INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The qnality of this reproduction is dependent upon the quality of the
copy submitted. Broken o r indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margin*,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright m aterial had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand corner and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. H igher quality 6” x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.

A Bell & Howell Information Company
300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA
313/761-4700 800/521-0600

ASSESSMENT OF ATTITU DES TOWARD DIETARY CHANGE
OF PARTICIPANTS IN THE MICHIGAN EXPANDED
FOOD AND N U T RI TIO N EDUCATION PROGRAM

By
Anne Seymour Murphy

A DI SSE RTATION

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

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1987

UMI Number: 9537249

Copyright 1987 by
Murphy, Anne Seymour
All rights reserved.

UNI Microform 9537249
Copyright 1995, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.

UMI
300 North Zeeb Road
Ann Arbor, MI 48103

ABSTRACT
A S SESSMENT OF ATT I T U D E S TOWARD DIETARY CHANGE
OF P AR TIC I P A N T S IN THE MIC HI GAN EXPANDED
FOOD AND N U T R I T I O N EDUCATION PROGRAM
By
Anne Seymour Murphy

As se ssment of attitudes toward dietary change has not
been conducted in pri or evaluations of the Expanded Food
and N utrition E d u c a t i o n Program

( E F N EP) . This research was

conducted in the M i c h i g a n EFNEP to determine

(1) if

attitudes toward d i e t a r y change of participa tin g ho m e ­
makers improved as a result of prog ra m participation,
(2) how pretest a tti t u d e scores w ere related to change in
food recall scores,

(3)

if change in attitude and food

recall scores of h o m e mak er s were related,
change of homemakers was p redicted by:
scores of homemakers or instructors,
or change scores of homemakers,

(4)

if attitude

pretest attitude

food recall pretest

instructors'

years of

experience,

locus of control pretest or change scores of

homemakers,

and

be predicted by:
homemakers,

(5)

if die tar y

(food recall)

change could

locus of control changes scores of

instructors'

years of experience,

change or pret es t food recall scores,

attitude

or pretest attitude

scores of homemakers or instructors.
A Likert scale was used to assess attitudes toward
dietary change and locus of control.

A 24-hour food recall

was used to m e a s ure self-re por te d d i e ta ry intake. Content

and contruct v ali d i t y and reliab il ity of the attitude and
locus of control

instruments was established.

collected from 195 low-income homemakers,

Data were

47 program aides

and a comparison group of 66 wo men enrolled in the S u p p l e ­
mental Food P r o g ra m for Women,

Infants,

and Children.

Results of A NO VA and t-test analyses indicated that
attitudes toward die tar y change of EFNEP participants
improved s ignificantly

(p <.001)

from pre- to po sttesting

and that posttest attitude scores were significantly
higher

(p <.001)

than those of the co mpa ri son group.

Subjects with high pretest attitude scores did not have
significantly grea ter change in food recall

scores than

homemakers with low pretest attitude scores

(p=.126).

Change in attitude and food recall scores of homemakers
were not correlated.

Results of multiple regression

analyses indicated that pretest food recall

scores

accounted for 57.88% of the variance in food recall change
of homemakers.
variables,

Due to co rr elations between the independent

two separate reg ression runs were conducted to

find predictors of attitude change.

Results of the

separate r eg ression analyses indicated that va ria nc e in
attitude change could be predicted by p retest locus of
control

(9.83%), change in locus of control

pretest attitude scores of homemakers

(51.35%), and

(40.28%).

Copyright by
ANNE SEYMOUR MURPHY
1987

To my parents,

Seymour and Harriet

And my children,

Lila and Katiejane

v

ACKN OW LED GME NT S

I would
advisor,

like to acknow led ge the assistance of my

Dr. Jenny Taylor Bond,

whose trust in my abilities

and respect for me as a person provided the enc ouragement I
needed to complete this research and dissertation.
director of this research,
the planning stages,
processing data,

As the

she provided food for thought in

timely suggestions for collecting and

a thorough r e v ie w of dissertation drafts,

and unending support in all phases of the project.

Dr.

Bond is a very "human" scholar.

I am grateful to her for

guidance related to my doctoral

studies,

but more

importantly for her influence and concern in my development
as a person.
I also app reciate the individual and combined efforts
of my other commit tee members,
Nierman,

Judith Anderson,

guidance in planning,
project.

A special

Drs.

Sharon Hoerr, Linda

and Fred Ignatovich, who provided

implementing,

and evaluating this

thanks to Dr. Linda Nierman, who allowed

me to di rect this resear ch p r o jec t with freedom and
flexibility.
vi

The contribution of the Michigan EFNEP extension home
economists and p r o gra m aides,

and Doris Showerman and the

Ingham County WIC Department staff, who assisted in the
collection of research data is also appreciated.

A special

thanks to Win -Jing Chung for his prompt and efficient
assistance with data analysis and to Bethany Ledford,
Johns,

Kim

and Janeen Hunt for their role as support staff.

I sincerely appreciate the support and suggestions
from friends,

especially Kendra Anderson, Barb Mutch,

Lynch,

and John Kallas.

Cathy,

for providing spiritual guidance at appropriate

Pat

I also am thankful to my sister,

intervals during this undertaking and to my sisters Maria
and Carol for providing encouragement.
I am grateful for the financial support, provided by
the Michigan Expanded Food and Nutrition Education Program,
the Michigan State University Department of Food Science
and Human Nutrition,

the College of Human Ecology,

and the

George D. Searle and Mead Johnson Scholarships which were
awarded through the Ame rican Dietetic Association.
I am grateful

to my father,

to pursue a doctoral degree.
provided a continual

Seymour, who inspired me

His confidence in my abilites

source of "silent" support and his

lifetime of achievements provided footsteps to follow.

I

sincerely appreciate my mother's continual assistance wit h
my family responsibilites and for providing emotional and
financial support, which allowed me to channel my efforts
into writing this dissertation.

I would also like to acknowledge the support provided
by my husband, Luke, whose willi ngn ess to assume the roles
of mother and father to our children,
me, expedited my efforts
Lastly,

I would

and c heerleader to

in completing this dissertation.

like to a cknowledge the roles of my

daughters, Lila and Katiejane.
giving me the time and peace

Lila was a good sport about

I needed to "do my homework"

and Katiejane taught me the meaning of "strength" and how
it can be applied to accomplish the "impossible".
To these people,
and encouragement,

and others who provided suggestions

I am sincerely grateful for their

contributions in assis tin g me in completion of this
research and dissertation.

viii

TABLE OF CONTENTS

LIST OF T A B L E S ............................................. xiii
LIST OF F I G U R E S .............................................. X V
I N T R O D U C T I O N ................................................... 1
Statement of the P r o b l e m ............................... 2
Research O b j e c t i v e s ..................................... 6
Research H y p o t h e s e s ..................................... 7
R E VIE W OF THE L I T E R A T U R E ..................................... 9
As se ssment of Change in Attitudes of EFNEP
P a r t i c i p a n t s ............................................. 9
Assessment of Attitude Change of Participants in
Nutrition Educat io n P r o g r a m s ......................... 11
The Correlational Re la tionship Between Attitude
and Behavior (Dietary C h a n g e ) ........................ 16
Correlational I nconsistency Between Att itude and
Behavior (Dietary C h a n g e ) .......................... 17
Correlational C on si ste nc y Between Attitudes and
Behavior (Dietary C h a n g e ) .......................... 21
Summary: The Rela ti ons hip B etween Attitudes and
B e h a v i o r .............................................. 29
Models of the R e la ti ons hip Between Attitude and
B e h a v i o r .................................................31
Model One: Attitudes Predict Behavior

(A-B)..... 31

Model Two: Behavi ors Predict Attitude

(B-A)..... 44

Locus of Control as a Predictor of Dietary
C h a n g e .................................................48
ix

Job Exp eri en ce as a Predictor of Dietary C h a n g e . 52
Attitudes of Instructors as a Predictor of
Dietary C h a n g e ....................................... 53
Food Recall as a Predictor of Dietary C h a n g e . . . . 54
M E T H O D S ........................................................61
Approval

to C o n duc t the S t u d y ........................ 61

Sub j e c t s .................................................62
Research S i t e s ..........................................62
Program D e s c r i p t i o n .................................... 63
Research Design:

Hypotheses T e s t i n g ................. 64

Development of the Attitude I n s t r u m e n t ............. 67
Development of the Locus of Control

I n s t r u m e n t . ... 72

Development and Justification for Use of the
24-Hour Food Recall I n s t r u m e n t ....................... 73
Training of Data C o l l e c t o r s .......................... 74
Data C o l l e c t i o n ........................................ 75
RESULTS AND D I S C U S S I O N ...................................... 78
Distribution of Scores and R e l i a b i l i t y ............. 78
The Effect of P r e t e s t i n g .............................. 80
The Effect of EFNEP Pa rticipation on Attitudes
Toward Dietary Change of Enrolled H o m e m a k e r s ...... 81
Equivalence of Pretest S c o r e s ......................81
Results of Item Analysis of Homemaker
Attitude S u r v e y s ..................................... 83
Results of G rou p Data Analyses of Homemaker
Attitude S c o r e s ...................................... 84
Change in A ttitudes of Experimental vs.
Comparison G r o u p s .................................... 86
x

The Effect of Pret es t Attitude Scores on Food
Recall Change of EFNEP H o m e m a k e r s

90

The R ela tionship Bet wee n Attitude and Food
Recall Change Scores of EFNEP H o m e m a k e r s ........... 91
Correlational A n a l y s i s .............................. 91
The Ability of Independent Variables to Predict
Chang e in Att it ude and Foor Recall S c o r e s .......... 94
Regression A n a l y s i s ................................. 94
Results of Data Analysis Related to A t t r i t i o n . ... 100
S ummary of R e s u l t s ....................................102
STRENGTHS AND L I MIT AT ION S OF THE S T U D Y .................. 104
Vali d i t y and R el ia bil ity of the M easurement
I n s t r u m e n t ............................................. 104
Threats to Internal V a l i d i t y ......................105
Threats to External V a l i d i t y ......................109
Weaknesses of the Likert S c a l e ......................112
Rationale for Use of the Likert S c a l e ............. 114
R E C O M M E N D A T I O N S ............................................. 116
Implementation of Nut r i t i o n Educat io n P r o g r a m s ......... 116
Instructor T r a i n i n g .........................................118
Ass ess me nt of Attitudes: Future Research
I n v e s t i g a t i o n s .............................................. 118
Use of Multiple Factors To Predict Dietary Change:
Future Research I n v e s t i g a t i o n s ........................... 120
APPPEND IC ES
APPENDIX A: A s s e ss men t of the R el ationship between
Attitude and Behavior (A Summary of the
Li te rature C i t e d ) .............................123
APPENDIX B: Approval to Conduct the R e s e a r c h ........... 127
Con sen t Form: Program P a r t i c i p a n t s ...... 128
Cons en t Form: Program I n s t r u c t o r s ........129
xi

APPENDIX C: Demographic I n f o r m a t i o n . .................... 130
APPENDIX D: Program D e s c r i p t i o n .......................... 132
Data Col lec ti on Procedures (An
Inservice H a n d o u t ) ......................... 134
Effective Aide Techniques (Newletter
Article, September 1 9 8 6 ) .................. 135
APPENDIX E: Pr eliminary Research Interview Script
and R e s u l t s ....................................136
APPENDIX F: Test Items for R e v i e w ........................ 141
APPENDIX G: Specifications for Development of the
Likert S c a l e .................................. 146
APPENDIX H: Attitude Qu estionnaire Revi ew F o r m .........147
APPENDIX I: EFNEP Survey

(First P i l o t ) .................. 148

APPENDIX J: Pilot Test Instructions for Program A i d e s . 156
APPENDIX K: R ev iew Form:

Program A i d e s .................. 158

APPENDIX L: EFNEP Survey

(Second Pilot-Two F o r m s )

159

APPENDIX M: Pilot Test Discrimination and Difficulty
I n d i c e s ........................................ 165
APPENDIX N: EFNEP Survey

(Final I n s t r u m e n t ) ............ 167

APPENDIX O: Locus of Control

I n s t r u m e n t s ................169

APPENDIX P: 24-Hour Dietary Food Recall I n s t r u m e n t .... 174
Scoring Table for 24-Hour D i e t ............. 175
APPENDIX Q: Pretest Item Analysis Results: Homemaker
Attitude S u r v e y ............................... 176
Posttest Item Analysis Results: Homemaker
Attitude S u r v e y ............................... 177
Posttest Item Analysis for H o m e m a k e r s
178
Pretest Item Analysis Results: Aide
Attitude S u r v e y ............................... 180
Item Analysis Results: Aide Posttest
S u r v e y ......................................... 181
Posttest Item Analysis for Program A i d e s . . 182

LIST OF REFERENCES

184

LIST OF TABLES

1.

Pilot test means and ranges of attitude scores for
upper and lower scoring g r o u p s ........................ 70

2.

Results of the second pilot test of the attitude
i n s t r u m e n t ................................................71

3.

Distribution of attitude, locus of control, and
food recall s c o r e s ...................................... 79

4.

Posttest scores of p retested and non-pretested
g r o u p s .................................................... 80

5.

Results of t-tests to deter min e the effect of
p r e t e s t i n g ................................................81

6.

Comparison of p r e tes t attitude means of the
experimental and co mparison g r o u p s .....................82

7.

Initial equality of comparison vs. experimental
group pretest at titude s c o r e s ......................... 82

8.

Homemaker mean item scores for the pre- and
posttest attitude s u r v e y ............................... 83

9.

Attitude pre- and posttest results of the
experimental g r o u p ...................................... 84

10. Differences in pre- and po sttest scores for the
experimental g r o u p ...................................... 85
11. Attitude pre- and po sttest results of the co mpa ris on
group (n=36) ............................................. 86
12. Differences in pre- and po sttest attitude
scores of the c o m par is on g r o u p ........................ 87

xiii

13. Comparison vs. experimental group attitude
posttest means (effect of program p a r t i c i p a t i o n ) ... 87
14. Two-way ANOVA of attitude postte st scores to test
for the effects of EFNEP and p r e t e s t i n g ............. 88
15. A comparison of change in food recall scores for
groups with high vs. low pretest attitude s c o r e s . . . 90
16. T-test analysis of food recall change in homemakers
with high vs. low pretest attitude s c o r e s ........... 91
17. Correlations b e t we en h omemaker attitude, locus of
control, and food recall s c o r e s ....................... 93
18. Results of the first regression analysis run to
predict variance in food recall c h a n g e ...............96
19. Results of the first regression analysis run to
predict variance in attitude c h a n g e .................. 97
20. Results of the second regression analysis run to
predict v ariance in attitude c h a n g e .................. 9 8
21. A comparison of pr ete st means for subjects that
did, and did not, take the p o s t t e s t ................. 100

xiv

LIST OP FIGURES

1.

Models related to the d i e t a r y change sequence
(Sims, 1 9 7 8 ) ............................................ 31

2.

The rel ati on shi p between nutritional attitudes,
knowledge, and practices (Sims> 1 9 7 8 ) ...............32

3.

The information pro cessing model
Sims, 1980)

(Olson and
34

4.

Four possible models of the i nte rrelationship
between nutritional knowledge, attitudes, and
practices (Schwartz, 1 9 7 5 ) ........................... 35

5.

The McGuire pe rs uasion matrix

6.

Theory of reasoned action (Azjen and Fishbein,
1 9 8 0 ) .................................................... 39

7.

A food-choice behavior model adapted from the
theory of reasoned action (Dalton et al., 1986)...40

8.

Model of the atti tu de- to- be hav io r process proposed
by Fazio (1986)........................................ 42

9.

Two-way ANOVA:

(Flay et al.,

1980).37

Solomen four group d e s i g n ............65

10. Regression analysis model to assess the ability of
independent v ariables to predict food recall
c h a n g e ................................................... 95
11. R egression analysis model to assess the ability of
independent variables to p redict attitude c h a n g e . . 97

xv

C H A P TE R 1
INTRODUCTION

The Expanded Food and Nutr it ion E ducation Program
(EFNEP)

is a federally funded program whic h provides basic

nutrition education to low-income families,
those with young children.

EFNEP was initiated in 1968 by

the United States D epa rtment of Ag riculture
promote positive changes
attitude,

skills,

families.

especially

(USDA)

to

in nutritio n-r el ate d knowledge,

and behaviors of limited income

The prog ram is implemented under the d irection

of the Cooperative E x t ens io n Service in fifty states,
District of Columbia,
American Samoa,

Guam,

the

Puerto Rico, Micronesia,

and the Virgi n Islands.

In EFNEP, para pr ofe ssi on als who are often indigenous
to the community in w h i c h they work, are employed as
instructors

(program aides)

to deliver the educational

program to low-income homemakers with young children.
Although some teaching takes place in groups or by selfinstruction,

the primary delive ry method is individual

instruction by the p r o gr am aide in the client's home.
Aides are trained in areas of basic foods and
nutrition such as food selection and preparation,

meal

planning,

food

food safety,

food sources of nutrients,
1

preservation, gardening, weight control,
maternal/infant nutrition.

and

Since the initiation of this

study, Michigan EFNEP program aides have also received
training in methods for promoting positive changes in
1
homemakers' attitudes toward d ietary change .
Statement of the P roblem
To provide accoun tab ili ty data,

program evaluation

should be conducted to determine whether EFNEP program
2
objectives (improved knowledge, skills, attitudes , and
3
behaviors ) are achieved. Cooperative Extension Service
(1981)

published "Program Evaluation in Extension: A

Comprehensive Study of Methods,
Procedures."

Recommendations

Practices,

and

listed in the report

include:
1.

development of a system for acc ou ntability and

evaluation to provide greater specificity and clarity
about national ac cou ntability needs.
1
"Attitudes toward dieta ry change" is defined in this
dissertation to mean evaluative responses in regard to
making changes in food consumption practices.
2
"Attitudes" are defined as an evaluative position with
regard to a fact or issue based on an individual's
beliefs
3
The term "behavior" is commonly used in the literature
cited in this di ss ertation to refer to self-reports of
foods consumed. When authors use "behavior" in this
manner, it will be followed by "(food recall)" for the
purpose of clarification.
Whe n "behavior" is not
followed by "(food recall)" it is used in the more
general sense to mean manner of acting.
2

2.

implementation of a more systematic approach to

evaluation at all org anizational
3.

levels.

establishment of greater clari ty r egarding the

needs and purposes of evaluation among different.
program areas and o rganizational
4.

levels.

assessment of the ava ilability and method of

organization of e v a l ua ti on resources.
5.

provide staff d e v e l o p m e n t in the area of program

evaluation.
Additionally,

a r e c o m m en dat ion from the General

Ac c o un tin g Office included in the Report to the Secretary
of Ag riculture

(1980,

p . 8)

stated "We recommend that

Coopera ti ve Extension S e r v i c e ... deve lop

(1) objective and

measurable standards for judging pr ogr am effectiveness and
(2)

the evaluation and feedback tools needed to measure

program performance agai nst such standards."
Comprehensive p r o g r a m evaluative efforts are needed
to determine if the objectives of the EFNEP program are
being met.

It is the goal of the Expanded Food and

N u t r i t i o n Education P r o g r a m to assist low-income families
to "acquire the knowledge,

skills,

attitudes,

and changed

behavi or s necessary for nu tr iti ona ll y sound diets"

(USDA,

1976, p . 6). As in most nutrition ed ucation programs,
m e asu rem en t has been conducted to evaluate change in
knowledge and dietary change
instrument)

(using the food recall

of EFNEP homemakers

(e.g.,

Marke tin g Economics Re sea rch Service,
3

Hutsey,

1970;

1972; Napier

and Wharton,

1974;

Kerr et al.,

1983; Wagne r et al.,

1983).

1979; N i e rm an et al.,

A lthough EFNEP names positive

attitude change as a component of the p r o g ram objective,
only a few studies

(e.g., Baird and S c h u t z , 1976;

Kaplowitz and Olson,

1983; Block et al.,

1984)

have been

conducted to assess attitudes of homemakers and/or aides.
Attitudes toward dietary change of EFNEP participants or
instructors have not been evaluated prior to this study.
Nu trition education programs need to provide data
which indicate whe th er positive changes in attitudes
toward dietary change have occurred in addition to
conducting assessment of knowledge and behavior
recall)

(McKenna,

1983).

(food

Results related to attitude

change can be used in formative e val uation to assess the
need for instructor training and in summative evaluation
to determine the effect of program participation on
attitude change.

In additi on to assessment of attitude

change due to p r o gr am participation,

it is desirable to

examine the relationship between attitudes and dietary
change. Carruth et a l . (1977)

proposed that if the

relationship between attitudes and dietary

(food recall)

change is important in terms of m odifying future dietary
intake, the results of attitude ass essment might be useful
to predict the potential amount of dietary change that can
be anticipated by p r o gra m participation.

They suggested

that more definitive research is needed to determine how
attitudes are modified by nutrition education programs and

4

to wha t extent attitude change is related to d i e tar y
change.
N u t rit io n educators have often assumed that if
positive change in knowledge occurs, positive change in
attitude and behavior

(food recall)

a ut om a t i c a l l y follows.

A primary reason for measu rem en t of attitudes in the past
has b ee n because of an assumed
self-reported diet ary change.

link between attitudes and
It cannot,

however, be

assumed that these v ariables are po sit iv ely correlated.
The consistency between attitude and b ehaviors has
been challenged by several investigators
1934;

Kutner et al.,

(e.g., LaPiere,

1952; Hard ing et al.,

1954). Research

has b e e n conducted to determine if a significant
correlation between nutri ti on- rel at ed attitudes and
dietary intake
recall

(primarily assessed using the 24-hour food

instrument)

exixts.

1976;

Sims,

1978; Schafer,

1982;

Perron and Endres,

(e.g., Picardi and Porter,
1978;

1985).

Dae lhousen and Guthrie,
Pre- and po stt es t

attitude and food recall scores have been assessed,
correlations computed and reported,

followed by

conclusions regarding the strength of the relationship.
Some researchers
housen and Guthrie,

(Picardi and Porter,

1982;

1976;

Perron and Endres,

Da e l ­

1985)

have

indicated that there is an inconsistency b e t wee n these two
variables.
Schafer,

Other investigators

1978;

Kok et al.,

(e.g. Schwartz,

1982)

1976;

have reported positive

correlations between attitude and food recall scores.

If mod erating factors had been included as
independent variables

in these studies rather than

computing only correlations,

data might have been

g enerated to provide specific information to determine if
attitude scores predict dietary change
food recall

(food recall)

scores predict attitude change.

or if

Results of

correlational and regression analyses which is based on
inclusion of multiple variables provides more information
about the rel ationship between attitudes and dietary
intake

(food recalls)

than the single correlational

analyses used in the reports cited.
information

(1)

Based on the need for

to determ in e if there is a change in

attitudes of EFNEP participants,

(2) to determine if there

is a correla tio n between attitude and behavior change,
(3)

and

to assess the pr edictive potential of several

independent variables on attitude change and/or food
recall change,

the f ollowing research objectives and

hypotheses were proposed.
Research Objectives
The objectives of this study were:
1.

To d etermine if attitudes toward dietary change of

EFNEP homemakers

(a) change from p r o gra m enrollment to

prog ra m com pletion

(b) have si gnificantly greater change

compared to a co mp ari son group and

(c) result in more/less

change depending on pretest attitude scores

6

(low vs high);

2.

To determine if change in attitude and food recall

scores of homemakers enrolled in EFNEP are correlated;
3a.

To determine if instructors'

years of instructor experience,
or pretest attitude scores,
scores,

homemaker attitude change

change in locus of control

or pretest food reecall

predict change in food recall
3b.

pretest attitude scores,

scores of homemakers

scores of homemakers;

To determine if instructors'

pretest attitude scores,

instructor years of experience,

pretest or change in

homemakers'

food recall

pretest or change in

homemakers'

locus of control

scores,

scores, or homemakers'

pretest attitude scores predict change in attitude scores
of homemakers.
Research Hypotheses
la. There is no di ff erence between pre- and posttest
attitude scores of EFNEP participants;
lb. There is no difference between attitude change scores
of the comparison and experimental groups;
lc. There is no difference between the amount of food
recall change of groups of homemakers with high vs

low

pretest attitude scores;
2. Attitude change is not positively correlated with food
recall change of EFNEP homemakers;

7

3a. Food recall change is not predicted by: change in
locus of control scores of homemakers,

instructors'

of job experience, pre te st instructors'

years

attitude scores,

food recall pretest scores of h o m e m a k e r s ,or pretest or
change in attitude scores of homemakers;
3b. Attitude change is not predicted by: h omemaker pretest
or food recall change scores,

pretest or change in locus

of control scores of homemakers,
experience,

instructors'

homemaker pretest attitude scores,

attitude scores of instructors.

8

years of job
or pret est

CHAP TE R 2
R EV IEW OF THE LITERATURE

This chapter provides a review of the existing
literature related to the hypotheses being tested.

The

first section reviews research reports regarding
assessment of attitude change in EFNEP and other
nutrition e ducation programs.

The second section includes

a review of the literature in which correlations between
attitude and d ietary intake

(food recall)

are reported.

The third section is a revi ew of models which have been
proposed for purposes of predicting attitude or behavior
(dietary)

change.

Assessment of Change in Attitudes of EFNEP Participants
Although results related to assessment of "attitudes
toward d ietary change" of EFNEP participants have not
been reported,

some research has been conducted to assess

changes in general attitudes of EFNEP participants
and S c h u t z , 1976;

Kaplowitz and Olson,

(Baird

1983; Block et a l .,

1984).
Block et al.

(1984)

included twelve attitude

statements as a c omp one nt of an instrument used to assess
long-term effects of prog ra m participation in the

California EFNEP on knowledge,
change of homemakers.

attitudes,

and food recall

A si gn ifi ca nt positive change was

reported in food recall scores with the gr ea tes t
improvement in consumption of milk/milk products

(a 15.6%

increase in number of persons consuming two or more
servings)

and fr ui t/vegetable consumption

(21% increase in

persons reporting c o nsu mp tio n of four or more s e r v i n g s ) .
Improvements were reported

in cooking skills

(65% i m p r o v e m e n t ) , k n ow led ge
attitudes

(9% increase)

(47% i m p r o v e m e n t ) , and

fro m pre- to posttesting.

of this study indicate that participants

Results

showed great

improvements in knowledge and food recall, but that only
slight increases in attitude scores occurred.
Kaplowitz and Olson

(1983)

attitudes toward breastfeeding,

evaluated knowledge and
and incidence and du ration

of breastfeeding of women enroll ed in EFNEP.

They found

that knowledge scores increased for women wh o were
predisposed to bottlefeed
method of infant feeding

(p<.05)
(p<.001)

or undecided regarding
but not for subjects who

had decided to breastfeed at the time of the pretest.

The

program did not re sult in improvements of attitudes toward
breastfeeding, an increase in the number of women who
actually breastfed,

or the duratio n of br eastfeeding

(compared to a control g r o u p ) . The authors attributed
these results to the small
personal educational
Baird and Schutz

sample size and use of a n o n ­

approach
(1976)

(mailing p a m p h l e t s ) .
assessed the relationship

10

between attitudes rega rd ing food use and d i e t ar y intake
of foods as measured by 24-hour food recall
homemakers part ic ipa tin g in EFNEP.

scores of

The respondents

indicated their a ttitudes related to use of 20 foods given
situations or d escriptors such as "a family favorite" or
"easy to make."

A si gnificant relationship was found

between attitudes regarding uses of foods and the use of
foods as indicated by self-reports of food consu mp tio n
(food recalls)

(p<.05).

Fifty-one percent of the variation

in food recalls was accoun te d for by attitude scores.
Problems reported by respondents as interfering with
making dietary changes included:

trouble with food

preparation, dige sti on difficulties,

lack of knowledge

about nutrition, and diet ary monotony.

The authors

suggested that a ssessing food-related attitudes can be
useful to predict food recall

scores.

However,

it is

apparent from the r e s p o n d e n t s ' remarks that food choices
are affected by several

factors in addition to attitudes

regarding use of foods.
Assessment of Attitude Change of Participants in Nutrition
Education Programs
A factor cited by Sunseri et al.

(1984) whi ch affects

behavior change is the readiness to change of program
participants.
learned,

They claimed that when new information is

formation of positive attitudes and dietary

improvements do not a u t o mat ic all y follow because of
resistance and/or lack of motivation to change existing
11

habits.

They investigated changes between pretest,

posttest,

and follow-up test scores of sixth-grade

students participating in a c ard iovascular risk reduction
program.

Results indicated that although knowledge scores

increased,

attitude

(toward h e a r t -h eal th y eating

practices)

scores decreased s ignificantly between posttest

and follow-up testing

(pC.Ol). Di eta ry scores decreased

from pre- to posttests
follow-up testing

(p<.001)

and from posttesting to

(p<.05). The authors stated that

knowledge functions as a tool only if and when people are
ready to make changes.
alter food intake,

If a health education program is to

it should provide opportunities for the

participants to experience the desired behavior.
authors also indicated that family involvement,
the program,

The
length of

and reading level might have affected

results.
Byrd-Bredbenner et al.

(1984)

assessed the affect of

p a rticipation in a health education progr am on n utrition
knowledge,
new foods",

attitudes

("caring about nutrition",

"nutrition affects health",

about nutrition"),

and dietary change

food frequency i n s t r u m e n t ) .
significantly increased.
senior high students,

"eating

and "learning

(as measured by a

Knowledge scores

Attitude scores increased for

but not for subjects in junior high

school. Neither k nowledge nor attitude scores were
significant predictors of food frequency change.

The

authors attributed the differences in attitude score
12

change between the two groups of students to di fferences
in maturity and ex posure to n u tr iti on information.
The authors,

however,

did not specify age levels of junior

vs senior high students or consider the overlap in
maturity level b e t w e e n the two groups.

Food frequency

scores did not increase for students at either junior or
senior high level. The y suggested that the lack of
significant change in food frequency scores from pre- to
po sttesting might have been due to instrumentation
of sensitivity in the food frequency tool used)

(lack

and to

lack of control over food choices in this age group.
Shortness of the p r o gr am

(six to ten weeks)

was also named

as a primary factor co ntr i b u t i n g to lack of changes in
attitudes and food frequency scores of students.
Ries and Schoon

(1986)

eval uat ed the ef fectiveness of

a caf ete ri a-b ase d nutr it ion educa tio n prog ra m to improve
the knowledge and attitudes of c ollege students.
Knowledge regarding the r el ati o n s h i p between nutr it ion and
health and attitudes toward nutrition and health were
assessed.

Knowledge scores increased signifi can tl y from

pre- to po sttesting

(p<.001),

but no significant change in

attitudes of the students occurred.

The authors attributed

the lack of improvement in attitudes to shortness of
program

(eight w e e k s ) , effect of pretesting,

pretest scores in the control
(81.25%)

groups.

(84%)

and high

and experimental

That changes in attitude and dietary

scores might be influenced by hig h pretest scores was also
13

indicated by results reported by Daelhousen and Guthrie
(1982).
The goal of an affectiv e-b as ed educational program
implemented by Brush et al.

(19 86) was to increase the

flexibility of attitudes of participants.

Flexibility vs

rigidity of attitudes was assessed using a six-point
Likert scale.

Participants used the nominal group

technique to develop and rank objectives for the
educational program according to priority.

The program was

based on learner needs as de te rmi ned from the nominal
group procedure.

The progr am included self-awareness and

value clarification techniques.

Nutrition knowledge

increased in experimental and control groups;

dietary

intake,

improved in

as measured by a 24-hour food recall,

the experimental group only.

F lexibility of nutrition

attitudes did not improve in either group.

The authors

attributed the lack of significant improvement in
attitudes to high prete st scores and the length of the
program.
In a study by Ross

(1984), k nowledge of nutrition

principles and attitudes of nursing students were assessed
before and after par tic ipa ti on in a nutrition education
program.
(p<.001).

Knowledge scores significantly increased
The author reported that there were no

significant changes in scores on thirteen of fifteen
attitude statements

(p.<05). Constructs represented were:

"role of the nurse in nutrition education",
14

"the role of

dietitians in the hospital"/
nursing",

"nutrition e ducation in

and "general nutrition."

The reason given for

the lack of positive improvement in attitudes of the
nursing students in this study was r egr ession toward the
mean on po sttest scores

(high pretest s c o r e s ) .

Attitude change might be affected by characteristics
of the educational message.

Looker and Shannon

(1984)

investigated the effect of threat vs benefit appeal of an
educational message re garding n utrition and health.

Two

types of pamphlets wer e provided, one wh ich emphasized the
benefits of dietary change and the other wh i c h focused on
the negative co nse quences of not complying wi t h dietary
recommendations.

