpful for investments, and even "gets me coupons for grocery shopping". For example: They have a lot of articles that I could use for work. Sometimes when I need examples and case studies for my teaching I just go there and search for articles I could use. So eventually I was like, you know what, I'll just pay for it because it's worth the money (Participant 22). I never really thought about not paying for the Wall Street Journal. I mean, I need it for my investments. So I will just pay for it (Participant 21). Before COVID I do some temp[orary] jobs here and there to make some extra bucks. They have a good classified ads section that is very useful to me (Participant 10). 31 I'm a huge coupon clipper. It [the news subscription] gets me coupons for grocery shopping (Participant 13). In addition, some participants also reported that they pay for news subscriptions because the content fulfills their interests and desire for exclusive and original content: [I paid for the news subscription because] I like their comic section and sports. Haven't missed a single day in 20 years (Participant 15). I was tired of reading all the news criticizing Trump. To me, the media has slanted far too liberal. I want to read some news that fits my values and stance more (Participant 1). I chose [to subscribe to] this one because they always have the insider tips and analysis of the international trade market that you can't find from elsewhere. I could be paying for two or three other sources but still missing information I can get from this one. So why not just pay for this (Participant 14)? A big pull for me [to pay] is that they have original stories about the president and investigative pieces about China (Participant 9). Journalism Quality. Another common motivation expressed by participants is to get quality news. Several participants mentioned they decided to subscribe because the news source is fair, credible, provides balanced viewpoints, accurate information, and in-depth analysis. For example: What matters to me the most is [the news source] does fair reporting and is not slanted. A few years ago the newspaper got really slanted so I canceled the subscription. Recently 32 they have more balanced and fair reporting so I started [the subscription] again (Participant 7). There's a lot of fake news out there and I want to make sure I get accurate information. I like how they hyperlink to official websites, update corrections, etc so I know they take journalism seriously and their news is credible (Participant 5). I feel it's easy to get basic information online but it's not enough. Like you know how they report the numbers of covid cases or even the election polls, these are basic. But I also want some in-depth analysis and explanations of what this information means and how things will turn out. So getting The Washington Post helps me with that (Participant 20). Affordability. Affordability is the most frequent subscription motivation stated by participants. Several news subscribers were driven by discounts, reasonable price, and value of their money in comparison of the quality they are getting: I remember it was around Black Friday and they had a sale going on. I had been thinking about getting a subscription for a while, so I just took advantage of the promotion (Participant 5). There's a lot of free content online so to me the price has to be reasonable. If the subscription is expensive, it's harder to pay money for news when I can definitely get news for free (Participant 3). 33 It also helped [decide to pay for the subscription] when we found out the paper in next town costs about $400 more a year. [In comparison] The Charles City Press is cheaper and more affordable (Participant 7). It’s also a good deal. I’m paying about $7 a month for the whole package. Worth every penny I spend (Participant 10). Convenience. The narrative of convenience emerged when participants stated their desire to gain easy access to the content whether it is in digital or print format. For example: If I pay, I will get full access to their content...I can go to their website, download the app on my phone and tablet, and just read the news articles anytime I want (Participant 5). I kind of just paid so I can use the app on my phone. It's way easier for me to get news that way (Participant 8). I bought the print subscription so I can get the newspaper delivered to my door. Otherwise, I have to go to the post office that's 15 minutes away to get the paper (Participant 2). Hitting the Paywall. Encountering and getting around the paywall was also one of the motivations reported by participants. There were multiple accounts associated with the paywall, for instance: I constantly use up the free articles and the rest are locked behind the paywall. So I just decided to pay so I don't need to spend all the time searching for other free sources (Participant 4). 34 I have to pay because that's the only way I can get around the paywall and read the articles I want [to read] (Participant 18). Surveillance. The qualitative data also showed that respondents purchase news subscriptions to satisfy their needs for surveillance at both local and national levels. Some participants said it is important to subscribe to the local newspaper so they can know what is happening within the community. At the national level, several participants also expressed desires to get the latest updates and stay updated on current events. For example: I guess you can say I'm nosy. I spend my whole life in this town and I want to know what's going on and what happened to whom...and often these are people I know [in real life] (Participant 2). I think it's important to understand how the local government works and how are some decisions are made at the city council and so on. These are decisions that might affect me personally so I want to keep an eye on them (Participant 3). I think that's [staying updated on current events] really the main reason why I signed up. For things like the event at Capitol on January 6 and the Biden inauguration, I just want a real-time reaction to things, and that's really what I'm paying for (Participant 5). Being A Good Citizen. The motivation of being a good citizen emerged when I asked participants if there was anything they would like to add regarding why they decided to purchase a news subscription. One participant brought up this aspect (Participant 20) and was later confirmed by a few other participants. Participants emphasized their desire to stay well- 35 informed, and based on the information they have, they can be more involved with civic duties. For example: It's about being a good citizen [why I decided to purchase a subscription]. You have to stay well-informed [...] and then act upon the information like who to vote, what issues to advocate, or even whether tell people to wear a mask or not. It's my obligation as a citizen to know what's going on and do something about it (Participant 2). I would say it [being a good citizen] is very important for me [to subscribe to LA Times]. The Hispanic community was hit by COVID really hard and without LA Times I wouldn't know what's going on and how to get involved [...] how to help and do my part (Participant 6). It [being a good citizen] definitely played a part when I decided to pay, or at least it's relevant. I know some people are like "I just want to mind my own business," but I think it's really important to be involved. Help work with folks to make stuff happen or stop or whatever it is that should be being involved. You can't do that, or at least do it right, without getting news from the local paper and know what's going on (Participant 13). Brand Reputation. Participants also noted that the brand reputation is one of the deciding factors when it comes to purchasing a news subscription: I want news, and NYT [The New York Times] is the best out there. I mean, when you think about newspapers, NYT is the first name that pops up. So it wasn't hard for me to pay for it (Participant 4). 36 One of the reasons [that I subscribed] is that my coworker said it [CNBC] has the best news in our business [...] I probably will shop around a bit more if she didn't recommend it to me (Participant 14). I get the paper because it's the paper to get. It's been around for decades and everyone knows about it. Want news about the town? This is the one to get (Participant 10). Supporting Journalism. One concept that emerged from the interviews is participants' motivation to support news organizations and journalists: The second reason [I purchased a subscription] is that I want to support the journalists who work there. It's about respecting their work. I have a lot of friends that are authors and artists and they talk about people should be paid for what they do. And after a while, it dawned on me that if I object to the newspaper, having a paywall, then I'm not helping support those journalists. And I should be willing to support them too because they are authors and they spend time and effort working on the news (Participant 2). I think that the journalism industry is incredibly important and I do think that it's an industry that has suffered in the last decade. Those jobs are paid less and harder to maintain. So I do believe in supporting journalism like truth and getting the truth out there is a really worthy cause. So it is important to me to make sure that that continues to happen, especially as it starts to feel more threatened (Participant 4). Part of it [the decision to subscribe] is to support the economy of the local community, you know, help the newspaper stay in business and journalists keep their jobs. I used to 37 get the paper every day and then five times a week and now twice a week. [...] compare to other towns it's nice to know that they are still in business. Otherwise, we won't get coverage that's specifically about our town (Participant 7). Construct Development/Definitions In this study, News Subscription Motivation is defined as the factors that determine the reasons why people purchase a news subscription. These reasons are underlying motives that reflect individuals’ unmet needs. And through paying for a news subscription, these needs can be fulfilled. During the qualitative interview, several collective narratives emerged. I have labeled them as: 1) content utility; 2) journalism quality; 3) price/value; 4) convivence; 5) hitting the paywall; 6) surveillance; 7) being a good citizen; 8) brand reputation; and 9) supporting journalism. Each dimension is defined as the following: Content utility refers to audiences' motivation to subscribe to a news source for content that satisfies their specific needs for news information. This dimension measures the individuals’ desire to get news that is useful for them or fulfill their interests. Useful news and news of interest include specific topics, original content, and content is exclusive to the news sources. Journalism quality refers to people’s motivation to subscribe to a news source for superior quality in news coverage. This dimension measures individuals’ desire for higher quality information. News subscribers show a need for information that aligns with quality journalism such as accuracy, credibility, depth of analysis and investigative efforts, and objectivity. Affordability refers to audiences’ motivation to get the best value for their money when subscribing to a news source. This dimension measures the individuals’ desire to get a good deal. 38 They would prefer to pay for a news subscription when the price is reasonable, when it is relatively cheap compared to other options, and when it involves a discount. Convenience refers to audiences’ motivation to get easier, more effortless access while paying for a news subscription. This dimension measures the individuals’ desire to have their news consumption experiences and processes as convenient and easy as possible. This includes features like easy access to news content whether it’s digital or print, and ease of finding desired information. Hitting the paywall refers to audiences’ motivation to pay for a news subscription after encountering a paywall requiring to pay for the news or requiring them to subscribe. This dimension measures individuals’ decision to pay for news when they are challenged with a paywall. This includes running out of free articles, wanting to read what’s behind the paywall, and getting tired of trying to around the paywall. Surveillance refers to people’s motivation to subscribe to the news because of their need to survey the environment. This dimension measures the individuals’ desire to keep up with what’s going on in the world or their local community. Being a good citizen refers to people’s desire to behave consistently with the good citizenship norms. This dimension measures individuals’ desire to act on the good citizenship norms, such as being a well-informed citizen and being engaged in social and community issues. Brand reputation refers to audiences’ motivation to pay for a news subscription because the news sources have a good reputation. This dimension measures individuals’ motivation to pay for a news source that is a well-known brand or is well-known for its reputation. 39 Supporting journalism refers to audiences’ motivation to pay for a news subscription for supporting journalisms and news organizations. This dimension measures the individuals’ desire to support the welfare of journalists, news organizations, or the press in general. 40 APPENDIX 41 A Qualitative Study Examining News Subscription Motivations Protocol - This scale measures the individuals’ reasons why they decided to pay for news Recruit participants from professional survey platforms: - In the U.S. - 18 years and older - Have at least one active subscription of a news source Send out a short pre-screening survey first ask demographics, whether they are paying for news, what sources they are paying for and how much, and then invite those subscribers back to zoom interview. Zoom interview with participants: Hi, Thank you so much for participate in this study. I really appreciate you making time for this interview and share your insights with me. My name is Weiyue Chen and I’m from Michigan State University. I’m a graduate student working with Dr. Esther Thorson at Michigan State University. The purpose of this study is to discover people’s motivations of paying for news subscriptions. I’m interested in how people decide to pay for news sources or not, and what are the reasons they decided to pay for news. People have a lot of different ideas about how to get the news, what kind of news they want, so there’s no right or wrong answer here, just what your personal experiences and insights are. I would love to hear your insights in this topic. Your participation in this study would only take about 30 minutes. Thank you again for your time and participation. The information will only be accessible by the researchers. Your responses will be anonymous. You can withdraw from this interview at any time. You must be over 18 years of age to participate. Do you mind if I record your answers for the purpose of taking notes? (TURN ON RECORDER. Say something, anything, to make sure it’s working. Interview follows. Let them know you’re listening and encourage them to speak, but don’t get involved in the conversation. Do ask follow-up questions on concepts and their meaning. Don’t aim for a long interview, but if they want to go on, and the material’s good, go with the flow. At the end…) Opening questions (warm-up): 1. Your participants ID [on the online surveying platform]? 2. What is your name? 3. What is your current occupation, or in other words, what is your current job? 4. How did COVID-19 affect your job ? 42 Questions: What are they paying and how much Be sure to ask for elaboration on concepts and what they mean by them. The goal of this study is to identify dimensions and items/item wording for those dimensions. Use “What does that mean to you?” or “what does that word mean to you?” 1. [Subscription] a. What news publications do you subscribe to? Online? Print? a. How long have you been paying for each of these sources? b. How much do you pay for each per month? (if can’t remember monthly payments, ask per year) b. Why did choose to pay for a subscription for news? a. What content do you feel makes your subscription a worthy investment? c. What are your motivations to continue your subscription? 2. [Donation] Some news organizations don’t require payment to use them, but they ask for donations. For example, the Guardian newspaper in London just asks for donations. Wikipedia doesn’t charge you to use its information, but it requests contributions. a. In the past year, have you made contributions/donations to any independent news sources (like Michigan Bridge, ProPublica etc)? a. Which news publications/organizations? b. If so, about how much in total for [insert news sources] in the past year? b. What motivated your contribution? c. About how long have you been donating to each of these sources? Dimensions from Literature Review. Skip the section if they already mentioned in their reasons. Check those they didn’t mentioned in a logical order. Sometimes people will talk about the news sources they are not paying for. Make sure to redirect to the news sources they are paying for whenever it is possible. If they indicated one factor is important, ask them why. 1. [Content] How does the content have an impact on whether you decide to pay for news? a. How do you decide what content is worth paying for? 2. [Price/Value] 43 People often pay different amounts for subscriptions to news. For example, students often get discounted rates. Some news organizations offer special introductory deals when you first start to subscribe. a. Are you getting any discount for any publications? b. How important was a discount in determining why you decided to pay for news? c. What about being affordable in factoring in your decision to pay for news? 3. [Surveillance/Immediacy] a. How important was a getting latest update in determining why you decided to pay for news? 4. [Convenience] – delivery, easiness of getting information/news package, payment a. How important was convenience in determining why you decided to pay for news? b. Why you think about “convivence,” what comes to your mind? 5. [Hitting the Meter] a. I’m going to give you an example in real life. Let’s say that you’re wanting to read some news stories online, and after reading a few stories on the same website, then a window bumps up saying “to read more stories, you need to pay for the subscription.” This is what some people call “hitting the meter.” How important was “hitting the meter” in determining your decision to subscribe to news sources? 