LIBRARY
Michigan Stale
University

 

 

 

PLACE Ii RETURN BOX in roman thin checkout hum your record.
TO AVOID FINES Mom on 0' More data duo.

DATE DUE DATE DUE DATE DUE
-4- z Al'fli-é

; 013 0}
72003

 

' T“
*3222 I I
w.-

I

 

 

 

IL

 

 

 

 

 

 

 

 

 

 

[:ij

MSU icAnNiirmctivc ActioniEqui Oppomnily initiation
m:

 

 

AN EVALUATION OF THE READABILITY AND ACCEPTABILITY OF
MATERNAL NUTRITION EDUCATION MATERIALS FOR EXPANDED
FOOD AND NUTRITION EDUCATION PROGRAM CLIENTS

BY

Karen Jane Schmitz

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1994

ABSTRACT

AN EVALUATION OF THE READABILITY AND ACCEPTABILITY OF
MATERNAL NUTRITION EDUCATION MATERIALS FOR EXPANDED
FOOD AND NUTRITION EDUCATION PROGRAM CLIENTS

BY

Karen Jane Schmitz

Written materials are often used in nutrition education
in an attempt to increase a client’s knowledge. However,
often the readability of these materials and the reading
level of the population are not assessed and compared. In
this study, the subjects were English speaking female
Expanded Food and Nutrition Education Program (EFNEP)
clients between 18 and 40 years of age. Michigan EFNEP
materials (flipchart and handouts) for pregnancy, titled
"Eating Right...for Two" (1992) were assessed for reading
grade equivalency and other characteristics of readability
using a scorecard developed by the investigator. In the
preliminary phase of this research, the reading level of a
sample of the Michigan EFNEP population was assessed using
the Slosson Oral Reading Test (SORT). The readability of

the printed materials and client reading level were compared

to determine the need for changes in readability of the
materials.

A new flipchart was developed (Client—Adapted Language
flipchart) using EFNEP-eligible women’s remarks and written
at the median reading grade level of the EFNEP women who
were involved in the reading level assessment. Several
surveys were used to assess client knowledge change and
acceptability of materials.

The results of the SORT indicated that the median
reading level of a sample of clients was 8.3 grade level
equivalency. Survey results indicated that a greater
percentage of clients agreed that "Eating Right...for Two"
was easier to read and less insulting compared to the
Client-Adapted Language flipchart. However, clients who
used the Client-Adapted Language flipchart had a
significantly greater increase in nutrition knowledge
compared to the women who used "Eating Right...for Two."
The method of allowing the target audience to produce the
wording for their own reading materials might result in more

understandable material.

Copyright by
Karen Jane Schmitz

1994

ACKNOWLEDGEMENTS

I would like to acknowledge my major professor, Dr.
Jenny Bond, for her guidance through this research project,
her encouragement in my efforts to begin a teaching career
and her interest in me as a person. Also, I would like to
thank the members of my committee (Dr. Wanda Chenoweth, Dr.
Anne Murphy, Dr. James Snoddy, Dr. Sandy Andrews) for
finding time in their busy schedules to read my proposal and
attend several meetings over the past few years. Dr. Anne
Murphy deserves a very special thank you for assistance with
designing the research plan, reviewing surveys, and planning
the pilot testing. In addition to these members of my
committee, I would like to thank Mrs. Stella Cash who
stepped on board at the last minute after the departure of
Dr. Andrews and took the time to read my dissertation and
attend my final defense.

I would like to thank several individuals for
assistance with the necessary details to complete this
project. Debie Lecato deserves my thanks forrelaying phone
messages, forwarding mail and formatting handouts. A
special thank you to undergraduate researchers: Susan

Miller, Paula Montgomery and Ann Shanks for checking data,

vi
stuffing envelopes and helping with the readability
scorecard. I’d also like to thank my fellow student, Donna
McLean, for encouraging me to continue with the doctoral
program when things became difficult.

I appreciate the support of everyone at the Expanded
Food and Nutrition Education Program at Michigan State
University, especially Barb Mutch and Char Mckay. Barb’s
interest in this project and willingness to provide
materials were extremely valuable during the data collection
phase of this research. Char provided a link to the
instructors in the counties as well as the motivation to
keep people working on this project month after month. Char
was organized, interested, and assisted in keeping the
project on a timeline.

I can’t forget the county Extension home economists and
instructors from Ingham, Kent, Kalamazoo, Oakland and
Genesee Counties who put in endless hours contacting new
clients, using the materials and collecting the survey data.
A special thank you to Ann Niewenhuis, former county
Extension home economist in Kalamazoo county, who took an
interest in the preliminary phase of this study. Her
assistance in the preliminary data collection and in the
development of the Client-Adapted Language flipchart was
extremely valuable in getting this project off the ground.
In addition, I would like to thank the individuals and

organizations responsible for funding my graduate studies

vii
and/or this project: The Michigan Agricultural Experiment
Station and the Department of Food Science and Human
Nutrition at Michigan State University.

Finally, my greatest thanks to my husband and best
friend, Pete, who spent many hours playing single father and
housekeeper. Without his support this project would never
have reached completion. My young son, Jeffrey, deserves a
thank you for being so good as he spent the first year of
his life being driven to and from meetings and seminars.
Also, I would like to thank my parents, Ray and Marilyn
Duve, and my grandfather, Mr. A.B. Kriz for their constant

support of my education and all of my endeavors in life.