Investigators assessed knowledge about

nutrient density,

attitudes regarding learning about

nutrition and use of nutrie nt dense foods,

and food choice

behavior regarding selection of nutrient dense foods.

The

authors indicated that the purpose of their research was
to identify strategies that posi ti vel y influence
n u trition-related attitudes and behaviors,
knowledge.

as well as

They found that there was no significant

difference in attitude scores related to the type of
message p resented

(threat vs b e n e f i t ) , but the group that

received n utrition educa ti on showed greater increases in
knowledge scores than the control group.

No improvements

in food choice b ehavior of either group were reported.
The authors c oncluded that the threat message may have
been too mild so that there was not enough difference in

the materials to result in significant di fferences in
attitude scores.

The authors also indicated that a high

attrition level and high pretest scores affected posttest
scores.
Research was cond uct ed by Rosander and Sims

(1981)

to

determine if food/ nut ri tio n knowledge regarding food's
affect on health,

attitude related to control over eating

habits, and behavior

(food frequency)

of low-income women

increased after pro gr am participation.

They developed an

affective-based nutri tio n intervention to be used in the
Special Supplemental
Children

(WIC).

Food Program for Women,

Infants,

and

The intervention consisted of a series of

three lessons, d elivered one month apart.

The instrument

developed to measure the affective domain was a five-point
Likert scale repres en tin g two constructs:

"personal

control over eating habits" and "foods effect in health
and feelings."

The respondents had significantly higher

posttest scores on both attitude scales.
scores,

as measured by a food frequency,

The dietary
showed

statistically significant improvement for the instructed
group.

Authors concluded that affective -b ase d nutrition

education is successful

in improving knowledge,

attitudes,

and food frequency scores of low-income women.
The Correlational R e l ati ons hi p Between Attitude and
Behavior (Dietary Change)
Numerous studies have been conducted in which
attitude and behavior were assessed for the purpose of
16

de termining if these two variables were significantly
correlated

(Appendix A). Conflicting results have been

reported and are summarized in the following review which
relates to the second research hypothesis.

Correlational Inconsistency Between Attitude and Behavior
(Dietary Change)
Even though many investigators
Doob,

1947; Green,

1954; Campbell,

(e.g. Allport,
1963)

as including a behavioral component,

1935;

defined attitudes

a wea k relationship

between attitude and behavior has been reported by
LaPiere

(1934),

and Guthrie

Picardi and Porter

(1976), Daelhousen

(1982), and Perron and Endres

A study by LaPiere

(1934)

(1985).

was conducted to

investigate the level of c onsistency between attitude and
behavior.

In this study the incidence of acceptance or

rejection on the part of proprietors at several inns to
provide

lodging to a C hinese couple was assessed.

were refused accommodations at only one place.

They

A follow-

up q uestionnaire was sent to the places at which the
couple had sought lodging. Most proprietors

(71.8%)

indicated on the question na ire that they would refuse
lodging to Chinese people;

indicating a d iscrepency

between self-reported attitudes
to Chinese people)

(toward providing lodging

and actual behavior.

LaPiere

(1934)

concluded that there was unr el iab ili ty of the attitude
responses to predict d i scr imi na tor y behavior.
however,

had several weaknesses.
17

The attitude

This study,

questio nn air e was sent months after the traveling
situation occurred increasing temporal instability of the
measurement.

Because the attitude q ues ti onn ai re was used

after behavior wase observed,

this study a ctually assessed

ability of behavior to predict attitude.

There was no

indication that the perso n who accep te d/r eje ct ed the
travelers was the same person w ho filled out the
que sti on nai re or that the employee was able to identify
the natio na lit y of the couple.

Alt hough limitations of

the study seriously w e a k e n LaPiere's claim that there is
inconsistency b e t we en attitudes and behavior,

this study

is fre quently cited as landmark research related to the
rela tionship between these two variables.
In contrast to LaPiere's
et al.

(1954)

(1934)

conclusions,

Harding

stated that the literature has confused

correlational

inc onsistency with situational threshold

differences and has thus exaggerated the inconsistency
between attitude and behavior.

These authors claim that

attitude has a lower threshold than behavior;

i.e.,

it is

easier to agree with a tt itude statements than to perform
the parallel behavior.

In L a P i e r e 's study

(1934),

it was

easier for ques ti onn air e respondents to state that they
would refuse to give the couple lodging than to actually
turn them away in person.
(1956)

According to Harding et al.

there would only be inconsis te ncy if a behavior

were performed
the attitude

(couple turned away)

that conflicted with

(accept Chinese people as g u e s t s ) . The
18

authors concluded that threshold differences accounted for
inconsistencies reported in the literature between
attitudes and behavior.
Kutner et al.

(1952)

conducted a study to examine how

threshold differences affect the co nsistency between
attitude and behavior.

In this study,

two white women

were seated in a re st aurant and were then joined by a
black women.
incident.

In all cases,

they were served without.

Then a requ es t for a group reservation was sent

to each establishment.

It was explained that the party

would include black and w hit e persons.
received.

No replies were

When the investigator placed follow-up phone

calls to the restaurants,

eight of eleven managers told

him no request had been received.

Most refused to

accomadate a party that was racially mixed.
study,

the behavior

(serving the women)

which conflicted with the attitude

In this

was performed

(refusal to serve a

mixed party of black and white p e r s o n s ) .
support those of L a P i e r e 's research;

These results

attitudes of

restaurant personnel regarding serving blacks were
negative but the co nflicting behavior

(serving blacks)

was

performed.
Several factors have been indicated to explain the
lack of a positive correl ati on between attitude and
behavior.

These are summarized in the following review.

Length of prog ram p ar ticipation was named as a factor
that affected the consi ste ncy between attitude and dietary
19

change in a study by Picardi and Porter

(1976).

conducted an ass essment of nutrition knowledge,

They
attitude

(health c o n c e r n s ) , and food choice behavior of high school
students p articipating in a fo od/ nutrition minicourse.
These researchers found no significant correl ati on between
knowledge,

attitude,

or food choice behavior.

There was a

significant improvement in knowledge, but not attitude or
food selection scores.

The authors attributed the results

to the shortness of progr am participa tio n

(30 h o u r s ) .

In an evaluation of a self-instructional nutrition
education program for pregnant women, D ael housen and
Guthrie

(1982)

used a five-point Likert scale to determine

changes in attitudes regarding importance of nutrition
during pregnancy.

Dietary change was assessed using a 3-

day food record and 24-hour food recall.

The self-

instructional prog ram c onsisted of a 16-minute vide o which
included information r egarding nutritional needs and
selection of an adequate diet to meet nutrient needs
during pregnancy.

The video included information designed

to e ncourage positive attitudes toward nutr it ion during
pregnancy.

The reported c orrelation between attitude and

dietary scores was not significant

(r=.08).

The authors

attributed the lack of a significant c or rel ati on to
extremely high pretest scores for both control
experimental

(9 0.8%)

(89.6%)

groups which limited the amount of

positive change possible.
Existing attitudes and lack of control over food
20

and

choices were named as variables that influenced dietary
scores in a study by Perron and Endres

(1985). These

authors investigated the rel ationship between nutritional
knowledge,

attitudes toward "general nutrition" and

"nutrition for the athlete", and food intake of adolescent
female athletes.

Food intake was measured using a 24-hour

food recall and a 48-hour food record.

The attitude

instrument was a 5-point Likert scale.

Knowledge and

attitude were sig nificantly c or related

(r=.52). No

significant, co rrelation was found between attitude and
dietary scores.
population

The authors concluded that in this

(female a t h l e t e s ) , food intake may be shaped by

a strong desire to be thin,
information.

rather than by nutrition

They felt that the adolescents'

control over foods served,

in addition to the existing

attitude of concern for weight,
correlational

lack of

accounted for the

inconsistency between attitudes and dietary

scores.
Correlational Consi ste ncy Between A tt itude and Behavior
(Dietary Change)
Several researchers

(Schwartz,

1977; Fazio and Zanna 1978;
1982)

Schafer,

1976;
1978;

Regan and Fazio,
Kok et al.,

have reported positive correlations between

nu tr ition-related attitudes and self-reports of food
intake.

The strength of the co rrelation was attributed to

various factors wh ich are presented in the following
review.
21

Schafer

(1978)

self concept,

assessed personal

beliefs)

factors

and social factors

friends and family members,
and educational programs)

(attitudes,

(influence of

advertisements,

TV programs,

that influenced diet ar y adequacy

and "empty calories" of self-reported food intake
food recall).

(24-hour

The most influential factor related to food

consumption for men and women were personal food
preferences.
for women,

Factors which influenced food preferences

in order of importance, were:

cost, health,

convenience,

habit,

nutrition,

taste,

and w e igh t control.

Men's food choices were influenced by the following:
taste, nutrition,

cost, health,

appearance,

convenience,

and habit. The author considered women's ranking of
"nutrition" as the primary factor in d ete rm ini ng food
preferences to be co nsistent with their traditional role
in selecting and p roviding nutritious foods for the
family.

Children's food preferences influenced the food

choices of women more strongly than men.

Health of family

members was ranked as a more important factor than self
health by men and women. Educational programs

(consumer

education and Exte ns ion Service)

but greater

had a small,

effect than the influence of media,
women.

Schafer

(1978)

between attitudinal

on foods selected by

conducted correlational analyses

factors and dietary adequacy.

Results

indicated that "personal preferences" and d i e tar y adequacy
were positively correlated for women
not for men

(r=-.03).

(r=.23, p<.01)

When cost was reported as a
22

but

significant factor of food choices,
were consumed

(r=.17, p<.05)

more empty calories

and diets were

less adequate

(r=-.19,p < .05). There was a po si tiv e correl at ion between
health as a factor and quali ty of diet for men
p < .05)

and women

(r=.19, p<.05).

(r=.16,

The more men perceived

women to influence their diets,

the fewer empty calories

they consumed

(r=.32),

and the higher the quality

of their diets

(r=.18, p<.05).

p<.001)

Pa rti ci pat ion in

educational programs was invers el y related to empty
calories consumed

(r-.19, p<.05)

with diet ar y ade quacy

and po sitively correlated

(r=.27, p<.01)

for women.

"Information" was posi ti vel y c o r r e lat ed with dietary
quality for men

(r=.18, p<.05)

and w o m e n

(r=.21, p<.01).

The author concluded that wh e n w o m e n influenced their own
or their husband's food choices,
high;

di et ary a dequacy was

the more c o n cer ne d men and women were about the

health of family members,

the bett er the quality of the

diet; when cost was co nsi de red an important factor,
quality of the die t was
were consumed.

low and more empty calories

P a r t ici pa tio n in nutri ti on education was

po s iti ve ly related to dietary a de qua cy for women, but not
for men.
The research by Schafer

(19 78)

investigated the

affect of several factors on d i e ta ry intake.

This type of

research where the r e la tio ns hip between multiple
attitudinal
more helpful

factors and dietary scores is investigated is
in u n d e r s ta ndi ng the process of dietary
23

change than research that assesses and correlates only
attitude

(usually towards the importance of nutrition)

and

food recall scores.
A study was condu ct ed by Kok et al.
knowledge about c a rd io vas cul ar disease

(1982)

to assess

(CVD), attitudes

toward CVD, and die tar y intake of subjects in the
Netherlands.

Attitudes were assessed using a five-point

Likert scale.

The reported corre lat ion between attitude

and food intake

(measured by 24-hour food recall)

(p<.001), betwe en knowl edg e and attitude was
and

.12

(p<.001)

.19

was

.20

(pC.OOl),

betw ee n know le dge and food recall

scores.

There was a stronger corre lat io n between attitudes and
food intake for those wh o had hig h food recall pretest
scores than for those that had
food intake measure.

low prete st scores on the

The authors

suggested that multiple

factors affect food recall change such as:
significant others,

influence of

dif fic ult y in chang ing food habits

(30% of respondents said that it would be difficult to
change food h a b i t s ) , and insufficient knowledge

(23%

indicated that they did not have sufficient k nowledge to
determine

"what kind of diet was healthy").

"Taste"

emerged as a p r i mar y reason for food choices.
suggested that attitudes,

The authors

and factors that influence

attitudes and food selection,

should be addressed in

planning and i mp lementing educational programs.
The manner of attitude formation
experience)

(direct or indirect

is important in the attitude- be hav io r
24

relationship.

Fazio and Zanna

(1978)

on the basis of

their research stated that there is increased c onsistency
between attitude and beh avior if the individual has formed
attitudes on the basis of direct,
experience.

rather than indirect,

A group of students was asked questions to

assess their attitude and behavior regarding participation
as subjects in experimental

research.

They were

classified according to past experience as subjects
(minimal,

intermediate, high).

The correl ati on between

attitude and behavior was calculated.
hig h level of previous par tic ip ati on

The group that had a
(direct experience)

had a high degree of consistency between attitude and
be havior

(r=.42).

The group with an intermediate

level of

experience also had a significant positive correlation
between attitude toward particip ati on in research and
actual participation

(r=.36). Attitude and behavior in the

group with minimal previous experience were not
significantly correlated

(r=-.03).

The authors concluded

that the direct ex pe rience of being research subjects
strengthened the correla ti on between attitude and behavior
because the experience focused the subjects'

thinking

which increased the acc es sib ili ty of the information from
memory.
A study by Regan and Fazio

(1977)

compared behavior

(signing of petitions regarding changing housing
regulations)

of college freshmen who were refused housing

due to a shortage of dorm ito ry rooms w ith behavior of
25

freshmen who were pr ovided wi t h dorm ito ry rooms.

The

students who were required to use temporary housing were
considered to be the gr ou p that used direct experience in
forming attitudes regarding un ive rs ity h ousing problems.
Those that were pr ovided w i t h perman ent housing were
considered to have formed attitudes regarding problems
with housing from indirect ex periences
discussions with o t h e r s ) .
both groups were similar,

(campus paper,

Al though the attitude scores in
a t t itu de -be hav io r c onsistency

was much greater in the direct experience group.

These

authors concluded that although individuals may hold
similar attitudes,

the strength of the attit ude -b eha vio r

consistency depends on whether dire ct experience was used
in formation of these attitudes.
Bern (1970)

promoted a theory that individuals have

difficulty drawing associations bet we en the attitude
object and evaluation categories unless they have been
involved in some b ehavior

(experience)

toward the object.

He claimed that previous exp erience strengthens the
evaluative component of the attitude,

increasing the

likelihood of accessing the attitude to guide the
individual

in eval ua tio n of the a tt itu de object.

Most

nutrition education programs do not include direct
experience
educational

(active participation)
intervention.

as part of the

It is po ssible that greater

positive change in attitudes would occur if client
p a rticipation

(direct experience)

w er e included in

nutrition education programs such as in EFNEP.
Schwartz

(1976)

evaluated changes in attitudes

after partic ip ati on in nutrition e ducation that included
the affective domain.

She assessed the relationship

between nutritional knowledge,
(eating habits,

nutritional counseling,

food preparation)
health nurses.

attitudes toward nutrition
meal planning,

and counseling practices of public

The instructional prog ra m included

encouragement to identify and explore feelings about
food/nutrition.

The attitude asse ss men t was based on

program objectives r e pre se nti ng the affective domain.
This intervention dif fered from many others in that the
affective domain was included as part of the educational
program.

Respondents indicated w h e t he r they agreed or

disagreed w it h opin ion statements and the degree of
certainty of their responses.

Practices were assessed by

indicating the frequency the behavior occured
frequently,

sometimes,

n e v e r ) . Factors that were

significantly c orr elated
were age

("over forty")

(always,

(p<.05)

w i t h positive attitudes

and "cooks for family and

friends." Factors p osi tively co rrelated with positive
counseling behaviors wer e age
health agency publications",
economist",

("over forty"),

"use of

"consultation with a home

and "cooks for family and friends."

Significant positive correlations were found between
know le dge-attitude and attitud e- pra cti ce s

(beha vio rs ).

The theory of cognitive dissonance states that people
27

tend to behave in a manner that is c on sistent with
existing attitudes and beliefs
Yetley,

(Bern, 1970; Schafer and

1975). Behaviors are not performed that conflict

with the existing cogni ti ve framework.
Schafer and Ye tley

(1975)

applied the theory of

cognitive dissonance to the area of nutritional behav io r
change.

They claimed that people prefer to d evelop and

maintain stability in diet ary patterns that are con sistent
with their values and percep tio ns of their environment.
Tension is produced w h e n stability is disturbed.
Attitudes are seen as anchors w i th in the frame of
reference used to evaluate the environment.

The authors

indicated that food choices are based on the need to
maintain con sistent environmental
dissonance.

When events occur

nutrition information)

patterns to avoid

(such as acquisition of

to disrupt the existing pattern,

r e structuring of the cogn iti ve belief system occurs only
if the new information does not conflict w ith existing
values/beliefs.

If the n ew information is not c on sistent

with e xisting attitudes or beliefs,

it is not likely to be

integrated into the individual's existing cognitive
framework.

The authors indicated that the ef fectiveness of

nutrition educ ati on to af fect behavior change can be
increased by assessing the exist ing values and beliefs of
learners. Nu t r i t i o n e ducators should be aware that if
information they provide is disson ant w ith the
individual's existing patterns, b eh avior change is not

likely to occur;

the instructor's message may be perceived

as th reatening and could result

in d efensive or avoidance

behavior in the learner.
S u m m a r y ; The Rela ti ons hip Betw een Attitudes and Behavior
In reporting results of researc h in wh i c h attitude
and dietary intake have been assessed,

authors have

indicated that several factors in addition to attitude
affect dietary change
existing habits
1985),
et al.,
change

(Appendix A ) . These factors include:

(Kok et al.,

1984;

Perron and

Endres,

(Ramsey and Cloyd,

pretest scores
Ries and Shoon,

1984;

1986),

Vermeersch,

Ries and Schoon,

locus of control

previous ex periences

1985),

1981;

Ross,

(Sims,

1979;

Picardi and Porter,

1976;

1984; Hollis et

(Yperman and
(Maiman et al.,

(Schafer,

Kok et al.,

1978;

1983;

(Schafer,
1982),

(Harding et al.,

Kok et al.,

1982),

Kok et al.,

1954;

taste

1982; Dalton et

1986), and discrep anc ies bet wee n measurement

29

1984;

1976;

Penner and Kolasa,

problems with attitude m eas ur eme nt

al.,

1986), high

influence of significant others

1978; Yperman and Vermeersch,

1984),

Bryd-

1982;

1978; Eden et al.,

1979; O'Connell et al.,

preferences

1976;

attitudes of the instructor
1979)

readiness to

Sunseri et al.,

(Daelhousen and Guthrie,

Wallston and Wallston,
1986),

1979;

(Byrd-Bredbenner

1985),

(Picardi and Porter,

Bredbenner et al.,

Reames,

Perron and E n d r e s ,

lack of control over food choices

length of program

al.,

1982;

instruments and prog ra m content
Schwartz,
Guthrie,

(Peterson and Kies,

1976; Rosander and Sims,
1982;

Eden et al.,

1984;

1981; Daelhousen and
Brush et al.,

1986).

Usability of the results reported depends on,
a great extent,

1972;

to

the val id ity and reliability of the

instrument used to collect the data.

Indications that

validity and re liability were d et ermined were provided
only a few of the studies reviewed
1976;

Carruth et al.,

1981;

Kaplowitz and Olson,

and Kolasa,
Shannon,

1977;

Sims,
1983;

(Picardi and Porter,
1978; O ’Connell et al.,

Kok et al.,

1983; Byrd -Br edb en ner et al.,

1984;

Ries and Schoon,

Guiry and Bisogni,

in

1982; Penner

1984; Looker and

1986; Brush et al.,

1986; Dalton et al.,

1986).

1986;

Use of

research results to make program planning and
implementation decisions should be limited to research

in

which a valid and reliable instrument was used.
A primary reason for m eas ur eme nt of attitude in the
past has been because of the assumed link between attitude
and b eh avior

(dietary c h a n g e ) . This

literature review

indicates that the dire ct c o rre la tio n between attitude and
die tar y scores might depend on one or more of these
factors.

30

Models of the Relationship Between Attitude and Behavior
There has been disagreement in the literature
regarding the "correct" sequence b etween attitude
behavior change

(A) and

(B). Regression analysis has been used to

determine if attitude scores predict behavior change or if
behavior scores predict attitude change.

A review of

research which tests these and other models follows.
Model One:
Sims

Attitudes Predict Behavior
(1978)

(A-B):

tested the assumptions of two models

wh ic h proposed different sequences in which attitude
preceeded behavior change,
beverage records

as measured by one-day food/

(Figure 1). Both models propose that

attitude change preceeds dietary change but the effect of
knowledge differs in its relationship to attitude.

KNOWLEDGE

BEHAVIOR

ATTITUDES

Altitudes ore intervening vonobles between knowledge ond bchovior

ATTITUDES

BEHAVIOR

KNOWLEDGE

Know ledge a c ts to in tervene between o ttitu d e * ond behavior.

Figure 1. Models related to the dietary change sequence
(Sims, 1978)

31

She assessed the knowledge, attitude,
intake of lactating women.
constructs:

Sims assessed four attitude

"nutrition is important",

are necessary",

and dietary

"vitamin supplements

"meal planning is important",

and "meal

preparation is enjoyable." Dietary intake was assessed
using three one-day food recalls. Total calories, protein,
calcium,

vitamin A, thiamin,

riboflavin,

ascorbic acid were calculated.

Results

niacin, and

indicated that

nutrition knowledge was positively corr el ate d with
"nutrition is important"
(r=.55, p<.01).

(r=.31, pC.Ol),

and education

"Vitamin supplements are necessary" was

negatively correlated with education

(r=-.37, pC.Ol).

"Nutrition is important" was positively correlated with
protein intake

(r=.25, p<.01). There we r e no other

significant correlations between the att itude constructs
and any of the die tar y factors.

BEHAVIOR
Protein Group
Inoei
(p ro .v it A . fe . th io . mo)

Vitam in C

Figure 2. The relationship between nutritional attitudes,
knowledge, and practices (Sims, 1978)

32

Path analysis indicated that nutrition knowledge was an
intervening variable b e t wee n attitudes and dietary
adequacy which supports Model

II

("attitudes cause

knowledge change w h i c h causes behavior")

rather than Model

I ("knowledge leads to attitudes, wh ic h lead to nutrient
intake")

(Figure 2).

On this basis,

Sims c oncluded that positive attitudes

toward nutrition increased subjects'
knowledge which led to die tar y change

interest, in acquiring
(A-K-B). She

recommended that educational programs focus initially
on attitude change to promote co mpr ehe ns ion of cognitive
information

(knowledge c h a n g e ) , resulting in dietary

change.
Olson and Sims

(1980)

p resented a model of

information processing that outlined several stages from
the point of receiving nutrition information to
de monstration of changes in food select io n and menu
planning behaviors

(Figure 3).

In this model exposure to

and c omprehension of the message and formation of
attitudes and intentions preceed behaviors
selection). N ew information is acquired,

(food

stored,

retrieved for later de cis ion - m a k i n g purposes.

and

The

information must be processed through the various stages,
including formation of attitudes,
(food selection)

behaviors.

33

to result

in a change of

INrORMATCN
mhgrato* m
h c i& i o n w a « n g
Caposure
A T T E N T IO N

-m-

c o m p r e h e n s io n

R EHEARSAL - * » -

knimiEdge

STRUCTURES
CONCEPTS AND

Kuicrs •
s tru c tu re or
PROPOSITIONS
ATTITUDES AND
BEHAVIORAL
INTENTIONS
DECi S'ON-MAKINC
PROCESSES
SELECTION

or root
menu

Figure 3. The information processing model
Sims, 1980)

Schwartz
model

(1975)

ttpes

P l a n n in g

(Olson and

conducted a study to determine which

(Figure 4) best explained the relationship between

knowledge, attitudes, and practices of college students
who had enrolled in high school home economics courses.
Knowledge and attitude, and attitudes and practices were
significantly correlated; knowledge an d practices were
not.

Based on these findings, Schwartz accepted Model

wh i c h indicates that attitudes influence knowledge and
practices,

to describe the re lat ionship between the

variables.
34

I,

K
K -A-P Model I

K

A *

K-A‘ P Model 3
A

K
K -A -P

P

A

Model 2

K-A-P Model A
A

Figure 4.
Four possible models of the interrelationship
between nu tritional knowledge, attitudes, and practices
(Schwartz, 1975)

Peterson and Kies

(1972)

inv es tigated nutrition

knowledge,

attitudes toward teaching n utrition and toward

the school

lunch program, and teaching practices of K-3

teachers rel ate d to teaching nutrition.

Sixty-three

perc en t of the teachers surveyed felt tha t nutrition
should be taught as an integral part of the elementary
school c u r r i c u l u m and 53% reported t eac h i n g nutrition in
this manner. Al tho u g h this appears to indicate a high
level of c o n s i s t e n c y between attitudes toward teaching
nutrition and teachi ng practices,

actual

teaching

behaviors were not evaluated. Ni n e t y - o n e percent of
teachers thought that promoting a f avo ra ble attitude
towards nutrit io n was more important than teaching factual
in formation in achieving positive e a t i n g habits.

There

wer e no p os itive correlations between knowledge and
attitudes toward teaching nutrition or between the

35

importance of n u t rit io n in p romoting d esirable dietary
habits and the school

feeding program.

The authors

concluded that nutr it ion knowledge of el ementary school
teachers was not related to attitudes toward teaching
nutrition but that teacher attitudes toward the importance
of teaching nutrition pred ic ted teaching behaviors.
Results found in r es earch by Gui ry and Bisogni

(1986)

illustrate h ow the c o r r e l ati on between attitudes and
behavior is affected by specificity of attitude
measurement.

They rep orted that the attitude constr uc t

"It

is hard to limit coffee consumption" was a strong
predictor of caffeine consumption.

This attitude explained

42.4% of the va riance in ca ff ein e consumption.

They also

used a more general scale to assess attitudes toward the
importance of nutrition.

This measure did not predict

caffeine consu mp tio n as measured using a 24-hour bev erage
recall and a beverage freq ue ncy list.

If this study had

assessed only the a ttitudes related to "importance of
nutrition",

no r e la tio ns hip bet wee n attitude and caffeine

co nsumption w oul d have been found.

The attitude construct

"It is hard to limit coffee consumption"

is more specific

related to caffeine c o n s u m pt ion than the general construct
"importance of nutrition."
A model by McGu ire
model by Olson and Sims

(Flay et. al.,
(1980)

1980)

supports the

in that it assumes

modifications in behav ior follo w changes in knowledge,
attitudes and intentions.

A series of progr es siv e steps

36

leading to behavioral change is out l i n e d in the McGuire
Persuasion M a t rix

(Flay et al.,

1980).

This model

(Figure

5) concep tua liz es the change process and the complexities
of the re lat i o n s h i p s between knowledge,

attitude, and

behavior in the communication process.

The message is

presented

( e x pos ur e); if the message is recognized

(awareness)
results.

and comprehended, a c ha nge in knowledge

If the message is accepted,

there is a yielding

to the m e s s a g e and a change in belie fs and attitudes
occurs.

If the n e w beliefs and a tti tud es are reinforced

and if a p p r o p r i a t e

skills are present,

the belief/attitude

will p e r si st and result in behavior change.

exposure to the m e s s ag e
awareness

(recognition of the message)

co mp reh ens io n of the m e s sa ge
knowledge change
c ha nge in beliefs

(acceptance of the message)

attitude change
re in for cem e n t / p e r s i s t e n c e of attitude
behavior change
maintenance of c h ang e
Figure 5. The McGuire persuasion m a t r i x
1980)

(Flay et al.,

Program in effectiveness in influencing behavior change may
be due to inc om ple te processing at a n y stage in the model.
37

For instance,

the target audience may not be reached

of exposure);

the message may not be attended to

a w a r e n e s s ) ; the message may not be u n d e rs too d
in knowledge)

or y ielded to

(lack

(lack of

(no change

(no change in b e l i e f s ) ;

attitudes might not be changed becau se beliefs are not.
retained,

and su bsequent behavior might not occur because

the attitude is not reinforced. When behavi or change is
assessed in p r o gra m ev alu at ion procedures,

the educational

program should be examined to d e ter mi ne if the learner is
allowed/ enc ou rag ed to proceed through the steps included
in this model.
Ajzen and Fishbein's
action,

(1980)

theory of reasoned

proposes that b eh avior is det erm ine d by attitudes

(A) and norma tiv e beliefs

(NB)

(Figure 6). N ormative

beliefs are perce pt ion s about wha t people whose
opinions are valued think about the behavior.
beliefs are Influenced by social values.
influenced by perc eiv ed advantages,

N ormative

Attitudes are

disadvantages,

and

consequences of per for mi ng the behav io r which is
considered to be a social,

factor.

Weights are assigned

(beta weights from the regression equation)
proportional

w h i c h are

to the relative importance of the predictor.

38

A

+ NB
B(wl)

= B I --------------- *B
-time
-skilIs
-existing habits

(w2)

Where:
A

= attitude toward the behavior
B

NB = normative beliefs
BI = behavioral intentions
B

= behavior

wl = weight applied to attitudes
w2 = weight applied to

subjective norms

Figure 6. Theory of reasoned action
1980)

Ajzen and Fis hbein

(1980)

(Ajzen and Fishbein,

propose that other

variables might i nfluence behavioral

intentions

by influencing at titudes or normative beliefs.
connection b etween intentions and
by time,

skills needed

the

less

(the more

the less likely the be havior is

to be p e r f o r m e d ) , and existing habits
existing habits,

The

behavior is influenced

to perfo rm the behavior

complex the skill needed,

indirectly

(the stronger the

likely they will be modified)

(Figure 6).
Dalton et al.

(1986)

c onducted research to assess the

ability of several dete rm ina nt s to predict food choice
behavior.

They tested the model of reasoned action

proposed by Ajzen and F ishbein

39

(1980)

(Figure 7).

irHAvotA! nuns
IO O O C m C K C I

1 HftSf'KVt>k*
E*p*rf-*rKy

7
G tN E R A l fO O O

3 S n i o r App*o!

CHOICE DETERMINANTS

SO C lAt

A tM u d * lo r o r A p O fn tl

food

7

C K o *c*

<

ft>odCfo-t*

CUlTURAl

Aituo’loorfGwa

PSYCHOLOGICAL
PHfS»OlOG»CAt

R * * c t * * d lr t l k * r < »
TKo» O iK * » i

o f 0 * K « r»

lK.nl SKould/SKoifld
Nq'CKoow food

Figure 7. A food-choice behavior model adapted from the
theory of reasoned action (Dalton et al., 1986)

Attitudes toward food choices,

intended food choices, and

actual food choices of employees at their workplace were
assessed.

They reported that employees who made food

choices that were consistent with intentions had more
positive attitudes

(p<.05)

toward foods they selected than

persons who made food choices that were
stated intentions.

inconsistent with

"Sensory appeal" was the best predictor

of intentions and actual food choices.

"Health value" was

the second strongest predictor of intentions, but did not
predict food choices.

These results indicated that

attitudes were more important than norm at ive beliefs
(perceived influence of others)
actual food choices.

in pre dic ti ng intended and

Based on these findings,
40

the authors

suggested that nutrition educators should consider taste
(sensory appeal)

as a p rimary factor affecting intentions

and food choice behaviors.

They suggested that educational

programs which attempt, to modify food choices could
increase e ffectiveness if factors

(e.g.

beliefs,

underlying food choices

and influence of others)

taste, health

are investigated.
Matheney et al.

(1987)

applied the theory of reasoned

action to dete rmi ne the predictive value of attitude and
normative beliefs on mothers'

beliefs,

intentions, and

actions regarding method of infant feeding.
this study indicated that attitudes,

Results of

based on beliefs,

were more predictive of actions than normative beliefs
(perceived influence of others).

Based on these results,

the authors re commended that nutrition e ducation programs
should assess attitudes and beliefs because of their
potential

importance in predicting food choice behavior.

A model

(Figure 8) which includes several factors to

explain the process by which attitudes guide behavior has
been proposed by Fazio

(1986)

41

of Indiana University.

Attitude------- *■------- direct experience
Activation-----f- ------- repeated ex pression of
the attitude
v
Selective-------«------- situational cues
Perceptions
v
Immediate
--------cr ed ibi lit y of the source
Perceptions
«-------- perceived value of the
information
v
Definition of
* ------- norms
the Event
Behavior
Figure 8. Model of the a t ti tud e- to- be hav ior process
proposed by Fazio (1986)

Attitude act iv ati on is the first component of Fazio's
Model

(Figure 8). Fazio claimed that existing attitudes

need to be recalled

(activated)

from memory and considered

before the new attitude can be integrated into the
individual's cognitive structure.

The strength of the

attitude affects attitude activation.

The stronger the

attitude the more easil y it is accessed from memory.