6. [Support] Some people also say that they pay for news to support good journalism or local news organizations. a. How important was supporting good journalism in your decision of paying for a news subscription? i. If important - What does “good journalism” mean for you? b. How important was support local journalism in your decision of paying for a news subscription? 7. [Follow a particular journalist] – added this because this was one of the top reasons found by Reuters Digital News Report (35% of U.S. respondents) 44 a. Some people pay for a news subscription because they want to follow the stories from a particular journalist they like. How important was following a particular journalist in your decision of paying for news subscriptions? 8. [Being a good citizen] - added this because one participants talked about this when I asked if there’s anything to add 9. Anything else you want to add? Thank you again for your time and insight. It’s appreciated. (TURN OFF RECORDER. TRANSFER AUDIO FILE TO EXTERNAL DRIVE. NAME THE FILE WITH LAST NAME. Mark in file notes the day, time, length of interview and the full name of person and organization. Make a backup of the audio file. In your notes, jot down what you think were the main points of what the participant said.) QUESTIONS THAT MIGHT COME UP: Will this be published? - Yes, that is our goal. We plan to try to get published in a scientific journal. Will my name be used? - No. We will identify respondents by a number, but we will not specifically say who you are or which organization you’re with. If I think of something else, may I contact you? - Certainly. Again, I’m Weiyue Chen, and my e-mail is chenwe47@msu.edu;. If you have any questions about the study, my supervisor Dr. Thorson is also available at ethorson@msu.edu. 45 REFERENCES 46 REFERENCES American Press Institute. (2018). Respondents’ demographics and news behaviors. In Paths to Subscription: Why recent subscribers chose to pay for news. American Press Institute. Retrieved from: https://www.americanpressinstitute.org/publications/reports/survey- research/subscribers-appendix-1/ Archibald, M. M., Ambagtsheer, R. C., Casey, M. G., & Lawless, M. (2019). Using zoom videoconferencing for qualitative data collection: perceptions and experiences of researchers and participants. International Journal of Qualitative Methods, 18, 1609406919874596. Boateng, G. O., Neilands, T. B., Frongillo, E. A., Melgar-Quiñonez, H. R., & Young, S. L. (2018). Best practices for developing and validating scales for health, social, and behavioral research: a primer. Frontiers in public health, 6, 149. Carpenter, S. (2018). Ten steps in scale development and reporting: A guide for researchers. Communication Methods and Measures, 12(1), 25-44. Carpenter, S., Takahashi, B., Lertpratchya, A. P., & Cunningham, C. (2016). Greening the campus: a theoretical extension of the dialogic communication approach. International Journal of Sustainability in Higher Education. Giesen, L., & Roeser, A. (2020). Structuring a Team-Based Approach to Coding Qualitative Data. International Journal of Qualitative Methods, 19, 1609406920968700. Lacy, S., & Rosenstiel, T. (2015). Defining and measuring quality journalism. New Brunswick, NJ: Rutgers School of Communication and Information. Marques, I. C. D. S., Theiss, L. M., Johnson, C. Y., McLin, E., Ruf, B. A., Vickers, S. M., ... & Chu, D. I. (2020). Implementation of virtual focus groups for qualitative data collection in a global pandemic. The American Journal of Surgery. Morris, M. D., Neilands, T. B., Andrew, E., Maher, L., Page, K. A., & Hahn, J. A. (2017). Development and validation of a novel scale for measuring interpersonal factors underlying injection drug using behaviours among injecting partnerships. International Journal of Drug Policy, 48, 54-62. Shah, S. K., & Corley, K. G. (2006). Building better theory by bridging the quantitative– qualitative divide. Journal of management studies, 43(8), 1821-1835. Stokes, D., & Bergin, R. (2006). Methodology or “methodolatry”? An evaluation of focus groups and depth interviews. Qualitative market research: An international Journal. 47 Vaismoradi, M., Jones, J., Turunen, H., & Snelgrove, S. (2016). Theme development in qualitative content analysis and thematic analysis. 48 CHAPTER 4: OPERATIONALIZING NEWS SUBSCRIPTION MOTIVATION In this chapter, I aim to demonstrate the scale development process of News Subscription Motivation. First, I review existing scales that can be used in operationalizing News Subscription Motivation. Second, I generate and revise the initial item pool based on steps recommended by Carpenter (2018). Third, I conduct a pilot study to "rehearse" the full survey and identify possible problematic questions. Then, I submit the measurements to fully launch surveys for the Exploratory Factor Analysis and Confirmatory Factor Analysis. Finally, based on the results of the quantitative studies, this dissertation provides a rationale for identifying a News Subscription Motivation. The Initial Process of Scale Development Existing scales. Previous literature does not have specific and direct measures of News Subscription Motivation. However, several studies have developed measurement scales for price/value (Sproles & Kendall, 1986), surveillance (Vincent & Basil, 1997), and being a good citizen (Copeland, 2014). These measures (Table 3) were also taken into consideration while creating the initial items pool. Table 3. Prior subscales developed in scholarly research Authors Items Price/Value Sproles & I buy as much as possible at sale prices. Kendall, 1986 The lower price products are usually my choice. I look carefully to find the best value for the money. Surveillance - so I can understand the world Vincent & - to find out things I need to know about daily life Basil, 1997 - it makes me want to learn more about things - because it helps me learn things about myself and others - it shows me what society is like nowadays - so I can learn about what might happen to me 49 Table 3 (cont’d) - it helps me judge what political leaders are really like - so I can keep up with what the government is doing - so I can talk with other people about what's covered - it helps me satisfy my curiosity - so I can learn what is going on the country and world Engaged citizenship norm Being active in politics. Copeland, 2014 Form my own opinion. Vote in elections. Help people who are worse off. Item generation. An initial set of 82 items were generated based on participants' response during the interview. I extracted words, phrases, and statements that fit each dimension. Then, I added existing measures from previous studies to the items pool. Expert feedback and pre-test. Once I gathered an initial pool of potential items, I sent them to three experts in scale development and news consumption to elicit feedback. Each expert responded through open-end feedback to suggest possible revisions. After the revision process, I had 73 items in the initial items pool. A pre-test is also essential in the scale development process (Carpenter, 2018). Based on recommendations by Boateng et al. (2018), I followed the procedures by Morris et al. (2017) and conducted additional cognitive interviews on March 24 and March 25, 2021. Five more participants were recruited to participate in zoom interviews and provided their feedback for assessing the scale items. These interviews lasted from 17 minutes to an hour. In total, participants noted six items they found confusing or do not fit well in their dimensions (these items are #7, #19, #32, #50, #52, and #57). I marked these items and made further revisions. Scale response categories and lead-in question. Weijters, Cabooter, and Schillewaert (2010) note that scales are fully labeled and with a midpoint provide more clarity to respondents. 50 The optimal numbers of response categories vary across studies. For example, some suggest that 7-point scales are optimal (e.g., Krosnick & Presser, 2010; Preston & Colman, 2000), while others suggest 5-point scales can yield better data quality (e.g., Contractor & Fox, 2011). Additionally, Dawes (2008) also notes that, in terms of data characteristics (mean scores, variances, skewness, and kurtosis), there is no statistically significant difference between data collected from a 5-point scale and a 7-point scale. I chose to employ a 5-point response to improve my survey response rate. According to American Press Institute (2018), about 60% of news subscribers are over 60 years old, and about 90% of them are above age 40. Given the online format and a large number of survey questions, a 7-point scale would easily wear out the respondents, make them impatient, and more likely to withdraw from the study. Thus I argue a 5-point scale would be optimal for my research. I also choose Item-Specific (IS) scales over Agree-Disagree (AD) scales. Research shows that participants exhibit deeper processing for IS scales than AD scales (Dykema et al., 2019; Höhne & Lenzner, 2017), and thus might lead to a more thoughtful response. Additionally, Hanson (2015) also find that IS scales generate more reliable responses than AD scales. Therefore, participants will be responding to five categories of item-specific options: 1= Not at all important; 2= Slightly important; 3= Moderately important; 4= Important; 5= Extremely Important. The lead-in question is "We want to know more about what motivates you to pay for a news subscription. Think about the news subscription(s) you are paying for right now or have purchased in the past. For each of the following statements, please rate the importance of each factor in explaining why you paid for it/them." The entire items pool is listed in Table 4 below. 51 Table 4. Proposed items for developing news subscription motivation scale Themes of each dimension Content utility = benefits to self; Journalism quality = JRN Quality; Affordability=affordable price, discount; Convenience=effortless; Hitting the paywall=paywall; Surveillance= surveillance; Good citizen=be well-informed, engaged good citizen; Brand reputation=reputation of the publication; Support journalism=support news organizations and journalists’ welfare. Moderately important Extremely important We want to know more about what motivates you to pay for a Not at all important Slightly important news subscription. Important Think about the news subscription(s) you are paying for right now or have purchased in the past 12 months. For each of the following statements, please rate the importance of each factor in explaining why you paid for the subscription(s). Access information that is useful to me. 1 2 3 4 5 Read classified ads that are useful to me. 1 2 3 4 5 Collect coupons that are useful to me. 1 2 3 4 5 Acquire information that I find interesting. 1 2 3 4 5 Obtain information that I enjoy reading. 1 2 3 4 5 Learn about different cultures. 1 2 3 4 5 Acquire local news that I care about. 1 2 3 4 5 Expands my worldview. 1 2 3 4 5 Access unique content on topics I am unable to find elsewhere. 1 2 3 4 5 Obtain original content that I cannot find from other news sources. 1 2 3 4 5 The accuracy of information. 1 2 3 4 5 The trustworthiness of information. 1 2 3 4 5 The truthfulness of information. 1 2 3 4 5 The credibility of information. 1 2 3 4 5 Thorough reporting. 1 2 3 4 5 Investigative reporting. 1 2 3 4 5 Balances both sides of an issue. 1 2 3 4 5 Fairness in reporting. 1 2 3 4 5 Unbiased approach to news reporting. 1 2 3 4 5 Receive a discount to start the subscription. 1 2 3 4 5 Receive a discount for continuing the subscription. 1 2 3 4 5 52 Table 4 (cont’d) Get a good deal. 1 2 3 4 5 The price is reasonable. 1 2 3 4 5 The price is affordable. 1 2 3 4 5 The price is cheaper compared to other news subscriptions. 1 2 3 4 5 The price fits my budget. 1 2 3 4 5 The subscription package is a good value for money. 1 2 3 4 5 Accessing the news is an easy process. 1 2 3 4 5 The print publication is directly delivered to me. 1 2 3 4 5 The publication’s website is effortless to use. 1 2 3 4 5 The publication’s mobile application is easy to use. 1 2 3 4 5 Effortless to know what’s going on when they send me email 1 2 3 4 5 notifications summarizing the news. Painless to determine what information is true. 1 2 3 4 5 Easy to decide what information is important to know. 1 2 3 4 5 The subscription package (for both digital and print) is convenient. 1 2 3 4 5 The payment process is trouble-free. 1 2 3 4 5 The publication only offers a few free articles. 1 2 3 4 5 The paywall popped up when I was reading an interesting article. 1 2 3 4 5 I kept finding articles that I wanted to read were behind a paywall. 1 2 3 4 5 I want to read what is locked behind a paywall. 1 2 3 4 5 I found it requires too much effort to search for articles that were 1 2 3 4 5 behind a paywall. I got tired of trying to get around the paywall. 1 2 3 4 5 I used up my quota for free articles. 1 2 3 4 5 I want to know what’s going on in the world. 1 2 3 4 5 I want to keep up with what’s happening in my local community. 1 2 3 4 5 I am usually curious about the latest events. 1 2 3 4 5 I want to understand my local government’s decisions. 1 2 3 4 5 I want to understand how things work in the world. 1 2 3 4 5 I want to know about the possible changes that might affect me 1 2 3 4 5 personally. I want to know what society is like nowadays. 1 2 3 4 5 I want to find out things I need to know about daily life. 1 2 3 4 5 I want to talk competently about my community. 1 2 3 4 5 Be a well-informed citizen. 1 2 3 4 5 Get involved in the society. 1 2 3 4 5 53 Table 4 (cont’d) Be active in politics. 1 2 3 4 5 Get involved with my local community. 1 2 3 4 5 Form my own opinion independently of others. 1 2 3 4 5 Help those who are worse off me. 1 2 3 4 5 Stay informed on how to vote in elections. 1 2 3 4 5 Make decisions based on rational evaluations of the situation. 1 2 3 4 5 The publication is a well-known news brand. 1 2 3 4 5 The publication has a good reputation for its journalism work. 1 2 3 4 5 My family and friends recognize this publication. 1 2 3 4 5 My family and friends speak highly of this publication. 1 2 3 4 5 The publication has been around for many decades. 1 2 3 4 5 The publication has been known for its national or local stature. 1 2 3 4 5 The news organization has a good standing. 1 2 3 4 5 Support journalists’ important work. 1 2 3 4 5 Pay respect to journalists’ hard work. 1 2 3 4 5 Help journalists keep their jobs. 1 2 3 4 5 Help news organizations stay in business. 1 2 3 4 5 Empower news organizations to continue monitoring power. 1 2 3 4 5 Support news organizations’ coverage of my local community. 1 2 3 4 5 Empowering news organizations to continue monitoring power. 1 2 3 4 5 Protect freedom of the press. 1 2 3 4 5 Ensure the continuance of local news. 1 2 3 4 5 Pilot Study Carpenter (2018) suggested that a pilot study of 50-100 people can be conducted to identify potential problematic questions. A pilot study was launched from March 25 to April 2 on Prolific. The full survey can be found in Appendix A. 121 news subscribers were invited to participate in the study, and 78 participants responded. 10 subjects did not pass the attention filter embedded in the survey, thus left us with 68 responses and a final response rate of 56.2%. A summary of participants’ demographic is shown in Table 5. 54 Table 5. Summary of demographics of participants in the pilot study Gender Education Male 61.8% High school/GED 13.2% Female 36.8% Some college 22.1% Non-binary 1.5% Bachelor's degree 47.1% Age Postgraduate/professional degree 17.6% 18-29 50% Area of living 30-39 17.6% Rural area 13.2% 40-49 7.4% Suburban area 52.9% 50-59 16.2% Urban area 33.8% 60-69 4.4% Partisanship 70-79 2.9% Strong Democratic 22.1% 80 and above 1.5% Democratic 27.9% Race Independent leaning democratic 17.6% Caucasian 64.7% Independent 13.2% African American 8.8% Independent leaning Republican 8.8% Hispanic 16.2% Republican 10.3% Asian 8.8% Strong Republican 0% Other 1.5% Income Less than $25,000 11.8% $25,000 - $49,000 19.1% $50,000 - $74,999 32.4% $75,000 - $99,999 14.7% $100,000 and above 22.1% Because the pilot study had more items than valid respondents, the statistical software could not run the common factor analysis. To make the data valid for factor analysis, I deleted six items that were noted to be problematic in the pre-test (items #7, #19, #32, #50, #52, #57). I first proceeded with examinations of data quality and did not find missing data or outliners. Then I examined the factorability of the data and found KMO value was .50, which is lower than the recommended value of .60 (Carpenter, 2018). This indicates the sample size is not adequate for the factor analysis. The Bartlett’s test had (χ2=2278; p < .00), which rejects the hypothesis that the correlation matrix is an identity matrix. This suggests that factor analysis 55 might be useful for dealing with current data. Given the contradictory indications of KMO and Bartlett's test, the results of the pilot study will have limited insights for predicting factors. I conducted a Parallel Analysis through JASP software, and the scree plot suggests the data contains five factors. As shown in Figure 1 blow, five of the eigenvalues are greater than the average eigenvalues in the Parallel Analysis line, thus makes the five-factors solution the optimal solution. Figure 1. Plot from parallel analysis I then conducted Exploratory Factor Analysis to identify factors and items. Given that the data was not normally distributed, the principal axis factoring method was employed as the extraction method recommended by Costello and Osborne (2005). The Promax method was selected for rotation because oblique methods allow factors to correlate and is more suitable for communications research (Carpenter, 2018). I set the factor numbers as five based on results from the scree plot. Following the guidelines from Yong and Pearce (2013), I first removed items that were cross-loading. Then I removed items that had a loading below .32 based on the recommendation by Carpenter (2018). After this step, results showed that factor 1 and factor 4 56 contained items from two or more originally proposed dimensions, which increased the difficulty to interpret and summarize the factors. Since other researchers (Costello & Osborne, 2005) also suggest loadings above .50 indicate stronger loaders, I used .50 as the cut-off for minimum factor loadings and eliminated three more items. The EFA results are illustrated in Table 6 below. Table 6. EFA results for pilot study with factor loadings above .50 Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 I want to find out current events I .860 need to know about daily life. I want to be informed about the .753 latest news events. I want to keep informed about governmental decisions and .752 changes that might affect me. I want to know what’s going on in .724 the world. Access information that is useful to .670 me. The truthfulness of information. .646 Make decisions based on rational .624 evaluation of the situation. It is painless to discern what .553 information is true. Thorough reporting of news events .548 and issues. The credibility of information. .522 I kept finding articles that I wanted .839 to read were behind a paywall. I got tired of trying to get around .836 the paywall. I found it required too much effort to search for articles that were .824 behind a paywall. I used up my quota for free articles. .779 I want to read what is locked .751 behind the paywall. 