TABLE OF CONTENTS

List of Tables .......................................... xii
List of Figures ......................................... xiv
I. Introduction .......................................... 1
Statement of Problem ...................... 1
Research Questions ........................ 2
Research Objectives ....................... 3
II. Review of Literature ................................. 4
Review of Literacy ........................ 4
Nutrition Education ....................... 41
Nutrition in Pregnancy .................... 46
III. Materials and Methods .............................. 82
Approval to Conduct Study ................. 82
Research Procedures ....................... 82
Sample Selection .......................... 87
Research Sites ............................ 90
Program Description ....................... 91
Selection of the Reading Levels Test ...... 92
Instrument Development .................... 93
Training of Data Collectors .............. 104
Data Collection .......................... 106

viii

ix

IV. Results and Discussion .............................. 115
A. Preliminary Research Phase ................. 115
1. Client Reading Levels ................. 115
2. Correlation Coefficients .............. 117
3. Readability of EFNEP's Written
Materials ...................... 119
B. Material Development Phase ................. 121
1. Client—Adapted Language Flipchart ..... 121
2. Readability of the Client—Adapted
Language Flipchart ............. 131
3. The Readability Scorecard
Used with Flipcharts and Handouts...132
4. Pilot Test Results:
Wording of Surveys ............. 135
5. Pilot Test Results: Item Analyses ..... 142
6. Readability of Surveys ................ 144
C. Data Collection Phase ...................... 146
1. Pregnancy Surveys I and II ............ 146
2. Material Acceptability Survey ......... 164
3. Materials Comparison Survey ........... 185
V. Benefits, Strengths and Limitations of the Study....197
Benefits ...................................... 197
Strengths and Limitations ..................... 198
A. Research Design ....................... 198
B. Test Instruments ...................... 200
C. Other Materials Developed by the
Primary Investigator ........... 203

X

VI. Recommendations ..................................... 206
A. Working within EFNEP .................. 206

B. Pilot-Testing of Materials ............ 207

C. Instructor Training ................... 208

VII. Summary ........................................... 210
VIII. Appendices ........................................ 213
A. Fry Graph ............................. 213

B SORT Test ............................. 214

C. Irwin/Davis Checklist ................. 216

D Approval to Conduct Study ............. 217

Approval of Preliminary Phase ....... 217

Approval of Data Collection Phase...218

E. Questions to Obtain Client Language...219
F. Knowledge Survey (First pilot) ........ 222
G. Knowledge Survey (Final Instrument)...227
H. Characteristics of Readability:

Scorecard ............ 233

I. Materials Acceptability Survey
(First pilot)....238

J. Materials Acceptability Survey
(Final Instrument) ........... 243
K. Materials Comparison Survey

(First pilot) ....... 249

L. Materials Comparison Survey
(Final Instrument)..252

M. Directions for instructors ............ 256
N. Consent Forms ......................... 258

O. EFNEP Handouts ........................ 260

xi

P. Pica Handout (Pilot form) ............. 264
(Final form) .............. 265
Q. Comments of EFNEP Clients ............. 266

XI. List of References ............................. 277

10.

11.

12.

13.

14.

15.

16.

17.

LIST OF TABLES

Flesch Reading Ease Scores ............................ 10
Dale-Chall Reading Grade Scores ....................... 12
Characteristics of Readability Formulas ............... 18
Recommended Dietary Allowances for Pregnant Women ..... 64

Characteristics of Readability: Definitions and
Recommendations ................................. 96

Characteristics of Readability: Assessment ........... 99
Mean Scores and Range of Scores, or Descriptions of
Characteristics of Readability Used for
Development of the Readability Scorecard ....... 101
Preliminary Research Results: The SORT Data .......... 116
Preliminary Research Results: Summary of SORT Data...117

The Reading Grade Levels of EFNEP’S Written Materials
for Nutrition During Pregnancy ................. 120

Wording of the Two Flipcharts ........................ 124

The Reading Grade Level Equivalents of the
Client-Adapted Language Flipchart .............. 131

Using the Readability Scorecard: Assessment of
Characteristics of Readability of Written

Materials ...................................... 133
Number of Correct Responses, Difficulty and
Discrimination of Knowledge Survey Items ....... 143
The Readability of the Pilot Survey Instruments ...... 145
Response Frequencies on the Knowledge Survey
(Pretest/Posttest) ................... 147
Response Frequencies for Correct Posttest Answers....150

xii

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

xiii

T-Test Results, Mean Scores and Range of Scores
for the Knowledge Survey ............ ' ........... 151

Number of Participants per County .................... 153

Mean Knowledge Survey Scores by Age of
EFNEP Clients .................................. 155

Mean Knowledge Survey Scores by Pregnancy Status of
EFNEP Clients ........................ 157

Mean Knowledge Survey Scores by Number of Babies Born
to the EFNEP Client .................. 158

Mean Knowledge Survey Scores by Race or Origin of
EFNEP Clients ........................ 161

Mean Knowledge Survey Scores by Level of Education of
EFNEP Clients ........................ 163

Frequency of Responses
(Material Acceptability Survey, Questions 1-8) ...... 165

Differences in Frequency of Responses
(Group 1 minus Group 2) ................... 167

Frequency of Responses (Material Acceptability Survey,
Questions 9-11, Length, Appearance and Print Size)..174

Frequency of Responses (Material Acceptability Survey,

Question 12--Pictures) ............... 177
Written Responses about Flipchart Pictures ........... 179
Written Responses about Improving the Flipcharts ..... 181