If an

evaluative opinion is strongly ass ociated with an object,
the attitude is e a si ly accessed and is therefore available
to guide behavior.

Behavior is not likely to be guided by

existing attitudes unless they are strong enough to be
accessed from memory.

If accessed attitudes are positive,

then the positive components of the attitude issue are
likely to be considered.

A negative attitude would prompt

the individual to focus on the negative aspects of the
attitude issue.

The activated attitude functions to guide
42

information processing,

attitude formation or change,

and

related behavioral change.
Fazio

(1986)

claimed that attitude a ct ivation is

influenced by two factors whic h affect the strength of the
attitude:

direct e xp erience in the formation of the

attitude and repeated e xpr ession of the attitude.

He

thinks that attitudes formed involving direct experience
are stronger pr edictors of su bs equent be havior than those
based on indirect experi enc es because attitudes are more
likely to be s tre ngthened resulting in increased
a c cessibility from memory.
Selective p e r c ep tio n occurs w hen people selectively
recall

information wh e n c onf ronted with a new situation.

The information re cal le d influences immediate perceptions
related to the at ti tud e object encountered.

Selective

perception is influenced by situational cues

(verbal,

printed)

which prompt an individual

to access existing

attitudes which may influence their d e f i nit io n of the
situation.

Selective pe rceptions such as opinions about

the credib il ity of the source and expectations regarding
the value of the inform at ion to be received,

affect

immediate perceptions.
The individual's

selective and immediate perceptions

influence the d e f i n i t i o n of the situation wh i c h guides
behaviors.
positive,

If the d e f i n i t i o n of the situation is
then a pp roa ch behaviors are likely to result;

the def inition of the event is negative,
43

then avoidance

if

behavior occurs

(Fazio,

1986).

This model proposed by Fazio

(1986)

provides one

expla na tio n of the process by w h i c h attitudes guide
behavior.

Whereas early investigations of the attitude-

behavior r e lat io nsh ip focused on determin ing whether
there was c on si ste ncy between attitude and behaviors
(correlational a n a l y s i s ) , this model attempts to explain
how the method of at titude formation and strength of
existing attitudes affect the individual's def ini ti on of
the attitudinal

issue and subsequent behavior change.

Fazio suggested that future research investigations should
assess the ability of several factors

(independent

variables)

(dependent variable)

to p redict be havior change

rather than to assess only attitude and behav ior and
compute a single c o r r e l at ion between them.
Model T w o ; Be havior Predicts Attitude

(B-A)

Some inv estigators think that behaviors are accurate
predictors of att itude change
al.,

1979; O'Connell

et al,

(e.g. Brehm,

1956; Maiman et

1981). This model

is

ab breviated as B-A.
Br eh m

(1956)

suggests that ac qui si tio n of new

information leads first to behavioral

improvements which

result in formation of positive attitudes.

Bre hm measured

attitudes and b e h avi ors regarding certain consumer
products.

He asked c onsumers to evaluate several products

and then told them they could select one item in payment

44

for their participation.

He found that evaluation of the

item selected as the reward increased in f a v o r a b i l i t y . The
evaluative opin io n toward the other items was less
favorable on the second evaluation.

B r e h m concluded that

the behavior of selecting a reward influenced attitudes
toward consumer products
Several
et al.,

investigators

1981;

Penner

that attitude change
experiences

(B-A).
(Maiman et al.,

and Kolasa,

1983;

1979; O'Connell

Reames,

1985)

found

depends in part on previous

(behavior)

of subjects.

A study was cond uct ed by Penner and Kolasa

(1983)

to

investigate attitudes of secondary school teachers towards
teaching nutrition,

nutrition knowledge,

experience teaching nutrition.

and previous

Atti tud es toward teaching

nutrition were as se sse d using a fi ve-point Likert scale
and two semantic differential

scales

("my own nutrition"

and "my teaching food and nutrition").
that teachers of health,
economics,

science,

Researchers found

physical education,

home

and social science w h o had taught some

aspect of nu trition had more positive attitudes than those
who had not taught nutrition.

Teachers

that indicated

they had previous ex per ie nce teaching nutri tio n were more
likely to have taken

f ood /n utrition courses and had higher

scores on the nutri ti on knowledge test. W h e th er the
positive attitudes of teachers with previous experience
teaching nutrition influenced su bsequent teaching
behavior was not reported.

45

O'Connell et al.

(1981)

investigated the r el ationship

between attitudes of teachers

("nutrition is important"

and "favors n utrition e ducation in schools")
behavior

and teaching

("commitment to teaching nutrition").

Scores on

the attitude a ssessments were si gnificantly correlated
(r=0.23 p<.05 and r=0.41 p<.001)
teaching nutrition."

Teachers w ho had taught nutrition

had signifi can tly higher means
nutrition education"

wit h "commitment to

(p<.05)

on "favors

than those who had not.

the "commitment to tea ching nutrition"

Scores on

scale were

influenced by c o m p e ti tio n for teaching time from other
subjects and high pret es t scores.

The authors concluded

that the exp eri en ce of teaching nutri tio n might result in
more favorable attitudes toward including it in the
curriculum.
A study by Reames

(1985)

was conduc te d to determine

level of c o n s i ste nc y b e t we en attitudes and practices of
physicians regarding infant feeding practices.

Scores

revealed that ph ysicians had strong positive attitudes
regarding

(1)

importance of bre as t f e e d i n g on the w e l l ­

being of infants and

(2) b r e a st fee di ng as the best form of

nourishment for infants.

However,

recommend ati on s of the

physicians re garding bre as t f e e d i n g and infant feeding
practices were not c o n s i st ent with successful
breastf ee din g practices.

Physicians whose children had

been b reastfed felt bre as t f e e d i n g was more important and
recommended b r e a s t f e e d i n g to patients more often,

46

than

physicians whose children had not been breastfed.
experiences

(having children w h o were breastfed)

Previous
seems to

have influenced the formation of positive attitudes and
subsequent related behavior

(infant feeding

recommendations).
Maiman et al.

(1979)

reported that dietitians'

own

successful weight loss experiences were significantly
correlated

(r=.38, p<.05)

with attitudes toward obesity

(self image, weight l o s s ) .

Authors concluded that

personal experience was a better predictor of attitudebelief consist en cy than knowledge or job experience.
The debate related to whet he r attitude scores predict
dietary change or whether dietary scores predict attitude
change is important to nutrition educators.

If attitude

change accounts for variance in dietary change,
implementation of educational programs should address the
affective domain to facilitate di et ary change.

If attitude

change is not necessary for dietary change to occur,
promotion of positive attitudes need not be considered as
ma ndatory in programs which promote die tar y change.

In

examining the models presented it can be seen that
investigators have reached diffe ren t conclusions based on
results of their research.

It is likely that the

relationship between attitudes and behavior
change)

(dietary

depends on other factors whic h were not included

in these models.

Instead of asking whether attitudes

predict behavior change or vise versa it seems more

47

important to investigate additional factors,
of control,

attitude of the instructor,

such as locus

and job e xp erience

of the instructor, w h i c h would indicate the ability of
several

factors to predict attitude or die ta ry change.

Locus of Control as a Pred ict or of Diet ary Change
Locus of c o n t r o l / p o wer les sn ess has been assessed in
attitude research investigations because of its proposed
relationship to behavioral change
19 76; W a l l s t o n and Wallst.on,
Hollis et al.,

1986).

(Haggstrom,

1966;

19 78; Eden et al.,

Sims,

1984;

People w h o believe events in their

lives are beyond their control are considered to have an
external

locus of control

(high powerlessness).

Ext ern al ly

c o ntrolled persons can further be classified into two
groups

(a)

those who expe ct that fate,

control events in their lives

(b)

luck, or chance

those w h o believe others

are respo ns ibl e for c ont ro lli ng events in their lives.
Persons with an internal
powerlessness)

locus of control

(low

perceive themselves as having control over

events in their lives.
H a g gst ro m

(1966)

suggested that, low-income persons

usually have an external

locus of control due to lack of

support and reinforcement.
feeling victimized,

He indicated that a sense of

due to limited income, was congruent

with a hig h level of powerlessness.

He suggested that if

low-income persons gain a feeling of increased control due
to p a r t ic ip ati on in educational programs,

the feeling of

being v i c t im ize d and powerless m a y be lessened and
48

po si t i v e behavioral change might be increased as a result
of a more internal

locus of control.

W a lls ton and Wal l s t o n

(1978)

defined locus of control

as "one's beliefs about the rela ti ons hi p between one's
behavior and its outcomes."
locus of control are more
dietary behaviors.

Persons with an internal

likely to adopt pre ventative

The y exhibit "approach" b eh avior and

are motivated by long-term benefits resulting from healthoriented behaviors.

The authors suggested that the outcome

of adapting positive health behaviors acts as
re inforcement to the person with an internal
control.

Persons with an external

locus of

locus of control do not

think that their immediate behaviors can influence
term health status.

long­

They need more reinforce me nt from

others to adopt and m ai ntain positive attitudes and
behaviors.

The authors further recommend that information

from powerlessness scales be used in

planning and

implementing nutrition education programs.
Eden et al.

(1984)

re commended that locus of control

scores be used to help plan programs in d ietary behavior
change.

They suggested that subjects with an internal

locus of control are likely to be more motivated to adopt
positive attitudes and behaviors with minimal instruction
and support than clients w ith an external
who require more support,

locus of control

use of motivational

techniques,

and more assistance in behavioral change.
Hollis et. al.

(1986)

used a semantic differential
49

attitude instrument wh i c h included a "helpless and
unhealthy"

(HU)

scale.

serum triglycerides,
pressure),

C a r d i o v asc ula r risk

(cholesterol,

lipoprotein levels and blood

medical risk,

psychological risk, and diet

habits were also assessed.

Wo m e n scored significantly

higher than men on the HU factor suggesting that women
pe rceived themselves to be less in control of their
dietary habits and less h e a lt hy than men.

High scores

were significantly n e g a t ive ly co rr elated with age
pC.Ol)

for men, not women.

"helpless and unhealthy"
medical

(r=.16, p<.05)

pC.Ol).

Men who scored high on the

scale had a high incidence of

and psychological

risk factors as did w o m e n

(r=-.31,

(r=.25 and

(r=.18, p<.05)

.25 resp ec tiv ely at

Significant p os iti ve correlations were found

between HU scores and the general

severity of total

distress as measured by the Hopkins Symptoms Checkl is t
for men

(r=.21, p<.01)

Scores on this factor
blood cholesterol
p<.01);
women:
p<.01),

and for wo m e n

(HU) were also pr edictive of high

(men: r=.18,

p<.05 and women:

low d ensity lipoproteins
r=.33,

pC.Ol);

(men: r=.21,

and serum triglycerides

diasto li c blood p ressure

density lipoproteins

(r=.24, p<.01).

r=.33,

p<.05 and
(r=.30,

(r=.25, p<.01),

(-.24, p<.01)

in women.

and high

"Helpless

and unhealthy" may be an indicator of low self-efficacy
and be indicative of inability of the individual to form,
modify,

and maintain h e a l t h - p r o m o t i n g diet ar y habits.

A study was c onducted by Ramse y and Cloyd
50

(1979)

to

assess EFNEP homemakers pre- and posttest
locus of control,
agencies,

(powerlessness)

o r i e n ta ti on toward use of professional

and d i e ta ry change as measured by 48-hour food

recalls of the p ar ti c i p a t i n g homemaker and one child.
Results indicated that powerl ess ne ss increased for both
EFNEP participants and the control
(i.e.,

(non-EFNEP)

group

locus of control became more external after program

participation).

S ig nif ic ant improvements

reported in the chi ldren's diets,
agencies to solve problems.
homemakers'

(p<.05)

were

and homemakers'

However improvements

use of
in the

food recalls were not significant.

Powerlessness was inversely related to the husband's
educational

level

(Tau C=-.40,

p<.001).

The researchers

stated that positive changes were greater

if the

individual was alre ady in the process of change and if
p r o g ram content was similar to program objectives.
Additional

factors suggested by the authors that could

have affected results of this study were strength of the
learner-teacher relat ion shi p and readiness to learn.
Davie et al.

(1973)

b etween 83 pro gr am aides'

investigated the relationship
level of powerlessness and their

prog ra m success in W a s h i n g t o n State EFNEP.

Program

success of aides was m ea sured using performance rankings
by CES Agents and other aides and improvement in 24-hour
food recalls of homemakers.

The pow erlessness scale used

included response pairs in wh ich one choice represented a
situation where a person indicated control over events in
51

their lives

(internal

alternative response

locus of control)

and the

indicated a belief that their

environ me nt was controlled by fate, chance,
powerful others

(external

luck, or

locus of c o n t r o l ) . The score of

aides indicated a low level of powerlessness
internal

locus of control).

(high

The authors suggested that an

awareness of the s oc ially a cce ptable responses might have
affected results.

There was no significant difference

between powerlessness scores of aides rated by agents or
peers as successful com pared to those rated as
unsuccessful.

However,

positive changes in food recall

scores of homemakers taught by aides with low
powerlessness scores we r e greater than for homemakers
taught by aides with high powerlessness scores.

The

authors suggested that if powerlessness scores were used
in the selection of p r o g ram aides,

the amount of

improvement on the food recalls might be increased.
Job E xpe rience as a Predictor of Dietary Change
Hunger

(1971)

co nducted a study of the national EFNEP

to determine if demogra phi c character ist ic s
years of job experience,

ethnicity)

(education,

of aides were related

to improved dieta ry recalls of homemakers.

No

relationship was found between the level of education,
years of em ployment in EFNEP,

or ethnic background of the

aide, and improvement in food recall

52

scores of homemakers.

Attitudes of Instructors as a Predictor of D i e tar y Change
The attitude of the teacher,

parent,

or healt h care

professional might influence the amount of change in
attitude and/or diet ary change of participants in
nutrition education programs.

C arruth et al.

Yperman and Ve r m e e r s c h

have investigated the

(1979)

(1977)

and

r e lat ionship between attitudes of persons in an
instructional role

(teacher,

parents)

and the attitude

and dietary change scores of learners.
Yperman and Ver m e e r s c h
attitudes

(19 79)

found that parents'

toward "importance of nutrition"

mother's educational

level

(p<.05)

(p<.01)

and

were the strongest

predictors of variety in the diets of children.

However,

food pr eferences of parents were sig nificantly di fferent
from preferences of their children

(p<.001)

indicating

that parents preferences did not d irectly account for
variance in food preferences of their children.
Carruth et al.

(1977)

conducted an a ss essment in the

Missouri EFNEP to determ in e h ow f l exi bi lit y to change,
p e rsonality factors, nutri ti on knowl edg e related to weight
modification concepts,

and requests for literature

affected ob served and verbal n u t ri tio n- rel ate d behaviors
of program aides.
Anderson

(1977)

in knowledge

Results reported by Carruth and

indicated that signifi can t gains occurred

(p<.05)

of aides that part ic ipa te d in

training sessions rel ate d to w ei ght management,

but

knowledge was not an a cc urate p redictor of behavior.

53

Twen ty-eight percent of aides agreed with the statement,
"Knowing something is good for me has little or no
influence on w ha t I choose to eat."

Eighteen percent

agreed that "traditional ways of prepar in g foods are the
best ways."

Sixty-s ev en percent agreed that "The basic

four food groups are the only usable tools for planning
an adequate diet" and 35% agreed with the statement,
"Restricting my meal patterns to familiar foods ensures
that I enjoy w h a t I eat"

indicating that aides'

attitudes

were not con sistent wi t h program content that they were
re sponsible to teach

(importance of including a varie ty of

foods in the diet. Att itude
.25, pC.Ol)

(r=.79, p<.01)

and age

(r = -

were si gnificantly correlated with aides'

nutrition-r el ate d behaviors.

Results of regression

analyses indicated that scores on these two variables
accounted for 75% of the var iance in behavior.

The authors

concluded that a lt hough attitudes of aides toward
nutrition tended to be flexible,

attitudes regarding

teaching methods and personal d i e tar y change were
inflexible.

This research indicated that the assumption

that nutrition educat io n instructors have positive
nutrit io n-r ela te d attitudes is not necessar il y valid.
Food R e c a l 1 as a Predic to r of D i e tar y Change
Limitations of using the food recall

instrument to

assess dietary intake are p resented by Madden et al.,
(1976); B eaton et al.,
Guthrie

(1979); Axelson,

(1984).
54

(1984); and

Axelson

(1984)

reported that me asurement error

influences results w h e n the 24-hour food recall
assess dietary intake.

is used to

She found that mean scores for

most nutrients increased

(but not significantly)

from pre-

to post-recalls in a study where no intervention was
included between the two assessments.

Intra-individual

v ariation

in daily nutrient

(variation w i th in subjects)

intake was reported in this study.

Axelson also suggests

that the experience of p a r t ic ip ati ng in the food recall
might account for some of the di ff ere nc e in p re -posttest
results.

The "flat-slope syndrome" also threatens

va lidity of food recall results.

This is the tendency for

persons who consume large q ua ntities of food to
underestimate amounts of foods consumed,

and for

light

eaters to ove re stimate qu ant iti es eaten.

Axelson suggests

that caution be used in c on c l u d i n g that differences
posttest food recall

in

scores are due to partici pat io n in

nutri tio n e ducation programs.
Presence of the "flat-slope syndrome" was also
reported by Ma dden et al.

(1976)

also indicated the

occurance of this syndrome in die ta ry recall reports from
elderly persons p a r t ici pat in g in special
Guth ri e
recall

(1984)

lunch programs.

indicated that the ability of the food

interviewee to estimate serving size, even with the

assistance of m e as ur ing devices and food models, was poor.
Men tended to overestimate,

and wo m e n underestimated,

serving sizes.
55

B e a ton et al.

(1979)

conducted a study to d etermine

the amou nt of va ri a n c e in the 2 4-hour food recall
hi gh ly- tra in ed investigators
six times.

in which

interviewed 60 subjects,

Based on results of their study,

each

they

indicated no effe ct of training and no d iff erence between
interviewers,

but re ported signi fi can t effect of sex and

"day of the week" for women. W hen investigators controlled
for differ enc es in energy, neither of these effects was
significant.
subjects,

Th e y rep orted that the larger the number of

the lower the b e t w e e n - g r o u p variability;

more observ ati on s per subject,
individual variation.

the

the lower the int ra ­

They c oncluded that high in tr a­

individual v ari a t i o n limits the usefulness of a one-day
food recall

to assess food intake.

The investigators

presented a formula by whi ch to es timate the minimum
number of days needed to reduce the va ria nc e to the point
where n ut rie nt estimates were valid.
Strengths of the food recall

instrument are presented

by Young et al.,

(1952); Alford and Ekvall,

Sorenson et al.,

(1985); and Pao et al.,

al.

(1985)

(1984);

(1985).

Pao et

compar ed one- and t hree-day food recalls of

subjects p a r t i c i p a t i n g in the USDA N a t i on wid e Food
Consump ti on Survey

(1977-78).

When co m p a r i n g one and

three-day food recalls there were no significant
differences

(less than 2% variation)

fat, and carbohydrate;
calcium,

b e t we en k i l o c a l o r i e s ,

less than 3% v ar i a t i o n between

iron, magnesium,

and phosphorus;

and less than 5%

variation in thiamin,

riboflavin,

niacin, and

Levels of vitamin C showed little
18 year old males and

vitamin B6.

variation except in 15-

19-22 year old females.

Mean vitamin

A values were higher for the three-day recall period than
for the one-day a s s es sm ent period.

These authors concluded

that use of the 24-hour recall was adequate to determine
self-reported intake of all nutrients assessed in this
study,

except Vitamin A.

Young et al.

(1952)

compared

the 24-hour

food recall

to a 7-day food record in three different populations:
seventh and eighth grade students,
students,

and pr egn ant women.

high school and college

Results from this study

indicated that the two methods were significantly
diffe ren t in es tim at ing kilocal or ie level of diets and
adequacy of nine nutrients
group results,

in individuals.

In comparing

there we r e no significant differences

b etween results of the 24-hour recall and seven-day record
for any of the groups except for niacin.
findings the authors

Based on these

suggest that the 24-hour recall

is as

valid as the seven-day food record in assessing selfreported dietary intake for groups, but less expensive and
time consuming.

These authors estimated that me asurement

error accounted for ten per ce nt of the variance in dietary
intake results.
Sorenson et al.
recall
intake,

(1985)

compared the 24-hour food

to a two-day food record,
and food frequency.

seven-day usual food

Results of the 24-hour food
57

recall for fiber,

calcium,

vitamins A and C wer e

phosphorus,

potassium,

and

lower for the 24-hour food recall

compared to results from the other three methods.

The

food frequency gave the highest mean values, but not after
results were adjusted for ener gy level.

Nutrients that

were most c o n si ste nt ly estim ate d when comp ari ng all
methods of as sessment were:
sodium,

iron,

thiamin,

fat,

and niacin.

values were obtained for fiber,
and C.

cholesterol,

Least consistent

calcium,

and vitamins A

There was more va r i a b i l i t y in nutrient intake

when using the sh ort-term instruments
record,

protein,

food recall)

(two-day food

than when using the food frequency or

seven-day usual food intake instruments.
Alford and Ekvall

(1984)

assessed inter-rater

reliability between nu tritionists and students,

and

between und ergraduate and g rad ua te students who conducted
diet historys.

The data co llectors reported subjects'

frequency of c o n su min g servings of the four food groups.
A s ignificant dif fe ren ce was found betw een graduate and
un dergraduate students'

reports for servings of meats and

for calcium, V i t am in D, and total calorie values.

There

were no differences b e t we en results of the combined
student groups and nutritionists.

The authors concluded

that v ari ab ili ty b e t w ee n i nvestigators is not a
significant co nt rib uto r
variance)

(accounts for <10 percent of the

to results of self-reports of foods eaten.

of diffe ren t food c o m p o s iti on tables pro bably also
58

Use

contr ib ute d to the var ia b i l i t y between investigators.
Sanjur

(1982)

provides a r evi ew of the strengths and

limitations of the food recall
that major
include:

instrument.

limitations of the food recall

She indicates
instrument

lack of accurate qua nt itative information,

inaccurate r e p r e s en ta tio n of usual food intake,
reliance on hone st y and memory of the subject.

and
She

cautions that d i e t a r y information re sulting from the 24hour food recalls shoulg not be used as an indicator of
nutritional status of subjects.

Sanjur suggests that

investigators should con sider these limitations when
using the food recall

in collection,

interpretation of results.

reporting,

She suggests that food recalls

are of some use bec aus e they are inexpensive,
qu alitative dietary information,
administer.

and

provide

and are quick and easy to

The t wenty-four hour food recall

is considered

to be valid for the as ses sme nt of die tar y intake of
groups,

but not individuals.

An objective of the d i e ta ry as se ssment is to provide
accurate information regarding eating habits of the
po pu lation surveyed.

However,

results of the food recall

instrument represent only reports of foods consumed in
response to questions from prog ram aides, not information
regarding dietary b ehaviors of subjects.

It simply

provides information wh i c h can be used to compare
differences in self-reports at two diffe re nt points in
time

(program entry and c o m p l e t i o n ) .

It cannot be assumed

that d ifferences in scores are influenced by program
partici pa tio n unless r andom ass ign me nt of subjects to
control and experimental groups is used.
Due to the importance of exam in ing multiple factors
that play a role in d i e tar y change,

several determinants

were selected as independent varia ble s for this study
including:

attitudes of instructors toward d ietary change,

instructors'

years of experience in E F N E P , locus of

control of the EFNEP prog ram part ici pa nts and instructors,
pretest and attitude change scores of homemakers.

By

c a lculating the correl ati on bet wee n attitude and food
recall change and by regressing the independent variables
on the d e p end ent variables
change of homemakers)

(attitude and food recall

information regarding attitude and

dietary change of EFNEP participants is provided.

60

CHAPTER 3
METHODS

The objectives of this research was to evaluate the
effectiveness of M ic hi g a n EFNEP in promoting positive
changes of pa rti cipants
change,

in attitudes toward dietary

to assess the relat ion sh ip betw ee n attitude

change and food recall change of enr olled homemakers,

and

to investigate the a b i l ity of indep en den t variables to
predict change in attitude and food recall scores.

Methods

which were selected and used to provide data to fulfill
these objectives are desc rib ed in this chapter.
Approval

to Cond uct the Study

Permission to seek co nse nt to p articipate from
homemakers enrolled in EFNEP and to collect data from them
was gran ted by the Assoc iat e P rogram D irector of Home
Economics Programs.
Approval

to conduct this study was granted by the

University Commi tte e on Research Involving Human Subjects
in March,

1986 prior to the co lle cti on of data

B) .

61

(Appendix

Subjects
The subjects were homemakers who initiated enrollment
in EFNEP during Augu st and September of 1986 and aides who
provided instruction to clients enrolled during that time
period.

The Unit Report Summary for this period of

homemaker enrollment pro vided demographic information
which is presented in A pp endix C.
The co mp arison gr ou p of subjects included
women w ho were enrolled
Program for Women,

low-income

in The Special Supplemental

Infants,

and Children

(WIC)

Food

in Ingham

County during the time the study was conducted. Women who
were,

or had been,

e nrolled in EFNEP were excluded from

the WIC c omp arison g rou p sample to avoid possible
contamination.
The attitudes and behaviors of pro gr am aides in all
counties were assessed at the onset and co mpletion of the
data collection period.

A descri pti on of demographic

characteristics of p r o gr am aides is provided in Appendix
C.
Research Sites
Data were collected by pr og ram aides under the
direction of County Extens io n Home Economists in ten
Michigan Counties:
Kent,

Kalamazoo,

attitude survey,

Berrien,

Wayne,

Muskegon, Macomb,

Oakland,

Ingham,

locus of control scale,

Saginaw,

and Genesee.

The

and 24-hour food

recall were administered by aides to subjects in their
homes.

The data for the comparison group were collected by
62

WIC nutritionists at the Ingham County Public Health
Services Building in Lansing,

Michigan.

Program Descrip tio n
A des cri pt ion of the p r o g r a m components is included
in A pp endix D.

The c u r ri cul um used to teach lessons

during the time of the study was

"Eating Right is Basic 2"

(Michigan State Un ive rsi ty C o ope ra tiv e Extension Service,
1986).

These teaching materials were developed by the

Michigan EFNEP staff and are used nationally.

The lesson

activities are de signed to teach basic nutrition
principles and skills related to nut ri ent needs,
preparation,
planning.

food storage,

food selection,

food

and meal

This c u r r ic ulu m does not sp ec ifically contain

infor mat io n/a cti vi tie s to promote positive attitude
change.

A lt hough attitude improvement is named as a major

program goal area,

no national prog ra m guidelines are

provided for i ncluding attitudinal

infor mat ion in

c u rricular materials.
The number of lessons taught to individual homemakers
depends on results of a co mp ete ncy -b ase d pretest which is
administered upon enrollment.

The as ses sm ent includes one

item to measure each of 126 competencies w hi ch were judged
to represent con ten t information that was essential
EFNEP homemakers to master.
an individual

for

If all items that represent

lesson are answered corre ctl y the homemaker

"passes out" of that lesson;

that is they do not receive

63

instruction in that area.

This process

is used to shorten

program enrollment by targeting instruction to specific
areas of need for individual homemakers.
R e search Design: H y p ot he ses Testing
To test the first hypothesis t-tests were used

(1)

to

determine if the di ffe re nce between pre- and posttest
attitude scores of ENFEP participants were significantly
different

(2) to deter min e if the amount of change in

attitude scores of the experimental group was
significantly greater than change in the attitude scores
of the c om parison group

(3)

to determine if homemakers

with high attitude pretest scores had a significantly
di fferent amount of change in food recall scores than
homemakers with

low pret est attitude scores.

The t-test statistic was used with pretest scores to
determine the initial equiva len ce of the control and
experimental groups.

The range,

variance of scores were computed.

standard deviations,

and

Frequencies of pre- and

posttest scores of control and experimental groups were
calculated to d etermine normal ity of distribution.
Two-way ANOV A

(Solomen Four Group Design)

was used to

assess the main effects of prete sti ng and program
partic ipa ti on while c on tro ll ing for the effects of
maturation and his tor y by using the c om parison group
(Campbell and Stanley,

1963).

In this design

(Figure 9),

the column means re present the main effect of EFNEP
part ic ipa tio n and the row means repr ese nt the effect of
64

pretesting.

Factor A:
Program Participa ti on
Factor B: Pre testing

EFNEP

Pretested
Group
No n- pre tes te d
Group

Figure 9. Two- way ANOVA:

non-EFNEP

n = 98

n = 36

n = 97

n = 30

Solomen four group design

In attitude research,

the pretest focuses the

attention of the r esp ondent on the attitudinal
poses a threat to internal validity,

issue. This

called test effect,

because the e x p e r ien ce of responding to the pretest
attitude items migh t influence posttest scores
and Morris,

1983;

Fazio et al.,

1983; Fazio,

(FitzGibbon

1986).

To minimize test effect and maximize the
ge ne ral iza bi lit y of the results,

a no n-p re tes ted group

wi thin the experimental and control groups was added to
the traditional

two- gro up experimental de sign

(Figure 9).

Results using the traditional p ret est - p o s t t e s t control
group d esign can only be genera liz ed to prete ste d groups.
Five of the ten partici pat in g counties were randomly
selected to use the att itude pretest and po stt es t These
subjects were the EFNE P "pretested g r o u p . ”

65

The staff in

the other five counties admin is ter ed only the posttest
attitude survey to h om emakers

(non-pretested g r o u p ) .

The second hy pothesis was tested by computing
correlations between a tt itude change and food recall
change scores of EFNEP participants.
The third hy pot he sis was tested by conducting
multiple regression analyses to determine whic h factors
accounted for var ian ce in food recall change scores

(the

depend en t v a r i a b l e ) . Independent variables were entered
into reg res si on analysis using the stepwise procedure.
Independent variables

included:

homemakers and instructors,

pretest attitude scores of

pretest

locus of control

scores of ho memakers and instructors,

locus of control

change scores of ho me makers and instructors, attitude
change of homemakers,
experience.

and instructors'

years of

To avoid using i ndependent variables that were

correlated in the same regression analyses,

two runs were

conducted using pre tes t scores in one c omputation and
change scores in a separate run.
To d e t er min e w h i c h factors pr edicted change in
attitude scores

(the d e pen de nt variable)

the following

independent v a ria ble s were entered into the regression
equation:

pret est food recall

scores of homemakers,

pretest at titude and locus of control scores of
homemakers and instructors,

locus of control change scores

of homemakers and instructors,
instructors,

years of experience of

and food recall change scores of homemakers.

66

To determine if aides'

years of job exp

ience in

EFNEP or attitude scores were related to amount of change
of homemakers on attitude and food recall measures,
tests we r e conducted,

t-

in additi on to regression analyses,

after di viding aide scores into "high" and "low" groups
according to pretest attit ud e scores and years of
experience.
Devel op men t of the Att itude

Instrument

The purpose of the attitude instrument used in this
study was to assess the subjects'

attitudes toward dietary

change so that the p ro posed hypotheses could be retained
or rejected.

The following steps were followed to develop

an instrument that was valid,

reliable,

and usable for the

purpose of assessing attitudes toward dietary change:
1. Existin g instruments were reviewed by the primary
research investigator to determine their usability in this
research.

Al though some items from existing instruments

were included in the initial

list of items to be reviewed,

an instrument that could be used w ith the low-income
po pu lation to assess

"attitudes toward dietary change" was

not found.
2. Two interviews w ith program aides in Ingham and Genesee
Counties were conducted by the primary research
investigator to elicit information related to the
attitudes toward dieta ry change of the target, audience
(homemakers pa rt ici pa tin g in E F N E P ) .
67

Questions were asked

by the primary res earch investigator to obtain information
from program aides regarding characteristics of homemakers
wit h positive or n ega ti ve attitudes.

Information from the

interviews was used to deve lo p items to specifically
measure the c o ns tru ct "attitudes toward dietary change" in
the EFNEP population.

By finding out h o w positive vs.

negative attitudes are demo ns tra ted in this population,

an

instrument specific to the given construct could be
developed for use w ith this population. A summary of the
interviews

is reported in A pp endix E.

3. Based on the results of the prelim in ary research
interviews,

a list of items was developed by the primary

research investigator using a c ombination of existing
and newly develo pe d items

(Appendix F ) . Specifications for

devel opm en t of a L ikert scale

(Likert,

1978)

were followed

in the co ns truction of items.

These guidelines are

presented in Ap pen dix G.
4. Co nte nt v ali di ty was determined by submitting test
items to expert reviewers for the purpose of assessing the
ability of items to repr ese nt the const ruc t "attitudes
toward diet ary change".