57 Table 6 (cont’d) The publication only offered a few .711 free articles. The paywall popped up when I was .675 reading an interesting article. Help journalists keep their jobs. .855 Support journalists’ important .853 work. Help news organizations stay in .798 business. Pay respect to journalists’ hard .795 work. Empower news organizations to .787 continue monitoring power. Protect freedom of the press. .702 Empowering news organizations to .674 continue monitoring power The price is affordable. .950 The price fits my budget. .807 The subscription package is a good .672 value for money. The price is reasonable. .657 Get a good deal for the .588 subscription. I want to keep up with what’s .902 happening in my local community. Ensure the continuance of local .869 news. Support news organizations’ .796 coverage of my local community. Collect coupons that are beneficial .572 to me. Factor 1 mostly includes items from the originally proposed dimension of surveillance, content utility, journalism quality, convenience, and being a good citizen. It includes items such as "I want to find out current events I need to know about daily life;" "I want to be informed about the latest news events;" "I want to keep informed about governmental decisions and 58 changes that might affect me;" "I want to know what’s going on in the world;" "Access information that is useful to me;" "The truthfulness of information;" "The credibility of information;" "Thorough reporting of news events and issues;" "It is painless to discern what information is true;" and "Make decisions based on rational evaluations of the situation." The Cronbach's alpha of this factor equals .90. It is worth noting that item #34 - "It is painless to discern what information is true" - an item originally from the proposed dimension of convenience clustered together with "The truthfulness of information" and "The credibility of information" from journalism quality. Therefore, I also marked this item to be problematic and will take it out from the final version of surveys for Explanatory Factor Analysis and Confirmatory Factor Analysis in later research stages. Factor 2 reflects the "hitting the paywall" dimension, which includes items like "The publication only offered a few free articles;" "The paywall popped up when I was reading an interesting article;" "I kept finding articles that I wanted to read were behind a paywall;" "I want to read what is locked behind the paywall;" "I found it required too much effort to search for articles that were behind a paywall;" "I got tired of trying to get around the paywall;" and "I used up my quota for free articles." Cronbach's alpha equals .91 Factor 3 echoes the "support journalism" dimension proposed in conceptualization. It includes items such as: "Support journalists’ important work;" "Pay respect to journalists’ hard work;" "Help journalists keep their jobs;" "Help news organizations stay in business;" "Empower news organizations to continue monitoring power;" and "Protect freedom of the press." The Cronbach's alpha is .91. 59 Factor 4 reflects the proposed dimension of affordability. Items in this factor contain "Get a good deal for the subscription;" "The price is reasonable;" "The price is affordable;" "The price fits my budget;" and "The subscription package is a good value for money." Cronbach's alpha equals .85. Factor 5 contains items from the proposed content utility dimension, surveillance dimension, and support journalism dimension. Three of the items indicated news subscribers longing for local news coverage. This factor includes items such as "Collect coupons that are beneficial to me;" "I want to keep up with what’s happening in my local community;" "Ensure the continuance of local news;" and "Support news organizations’ coverage of my local community." Cronbach's alpha equals .77. To explore other possibilities, I also did a further cleaning of the factors using .70 as the cut-off point for minimum loadings. As a result, there were slight changes in the factor structure. The EFA results are shown in Table 7 below. Table 7. EFA results for pilot study with factor loadings above .70 Factor Factor Factor Factor Factor 1 2 3 4 5 Help journalists keep their jobs. .879 Help news organizations stay in business. .839 Support journalists’ important work. .820 Pay respect to journalists’ hard work. .791 Protect freedom of the press. .749 Empower news organizations to continue .706 monitoring power. I got tired of trying to get around the paywall. .835 I kept finding articles that I wanted to read were .813 behind a paywall. I used up my quota for free articles. .812 I found it required too much effort to search for .787 articles that were behind a paywall. 60 Table 7 (cont’d) I want to read what is locked behind the paywall. .736 The publication only offered a few free articles. .727 I want to keep up with what’s happening in my .969 local community. Ensure the continuance of local news. .847 Support news organizations’ coverage of my local .825 community. I want to be informed about the latest news events. .834 I want to find out current events I need to know .763 about daily life. I want to keep informed about governmental .747 decisions and changes that might affect me. The price fits my budget. .959 The price is affordable. .835 Factor 1 echoes the "support journalism" dimension proposed in conceptualization. It includes items such as: "Support journalists’ important work;" "Pay respect to journalists’ hard work;" "Help news organizations stay in business;" "Empower news organizations to continue monitoring power;" and "Protect freedom of the press." The Cronbach's alpha for factor 1 is .92. Factor 2 reflects the "Hitting the paywall" dimension, which includes items like: "The publication only offered a few free articles;" "I kept finding articles that I wanted to read were behind a paywall;" "I want to read what is locked behind the paywall;" "I found it required too much effort to search for articles that were behind a paywall;" "I got tired of trying to get around the paywall;" and "I used up my quota for free articles." Cronbach's alpha equals .91. Factor 3 contains items from the proposed surveillance dimension and support journalism dimension, but clustered items indicate news subscribers longing for local news coverage. This factor includes items such as "I want to keep up with what’s happening in my local community;" 61 "Ensure the continuance of local news;" and "Support news organizations’ coverage of my local community." Cronbach's alpha equals .92. Factor 4 reflects the proposed dimension of surveillance needs. Items in this factor are: "I want to be informed about the latest news events;" "I want to keep informed about governmental decisions and changes that might affect me;" and "I want to find out current events I need to know about daily life." The Cronbach's alpha for factor 4 is .84. Factor 5 demonstrates that news subscribers are also driven by the monetary price they pay for news subscriptions. Items in this dimension are: "The price is affordable;" and "The price fits my budget." Cronbach's alpha equals .89. Discussion of Pilot Study Results The purpose of conducting a pilot study is two-fold. First, researchers can identify skipped questions by participants and problematic items in the scale. I did not find any skipped questions by the participants. This is a direct result from setting “force response” function in the survey distribution software, which prevents participants from submitting their answers if there is a missing question. However, the pilot study does suggest one problematic item: “It is painless to discern what information is true.” This item was originally designed to be included in the convenience dimension. However, after conducting factor analysis, it grouped with two items in the journalism quality dimension that describes trustworthiness and credibility. Therefore, I decided to eliminate this item from the final survey of the full launch for EFA. The second goal of this pilot study is to identify possible dimensions. From two versions of EFA results, four factors consistently emerged: price/value, hitting the paywall, support journalism, and local news. Thus, I expect to see these four dimensions appear again in the full launch for EFA. In addition, I also expect to see more factors emerge, given the limited sample 62 size (N= 68) for the pilot study and an inadequate result of KMO test (=.50). The full launch of the EFA survey will have 400 participants and hopefully increase the data's factorability. It also came to my attention that during the cleaning processes of factor analysis, items originally designed for “content utility” tend to cluster with other dimensions. While this can be explained by the smaller sample size (N= 68) for pilot the study and consequently the factorability of data, I decided to further revise items in this dimension. The theme of the “content utility” essentially highlights people’s motivation to pay for news because they see personal benefits in getting a news subscription. The new proposed items are listed below in Table 8, and will be included in the next round survey for EFA. Table 8. Revised items for content utility dimension Conceptual Definition Content utility refers to audiences' motivation to subscribe to a news source for content that satisfies their specific needs for news information. Operational Definition This dimension measures the individuals’ desire to get news that is useful for them or fulfill their interests. Useful news and news of interest include specific topics, original content, and content is exclusive to the news sources. Theme: personal benefits Items Access information that is useful to me. Read news that are helpful to my work. Gather information that is valuable to my daily life. Read classified ads that are helpful to me. Collect coupons that are beneficial to me. Acquire information that I find interesting. Obtain information that I enjoy reading. Learn about different cultures. Get news articles that are rewarding to my personal growth. Access unique content on topics I am unable to find elsewhere. Obtain original content that I cannot find from other news sources. 63 Exploratory Factor Analysis (Study 2) After the pilot study, I revised the pool of items and submitted 77 items to conduct an Exploratory Factor Analysis. This study is labeled as Study 2 in my dissertation research. Sample. I employed a professional survey company, Dynata (www.dynata.com) to recruit survey participants. The project manager from Dynata used email, text, phone alters and in- platform messaging to invite participants from their U.S. national panel. The incentives for participants were based on participants’ completion rate, time, and quality. Participants who passed the quality screening by Dynata were provided cashable Dynata credits that are suitable for a 10-minute survey. The survey used in this study can be found in Appendix 2. To maximize the generalizability of the final scale, the goal at this step is to get a sample quota that is relatively representative of news subscribers in U.S. From recent research on news subscribers, I selected the study by the American Press Institute (2018) for its large scale. In this study, API collaborated with 12 newspaper publishers and 90 newspapers from 47 states in the U.S., and they were able to collect responses from news subscribers across the country (N=4,113). Therefore, I used demographic quotas provided by API (2018) while recruiting participants. A total of 1,174 subjects participated in the survey. An initial screening question was used to eliminate participates who are not news subscribers, which screened out 601 participants. Two attention checks also eliminated 156 invalid responses, which left 417 participants in the final sample (Response Rate = 35.5%). The size of 417 is above the five to one ratio for sample size suggestion made by Carpenter (2018). The demographic quotas roughly match the API study with a younger sample. Table 9 illustrates the demographic details of the API study, and Table 10 shows the summary of participants in Study 2. 64 Table 9. Respondents’ demographics from the API (2018) study (N=4,113) Age Education 18-39 7% High school or less 5% 40-59 28% Some college 26% 60 or older 65% College degree and above 67% Gender Race Male 51% White 88% Female 47% Other 12% Income Living Area under 50k 34% Rural 17.00% 50k-100k 33% Suburban 55.00% above 100k 33% Urban 28.00% Table 10. Summary of participants in Study 2 (N=417) Age Education 18-29 12.50% Less than high school 0.20% 30-39 13.20% high school or GED 12.50% 40-49 16.30% Some college 24.90% 50-59 7.70% Four-year college 33.80% 60-69 18.50% Postgraduate 28.50% 70-79 28.50% Race 80 or older 3.40% White 83.00% Gender Black 6.20% Male 54.40% Hispanic 4.60% Female 45.60% Asian 3.60% Income Indigenous/Alaskan Native 1.00% under 25k 7.90% Multiracial 1.00% 25k-49,999 24.70% Other 0.70% 50k-74,999 22.10% Partisanship 75k-99,999 14.60% Strong Democratic 21.80% above 100k 30.70% Democratic 20.40% Living Area Independent leaning Democratic 13.20% Rural 18.50% Independent 15.30% Suburban 48.70% Independent leaning Republican 7.20% Urban 32.10% Republican 11.80% Missing 0.70% Strong Republican 10.30% 65 Verifying data quality. According to recommendations by Carpenter (2018), I first examined the correlation matrix for all 77 initial items, ensuring all correlation numbers are above .30. I then submitted the data to Bartlett’s Test of Sphericity and Kaiser-Meyer-Olkin (KMO) Test for Sampling Adequacy. The result for Bartlett’s Test was significant (χ² = 9493.25, p<.001), and the KMO value equals .95. These results meet the recommended thresholds (p<.05 and KMO ≥ .60), which indicates adequate data quality for conducting factor analysis. Determining optimal number of factors. A Parallel Analysis was conducted to determine the best number of factors, following the suggestions by Carpenter (2018) and Watkins (2006). Parallel Analysis is considered a more suitable method for two reasons. First, the cut-off point for the scree plot is relatively subjective (Carpenter, 2018). Second, scholars argue that using eigenvalues greater than one often result in inaccurate extractions of factors (Carpenter, 2018; Kline, 2013). The result of Parallel Analysis indicated the six-factor solution is optimal. To obtain the most accurate results, I also tested the five and seven factors in the EFA. However, the EFA set for five factors resulted in a single-item factor, and the EFA for seven factors resulted in an empty factor. Based on prior scale development recommendations, a factor needs to contain at least three items (Carpenter, 2018). Therefore, I determined six factors were to be extracted. Factor extraction and rotation methods. Costello and Osborne (2005) suggest that the principal axis factoring method should be employed when the data was not normally distributed, and maximum likelihood method should be used when data fit normal distribution. I conducted a Mardia’s Test for multivariate normal distribution in R, which rejected the null hypothesis (Skewness: β = 1695.80, p<.001; Kurtosis: β = 6911.75, p<.001). A Q-Q plot was also produced 66 in R to illustrate that data from Study 2 does not fit multivariate normal distribution (Figure 2). Therefore, principal axis factoring method was employed for factor extraction. Figure 2. Q-Q plot from multivariate normality test The Promax method was selected for rotation because oblique methods allow factors to correlate and is more suitable for communications research (Carpenter, 2018). I set the factor numbers as six based on results from the Parallel Analysis and submitted the data to EFA. The EFA was conducted through SPSS. Item reduction. Following the guidelines from Yong and Pearce (2013), I first removed items that were cross-loading. Then I removed items that had a loading below .32 based on the recommendation by Carpenter (2018). After these steps, 39 items were eliminated. To further clean the scale, I followed the recommendation by Costello and Osborne (2005), which suggests factor loadings above .50 are more robust. Seven more items were deleted from this step. 67 To further reduce the number of items in the scale, I evaluated items in each factor based on face and content validity. Items that did not match the conceptual definition were also eliminated. For items that are similar with each other, I compared their factor loadings and retained the ones that has stronger loadings. In the end, a final scale of 19 items were attained, with three items left for Factor 1 to Factor 5, and four items were retained for Factor 6. DeVellis (2016) also suggests researchers inspect the internal consistency of each factor. I calculated Cronbach’s α, average variance extracted (AVE) and composite reliabilities (CR) for each factor. Alpha values for all six factors ranged from .78 to .87, which meet the requirement of .70 (DeVellis, 2016). Fornell and Larcker (1981) recommend AVE to be .50 and higher, and Hair (1997) suggest CR to be above .70. The first five factors showed good internal consistency of AVE ranging from .54 to .70, and CR ranging from. 70 to .88. Factor 6 (content utility) demonstrate a good CR of .72, yet the AVE value (.40) is below recommendation. It is also worth noting that Fornell and Larcker (1981, p.41) contend that AVE is “a more conservative measure,” and based on adequate CR values alone, researchers can claim convergent validity of the construct is reached (Lam, 2012). Since CR value for all six factors are above .70, I retained Factor 6 in the final scale. Final items. Table 11 demonstrates the final scale generated from EFA. Correlations between factors are shown in Table 12. All factor loadings are above .50, with communalities ranging from .44 to .75. The factor structure accounted for 60.67% cumulative variance. Factor 1 to Factor 6 were labeled as: Supporting Journalism (SJ), Journalism Quality (JQ), Triggered by the Paywall (PW), Community Attachment (CA), Price (PR), and Content Utility (CU). 