Frequency of Responses (Material Acceptability Survey,
Questions 14-19--Handout Acceptability) ........ 184

Frequency of Responses (Comparison Survey) ........... 185

Responses to Comparison Survey Questions, Reported
by the Order of Use of the Materials ...... 189

Written Responses about Differences
Between Flipcharts ................... 191

Frequency of Overall Flipchart Scores ................ 193

LIST OF FIGURES

Research Procedures .................................. 84

Example of Assignment of EFNEP Clients to

Research Groups ............ 90
Client Educational Level vs. Client SORT Scores ..... 118
Client Age vs. Client SORT Scores ................... 118

Comparison Survey Overall Scores for "Eating Right
...for Two" vs. Knowledge Survey Posttest Scores...194

Comparison Survey Overall Scores for the Client-
Adapted Language Flipchart vs. Knowledge
Survey Posttest Scores ......... 195

Comparison Survey Overall Scores for "Eating Right
...for Two" vs. Knowledge Survey Change Scores ..... 195

Comparison Survey Overall Scores for the Client—

Adapted Language Flipchart vs. Knowledge
Survey Change Scores ........... 196

xiv

INTRODUCT I ON

A. Statement of the Problem

Several studies have indicated that low—income, low—
literate individuals are the people least likely to
understand the message in written health materials
(Freimuth, 1979; Doak and Doak, 1980; Doak et al., 1985;
Meade et al., 1989). In a 1985 study, Doak et al. found
less than 10% of health-related educational pamphlets were
written below a seventh grade reading level. The low-
literate adult who reads below the seventh grade level is
left with little comprehensible written material on health
and nutrition.

Participants in this study will be a sample of the low-
income women ages 18-40 who participate in the Expanded
Foods and Nutrition Education Program (EFNEP) in Michigan.
The reading level of the average EFNEP participant has never
been reported and there is no documentation of measurement
of the reading grade level equivalent of Michigan's printed
EFNEP materials. Preliminary tests by the investigator of
written materials ("Eating Right is Basic" and "Eating
Right... for Two" flipchart and handouts) have shown a wide

variation in readability within the different sections from

2
grade seven to grade thirteen using a SMOG readability test.
By measuring the reading level of a sample from this
population and the readability of the written text, it can
be determined whether the existing materials for this
population are written at the appropriate grade level
equivalency. The target population, women in Michigan
between the ages of 18 and 40 who participate in EFNEP, will
benefit by having access to readable literature to increase

their knowledge of nutrition in pregnancy.

B. Research Questions

1. What is the average reading level of a sample of
women between the ages of 18 and 40 enrolled in
Michigan EFNEP?

2. What is the readability of EFNEP's "Eating
Right...for Two" flipchart and handouts?

3. What is the prevalence of characteristics that
affect readability (but are not measured by
formulas) in "Eating Right...for Two": number of
abbreviations, percent white space, font type
(Roman serif, Helvetica, etc.), type size (12 pt.,
14 pt., etc), type of letters (capital vs. lower
case), use of voice (passive vs. active), emphasis
on action (positive vs. negative) use of symbols,
number of headers used and leading size.

4. Which written materials produce the greatest
increase in knowledge of nutrition during
pregnancy:

a) Michigan EFNEP's "Eating Right...for
TWO? "

b) Michigan EFNEP's "Eating Right...for
Two" rewritten by nutrition-trained
researchers using clients’ original
language and written at the median
reading grade level of female EFNEP
clients in Kalamazoo County?

5.

3

Which of the two sets of written materials are
perceived to be the most informative, acceptable
and interesting among female EFNEP clients?

C. Research Objectives

1.

To determine the average reading level of women
between the ages of 18 and 40 enrolled in the
Michigan EFNEP program by using the SORT test.

To determine the readability of Michigan EFNEP
materials "Eating Right...for Two" flipchart and
handouts by using readability formulas.

To assess Michigan EFNEP maternal nutrition
materials for characteristics that affect
readability: number of abbreviations, percent
white space, font type (Roman serif, Helvetica,
etc.), type size (12 pt., 14 pt., etc.), letter
type (capital vs. lower case), use of voice
(passive vs. active), emphasis on action (positive
vs. negative), use of symbols, number of headers
used and leading size.

To determine which written materials produce the
largest increase in knowledge about nutrition
during pregnancy among a sample of female EFNEP
clients:

a) Michigan EFNEP’s "Eating Right...for Two"

b) Michigan EFNEP's "Eating Right...for Two"
rewritten by researchers using clients’
original language and written at the
median reading grade level of EFNEP
clients in Kalamazoo County.

To determine, by means of two surveys, which
written materials are perceived by the EFNEP women
as the most informative, acceptable and the most
interesting: a or b above.

REVIEW OF LITERATURE

This review of literature is a report of an
investigation of research in three major areas. The first
portion of this review focuses on literacy and the impact of
literacy on nutrition education in low-literate populations.
The second section reviews aspects of nutrition education
programs with a focus on the Expanded Foods and Nutrition
Education Program (EFNEP). Finally, the third portion of

this review will address research findings in the area of

nutrition during pregnancy.

Literacy

Illiteracyl is a complex cultural and social problem
that often isolates groups at risk: the poor, those who
live in the inner city, recent immigrants, racial or ethnic
minorities and high school drop—outs or push-outs (Nitzke,
1989). For many people lacking in literacy skills,
illiteracy is just one of many social problems complicating

their lives. Others include: illness, large families,

 

1 The definition of literacy as used in this dissertation is
having the ability to read, write and comprehend text written at

the fifth grade level.