Six reviewers w ere selected

because of their expertise in one of the following areas:
program evaluation,

att itude assessment,

or experience

with EFNEP. A r e v i e w f orm was provided for evaluation of
individual items

(Appendix H ) .

68

5.

The items wer e revised based on the results of the

expert review.
6. The first draft of the attitude instrument

(Appendix I)

was pilot tested in Kent County with twenty enrolled
homemakers.

The survey included sixty items and used a

seven-point L ikert scale. Complete instructions including
procedures to follow to conduct the pilot test are
included in A ppendix J.
7. An item analysis was conducted to determine construct
validity and reliabil ity of the instrument with this group
of subjects.
a.

Constr uct val idi ty of items was de te rmined by

comparing item means of high scorers
scorers

(lower 18%).

(upper 18%)

If the items are valid,

vs.

low

the mean

scores between high and low groups should differ in the
expected direction.

To determine w hether individual items

discriminated between homemakers with positive vs.
negative attitudes toward d ietary change,

the

d i scr imination indices were calculated for each item. The
d i scr imination index reports the difference in mean
percent scores b e t w e e n the upper and lower groups of
scorers and ranges from -1.00 to +1.00. Only items with a
di sc rimination index of >.33 were considered to adequately
di stinguish between high and low scorers.

The

discri mi nat ion index applied in this item analysis was:

69

X

H

- X
L

Where:
X = mean item score of homemakers
H
group (upper 18%)

in the high scoring

X = mean item score of homemakers in the low scoring group
L
(lower 18%)
b. The di ffi cu lty index is the mean percentage of
respondents who selected the most positive response.
range for this index is 0-100.
be too "hard"
"easy"
c.

An item

The

was considered

if the di ffi cul ty index was

to

< 20% or too

if the d i f f ic ult y index was > 80%.

In comp ari ng total mean scores of the

scoring group of respondents,
66.37 respectively

(Table 1).

upper vs.

lower

the means were 88.10 and
The di ff erence between

these two means is 21.73.

Table 1. Pilot test means and ranges of attitude scores
for upper and lower scoring groups
Mean Score

Group

High Scorers
L o w Scorers
All Subjects

(n=3)
(n = 3 )
(n=16)

Range of
Scores

88.10

77.1 - 89.0

66.37

63.1 - 67.9

75.65

63.1 - 89.0

8. Using the criteria indicated p r e vi ous ly for acceptable
d i scr imination and d i f f ic ul ty indices for individual
items,

a second draft of the instrument was developed.
70

Aides that adm in ist er ed the initial pilot test were asked
to complete a r e vi ew for m regarding the instrument's
u sability w i t h the EFN EP popu lat io n

(Appendix K ) . Comments

from these reviews were also used to revise the format
of the instrument after the first pilot.
recommend at ion s from pro gra m aides,
scale was used

Based on

a five-point Likert

(instead of seven response categories as in

the pilot test instrument)

and the number of items was

reduced from 60 on the first pilot to 33 for the second
pilot.
9.

A second pilot test was condu cte d in Oakland County

with 25 h om emakers using an att itude instrument which
included two di ff e r e n t response formats

(Appendix L ) . An

item analysis of the second pilot test was conducted.
Difficulty and d i s c r i m i n a t i o n indices which were
calculated using the same procedures as with the first
pilot test are presen ted in A ppe nd ix M. Total mean scores
for the second pilot test of the attitude instrument are
presented in Table 2.

Table 2. Results of the second pilot test of the attitude
instrument
Group

Mean

Range

High Scorers

(n=4)

86.87

80.8 - 92.9

L o w Scorers

(n = 4)

57. 58

52.0 - 61.1

All Subjects

(n=25)

69.51

52.0 - 92.9

71

10.

Items which met standards set for d i scr im ina ti on and

di ff icu lty indices and that were approved by p r o g ram aides
regarding usability were compiled into a final draft of
the instrument whi ch was used for data collection in this
study

(Appendix N ) .

Development of the Locus of Control

Instrument

One of the independent variables used in regression
1
analyses in this study was the locus of control
of
homemakers and aides.
O) which assessed

Two existing instruments

(Appendix

locus of control were pilot tested in

Mu skegon and Berrien Counties to d etermine w hic h items
were valid,

reliable,

and usable with this population.

Indices of d i s c r i min ati on and diff ic ult y were computed
using the same standard procedures that were used during
the attitude scale pilot tests.

Items which met the

standards specified in this research

(discrimination index

of >.33; dif fic ult y index of

.20-.80)

second pilot test instrument

(Appendix L) and tested in

Oakland County.

were included in the

Items whi ch were determined to be valid

(discriminated be tw een the upper and lower 18% of scorers)
and that w ere cons ide re d by progr am aides to be usable
wit h EFNEP homemakers were included as items 1-5 on the
survey

(Appendix N) and scored as a separate scale.

1
locus of control is defined as the perceived level of
control individuals have over their environment. Individ­
uals with an internal locus of control perceive themselves
to be in control over their environment; an external locus
on control is charact er ize d by a p ercieved lack of control
over environmental situations (Eden et al.
1984).
72

Develop me nt and J u s t i fic at ion for Use of the 24-Hour Food
R e c a l 1 Instrument
Although c o n t r ove rs y exists regarding the validity of
the 24-hour food recall for individuals,

it has

co ns istently been used in EFNEP to assess die tar y adequacy
of enrolled ho memakers and is c ons idered valid for groups.
It is used because it is quick,
homemakers and instructors,

easy, we ll- ac cep te d by

and is not as intrusive as

clinical or biochemical a sse ssment procedures.

Although

the primary research investigator considers this
instrument to have severe

limitations in a ssessing actual

dietary intake or nutritional

status,

it was used as a

general measure of se lf-reported dieta ry intake of
homemakers as a group because it was the instrument in
place durin g the period of time that the attitude and
locus of control

instruments were used to collect data.

It was not an option for the prim ary research investigator
to implement the use of additional d ietary assessment
instrument or to mo dify the existing measure by taking
recalls of additional days.
The food recall instrument used the method of scoring
developed by the Synetics Scoring System for use in EFNEP
by the USDA
numerical

(Jones et al.,

food recall

on the q uan tit y

1975).

A co mprehensive

score from 0-100 was assigned based

(number of total servings)

and variety as

measured by number of servings reported from each food
group as follows:

two servings of milk/milk products,

servings of meat or meat alternates,
73

four servings of

two

fruits or vegetables,

and four servings of grains.

Additional points wer e assigned if 50% of the recommended
servings were consumed for all food groups.
The form used and procedure for scoring the food
recall used in this study to collect data related to selfreported fo od/beverage

intake is presented in Append ix P.

Training of Data Collectors
There are ten counties
program is conducted.

Aides

in Michigan in which the EFNEP
in all counties attended an

inservice conducted by the primary research investigator
at whi ch the purpose of the research and procedures for
coll ecting data wer e explained.

(A summary of this

inservice pro gr am is provided as a component of "Program
Description",

Ap pendix D ) . Aides were encouraged to

participate in the study.
aides that their scores,

It was explained to program
and the scores of the homemakers,

would not be used to eva luate their individual job
pe rformance and that the scores would be coded to provide
anonymity.

The Consent Form

and reviewed.

(Appendix B) was distributed

Aides that were wil li ng to participate were

asked to sign the form and to complete the attitude and
locus of control surveys

(pretests). Appro pr iat e use of

the a tti tude/locus of control

instrument and procedures

for compiling and retu rn ing surveys was presented.
Aides had prev iou sl y been trained in the procedure
for conducting 24-hour food recalls.

74

To maximize inter-

and intra-rater reliability,

the training included use of

"food recall kits" w h i c h include dishes wi t h serving sizes
marked so that acc ur acy of estimating portion sizes could
be i m p r o v e d .
Data C oll ection
Data w ere co llected by all p araprofessional aides who
were employed during the time period that the study was
conducted

(August 1,

1986 to March 1, 1987).

The

paraprofessional aides w ere chosen to coll ec t the data
because they were not considered to be as threatening to
the homemakers as a re search investigator might be.
A consent form
potential

(Appendix B) was prese nte d by aides to

subjects on an individual basis when they

enrolled in the program.

Questions regarding the survey

were elicited and an swered by progr am aides.
locus of control,

The attitude,

and 24-hour food recall measures were

ad mi nistered at the first lesson to all consenting
homemakers who enrolled in the prog ra m between A u gu st 1
and September 30 of 1986. The p osttest was administered to
individual homemakers as they completed par tic ip ati on in
EFNEP.
The pro gra m aide read the a ttitude/locus of control
items to subjects and then asked them to indicate their
responses by selecting one of the response choices.
Al though aides were e nc ouraged to use computer forms for
recording responses to save the step of transferring
answers from the survey to the computer form for scoring,
75

some were not co mfortable using this form.

Therefore,

whet he r answers were recorded on the computer forms or
di rectly on the survey was

left up to the individual aide.

If the surveys were used as answer sheets,

responses were

transferred by the s ecretary at the county level or by the
primary resea rch investigator after they were sent to the
EFNEP State Office.

A score for each item was obtained by

assigning a value of 1-5 with the score of five
re pr esenting the most positive response.

The total score

for individual respondents was ac hieved by summing item
scores.

Total scores we r e converted to percentages by

dividing the raw score by the total possible points.

The

item analyses were conducted by the MSU Scoring Office to
provide indices of d i s c r i mi nin at ion and di fficulty for
individual

items and to d etermine the co efficient of

reliability.
ANOVA,

To compute correlational analyses,

t-tests,

and regression analyses of group scores the SPSSX

pro gr am for m ic ro com pu ter s was used

(SPSS,

1983).

Tw enty-four hour food recalls of EFNEP homemakers
were c onducted by p r o g ra m aides at the b eginning and
co mp letion of homema ker en ro llment

(Appendix P ) . The

homema ke r was p rompted by the prog ram aide to report all
foods and beverages comsumed w ithin the previous 24-hour
period.

To increase the acc uracy of serving size

estimates,

serving sizes were marked on sample glasses,

cups, bowls,

and plates whi ch were used by aides to

conduct food recalls.

The dietary items named by
76

homemakers were recorded by program aides and then
cl as sified according to food groups and coded for computer
analysis by clerical
co mprehensive score

staff in the counties.
(percent)

A

was assigned, based on

computer analysis w h i c h represents the total number of
servings reported and number of servings w ithin the four
food groups compared to the r ecommended number of servings
(Appendix P ) . The score assigned to individual homemaker
recalls was det ermined as specified previously.
The attitude and locus of control pre- and posttests
were also admin ist er ed to the compa ri son group
WIC

(consenting

clients who were enrolled d uring the same time

period)

using the same procedures as wit h the experimental

group. The 24-hour food recall

instrument was not used

with the co mpa ris on group because the purpose of this
study was not to determ in e if the experimental group
showed greater change in food recall
comparison group.

scores than the

Pr ogr am aides in all the counties

completed the attitude and locus of control assessments at
the begi nn ing and end of the data col lection period so
that their scores could be used as independent variables
in regression analyses.

77

C H A P TE R 4
RESULTS AND DI SCU SS ION
The results of data analyses are reported and
di scussed in this chapter.

Information related to the

distr ibu ti on of scores and the effect of pretesting are
reported initially.

The results/

and a disc uss io n of the

results follows and are p resented according to the three
hypotheses of this study.
Di st ribution of Scores and Re liability
Re liability coe ffi cie nt s using the Kud er -Richardson
20 method

.803 and

.744

for the attitude pre- and posttests respe cti ve ly and

.476

and

(Kuder and Richardson,

1937)

were

.509 for locus of control pre- and posttests.

Since

this method scores the items d i c h o t o m o u s l y , only the most
positive response was

scored as correct resulting in an

underest ima ti on of reliability.
Frequency analysis was c onducted on all variables to
assess distribution,
of scores

standard deviation,

(Table 3).

78

skew,

and range

Table 3. Distribution of attitude,
food recall scores
Range

Variable

EFNEP Homemakers and WIC Clients

locus of control,

and

Mean

Skew

+_

SD

(n=261):

Attitude pretest
Attitude posttest

43-94
44-95

71.19 + 9.40
74.22 + 9.41

-0.038
-0.513

Locus of Control pretest
Locus of Control posttest

44-96
44-96

69.13 + 11.72
71.96 + 11.40

-0.071
-0.230

4-100
10-100

51.81 + 26.16
84.74 + 19.00

+0.273
-1.471

63-94
63-92

82.82 + 6.60
79.11 + 6.37

-0.583
-0.642

EFNEP Homemakers

(n=195):

Food Recall pretest
Food Recall posttest
Program Aides

(n=67):

Attitude pretest
Attitude posttest

The food recall

scores were well d istributed across

the full range of possi bl e values.
deviation, within groups,

There was a much larger

and a more normal ly distributed

curve for the food recall measure than for the attitude
and locus of control

surveys.

The values on the attitude

assessment, p a r t i c u l a r l y for p rogram aides, were

limited

to the top half of the score range as indicated by the
range and skew values presented in Table 3. This author
suggests that the occur enc e of high scores
negative skew)
response set

(and resulting

on the attitude assessment were due to

(i.e., the desire to please the investigator

and the ability of respondents to choose the socially

79

acceptable response)

(See "Strengths and Li mitations of

the S t u d y " ) . Food recall scores piled up at the value of
100%.

This author believes that the ceiling effect at the

highest score value on the food recall was due to
overrepor ti ng of foods consumed.
food recall

A weakness of the w ay the

is scored in EFNEP is that a high score can be

achieved by reporting consum pti on of several,
amounts of foods.
the food recall

or large

This is one of the weaknesses of using

instrument,

as scored in EFNEP,

measure of food intake in this study.

as a

Other limitations

and strengths of this measure are reported in the section
"Strengths and L imitations of the Study".
The Effect of Pretesting
Subjects wer e randomly assigned to "pretested" or
"non-pretested" groups.
pretesting,

To assess the effect of

mean postte st scores for the pretested

vs non-pre te ste d

(n=127)

(n=133)

groups were computed and are

presented in Table 4.

Table 4. Posttest scores of pretested and non-pretested
groups
Mean Score
Pretested
Non-pre te ste d
Group
Group

Variable

Attitude
Food Recall
Locus of Control

74.62
86.97
71.85

80

73.81
82.08
72.08

The effect of pre testing was analyzed by co mputing ttest values for the groups that completed pre- and
posttests vs. groups

that took only the po sttest

(Table

5) .

Table 5. Results of t-tests to determine the effect of
pretesting
Variable

Degrees of
Freedom

t-test
Value

P
Value

Attitude

257.81

.64

.524

Locus of Control

257.33

-.16

.871

Food Recall
(EFNEP only)

155.53

1.65

.100

Groups that wer e pretested did not score
s i gnificantly di f f e r e n t than non-pretested groups on any
of the variables.

The t-test results indicate that test.

effect was not a threat to the v alidity in this study.

The Effect of EFNEP Particip ati on on Attitudes Toward
Dietary Change of E nro ll ed Homemakers
HYPOTHESIS la: There is no di fference between pre- and
posttest attitude scores of EFNEP participants.
Eguiv ale nc e of P r e tes t Scores
It is important to determine equiv ale nc e of pretest
scores bet wee n the com pa ris on and experimental groups to
qual if y inte rpr eta ti on of posttest results.

If group

pretest means were si gnificantly different,

c omparing

81

posttest scores wo u l d not be an accurate method for
de termining the effect of program participation.

The

attitude pretest means for experimental and c omp arison
groups were similar

(Table 6).

Table 6.
Co mpa ri son of pretest attitude means of the
experimental and c o m p a ri son groups
Mean
Experimental
Group (n=98)

Variable

Comp ar iso n
Gr ou p (n=36)

71.42

Pretest attitude

70. 58

A t-test was used to compare the attitude pretest
scores of the co mp ari so n

(n=98)

and experimental

(n=36)

groups to determine if the difference was significant
(Table 7).

Table 7. Initial equ ality of c omp arison vs experimental
group pretest attitude scores
Variable
Pretest attitude

Degrees of
Freedom
78.26

t-test
Value
0.51

P
Value
.515

The t-test value for the attitude test was not
significant at the

.05 level

indicating that the

experimental and control groups were not significantly
diffe ren t at the onset of the study.

82

Results of Item An alyses of Homemaker Attitude Surveys
A co mparison of pre- and po sttest item means for the
attitude measure is provid ed in Table 8.

Table 8. Ho memaker mean item scores for the pre- and
p o sttest attitude survey
Pretest

Item
In general,
changes.

Posttest

I don't like to make
3.63

3.50

Improving the w a y I eat is important
to m e .

4.42

4.47

I can't do much to improve the way
I eat because of my income.

3.27

3.62

I often talk about nutriton with
my family or friends.

3.22

3.82

Nutri tio n is not a p rio ri ty in my life . 3.49

3. 78

I am motivated to prepare nutritious
meals.

3.59

4.04

The way I eat now d o e sn 't need
improvement.

3.61

3.35

It would benef it me to improve the
way I eat.

4.10

3.99

Nutri tio n is a boring topic to me.

3.89

4.12

There are not enough advantages to
improving my diet to make it worth
the effort.

3.80

4.07

There are many things I am more
concerned wit h than improving my
eating habits.

2.93

3.47

How con ven ie nt a food is to prepare
affects whe th er I use it.

2.85

3.18

The price of a food affects wheth er
I buy it.

3.74

3.69

83

Mea n scores on 9 of the 13 items showed positive change.
In summary,
(Table 8)

the item analyses of pre-and posttest means

indicated that homemakers:

felt less constrained

by income in making die ta ry changes;

believed that making

dietary changes woul d be beneficial;

felt that dietary

improvement was important;
family and friends;
a priority;
meals.

talked more about nutrition to

agreed more strongly that nutrition is

and were more motivated to prepare nutritious

The percent of persons selecting each response

category and indices of d i f fi cul ty and di sc rimination are
provided in Ap pendix Q.
Results of Group Data Analysis of Homemaker Attitude
Scores
The attitude pretest mean for participants in the
EFNEP program was 71.42;

the posttest mean was 75.79

(Table 9).

Table 9. Attitude pre- and posttest results of the
experimental group
Mean

Range

Variable
Attitude pretest (n=98)
A t titude post.test (n=195)

71.21
75.79

43-94
44-95

Standard
Deviation
9.93
8.46

A paired sample t-test was conducted to determine if
the differences bet wee n pre- and posttest attitude scores
of the experimental group were significant

(Table 10).

The

t-test value re pre senting differences in pre- to posttest
84

attitude scores was s i g n i fi can t

Table 10. Differences
experimental group

(p <.001).

in pre- and posttest scores for the
Degrees of
Freedom

Variable
Attitude change

(n=97)

96

t-test
Value

P
Value

-3.65

.000*

* p < .001

These results

indicate a significant improvement in

attitude scores of p ar ticipating homemakers from the time
of program entry to graduation from the Michigan E F N E P .
This author attributes these results to
characteristics of the EFNEP program.
EFNEP,

In the Michigan

attitude improvement is co nsi der ed an important

objective in addition to promotion of positive changes in
food recall and k nowledge scores.
information program.

Homemakers are taught skills which

assist them in using cognitive
make dietary changes.

EFNEP is more than an

(factual)

information to

Program content is broken down into

manageable tasks represented by 126 competencies
State Un iversity Co operative E xtension Service,
c u rriculum materials,

and related training,

(Michigan
1986). The

focus on

mastery of information and skills needed to implement
changes in food selection and prepar ati on methods which
are likely to result in improved food recall

scores

(Michigan State University Cooperative Exten si on Service,

85

1986).

Although none of the competencies specifically

addresses formation of positive attitudes, en couragement
and reinfo rce men t pr ovi ded by p r o gra m aides probably
result in positive attitude change as indicated in this
study.
Based on these findings the first null hypothesis
(la) can be rejected.
Change in Attitudes of Experimental v s . Co mp ari son Groups
HYPOTHESIS lb:
There is no difference between attitude
scores of the com par is on vs. experimental groups.
Pretest and posttest means for subjects in the
comparison group

(n=66)

are reported in Table 11.

Table 11. Attitude pre- and posttest results of the
comparison group
Mean

Range

Variable
Attitude pretest
Attitude posttest

70.58
71.08

Standard
Deviation

55-88
51-89

7.87
8.76

The t-test values comparing pre- and posttest
attitude

(p=.422)

means of the com parison group subjects

were not s ignificantly d ifferent

(Table 12).

indicates that there was no improvement

This finding

(change)

between

pre- and posttest att itude scores of the c o m p ar iso n group.

86

Degrees of
Freedom

Variable

35

p
Value
.422

•

Attitude change

t-test
Value
i
o
00

Table 12.
Differences in pre- and posttest, attitude
scores of the c omp arison group

A t-test was conducted to d etermine if posttest
scores,

and amount of change from pre- to posttests

between groups was significant
showed that the experimental

(Table 13).

(n=195)

T-test results

and comparison

(n=66)

group scores were sig nificantly diff er ent on the posttest
attitude measure

(p<.05)

and the amount of change on the

attitude measure

(posttest score minus pretest, score)

also significant

(p<.05).

was

Table 13.
Comparison vs. experimental group attitude
posttest, means (effect of program participation)
Variable

Degrees of
Freedom

Attitude posttes t.
Attitude change

121.34
127.42

t-test.
Value

P
Value

3.34
2.48

.001**
.014*

*p <.001

A two-way analysis of variance
(n=261)

(ANOVA)

was conducted

to test, for the main effects of program

participation

(EFNEP)

and pretesting

(Table 14). F-ratios

were computed to dete rmi ne the significance of F
value)

(p-

to test for the effect of p r o gr am p ar ticipation at

87

the

.01 level of significance.

The F-ratio represents how

much group means va r y compared with h o w much means within
a group vary.

A si gnificant F-value indicates that the

variance is great b e t we en the groups and minimal within
the groups.

Error is the un explained varianc e

and Norpoth,

1976).

(Iversen

Table 14.
Two-way A N O V A of attitude postte st scores to
test for the effects of EFNEP and pr etesting
Source of
Variat io n

Degrees of
Freedom

Main Effects
pretesting
EFNEP

Sum of
Squares

Mean
Square

F
Value

1
1

58.829
892.863

58.829
892.863

.685
10.396

Interaction Effect
EFNEP by
1
pretesting

.437

.437

.005

Explained
Residual

3
256

935.453
21987.608

311.818
85.889

3.630

Total

259

22923 .062

88.506

* p <.05

P
Value

.409
.001**
.943
.014*

** p <.001

The effect of p r o gr am p art i c i p a t i o n on attitude
po st tes t scores was s i gni fic an t at the

.001 level

(p=.001)

indicating that a significant amount of the variation in
attitude scores bet wee n the experimental
c o mparison groups
part ic ipa tio n

(n=66)

(n=195)

and

can be att ributed to program

(Table 14). A N O V A results indicated that the

effect of pre testing on attitude po sttest scores was not
si gnificant

(i.e., test effect was not a threat in this

88

study).

Based on these findings,

Hypothesis

lb can be

rejected.
This author attributes the significant differences in
attitude scores between the groups to differences in
characte ris ti cs of the programs.

The WIC program includes

a nutri ti on education component w h i c h is primarily
cognitive based.

If knowledge changes had been evaluated,

it is possible that the co mparison group would have shown
improvement on that variable.
with the WIC program,

Based on personal experience

it the ob servation of the principal

research investigator that the time involved in nutrition
education in WIC is minimal.
to improve the nutritional

The primary purpose of WIC is

status of participants

determined to be at nutritional risk by providing coupons
which can be redeemed for foods.

The educational component

is a minor component of the program which provides
information related to nutrient needs during pregnancy and
infancy.

Attitude change is not a prima ry goal of WIC.

In contrast,

EFNEP is an educational program rather than a

supplemental

food program.

The clients spend several

hours in on e-to-one instruction. Goals of the EFNEP
pro gra m are to improve knowledge,
behaviors of low-income homemakers

skills,
(UDSA,

attitudes,

and

1976). EFNEP

aides in Mi chigan receive training to assist them in
improving attitudes and dietary ad equacy of clients.

Such

training has not been provided for WIC Nutritionists who
participated in this study.

89

The Ef fect of Pretest Attitude Scores on Food R e c a l 1
Change of EFNEP Homemakers
HYPOTHESIS lc: There is no difference between food recall
change of groups wi t h high vs. low pretest attitude
scores.
Pretest attitude scores of homemakers
divided at the media n into two groups.

(n=90) were

The range of

pretest attitude scores in the high scoring group was 4369. The range of scores in the low attitude group was 7194.

A comparison of food recall means is p resented in

Table 15.

Table 15. A com par is on of change in food recall scores for
groups with high vs. low p retest attitude scores
Group

Mean Food Recall Change +_ SD

Low pretest attitude
High pretest attitude

(n=39)

33.10 + 23.57

(n=51)

41.49 + 27.81

When amount of food recall change between the groups
was compared using t-test analysis,

results indicated that

although the amount of change on the food recall measure
was significant

(p< .001)

for all subjects as a group,

when subjects were divi ded a ccording to high vs.
pretest attitude scores,

low

there was no diffe ren ce between

mean food recall change scores.

90

Table 16. T-test analysis of food recall change in
homemakers with high vs. low pretest attitude scores.
Variable

Degrees of
Freedom

t-test
Value

p Value

Food Recall Change
High vs. low pretest
attitude groups

87.02

-1.55

All subjects

89.00

-13.68

.126
.000*

* p <.001

These findings indicate that Hypothesis

lc can be

retained i.e., there is no dif ference in amount of food
recall change between groups related to pretest attitude
scores.

The large standard deviations

(Table 15)

co ntributed to the t-test analysis finding of "no
d i f f e r e n c e s ".

The Re la tionship Between Attitude and Food Recall Change
Scores of EFNEP Homemakers
HYPOTHESIS 2: Attit ud e change of homemakers is not
correlated with food recall change of EFNEP homemakers.
Correlational Analysis
C o rre lations were computed to determine the strength
of the relati ons hi p betw een attitude and food recall
change scores of homemakers.

The c o rre la tio n between

attitude and food recall change was
significant at the
study,

.15 which is not

.01 level. Based on the results of this

it ca nnot be concluded that changes in attitude and

91

food recall

scores are related.

A report of all

co rr elations between attitude and food recall variables is
prese nt ed in Table 17.

92

Table 17. Correlations between homemaker attitude, locus
of control, and food recall scores.

Variable

1

2

3

4

5

6

7

Locus of Control
.266
1. Pretest
1 .000
.605** .420** .101 - .325* «.157
.266 1 .000
2. Posttest
.610** .205
.483** .209 •.068
.605** .610** 1.000 - .176
3. Change
.316*
.439**
Attitude
4. Pretest
5. Posttest
6. Change

.420
.101
.325*

.205
-.176
1 .000
.406**- .635**-048
.483**
.316*
.406**1 .000
.448**- 320
***
.209
.439**- .635** .448**1 .000

8

9

.126
.048

-.050
-.041
.007

.005
-.166
***

.017
.164
.122

.251

***
***

Food Recall
7. Pretest
8. Posttest
9. Change
* p <.01

.157
.126
.050
** p< .001

- .068
.048
- .041

***
***
.007

-.048
.005
.017

- .320
.166
.164

***
***
.122

*** correlation not calculated

,000
.251
***

1.000
***

1.000

A summary of studies cited in the review of the
literature,
be havior

reporting correlations between attitude and

(food recall)

scores is presented in Append ix A.

Some investigators reported significant positive
correlations between n u t rit ion -r ela te d attitudes and food
intake measures
Sims,

1956; Baird and Schutz,

1978; O'Connell et al.,

others did not
and Guthrie,
1984;

(e.g. Brehm,

1981;

Dalton et al.,

(e.g. Picardi and Porter,

1982; Ross,

Sunseri et al.,

1984;

1976;

1986);

1976; Daelhousen

Byrd-Bredbenner et al.,

1984; Ries and Shoon,

1986).

Results

from this study regarding the relat io nsh ip between
attitude and dietary change of homemakers show that these
factors are not correlated.

Based on this finding the

second research hypothesis can retained.
The Ability of Independent Variables to Predict Change in
Attitude and Food R e c a l 1 S c o r e s .
HYPOTHESIS 3a: Food recall change is not predicted by:
instructor years of experience, pretest attitude scores of
instructor, homemakers' locus of control change score,
attitude pretest or change score of homemakers, or pretest
food recall scores of homemakers.
HYPOTHESIS 3b: Attitude change is not predicted by:
instructor years of experience, pretest attitude of
homemakers, homema ker pretest or change in food recall,
instructors' p retest attitude, homemakers' pretest or
change in locus of control scores.
Regression Analysis
Regression analyses were conducted to dete rmi ne if
change in attitude or food recall scores depended on
several independent variables.

The results are reported

related to the per cen ta ge of variance
94

(R-squared)

accounted for by each independent variable.
of all independent variables

(total variance)

The variance
represents

the degree of pre dic ti on of the de pendent variable.
each phase of the stepwise procedure,

At

the "fit" of the

independent variables was analyzed while co ntrolling for
the variance contr ib ute d by the other independent
variables.

D ependent Variable:
Food recall change of homemakers
(posttest minus p retest score)
Independent Variables
First Regression Run
Instructors pretest attitude
Homemaker attitude change
Homemaker pretest food recall
Second Regression Run
Homemaker p retest attitude
Homemaker locus of control change
Instructor years of experience
Figure 10. Regression analysis model to assess the ability
of independent variables to predict food recall change

Independent variables were r egressed on food recall
change using the stepwise procedure in two separate
analyses to avoid use of variables that were correlated
withi n the same analysis

(Figure 10).

Results of these two regression runs indicated that
only one factor,

homemaker pretest food recall scores,
95

entered into the re gr ession equ ation

(Table 18).

Table 18. Results of the first regression analysis run to
predict variance in food recall change
2
Step/
V a riable

Multiple
R

1. Homemaker food
recall pretest

.7608

R

.5788

FValue

Signicance
of F

119.54

.000

Beta

7608

This factor accounted for 57.88% of the v ariance in food
recall change
variables

(p<.001). None of the other independent

included in this run,

or the second regression

run, a ccounted for a significant amount
variance

in food recall change.

(p < -05)

of the

Based on these results

Hypothesis 3a can be retained for all variables except
pretest food recall scores of homemakers.

These findings

indicate that 57.88% of the change in food recall scores
from beginning to end of program p ar tic ipa ti on in Michigan
EFNEP can be predicted by the pret est food recall score.
Independent variables that w ere entered into the
re gr ession equation to predict attitude change
variable)

(dependent

are listed in Figure 11. Two separate regression

runs were conducted so independent va riables that were
correlated

(Table 17) would not be entered into the same

regression equation.

96

De pendent Variable:
At ti tud e change of homemakers
(posttest minus pretest score)
Independent Variables
First Regression Run
Instructor pretest attitude
Instructor years of experience
Homemaker pre tes t locus of control
Homemaker pretest food recall
Second Regression Run
Homemaker locus of control change
Homemaker pretest attitude
Homemaker food recall change
Figure 11. Regression analysis model to assess the ability
of independent v ariables to predi ct attitude change

Results of the first reg ression analysis run
indicated that pretest locus of control accounted for
9.83% of the variance in attitude change

(Table 19).

Table 19. Results of the first reg re ssi on analysis run to
predict va riance in attitude change
2

Step/
Va riable

Mu ltiple
R

1. Homem ak er pretest
locus of control

.3136

R

.0983

FValue

10.25

Signicance
of F

.002

Beta

-.3136

The other independent variables that were included in this
run

(pretest food recall,

instructor years of experience,

97

and instructor p r e te st attitude score)
significance

did not have a

level w h i c h was high e nough

(p <.05)

for

entry into the re gre ss ion equation.
Results of the second reg res sio n run to predict
va riance in attitude change showed that pretest attitude
scores of homemakers accounted for 40.2 8% of the v ari an ce
of attitude change and that locus of control change scores
predicted 51.35%
varia bl e

(pC.OOl)

of the v ari anc e in the depend ent

(Table 20).

Table 20. Results of the second r egr ession analysis run to
predict variation in a ttitude change
2
Step/
Variable

Mu lt ipl e
R

R

FValue

Si gn ificance
of F
Beta

1. Homemaker
attitude pretest

.6347

.4028

58.69

.000

-.5751

2. Homema ker locus
of control change

.7166

.5135

45.38

.000

.3379

The other independent va riable wh i c h was included in the
second r egr ession run

(food recall change of homemakers)

did not have a sig ni ficance level

(p C.05)

high enough to

result in entry into the r egr ession equation.