68 Table 11. Exploratory factor analysis results (final scale) Item SJ JQ PW CA PR CU Communalities Mean SD Factor 1: Supporting Journalism SJ1 Support journalists’ important work. .88 .03 -.03 -.05 -.03 .03 .75 3.55 1.20 SJ2 Help journalists keep their jobs. .82 .00 .03 -.09 -.02 .04 .66 3.35 1.29 Help news organizations stay in SJ3 .81 .01 -.01 .13 .05 -.13 .67 3.42 1.24 business. Factor 2: Journalism Quality JQ1 The credibility of information. -.06 .92 .02 .03 -.02 -.05 .75 4.36 .89 Unbiased approach to news JQ2 .04 .78 .03 .01 .03 -.03 .64 4.20 .98 reporting. JQ3 Thorough reporting. .17 .51 -.06 -.06 -.01 .23 .56 4.13 .97 Factor 3: Triggered by the Paywall The paywall popped up when I was PW1 -.07 .04 .89 -.01 -.08 .02 .69 2.37 1.39 reading an interesting article. PW2 I used up my quota for free articles. .04 .00 .82 -.04 -.05 -.04 .61 2.29 1.39 I found it requires too much effort PW3 to search for articles that were .03 -.03 .59 .09 .16 -.05 .50 2.78 1.37 behind a paywall. Factor 4: Community Attachment I want to keep up with what’s CA1 -.11 .03 -.07 .90 .02 .02 .74 3.85 1.14 happening in my community. I want to understand my local CA2 .00 .04 .01 .66 -.03 .12 .55 3.84 1.08 government’s decisions. I want to get involved with my local CA3 .18 -.08 .14 .61 -.03 .01 .57 3.33 1.23 community. Factor 5: Affordability PR1 The price is reasonable. .02 .06 -.08 .12 .85 -.16 .65 3.89 1.06 69 Table 11 (cont’d) Item SJ JQ PW CA PR CU Communalities Mean SD PR2 The price is affordable. .01 -.08 -.04 -.09 .79 .10 .59 3.92 1.02 Get a good deal when starting or PR3 -.07 .04 .23 -.09 .52 .16 .51 3.42 1.18 continuing the subscription(s). Factor 6: Content Utility Gather information that is valuable CU1 -.09 -.01 -.05 .11 -.04 .79 .56 3.88 .97 to my daily life. Obtain original content that I cannot CU2 .04 .06 -.01 .01 -.01 .60 .44 3.65 1.16 find from other news sources. Obtain information that I enjoy CU3 .00 .11 -.01 .02 .08 .59 .53 4.05 .90 reading. Get news articles that are rewarding CU4 .18 -.07 .07 .06 .08 .51 .54 3.49 1.16 to my personal growth. Cronbach's α .83 .80 .87 .81 .78 .79 Explained Variance 36.58% 8.84% 5.39% 4.94% 2.94% 1.99% Total 60.67% Eigenvalues 7.34 2.05 1.39 1.29 .91 .80 Note: Factor loadings presented here are from the pattern matrix. Extraction Method: Principal Axis Factoring. Rotation Method: Promax with Kaiser Normalization (Kappa=4). Rotation converged in 7 iterations. The question battery was: “Think about the news subscription(s) you are paying for right now or have purchased in the past 12 months. For each of the following statements, please rate the importance of each factor in explaining why you paid for the subscription(s).” 70 Table 12. Correlations between factors extracted from EFA SJ JQ PW CA PR CU SJ -- JQ .49** -- ** PW .37 .08 -- ** ** CA .53 .35 .36** -- ** ** ** PR .40 .36 .42 .40** -- ** ** ** ** CU .58 .55 .38 .62 .56** -- Note: Correlations were calculated using pairwise deletion and Spearman’s rank-order since the data were not normally distributed. ** indicates significance at the 0.01 level (2-tailed). Factor 1 was labeled as Supporting Journalism as it contains three items that reflect participants’ desire to support the journalistic effort, journalism workers and organizations. Items include: “Support journalists’ important work;” “Help journalists keep their jobs;” and “Help news organizations stay in business.” Items in Factor 2 were anchored towards the quality aspect of journalism such as credibility, objectivity and thoroughness of news reporting. Hence this factor was labeled as Journalism Quality. Items include “The credibility of information;” “Unbiased approach to news reporting;” and “Thorough reporting (of the publication).” Factor 3 was defined as Triggered by the Paywall. This series of items described participants’ purchase behavior when they encountered a paywall. There were other items in the initial pool that embed negative emotions (e.g., frustration of paywall), but items retained in Factor 3 demonstrated relatively neutral sentiment. Subscribers’ motivation to pay was triggered when they encountered a paywall. Thus, I used Triggered by the Paywall as the conceptual label for this subscale. Items included: “The paywall popped up when I was reading an interesting article;” “I used up my quota for free articles;” and “I found it requires too much effort to search for articles that were behind a paywall.” 71 Items in Factor 4 mostly reflected news subscribers’ need for strong community connection, with a combination of keeping up with current events and local governments and taking actions to be involved in the community. Items include: “I want to keep up with what’s happening in my community;” “I want to understand my local government’s decisions;” and “I want to get involved with my local community.” Factor 5 was labeled as affordability since it described cost-related motivations for subscribers to pay. Items include: “The price is reasonable;” “The price is affordable;” and “Get a good deal when starting or continuing the subscription(s).” Factor 6 was described as Content Utility as it reflected a common theme of individuals using the news subscription to fulfill their needs other than getting quality news. The information news subscription offered are deemed valuable and useful to individuals’ daily life (e.g., original content, leisure, personal growth). Items included: “Gather information that is valuable to my daily life;” “Obtain original content that I cannot find from other news sources;” “Obtain information that I enjoy reading;” and “Get news articles that are rewarding to my personal growth.” Discussion of Exploratory Factor Analysis Results The supporting journalism factor reflected people’s recognition of the crisis news industry is facing. I did not label this factor as altruism since the motivation for supporting journalists or news organizations may not be entirely self-less. Furthermore, this factor also might be vulnerable to social desirability bias. Although there is little empirical evidence of whether news consumption is perceived as socially desirable, further research is needed to explore the correlation between motivations for subscribing to news and social desirability. 72 Items of journalism quality demonstrated American consumers’ needs in the current state of journalism practice. The credibility item is a reflection of the proliferation of fake news (Pew Research Center, 2020). The unbiased reporting item is a reflection of the public’s perception of increased media slant (Asbury, 2020) amplified needs for neutral coverage (Mitchell et al., 2018). Moreover, the inclusion of three items provides a more specific inspection of what matters to news audience. It also advances the measurements of “what journalism quality drives paying,” comparing to the broader index of journalism attributes and writing quality in previous research (Chen & Thorson, 2021). For the paywall factor, it is worth noting that the sentiments reflected in the items are neutral, not negative emotions such as frustration or anger. This might result from people’s growing acceptable of paywall, which is consistent with the increased digital subscriptions in the past two years (Kalim, 2020). In addition, marketing research also suggested that some consumers are more impulsive when it comes to shopping (Sproles & Kendall, 1986)., and have a stronger need for instant gratification (Barbopoulos, & Johansson, 2016). Thus I speculate subscribers who are more likely to be triggered by the paywall would also exhibit other impulsive purchase tendencies. However, more empirical evidence is needed to support this assumption. Findings of community attachment and price are consistent with previous studies in news consumption (Pew Research Center, 2019; Goyanes, 2020; Burger et al., 2015) and marketing research (Barbopoulos, & Johansson, 2016; Sproles & Kendall, 1986). For local news organizations, the community attachment factor also emphasizes their involvement in the local community and journalism’s role in building a democratic society. 73 The content utility factor is a relatively weak factor with a lower AVE value and an explained variance below 2%. The included items described more abstract consumer needs (e.g., “valuable to my daily life,” “rewarding to my personal growth”) rather than concrete functionality of the content. Readers should consult the previous qualitative study to better understand this factor. Since this is an exploratory study, I decided to keep this factor in the final scale. More data is needed to test the factor model again for scale validation. Scale Validation: Confirmatory Factor Analysis (Study 3) To provide additional validity for the final scale generated from EFA, it is standard practice for scale developers to conduct another CFA, ideally with data from a new sample (Worthington & Whittaker, 2006). The goal at this step is to assess whether the measurement of items, factor structure, and function stand true with republication using a new sample or data from different time points. In other scholarship, this step is also labeled as “the test of dimensionality” (e.g., Boateng et al., 2018). Therefore, I collected another dataset to conduct CFA. Sample and procedure. A second wave of the survey was launched through Dynata in September 2021, following the quota and recruitment methods for Study 2. However, to ensure the independence between two samples, Dynata also used a blocklist to prevent Study 2 participants from partaking Study 3. For participants who provided valid responses, Dynata issued cashable credits that are suitable for a 15-minute survey. The survey used in this study can be found in Appendix B. A total of 2,059 subjects consented to take the survey. Two initial questions were set at the beginning to make sure only news subscribers’ responses are collected. 1,217 individuals were eliminated from initial screening. The first attention check rejected 119 invalid responses, 74 and the second attention check also removed 203 participants. Finally, I further removed 14 responses that were completed within four minutes, which is deemed implausible given the survey length. Thus, the final sample consisted of 506 subjects (Response Rate = 24.6%). The demographic quotas also roughly match Study 2, and a summary of sample participants is illustrated in Table 13 below. Table 13. Summary of participants in study 3 (N=506) Age Education 18-29 6.3% Less than high school 0.8% 30-39 13.0% high school or GED 13.2% 40-49 17.2% Some college 31.6% 50-59 9.3% Four-year college 30.6% 60-69 20.0% Postgraduate 23.7% 70-79 29.6% Race 80 or older 4.5% White 83.6% Gender Black 9.1% Male 53.4% Hispanic 3.0% Female 46.6% Asian 2.4% Income Indigenous/Alaskan Native 0.2% under 25k 8.3% Multiracial 0.8% 25k-49,999 27.3% Other 1.0% 50k-74,999 19.8% Partisanship 75k-99,999 18.2% Strong Democratic 24.9% above 100k 26.5% Democratic 16.2% Living Area Independent leaning Democratic 8.9% Rural 16.4% Independent 17.6% Suburban 56.3% Independent leaning Republican 9.5% Urban 27.3% Republican 13.2% Strong Republican 9.7% Results of CFA. I used lavaan package in R to conduct Confirmatory Factor Analysis. The results are further shown in Figure 3. In terms of fit indices, Boateng et al. (2018) recommend researchers report the chi-square test of exact fit, Root Mean Square Error of Approximation (RMSEA), Tucker Lewis Index (TLI), Comparative Fit Index (CFI), and 75 Standardized Root Mean Square Residual (SRMR). For adequate model fit, RMSEA needs to be .06 or lower, and SRMR should be .08 or lower. TLI and CFI values above .95 are also considered indicators of good fit. The six-factor solution showed good fit in the sample of Study 3. The chi-square test for the factor model was significant (χ2 (137) =193.22, p < .001), with a χ2/df ratio of 1.41. Both CFI and TLI exceeded the recommended value of .95 (CFI= .99, TLI= .99), which indicate excellent model fit. RSMEA= .03, and SRMR= .05, which also indicate adequate model fit. Thus, the six- factor model of the NSM scale demonstrates validity across different samples. 76 Figure 3. Confirmatory factor analysis results 77 Convergent and discriminant validity. The purpose of testing convergent and discriminant validity of the final scale is to establish the construct validity of News Subscription Motivation. Netemeyer et al. (2003) suggest the convergent validity is established when correlations are found between independent measures of the same construct. Since NSM is a newly defined construct, here I aim to demonstrate the convergent validity of NSM subscales. I looked for similar concepts for each factor of NSM, and the details are presented in Table 14 below. Table 14. Similar subscales for factors of news subscription motivation NSM Factors Similar Concept(s) Measures People can ask me for help if necessary. People can count on my help if they have Helping Behaviors (ten difficulties. Supporting Brummelhuis, van der Lippe, Journalism I often help people in need. & Kluwer, 2010) If someone is absent, I’m willing to take over the work. Getting very good quality is very important to me. High-Quality Conscious Journalism Consumer, from Consumer When it comes to purchasing products, I Quality Style Inventory (Sproles & try to get the very best or perfect choice. Kendall, 1986) In general, I usually try to buy the best overall quality. I should plan my shopping more carefully Impulsive Consumer, from than I usually do. Consumer Style Inventory I am impulsive when purchasing things. (Sproles & Kendall, 1986) I often make careless purchases I later wish I had not. Triggered by the Paywall Get something that I wanted or needed for now. Instant Gratification from Consumer Motivation Scale Satisfy immediate needs. (Barbopoulos & Johansson, Choose an option that increases my 2016) immediate comfort. Act in a way that was comfortable. 78 Table 14 (cont’d) NSM Factors Similar Concept(s) Measures I feel like I belong in my community. Overall, I am very attached to my community. I like to think of myself as similar to the people who live in my community. I think I agree with most people in my Community Community Attachment community about what is important in life. Attachment (Theodori & Theodori, 2015) I plan to remain a resident of this community for a number of years. If the people in this community were planning something, I’d think of it as something WE were doing rather than THEY were doing. Price Conscious Consumer, I buy as much as possible at sale prices. from Consumer Style I look carefully to find the best value for Affordability Inventory (Sproles & the money. Kendall, 1986) I carefully watch how much I spend. For news that fits into my busy schedule. To get latest updates on news stories. To obtain news at the times I want. Gratification-Opportunities For stories on a variety of topics. Content Utility (Dimmick, Chen, Li, 2004) To get information as quickly as possible. To use my time wisely. For a variety of choices in news coverage. For convenient access to news. Bivariate correlations were calculated between each of the six NSM factors and their corresponding subscales. Spearman's Rank-Order Correlation was employed, and the analysis demonstrated all six dimensions of NSM are positively correlated with their corresponding scales at the .01 level. The correlation coefficient rs ranged from .29 to .56. Thus the convergent validity is established. 