5
immobility, inadequate housing, joblessness, low self-
esteem, high degree of welfare dependency and a general
sense of powerlessness (Harman and Hunter, 1979; Nitzke,
1989). This portion of the review of literature will look
at several aspects of literacy and the relationship between
literacy and health and nutrition education materials
including: formulas used to measure the readability of
text, considerations and concerns of using readability
formulas, factors not measured by formulas that might affect
the reader’s ability to comprehend text, the problems of
unreadable health education materials and guidelines for
producing readable text.

Harman (1987) suggests that three types of data need to
be looked at when describing the literacy levels of a group
of people: the results of national surveys on literacy and
functional literacyg, the results of readability studies
done on similar groups of people and the average grade level
completed in school. The most widely used data on American
literacy was collected by the U.S. Census Bureau (1986) and
suggest that 13% of Americans above the age of 20 are
illiterate (Harman, 1987).

One investigation of the prevalence of illiteracy in

the U.S., known as the Adult Performance Level (APL) Study,

 

2 The definition of functional literacy as used in this
dissertation is having the abilities of basic reading and writing
needed.to obtain and use information to solve problems of everyday
life.

6
was a joint project of the University of Texas and the U.S.
Office of Education (Kazemek, 1985). This four year study
of 10,000 people showed that over 3,000 of these adults
could not follow instructions, could not comprehend graphs
or bar charts and would need to read at the tenth grade
level to understand the directions on a bottle of aspirin
(Doak et al., 1985). Nearly 33% of the test population
lacked the ability to fill out job applications, address
envelopes or use a telephone directory and were, therefore,
labeled functionally illiterate (Harman, 1987).

According to Doak and Doak (1980), in the future
education will play an increasing role in health
maintenance. If we are to improve health for the population
as a whole, we must pay particular attention to individuals
with low literacy skills. Mohammed (1964) reported that the
well-informed adult generally participates in self-care more
effectively than the poorly-informed individual. Written
materials are often used to attempt to increase a client's
knowledge, but the effectiveness of the materials depends
upon the suitability of the reading level for that
particular individual (Mohammed, 1964). Readability
formulas can serve as tools to assess the reading level
required for comprehension of written materials, so
information can be more effectively communicated to the
client.

In the past, reading competency has been assessed by

7
equating it with the last grade of school completed (Doak
and Doak, 1980). However, a study by the National
Tuberculosis Association in 1956 showed little correlation
between education levels claimed by adults and actual
reading ability (Mohammed, 1964). The results of this study
indicate that even if printed material were rated by grade
level, the problem of matching the reader to the appropriate

material would remain.

Readability formulas for assessing text reading difficulty

Readability is often measured by one of over 40
different formulas for assessing the degree of difficulty of
reading material. Some of the more widely used formulas are

outlined below.

Lor e est Lorge (1939, 1948) was the first investigator
to base a readability formula on the McCall-Crabbs Standard
Test Lessons in Reading (McCall and Crabbs, 1925).
Originally, these passages were not designed to be used with
readability formulas, but were convenient for this use
because they were already graded for difficulty by the
authors on the basis of comprehensibility of the questions
at the end of each passage. Lorge designed his formula for
children in grades three to twelve and used the criterion of
50% correct answers to the questions following a passage of

the McCall-Crabbs test. After counting independent

8

variables in each passage, Lorge used multiple regression to
correlate the variables with grade level (R=.77) and derive
a formula. Lorge's sequence of steps for formula
development became very popular and similar steps were used
by Flesch (1948), Dale and Chall (1948) and SMOG (1969).

The procedure for using,the Lorge test contains the
following steps (Lorge, 1948):

a) Choose a random sampling of paragraphs from the

text to be tested (3 to 5 for an article, 25 to 30
for a book).

b) Count out the first 100 words in each sample.

c) Calculate the average sentence length in words for
the 100-word samples.

d) Count the number of prepositional phrases per 100—
word sample.

e) Count the number of words in each sample that are
outside the Dale list of 769 words (Dale, 1931).

f) Calculate the final score.

Grade Level = .06 x (average sentence length in words)

+ .10 x (number of prepositional phrases per 100 words)
+ .10 x (number of words outside the Dale list of 769
words per 100 words) + 1.99

Flesch test Grade level measurement that estimates the
reading difficulty of written material was used by Flesch in
his 1943 formula which he revised in 1948. He established
the Flesch formula based on the measurement of two
components of comprehensibility: reading ease and human
interest of the passage. By measuring human interest
factors, Flesch was testing the conversational quality of
the passage (Flesch, 1948). Other researchers had been

critical of the human interest factor in his 1943 formula,

9

claiming that this portion of the equation diluted the value
of the result (Klare, 1988). Therefore, in 1948 Flesch
separated his formula into two different equations, one for
measuring reading ease and another for determining human
interest. Based on the McCall-Crabbs Standard Test Lessons
in Reading, the reading-ease formula predicts the average
grade level of a child who could answer correctly 3/4 of the
test questions asked about a given passage.