The total

variance in attit ude c hange ex plained by the independent
variables in the first r egr ession run was 9.83%;

the total

variance predicted in the second regr es sio n run was
91.63%.
Several studies w e r e reviewed in this di ss ertation
98

that proposed two models:

"attitudes predict dietary

change" or "dietary intake predicts attitude change".
Results from this study show that most of the va riance in
attitude change was accounted for by pretest locus of
control

scores,

locus of control change scores,

and

pretest attitude scores, but not food recall scores.
Therefore the model wh i c h proposes that food recall scores
predict attitude change is not supported by this research.
These results indicate that nu trition e ducators/evaluators
can p redict change in attitudes from pre- to posttesting,
using pretest attitude, pretest

locus of control,

and

locus of control change scores of homemakers as
predictors.
Varia nc e in food recall change was not accounted for
by attitude scores

(although pretest food recall scores

were a significant predictor)

indicating that the

"attitude predicts food recall change" model does not
apply to this sample of low-income homemakers
participating in the Michigan EFNEP program.

These results

indicate that most of the v ariance in food recall change
was not predicted by variables included in this study.
Variables other than those included in this research
should be assessed in an attempt to find predictors of
dietary change.

Suggestions regarding such research will

be discu sse d in more detail in Chapter 5.

99

Results of Data Analysis Related to Attrition
If the pretest means of subjects that dropped out of
the progr am are significantly higher or lower than persons
who completed the prog ram
affected.

(posttest),

results will be

By comparing the number of cases

(percent)

lost

from the experimental and co mparison groups and the
pretest means of subjects that did vs.
complete program participation,

those that did not

the potential threat of

attrition can be determined.
The pretest means for subjects that dropped out. of
the pro gra m compared to pretest means of respondents that
did complete the program

(and posttest)

are presented in

Table 21.

Table 21. A comparison of pretest means for subjects that
did, and did not, take the posttest.
Gr ou p/Variable

Attrition Rate

Aides:
Attitude/
Locus of Control

20.90%

Homemakers:
Attitude/
Locus of Control

22.10%

79.71

69.21

Food Recall
Comparison Group:
Attitude/
Locus of Control

Pretest Means
Graduates
Dropouts

13.50%

100

80.73

71.21

44.48

48.75

70.06

70.56

There was a higher percentage of dropouts in the
experimental

(22.10%)

than the comparison groups

This is probably due to differences

in the programs.

the WIC program the client receives food;
participants receive education.

(13.50%).
In

in EFNEP

Food coupons might

provide a stronger incentive to remain in the program than
education.
Although there was a d if ference of 8.6% between the
number of persons that dropped out of EFNEP compared to
the non-EFNEP co mparison group,

results were not affected

by pretest scores of persons who terminated because only
matched pairs of

scores were used in analyses of data.

pretest means of

homemakers that did and did

the posttest. wer e similar

The

not complete

(Table 21).

The pretest attitude means of dropouts vs. completers
in the c omp arison group were very similar

(Table 21).

Pretest scores of WIC clients who did not take the
posttest were:
low scores

83.3,

66.7,

(21.5 and 12.0)

on the pretest mean.

60.0,

21.5, and 12.0.

The two

had a dramatic lowering affect

These results

indicate that the

pretest mean scores for the co mpa ri son group may have been
lower if the persons that did not take the posttest had
been included in

data analyses.

This was not

this study because pretest scores of

persons

a threat in
w ho did not

take the posttest w ere not inlcuded in data analysis.
The pretest attitude means of the program aides that
terminated employment before the attitude posttest was

101

conducted were similar to pretest means for aides that
completed the pre- and posttests

(Table 21).

In summary, a ttrition was not a threat to results of
the data analysis in this study because pretest scores
were not included unless a matched po sttest score was
available and percentages of persons that terminated
program par ticipation were not great enough to affect
sample sizes.
Summary of Results
There were three research hypotheses in this study.
The first hypothesis stated that

(a)

there was no

difference between pre- and postte st attitude scores of
the experimental group

(homemakers enrolled in E F N E P ) , (b)

there was no di fference between posttest attitude scores
of the experimental vs. c om parison group and
no d iff erence between change in food recall

(c) there was
scores of

homemakers with low vs. high pretest attitude scores.
Results of t-tests and ANOVA indicated that homemakers
attitude

(p=.001)

and food recall

scores

(p=.001)

showed

significant improvement from pre- to posttests and that
attitude posttest scores
si gnificantly greater

(p=.001)

(p=.05)

than p os ttest or change

scores of the comparison group.
study,

parts

rejected.

(a) and

and change scores were

Based on findings of this

(b) of this hypothesis can be

Results of t-test analyses indicated that

homemakers w ith low pretest attitude scores did not show
significantly more change on the food recall posttest
102

(p=

.126).

Based on this finding part

(c) of the first

hypothesis can be retained.
The second re search hypothesis stated that attitude
and food recall change scores of EFNEP homemakers were
not correlated.

C or relational

analyses indicated that

homemaker change scores on attitude and food recall
variables were not corre lat ed indicating that the second
research hypothesis can be retained.
The third research hypothesis stated that independent
variables entered into the regression e qu ation would not
account for variance in
attitude change.

Part,

(a)

food recall change or

(b)

(a) of this hypothesis can be

retained for all independent variables except pretest food
recall

scores of ho memakers whi ch accounted for 57.88% of

the variance in food recall change.
Part

(b) of the hy po thesis can be rejected for

pretest and change in locus of control

scores and pretest

attitude scores of homemakers and retained for instructor
years of experience,

pre te st food recall of homemakers,

pretest attitude scores of instructors,
change of homemakers.

and food recall

The amount of v ari anc e in attitude

change whi ch was accounted for in the first reg ression run
was 9.83%.

The total pre di cti ve abil it y of the variables

entered into the re gre ss ion equation in the second run was
91.63%.

Only v ariables that were not correlated w ere used

wi t h i n a single reg res si on run.
103

STRENGTHS A N D L IM IT ATI ONS OF THE STUDY
The usability of results of attitude assessment
depends on the val idi ty and reliability of the instrument
used.

Matheney et al.,

(1987)

suggested that failure to

verif y va lid it y of attitude measures has severely limited
the interpretation of research findings in the area of
attitude assessment.
measurement

Steps must be taken to insure that

instruments address the c onstruct of interest

by including expert r e vie w or factor analysis in the
instrument d e v e l op me nt stages.
deter min e if the instrument

It is also necessary to

is usable with the identified

population and if it dem ons tra te s reliability.
Sims

(1981)

of attitudes.

examined issues regarding the measurement

She suggested that attitude assessment can

be strengthened by using instruments that measure a common
dimension,

are reliable

(consistent), and valid

(measure

what they are supposed to measure) . Sims stated that,
methodological

factors such as re liability and validity

also influence the strength of the co rrelation between
attitude and behavior

(dietary c h a n g e ) .

Va li dit y and Reliabi li ty of the Measu re men t Instrument
A strong point of this study was that appropriate
steps were taken in the p re li min ar y research and pilot,
test stages of this study to insure that the attitude
and locus of control

instruments were valid,

a p propriate for use with EFNEP clients.
104

reliable,

and

Content va lidity is the degree to wh ich test items
represent a specific domain
1967).

In this study,

(construct)

(Shaw and Wright,

content val idity was verified in the

development stages of the attitude instrument by expert
reviewers.
Construct v al idi ty is demonstrated if persons w ith
differing attitudes a ctu all y respond d i ffe ren tl y on the
attitude measure i.e.,

the measure discriminates between

defined criterion groups in the expected direction
and Wright,

(Shaw

1967) . A val id scale consists of items that

yield different scores for high vs.

low scoring groups

(a

high di scr imination i n d e x ) . Construct val id ity in this
study was determined during pilot testing by comparing
item means of homemakers with high vs

low scores.

Threats to Internal Va li dit y
Threats to internal validity include history,
maturation,

and selection

(Campbell and Stanley,

1963).

Use of the comparison group minimized the threats of
history and maturation.

A control group could have been

formed by delaying e nro llment to a grou p of eligible
persons and adminis ter in g the assessment to this group
when the experimental group was tested.

Ra ndom assignment

of subjects to such a control group would have controlled
for the threat of selection.
in this study because

This procedure was not used

it was considered to be unethical to

keep potential EFNEP clients from entering the program for

105

the eight-month period that data were collected.

It is

likely that negative feelings wou ld be formed by
w i thh old in g enrollment, which could affect scores of the
control group.

Attr it ion of the "delayed entry group"

would probably be high due to loss of interest during the
w aiting period.

For this reason,

the threats of history

and ma tu ration were con trolled by using a comparison group
formed using clients from the Supplemental Food Program
for Women,

Infants,

and Children

selection into EFNEP vs.

(WIC). The threat of

non-EFNEP groups could not be

controlled because subjects were self-selected rather than
randomly assigned to co mp ari son and experimental groups.
The threat of selection into pretested vs.

non-pretested

groups was controlled by randomly assigning subjects

(by

county for the experimental group and by alternation for
the comparison g r o u p ) .
To determine if "selection" confounded results,
pretest scores of the experimental and comparison groups
should be compared

(Campbell and Stanley,

were in this study.

Results of the t-test

1963), as they
(Table 5)

indicated that p retest attitude and food frequency scores
for the experimental

and com parison groups were not.

significantly d i f fer en t indicating initial equivalence of
groups.
A limitation of all attitude m eas ure me nt which poses
a threat to internal va li d i t y is the time lag between
the message

(educational intervention),a formation of new

106

attitudes,
changes.

intent to change,

and subsequent dietary

These steps related to the change process might

or might not occur w it hin the time of the educational
intervention and assessment.

Changes in attitude or food

recall scores are used as evaluative measures to determine
program effectiveness.
from the educational

However,

attitude change resulting

experience continues to evolve after

program pa rticipation has ended.
acted as the initial

stimulus.

attitude and food recall

The program has only

Therefore,

posttest

scores might underestimate

positive changes resulting from prog ram participation.
Connell et al.
scores

(1985)

suggested that posttest attitude

"are the result of a relatively brief exposure to

instruction."

Posttest as sessment scores may

underestimate the actual effect of the intervention on
attitude and dietary change for this reason.

"Holding the

program accountable for having a strong immediate effect
would be unfair"

(Henerson et al.,

1978).

The assessment of attitudes uses indirect
measurement.

Campbell

and Stanley

(1963)

stated that

"Attitude is a hypothetical or latent variable,

rather

than an immediately observable variable. Attitudes
constantly fluctuate depending on the time of day,
influence of others,
other factors."

changes in related attitudes, and

"An attitude is not something we can

examine and measure in the same way we can measure the
heart rate.

We can only infer that a person has attitudes

107

by her words and actions"

(Henerson et al.,

1978).

Since it is not possible to measure attitudes
directly,

assessment of them is based on self-reported

responses given item statements as a stimulus.

These self-

reports vary d epending on the hone sty of the respondent
and their ability to understand the statement and to
categorize their attitu de s into fixed response choices.
There is an as sumption that it is valid to accept a
person's responses about their own attitudes as accurate
indicators of the attitude.
obtained overestimate

It is likely that scores

true attitudes because of the

tendency for people to represent their attitudes as more
positive than they ac tu all y are.
The food recall
food intake.

instrument is an indirect measure of

The a cc uracy of this measure depends on

honesty of s el f-reported food consumption,
the respondent to recall

foods consumed,

the ability of

and the accur ac y

of the data collector to e stimate number and size of
servings of food consume d based on the information
supplied by the respondent..

It is not. possible to

disti ngu is h whether im provement in food recall

scores

represents actual d i e ta ry change or whet he r nutritional
kn owledge which was acquired in the educational program is
applied to the reporting of foods consumed i.e., the
desire to please the investigator by reporting consumption
of foods known or p e r cei ve d to be

"nutritious".

The

va lidity of these instruments also depends on whether the

108

data collectors

introduced bias when obtaining or

reporting information from subjects.

A lt hough aides were

told that scores of homemakers would not affect evaluation
of their job performance,

there might have been a tendency

to encourage positive responses on the attitude measure.
Interpretation and usabi lit y of the results of data
collected from the food recall instrument is limited due
to these we akn ess es w h i c h pose a threat to internal
validity of the food recall

instrument.

The procedures for

scoring and the jus tification for use of the 24-hour food
recall

instrument are presented in Cha pte r 3, pp.

72-73.

Threats to External Val id i t y
External va li dit y is the extent to which results can
be g eneralized to, or across,

persons,

settings,

or times.

External va li dit y can be maximized by ra ndom selection of
subjects from the unive rs e to wh ic h the investigator plans
to generalize.

In this study, data were collected from all

consenting pa rti cipants who were e nrolled in the Michigan
EFNEP during a specific time period.

Although it is

desirable to ge neralize results of this study to EFNEP
clients in all states,

results can only be general iz ed to

participants in the Michi gan progr am because subjects were
not randomly selected from all states.
Response set is a threat to external validity. Orne
(1962)

stated that any feature of a research setting or

procedure that might indicate to the subject h o w they
should respond affects responses
109

(response s e t ) . Answers

that will please the investigator might be given instead
of true responses.
(acquiescent),

o r

If someone wa nted to appear agreeable
if they want to "go along with"

socially acceptable response,

the

they might select primarily

the "agree" response category.
Another type of response set is when the respondent
does not read the item statements but simply indicates
their reponses by marking a single response c at egory for
all items.

This type of response set can be minimized by

wording items pos itively and n ega tively and placing them
randomly on the questionnaire.

The surveys can be

spotchecked to deter mi ne if this type of response set is a
threat to validity.

Review of the questi on nai re s by the

primary research investigator indicated that multiple
selection of a single response did not occur on the
attitude assessment.

However,

it is ve r y likely that

acquiescence and pe rceived social d es ira b i l i t y of
responses were confounding factors in this research.

The

socially acceptable answer is frequently obvious on an
attitude measure,

and that response might be selected even

though it does not represent the respondents'

attitudes.

Interest in p le asing the investigator mi ght have affected
results,

esp ecially on the p os ttest because of the close

relationship that might have bee n e st abl ish ed between the
aide and the homemaker.
The measu rem en t setting was not h i ghl y standardized
and poses a threat to the external v ali dit y of the study.
110

Although instructors wer e provided w ith group training
regarding data c o l lec ti on procedures,

a d mi nis tr ati on of

the survey was cond uct ed on an individual basis in the
clients'

homes.

Many factors could have affected

homemaker responses including:

interruptions from

children,

fluctuations

neighbors,

or phone;

environmental conditions or noise
of the week;

level;

in

time of day; day

and events that occurred in the home on the

day of the testing.

Aides conducted data c ol lection at

various times of the day and week.
be controlled.

This v ari abl e could not

It is possible that time of the month

related to receipt of food stamps may have co nf ounded the
results of the 24-hour food recall.
Reliability is the ability of the m ea su rem ent
instrument to ma intain stability and consist en cy over
repeated use.

Stab ili ty is measured by correl at ing scores

from subsequent tests.

Internal c ons is ten cy is measured by

correlating item means withi n a single test
a dministration

(Carmines and Zeller,

coefficient of rel iab il ity

1979).

The K-R 20

(Kuder and Richardson,

1937)

was used in this study to deter min e the r eli ab ili ty of the
attitude measure.

Beca use the K-R 20 method is based on a

dichotomous system of scoring,

it underestimates the true

reliability of the instrument.

The Cr on bac h alpha method

of assessing reliab il ity might have been a more
appropriate statistical procedure to use.

Ill

We ak nesses of the Like rt Scale
The Likert scale does not have a true zero point
(point at whic h attitudes change from positive to
negative)

and response categ ory intervals are not

demonstrated to be equal

(Shaw and Wright,

1967) .

The "undecided" response choice is not a true zero
point.

A respondent might choose this response category

for a varie ty of reasons:

because they have no opinion

regarding the statement, because they have some positive
and some negative feelings regarding the statement,

or if

they have not thought enough about the issue to select a
response that indicates agre em ent or disagreement.
The Likert scale makes the assumption that the
distances between response categories
agree,

no opinion,

equal.

disagree,

(strongly agree,

and strongly disagree)

are

It can be assumed that "strongly agree" has more

"agreement"

than other response categories,

but it. is not

known if the distance between all response categories

is

equal.
Because equal intervals cannot be verified,
Likert scale is an ordinal,

rather than interval,

the
scale.

An assumption in the use of parametric statistics is that
the scale is interval or ratio level
However,
analysis,

parametric statistics
ANOVA,

commonly used,

(not ordinal).

(t-tests,

regression

Pierson product moment correlation)

are

rather than the less powerful non-

parametric tests,

in analy zin g data coll ect ed using the
112

Likert scale.

N on - p a r a m e t r i c tests do not assume that the

scale has equal

intervals

use with ordinal

(i.e.,

level s c a l e s ) .

parametric statistics are used,
statistic,

they are appropriate for
But when the nonthe strength of the

and ther ef ore the c onclusions made from the

results, are weaker.

More specific information can be

obtained when p a r a m etr ic statistics are used,

but the

assumptions u n d e rly ing the use of the tests are not
co mp letely met

(Blalock,

1974).

"The appro pr iat en ess of a given statistic is
co nditioned by the nature of the scale against which
measurements are m a d e ... Having measured a set of items by
making numerical a ss ign men ts in accordance with a set of
rules, we are free to change the assignments by whatever
transformations will p re serve the empirical

information on

the scale...the empirical operations that underl in e the
scale d etermine what tra nfo rm ati on can be made without
sacrifice of information,
transformations determine,
statistical measures"

and the permissable
in turn,

(Blalock,

On the other hand,

the appropriate

1974).

B o h r n ste dt and Carter

(1971)

argue

that regression analysis is so robust that the assumption
of equal

intervals is not a serious limitation of data

analysis when the L ike rt scale is used.
It is the o pinion of this r es earcher that use of
pa ra metric statistical p roc edures
regr ession analysis,

(t-test, ANOVA,

Pear son p r odu ct- mo men t correlations)
113

is justified to analyze these data even though
Likert scale intervals have not been demon st rat ed to be
equal.

The ordering among response categories is

preserved by applying weights

(e.g.

1-5)

and using the

scores obtained in data analysis procedures.
Rationale for Use of the Likert Scale
In the second field test of the attitude instrument,
two d i ffe re nt response formats were used,

a five-point

Likert scale and a magnitude e sti mation scale

(Appendix L)

The same item statements were used on both versions of the
survey;

only the method for selecting responses varied.

These two versions of response format were used to
determine w h i c h scale would result in a minimum loss of
information and be easiest for aides and homemakers to
use. With the m agnitude estimation scale,

the homemakers

could assign any value from zero to ten to their response
depending on strength of a greement w ith the item
statement.

This type of scale has the highest potential

for c ol lecting exact response information.
require the re spondent to "force"

It does not

their answers into fixed

response categories as does the Likert scale.

Another

advantage of the magnitude e sti mation scale is that it is
a true interval

scale wh ich satisfies the assumption

needed to justify use of parametric statistics
1981). The Likert scale is an ordinal
intervals are assumed.

In theory,

114

scale;

(Lodge,

equal

the magnitude

estimation scale is superior to the L i ke rt scale, but this
was not the case when it was used w ith this population.
Although homemakers w er e instructed to select a number
from 0-10 to express their level of a greement or
disagreement w i t h an item,

they almost exclu siv el y

selected 0, 5, or 10. This limitation resulted in a
greater loss of true response information than with the
Likert scale which included five fixed response
categories.
Although the magn itu de es timation scale has the
advantage of being interval rather than ordinal,
the potential
information,

and has

to c o l lec t the most accurate response
it is not as effective in e liciting response

information from the EFN EP po pulation as the Likert scale.
Therefore,

a five-point Likert scale was selected for use

in this study.

115

C H A P TE R 5
REC OMM EN DAT ION S
Based on the findings of this research,

this author

has several re commendations to make r egarding inclusion of
the affective domain in instructor training and
implementation and eval ua tio n of n utrition education
programs.

In addition,

recommenda tio ns related to the

assessment of attitudes and die ta ry change are presented.
Implementation of N ut r i t i o n Education Programs
To effectively promote improvement in attitudes
toward dietary change,

the nutrition education

intervention should include an a ffective-based component.
An objective of instructors and supporting materials
should be to assist clients w it h formation of positive,
and replacement of negative,

attitudes.

Instructors can as sist homemakers through the process
of improving attitudes toward d ietary change and in making
dietary improvements using a c ombination of the following
methods.

Planning sessions w her e the learner and

instructor specify goals to be reached by program
co mp letion related to dietary change could be c onducted at
the begi nni ng of the lesson series. The food recall
pretest could be used for d i s cu ssi on purposes during this
116

goal-setting procedure.

Circumstances wh i c h influence

current dietary patterns such as environmental cues which
trigger positive or negative diet ary behaviors should be
identified.

Instructors can provide e nc ouragement and

reinforcement to facilitate the m o d ifi cat io n of existing
attitudes and dietary behaviors.

In EFNEP,

integration of

new attitudes and behaviors into the existing cognitive
framework can be promoted by encou ra gin g the participant
to discuss attitudes using the compl ete d attitude/locus of
control pretest as a stimulus for such a discussion.
Ev al uation of changes in attitude and dietary
behaviors should be conducted after pa rticipation in
programs which include an affective- and b eh avior-based
intervention.

It is likely that greater changes in

attitude would result if instructors were trained in use
of methods to promote positive changes
dietary intake.

in attitudes and

In the past many programs have taught

p rimarily c og nitive-based information, but have used the
food recall

instrument in addition to a knowledge

assessment to determine impact of program participation.
Changes on the food recall from pre- to p o stt es tin g might
be due to several factors.

Additional ass essment using

methods which assess diet ary behavior,

rather than reports

of "foods eaten yesterday" are needed.

It is the

recommendation of this author that it is time to go beyond
the use of the food recall instrument as a measure of
dietary change,

and include instruction and assessment
117

related to die tar y behav ior change.
Instructor Training
Instructors need a strong system of support in
working to promote positive changes in attitudes and
dietary behaviors.

T hey might be fru strated from w o r ki ng

in a situation where attempts to promote attitude and
dietary change are neg at ed by many factors whi ch are not
within the control of the instructor.

Instructors need to

be reminded that their teaching r es pon si bil ity is to
provide information,

activities,

and en cou r a g e m e n t to

promote positive changes in attitudes,

skills,

and dieta ry

behaviors, but they c a nn ot assume the ch ange process for
the learners and should not feel def eated when clients do
not choose to implement changed attitudes and behaviors.
Training for instructors should include information
regarding the r e l at ion shi p be tw een att itude and dietary
change and include training in use of motivational
techniques to assist clients to improve attitudes and
behaviors.

Techniques to a ssist in ma nag in g job-related

frustrations should be included for instructors,
EFNEP aides.

such as

This type of infor ma tio n/ sup por t has been

provided through inservice w o rk sh ops in Mi chigan EFNEP.
Ass ess me nt of A t t i t u d e s : Future Re search Investigations
Several investigators

(Appendix A)

have reported that

attitudes are one of the factors that influence diet ar y
change.

But quan tit at ive a ttitude as se ssm ent data in EFN EP
118

which has resulted in insufficient p r o gra m ev alu ati on data
needed to determine if this program is improving attitudes
of participants.

E va l u a t i o n of major objectives

in

Ex tension programs is increasingly important as federal
a p propriation of f unding is steadily threatened.
situation demands tha t funded programs,
valid and reliable a tt itude instruments,
used in this study,

This

such as EFNEP,

use

such as the one

to col lec t specific information on a

regular basis to d e te rm ine if the pro gra m results in
positive attitude changes in partic ipa ti ng homemakers.

The

results of this study should be used to formulate
recommendations add r e s s i n g internal a c c o u n ta bil ity needs
regarding ongoing p r o gr am management d ecisions such as
training needs of instructors,

need to assess attitudes of

potential instructors and clients,

and to de termine the

need for mod ifi ca tio n in the teaching methods and
materials to promote po sitive attitude and die tar y change
of prog ram participants.

Additional resear ch is needed to

determine w hat factors influence the abi lit y of EFNEP
instructors to w o r k eff ec tiv el y with this population
(i.e., promote positive changes in attitudes and dietary
behavior).
Results of this study p rovide ev idence regarding the
effect of p r o g ram par tic ip ati on on at ti tud e improvement
related to the general cons tr uct "attitudes toward dietary
change."

It appears that there might be more than one

construct w i t h i n this general construct such as:
119

moti va tio n to change dietary behaviors,
importance of making

p rc ived

improvements in d ietary behaviors,

and pe rceived benefits of making dietary improvements.

A

factor analysis could be conducted to dete rmi ne if this is
true.

If other factors emerge,

m ean item scores could be

grouped according to these constructs w h i c h wo u l d provide
additional

information for use in prediction of diet ary

change.
Use of Multiple Factors to Predict Dietary C h a n g e ; Future
Research Investigations
Several models have been p roposed
Carlson,
1978)

1956;

Rosenberg,

(dietary change)

predict attitude change.

and Sims,

A c cor di ng to results of this

attitudes did not account for any of the varia nc e

investigator,

recall

1975;

1956;

and for die tar y scores to

in food recall change scores.

this,

Schwartz;

Brehm,

to explain the abi lit y of a ttitudes to predict

behavior

study,

1960;

(e.g.

It is the opinion of this

based on results of regression analysis from

and other studies,

that most of the va riance in food

scores is accoun ted for by factors other than

attitude.

Altho ugh a tt itude ch ange may be an important

objective in nutrit io n education programs,

these results

indicate that dietary change is influenced by various
de te rminants in a ddition to at ti tud e that were not
assessed in this study.
that locus of control
in attitude.

Results of this research indicate

is one factor that influences chang e

Other factors might include:
120

past ex per ien ce

(failure)

in implementing positive dietary changes,

poor

self esteem, perce iv ed importance and benefits of making
dietary changes,
action).

and normative beliefs

This author recommends inclusion of items to

evaluate these factors
Heider
(1975),

(Theory of reasoned

in future research.

(1958), Bern (1970), and Schafer and Yetley

suggested that tension

(cognitive dissonance)

is

produced when stability related to existing attitudes and
behaviors are disturbed.

It is likely that homemakers'

resistance to change attitudes or die tar y behaviors is to
avoid conflic t bet wee n existing attitudes and behaviors
and those promoted by the instructor or program materials.
Although subjects might agree with attitude item
statements regarding the importance and perceived benefits
of changing,

r elu ctance to adapt new behaviors may account

for the lack of a positive correl at ion between attitude
and food recall change scores.
Expecting signi fi can t positive attitude and die tar y
change might not be realistic,

esp ec ial ly if program

pa rticipation is short and if the progr am does not train
instructors in use of methods for promotion of attitude
and dietary change.
transportation,

Immediate problems

famil y interactions,

(e.g., housing,

and healthcare)

often

demand the immediate attent ion of p rogram participants
with limited incomes;
priority.

die ta ry change may not be a

It is the opinion of this author that whet her

learners improve attitude and food recall scores after
121

participating in nu tr i t i o n education programs,
EFNEP,

such as

depends on the level of stability in their

environment at the time of participation.
Valid and reliable assessments to measure factors
such as cognitive dissonance,

self esteem, and normative

beliefs should be d e v el ope d so that a more complete
investigation of factors that influence change in
attitudes, dietary intake,
conducted.

and d i e ta ry behaviors can be

Results from such an investigation could

provide information that could be used,
to results from this study,

in addition

to further understand the

relationship bet wee n attitudes toward d ietary change and
dietary improvements.
In summary,

factors w h i c h were included in this

research to predict die tar y change were: pretest attitude
of the instructor,

locus of control of the homemakers

(pretest and change s c o r e s ) , attitude of the homemakers
(pretest and change s c o r e s ) , years of experience of the
instructor,

and prete st food recall

scores.

Future

investigations should evaluate the ability of additional
factors to predict diet ary change such as: normative
beliefs,

self esteem,

motivation,

benefits of making improvements

perceived importance and

in the diet.

Instruments

used to assess these factors should be valid and reliable.
Results of this study and future research should be used
in planning,

implementing,

and ev aluating effectiveness of

nutrition education programs to promote dieta ry change.

122

APPENDICES

APPE ND IX

ASSESSMENT OF THE RELATIONSHIP BETWEEN ATTITUDE AND BEHAVIORS
A SUMMARY OF LITERATURE CITED
AUTHOR

FACTORS INDICATED**

ASSESSMENT MEASURES

food preparation, ethnicity,
perceived food use, knowledge

Baird and
food use (a)
Schutz (1976)* 24-hour food recall (b)
food habit change (b)

consistency between attitudes
and behavior

Brehm (1956)*

evaluation of products
selection of a product

(a)
(b)

Brush et
al. (1986)

nutrition knowledge (k)
flexibility of attitudes
toward nutrition (a)

length of program
high pretest scores

ByrdBredbenner et
al. (1984)

caring about nutrition (a)
eating new foods (a)
nutrition affects health (a)
learn about nutrition (a)
food frequency (b)

measurement limitations,
length of program,
lack of control over food
choices

Carruth et
al. (1977)*

flexibility to change (a)
flexibility of attitudes
age
nutrition practices (b)
observed nutrition practices
requests for literature (b)
knowledge about weight loss (k)
personality questionnaire

Cosper and
Wakefield
(1975)

trying new foods
motivation (a)
food choices (b)

preferences of husband/child
personal food preferences
recommendations of health
care professionals

(a)

Daelhousen and 3-day food record (b)
Guthrie (1982) 24-hour recall (b)
nutrition during pregnancy

high pretest scores
(a)

Dalton et
al. (1986)*

food choices (a,b)
intended food choices

specificity of the measures,
influence of others, taste,
health beliefs

Davie et
al. (1973)*

powerlessness-teacher (a)
food recal1-learner (b)
job experience (b)

job experience,
measurement limitations

*
**
(a)
(b)
(k)

a positive correlation was reported between attitude and behaviors
factors accounting for results reported in studies cited
= assessment of attitudes
= assessment of food/nutrition behaviors (dietary change)
= assessment of knowledge

123

AUTHOR

ASSESSMENT MEASURES

FACTORS INDICATED

Fazio and
Zanna (1978)*

experience as a subject (b)
attitudes toward research
participation (a)

direct exp

ience

24-hour beverage recall (b)
specificity of the
Guiry and
B iso gni (1986)* beverage frequency list (b)
measurement instrument
limiting coffee consumption(a)
caffeine consumption during
pregnancy (a)
knowledge about caffeine (k)
Hollis et
al. (1986)*

diet habit survey (b)
powerlessness (a)
cardiovascular risk
medical symptoms

age, powerlessness
medical risk
cardiovascular risk

Kaplowitz and
Olson (1983)

breastfeeding

small sample size
non-personal program

Kok et al.
(1982)*

cardiovascular disease (a)
knowledge about cvd (k)
24-hour food recall (b)

significant others, knowledge
difficulty in changing habits
taste, existing attitudes and
behaviors

Kutner et
al. (1952)

willingness to serve
a racially mixed party (a)
serving a racially mixed
party (b)

inconsistency between
attitudes and behavior

LaPiere
(1934)

providing lodging to
inconsistency between
Chinese people (b)
attitudes and behavior
attitudes toward providing
lodging to Chinese people (a)

learning
Looker and
Shannon (1984) density
nutrient
nutrient

(a,k,b)

about nutrient
(a)
dense foods (k)
density of foods

message characteristics
high pretest scores
attrition
(b)

Maiman et al.
(1979)*

weight loss experience (b)
ways to lose weight (k)
attitudes about obesity (a)

O'Connel1
et a l .(1981)*

Commitment to
teach nutrition (b),
importance of nutrition (a)
favors nutrition education(a)

Penner and
teaching nutrition (a)
Kolasa (1983)* my teaching nutrition (b)
my own nutrition (b)

124

personal experience
experience teaching nutrition
time available
high pretest scores
experience teaching nutrition

FACTORS INDICATED

AUTHOR

ASSESSMENT MEASURES

Perron and
Endres (1985)

24-hour food recall (b)
48-hour food record (b)
attitudes about nutrition
nutrition for the athlete

Peterson and
Kies (1972)*

teaching nutrition (arb)
school lunch (a)
general nutrition (k)

consistency between attitudes
and behavior

Picardi and
Porter (1976)

health concerns (a)
food choices (b)
nutrition knowledge

readiness to change
length of program

existing attitudes/behavior,
lack of control over food
(a) choices
(a)

(k)

Ramsey and
Cloyd (1979)

powerlessness (a)
use of agencies (b)
48-hour recall, mother
and child (b)

Reames (1985)

Infant feeding
knowledge
previous experience
recommendations (b)
attitudes about breastfeeding

Regan and
Fazio (1977)*

petition signing

Ries and
Schoon (1986)

nutrition and health (a,k)

length of program
high pretest scores
effect of pretesting

Rosander and
Sims (1981)*

locus of control (a)
food frequency (b)
food's effect on health

affective-based intervention

Ross

general nutrition (a)
high pretest scores
nutrition education in
(regression toward mean)
nursing school (a)
role of the nurse in nutri­
tion education (a)
role of hospital dietitians ( )
nutrition principles (k)

(1984)

Schafer
(1978)*

direct experience

(b)

dietary adequacy (b)
empty calories consumed
personal/socia1 factors

125

husband's education,
readiness to learn (change),
teacher-learner relationship

(b)
(a)

personal food preferences,
self-concept, values/beliefs,
cost, convenience, knowledge,
health of the family, media,
weight/appearance, self­
health, nutritional value

FACTORS INDICATED

AUTHOR

ASSESSMENT MEASURES

Schwartz
(1976)*

attitudes toward counseling,
counseling practices (b)
meal planning/preparation,
use of publications (b)
affective-based intervention
attitudes toward meal
planning, food preparation,
eating habits, counseling (a)

Sims

general nutrition (k)
nutrition is important (a)
vitamin supplements (a)
meal planning is
important (a)
meal preparation is
enjoyable (a)
one-day food record (b)

(1978)*

Sunseri
et al. (1984)

nutrition knowledge (k)
attitudes about nutrition
heart-healthy eating and
shopping practices (b)

Yperman and
Vermeersch
(1979)

importance of nutrition(a)
variety of foods eaten (b)
nutrition knowlege (k)
parents' food preferences
friends' food preferences

126

knowledge

readiness to change
(a) length of program
reading level
family involvement
mother's education,
family/friends preferences
parents attitudes

APPE N D I X B

M ICHIGAN STATE UNIVERSITY
tm iv ia s iY c o M K rrm

on

n

u u

i

■AST LANSING • M ICHIGAN • « N IM

in v o l v in g

HUMAN (U ^IC T S (UOHHS)
XM ADMlNmMATION M U M N C
(SIT) SSS-1 I M

April 3. 1986

ficCciVrD
*.:-8 4 i-JW

Ms. Anne Murphy-Roy
202 Wills House
Expanded Nutrition
Dear Ms. Murphy-Roy:
Subject:

Proposal Entitled, "The Measurement of Attitudes and
Behavioral Intentions of Participants and Instructors
in The Expanded Food and Nutrition Education Program"

I am pleased to advise that I concur with your evaluation that this
project is exempt from full UCRIHS review, and approval is herewith
granted for conduct of the project.
You are reminded that UCRIHS approval is valid for one calendar year. If
you plan to continue this project beyond one year, please make provisions
for obtaining appropriate UCRIHS approval prior to April 3, 1987.
Any changes in procedures involving human subjects must be reviewed by the
UCRIHS prior to initiation of the change. UCRIHS must also be notified
promptly of any problems (unexpected side effects, complaints, etc.)
involving human subjects during the course of the work.
Thank you for bringing this project to my attention.
future help, please do not hesitate to let me know.
Sincerely

Henry E. Bredeck
Chairman, UCRIHS
HEB/Jms
cc:

Dr. Jenny T. Bond

MSU i t mt A fftrm s tn * M tto n /E q u s i Q pponm uty

127

hiMaiiM

If I can be of any

COOPERATIVE.
EXTENSION
SERVICE
MICHIGAN STATE UNIVERSITY • U.S. DEPARTMENT O P AGRICULTURE * COUNTIES COOPERATING
Expanded Nutrition Program

203 WWa Ho u m
Em i LanMng. Ml 46S24
317-3534102

Consent Form: Program Participants
I, __________________________ , agree to participate in a project
to determine poeoples’ feeling about nutrition which Is being conducted by
the Expanded Food nd Nutrition Education Program (EFNEP). The purpose of this
project Is to learn more about opinions of people 1n this program regarding
changing food habits.
I understand I will be asked to respond to several questions about how
I feel about making changes 1n my life, especially related to food habits.
I understand that I am free to decide not to participate.