79 Definitions and assessments of discriminant validity vary across the literature (Rönkkö & Cho, 2020). Since NSM is a newly developed construct without alternative measurements, the goal of establishing discriminant validity here is to provide evidence that NSM factors represent theoretically different concepts. I followed the suggestion of Henseler, Ringle and Sarstedt (2015) and computed the heterotrait-monotrait ratio of correlations (HTMT). The HTMT values among different factors (latent variables) ranged from .10 to .57, meeting the recommendation of below .85 (Henseler et al., 2015). Therefore, the discriminant validity of NSM is established. Conclusion To operationalize News Subscription Motivation, I generated a pool of potential items from the qualitative research, revised the items based on expert feedback, pretest, and pilot study. I also collected two samples of news subscribers in the U.S., developed and tested the scale through EFA and CFA. To further validate the scale, A total of six factors emerged, and the final scale showed robust fitness indices and strong evidence for convergent and discriminant validity. Therefore, I conclude the measurement items of NSM are reliable. However, in a broader context, more research is needed to investigate where the construct of News Subscription Motivation fits into the theoretical framework of consumer behavior of news subscribers. In the next chapter, I tested how factors in News Subscription Motivation work as predictors. Through additional analysis, I aimed to demonstrate the importance of News Subscription Motivation and its essential connections of predicting and interpreting people’s purchase behavior and intention when it comes to paying for news subscriptions. 80 APPENDICES 81 Appendix A: Survey - Exploratory Factor Analysis (Study 2) Screening Question The following question is asked at the beginning so we could make sure we are surveying people who are news subscribers. This survey is designed for people who are currently paying for a news subscription, or have paid for a news subscription in the past 12 months. Are you currently paying for a news subscription, or have you paid for a news subscription in the past 12 months? (By “paying/paid,” this study means “purchase,” not “donate”) o Yes, I am paying/ I have paid for a news subscription (1) o No, I do not pay for news (2) o I’m not sure (3) Data Quality Question We care about the quality of our data. For us to get the most accurate measures of your knowledge and opinions, it is important that you thoughtfully provide your best answers to each question in this survey. Do you commit to thoughtfully provide your best answers to each question in this survey? o I will provide my best answers (1) o I will not provide my best answers (2) o I can’t promise either way (3) Subscription Details What news publication(s) are you currently paying for? ________________________________________________________________ Please indicate the MONTHLY PRICE you are paying for the following sources (If you are not paying any money, please just fill in “0”): - National newspaper (e.g. The New York Times, The Washington Post) --Digital only subscription ________________________________________________________________ 82 - National newspaper (e.g. The New York Times, The Washington Post)--Print and digital subscription bundle ________________________________________________________________ - Local newspaper in your own town/community--Digital only subscription ________________________________________________________________ - Local newspaper in your own town/community--Print and digital subscription bundle ________________________________________________________________ News Subscription Motivation Scale We want to know more about what motivates you to pay for a Not at all important Slightly important Moderately important Extremely important news subscription. Important Think about the news subscription(s) you are paying for right now or have purchased in the past 12 months. For each of the following statements, please rate the importance of each factor in explaining why you decided to purchase the news subscription(s). Access information that is useful to me. 1 2 3 4 5 Read news that are helpful to my work. 1 2 3 4 5 Gather information that is valuable to my daily life. 1 2 3 4 5 Read classified ads that are helpful to me. 1 2 3 4 5 Collect coupons that are beneficial to me. 1 2 3 4 5 Acquire information that I find interesting. 1 2 3 4 5 Obtain information that I enjoy reading. 1 2 3 4 5 Learn about different cultures. 1 2 3 4 5 Get news articles that are rewarding to my personal growth. 1 2 3 4 5 Access unique content on topics I am unable to find elsewhere. 1 2 3 4 5 83 Obtain original content that I cannot find from other news sources. 1 2 3 4 5 The accuracy of information. 1 2 3 4 5 The trustworthiness of information. 1 2 3 4 5 The truthfulness of information. 1 2 3 4 5 The credibility of information. 1 2 3 4 5 Thorough reporting. 1 2 3 4 5 Investigative reporting. 1 2 3 4 5 Balances both sides of an issue. 1 2 3 4 5 Fairness in reporting. 1 2 3 4 5 Unbiased approach to news reporting. 1 2 3 4 5 Receive a discount to start the subscription. 1 2 3 4 5 Receive a discount for continuing the subscription. 1 2 3 4 5 Get a good deal. 1 2 3 4 5 The price is reasonable. 1 2 3 4 5 The price is affordable. 1 2 3 4 5 The price is cheaper compared to other news subscriptions. 1 2 3 4 5 The price fits my budget. 1 2 3 4 5 The subscription package is a good value for money. 1 2 3 4 5 Accessing the news is an easy process. 1 2 3 4 5 The print publication is directly delivered to me. 1 2 3 4 5 The publication’s website is effortless to use. 1 2 3 4 5 The publication’s mobile application is easy to use. 1 2 3 4 5 Easy to decide what information is important to know. 1 2 3 4 5 The subscription package (for both digital and print) is convenient. 1 2 3 4 5 The payment process is trouble-free. 1 2 3 4 5 The publication only offers a few free articles. 1 2 3 4 5 The paywall popped up when I was reading an interesting article. 1 2 3 4 5 I kept finding articles that I wanted to read were behind a paywall. 1 2 3 4 5 I want to read what is locked behind a paywall. 1 2 3 4 5 I found it requires too much effort to search for articles that were behind a paywall. 1 2 3 4 5 I got tired of trying to get around the paywall. 1 2 3 4 5 I used up my quota for free articles. 1 2 3 4 5 I want to know what’s going on in the world. 1 2 3 4 5 I want to keep up with what’s happening in my local community. 1 2 3 4 5 I am usually curious about the latest events. 1 2 3 4 5 84 I want to understand my local government’s decisions. 1 2 3 4 5 I want to understand how things work in the world. 1 2 3 4 5 I want to know about the possible changes that might affect me personally. 1 2 3 4 5 I want to know what society is like nowadays. 1 2 3 4 5 I want to find out things I need to know about daily life. 1 2 3 4 5 I want to talk competently about my community. 1 2 3 4 5 Be a well-informed citizen. 1 2 3 4 5 Get involved in the society. 1 2 3 4 5 Be active in politics. 1 2 3 4 5 Get involved with my local community. 1 2 3 4 5 Form my own opinion independently of others. 1 2 3 4 5 Help those who are worse off me. 1 2 3 4 5 Stay informed on how to vote in elections. 1 2 3 4 5 Make decisions based on rational evaluations of the situation. 1 2 3 4 5 The publication is a well-known news brand. 1 2 3 4 5 The publication has a good reputation for its journalism work. 1 2 3 4 5 My family and friends recognize this publication. 1 2 3 4 5 My family and friends speak highly of this publication. 1 2 3 4 5 The publication has been around for many decades. 1 2 3 4 5 The publication has been known for its national or local stature. 1 2 3 4 5 The news organization has a good standing. 1 2 3 4 5 Support journalists’ important work. 1 2 3 4 5 Pay respect to journalists’ hard work. 1 2 3 4 5 Help journalists keep their jobs. 1 2 3 4 5 Help news organizations stay in business. 1 2 3 4 5 Empower news organizations to continue monitoring power. 1 2 3 4 5 Support news organizations’ coverage of my local community. 1 2 3 4 5 Empowering news organizations to continue monitoring power. 1 2 3 4 5 Protect freedom of the press. 1 2 3 4 5 Ensure the continuance of local news. 1 2 3 4 5 Demographic Questions What is your age? o 18 - 29 (1) o 30 - 39 (2) o 40 - 49 (3) o 50 - 59 (4) o 60 - 69 (5) o 70 - 79 (6) 85 o 80 or older (7) What is your gender? o Male (1) o Female (2) o Non-binary (3) o Other (4) What is your race? o White (1) o Black or African American (2) o Asian (3) o Hispanic (4) o Other (5) Please indicate your annual household income before taxes. o Less than $10,000 (1) o $10,000 - $19,999 (2) o $20,000 - $29,999 (3) o $30,000 - $39,999 (4) o $40,000 - $49,999 (5) o $50,000 - $59,999 (6) o $60,000 - $69,999 (7) o $70,000 - $79,999 (8) o $80,000 - $89,999 (9) o $90,000 - $99,999 (10) o $100,000 - $149,999 (11) o More than $150,000 (12) What is the highest level of school you have completed or the highest degree you have received? o Less than high school (1) o High school graduate or GED (2) o Some college (including tech/vocational, some community college, associate’s degree) (3) o Four year college degree/bachelor’s degree (4) o Postgraduate or professional degree, including master’s, doctorate, medical or law degree (5) Where do you live? o Rural area (1) o Suburban area (2) o Urban area (3) Which of the following best describes your party affiliation? o Strong Democrat (1) 86 o Democrat (2) o Independent leaning Democrat (3) o Independent (4) o Independent leaning Republican (5) o Republican (6) o Strong Republican (7) 87 Appendix B: Survey - Confirmatory Factor Analysis (Study 3) Screening & Intro Q1 This survey is designed for people who are paying for a news subscription, or have paid for a news subscription in the past 12 months. Are you currently paying for a news subscription, or have you paid for a news subscription in the past 12 months? o Yes, I am paying/ I have paid for a news subscription. (1) o No, I do not pay for news. (2) o I'm not sure. (3) Q2 Consent Form Q3 We care about the quality of our data. For us to get the most accurate measures of your knowledge and opinions, it is important that you thoughtfully provide your best answers to each question in this survey. Do you commit to thoughtfully provide your best answers to each question in this survey? o I will provide my best answers (1) o I will not provide my best answers (2) o I can’t promise either way (3) Paying for News P0 Please indicate the MONTHLY $ PRICE you are paying for the following sources. (Please only fill in numbers. If you are not paying any money, please just fill in “0”): P1 National news publications (e.g. The New York Times, Newsweek) --Digital only subscription ________________________________________________________________ 88 P2 National news publications (e.g. The New York Times, Newsweek) --Print and digital subscription bundle ________________________________________________________________ P3 Local news publications in your own town/community--Digital only subscription ________________________________________________________________ P4 Local news publications in your own town/community--Print and digital subscription bundle ________________________________________________________________ P5 In the past 12 months, about how many different news publications have you paid for a subscription fee? 0 1 2 3 4 5 6 7 8 9 10 # of paid news publications PF0 We would also like to know more about your possibilities of continuing your current news subscription(s). PF1 Thinking about the primary news publication you are currently paying for, how would you rate your likelihood of continuing the subscription in the future? (e.g., your answer of 35 would indicate a 35% of chance to continue the subscription) 0 10 20 30 40 50 60 70 80 90 100 In the next 3 months In the next 6 months In the next 12 months 89 News Subscription Motivation We want to know more about what motivates you to pay for a news subscription. Think about the news subscription(s) you are paying for right now or have purchased in the past 12 months. For each of the following statements, please rate the importance of each factor in explaining why you paid for the subscription(s). Response Categories: 1 = Not at all important, 2= Slightly important, 3= Moderately important, 4= Important, and 5= Extremely important Thorough reporting. (1) Unbiased approach to news reporting. (2) The credibility of information. (3) Understand my local government’s decisions. (4) Keep up with what’s happening in my local community. (5) Get involved with my local community. (6) I found it requires too much effort to search for articles that are behind a paywall. (7) The paywall popped up when I was reading an interesting article. (8) I used up my quota for free articles. (9) Help journalists keep their jobs. (10) Support journalists’ important work. (11) Help news organizations stay in business. (12) The price is reasonable. (13) The price is affordable. (14) Get a good deal when starting/continuing my subscription(s). (15) Gather information that is valuable to my daily life. (16) Obtain information that I enjoy reading. (17) Get news articles that are rewarding to my personal growth. (18) Obtain original content that I cannot find from other news sources. (19) Please select "Slightly important" for this question. (20) Corresponding Subscales CS0 We are also interested in your life outside of the news world. Please share your opinion with us. CS1 Please tell us how well each statement describes you. Response Categories: 1= Does not describe me at all; 2=Does not describe me very well; 3= Describes me somewhat; 4= Describes me well; 5= Describes me very well. Getting very good quality is very important to me while shopping. (1) When it comes to purchasing products, I try to select the very best or perfect choice. (2) In general, I usually try to buy the best overall quality. (3) I feel like I belong in my community. (4) Overall, I am very attached to my community. (5) I like to think of myself as similar to the people who live in my community. (6) 90 I think I agree with most people in my community about what is important in life. (7) I plan to remain a resident of this community in the future. (8) If the people in this community were planning something, I’d think of it as something WE were doing rather than THEY were doing. (9) I should plan my shopping more carefully than I usually do. (10) I am impulsive when purchasing things. (11) I often make careless purchases I later wish I had not. (12) People can ask me for help if they need it. (13) People can count on my help if they have difficulties. (14) I often help people in need. (15) If someone is absent, I’m willing to take over the work. (16) I buy as much as possible at sale prices. (17) I look carefully to find the best value for the money. (18) I carefully watch how much I spend. (19) Please select "Describe me well" for this question. (20) CS2 When you decided to use your money, how important is it for you to… Response Categories: 1 = Not at all important, 2= Slightly important, 3= Moderately important, 4= Important, and 5= Extremely important. Get something that I wanted or needed for now. (1) Satisfy immediate needs. (2) Choose an option that increases my immediate comfort. (3) Act in a way that is comfortable. (4) CS3 Thinking about the news publication(s) you have paid for, please rate how helpful are these publications in satisfying your needs as listed below: Response Categories: 1 = Not at all important, 2= Slightly important, 3= Moderately important, 4= Important, and 5= Extremely important. For news that fits into my busy schedule. (1) To get latest updates on news stories. (2) To obtain news at the times I want. (3) For stories on a variety of topics. (4) To get information as quickly as possible. (5) To use my time wisely. (6) For a good variety of choices in news coverage. (7) For convenient access to news. (8) Demographics 91 D1 What is your gender? o Male (1) o Female (2) o Non-binary (3) o Other (4) D2 What is your age? o 18 - 29 (1) o 30 - 39 (2) o 40 - 49 (3) o 50 - 59 (4) o 60 - 69 (5) o 70 - 79 (6) o 80 or older (7) D3 Which race or ethnicity do you identify with the most? o White (1) o Black or African American (2) o Hispanic or Latino (3) o Asian American (4) o Indigenous/Alaskan Native (5) o Multiracial (6) o Other (7) 92 D4 Please indicate your annual household income before tax. o Less than $25,000 (1) o $25,000 to $49,999 (2) o $50,000 to $74,999 (3) o $75,000 to $99,999 (4) o $100,000+ (5) D5 What is the highest level of school you have completed or the highest degree you have received? o Less than high school (1) o High school graduate or GED (2) o Some college (including tech/vocational, some community college, associate’s degree) (3) o Four year college degree/bachelor’s degree (4) o Postgraduate or professional degree, including master’s, doctorate, medical or law degree (5) D6 Where do you live? o Rural area (1) o Suburban area (2) o Urban area (3) 93 D7 Which of the following best describes your party affiliation? o Strong Democrat (1) o Democrat (2) o Independent leaning Democrat (3) o Independent (4) o Independent leaning Republican (5) o Republican (6) o Strong Republican (7) 94 REFERENCES 95 REFERENCES Avsbury, N. (2020). A majority of Americans believe the news’ role is vital but see an increasing level of bias in coverage, poll finds. Poynter Institute. Retrieved from: https://www.poynter.org/ethics-trust/2020/a-majority-of-americans-believe-the-news- role-is-vital-but-see-an-increasing-level-of-bias-in-coverage-poll-finds/. American Press Institute. (2018). Respondents’ demographics and news behaviors. In Paths to Subscription: Why recent subscribers chose to pay for news. American Press Institute. Retrieved from: https://www.americanpressinstitute.org/publications/reports/survey- research/subscribers-appendix-1/. Barbopoulos, I., & Johansson, L. O. (2016). A multi-dimensional approach to consumer motivation: exploring economic, hedonic, and normative consumption goals. Journal of Consumer Marketing, 33(1), 75-84. Boateng, G. O., Neilands, T. B., Frongillo, E. A., Melgar-Quiñonez, H. R., & Young, S. L. (2018). Best practices for developing and validating scales for health, social, and behavioral research: a primer. Frontiers in public health, 6, 149. Berger, B., Matt, C., Steininger, D. M., & Hess, T. (2015). It is not just about competition with “free”: Differences between content formats in consumer preferences and willingness to pay. Journal of Management Information Systems, 32(3), 105-128. Chen, W., & Thorson, E. (2021). Perceived individual and societal values of news and paying for subscriptions. Journalism, 22(6), 1296-1316. Carpenter, S. (2018). Ten steps in scale development and reporting: A guide for researchers. Communication Methods and Measures, 12(1), 25-44. Contractor, S. H., & Fox, R. J. (2011). An investigation of the relationship between the number of response categories and scale sensitivity. Journal of Targeting, Measurement and analysis for Marketing, 19(1), 23-35. Copeland, L. (2014). Conceptualizing political consumerism: How citizenship norms differentiate boycotting from buycotting. Political Studies, 62, 172-186. Costello, A. B., & Osborne, J. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical assessment, research, and evaluation, 10(1), 7. Dawes, J. (2008). Do data characteristics change according to the number of scale points used? An experiment using 5-point, 7-point and 10-point scales. International journal of market research, 50(1), 61-104. 96 DeVellis, R. F. (2016). Scale development: Theory and applications. SAGE Publications. Dimmick, J., Chen, Y., & Li, Z. (2004). Competition between the Internet and traditional news media: The gratification-opportunities niche dimension. The journal of media economics, 17(1), 19-33. Dykema, J., Garbarski, D., Wall, I. F., & Edwards, D. F. (2019). Measuring trust in medical researchers: Adding insights from cognitive interviews to examine agree-disagree and construct-specific survey questions. Journal of Official Statistics, 35(2), 353-386. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of marketing research, 18(1), 39-50. Goyanes, M. (2020). Why do citizens pay for online political news and public affairs? socio- psychological antecedents of local news paying behaviour. Journalism Studies, 21(4), 547-563. Hanson, T. (2015). Comparing agreement and item-specific response scales: results from an experiment. Social Research Practice, 1, 17-25. Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the academy of marketing science, 43(1), 115-135. Höhne, J. K., & Lenzner, T. (2018). New insights on the cognitive processing of agree/disagree and item-specific questions. Journal of Survey Statistics and Methodology, 6(3), 401-417. Kalim, F. (2020). 300% growth in subscriptions for digital news and media: Time for publishers to reassess their business models?. What’s New in Publishing.com, Retrieved from: https://whatsnewinpublishing.com/300-growth-in-subscriptions-for-digital-news-and- media-time-for-publishers-to-reassess-their-business-models/. Kline, R. (2013). Exploratory and Confirmatory Factor Analysis. In Applied quantitative analysis in education and the social sciences (pp. 183-217). Routledge. Krosnick, J. A., & Presser, S. (2010). Question and Questionnaire Design. Handbook of Survey Research, 263. Lam, L. W. (2012). Impact of competitiveness on salespeople's commitment and performance. Journal of Business Research, 65(9), 1328-1334. Mitchell, A., Simmons, K., Matsa, K.E., & Silver, L. (2018). Publics Globally Want Unbiased News Coverage, but Are Divided on Whether Their News Media Deliver. Pew Research Center. Retrieved from: https://www.pewresearch.org/global/2018/01/11/publics- 97 globally-want-unbiased-news-coverage-but-are-divided-on-whether-their-news-media- deliver/. Morris, M. D., Neilands, T. B., Andrew, E., Maher, L., Page, K. A., & Hahn, J. A. (2017). Development and validation of a novel scale for measuring interpersonal factors underlying injection drug using behaviours among injecting partnerships. International Journal of Drug Policy, 48, 54-62. Netemeyer, R. G., Bearden, W. O., & Sharma, S. (2003). Scaling Procedures: Issues and Applications. SAGE. Pew Research Center. (2019). For Local News, Americans Embrace Digital but Still Want Strong Community Connection. Pew Research Center. Retrieved from: https://www.pewresearch.org/journalism/2019/03/26/for-local-news-americans-embrace- digital-but-still-want-strong-community-connection/. Pew Research Center. (2020). Misinformation. Pew Research Center. Retrieved from: https://www.pewresearch.org/topic/news-habits-media/media-society/misinformation/. Preston, C. C., & Colman, A. M. (2000). Optimal number of response categories in rating scales: reliability, validity, discriminating power, and respondent preferences. Acta psychologica, 104(1), 1-15. Ronkko, M., & Cho, E. (2020). An Updated Guideline for Assessing Discriminant Validity. Organizational Research Methods, 1, 42. Sproles, G. B., & Kendall, E. L. (1986). A methodology for profiling consumers' decision‐ making styles. Journal of Consumer Affairs, 20(2), 267-279. ten Brummelhuis, L. L., van der Lippe, T., & Kluwer, E. S. (2010). Family involvement and helping behavior in teams. Journal of Management, 36(6), 1406-1431. Theodori, A. E., & Theodori, G. L. (2015). The influences of community attachment, sense of community, and educational aspirations upon the migration intentions of rural youth in Texas. Community Development, 46(4), 380-391. Vincent, R. C., & Basil, M. D. (1997). College students’ news gratifications, media use, and current events knowledge. Journal of broadcasting & electronic media, 41(3), 380-392. Watkins, M. W. (2006). Determining parallel analysis criteria. Journal of modern applied statistical methods, 5(2), 344-346. Weijters, B., Cabooter, E., & Schillewaert, N. (2010). The effect of rating scale format on response styles: The number of response categories and response category labels. International Journal of Research in Marketing, 27(3), 236-247. 98 Worthington, R. L., & Whittaker, T. A. (2006). Scale development research: A content analysis and recommendations for best practices. The counseling psychologist, 34(6), 806-838. Yong, A. G., & Pearce, S. (2013). A beginner’s guide to factor analysis: Focusing on exploratory factor analysis. Tutorials in quantitative methods for psychology, 9(2), 79-94. 99 CHAPTER 5: TESTING NEWS SUBSCRIPTION MOTIVATIONS AS PREDICTORS In this chapter I show the nomological validity of the News Subscription Motivation construct. Specifically, I want to demonstrate how factors of NSM work as predictors and fit into the broader theoretical framework of paying for news research. In the following sections, I first develop research questions that employ NSM construct. Then I clarify the dependent measures and submitted NSM factors as independent variables in the data analysis. Finally, I present the results and discussed the findings. Research Questions Previous research of paying for news have investigated several dependent measures: intent to pay for news (Goyanes, Artero, & Zapata, 2021; Chyi, 2012), willingness to pay for news (Berger, Steininger, & Hess, 2015; Chyi, 2005), whether currently paying for news or not (Goyanes, 2020), and the monetary amount paid for news (Chen & Thorson, 2021). In the present study, I argue that NSM serves as primary driving factors in people’s decision to purchase a news subscription. The stronger the motivations are, the more likely people end up buying the news subscription. Yet, given NSM contains several factors, I also expect a more complex relationship between different subscription motivations and purchase behaviors. Moreover, the purchase behavior of news consumption can be further specified as: 1) the monetary amount people are paying for news; 2) the quantity of publications a news subscriber is paying simultaneously; and 3) the likelihood/intent to continue the subscription in the future. Therefore, the present study asks: RQ1: What are the relationships between news subscription motivations and the monetary amount people pay for news subscriptions? 100 RQ2: What are the relationships between news subscription motivations and the quantity of people’s news subscriptions? RQ3: What are the relationships between news subscription motivations and subscribers’ intent to keep their subscriptions in the future? Particularly, this present study investigates: RQ3a. What are the relationships between news subscription motivations and subscribers’ intent to keep their subscriptions in the next 3 months? RQ3b. What are the relationships between news subscription motivations and subscribers’ intent to keep their subscriptions in the next 6 months? RQ3c. What are the relationships between news subscription motivations and subscribers’ intent to keep their subscriptions in the next 12 months? Additionally, given the increasing differences between national and local news consumption (Sands, 2019), it is also necessary to examine how different motivations lead to the purchase of local and national news. Thus we ask: RQ4: Are determining motivations for subscribing to national news and local news different? Dependent Measures Paying for news. Four items were used to measure for paying for news. I asked subjects to fill in their monthly payment for (1) national news publications– digital-only subscriptions, (2) national news publications – print and digital bundle, (3) local news publications – digital-only subscription, and (4) local news publications – print and digital bundle. Paying for news in total was measured by summing up all four items; Paying for print and digital bundles was measured as the sum of items (2) and (4); Paying for digital-only subscription was measured as the sum of item (1) and (3); Paying for national news was measured by summing up (1) and (2); and Paying 101 for local news was measured as the sum of the following two items: (1) local newspaper – digital-only subscription and (2) local newspaper – print and digital bundle. Subscription of national/local news. Subscription of national/local news was indexed as categorical variables with 1=Yes and 0=No. Responses of paying $0 for national or local news subscriptions were recoded as 0, and any other dollar amount paid were recoded as 1. Quantity of paid news publications was measured by one single item. Participants were asked to report to the question “In the past 12 months, about how many different news publications have you paid for a subscription fee?” Intention to keep the news subscription had three separate measures for different time range of next 3 months, 6 months, and 12 months. Participants were asked to answer the question “Thinking about the primary news publication you are currently paying for - how would you rate your likelihood of continuing the subscription in the future?” and they were given a sliding scale to rate their likelihood of keeping the news subscription from 0% to 100%. A summary of descriptive statistics of all the dependent variables in this chapter can be found in Table 15. Table 15. Descriptive statistics of dependent variables Continuous Variables Mean SD Min. Max. Range Monthly Payment for News Subscriptions in Total 32.94 43.10 0.00 350.00 350.00 Monthly Payment for National News Subscriptions 14.89 32.02 0.00 350.00 350.00 Monthly Payment for Local News Subscriptions 18.05 26.11 0.00 240.00 240.00 Monthly Payment for Print & Digital Bundle 21.10 33.21 0.00 240.00 240.00 Monthly Payment for Digital-Only Subscriptions 11.84 22.89 0.00 200.00 200.00 Number of News Subscriptions 2.00 1.77 1.00 10.00 9.00 Subscriber Retention - Next 3 Months 79.00 30.17 0.00 100.00 100.00 Subscriber Retention - Next 6 Months 77.35 30.73 0.00 100.00 100.00 Subscriber Retention - Next 12 Months 74.58 31.91 0.00 100.00 100.00 Categorical Variables Response Percent Yes=1 50.9% Subscription of National News No=0 49.1% 102 Table 15 (cont’d) Yes=1 75.3% Subscription of Local News No=0 24.7% Results RQ1 asks the relationships between news subscription motivations and the monetary amount people pay for news subscriptions. Table 16 illustrates the multiple linear regression results of predicting how much people are paying for news subscriptions. Here I have reported the beta value (standard coefficients), p-value (statistical significance of coefficients), and explained variance (Adjusted R2 for each model). Triggered by the paywall is positively associated with how much subscribers pay for national news subscriptions (β= .12, p<.05) and digital-only subscriptions (β= .14, p<.01). Community attachment is positively related to how much people pay for local news subscriptions (β= .15, p<.01). And affordability motivation is negatively associated with the monetary amount people pay for news subscription in total (β= -.11, p<.05), local news subscriptions (β= -.11, p<.05), and print and digital bundle (β= -.11, p<.05). However, it is worth noting that the Adjusted R2 for each model are below 6%. This indicates comparing to NSM, other variables might be better at predicting how much people pay for news subscriptions. Table 16. Predicting the monetary amount subscribers are paying for news subscriptions Print News National Local News & Digital-only Subscriptions News Subscriptions Digital Subscriptions in Total Subscriptions Bundle β β β β β Age .01 -.08 .08 .08 -.12* Income .15** .10# .13* .14** .09# Education .08 .14* .00 .09 .03 Female .01 -.01 .02 .05 -.07 103 Table 16 (cont’d) White .01 -.01 .03 .02 -.01 Partisan -.06 -.06 -.04 -.06 -.04 Supporting .02 .08 -.04 -.02 .07 Journalism Journalism -.04 .01 -.06 -.04 -.01 Quality Triggered by the .08 .12* .01 .02 .14** Paywall Community .08 -.04 .15** .11# .01 Attachment Affordability -.11* -.06 -.11* -.11* -.06 Content Utility .07 .02 .08 .08 .02 2 Adjusted R 4.4% 5.6% 3.7% 5.8% 5.4% # * ** *** Note: p < .10, p < .05, p < .01, p < .001. RQ2 investigates the relationships between news subscription motivations and how many news subscriptions people are paying at the same time. Table 17 demonstrates the hierarchical linear regression results of predicting the quantity of paid news publications. Results shown that the more motivated people are in supporting journalism (β= .11, p<.05), being triggered by the paywall (β= .26, p<.001)., and being involved with the local community (β= .10, p<.05), the more news subscriptions they have. Table 17. Predicting the quantity of publications subscribers are paying for Number of Publications Model 1 Model 2 β SE β SE *** *** Age -.32 .05 -.19 .05 Income .04 .07 .04 .06 * Education .13 .09 .12* .09 Female -.11* .16 -.10* .15 * White -.09 .22 -.05 .21 Partisan -.05 .04 -.03 .04 ΔR2 14.0% *** 104 Table 17 (cont’d) Adjusted R2 12.9% Supporting Journalism .11* .03 Journalism Quality -.03 .04 Triggered by the Paywall .26*** .02 Community Attachment .10* .03 Price -.04 .03 Content Utility .00 .03 ΔR2 9.0%*** Adjusted R2 21.1% Note: # p < .10, * p < .05, ** p < .01, *** p < .001. RQ3 asks the relationships between news subscription motivations and subscribers’ intent to keep their subscriptions in the next 3 months, 6 months, and 12 months. The hierarchical linear regression results are presented in Table 18. As shown in Table 18, triggered by the paywall and price demonstrate consistent negative associations with intention to keep the news subscription in the next 3 months (Paywall: β= -.14, p<.001; Price: β= -.14, p<.01), 6 months (Paywall: β= -.13, p<.01; Price: β= -.15, p<.01), and 12 months (Paywall: β= -.12, p<.01; Price: β= -.18, p<.01). Other motivations, however, have exhibited positive relationships with the intention to keep the news subscriptions. Content utility was found to be a significant predictor for the likelihood of continuing the news subscription in the next 3 months (β= .19, p<.01), 6 months (β= .17, p<.05), and 12 months (β= .13, p<.05). Supporting journalism showed positive relations to intention to keep paying for news in the next 6 months (β= .11, p<.05) and 12 months (β= .12, p<.05). Additionally, the positive effect of journalism quality started to show at the 12 months mark (β= .13, p<.05). RQ4 asks if there are different driving factors for subscribing to national news and local news. Table 19 illustrates the logistic regression results of predicting the purchase of national 105 and local news subscriptions. Subscripting to national news was positively related to supporting journalism (β= .12, p<.01) and journalism quality (β= .18, p<.01), but negatively associated with community attachment (β= -.29, p<.001). Subscripting to local news exhibited a positive association with community attachment (β= .48, p<.001), but a negative relation to price ((β= -.12, p<.05). 106 Table 18. Predicting subscribers' retention in the future In the Next 3 Months In the Next 6 Months In the Next 12 Months Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 β SE β SE β SE β SE β SE β SE *** *** *** *** *** *** Age .28 .81 .25 .86 .28 .83 .26 .87 .26 .86 .23 .91 Income .01 1.14 .01 1.11 -.01 1.16 -.01 1.13 -.01 1.21 -.01 1.18 Education -.03 1.55 -.06 1.53 -.01 1.58 -.04 1.55 .02 1.65 -.01 1.62 * * # * * Female -.07 2.67 -.09 2.62 -.08 2.71 -.11 2.66 -.07 2.83 -.10 2.78 White .00 3.71 .01 3.64 .00 3.77 .02 3.69 -.04 3.94 -.03 3.85 # # # Partisan -.08 .64 -.05 .64 -.08 .65 -.04 .65 -.09 .68 -.05 .68 2 *** *** *** ΔR 8.6% 9.1% 7.8% 2 Adjusted R 7.5% 8.0% 6.7% Supporting Journalism .09 .44 .11* .44 .12* .46 * Journalism Quality .09 .66 .10 .67 .13 .70 *** ** * Triggered by the Paywall -.14 .42 -.13 .42 -.12 .44 Community Attachment -.02 .51 .00 .52 .02 .54 ** ** ** Price -.14 .55 -.15 .56 -.18 .58 ** * * Content Utility .19 .57 .17 .58 .13 .61 2 *** *** *** ΔR 6.8% 7.3% 7.5% 2 Adjusted R 13.4% 14.4% 13.2% # * ** *** Note: p < .10, p < .05, p < .01, p < .001. 