According to Flesch (1948) the procedure could be
completed in six minutes:

a) Choose a random sampling of paragraphs from the

text to be tested (3 to 5 for an article, 25 to 30
for a book).

b) Count out the first 100 words in each sample.
c) Count the syllables in each 100 word sample.
d) Calculate the average sentence length in words for

the loo-word samples. It might be necessary to
count 95 or 105 words to make complete sentences
near 100 words.

e) Figure the number of personal words per 100 word
sample. Include: all pronouns referring to
people, all words with masculine or feminine
gender, and the words people or folks.

f) Figure the number of personal sentences per 100
sentences in all of the samples. Include: spoken
statements, questions, commands, requests,
exclamations and grammatically incomplete
sentences whose full meaning is hidden in the

text.
9) Calculate the final score.
Reading ease = 206.835 - .846 x (word length) — 1.015 x

(sentence length)

Human interest = 3.635 x (no. of personal words) + .314
x (no. of personal sentences)

10
Table 1: Flesch Reading Ease Scores

 

 

Readinq Ease Score Grade Level
90-100 5
80-90 6
70-80 7
60-70 8 and 9
50-60 10 to 12
30-50 13 to 16

0-30 college grad

A low score of zero is interpreted as virtually
unreadable text for the average reader or as text with no
human interest, whereas, a top score of 100 indicates that a
passage is easy-to-read (at the fifth grade level) or the
text is full of human interest (See Table 1).

According to Flesch (1948), the point value of 100
corresponds to the prediction that a person who has
completed the fourth grade would be able to answer 3/4 of
the test questions. This person is barely literate
according to the U.S. Census Bureau (Harman, 1987),
therefore, even text with a score of 100 might be a struggle
for the poor readers. The Flesch formula is valid for use
in assessing news reports, adult education materials,
government publications (Pichert and Elam, 1985), psychology
textbooks and advertisements (Flesch, 1948). The Flesch
test has recently been used in computer software programs

for readability measurement, such as Grammatik III

11
(Reference Software, Inc., 1989) or
Rightwriter (Que Software, Inc., 1992). According to Klare
(1974), the Flesch formula has become the most widely used

formula in the history of readability measurement.

Dale—Chall test Dale and Chall developed a formula for
use with general audiences after they reviewed the relation
of the various elements of reading difficulty: word length,
sentence length, percent of personal words or personal
sentences, the number of syllables, number of affixes,
number of prepositional phrases and the proportion of
difficult words mentioned in word lists (Spache, 1953).-
They found that the best predictors of reading difficulty
are the number of difficult words and the average sentence
length.

These researchers developed a word list of 3,000 words
known by 80% of American fourth graders. According to Dale
and Chall, the reason for the development and use of the
larger word list was to increase the discrimination of the
formula at higher grade levels (above grade eight), so the
formula could be used with high school as well as children's
texts. Like Lorge before them, Dale and Chall used 50%
correct on questions from the McCall-Crabbs Standard Test
Lessons in Reading as the criterion on which to base their
formula (Dale and Chall, 1948). The procedure consists of

the following steps (Dale and Chall, 1948):

12

a) Choose a random sampling of paragraphs from the
text to be tested (3 to 5 for an article, 25 to 30
for a book).

b) Count out the first 100 words in each sample.

c) Calculate the average sentence length in words for
the 100—word samples.

d) Count the number of hard words (all words not on
the Dale list of 3,000 words) in each sample.

e) Calculate the final score using the equation
below.

Reading Grade Score = .1579 x (number of words outside

the Dale list of 3,000 words) + .0496 x (average

sentence length) + 3.6365
The interpretation of the reading grade score can be found
in Table 2. The multiple~correlation coefficient of the two
formula variables (sentence length and number of difficult
words) with the criterion is .70. Dale and Chall (1948)
report that if they add the Flesch human interest factor,
the correlation coefficient rises only to .7025, a

negligible increase.

Table 2: Dale-Chall Reading Grade Scores

Formula Score Grade Level
4.9 and below 4 and above
5.0-5.9 5 to 6
6.0-6.9 7 to 8

7.0—7.9 to 10

8.0—8.9 11 to 12

9.0-9.9 13 to 15
10.0 and above 16 and above

13
Gunning-Fog In 1950, Robert Gunning developed a readability
formula using sentence length and difficulty of vocabulary
as factors of readability (Gunning, 1968). The procedure
for using the formula includes the following steps:

a) Take 3 samples of approximately 100 words each.
Count to the end of the last complete sentence and
note the exact number of words in each passage.

Laubauch and Koschnick (1977) suggest avoiding sampling of
titles, headings, vertical lists and dialogue unless it is
typical of the content of the text.

b) Divide the number of words by the number of
sentences and round to the nearest tenth. This is
the average sentence length.

c) Count the total number of hard words in each
sample. A hard word is defined as any word of 3
or more syllables excluding proper names,

numerals, initials, acronyms, abbreviations or two
syllable words with the endings: s, es, ed, er,

est, ly.
d) Calculate the percentage of hard words.
Fog index = (percent hard words + average sentence length) x
0.4

According to Laubauch and Koschnick (1977), this
formula was developed for use with adult materials. Today
it can be found in several computer software programs
designed to measure readability such as Rightwriter (Que
Software, Inc., 1992) or Grammatik III (Reference Software,

Inc., 1989).

14

Spache test Like Dale and Chall, Spache chose average
sentence length as a predictor of reading difficulty
(Spache, 1953). He used also the Dale list of 769 words,
judging all words outside the list as hard words. Spache
(1953) developed an equation using multiple regression
techniques which determined the combination of these two
factors needed to estimate reading grade level. The
procedure for this test is as follows (Spache, 1953):

a) Choose a random sampling of paragraphs from the

text to be tested (3 to 5 for an article, 25 to 30
for a book).

b) Count out the first 100 words in each sample.

c) Calculate the average sentence length in words for
the loo-word samples.

d) Count the number of hard words (words outside the
Dale list of 769 words) per 100—word sample.

e) Calculate the final score using the equation
below.