I may choose not to

answer any or all of these questions and I will still be able to receive
nutrition lessons for the nutrition Instructor.
I understand that my responses will be treated condldentlally and that
all Information about me will be anonymous. Hy name will not be used In any
part of the project. General results of the project will be avialable to
me at my request.

Signed___________________________________
Date_____________________________________

HSUmmmAffirmttui

128

Offnmmtyhun$mtwm

I

[COOPERATIVE.
EXTENSION
I SERVICE
MICHIGAN STATE UNIVERSITY • U.S. DEPARTMENT OF AGRICULTURE 6 COUNTIES COOPERATING
Expanded N utrition Program
202 WKia Housa
Eaatlanaing, Ml 40824
Phooa: 517-353-0102

Consent Fora: Program Instructors
I, _____________________________ , agree to participate In the project
to measure attitudes of Instructors and participants 1n the Michigan EFNEP program.
I understand that the purpose of this project Is to determine how participation
In the program affects attitude of low>1ncome homemakers and to Identify how
attitudes are related to behavior change of participants.
As a participant In this project I realize that I will be asked to fill out
a written survey at the beginning and end of the project.
I understand that I am free to decide not to participate and 1f I choose not
to, It will have no Influence on p*y employment status with the Cooperative
Extension Service. I may also decline to answer any questions I find unacceptable.
I understand that my responses and all Information about me will be treated 1n
strict confidence and that 1 will remain anonymous (all Information will be
coded by number). General results of the study will be made available to me
at my request.
Signed ______________________________________
Date________________________________________

k tS U t i m

129

t A c d w / f f w r f Oppo n mm r y M i n t ages—

APPENDIX

DEMOG RA PHI C INFORMATION
EFNEP Unit Report Summary
(August 1986 - March 1987)
HOMEMAKERS:
PERCENT OF HOMEMAKERS

FACTOR
Average length of particip at ion
0-5 months
6-11 months
12-18 months
>18 months

64.02
29.49
5. 10
0. 29

Sex
Female
Male

97.71
2.29

Racial/Ethnic
White
Black
Hispanic
Indian
Asian/Pacific

52.74
38. 20
6.83
1.17
1.08

Islander

Type of Instruction
Individual
Group
Both

62.44
14.21
23.35

Place of Residence
Farms
Towns <10,000
10,000-15,000
suburbs
>50,000

1.10
11.66
29.59
13.09
44.55

130

PERCENT OF HOMEMAKERS

FACTOR
Number of Children

10.31
54.28
31.32
4.09

0
1-2
3-5
>5
Monthly Income

(?)

<438
439-588
589-738
739-888
889-1038
1039-1188
1189-1338
>1338

47.40
19.66
14.07
7.04
5.61
2.08
1.73
2.41

PROGRAM AIDES:
PERCENT OF AIDES

FACTOR
Sex
Female
Male

97.96
2.04

Race
White
Black
Hispanic
Indian
Asian/Pacific

41.84
44.90
10.20
3.06
0.0

Islander

131

APPE N D I X

PROGRAM DE SCRIPTION
The first research objective was to determine the effect
of program pa rticipation on attiutde toward dietary
change. The homemaker "program" is co nsidered to be the
instruction w hi ch occurred between pre- and posttest
measurements.
For instructors, the "program" consisted of
all information related to attitude change whi ch was
provided either through inservices or in written materials
at the initiation and throughout the study.
Instructor Program:
1. Inservice Workshop:

July,

1986

Ex planation of the purpose of the project:
-attitude improvement is an objective of EFNEP
-attitude change has not been assessed to date
-the effect of aides' attitudes on attitude and food
recall change have not been assessed to date
Obtaining informed consent

(Appendix B)

Assessment of instructors'

pretest attitudes

Training for data collection. Handout, provided for data
collection procedures ("Attitude Survey" p. 131)
Presentation of information related to the relationship
between attitude and dietary change
-effect of e xisting

(negative)

attitudes of homemakers

-attitude formation

(direct vs indirect experience)

-effect of the learning envir on men t on attitude change
-effect of cost, convenience, culture,
preferences on food selection

and taste

-effect of locus of control on behavior
-effect of instructors' attitudes on attitude and
dietary change of learners
2. Collection of Data
3.

(A u g u s t - S e p t e m b e r , 1986)

"Effective Aide Techniques"
September 1986, p. 132)

132

(Newsletter article,

4. A sse ssment of Instructors'

Attitudes

(posttest)

Homemaker Program:
1. First visit to the homemaker

(August-September 1986)

Explanation of the purpose of the survey
Elicitation of informed consent

(Appendix B)

Admi nistration of attitude, locus of control, and food
recall pretests (Appendices N and P)
2. Participation in individual or small group lessons
Number of lessons received depended on results of
competency-based assessment instrument
3. Posttest attitude, locus of control, and food recall
instruments administered d uring last visit
(Appendices N and P)

133

ATTITUDE SURVEY
Task
1. Attend inservice to receive trianing in
positive attitude change of homemakers.

Data Collection Procedures
To Be Completed By
la. EHE's
lb. program aides

When
1. July 29

2. Take survey.

2.

3. Begin using survey with aT[ homemakers that enroll
in August and September (pre-test).

3. program aides: Muskegon, Berrien, 3. August 1 Genesee, Ingham, and Kent counties
September 30, 1986

4. Mail all completed computer forms, a copy of the
Food Recall Form, and Family Record to Anne Roy
at the State Office

4a. EHE's of counties in #3
4. September 30, 1986
4b. EHE's of ocunties in #6; send
recall and Family Record form only

5. Send list of all homemakers who need to take the
post-test.

5.

Anne

5. October

6. Use survey with all homemakers that took pre-test
when they finish lessons (post-test).

6.

program aides: Muskegon, Berrien,
Genesee, Ingham, and Macomb

6. last lesson
(deadline:

program aides

counties
7. Use survey with all homemakers that began enroll­
ment in EFNEP during August-September when they
finish lessons (post-test only).

7.

8. Analyze results

8. Anne

NOTE:

program aides: Oakland, Wayne,
Kalamazoo, Kent, and Saginaw
counties

2. July 29

March 30, 1986)
7. last lesson

8. April-June, 1987

If a homemaker takes the pre-test but does not complete the program and takes the post-test, write the date
of withdrawl and the reason on the ocmputer form and recall form.

TEACHING
EFFECTIVE AIDE TECHNIQUES
•

Use • n o n - a u t h o r 1 1 a r 1 a n
approach

a

Smile!

Have

a good

sense

a

Avoid

a

A c c e p t the
they are

of

'know

It

all*

approach

homemakers

where

humor
a

a

a

a

Be

'human*

E x p l a i n t h a t t h e p r o g r a m Is
n o t p a r t of S o c i a l S e r v i c e s

a

Be
of

respectful
homemakers

Use 2 - way a p p r o a c h
from homemakers)

a

Don't

become

a

Don't

rush

a

Become Involved
lesson; Involve

a

Relate
saving

Praise

positive

a

Answer

honestly

a

Be

open

overly

through

minded

--Ann
EFNEP

a .

Graduate

XX

*'T>
K

k

TRIVIA ANSWERS

Be rr ien , Ge ne s e e , Ingham,
K a l a m a z o o , Kent, Nacomb,
Muskegon, Oakland, Saginaw,
and Wayne.
Eating

the

lesson

In t h e
the h o m e m a k e r

all l e s s o n
money

Roy,

XX

XX

2.

friendly

changes

*'T>

1.

considerate

(learn

E n c o u r a g e b e h a v i o r c h a n g e by
u s i n g food p r e p a r a t i o n that
reinforces lesson material

a

and

Right

Is

Basic

135

2

3.

4,300

4.

BOS

5.

160

families

material

to

Assistant,

APPENDIX

PRELIMINARY RESEARCH INTERVIEW SCRIPT AND RESPONSES

"The purpose of this Interview le to obteln Information about the
attitudes and behaviors of EFNEP homemakers. I will ask several
questions. Please answer the questions with the first response that cornea
to mind. The information will be used to design a survey which will be
used to learn more about the attitudes and behaviors of the people who
participate in EFNEP.*
QUESTION TOPIC:
la.

POSITIVE ATTITUDES

How would you describe a homemaker that you consider to have positive
attitudes about foods and nutrition?
- they are there when you make a home visit (not a "no show*)

lb.

- they

are 'ready* for the leaaon

- they

ask questions, act Interested

- they

are usually involved with their

- they

are confident and have a goodself-image in general

children

What do they (homemakers) say to indicate their positive feelings
about foods and nutrition?
- they ask the opinion of the aide regarding their food practices
- they want to know how to manage their money ao they can be less
dependent on social services

lc.

What do they (homemakers) do to Indicate these positive feelings
•about foods and nutrition?
- recommend the program to their friends/neighbors
- are open to new information, receptive
- have good eye contact during a lesson

Id.

What do homemakers say or do that indicates they think nutrition la
Important ?
- they ask questions about 'how to feed their kids right*

136

QUESTIOH TOPIC:

2a.

WEGATIVE ATTITUDES

How would you describe a hoaeaaker that haa negative attltudaa about
foods and nutrition?
- thay have a "huffy" attltuda
- thay don't want tha program but ara raqulrad to participate
- thay act Ilka tha laaaona ara "owed" to thaw
- thay ara defensive about thalr eating habits
- referrals from food banks and "cheese lines" are usually the least
motivated hoaeaakers

2b.

What do thay say to Indicate thalr negative feelings about food and
nutrition?
- tha nutrition lessons are a waste of tlae

2c. Vhat do they do to Indicate their negative feelings about foods and
nutrition?
- thay ara not there for scheduled lessons and do not call to re­
schedule
- they don't act Interested In the lessons (watch T.T.)
- they are not prepared for the lesson
QUESTIOH TOPIC:

POSITIVE BEHAVIORS

3a. Think about a hoaeaaker that has positive or good nutritional
practices. What does chat person say or do that denonstratea that
they have good nutritional habits or behaviors?
- they provide feedback regarding improved nutrltloal practices from
lesson to lesson
- they wake coaaents about their own food habits while they take the
knowledge quls
- they use recipes provided by prograa aides
- they are able to "control" the eating habits of their children
- they discuss food preferences

137

3b.

What do hoaeaakera lay or do that ladlcataa that thay think laprovlng
food praparatlon akllla la laportantT
— thay want to know how thay can change food preparation aethoda ao
that food ataaps will laat through tha aonth

QUESTION TOPIC: NEGATIVE BEHAVIORS
4.

Think about a hoaeaaker that haa poor nutritional practlcee. What
doaa Chat peraon aay or do thao deaonatratea that they have poor
nutritional habita or behavioral
- they are not lntereated in trying new foode/reclpea
- they don't aak the alde'a opinion about their food habita

QUESTION TOPIC:
S.

POOD SELECTION

What do hoaeaakera aay or do that lndlcatea that they think aelectlng
and aervlng nutrltloua fooda la laportantT
- they think that their chlldran'e behavior la related to fooda
eaten

QUESTION TOPIC:
6.

FOOD SAPETT

What do hoaeaakera aay or do that lndlcatea that they think food
aafety la laportantT
- they like the inforaatlon provided about food aafety if it la not
‘peraonal*
- they are "open" to the Inforaatlon In thla area becauae they think
it will help then to aave aoney by avoiding waate

QUESTION TOPIC:
7a.

PROGRAM PARTICIPATION

What do hoaeaakera aay or do that indicate! that they think learning
about nutrition la laportantT
- they aak queatlona when they are taking the qulzzea
— they aak "An I doing right when I do..."

7b.

What do hoaeaakera aay or do that lndlcatea that they think partici­
pating in EPNEP reaults in laproveaent of dietary habita?
- they aay they "eat better" and aave aoney
- they are aore open to trying new fooda aa the leaaona progreaa

138

QUESTION TOPIC:

8.

POOD BUDGETING

What do heaiBiktrt i«y or do that Indicates that thay think learning
how to manage food dollars le important?
- moat of them want budgeting information more than any of the other
information wa teach
- they aay that food atampa are lasting longer now than before they
began program participation
- sonatinas you can see that they have more food available

QUESTION TOPIC:
9.

MOTIVATION

What motivational techniques have you tried that have improved the
attitudes of homemakers you have worked with?
- use a non-authoritarian approach
- smile
- have a good sense of humor
- explain that the program is not part of social services so they
will not feel threatened by our presence in the home
- teach lessons using the 2-way approach (be open to learn as well as
teach)
- bring food items that the homemaker has been hesitant to try.
Trying will help them change their attitude towards the food,
whereas talking about it does not
- praise positive changes they make
- always answer honestly
- be open minded
- relate to the level the homemaker is at
- be "human"
- respect homemakers
- don't get "over-friendly''
- don't rush through the lesson just to get done
- get "involved" in the lesson
- have a positive attitude about food/nutrltlon
- offer support and encouragement

139

QUESTIOH TOPICt

10.

SUMMARY

How would you deaerlba tha difference In hoacatkari
motivated to laarn naw Inforaatlon?

that ara/are not

- Hoaeaakera can be divided Into two groupe: thoaa that ara really
Interested In aavlng aoney and improving food habita and thoaa that
faal tha gowernaent owaa thaa food etaape, commodity fooda, and
EFHEF leaaona.

140

APPE ND IX

TEST ITEMS FOE EEVIEW
Fltii* as* eh* attached fora to evaluate th* ability of th*s* lt*a* to
****** attltud** regarding 'dietary change.”
1.

I feel ready

to learn tor* about food* and nutrition.

2.

I have a lot

of question* about how to ”eat right.”

3.

I an Interested In learning new ways of doing thing*.

4.

I would like

5.

I consider ayself to be a confident person.

6.

I feel good about how ay life Is going.

7.

I wish I had nore confidence.

8.

Many areas of ay life ara ”out of control.”

9.

I do not feel in control of ay

to learn lore about nutrition.

life.

10. Learning how to aenage ay "food dollars” will help le to have nore
control In ay life.
11. I cook about the sane way ay aother did.
12. I consider ayself to be open-nlnded.
13. I generally like to hear new Inforaatlon.
14. I an receptive to new Inforaatlon.
15. I don't like to hear Inforaatlon that la new to ae.
16. I really don't like to change ay habita.

141

P(|« 2
17.

Beeiuit of ay low lncoai, I feel trapped In a bad altuatlon.

18.

If I try hard, 1 know X can Improve ay life.

19.

Learning new Inforaatlon la a waate of time.

20.

Old habita are hard to change.

21.

I don't have auch trouble "changing ay ways."

22.

I want to know If the way X aa feeding ay faally la OK.

23.

X like to taite new

fooda.

24.

I don't like to try

new reclpee.

25.

I have Improved the

way X eat Itfflng the paat year.

26.

X aa eating about the eame way now aa X did a year ago.

27.

The way I eat could use a lot of lnprovenent.

28.

I like to try foods

29.

I consider ayself open to new Ideas.

30.

There la a lot for ae to learn about nutrition.

31.

I do not feel like I aa In control of choosing the foods ay faally
eats •

32.

How I eat doesn't really affect ay health.

33.

X soaetlaes feel like It's too hard to feed ay faally nutritious
aeals.

34.

I enjoy the responsibility of selecting foods for ay faally.

I have never eaten.

142

F«|t 3

35.

I like to "toko eharti* of situations.

36.

I injor eating many types of food.

37.

I would raally Ilka to wake laproveaents In the way I eat.

38.

I've been eating the aame way for yeara.

39.

I wlah X knew aore about nutrition.

40.

I like to cook.

41.

I enjoy planning aenua.

42.

I like trying new fooda.

43.

My parent(a) were not very Intereated In aervlng nutrltioue aeala.

44.

I wlah ay faally waa aore concerned with nutrition.

45.

I like to aake ■y own declaions about wh a t I eat , but often I eat
what everyone elae la eating.

46.

For the aoat part, ay huaband/boyfrlend decldea what the faally
will eat.

47.

The food I eat haa nothing to do with the way I feel.

48.

I feel guilty when I eat "Junk” food,

49.

I think I know a lot about nutrition

50.

I don't feel very aotlvated to ahop for and prepare nutrltloua
aeala.

51.

I have enough Inforaatlon to aake good food

143

Page i

32.

1 consider ayself to b«

53.

Making decisions Is easy for ae.

5A.

The way I eat la too strong of a habit to change easily.

55.

I often buy certain foods ay faally wants, even If I don't think they
are nutritious.

56.

My friends opinions about ae are aore laportant than ay own.

57.

If snacks are available, I usually eat thea.

58.

I don't have such will power.

59.

When I "eat right" I am proud of ayself.

60.

I overeat aore often than I should.

61.

Time I have spent learning about nutrition has been well worth It.

62.

I have forgotten aost nutrition Information I have learned.

63.

Nutrition Is a boring topic to ae.

6A.

I consider myself to be someone who can manage a food budget well.

65.

I aa eager to learn new ways to Improve ay food habits.

a

parson who can aake good decisions.

66. I tend to ask questions about new information I hear or read.
67.

I aa not very interested in nutrition.

144

Page

68.

1

am proud

o 9.

1

easily

70.

I e n joy e a t i n g

fruits.

71 .

I e n joy e a t i n g

vegetables.

72 .

I

enjoy b e i n g

73.

I

enjoy d r i n k i n g

7A .

Someday

75.

I eat too many:
(check all that

of m y B e l f

when

forg e t w h a t

I l e ar n to do s o m ething

better

I learn.

active.

I hope

milk

I won't

or eating cheese.

need

food

76*
apply)

stamps.

I eat too few:
(check all that

fr u i t s

appl y )

fruits

vegetables

______ ve g e t a b l e s

meats

______ mea t s

milk/milk

products

______ m i l k / m i l k

products

breads/cereals

______ b r e a d s / c e r e a l s

sweets

______ n u t r i t i o u s

salty

foods

about

77 .

I often

talk

78.

I'm not

convinced

that

what

I eat

79.

Teaching

children

good

food

habits

80.

Eating

a variety

nutrition

of

foods

145

to my

foods

friends.

a f f e c t s my

is

is the key

health.

Important.

to good

nutrition.

AP PENDIX

S p e c i f i c a t i o n s for

Development of the L i k e r t Scale

The f ol l o w i n g steps should be i ncl ude d in development of a
L i k e r t scale (Likert, 1974 and Green, 1978)
1. L is t a large n u mbe r (75-100) of s tat ements concerning
the a t t i t u d e to be assesed.
L i k e r t (1974) states that
items shou ld be such that persons w i t h diffe ren t points of
v i e w r ega r d i n g a particular attitude, will respond
differently.
If people with d i f f e r e n t attitudes respond in
the same way, the item is uns atisfactory.
Items should
r e p res en t fe eli ng s or opinions, not facts.
Persons with
di f f e r e n t a t tit ud es may respond the same to factual
statements.
Ea c h item should be clear, concise, and
s t rai ght forward, w i t h no d o u b l e - b a r r e l e d statements or
d ou ble n eg a t i v e s w i t h i n the item.
2. C l a s s i f y e ach item as "favorable" or "unfavorable" w ith
r e gard to the att i t u d e construct.
It is desirable to have
ab o u t half of the items classified at ea c h end of the
c o n t i n u u m to avoid response set.
3. C o n d u c t a pilot test with a sub g r o u p of the population
in w h i c h the res p o n d e n t checks one of the following
descriptors:
a.
b.
c.
d.
e.

s tro n g l y agree
agree
undecided
d isa gr ee
s t r o n g l y disag re e

4. A ss ig n a numerical weight to each reponse

(e.g.

1-5).

5. C a l c u l a t e the total attitude score of the individual by
s umming the wei ght s associated with ea c h response.
6. C a l c u l a t e the dis cr imination index for each item. Select
items w h i c h d i s c r i m i n a t e between high and low scorers,
(greatest di f f e r e n c e in means b e t w e e n high and low scorers.
7. C a l c u l a t e reliability.
8. C o r r e l a t e the mean item scores a g a i n s t the total mean to
see if the numerical values are p r o p e r l y assigned
(item
d i f f e r e n t i a t i o n ) . If the c orr el a t i o n between the
individual item and the total item m ean is negative, the
w e i g ht s a s s i g n e d to responses should be reversed.
If the
co r r e l a t i o n is zero or very low, the statements may
repre se nt a different attitude construct, or may be factual
rather than a t t i t u d i n a l . ._

A P PE N D I X

ATTITUDE QUESTIONNAIRE REVIEW FORK
Item Number
Tes
1.

Is the Item clearly written
for the EFNEP audience?

2.

Is the Item free from
Irrelevant material?

3.

Are the gramnar and
punctuation correct?

4.

Is the Item worded In such
a way that the respondent
will feel compelled to
select the strongly agree
or strongly disagree
response (I.e. Is the
item written so that
there is clearly one
socially acceptable
response)?

5.

Does the item adequately
represent the purpose of
the Instrument (I.e. to
assess attitudes toward
dietary behavioral
changes)?

6.

Is the item appropriate
for the EFNEP homemaker?

Suggested Revisions:

Additional Coements:

147

______

Questionable
______

No
___

A P PENDIX

Aide ID Number:
EFNEP SURVEY
Directions for the Aide:
Before you give this survey to the homemakers, please Indicate whether you
think this homemaker has a
positive or a ___ non-positive attitude about
changing food behaviors (habits). Read all Items to the homemaker (you may
have to read some of the items twice). Record the responses on this form.
PART A: Directions (read to the homemaker):
I will read several statements about foods, nutrition, and changing habits.
These are opinion statements; there are no right or wrong answers. We need
to know how you feel about these topics so we can improve this program. There
are several answer choices.
Here 1s a card with the possible answers on 1t.
The neutral choice means you neither agree, nor disagree. Please choose
the
answer that 1s closest to the way you feel. Are you willing toparticipate in
this survey? (If yes, go on.
Ifno, go on with the lesson.)

1.

If I try hard, I know I can improve my life.

I I very strongly agree
I
2.

I disagree

1 1 strongly disagree

1 I disagree

I

I

I I neutral
I very strongly disagree

I

I strongly agree

I I strongly disagree

I

I I neutral

I very strongly disagree

I have a lot of questions about how to "eat right."
strongly agree

I I disagree

1 Istrongly agree

I I strongly disagree

I

| | neutral
I very strongly disagree

I would like to learn more about nutrition.

I I very

strongly agree

I I disagree
6.

1 I strongly agree

1 I strongly disagree

I very strongly agree

I I very

5.

f I very strongly disagree

I consider myself to be open-minded.

I I disagree
4.

I I neutral

I cook about the same way my mother did.

1 I very strongly agree

3.

I I strongly agree

1 1strongly agree

I I strongly disagree

I

1 Ineutral
I very strongly disagree

I think I am a confident person.
I I very

strongly agree

I 1 disagree

I Istrongly agree

I I strongly disagree

148

|

[ 1neutral
| very strongly disagree

Page 2
EFNEP Survey
7.

I feel good about how my
I 1 very strongly agree
| | disagree

8.

life is going.
I Istrongly agree

strongly disagree

I I neutral
very strongly disagree

Many areas of my life are "out of control."
I | very strongly agree
| 1 disagree

9.

d J strongly agree

strongly disagree

t ZU neutral

[ZU very strongly disagree

I generally like to learn new information.
I

| very strongly agree

| | disagree
10.

d j strongly agree

[^] strongly disagree

I

neutral

I very strongly disagree

I would really like to make improvements 1n the way I eat.
| |very

strongly agree

| |disagree
11.

I I strongly agree

I I strongly disagree

I

I

I neutral

Ivery strongly disagree

I feel guilty when I eat "junk" food.
I |very

strongly agree

I |disagree
12.

I L strongly agree

I I strongly disagree

1

Q ] neutral

1very strongly disagree

I think I know a lot about nutrition.
I |very

strongly agree

I |disagree
13.

I 1 strongly agree

1 1 strongly disagree

1

I

| neutral

Ivery strongly disagree

Generally, I make good decisions.
j |very

strongly agree

I 1 disagree
14.

I 1 strongly disagree

1

1

I neutral

1very strongly disagree

I feel I am not in control of my life.
I j very

strongly agree

I | disagree
15.

I 1 strongly agree

I I stcongly agree

1 I strongly disagree

I

1

I neutral

Ivery strongly disagree

It's important to eat nutritious food.
I 1very

strongly agree

I I disagree

I [ strongly agree

1 I strongly disagree
149

I

I

[ neutral

1very strongly disagree

Page 3
EFNEP Survey
16.
I

I've been eating the same way for years.
I very strongly agree

| | disagree

strongly agree

CZ] strongly disagree

1

1 neutral

I I very strongly disagree

17. I wish I knew more about nutrition.
| 1very

strongly agree

I | disagree

I | strongly agree

strongly disagree

1

I neutral

d ] very strongly disagree

18. I like to cook.
I | very

strongly agree

I I disagree
19. I wish I had
I |very

20.

I I strongly agree

strongly disagree

I

I neutral

d ) very strongly disagree

1 1 strongly agree

I j strongly disagree

strongly agree

1 Idisagree

I

| neutral

I I very strongly disagree

I I strongly agree

I I strongly disagree

1

1 neutral

1 1 very strongly disagree

I like to taste new foods.

1 Ivery

strongly agree

1 [ disagree

1 I strongly agree

I | strongly disagree

□

neutral

I | very strongly disagree

I have little trouble "changing my ways."

1 |very

strongly agree

I |disagree
24.

I very strongly disagree

I think I am open to new ideas.

I j.very

23.

I neutral

more confidence.

strongly agree

| Idisagree

22.

I

1

I easily forget what I learn.

1 |very

21.

Qj] strongly disagree

strongly agree

disagree

1 I strongly agree

1 1 strongly agree

1 [ strongly disagree

I

1 neutral

I I very strongly disagree

Old habits are hard to change.

I |very

strongly agree

I | disagree

I I strongly agree

I I strongly disagree
150

I

| neutral

| 1 very strongly disagree

Page 4
EFNEP Survey
25.

Making decisions is easy for me.

I | very

strongly agree

I | disagree
26.

\

strongly agree

I I strongly disagree

I

I )neutral
1 very strongly disagree

Because of my low income, I fee1 trapped in a bad situation.

I I very strongly agree
I I disagree
27.

I

I I strongly

I I strongly disagree

agree
I

I Ineutral
I very strongly disagree

I often talk about nutrition to my family or friends.

I I very

strongly agree

1 I disagree

I I strongly

| | strongly disagree

agree
I

I Ineutral
I very strongly disagree

28. What I eat affects my health.
I I very strongly agree
I I disagree
29.

strongly agree

I | disagree

I I very strongly disagree

agree

I

\

neutral

I I very strongly disagree

I I strongly

I I strongly disagree

agree
I

I I neutral

1 very strongly disagree

I am disinterested 1n nutrition.

| | very

strongly agree

| | disagree

I I strongly

I I strongly disagree

agree
I

I I neutral

I very strongly disagree

Learning how to manage my grocery money will help me to have more control
in my life.

| | very strongly agree
| | disagree
33.

1 I strongly

I I strongly disagree

strongly agree

| | disagree

32.

I neutral

Eating many different types of foods is the key to good nutrition.

I | very

31.

| | strongly disagree

I

Teaching children good food habits is important.

I | very

30.

I I strongly agree

I I strongly agree

1 I strongly disagree

t

I I neutral
I very strongly disagree

I think I ’m good at managing a food budget.

I 1 very strongly agree
| | disagree

I I strongly agree

I 1 strongly disagree
151

I

I I neutral
I very strongly disagree

Page 5
EFNEP Survey
34.

I am eager to learn new ways to improve my food habits.

| | very strongly
| | disagree
35.

I

1 I neutral

I very strongly disagree

I I strongly

I I strongly disagree

agree
I

I I neutral

I very strongly disagree

I often buy certain foods my family wants, even if I don't think they are
nutritious.

I | very

strongly agree

| | disagree

I | strongly

I j strongly disagree

agree
I

I I neutral

I very strongly disagree

My friends' opinions about me are more important than my own.

I I very

strongly agree

I | disagree

I I strongly

| I strongly disagree

agree
I

I I neutral

I very strongly disagree

I have very little will power.

I 1 very strongly
I | disagree
40.

neutral

I very strongly disagree

I I strongly agree

) I strongly disagree

strongly agree

| | disagree

39.

1

Q]

The way that I eat is a habit that is too strong to change easily.

I 1 very

38.

I strongly agree

I I strongly disagree

strongly agree

| 1 disagree

37.

1

I like to "take charge" of situations.

| | very

36.

agree

agree

I

[ strongly agree

| I strongly disagree

I

I

Ineutral

I very strongly disagree

Nutrition is a boring topic to me.

I I very strongly agree
I I disagree

I I strongly agree

1 I strongly disagree

[^]

I I neutral
very strongly disagree

41. Some people consider me tobe a stubborn person.
I I very strongly
I
42.

1 disagree

agree
I

I

I strongly agree

I strongly disagree

I

|neutral

very strongly disagree

I don't feel very motivated to prepare nutritious meals.

1 | very strongly agree
I 1 disagree

| 1 strongly agree

1 I strongly disagree

152

I

I | neutral

| very strongly disagree

Page 6
EFNEP Survey
43.
1

I wish my family would be more concerned with nutrition.
| very strongly agree

| | disagree
44.