107 Table 19. Predicting paying for national or local news subscription National News Local News B OR B OR *** Age -.50 .61 .07 1.07 Income .22 1.25 .01 1.01 *** Education .48 1.62 -.15 .86 # Female -.26 .77 -.50 .61 White -.15 .86 .45 1.57 Partisan -.10 .90 .00 1.00 ** Supporting Journalism .12 1.13 -.05 .96 ** Journalism Quality .18 1.20 -.13 .88 Triggered by the Paywall .07 1.07 -.03 .97 Community Attachment -.29*** .75 .48*** 1.62 * Price -.01 .99 -.12 .88 Content Utility .04 1.04 -.07 .93 2 Cox & Snell R 28.3% 20.8% 2 χ 168.02*** 118.11*** Note: Logistic regression beta and odds ratio are reported. # p < .10, * p < .05, ** p < .01, *** p < .001. Discussion The aim of this chapter is to test how factors for NSM work as predictors and fit into the broader theoretical framework of paying for news research. While all the other models demonstrated a good fit, the regressions predicting the monetary amount people pay for news resulted in inadequate total explained variance. Although the significant relationships can still add insight to current scholarship, there might be other unidentified variables better predict how much people are paying for news. Another possibility is that, since news subscriptions are generally cheaper than other media subscriptions, the dependent variable of how much people are paying could also have a relatively low variance. Thus, NSM has better ability explaining “paying or not paying” rather than “paying for how much.” 108 The present study also demonstrated different motivations for purchasing local and national news. Community attachment, of course, plays a vital role in subscribers’ decision to pay for local news. Yet for purchasing national news subscriptions, the key motivations are supporting journalism and journalism quality. This further reflects news audiences’ needs at different levels. At the local level, news consumption is tied up to one’s civic life. It is also possible that participants don't think local journalism has quality comparing to local news organizations. While at the national level, the importance of journalism – how it informs the public and the recognition of journalistic efforts – contributes more in consumers’ decision- making process. In terms of how many publications people are paying for, paywall, supporting journalism, and community attachment are shown to be the major driving factors. Community attachment, of course, weighs in when people add local news sources to their existing repertoires. On the other hand, the effect of paywalls can be explained by consumers’ impulsive purchase style and needs for instant gratification. In other words, it is not that paywalls have a general effect on people to get more news subscriptions, but they are particularly useful for a certain group of consumers. The most important finding of this model, therefore, is the effect of motivations for supporting journalistic efforts. While all NSM factors are important, subscribers’ needs can be fulfilled when they only pay for one or two news sources and find other information for free. People who recognize the importance of journalism are more likely to purchase more news subscriptions. The last research question, an investigation of people’s intention to keep their news subscription in the future, revealed different driving factors for short-term and long-term decision-making. While price and paywall are effective in getting people to subscribe at the beginning, the negative associations indicate they cannot ensure subscribers’ continuation of 109 purchases. It is also interesting to see how different motivation started to weigh in at different time marks. For example, content utility demonstrated effects as early as 3 months, supporting journalism mattered at the 6-month mark, yet the journalism quality factor only started showing a positive relation at the 12-month mark. This comparison also indicates the significance of journalism quality for sustaining newsrooms in the long run. 110 REFERENCES 111 REFERENCES Berger, B., Matt, C., Steininger, D. M., & Hess, T. (2015). It is not just about competition with “free”: Differences between content formats in consumer preferences and willingness to pay. Journal of Management Information Systems, 32(3), 105-128. Chen, W., & Thorson, E. (2021). Perceived individual and societal values of news and paying for subscriptions. Journalism, 22(6), 1296-1316. Chyi, H. I. (2005). Willingness to pay for online news: An empirical study on the viability of the subscription model. Journal of Media Economics, 18(2), 131-142. Chyi, H. I. (2012). Paying for what? How much? And why (not)? Predictors of paying intent for multiplatform newspapers. International Journal on Media Management, 14(3), 227-250. Goyanes, M. (2020). Why do citizens pay for online political news and public affairs? socio- psychological antecedents of local news paying behaviour. Journalism Studies, 21(4), 547-563. Goyanes, M., Artero, J. P., & Zapata, L. (2021). The effects of news authorship, exclusiveness and media type in readers’ paying intent for online news: An experimental study. Journalism, 22(7), 1720-1738. Sand, J. (2019). Local news is more trusted than national news - but that could change. KnightFoundation.org. Retrieved from https://knightfoundation.org/articles/local-news- is-more-trusted-than-national-news-but-that-could-change/ 112 CHAPTER 6: GENERAL DISCUSSION The overarching goal of this dissertation is two-fold. First, given the known predictors of paying for news are scattered in different pieces of journalism research, I aim to provide a theoretical framework that links them together. Thus, the construct of News Subscription Motivation was proposed to describe the fundamental motivations for paying for news. Second, I aim to provide a concise and easy-to-use scale for future research to measure people’s motivations for purchasing news subscriptions. A qualitative study of 22 in-depth interviews and two waves of surveys were conducted to develop and validate the new scale. In this chapter, I discuss the findings from my results and then elaborate on the major contributions from the perspectives of theoretical and practical implications. Discussion of the Scale Development Results In this dissertation, I have created a new scale of what drives people to pay for news subscriptions. The final scale resulted in 19 items and six factors in total, and each factor is labeled and defined as a news subscription motivation. These motivations are: supporting journalism, journalism quality, triggered by the paywall, community attachment, affordability, and content utility. In the following section, I discuss how news subscription motivations identified in this study link to and are different from previous research. Supporting Journalism has the highest explained variance in the present study. The items reflect subscribers’ desire to recognize journalists’ and news organizations’ efforts and help improve their financial security. Although previous research has touched on this aspect, prior analyses were descriptive in nature and emphasized journalism’s societal value as an institution. In particular, 31% of survey participants from the American Press Institute study (2018) indicated they subscribe to news in the hope to support journalism’s role in society, especially 113 when it provides “truth” and serves as a trustworthy source for the public. Reuters Institute (2020) also reported a “support” motivation in people’s reasons to pay for news. In this qualitative study, one participant suggested they pay for news subscriptions that support their political stance after President Trump’s election in 2016. Meanwhile, 51% of survey subjects from the U.S. indicated wanting to help fund good journalism. In quantitative research, Goyanes (2020) also found that if people perceive their local news organizations are not in good financial standing, they are also more likely to subscribe to local news. Despite some overlap with previous literature, my items in this factor are unique in highlighting people’s desire to improve journalists’ personal efforts and job security. Journalism Quality is also an important motivation that drives people to pay for news subscriptions. Some research addresses the quality aspect through the comparison with free information online (Reuters Institute, 2020). In the report by American Press Institute (2018), journalism quality was referred to as the publications’ accuracy, the publication of corrections, and fair reporting. In other studies, news quality is measured through newsroom investment (e.g., Chen, Thorson, & Lacy, 2005) and how well the article is written in general (Chen & Thorson, 2021). However, the present study highlighted three aspects that drive individuals to pay for news: credibility, an unbiased approach to reporting, and thorough reporting. Extensive scholarship has been done on this topic (e.g., Meyer, 1988; Gaziano, 1986; Gantz, 1981), but there has been little linkage between credibility and paying for news. Credibility studies also suggest multiple dimensions in measuring this concept, such as accuracy, authenticity, trustworthiness, fairness, unbiased, and completeness (Appleman & Sundar, 2016; Meyer, 1988; Gaziano, 1986). In this dissertation, the journalism quality measure had two items that touch on credibility: One item about an unbiased approach to reporting, and the other measure is a single 114 item with a broad label of “credibility of information.” Future research could usefully utilize more detailed credibility measures and further explore the relationship between news credibility and paying. Triggering by the paywall was the third factor that was identified as a news subscription motivation. This factor is consistent with previous research. According to American Press Institute (2018), about 50% of digital subscribers were triggered to pay for a news subscription when they hit a paywall meter. In my results, the items reflected subscribers’ desire to continue reading (the paywall popped up in midst of reading an article, or used up the quota for free articles), and desire to avoid extra effort getting around the paywall. While testing the convergent validity of the final NSM scale, the “triggered by the paywall” factor also showed positive correlations with instant gratification and impulsive consumer style from marketing research (Barbopoulos & Johansson, 2016; Sproles & Kendall, 1986). Thus, my findings here suggest that paywalls are particularly effective for people who demonstrate stronger needs for instant gratification - and in the news consumption context, stronger needs for continuing reading news articles on the spot. Factor 4 is labeled as Community Attachment. Goyanes (2020) found that the more local residents feel attached to their community, the more likely they would subscribe to their local newspaper. American Press Institute (2018) also noted that when participants explained reasons to subscribe in their own words, 30% of them expressed the desire to have access to local news and to stay connected to the community. Items in my results also reflected such attachment (get involved with the local community) but also demonstrated individuals’ surveillance needs for keeping up with what is happening and understanding local governments’ decisions. 115 It is also important to note that community attachment was not in the dimensions that emerged from the qualitative interviews. Items loaded in this factor are from dimensions of surveillance and being a good citizen, but have common wording “local” and “community” in the motivation statements. This suggests the importance of future and deeper examinations of drivers of community attachment and how attachment influences people’s motivation to pay for news. Affordability is the fifth factor in the NSM scale. It reflects news subscribers’ emphasis on getting a reasonable price or discount when starting or continuing a news subscription. Previous studies have also found people’s perception and evaluation of how expensive the product is to them before making the purchase decision, and the same is found in the willingness to pay for news (Berger et al. 2015). The more inexpensive consumers perceive a product, the more likely they will pay for it. Similar factors were labeled as “price/value-conscious” consumers in the Consumer Style Inventory (Sproles & Kendall. 1986) and “thrift” in the Consumer Motivation Scale (Barbopoulos & Johansson, 2016). In the present study, I decided to label this factor “Affordability” instead of “Price.” Based on the definition in economics, price is determined by and reflects the relationship between supply and demand (Stigler, 1966). However, the items loaded in this factor mostly demonstrated people’s desire to pay for news at a reasonable, affordable monetary amount or receive a discount. Therefore affordability is a better name than price. The final factor is content utility. The common theme of items reflected consumers' motivation that getting a news subscription is useful to them in some way (e.g., getting information valuable to daily life, obtaining original content, etc.). Although a previous study found that 40% of current news subscribers noticed numerous interesting and useful articles 116 (American Press Institute, 2018), researchers noted it to be a background factor that serves as a precondition for people to subscribe. However, the statistical results of the NSM scale showed a direct linkage between people’s motivation to pay for news and their perception of how useful or interesting the news content might be. The identification of the content utility factor also advances theories in media economics and content quality. Lacy (2000) uses the term “expected utility of media” to describe audiences’ perception of how the media information will meet their needs and wants. He also posits that increased utility to audiences would result in increased use of the news organization’s product, and thus improve the news organization’s subscription and advertising revenue. However, this proposition has never been tested by empirical research. Thus, the content utility factor in the NSM scale provides evidence that supports Lacy’s perspective. In sum, the factors identified in the NSM scale are consistent with previous research on paying for news. The final scale of NSW also advances the current scholarship by adding a theoretical framework of consumer motivation that links together different predictors of paying for news. Compared to existing scales that examine consumers’ decision-making styles and motivation to purchase, NSM is also unique in its context of news consumption. The scale development process was also subjected to rigorous methodology standards, and the final scale is concise and easy to use in future research. Discussion of NSM as Predictors of News Paying Behavior In Chapter 5, I aimed to test the nomological validity of the NSM scale. In doing so, I employed several regressions to investigate how different NSM factors predict subscribers’ news paying behaviors. 117 Explained Variance. Only one model failed to yield an acceptable explained variance in a targeted dependent variable. After controlling for demographic variables, multiple linear regression results showed only 3.7% to 5.8% variance for models using NSM factors to predict how much subscribers are paying for news subscriptions. One possible explanation for this is the current NSM scale is focused on current subscribers’ reasons why they subscribe, which measures a question of “pay or not pay,” rather than how much they pay. Another reason might be that news subscriptions are generally cheaper than other media products with monthly costs ranging roughly from $1 (e.g., most starter deals) to $10 (e.g., The Washington Post All-Access subscription). When calculating the regressions, the low range of subscription prices also affect the variability of the dependent variable, which might also result in the low explained variance of the regression models. In general, factors in NSM exhibited good predictability for other news paying behaviors, with total explained variance ranging from 13.2% to 28.3% (after including demographics). These news paying behaviors included how many publications subscribers are paying for at the same time, whether they are paying for national or local subscriptions, and the likelihood of maintaining their primary news subscription in the next 3-12 months. The R-square changes for adding NSM factors were also statistically significant. It is also important to note that in some regression models, demographic variables accounted for more explained variance than NSM factors. For instance, in the model of predicting the quantity of paid news subscriptions, demographic variables accounted for 14% of the variance, and news subscription motivations accounted for 9% of the variance. This pattern is also found in models predicting subscription retention in the next three, six, and 12 months, with R-squared differences between two blocks of predictors ranging from 0.2% to 1.8%. Similarly, 118 Lacy and Fico (1991) also found that the city population accounted for more variance than the index of journalism quality in predicting next year’s newspaper circulation. These findings suggest that most of the news paying behaviors have stronger associations with traditional structural variables than with news content or subscription motivations. Effect size. I calculated Cohen’s f2 to examine the effect sizes of multiple regression models (results in Table 16-18). According to Cohen (1988), the effect size is considered large when f2 is equal to or greater than .35; and the threshold for medium and small effect sizes are .15 and .02, respectively. For the model that predicts how much people pay for news subscriptions, the overall model effect sizes are small with Cohen's f2 ranging from .04 to .06. For the model predicting the number of publications subscribers are paying for, the overall model yielded a medium effect size (Cohen's f2 = .27), and NSM factors had a small effect size (Cohen's f2 =.10). Models for subscriptions retention in the next 3-12 months also demonstrated medium effect sizes (Cohen's f2 ranged from .15 to .17), and NSM factors had small effect sizes as well (Cohen's f2 ranged from .07 to .08). The small effect sizes of NSM factors further indicate that our findings on news subscription motivations are not the panacea to boost subscriptions, but provide more information for news organizations to move forward in this direction. Predictor Evaluation. I also calculated odd ratios to examine effect sizes for logistic regression models that predict people’s national/local subscriptions (results in Table 19). The odd ratios allowed me to compare the effect sizes among different predictors. According to Cohen’s (1988) guidelines, odds ratios below 1.44 are considered very small and odds ratios below 2.48 are considered small. 