Grade level of text: .141 x (average sentence length
per 100 words) + .086 x (words outside the Dale list of
769 words) + .839

An answer of 2.3 would designate a text equal in
difficulty to a level of second grade, third month. When
using this formula there is no need for a table to interpret
the results.

Spache (1953) reports that the multiple—correlation
coefficient obtained by combining sentence length and
percent of hard words to predict grade level of texts is
.818, one of the highest correlations found between a

readability formula and the criterion. According to Klare

15
(1974), the Spache formula is valid for use with children’s
materials for grades one to three, but has not been

validated for use with higher reading level materials.

Fry graph With the publication of the Fry graph in 1968,
readability formulas became available to greater numbers of
people (Vaughan, 1976). The Fry graph (Appendix A) permits
a direct estimate of reading grade level using two
variables: number of syllables per 100 words and sentence
length, but avoids time consuming formula calculations
(Klare, 1988). According to a University of Arizona Study,
the Fry scores consistently agree with those of the Dale-
Chall formula (Vaughan, 1976), but the Fry formula has the
advantage of being more simplistic (Fry, 1968). Both tests
are based on prediction criteria of 50-70% comprehension at
the grade level scored (Vaughan, 1976) and Fry reports
accuracy within a grade level (Fry, 1968). The procedure

for using this formula is as follows (Fry, 1968):

a) Select three 100 word passages from the beginning,
middle and end of the text, skipping all proper
nouns.

b) Count the total number of sentences in each 100

word passage to the nearest tenth of a sentence.
Average these three numbers.

c) Count the total number of syllables in each 100
word sample. Average the total number of
syllables for the three samples.

d) Plot on the Fry graph (See Appendix A) the average
number of sentences per 100 words and the average
number of syllables per 100 words. The graph will
identify the grade level of reading difficulty.

16
Due to its simplicity and accuracy below the sixth grade
level, the Fry formula is used by the Adult Performance
Level Project and Literacy Volunteers of America, Inc. to

evaluate written text (Laubach and Koschnick, 1977).

SMOG According to McLaughlin (1969), the SMOG
readability test is simpler and easier to use than the Fry
graph. Freimuth (1979) listed the advantages of this
procedure as being useful for a general audience, applicable
without a computer, only a two variable equation and no word
list is required. The procedure for using this formula is
written below (McLaughlin, 1969):

a) Count ten consecutive sentences near the beginning

of the text, ten near the middle and ten near the
end. Use these three samples for reading

assessment.

b) Count every word (in the 30 sentences) of three or
more syllables.

c) Estimate to the nearest whole number, the square

root of the number of polysyllabic words counted.

d) Add three to the approximate square root. This
gives you the SMOG grade, which is the reading
grade that a person must have reached to
understand the text.

SMOG grade = 3 + the square root of the number of
polysyllabic words
The author of this test reports that this formula will
predict the reading grade level of a text correctly within
one and one—half grades in 68% of the cases. It takes only
nine minutes to derive a SMOG grade based on a sample of 600

words, whereas, it takes the same time to find a Dale—Chall

17
prediction using a sample of only 100 words or a Flesch
score based on two loo-word samples (McLaughlin, 1969).

Freimuth (1979) reports that the SMOG formula was
tested on 64 university students using eight 1000—word
passages from a variety of periodicals. After reading the
passages, each student was tested for unaided recall to
measure comprehension. After the scores were controlled for
speed-reading, a perfect negative correlation was found
between polysyllabic word counts and the measures of reading
efficiency.

Published in 1969 by McLaughlin, the easy-to-use SMOG
formula produced scores that did not agree with the results
of the Fry graph or the Dale-Chall test. A University of
Arizona study (Vaughan, 1976) found that SMOG grades tended
to be two grades higher than the results of either the Fry
graph or the Dale-Chall test. However, results of a
Spearman rank correlation clearly show that the three
formulas are highly correlated. Vaughan (1976) reported
that there are differences in what the tests predict. The
Dale-Chall and Fry tests are based on prediction criteria of
50-70% comprehension at the grade level scored, whereas, the
SMOG formula is based on a predictive criterion of 90-100%

comprehension at the grade level scored.

18

Table 3: Characteristics of Readability Formulas

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Formula Year Variables Criterion2 Reggired
Measured1 word list

Lorge 1939 $1, pp, dw 50% correct yes

Flesch 1948 wl, $1 75% correct no

Reading-

ease

Dale-Chall 1948 dw, 51 50-70% yes
correct

Gunning- 1950 dw, 81 not no

Fog reported

Spache 1953 dw, sl not yes
reported

Fry graph 1968 sy,sl 50—70% no
correct

SMOG 1969 psy 90—100% no
correct

 

1. The variables measured are abbreviated as follows:
sl=sentence length, wl=word length, pp=number of
prepositional phrases, dw=number of difficult words,
pw=number of personal words, ps=number of personal
sentences, sy=number of syllables and psy=number of
polysyllabic words.

2. The criterion used for all formulas in the table is a
percentage of correct answers on the McCall-Crabbs Standard
Test Lessons in Reading. -

Therefore, if a text is given a Fry score of fifth
grade at a 50-70% comprehension level and a SMOG grade of 7
at a 90-100% comprehension level, both reading grades might
be correct. SMOG grades might disagree with the Fry graph
and the Dale—Chall test but they are not inaccurate

(Vaughan, 1976).