I

enjoy

I |very

f

f

planning

I I strongly

1 strongly disagree

|

| | strongly

I strongly disagree

f~ | very strongly disagree

agree
|

I

Ineutral

I very strongly disagree

agree
I

I

Ineutral

I very strongly disagree

agree
f

I

Ineutral

| very strongly disagree

menus.

strongly agree

I I disagree

1 I strongly

| strongly disagree

agree
I

I

Ineutral

I very strongly disagree

1 like trying new foods.

1 1 very

strongly agree

I I disagree

f

1 I strongly agree

I strongly disagree

I

Ineutral

[ ~~1 very strongly disagree

My parent(s) were not very interested in nutrition.

I I very

strongly agree

I 1 disagree

I

j ~| strongly agree

I strongly disagree

f

|

|neutral

I very strongly disagree

I like to make my own decisions about what I eat.

I I very

strongly agree

I [disagree
51.

I 1 strongly

1 strongly disagree

strongly agree

I | disagree

50.

1 neutral

Time I have spent learning about nutrition has been well worth it.

1 1very

49.

f

strongly agree

| [ disagree

48.

I

I overeat more often than 1 should.

f |very

47.

I strongly disagree

strongly agree

| |disagree

46.

I strongly agree

I have enough information to make good food choices.

| | very

45.

f

1

f

1 I strongly agree

I strongly disagree

I

I

Ineutral

I very strongly disagree

For the most part, someone other than myself decides what the family
will eat.

1 I very

strongly agree

I 1 disagree

f

I 1 strongly agree

i strongly disagree
153

f

|

|neutral

" ! very strongly disagree

Page 7
EFNEP Survey
52. I dislike trying new recipes.
I | very strongly
1
53.

agree

| disagree

□

I

\ strongly

strongly disagree

| | disagree

I

I

I 1 very strongly

agree

| disagree

□

1

□

neutral

1 very strongly disagree

£ Z ) strongly agree

□

neutral

d j very strongly disagree

I I strongly agree

strongly disagree

strongly agree

1 disagree

□

I 1 neutral
very strongly disagree

Q

1 I strongly agree

strongly disagree

1 1 neutral
very strongly disagree

Sometimes i t ’s too hard to feed my family nutritious meals.

j I very

strongly agree

1 | disagree

£ZI

I I strongly agree

strongly disagree

Q

I 1 neutral
very strongly disagree

There is a lot for me to learn about nutrition.

1 1 very

strongly agree

I 1 disagree
60.

I very strongly disagree

Eating habits are related to health.

1 | very

59.

1

neutral

1 am eating no better now than 1 did a year ago.

] | disagree

58.

agree

never eaten.

strongly disagree

1 | very strongly agree

1

very strongly disagree

d ) strongly agree

I strongly disagree

55. I like to try foods I have

57.

neutral

The way I eat could use a lot of Improvement.

1 1 disagree

56.

strongly

I strongly disagree

1 I very strongly agree

|

□

dJ

1 have improved the way I eat during the past year.

I | very strongly agree

54.

agree

f

I I strongly agree

I strongly disagree

I

1 I neutral

I very strongly disagree

1 enjoy the responsibility of selecting foods for my family.

I "1very

strongly agree

1 t disagree

1

I I strongly agree

I strongly disagree
154

1

1 I neutral
I very strongly disagree

Page 8
EFNEP Survey

PART B: Directions (read to the homemaker):
The next set of questions relates to your food choices! Please answer according to
how you usually eat.
61.

I like to eat fruits.

I Ivery

strongly agree

I I disagree
62.

1 I strongly

strongly disagree

agree
I

(

1neutral

| very strongly disagree

I like to eat vegetables.

I Ivery

strongly agree

I I disagree

I ( strongly agree

strongly disagree

j

1

|neutral

[ very strongly disagree

63. I like to eat neat.
I I very
1

64.

1 disagree

65.

Q ] strongly disagree

strongly agree

I disagree

agree

I

1neutral

[ I very strongly disagree

1 I strongly

agree

strongly disagree

I

Ineutral

verystronglydisagree

I like to eat cheese or drink nllk.

I [ very strongly agree
I I disagree

66.

! I strongly

I like to eat bread and other grain foods.

1 I very

I

strongly agree

I

I I strongly agree

| strongly disagree

[^]

I

I neutral

very strongly disagree

I like to eat many different types of foods.

1 1 very strongly agree
I I disagree

f

I I strongly agree

I strongly disagree

1

I neutral

I I very strongly disagree

67. I eat too many: (check all that apply)
1 eat too few: (check all that apply)
_ _ _ _ _ fruits
_ _ _ _ _ fruits
_ _ _ _ _ vegetables
_ _ _ _ _ vegetables
_ _ _ _ _ meats
_ _ _ _ _ meats
_ _ _ _ _ milk/milk products
_ _ _ _ _ mllk/nflk products
_ _ _ _ _ breads/cereals
_ _ _ _ _ breads/cereals
_ _ _ _ _ sweets
_ _ _ _ _ salty foods
_ _ _ _ _ nutritious foods

155

A P P ENDIX

PILOT TEST INSTRUCTIONS FOR PROGRAM AIDES
1.

You will be administering the attitude survey t o 4 o f y o u r regular homemakers
during the week of June 16, 1986. Choose 2 homemakers that you feel have a
positive attitude about changing food behaviors, and two that do not. It
does not matter how long the homemakers have been 1n the program — but 1t
Is very Important to choose 2 with positive attitudes and 2 with n o n - p o s i t i v e
attitudes. This information will help you decide:
A homemaker with positive attitudes toward changing food behavior:
• is there when you make a home visit
e Is prepared for the lesson
• acts Interested, asks questions, maintains eye contact
• asks your opinion about their food habits
• are confident about their homemaklng skills
• are open to new Information; do not act threatened or defensive
• tries new foods/redpes
A homemaker with non-positive attitudes toward changing food behavior:
•
•
•
•
o
o
o

is not home when you visit; does not call to reschedule
1s not prepared for the lesson
does not act interested in the lesson material
does not discuss her own food habits
is not confident about homemaking skills
is not open to new Information; acts threatened or defensive
does not try new foods/red pes

Think about these two descriptions, and select 2 homemakers with positive
attitudes (as described) that you will be visiting during the week of
June 16th. Select 2 homemakers that you feel have non-positive attitudes
(as described) that you are scheduled to visit.
2.

Before using this survey with the homemakers, take the survey yourself so
that:
(a) you will become familiar with 1t
(b) we can see if the test scores of Aides are higher than
those of homemakers (If the survey is a good measure
of attitude, Aides should score higher than homemakers.
We want to see If this is true.)

3.

Use the survey with the four homemakers you have selected. The survey will
take about 10 minutes. It should not replace the lesson — just use it
before or after your usual lesson. Directions are included on the survey.
Notecards with the answer choices are provided. They will help her/him to
remember the choices so that you will not have to read them after each item.
(The survey used in Muskegon county has just two answers — a or b. No
response cards will be used.) If the homemaker chooses not to participate,
proceed on with your scheduled lesson and select a different homemaker to
use the survey with.
156

4.

Read each Item once. Repeat the Item 1f they ask you to or If they have
not responded after several seconds. If they begin to talk about the
Item, repeat It a g a i n and ask them for their answer. Oo not discuss the
statement with them. Do not help them decide. If they do not give you
an answer, leave the Item blank. Do not In any way Indicate whether you
approve or disapprove of the answers they give. You will be completing
a review form on which you will be listing all Items that the homemakers
had difficulty with. Please circle or make marks directly on the survey
which will help you remember what problems the homemaker had.

5.

After comletlng the survey with 4 homemakers, complete the (pink)
Review Form. Your comments on this review form are extremely helpful in
revising the survey so that it Is appropriate for use with EFNEP home­
makers.

6.

Give the completed homemaker surveys and review forms to your EHE. All
information collected from this survey will be analyzed so that Aides
and Homemakers remain anonymous. No individual names or coirments will
be included 1n the results. THANK YOU for your time and cooperation with
this Important project. I know this pilot test 1s an extra task in
addition to your already busy schedule. It is appreciated! Results of
this pilot will be reported at the July 29th Inservice.

157

APPENDIX

REVIEW FORM:

PROGRAM AIDES

What changes need to be made regarding the format of the survey:
length _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
layout _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
size of type _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Which items did the homemakers have difficulty understanding? (list
item numbers and describe the problem)

How did homemakers feel about taking the survey (interested, bored,
apprehensive, etc.)

How did you feel about giving the survey? Were any items "touchy'1
or not appropriate for homemakers? Were you uncomfortable about
any of the procedures or item statements?

Are any of the directions, statements, or responses unclear?

Did homemakers have difficulty understanding how to choose an answer?

Are there any changes you would recommend to make this survey easier
for aides to give or for homemakers to take?

Thanks for vpur help!

APPENDIX L

Aide ID Number:
EFKEP

sunny

Directions for the Aide:
Before you give this survey to the homemakers, please Indicate whether you think this
non-pos1t1ve attitude about changing food
homemaker has a ____ positive or a
behaviors (habits). Read all Items to the homemaker (you may have to read some of
the Items twice). Record the responses on this form.
Part A: Directions (read to the homemaker):
1 will read several statements about food and nutrition. These are opinion statements;
there are no right or wrong answers. Me need to know how you feel about these topics
so we can Improve this program. There are several answer choices. Here Is a card with
the possible answers on 1t. Please choose the answer that Is closest to the way you
feel. Are you willing to participate In this survey? (If yes, continue with the_____
survey. If no, go on with the lesson.)

1.

It's hard for me to use new Ideas.

2.

Improving the way I eat is Important to me.

3.

Becoming a success 1s a matter of hard work; luck
has little to do with It.

4. I feel good about how my life 1s going.

5.

In general. 1 feel I am not In control of my life.

6.

It isn't easy to make changes In the way I eat.

7.

Other people are more 1n control of my life than I am.

8.

I've been eating the same way for years.

9.

I like to cook.

10. I easily forget what I learn.
II. Trying new recipes Is not worth the trouble.
12. I can't do much to improve my diet because of my
Income.
13. Mhen I am short of money, I feel trapped 1n a bad
situation.
1 often talk about nutrition with my family or
friends.

159

Page 2
EFNEP Survey

IS.

1 think I'm good at managing a food budget

16. 1 like to "take charge" of situations.
17. Nutrition Is a boring topic to me.
16. 1 am motivated to prepare nutritious meals.
19. 1 enjoy planning meals.
20.

1 Hke trying new foods.

21. 1 have Improved the way I eat during the past year.
22. I enjoy selecting foods for my family.
23. Many times 1 feel that 1t does not do any real good
to think about what to do.
24. A person who gets a good job Is just luck to be
at the right place at the right time.
25. Much of what happens to me 1s probably a matter of
chance or luck.
26. The things that happen to most people are outside
their own control.
27. It Isn't wise to plan too far ahead because most
things turn out to be a matter of chance anyhow.
28. When things are going well for me I usually
think of 1t as a run of good luck.
29.

1 have usually found that what Is going to happen
will happen no matter what 1 think or do about It.

30. Most of the things that have disappointed me
in my life have come because my luck ran out.
31. Success is mostly
32.

a

matter of getting good breaks.

Many times I feel that I have little Influence over
the things that happen to me.

33. Sometimes 1 feel that I don't have enough control
over the way my life Is going.

160

Pagt j
EFNEP Survey

35.

How many times per day do you eat:
one

two

three

four

five or more

vegetables
fruit or fruit Juice
soft drinks
cheese or milk
bread or other grain foods
meat, eggs, poultry, or fish
dried beans, split peas, nuts
or peanut butter
sweets (cookies, candies,
cakes, etc.)
36.

Do you take a supplement (vitamin/mineral pills):
yes

37.

no

If so, what type (I.e. vitamin C, protein, multiple, etc.):

Thank you for answering these questions.

161

He appreciate your help!

Aide ID Number:

EFXEP SURVEY

_

Directions for the Aide:
Before you give this survey to the homemakers, please Indicate whether you think
this homemaker has a
positive or a _______ non-positive attitude about changing
food behaviors (habltsT Rod all Items to the homemaker (you may have to read
some of the Items twice). Record the responses on this form.
Part A: Directions (read to the homemaker):
I will read several statements about food and nutrition. These are opinion statements:
there are no right or wrong answers. We need to know how you feel about these topics
so we can Improve this program. Choose an answer from 0 to 10. 0 ■ no agreement at
all; 10 ■ complete agreement. The more you agree with the statement the higher the
number you should give for your answer (up to 10). Five (5) represents a "middle"
level of agreement. Please choose the answer that 1s closest to the way you feel.
Are you willing to participate In this survey? (If yes, continue with the survey.
If no, go on with the lesson.)
10 ■ complete agreement
0 ■ no agreement
1.

It's hard for me to use new Ideas.

Answer:

2.

Improving the way I eat 1s Important to me.

Answer:

3.

Becoming a success Is a matter of hard work; luck has
little to do with It.

Answer:

4.

I feel good about how my life Is going.

Answer:

5.

In general, I feel I am not 1n control of my life.

Answer:

6.

It Isn't easy to make changes In the way I eat.

Answer:

7. Other people are more 1n control of my life than I am.

Answer:

8.

I've been eating the same way for years.

Answer:

9.

I like to cook.

Answer:

I easily forget what I learn.

Answer:

10.

11. Trying new recipes 1s not worth the trouble.

Answer:

12.

Answer:

I can't do much to Improve my diet because of my Income.

13. When I am short of money, I feel trapped In a bad
situation.

Answer:

14.

I often talk about nutrition with my family or friends.

Answer:

15.

I think I'm good at managing a food budget.

Answer:

162

EFNEP Survey

16.

1 like to "take charge" of situations.

Answer:

17.

Nutrition is a boring topic to me.

Answer:

16.

1 am motivated to prepare nutritious meals.

Answer:

19.

1 enjoy planning meals.

Answer:

20.

1 like trying new foods.

Answer:

21.

1 have improved the way I eat during the past year.

Answer:

22.

1 enjoy selecting foods for ny family.

Answer:

23.

Many times 1 feel that it does not do any real good to
think about what to do.

Answer:

A person who gets a good job 1s just luck to be at
the right place at the right time.

Answer:

Much of what happens to me is probably a matter of
chance or luck.

Answer:

The things that happen to most people are outside their
own control.

Answer:

It isn't wise to plan too far ahead because most things
turn out to be a matter of chance anyhow.

Answer:

When things are going well for me I usually think of
it as a run of good luck.

Answer:

1 have usually found that what is going to happen will
happen no matter what I think or do about 1t.

Answer:

Host of the things that have disappointed me in my
life have come because my luck ran out.

Answer:

31.

Success is mostly a matter of getting good breaks.

Answer:

32.

Many times I feel that 1 have little Influence over the
things that happen to me.

Answer:

Sometimes I feel that 1 don't have enough control over
the way my life Is going.

Answer:

24.
25.
26.
27.
28.
29.
30.

33.

Part B: DiretIons (read to the homemaker):
34. How much does each of the following affect whether you buy a food or not. There
are 5 choices. Use the number 1 for the answer that affects your food choices the
most. Continue to rank by using the numbers 2, 3, 4, end 5. The answer that
affects your choice the least should have the number S.
cost
taste
ease of preparation
family likes it
nutritional value
T63

Page 3
EFNEP Survey

Pert B: Directions (read to the homemaker):

34.

How much does each of the following affect whether you buy a food or not. There
are S choices. Use the number 1 for the answer that affects your food choices
the most. Continue to rank by using the numbers 2, 3, 4, and S. The answer
that affects your choice the least should have the number S.
_ _ _ _ _ _ _ _ cost_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ________ ease of preparation
'_ _ _ _ _ taste
_ _______ family Hkes 1t
_ _ _ _ _ _ _ _ nutritional value

35.

How many times per day do you eat:
vegetables- - - - - - - - - - - - - - - fruit or fruit juice
soft drinks
cheese or milk
bread or other grain f o o d s
meat, eggs, poultry, or fish
dried beans, spilt peas, nuts
or peanut butter
sweets (cookies, candies,
cake>, etc.)

36.

one
two
three
four
five or more
--------------------------------------------------------------------___________________________________
___________________________________
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ■—
___________________________________
___________________________________

Do you take a supplement (vitamin/mineral pills):
yes

37.

no

If so, what type (i.e. vitamin C, protein, multiple, etc.):

Thank vnu for answrrino thrsr ouestions.

164

Ue aonreciate vour heln!

APPENDIX

PILOT TEST: DISCRIMINATION AND DIFFICULTY INDICES
Number

Item

Difficulty
Index

Discrimination
Index

1.

It's hard for me to use new ideas

69.7

.30

2.

Improving the way I eat is
important to me

68.2

.08

3.

Becoming a succes is a matter of
hard work; luck has little to do
with it

71.2

.30

4.

I feel good about how my life is
going

61.7

.18

5.

In general, I feel I am not in
control of my life

65.2

.34

6.

It isn't easy to make changes in
the way I eat

53.0

.30

7.

Other people are more in control
of my life than I am

65.2

.35

8.

I've been eating the same way for
years

54.5

.17

9.

I like to cook

63.6

.13

10.

I easily forget what I learn

75.8

.38

11.

Trying new recipes is not worth
the trouble

80.0

.22

12.

I can't do much to improve my diet
because of my income

63.6

.42

13.

When I am short of money, I feel
trapped in a bad situation

14.9

.05

14.

I often talk about nutrition with
my family or friends

69.2

.27

15.

I think I'm good at managing a
food budget

78.5

.38

16.

I like to take charge of situations

74.3

.35

17.

Nutrition is a boring topic to me

68.2

.17

165

18.

I am motivated to prepare
nutritious meals

75.5

.17

19.

I enjoy planning meals

74.3

.25

20.

I like trying new foods

80.0

.08

21.

I have improved the way I eat
in the past year

74.3

.34

22.

1 enjoy selecting foods for my
family

80.0

.17

23.

Many times I feel that it does not
do any real good to think about
what 1 do

24.

A person who gets a good job is
just lucky

96.7

.38

25.

Much of what happens to me is
probably a matter of luck or chance

72.7

.30

26.

The things that happen to most
people are outside their control

74.3

.38

27.

It isn't wise to plan too far ahead 59.1
because most things turn out to
be a matter of chance anyhow

.18

28.

When things are going well for me
I think of it as a run of luck

69.7

.42

29.

I have usually found that what is

72.7

.34

72.7

.17

30.

Most of the things that have
disappointed me in my life have
come because my luck ran out

75.8

.43

31.

Success is mostly a matter of
getting good breaks

74.3

.50

32.

Many times I feel that I have
little influence over the things
that happen to me

75.8

-.43

33.

Sometimes I feel that I d o n ’t have
enough control over the way my
life is going

66.7

.43

166

APPENDIX N

Aide 1.0. No: _______
Aide Code No: _______
Homemaker l.D. ______
Lesson No:
____ _
Aide Experience (years}

County

EFNEP SURVEY
Directions:
This survey has three sets of questions. The first part deals with your opinions
about some general areas. The second part asks for your opinions about nutrition.
The last part relates to your usual pattern of eating. Please answer all questions
based on how much you agree or disagree with the statement. There are no right or
wrong answers since this survey 1s about feelings and opinions. All responses will
be treated confidentially.

PART A: General Questions
1. 1 feel good about how my life 1s going.
2. Success is mostly a matter of getting good breaks.
3. I like to "take charge" of situations.
4. Much of what happens to me is probably a matter
of chance or luck.
5. The things that happen to most people are outside
their own control.
6.

In general, I don't like to make changes.

PART B: Nutrition Questions
7.

Improving the way I eat 1s important to me.

8. I can't do much to Improve the way I eat
because of my Income.
9. 1 often talk about nutrition with my family
or friends.
10.

Nutrition 1s not a priority in my life.

11.' 1 am motivated to prepare nutritious meals.
12.

The way I eat now doesn't need improvement.

13.

It would benefit me to improve the way I eat.

14.

Nutrition 1s a boring topic to me.

IS.

There are not enough advantages to improving
my diet to make it worth the effort.

167

PART B: Nutrition Questions (cont.)
16. There are many things 1 am more concerned with
than Improving my eating habits.
17. How convenient a food Is to prepare affects
whether I use 1t.
18. The price of a food affects whether I buy It.

PART C: Food Habits
Write the number 1 next to the answer that 1s the most Important reason you choose
foods as you do; give a number " 2 " to the seocnd most Important reason; and use "3"
for the least Important reason.
19. How do nutritional value, taste, and cost affect your food choices? (1 ■ most
Important; 3 * least Important)
taste/flavor
____ nutritional value
____ cost
20. The Improvements I have made In the way 1 eat have been: (1 * best reason;
3 « worst reason)
to Improve my health
to save money
because of new Information I have learned
21. The biggest reason I have not made more Improvements in the way I eat is:
(1 * most important reason; 3 * least Important reason)
It's too hard to change my food habits.
1t probably won't result 1n any benefit to me.
1 am too busy with other things.
22. How many servings of each of these types of food do you usually eat every day?
1-2 times
0
per week
1 2
3
4
vegetables
I
I I
I
fruit or fruit juice
______________________________
soft drinks
______________________________
cheese/milk
______________________________
bread/pasta/grain foods
______________________________
meat, eggs, poultry, or fish
______________________________
dried beans, split peas, nuts,
peanut butter
______________________________
sweets (cookies, candies, cakes, etc.) I I
I
I I
I
23. Do you take a nutritional supplement (pill)?
yes
no
24.

If yes, what type? ____________ ;__________________________________
THANK VOU FOR YOUR PARTICIPATION!

168

AP P ENDIX O

Aide ID Number;

POttRLESSNESS SURVEY
Read to the Homemaker: We need your help 1n answering some questions that will
help us to Improve this program. I will read several statements about various
topics. They are opinions that have been collected from many different people.
There are no right or wrong answers; for each Item there Is a large number of
people that agree and disagree. Please show whether you agree or disagree by
choosing one of these responses: "completely agree" If you completely agree
with the statement. Choose "strongly agree” 1f you agree quite strongly with
the statement. Select "agree" If you are somewhat In agreement with the state­
ment. Choose "neutral" 1f you neither disagree nor agree with the statement.
Choose disagree 1f you are somewhat 1n disagreement with the statement. Choose
"strongly disagree" 1f you are in strong disagreement with the statement.
Choose "completely disagree" 1f you completely disagree with the statement.
I will not put your name on this survey. You will remain anonymous. Are you
willing to participate 1n this survey? (If yes. go on. If no. go on with
the lesson.)

1.

I think we will always have wars between
countries no matter what we do to try to
stop them.

2. If you are successful you will usually have
more good breaks than bad breaks.
3.

Many times I feel that 1t does not do any
real good to think about what to do.

4. A person who gets a good job Is just lucky
to be at the right place at the right tlaae.
S. I don't understand why other people act
toward me the way that they do.
6. Much of what happens to me 1s probably a
matter of chance or luck.
7. I feel I hove little Influence over the
way other people act.
8. It 1s very hard to figure out what the
future will be.
9. The ordinary person has very little
control over what politicians do.

169

10.

th e th in g s t h a t happen to most people are
o u ts id e t h e ir own c o n t r o l.

11.

I t i s n ' t w ise t o p la n to o f a r ahead because
most th in g s tu r n o u t to be a m a tte r o f
chance anyhow.

12.

You c a n 't r e a l l y t e l l how o th e r people
are going to a c t .

13.

When th in g s a re go in g w e ll f o r me I
u s u a lly t h in k o f 1 t as a run o f good
lu c k .

14.

Most people d o n 't r e a liz e how much t h e ir
liv e s are in flu e n c e d by th in g s t h a t J u s t
a c c id e n tly happen.

15.

1 have u s u a lly found th a t what is go in g to
happen w i l l happen no m a tte r what I th in k
o r do about i t .

16.

Most o f th e th in g s th a t have d is a p p o in te d
me in my l i f e have come because my lu c k
ran o u t.

17.

1 d o n 't r e a l l y b e lie v e th e sa yin g t h a t a
person can be "th e m aster o f h is f a t e . "

18.

Success i s m o s tly a m a tte r o f g e ttin g
good breaks.

19.

Hhat happens in th e w o rld seems to be
beyond the c o n tr o l o f most pe op le.

20.

1 fe e l th a t most people c a n 't r e a lly be
he ld re s p o n s ib le f o r t h e ir a c tio n s .

31.

Many tim es the way people a c t has
a b s o lu te ly no reason behind i t .

22 .

Success in w o rk in g w ith o th e r people depends
much more on th e way th e y fe e l than how 1 f e e l.

23.

Many tim es 1 f e e l th a t 1 have l i t t l e
over the th in g s th a t happen to me.

24.

Sometimes 1 f e e l th a t 1 d o n 't have enough
c o n tro l over th e way my l i f e is go in g.

25.

To get ahead you have to gamble on th in g s
th a t you are unsure o f .

in flu e n c e

Thank you f o r yo u r h e lp by c o m p le tin g th e su rv e y .

170

Aide ID Number:

WHAT DO YOU THINK?

We hope you will help us Improve our program by participating in a survey.
There are several Items here regarding how people feel about events 1n our
lives. Each Item has two choices lettered a. or_b. Choose the statement
which you actually believe to be most true as far as you are concerned.
Choose the one choice of each pair that represents how you feel. Are you
willing to participate? (If yes, go on. If no, go on with the lesson.)
This 1s a survey of personal opinions — there are no right or wrong answers.
Which statement 1n each pair do you agree with more?
1.

a.
b.

Many of the unhappy things 1n people's lives are partly due to bad luck.
People's misfortunes result from the mistakes they make.

a.
b.

In the long run, people get the respect they deserve.
Unfortunately, people often don't get credit for what they do no matter
how hard they try.

a.
b.

Without the right breaks, one cannot be a good leader.
Being a good leader depends on having many skills.

a.
b.

No matter how hard you try, some people just don't H k e you.
People who are not Hked, often do not try very hard to be liked.

a.

Becoming a success is a matter of hard work; luck has little or nothing
to do with It.
Getting a good job depends mainly on being 1n the right place at the
right time.

2.

3.

4.

5.

' b.
6.

a.
b.

The average citizen can have an Influence In government decisions.
This world Is run by the few people in power, and there Is not much the
little guy can do about 1t.

171

Page 2
What Do You Think?

7.
a. When I make plans, I am almost certain that I can make them work.
b. It Is not always wise to plan too far ahead because many things turn out
to be a matter of good or bad fortune anyhow.
8.

a. In my case, getting what I want has little or nothing to' do with luck.
b. Many times we might just as well decide what to do by flipping a coin.
9.

a. Who gets to be the boss often depends on who was lucky enough to be 1n
the right place first.
b. Getting "to the top" depends upon ability; luck has little or nothing
to do with 1t.
10

.
a. As far as world affairs are concerned, most of us are the victims of forces
we cannot understand or control.
b. By taking an active part 1n political and social affairs, the people can
Influence world events.

11.

a. Most people don't realize how much their lives are controlled by chance.
b. There really 1s no such thing as "luck."
12.

a. It Is hard to know whether or not a person really H kes you.
b. You can usually tell when a person H kes you.
13.
a. With enough effort, we can wipe out political corruption.
b. It 1s difficult for people to haye much control over the things politicians
do.
14.
a. Many times I feel that I have little control over the things that happen
to me.
b. It 1s Impossible for me to believe that chance or luck plays an Important
role 1n my life.

172

Page 3
What Do You Think?

15.
a. People are lonely because of their own personality.
b. There 1s not much use 1n trying too hard to please people; 1f they like
you, they like you.
16.
a. Host of the time I can't understand why politicians behave the way they do.
b. In the long run, the people are responsible for bad government.

173

A P PENDIX

m e ttle * *

> n u t r it io n education
EFNEP R ep o rtin g P ro je c t

expanded food

aro m a*

24-HOUR DIETARY FOOD RECALL

A1 (M

Ndae_

S

0«te

A io e 't to p __

Report P e rio d :

F i l l 1/ Neae_

C o a o u u r U se

f a a ll y 1OP__

Food R ecall NuaPer

AMOUNT

FOOD

I
|

HUMBER
OF
SERV1N6S

fooo

CODE

_ A p r l1
yes

A V C6.

HAT

no

BREAD i
CEREAL

f w it

MU

_Oc to p e r

Heal 1

1

S na ck

i

Neal 2

S na ck

Neal

3

1

1

t

S na ck

i
i
i
_

yei

no

•*

h o m ta a k e r p r e g n a n t o r l a c t a t i n g ?

8

th e

s
a

Is

1
1
T o ta l
R ia N te r
S e r v in g s

!

>

I

Niafcer o f S ervings

P ro te in

v m e in

Fat

Thiaaine

C arbohydrate
L ..
_ .
Revised 09/85
NCC1 P ro je c t

N ia c in

Alcohol
a

R ib o fla v in

j

SPOTCHECR REPORT

f i r c n t o f RM
C a lo rie s

•
1

<

C a lc i urn

v it a a in C
V ita a in

0

Iro n

Sodiua

F rw its
Meat
Other
Food Recall Score
CoM»uter Record Date

174

Vegetables

Bread 1 Cereal

P o ta ssiga
Zinc

i

Hi Ik

OTHER

SCORING TABLE FOR TWENTY-FOUR HOUR DIET
To find the Twenty-four Hour Diet score:
t. Select the (appropriate triale (below) on the b a jis o jth e number o tm ilk _ servings reported in Item 7, FAMILY flECORD-B
NOTE: C ircled nunbert I G). G ) I we the h ltfirs t score possible In e toodgioup. For maeher ol
eervlnga la rM J h a n tw circle d raariaer. use the circle d lum bar. Example, lo r 3 servings of
m ilk , use daem M ILK SERVINGS tria ls.

(0. 1 .(5 )o r more).

2.

Select the proper column ol the table on The basis ol Ihe number o l moot servings reported in Hem 8.

3.

Select the proper area o l the table on the basis ol the ntsnber of v t f f t t a b lt / f r u it servings reported in Item 9

4.

Find llae proper line ol Ihe lable on the basis ol the number ol bread/cwttl servings reporled in Item 10.

(0, I. 2.3.

(7)oraaoro).

The manlier lo the right ol this (in type style " 7 4 " ) is Ihe Twenty-four Hour D ial score. Enter the diet score at theAppropriate "auanlhs
in profyam" lime on the homemaker's FOOD AND NUTRITION PROGRESSION RECORD

0

Va*

0 MEAT

1 MEAT

S E R V IN G !

S O W IN G

Itia l

175

rrv h

0

1

WsaB

fin il

C * r « li

AI

m

k rn

vaariw it

ana*
C anal

0

1

0

1

to

1

Scar*

4

2

12

•

3

It

©
0

•
2

©
0

to

©
0

1

•

1

22

1

2

11

2

20

2

36

3

13

3

31

3

21

1

11

7

13

3

21

13
21

3

26

0

•

t

21

2

26

3

26
33“

2S
31

3

37

(« )

n
32

3
©

1 SKAT

0 MEAT
SERVINGS

@ M A T
S ER VIN G S

s e r v in g

3

41
4S

41
4t
64

(« )

vaa
F la il

B r**4
C * *« *l

S ca t*

va a
f r a il

* * a tf
C *r**<

Scats

0

3

0

11

0

M

1

i«

1

21

t

20

t7

2

12

2

27

2

17

3

IS

3

36

3

41

20

©

21

©

10

©

14

0

to

0

24

0

20

27

1

22

1

42

1

62

2

26

2

60

2

64

3*

3

31

3

64

3

00

41

©

17

©

64

©

S4

17

0

12

0

27

0

37

1

36

2

34

3

41

1

2

1

26

2

11

0

t

2

1

to

2

64

0

1

2

0

a

1

64

2

42

1

»*w
C ***-*1

0
l
2
3
©
0
I
2
3

©0
2

1
2

3

17

3

00

3

SO

3

©

41

©

64

©

70

26

0

16

0

36

0

41

©0

t

30
41

3

1

13

2

17

47

3

to

©

20
41

2

47

3

00

©

0

Va*
f r a il

47

2

0
1

37

S car*

26

©

21

W aat
Ca.aU

2

©
0

IS

1

0
1
2

2

41

2

©

0 MLR. SERVINGS

3

©
0

t2

1

3

29

1

32

2

©
0

•

1

1ST

2

26

2

©

©
0

4

21

rial)

S

2

0

V aa

M

3

1

MEAT

SERVING

i

2
0

•

S ER VIN G S

SER VIN G S

0

3

( 0

V*»-

Q

t

©
0

.1

Scar#

1 M U

0

©
0

2

M I L K S E R V IN G S

06

3

0)

1

64

2

SO

41

3

46

©

0

23

1

37

2

41

3

46

©

St

3

(« )

1

1

to

2

64

04

3

70

77

©

S6

0

30

0

46

©0

1

64

1

64

1

2

04

2

70

3

77

3

06

3

©

■2

©

01

©

3

(< )

3

(«)

2
1

2

1 MAT
SERVING

S ra ia

0
H
17
26
20
14
27
16
10
41
17
16
JO

41
47
26
30
41
47
60
20
41
47
SO
St

vaa
rm i

0

»**4
Cno d

0
1
2
3

3

SO
as

3

70
at
46
44
70
St
01

©0
1

(0

n

1
2

1
2
3

2
3

©

V a*
foasf

N
20

©0

1
2
3

©0
2

S<m«

4)
46
20
61
64
SO
S4
37
64
42
at
70
41

©0
1

© ma
SERVINGS

0

Biaaa
•c a r*
Canal

0
1
2
I

i

©
0
1
2
3

©0
I

7

2
3

©0
1

1
2
3

©0
1

(«)

2
3

©

21
74
43
17
61
10
ts
02
ts
a
41
02
ts
S2
SS
47
ta
02
ta
04
SI

to
OS

04
HO

APPENDIX Q

Pre Test Itea Analysis Results:

Homemaker Attitude Survey

(n • 130)

DIFFICULTY
INDEX

ITEM

DISCRIMINATION
INDEX

1.