119 The biggest driver for people to pay for national news is education (OR=1.62), but the effect size is considered small based on Cohen (1988). Significant NSM predictors all yielded very small effect sizes with OR ranging from .75 to 1.20. Journalism quality and supporting journalism were found to be the stronger motivation for people to get national news subscriptions. The quality factor suggests that people are drawn to national news publications for better content quality in terms of credibility, fairness, and comprehensiveness. The supporting journalism motivation demonstrated people’s recognition of journalism’s importance and how national news organizations play a part in it. On the other hand, the strongest motivator for subscribing to local news is community attachment, which indicated a small effect size (OR= 1.62), followed by being Caucasian (OR=1.57). This once again shows the strong relationship between how attached people feel toward their community and the need for them to get local news. In previous studies, researchers also found that people who are highly active in their local community also feel more connected to their local newspaper and consume more news about the development of the local community (Thorson, Chen, & Lacy, 2019; 2020). This finding is particularly important for local news organizations to survive, as more than 2,000 local newspapers closed since 2008, and the loss of local news organizations will pose danger to democratic practices within local governments (The Washington Post Magazine, 2021). It is interesting to see different motivations for getting national or local news subscriptions, and how community attachment is crucial for local news organizations to perform well financially. Theoretically speaking, competition for local news organizations is fiercer as they face several layers of intercity or “umbrella” competition (Lacy & Simon, 1993). George and Waldfogel (2006) also found that when The New York Times entered local markets, the 120 circulation of local newspapers dropped by 16% for highly-educated readers and 7% for less- educated residents. Thus, for local news organizations to survive, it is essential for them to highlight their connection to the community and local residents. I used beta weights and the significance of t-statistics to evaluate predictors in the multiple regressions. Higher-income and less desire for affordability are found to be the primary drivers of the total monetary amount people spend on the news. To get people to pay for more publications, triggered by the paywall, supporting journalism showed a positive impact. Being in older age groups showed consistent importance in determining to keep subscriptions in the next 3, 6, and 12 months. For subscriber retention in three months, content utility demonstrated a positive impact, whereas triggered by the paywall and affordability exhibited negative associations. Supporting journalism started to show a positive influence on retention in the next six months, and journalism quality became important to keep subscribers in the next year. The differences in what predicts retention across time, and differences in explained variance accounted for demographic and NSM, reveal different mechanisms in consumers’ short- term and long-term decision-making. Although triggered by the paywall and affordability motivate news audiences to get the subscriptions, they are negatively associated with retention in the future. Instead, my results showed journalism quality is crucial for keeping subscriptions in the long run. This is also supported by previous research at the macro level. Previous studies employed newsroom investment as an index measure for content quality and found that journalism quality positively related to newspaper circulation at year 1 and year 5, but the relationship declined after five years (Lacy & Fico, 1991; St. Cry, Lacy, & Guzman-Ortega, 2005). Thus, it is crucial for newsroom managers to continue investing in journalism quality in the long run. 121 Theoretical Implications Before summarizing the contribution of the NSM construct, it is important to clarify what NSM is not. Although both constructs suggest individuals’ needs and wants are fundamental to media consumption, the construct of News Subscription Motivation is distinct from the traditional model of Uses and Gratifications (U&G). U&G explains why people choose certain media sources and how often they use them (Rubin, 2009). But NSM goes beyond media use and explains why people pay for news subscriptions, which involves an additional financial commitment that is not required by free information. The difference between NSM and U&G is also reflected in my scale development process. In the qualitative interviews, multiple participants mentioned their surveillance needs as one of the reasons to pay for news, which is also a consistent dimension in various U&G research (e.g., McDonald & Glynn, 1984; Vincent & Basil, 1997; Diddi & LaRose, 2006). However, the surveillance dimension was not retained in the scale after the reduction process done by EFA. Surveillance is a need that many news audiences value, but does not predict paying for a subscription. People who demonstrate high surveillance needs may still show higher levels of news use, but it is possible for them to satisfy their needs through a free news source. Media habit is another important factor that is essential in media consumption but not reflected in the NSM construct. Previous research also noted a positive relationship between news habit strength and the monetary amount people pay for news (Chen & Thorson, 2021). However, habit functions separately from motivation and gratification in affecting human behaviors (Diddi & LaRose, 2006). According to LaRose (2010), habit depicts the unconscious aspect of human behavior, and is automatic and repetitive. Other scholars also contend that habit is a function of conserving cognitive and mental energy, and once the habit is formulated, it 122 requires no effort (Verplanken & Orbell, 2003). On the other hand, the development of NSM started by asking subscribers why they decided to pay for news, which investigates consumers’ conscious intentions to satisfy their needs through paying for news subscriptions. Thus, although habit is an important determinant for media consumption behavior, it is beyond the scope of this study. A comparison between news habit strength and news subscription motivations can be included in future research. The construct of News Subscription Motivation was presented as a central conceptual framework that unifies the drivers of paying for news subscriptions: motivations for supporting journalism, getting quality news, getting around paywalls, maintaining a strong connection with the local community, paying for a reasonable price, and utilizing the content in some way in one’s daily life. By conceptually explicating NSM, I offered a theoretical foundation for understanding the value of news subscriptions from the consumers’ perspective, which contributes to the current discussion of news product management in journalism research (e.g., Gordon, 2020; Royal & Kiesow, 2021). The NSM construct also provided a solid scale for further research on news consumers. This measurement scale is innovative and unique as it is specially designed in the context of news consumption. Rather than the general measures of paying for news and overall audience, NSM made an important distinction between subscription and donation, and solely focused on the existing consumers who are already paying for news subscriptions. Methodologically speaking, this scale was also subjected to an additional, rigorous qualitative study to generate and validate dimensions and items, which is also an advancement compared to existing scales of Consumer Style Inventory (Sproles, & Kendall, 1986) and Consumer Motivation (Barbopoulos, & Johansson, 2016). In the two following quantitative studies, I also followed best practices of 123 scale development suggested by communication, marketing, as well as social and behavioral research. Moreover, I also provided strong evidence for content, convergent, discriminant, and nomological validity, and thus I suggest the NSM scale to be a useful measurement tool for future research. The factors generated from NSM construct also provide useful variables for further examining purchase behaviors and consumers’ decision-making process when it comes to paying for news. In Study 3, I demonstrated the ability of NSM to predict news purchase behaviors such as the number of publications people are paying for, types of news subscriptions people are getting, and subscribers’ intention to keep the news subscription in different time frames. Moreover, factors of NSM can be also used to examine the mechanism of paying for news in the consumer’s decision-making process. It would also be interesting to investigate how different motivations might influence consumers’ decisions at each stage of the decision-making process. Practical Implications From a practical perspective, this project contributes to the industry in several ways. First, understating NSM is crucial for news organizations to navigate their financial crisis. With news organizations still fighting for their places in digital advertising, more and more news publications adapted to the all-subscription model to secure subscription revenue. In addition, previous research on newspaper revenues also demonstrated the positive associations between advertising sales and circulation revenue (e.g., Chen, Thorson, & Lacy, 2005). Therefore, understanding what motivates people to subscribe is fundamental for the financial outcomes of the news industry. Second, the final scale provides a useful tool for individual news publications to test among their own subscribers. The results of testing the NSM scale can provide useful insights to 124 understand their existing consumers and prevent them from canceling their subscriptions. The consumer insights also can provide guidance for news organizations to build business strategies to attract and convert potential consumers. For each individual newsroom, testing the NSM scale is also useful for the marketing departments to figure out strategies to design promotion messages and encourage subscription behaviors. Third, this project also provides a strong argument for newsrooms to prioritize journalism quality for their sustainability in the long run. As shown in chapter 5, journalism quality is the most important factor for subscribers to keep the subscriptions in the next 12 months. In previous scholarship, Lacy, Stamm, and Martin (2014) also note how short-term oriented decisions reduced the content quality, and then further led to the downfall of financial performance. Thus, it is essential for newsroom managers to adapt to long-term thinking and have a strong commitment to journalism quality. Finally, the significance of supporting journalism motivation also calls for media literacy training for Americans. Being one of the most important factors in NSM, it indicates people’s willingness to financially support news organizations. It is also interesting to note that, in a recent study that employed a sample of the general population, no relationship was found between paying for news and individuals’ perception of journalism’s contribution to society (Chen & Thorson, 2021). While the specifics of whether supporting journalism is a motivation for personal or public benefits, we might argue that the willingness to support, essentially, stems from people’s understanding of why journalism is important work. Therefore, more effort needs to be made to educate the public about the importance of the press. 125 Limitations This project, like other research, has limitations. First, the construct of News Subscriptions Motivation was developed focusing on existing subscribers and news audiences who had a history of subscribing in the U.S. Thus the scale is not applicable to the general U.S. population, and may not be directly useful to newsrooms in other countries. Given our samples leaned largely toward older people and Caucasians, news subscription motivations may also differ if given a sample with different demographics. Second, NSM only examined the more salient reasons for paying for news by directly asking participants what are the important factors that motivate them to pay. Some other factors might still have an impact on consumers’ decision to purchase the subscription, but their effects may be unconscious and unidentified by individuals. For instance, Chen and Thorson (2021) showed that paying for news is also positively related to people’s need to reinforce their social identity. NSM does not include possible underlying factors that news subscribers are not aware of themselves. Finally, the samples used in the scale development process were vulnerable to sampling bias and the results are thus can be criticized. Although I tried to match the demographics quotas from previous research, the samples are only a rough representation of current news subscribers in the U.S. In a practical sense, the final scale should be more useful while being tested in individual newsrooms, and assumptions of its generalizability should be cautioned. 126 REFERENCES 127 REFERENCES Appelman, A., & Sundar, S. S. (2016). Measuring message credibility: Construction and validation of an exclusive scale. Journalism & Mass Communication Quarterly, 93(1), 59-79. American Press Insititute. (2018). Paths to Subscription: Why recent subscribers chose to pay for news. American Press Institute. Retrieved from: https://www.americanpressinstitute.org/publications/reports/survey-research/subscribers- appendix-1/ Barbopoulos, I., & Johansson, L. O. (2016). A multi-dimensional approach to consumer motivation: exploring economic, hedonic, and normative consumption goals. Journal of Consumer Marketing, 33(1), 75-84. Berger, B., Matt, C., Steininger, D. M., & Hess, T. (2015). It is not just about competition with “free”: Differences between content formats in consumer preferences and willingness to pay. Journal of Management Information Systems, 32(3), 105-128. Chen, R., Thorson, E., & Lacy, S. (2005). The impact of newsroom investment on newspaper revenues and profits: Small and medium newspapers, 1998–2002. Journalism & Mass Communication Quarterly, 82(3), 516-532. Chen, W., & Thorson, E. (2021). Perceived individual and societal values of news and paying for subscriptions. Journalism, 22(6), 1296-1316. Cohen, J. (1988). The effect size. Statistical power analysis for the behavioral sciences, 77-83. Diddi, A., & LaRose, R. (2006). Getting hooked on news: Uses and gratifications and the formation of news habits among college students in an Internet environment. Journal of Broadcasting & Electronic Media, 50(2), 193-210. Gantz, W. (1981). The influence of researcher methods on television and newspaper news credibility evaluations. Journal of Broadcasting & Electronic Media, 25(2), 155-169. Gaziano, C., & McGrath, K. (1986). Measuring the concept of credibility. Journalism Quarterly, 63(3), 451-462. George, L. M., & Waldfogel, J. (2006). The New York Times and the market for local newspapers. American Economic Review, 96(1), 435-447. Gordon, R. (2020). Yes, Product Thinking Can Save Journalism. Six Reasons Why News Media Need Product Thinkers. Knight Lab Northwestern University, Retrieved from: 128 https://knightlab.northwestern.edu/2020/11/23/product-thinking-can-save-journalism- product-management-news-media/ Goyanes, M. (2020). Why do citizens pay for online political news and public affairs? socio- psychological antecedents of local news paying behaviour. Journalism Studies, 21(4), 547-563. Lacy, S., & Simon, T. F. (1993). The economics and regulation of United States newspapers. Ablex Pub. Lacy, S. (2000). Commitment of financial resources as a measure of quality. Measuring Media Content, Quality and Diversity: Approaches and Issues in Content Research. Turku, Finland: Turku School of Economics and Business Administration, 25-50. Lacy, S., & Fico, F. (1991). The link between newspaper content quality & circulation. Newspaper Research Journal, 12(2), 46-57. Lacy, S., Stamm, M., & Martin, H. (2014). Short-run decisions threaten papers' long-run viability. Newspaper Research Journal, 35(4), 6-20. LaRose, R. (2010). The problem of media habits. Communication Theory, 20(2), 194-222. McDonald, D. G., & Glynn, C. J. (1984). The stability of media gratifications. Journalism Quarterly, 61(3), 542-741. Meyer, P. (1988). Defining and measuring credibility of newspapers: Developing an index. Journalism Quarterly, 65(3), 567-574. Reuters Institute. (2020). Digital News Report 2020. Retrieved from: Reuters Institute for the Study of Journalism, Retrieved from: https://www.digitalnewsreport.org/survey/2020/how-and-why-people-are-paying-for- online-news/ Royal, C., & Kiesow, D. (2021). From Boundary to Bridge and Beyond: The Path to Professionalization of Product Roles in Journalism. Journalism Studies, 22(11), 1546- 1565. Rubin, A. M. (2009). Uses and gratifications. The SAGE handbook of media processes and effects, 147-159. Sproles, G. B., & Kendall, E. L. (1986). A methodology for profiling consumers' decision‐ making styles. Journal of Consumer Affairs, 20(2), 267-279. 129 St. Cyr, C., Lacy, S., & Guzman-Ortega, S. (2005). Circulation increases follow investments in newsrooms. Newspaper Research Journal, 26(4), 50-60. Stigler, G. J. (1966). Theory of price. Macmillan, New York. The Washington Post Magazine. The Lost Local News Issue. The Washington Post Magazine, Retrieved from: https://www.washingtonpost.com/magazine/interactive/2021/local-news- deserts-expanding/ Thorson, E., Chen, W., & Lacy, S. (2019). Civic participation and connectivity with a metro newspaper. Newspaper Research Journal, 40(3), 346-361. Thorson, E., Chen, W., & Lacy, S. (2020). National and local news consumption and indices of community participation in an American civic-journalism news environment. Journalism Studies, 21(4), 443-458. Verplanken, B., & Orbell, S. (2003). Reflections on past behavior: a self‐report index of habit strength 1. Journal of applied social psychology, 33(6), 1313-1330. Vincent, R. C., & Basil, M. D. (1997). College students’ news gratifications, media use, and current events knowledge. Journal of broadcasting & electronic media, 41(3), 380-392. 130