19
Cloze Procedure This test is not a formula, but the
results are known to conform to the results of the Flesch
test and the Dale-Chall test for estimating readability
(Taylor, 1953). The procedure requires deleting every fifth
word in a passage and asking respondents to fill in the
deleted words. The percentage of exact replacements is
considered to be the readability score (Farkas et al.,
1987). Taylor (1953) reported that to be successful at
filling in the blanks, the reader must know the meanings of
different combinations of letters and not just recognize
words. This procedure is commonly used and takes into
account factors that other formulas ignore, such as previous
knowledge or confusing sentence structure (Taylor, 1953).
Taylor cautions it is necessary to strike out enough words
when using this test so that the blanks will represent
proportionately all kinds of words to the extent that they
occur.

The Cloze test is superior to the use of multiple
choice exams because multiple choice questions that are
related to the passage being measured, are highly subjective
and the number of questions possible for a ISO-word passage
is less than the 30 blanks allowed by the Cloze procedure
(Sticht, 1975). Sticht reports also that because the Cloze
test has consistently shown high correlations with multiple
choice exams, it is a valid measure of reading

comprehensibility. Pichert and Elam (1985) recommend the

20
use of the Cloze Procedure for measuring the readability of

health related materials.

Considerations and Concerns of Using Readability Formulas

Various researchers have offered the following
suggestions for consideration when using readability
formulas (Klare, 1974; Freimuth, 1979; Klare, 1988; Zakaluk
and Samuels, 1988):

* . Remember that different formulas give different
grade level scores based on different percentage
levels of comprehension for the same piece of
text. Know what the formula is measuring.

* Use a formula only as a screening device.

* Choose a formula with two variables, one greatly
decreases predictiveness and three are very time
consuming to measure.

* Formulas are poor predictors at high grade
(college) levels.

* Be aware of other variables that might affect
comprehension (motivation, previous knowledge,
etc.) that a formula does not measure. For
example, variables such as previous knowledge can
be tested by administering exams, whereas,
motivation might have to be estimated by using a
Likert scale or might have its effects limited by
use of a randomized population.

* Formulas do not indicate how to write for maximum
comprehension.

Although readability formulas are as popular as ever
(Fry, 1988), Pichert and Elam (1985) caution that the use of

readability formulas should be controlled by three criteria:

21

a) The formulas should be supplemented by other means
of judging the quality of the printed material.

b) The readers for whom the text was intended should
be similar to those on whom the formula was
validated.

c) The text to be assessed should have been written

without reading ability formulas in mind.

Pichert and Elam (1985) added that educators cannot assume
that a patient can understand a text just because its
readability score matches the reader’s grade level. Prior
knowledge and motivation of the reader can affect
comprehension of the material.

Many formulas (Flesch, Fry, SMOG, Dale-Chall, Lorge)
are based on the McCall—Crabbs test lessons. These tests
were poorly standardized on a small group of New York City
school children (Stevens, 1980; Zakaluk and Samuels, 1988)
and were designed only as practical exercises in reading
(Pichert and Elam, 1985), not as a criterion for formulas.
Grade scores of the passages and questions were determined
by matching scores with student performance on the
Thorndike-McCall Reading Scales Performance Test. Despite
the shortcomings of the test lessons, until a more valid
criterion is proposed, the McCall-Crabbs Test Lessons in
Reading are the best criterion available (McLaughlin, 1968).

Irwin and Davis (1980) list other shortcomings of
reading formulas, admitting at the same time that no other
good replacements are available. Formulas do not examine:

the relationship between the reader’s background and the

22
difficulty of the text, the new ways in which concepts are
introduced, and the motivation or the organization of the
material.

In addition to the general problems that accompany
formula use, many formulas have specific drawbacks. The
Dale-Chall formula uses a list of 3,000 words that are
familiar to fourth graders, but are words unfamiliar to
fourth graders necessarily the same words that are
unfamiliar to adults (Pichert and Elam, 1985)? Many of the
formulas have not been validated on groups of low-literate
adults and might not allow for differences in socio-economic
status. The Fry test has been validated on tenth grade
students and novels they had read in English class, whereas,
the SMOG test was validated on University students using
adult periodicals (Pichert and Elam, 1985). However, the
Flesch formula is one of the equations that has been
validated using adult education materials among other
written works.

Although readability formulas are not perfect
predictors of reading difficulty of text, they do have
several advantages over other comprehension tests. Written
comprehension tests can be very time consuming and require a
willing population, whereas, readability formulas can be
applied straight to a text, are quick, easy-to-use, widely
recognized, highly reliable and valid for use with a

specific population or particular materials. According to

23
Klare (1974), formulas have their maximum predictive
validity when used for the purpose for which they were
designed. The applicability of the formula lies with the
user who must understand how the formula was developed, what
it measures and to what degree of accuracy the grade level

is predicted to correctly understand the results.

Measuring the reading levels achieved by the individual

The Slosson Oral Reading Test (SORT) This test attempts to
measure reading achievement and was designed to be
administered individually (Slosson, 1963). It is based on
the ability to verbalize written words at increasing levels
of difficulty. An individual is given a list of twenty
words at a known grade level to read (Appendix B). He/she
pronounces the words on the list to the best of his/her
ability and then advances to the next grade-level list until
he/she finishes the high school level list. An error is
made if a word is mispronounced, omitted or takes more than
five seconds to pronounce. The raw score is the total
amount of words pronounced correctly, whereas, the reading
level is half of the raw score divided by ten. For example,
if an individual pronounced 42 words correctly and,
therefore, had a raw score of 42, then his/her reading level
would be 2.1 or second grade first month.