I feel good about how my life Is going.

24

37

2.

Success Is mostly a matter of getting good breaks.

12

31

3.

I like to "take charge" of situations.

15

29

4.

Much of what happens to me Is probably a matter
of chance or luck.

18

43

8

20

5. The things that happen to most people are outside
their own control.
6.

In general, I don't like to make changes.

15

46

7.

Improving the way I eat 1s Important to me.

47

89

8.

I can't do much to improve the way I eat
because of my income.

9

26

9.

I often talk about nutrition with my family
or friends.

8

26

15

43

10. Nutrition is not a priority in my life.
11.

I am motivated to prepare nutritious meals.

18

51

12.

The way I eat now doesn't need improvement.

17

49

13.

It would benefit me to improve the way I eat.

35

71

14. Nutrition is a boring topic to me.

26

83

15.

There are not enough advantages to improving
my diet to make it worth the effort.

19

60

16.

There are many things I am more concerned with
than improving my eating habits.

5

20

17.

How convenient a food is to prepare affects
whether I use it.

26

14

18.

The price of a food affects whether I buy it.

22

40

Standard deviation ■ 3.25
Variance ■ 10.60
Mean item difficulty * 18
Mean item discrimination * 43

176

ITEM

Post Test Itea Analysis Results: Homemaker Attitude Survey
(n - 177)
DIFFICULTY
INDEX

DISCRIMINATION
INDEX

1.

I feel good about how my life is going.

28

47

2.

Success is mostly a matter of getting good breaks.

15

34

3.

I like to "take charge" of situations.

16

22

4. Much of what happens to me is probably a matter
of chance or luck.

12

33

5. The things that happen to most people are outside
their own control.

14

20

6.

In general, I don't like to make changes.

13

30

7.

Improving the way I eat Is Important to me.

52

78

8.

I can't do much to improve the way I eat
because of my Income.

18

51

9.

I often talk about nutrition with my family
or friends.

16

45

10.

Nutrition Is not a priority 1n my life.

27

56

11.

I am motivated to prepare nutritious meals.

31

70

12.

The way I eat now doesn't need improvement.

7

23

13.

It would benefit me to Improve the way I eat.

27

49

14.

Nutrition Is a boring topic to me.

32

72

15.

There are not enough advantages to improving
my diet to make it worth the effort.

31

70

16.

There are many things I am more concerned with
than Improving my eating habits.

13

34

17.

How convenient a food is to prepare affects
whether I use it.

8

21

18.

The price of a food affects whether I buy it.

21

41

Standard deviation - 3.39
Variance * 11.51
Mean item difficulty ■ 22
Mean item discrimination * 38

177

Aide ..0. Ho:
____________
Aide Code Ho: ________________
Homemaker l . D . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Lessor Ho:
--------Aide Experience (years) ______
Post Test Itea Analysts for homemakers
EFHEP SURVEY
n • 177
Directions:
This survey has three sets of questions. The first part deals with your opinions
about some general areas. The second part asks for your opinions about nutrition.
The last part relates to your usual pattern of eating. Please answer all questions
based on how much you agree or disagree with the statement. There are no right or
wrong answers since this survey is about feelings and opinions. All responses will
be treated confidentially.

/ S

/ &/''''*/

/ «> / & / g

PART A: General Questions
1 feel good about how qy life 1s going.

28* 51

Success is mostly a matter of getting good breaks.

12

6

3

24

17

36

15*

16* 48

24

10

2

8

1 like to "take charge" of situations.

•>/<

/ £ / £

Much of what happens to me is probably a matter
of chance or luck.

6

19

14

49

12*

The things that happen to most people are outside
their own control.

5

13

19

49

14*

4

19

11

53

13*

52* 44

3

1

0

14

8

53

18*

16* 61

14

9

1

11

10

46

27*

I air, motivated to prepare nutritious meals.

31* 53

9

6

0

12. The way I eat now doesn't need improvement.

7* 18

14

54

7*

27* 57

7

5

3

In general, I don't like to make changes.
PART B: Nutrition Questions
Improving the way I eat is Important to me.
8. 1 can't do much to Improve the way 1 eat
because of my Income.
9.

7

I often talk about nutrition with my family
or friends.

6

10. Nutrition is not a priority in my life.
11.

13.

It would benefit me to improve the way I eat.

14. Nutrition is a boring topic to me.
15. There are not enough advantages to improving
Hy diet to make it worth the effort.

178

|

2

5

6

56

32*

3

2

9

54

31"

PART •: Nutrition Questions (cont.)
16. There are many things 1 am more concerned with
than improving my eating habits.
17. How convenient a food is to prepareaffects
whether 1 use it.
16. The price of a food affects whether I buy it.

PART C: Food Habits
Write the number 1 next to the answer that is the roost important reason you choose
foods as you do; give a number "2" to the seoend roost important reason; and use "3“
for the least important reason.
19. How do nutritional value, taste, and cost affect your food choices? (1 * most
important; 3 • least important)
21
taste/flavor
45
nutritional value
25 cost
20. The Improvements 1 have made in the way I eat have been: (1 * best reason;
3 ■ worst reason)
im p ro v e

my

SO

to

27

b e c a u s e o f new i n f o r m a t i o n

h e a lth

14
1 h ave

to save

m oney

le a rn e d

21. The biggest reason I have not made moreimprovements
in the way 1 eat is:
(1 • most Important reason; 3 ■ least important reason)
52 it's too hard to change my food habits.
a
it probably won't result in any benefit to me.
?a I am too busy with other things.
22. How many servings of each of these types of food doyouusually
1-2 times
0
per week
1 2
vegetables
______________
fr u it

o r fr u it

ju ic e

24.

3

4

_________________________________________________

soft drinks
cheese/milk
bread/pesta/grain foods
meat, eggs, poultry, or fish
dried beans, split peas, nuts,
peanut butter
sweets (cookies, candies, cakes, etc.)
23.

eat every day?

_______________________________
_______________________________
_______________________________
_______________________________
_______________________________
[ I
I
I
I
I

Do you take a vitamin supplement (pill)?
yes

no

If yes, what type? ____________ ;______________________________ _
THANK YOU FOR YOUR PARTICIPATION!

179

Pre Test Itea Analysis Results:

Aide Attitude Survey

(n - 67)

DIFFICULTY
INDEX

ITEM
1. I feel good about how ay life 1s going.

DISCRIMINATE
INDEX

46

61

31

67

3. I like to "take charge" of situations.

25

22

4. Much of what happens to me 1s probably a matter
of chance or luck.

39

61

5. The things that happen to most people are outside
their own control.

22

44

6. In general, I don't like to make changes.

25

61

7. Improving the way I eat Is Important to me.

64

61

8. I can't do much to Improve the way I eat
because of my Income.

60

83

9. I often talk about nutrition with my family
or friends.

48

78

10. Nutrition 1s not a priority 1n my life.

46

77

11. I am motivated to prepare nutritious meals.

54

88

12. The way I eat now doesn't need Improvement.

13

16

13. It would benefit me to improve the way I eat.

40

45

14. Nutrition 1s a boring topic to me.

40

83

15. There are not enough advantages to improving
my diet to make It worth the effort.

48

67

16. There are many things I am more concerned with
than Improving my eating habits.

12

39

17. How convenient a food Is to prepare affects
whether I use 1t.

13

0

18. The price of a food affects whether I buy It.

25

22

2.

Success 1s mostly a matter of getting good breaks.

Standard deviation * 3.96
Variance ■ 15.73
Mean item difficulty ■ 37
Mean item discrimination * 54

180

Itea Analysis Results:

Aide Post Test Survey

(n ■ 44)

DIFFICULTY
INDEX

ITEM

DISCRIMINATION
INDEX

1. I feel good about bow my life is going.

34

27

2. Success 1s mostly a matter of getting good breaks.

20

55

3.

I like to "take charge" of situations.

14

9

4. Much of what happens to me 1s probably a matter
of chance or luck.

32

73

5. The things that happen to most people are outside
their own control.

23

64

6.

In general, I don't like to make changes.

9

16

7. Improving the way I eat 1s Important to me.

68

55

8. I can't do much to Improve the way I eat
because of my Income.

41

82

9. I often talk about nutrition with my family
or friends.

52

55

10. Nutrition Is not a priority 1n my life.

34

82

11. I am motivated to prepare nutritious meals.

50

91

12. The way I eat now doesn't need Improvement.

14

27

13. It would benefit me to improve the way I eat.

27

36

14. Nutrition Is a boring topic to me.

55

82

15. There are not enough advantages to improving
my diet to make it worth the effort.

55

91

16. There are many things I am more concerned with
than Improving my eating habits.

16

45

17. How convenient a food is to prepare affects
whether I use it.

16

45

18. The price of a food affects whether I buy it.

11

9

Standard deviation ■ 3.80
Variance ■ 14.46
Mean item difficulty ■ 30
Mean item discrimination • 45

181

Aide I.D. Mo:
Aide Code No:
Homemaker I.D.
Lesson No:
Aide Experience Irears)
~
Post Test Item Mialysls for Program Aides
CFNEP SURVEY
(n - 44)
Directions:
This survey hss three sets of questions. The first part deals with your opinions
about tome general areas. The second part asks for your opinions about nutrition.
The last part relates to y o u r usual pattern of eating. Please answer all questions
based on how much you agree or disagree with the statement. There are no right or
wrong answers since this survey Is about feelings and opinions. All responses will
be treated confidentially.

PART A: General Questions
1.

1 feel good about how my life is going.

2. Success is mostly a matter of getting good breaks.
3. I like to ‘take charge* of situations.
4. Much of what happens to me is probably a matter

34* 50

9

2

2

9

20

50

20*

14* 52

16

16

2

0

0

0

18

50

32*

5. The things that happen to most people are outside
their own control.

2

5

18

52

23*

6. In general, I don't like to make changes.

2 14

18

50

9*

11

43

41*

11

45

34*

18

59

14*

14. Nutrition is a boring topic to me.

39

55*

15. There are not enough advantages to Improving
my diet to make it worth the effort.

41

55*

of chance or luck.

PART B: Nutrition Questions
7. Improving the way I eat is important to

68*

25

8. I can't do much to Improve the way I eat
because of my income.
9. 1 often talk about nutrition with my family
or friends.

52* 45

10. Nutrition is not a priority in my life.
11. I am motivated to prepare nutritious meals.

50* 43

12. The way I eat now doesn't need improvement.
13. It would benefit me to improve the way I eat.

182

27* 57

PART I: Rutrltlon Questions (cont.)
16. There are many things 1 am more concerned with
than Improving my eating habits.

2

16

23

43 16*

17. How convenient a food is to prepare affects
whether 1 use It.

5

9

27

43 16*

11* 41

23

16

18. The price of a food affects whether I buy It.

9

PART C: Food mbits
Write the number 1 next to the answer that 1s the most important reason you choose
foods as you do; give a number "2* to the seocnd most Important reason; and use “S’'
for the least important reason.
19. How do nutritional value, taste, and cost affect your food choices? (1 * most
Important; 3 ■ least Important)
41
taste/flavor
aa nutritional value
q cost
20. The improvements 1 have made 1n the way I eat have been: (1 ■ best reason;
3 ■ worst reason)
57 to Improve my health
n
to save money
36 because of new information 1 have learned
21. The biggest reason I have not made more improvements in the way 1 eat is:
(1 * most Important reason; 3 ■ least Important reason)
50
it's too hard to change my food habits.
0 it probably won't result in any benefit to me.
34 1 am too busy with other things.
22. How many servings of each of these types of food doyou usually eat every day?
1-2 times
0
per week
1 2
3
4
vegetables
[ I I I
fruit or fruit juice
______________________________
soft drinks
______________________________
cheese/milk
______________________________
bread/pasta/grain foods
______________________________
meat, eggs, poultry, or fish
______________________________
dried beans, split peas, nuts,
peanut butter
______________________________
sweets (cookies, candies, cakes, etc.) [ I
I
I
I
I
23. Do you take a vitamin supplement (pill)?
yes
no
24.

If yes. what type?____________:_______________________ ____________
THANK YOU FOR YOUR PARTICIPATION!

183

LIST OF REFERENCES

LIS T OF REFERENCES
Ajzen, I., and M. Fishbein. U nde rst an din g attitudes and
pr e d i cting social b e h a v i o r , Engl ewo od Cliffs, N J : PrenticeHall, Inc., 1980, pp. 41-91.
Alford, H. and S. Ekvall. V a r i a b i l i t y of di eta ry assessment
values among nut r i t i o n students. Journal of the American
Dietetic A s s o c i a t i o n 84:71-74, 1984.
Allport, G.W. Attitudes, In A h and boo k of social
p s y c h o logy.
C.A. Murchison, ed. N e w York: Russell &
Russell, 1935, pp. 798-884.
Axelson, J.M. Rep eat ed measurements in evaluation.
of N u t ri tio n Ed u c a t i o n 16:12-14, 1984.
Baird, P.C. and H.G. Schutz.
to "selling" good nutrition.
8:13-17, 1976.

Journal

The marketing concept applied
Journal of N u t r i t i o n Education

Beaton, G . H . , J. Milner, P. Corey, V. McGuire, M. Cusins,
E. Stewart, M. de Ramos, D. Hewitt, P.V. Grambsch, N.
Kassim, and J.A. Little. Sources of va ri a n c e in 24-hour
dietary recall data: Implications for n u tr it ion study
design and interpretation. A merican Journal of Clinical
Nutrition 32:2546-59, 1979.
Bern, D.J. B e l i e f s , attitudes and h u m an a f f a i r s . Belmont,
CA: Brook and Cole, 1970, p p . 66-70.
Blalock, H.M. Jr. Beyond ordinal measurement: Weak tests of
strong theories. In Measur eme nt in the social s c i e n c e s :
T h e o r i e s and s t r a t e g i e s . Chicago: Al dine Publishers, 1974,
pp.
424-33.
Block, A., S. Laughlin, A. DelTredici, and C. Omelich.
Executive Summary: The Cal ifo rn ia EFNEP E v a l u at ion Study,
1984, 6pp.
Bohrnstedt, G . W . , and T.M Carter. Robustness in regression
analysis, In Sociological m e t h o d o l o g y . H.L. Costner, ed.
San Francisco: Jossey-Bass, 1971, pp. 116-22.
Brehm, J. P o s t - dec is ion changes in d e s ir abi lit y of
alternatives. Journal of Abnormal and Social Psychology
52:384-89, 1956.
184

Brush, K.H., D.M. Woolcott, and G. F. Kawash. Eval ua tio n of
an af fective-based adult nutr it ion e ducation program.
Journal of Nu trition Education 18:258-64, 1986.
B y r d - B r e d b e n n e r , C., L.H. O'Connell, B. Shannon, and J.M.
Eddy.
A n utrition cu rri cul um for health education: Its
effect on students' knowledge, attitude, and behavior.
Journal of School Health 54:385-88, 1984.
Campbell, D.T.
Social attitudes and other acquired
behavioral dispositions, In P s y c h o l o g y : a study of a
science.
S. Koch, ed. N ew York: McGraw-Hill, 1963, pp.
164.

94-

Campbell, D.T. and J.C. Stanley. Experimental and quasiexperimental d e s i g n for r e s e a r c h . Boston: Houghto n Miff li n
Co., 1963, pp. 24-25.
Carlson, E.R. Attitu de change through modi fic ati on of
attitude structure.
Journal of Abnormal and Social
Psy cho lo gy 52:256-61, 1956.
Carmines, E.G., and R. A Zeller. R el iab il ity and val idi ty
a s s e s s m e nt. J. Sullivan, Ed. B e v erl y Hills: Sage
Publications, 1979, p. 48.
Carruth, B.R., and H L. Anderson. Scaling criteria in
de ve loping and eva luating an instrument. Journal of the
American Dietetic A ss oci at ion 70:42-47, 1977.
Carruth, B.R., M. Mangel, and H.L. Anderson. Assessing
c hange-proneness and n u t r i t i on- re lat ed behaviors. Journal
of the Am erican Dietetic A sso ci ati on 70:47-52, 1977.
Connell, D.B., R.R. Turner, and C.F. Mason. Summary of
findings of the school health educ ati on evaluation: Health
promotion effectiveness, implementation, and costs. Journal
of School Health 55:316-21, 1985.
Coope rat iv e E xtension Service. Program evaluation in
Extension: A compre he nsi ve study of methods, practices,
procedures. 1981, pp 77-80.
Cosper,B.A., and L.M. Wakefield. Food choices of women.
Journal of the Am er ica n Dietetic A ss oci at ion 66:152-55,
1975.
Daelhousen, B.B., and H.A. Guthrie. A self-instruction
nu trition prog ram for pregnant women. Journal of the
American Dietetic Ass oci at ion 81:407-12, 1982.
Dalton, S.D., R.A. Linke, and M.D. Simko. W ork sit e food
choices: An i nvestigation of intended and actual sel ec ­
tions.
Journal of Nu tr i t i o n Educa ti on 18:182-86, 1986.
185

and

Davie, L.R., R.O. Butler, B. Williams, and S. Meiners.
Final Report: A study of c ertain personality, skill, and
demographic char act er ist ic s of the Expande d Food and
N utrition Educ ati on Pro gr am aides and h o m e m a ker s^
Wash in gto n State Uni ver si ty Co ope ra tio n E xte ns ion Service,
1973, 141 pp.
Doob, L.W. The be hav ior of attitudes.
Review 54:135-56, 1947.

Psychological

Eden, I., S.K. Kamath, M.B. Kohrs, and R.E. Olson.
Perceived control of n utrition behavior: A study of the
locus of control t heory among h e a l t h y subjects. Journal of
the Ameri can Dietetic A s soc iat io n 84:1334-39, 1984.
Fazio, R.H. "How do attitudes guide behavior?"
In The
Handbook of m o t i vat io n and c o g n i t i o n : Foundations of social
b e h a v i o r , R.M. Sor ren ti no and E.T. Higgins, Eds.
New York:
Guilford Press, 1986.
Fazio, R.H., M.C. Powell, and P.M. Herr. Toward a process
model of the a tt itu de -be ha vio r relation: Acces sin g one's
attitude u pon mere obser vat ion of the at titude object.
Journal of P e r s o nal ity and Social P s y c hol ogy 44:723-35,
1983.
Fazio, R.H. and M.P. Zanna. On the pred ic tiv e val id ity of
attitudes: The role of d irect exp eri en ce and confidence.
Journal of P e r s o nal ity 46:228-43, 1978.
Fishbein, M. Attit ude s and the p r e d i c t i o n of behavior, In
Readings in attitude theory and m e a s u r e m e n t . M. Fishbein,
ed.
N e w York: John W i l e y and Sons, 1967.
Fitz-Gibbon, D.T. and L.L. Morris. H ow to design a p rogram
evaluation. Bev erl y Hills: Sage Publications, 1983, pp. 2731.
Flay, B.R., D. Phil, D. DiTecco, and R.P. Schlegel.
Mass
media in h e alt h promotion: An a nalysis using an extended
in fo rma tio n- pro ces si ng model. Health E d uc ati on Quart er ly
7:127-47, 1980.
General A c c o unt in g Office. Areas needing i mprovement in the
Adult Ex pa n d e d Food and N u t ri tio n E d uc at ion Program.
Pu blication No. CED-80-138, 1980, 41 pp.
Green, B.F. At titude measurement. In H and boo k of social
psychology.
Reading, Mass.: A d d i s o n - W e s l e y , 1954, pp. 33569.
Green, P.E., and D.S. Tull. Research for ma rketing
d e c i s i o n s . Englewood Cliffs, N .J .:Prentice Hall, Inc.,
1978, pp 192-93.
186

Guiry, V.C., and C.A. Bisogni. C affeine knowledge,
attitudes, and pract ice s of young women. Journal of
Nu trition Education 18:16-22, 1986.
Guthrie, H.A. S election and q u a n t i fi ca tio n of typical food
portions by young adults. Journal of the Am erican Dietetic
A s sociation 84:1440-44, 1984.
Haggstrom, W.C. The power of the poor, In Poverty in
A m e r i c a . L.A. Rerman, J.L. Kornbluh, and A. Hobr, eds.
Ann Arbor: University of M ic higan Press, 1966, pp. 315-34.
Harding, J . , B. Kutner, H. Proshansky, and I. Chein.
Prejudice and ethnic relations, In Handbook of social
p s y c h o l o g y . G. Lindzey, ed. Reading, Massachusetts:
A d d i s o n - W e s l e y , 1954, pp. 1021-61.
Heider, F. The p s y chol ogy of interpersonal r e l a t i o n s . N e w
York: John W i l e y and Sons, 1958.
Henerson, M . , L.L. Morris, and C.T. FitzGibbon.
How to
measure a t t i t u d e s . B e v er ly Hills: Sage Publications, 1978,
pp. 131-150.
Hollis, J.F., T.P. Carmody, S.L. Conner, S.G. Fey, and J.D.
Matavazzo.
The n u tr iti on attitude survey: Associations
with d i e t ar y habits, psychological and physical well-being,
and coronary risk factors. Health Psy cho log y 5:359-74,
1986.
Hutsey, J. An eval uat io n of the Expand ed N utrition
Education Program in Pennsylvania. Un iv ersity Park:
Pennsylvania State University, 1970.
Iversen, G.R.,
Beverly Hills:

and Norpoth, H. Analysis of v a r i a n c e .
Sage Publications, 1976, pp 11-69.

Jones, E.M., S.J. Munger, and J.W. Altman. A field guide
for eva lua ti on of nutri tio n education.
Allison Park, PA:
Synectics Corporation, 1975.
Kaplowitz, D.D., and C.M. Olson. The effect of an education
program on the de ci s i o n to breastfeed. Journal of Nutr it ion
Education 15:61-65, 1983.
Kerr, M., M. Andrews, L. Nierman, and C. Kataregga.
E v aluation of the long-term effects of EFNEP in Michigan.
E. Lansing, Michigan: Mi chigan State University Cooperative
E xtension Service, 1979, 22 pp.
Kok, F.J., A.W. Matroos, J.G.A.J. Hautvast, and A.W.
Vandenban. D irection for nutrition education determined by
target group segmentation. Journal of the A me rican Dietetic
Assoc iat io n 80:566-73, 1982.
187

Kuder, G.F. and M.W. Richardson. The theory of the
e s timation of test reliability. Ps ychometrika 32:1-13,
1937.
Kutner, B . , C. Wilkins, and P.R. Yarrow. Verbal attitudes
and overt behavior involving racial prejudice. Journal of
Abnormal and Social Ps yc hol ogy 47:649-52, 1952.
L a P i e r e , R.T. Attitudes vs. actions.
13:230-37, 1934.

Social Forces

Likert, R. The method of constructing an attitude scale, In
S c a l i n g : A sourc ebo ok for beh avioral sc i e n t i s t s . G.M.
Maranell, ed. Chicago: Aldine Publishing Co., 1978, pp 23343.
Lodge, M. Magnitude Scaling: Q u a n t itative m e a surement of
opinions. Beverly Hills: Sage P u b l i c a t i o n s ’
, 1981.
Looker, A. and B. Shannon. Threat vs. benefit appeals:
Effectiveness in adult nutrition education. Journal of
Nutrition Education 16:173-76, 1984.
Madden, J.P., S.J. Goodman, and H.A. Guthrie. V ali di ty of
the 24-hour recall. Journal of the Am erican Dietetic
Association 68:143-47, 1976.
Maiman, L.A., V.L. Wang, M.H. Becker, J. Finlay, and M.
Simonson. Attitudes toward obesity and the obese among
professionals. Journal of the A merican Dietetic Association
74:331-36, 1979.
Marketing Economics Research Service. Impact of the
Expanded Food and Nutri ti on Education Program on low-income
f a m i l i e s : An indepth a n a l y s i s . W a s h ing to n D . C . : U.S.
Department of Ag riculture Economic R es earch Service,
Agricultural Economic Report No. 220, February, 1972, 73
pp.
Matheney, R . J . , M.F. Picciano, and L. Birch. Attitudinal
and social influences on infant-feeding preference. Journal
of N u tr iti on E d u ca tio n 19:21-31, 1987.
McKenna, C. E v a l u at ion for accountability.
Extension S e p t e m b e r / O c t o b e r :22-26, 1983.

Journal of

Michigan State Uni versity Cooperative E x te nsi on Service.
Eating right is basic 2. E. Lansing, MI: M ic hig an State
University Press, 1986.
Munger, S.J. Expanded Food and N ut rit io n Educ at ion Program:
A final ev al uation in the maturing program, April 1970 March 1971.
Allis on Park, Penn.: Synectic Corporation,
1971, 65 pp.
188

Napier, T.L., and C. Wharton. The E xpanded Food and
Nutrit io n E ducation Program: An exp eriment in behavior
change.
Ohio A g r i c u ltu re Re sea rc h and Develo pme nt Center,
Research Bu lletin No. 1070, 1974, 18 pp.
Nierman, L . , S. Walker, and K. Akpom. Eff ec t i v e l y teaching
foods and nutrit io n to low income families.
E. Lansing,
Michigan: Mic hi gan State Univ ers it y Coope rat iv e Extension
Service, August 1983, 69 pp.
O'Connell, L . H . , B. Shannon, and L. Sims. Assessing
n u t rit io n-r ela te d attitudes and bel ief s of teachers.
Journal of Nutrit io n Educ ati on 13:S80-85, 1981.
Olson, J.C., and L.S. Sims. A s ses sin g nutrit io n knowledge
from an infor mat ion pr oce ssi ng perspective.
Journal of
Nutrition Education 12:157-61, 1980.
Orne, M. On the social psyc hol og y of the p sy chology
experiment: W i t h p ar ticular reference to demand
characteristics and the implications. Am er i c a n Psychologist
17:776-83, 1962.
Pao, E.M., S.J. Mickle, and M.C. Burk. O n e -da y and 3-day
nutrient intakes by individuals - N a t i onw id e food
co nsumption survey findings, Spring 1977. Journal of the
American Dietetic A s soc ia tio n 85:313-24, 1985.
Penner, K.P., and K.M. Kolasa. Se c o n d a r y teachers'
nutrition knowledge, attitudes, and practices. Journal of
N utrition E duc ati on 15:141-45, 1983.
Perron, M. and J. Endres. Knowledge, attitudes, and dietary
practices of female athletes. Journal of the American
Dietetic A ss oci at ion 85:573-76, 1985.
Peterson, M.E., and C. Kies. N u tri ti on knowledge and
attitudes of ear ly e l e m en tar y teachers. Journal of
Nutri tio n E d uc at ion 4:11-15, 1972.
Picardi, S.M., and D. Porter. Multid ime nsi on al evaluation
of a food and nu t r i t i o n minicourse.
Journal of N utrition
Education 8:162-68, 1976.
Ramsey, C.E., and M. Cloyd.
Mul tip le o bje ctives and the
success of educational programs. Journal of Nutri tio n
Education 11:141-44, 1979.
Ramsey, C . E . , and M. Cloyd. An e val uation of EFNEP side
effects. Journal of E x te ns ion X I I I : 11-20, 1975.
Reames, E.S. Opinions of physicians and hospitals of
current br ea s t - f e e d i n g recommendations. Journal of the
Am erican Dietetic A s s o c i ati on 85:79-80, 1985.
189

Regan, D.T., and R.H. Fazio. On the consistency between
attitudes and behavior: Look to the method of attitude
formation. Journal of Experimental Social Ps yc hology 13:3845, 1977.
Ries, C.P., and D.K. Schoon. E v a lua tio n of a cafeteriabased e du ca tio n prog ram for col leg e students. Journal of
Nu tri tio n Education 18:107-10, 1986.
Rosander, K., and L.S. Sims.
Measu rin g effects of an
affective -b ase d nutr it ion e ducation intervention.
Journal
of N ut r i t i o n Educat io n 13:102-5, 1981.
Rosenberg, M.J. Cognit iv e r e or gan iz ati on in response to the
hypnotic reversal of attitudinal effect. Journal of
Pe rs ona lit y 28:39-63, 1960.
Ross, J.K. Nutri tio n attitudes and knowledge of nursing
students. Journal of the Americ an Dietetic As sociation
84:687-88, 1984.
Sanjur, D. Social and c u l tural p er spectives in n u t r i t i o n .
Englewood Cliffs, N.J.: P r e n t i c e - H a l 1, Inc., 1982 pp. 169231.
Schafer, R . , and E.A. Yetley. Social psyc ho log y of food
faddism. Journal of the A me rican Dietetic As soication
66:129-33, 1975.
Schafer, R.B. Factors affecting food behavior and the
quality of husbands' and wives' diets. Journal of the
American Dietetic Association, 72:138-43, 1978.
Schwartz, N.E. N u tri tio n
knowledge, attitudes, and
practices of high school
graduates. Journal of the Am eri ca n
Dietetics As soc ia tio n 66:28-31, 1975.
Schwartz, N.E. N u tri tio n
knowledge, attitudes, and
practices of Ca nadian public heal th nurses.
Journal of
N u t rit io n E duc at ion 8:28-31, 1976.
Shaw, M.E., and J.M. Wright. S c a l e s for the measur eme nt of
a t t i t u d e s . N e w York: McGraw-Hill B oo k Co., 1967, pp 1-21.
Sims, L.S. De mographic and attitudinal correlates of
nutrition knowledge.
Journal of Nu tr i t i o n Education
8:122-25, 1976.
Sims, L.S. Dietary status of lactating women.
the Ame ri can Dietetic Association. 73:147-54,

Journal of
1978.

Sims, L.S. Toward an u n d e r sta nd ing of attitude assessment
in nutr it ion research.
Journal of the America n Dietetic
A s soc iat io n 78:460-66, 1981.
190

Sorenson, A . W . , B.M. Calkins, M.A. Connally, and E.
Diamond. Com parison of nu trient intake d e t er mi ned by four
d i e t ar y intake instruments. Journal of Nutrit ion E d uca ti on
17:92-99, 1985.
SPSS,

Inc.

SPSSx u s e r 's guide.

Chicago:

M c G r a w Hill,

1983.

Stunkard, A.J., and M. Waxman. A cc u r a c y of self-reports of
food intake. Journal of the A mer ica n Dietetic Assoc iat ion
79:547-51, 1981.
Sunseri, A . J . , J.A. Alberti, N.D. Dent, J.A. S c h o e n b e r g e r ,
and T .A . Dolecek. Ingredients in n ut r i t i o n education:
Family involvement, reading, and race. Journal of School
Health 54:193-96, 1984.
United States De p a r t m e n t of Agriculture. Revised policy
guidelines and suggestions for co nd ucting the E xtension
Expanded Food and N u tr it ion Education Program.
Home
Economics Report No. 100, 1976, 19 pp.
Wagner M. M., M. H. Machover, L. Newman, and J.
Fairweather.
An experimental ev aluation of nutrition
education methods: Final Report, 1983, 177 pp.
Wallston, B.A., and K.A. Wallston. Locus of control and
health: A re view of the literature. Heal th Educa tio n
Monographs 6:107-17, 1978.
Young, C.M., G.C. Hagan, R.E. Tucker, and W.D. Foster. A
comparison of d i e t ary study methods: D i e tar y h i s t ory vs.
seven-day record vs. 24-hour recall. Journal of the
American Dietetic As s o c i a t i o n 28:218-21, 1952.
Yperman, A.M., and J.A. Vermeersch. Factors as soc iat ed with
children's food habits, Journal of N u tr iti on E d u cat io n
11:72-76, 1979.

191