This test was validated using the Standardized Oral

24
Reading Paragraphs (Gray, 1960) and a correlation
coefficient of .96 was obtained. Slosson claims that this
test can be given to an individual and scored in less than

three minutes.

Factors affecting readabilty that are not measured by

formulas

The purpose of readability formulas is to give an
indication of the reading grade level of a text in order to
match the text with the ability of the student (Lorge,
1939). Used alone, however, the formulas are not sufficient
for assessing readability (Farkas et al., 1987). Emphasis
should not be placed on sentence and syllable length without
considering the impact of variables such as: background
knowledge, motivation and other meanings which might be

derived from the material (Pichert and Elam, 1985).

Background knowledge Research by Recht and Leslie
(1988) showed that previous knowledge can greatly increase
the amount of information recalled from a passage and can
even compensate for poor reading ability. In an
investigation of the effect of prior knowledge on recall of
information, 64 seventh and eighth graders were tested for
their knowledge of baseball and ability to read a passage

about a series of plays in an inning. The students were

 

 

.‘Av

u...
.

 

,
on...
- I-

“~ ..-

up.v

5w».
.

.
\“ ‘9

"Owcc

‘o.

.,.'

 

-
.‘l.'

.
n
1

 

a. .
‘5
‘~
‘~.
.-
~‘ ‘5
u ‘
u-

 

 

 

I
‘.\, .
\

\

.

.0

.

\

25
divided into four groups: good readers with good knowledge
of baseball, good readers with poor knowledge of baseball,
poor readers with good knowledge of baseball and poor
readers with poor knowledge of baseball. Good readers were
defined as students who scored above the 70th percentile and
poor readers scored below the 30th percentile on the
Scientific Research Association Achievement Comprehension
Test (SRA Achievement Series, 1971).

Students were asked individually to read a fifth grade
level passage about an inning of baseball, then complete the
following three tasks. First of all, they were to move
figures around a board to demonstrate the action about which
they had just read. After the reenactment, students were
asked to summarize the text in their own words and thirdly,
they were to rank 22 sentences according to their importance
to the main ideas of the passage. Students were given _
points for each piece of information recalled correctly.

The researchers found greater recall scores of high
knowledge readers compared to low knowledge readers
(p<0.01). The reading ability of the students had no
significant effect on recall in this study.

In a similar study, (Taylor, 1979) evaluated the recall
of two groups of third and fifth grade children after
reading two passages. One passage was about a topic
familiar to all students, whereas, the second passage was

about an unfamiliar subject. The group of poor readers

 

 

«Au-
§

bung.

 

 

.
.v.i

‘ '
.'c.

«
-

 

26
consisted of third and fifth graders reading at the third
grade level and the good readers were the children from
either grade who were reading at the fifth grade level.
Regardless of reading level, all groups recalled more
details from the familiar passage than from the unfamiliar
one with good readers recalling more information than poor
readers from both passages. Taylor concluded that poor
readers’ comprehension suffers the most when background
knowledge is low.

Zakaluk and Samuels (1988) reviewed various research
articles that investigated the role of background knowledge
in reading retention. They found some similar results among
the studies:

a) Subjects familiar with a topic recalled more ideas
from the text than students unfamiliar with the
topic.

b) Well written text signals to the reader what
background knowledge must be used to process the
information.

c) Readers are able to construct meaning from
background knowledge which allows them to make
inferences and fill in missing ideas when a lapse
in recall of the text occurs.

Motivation Motivation to read a piece of text can
increase reading comprehension. When a reader’s interest is
high, the text might become more comprehensible to the
reader. For example, a fifth grade reader and avid football

fan might willingly read and comprehend details in a seventh

grade level book about the Detroit Lions, but not understand

27
a word of his fifth grade level science text.- As motivation
to read increases, readability formulas tend to overestimate
the difficulty of the text and when interest is low there
might be underestimation (Pichert and Elam, 1985). Fry
(1964) adds that a highly motivated student might have the
interest to work through some very difficult material,
whereas, Nitzke (1986) warns that even simple, readable
language cannot make up for a boring topic. After examining
36 experimental studies of the effect of different variables
upon reading comprehension, Klare (1976) concluded that
motivation can sometimes override the effect of readability
on comprehension. Therefore, raising motivation and keeping
it high are extremely important factors in teaching.

Klare (1976) found also that gain attributed to
improved readability was greater for a topic lower in
interest value than for one higher in interest value. This
finding is due to the observation that interest increases
readability without changing the reading difficulty of the
text. Klare concluded that it is more worthwhile to improve
the readability of low-preference material compared to high-

preference material.

Underlining The technique of underlining words or
phrases is often used to aid in the recall of information
for low-literate reader (Doak et al., 1985), but has been

shown to have only a limited effect on retention. Johnson

28
(1988) conducted a study to investigate the usefulness of
underlining. Two groups of below average college freshman
(those first year students who scored below the 16th
percentile on the Nelson-Denny Reading Test) were asked to
read a passage written at the seventh grade level. One
group was asked to underline one sentence per paragraph that
they thought contained the main idea of the passage,
whereas, the other group was asked to simply read the
passage. After both groups took a multiple choice exam, it
was discovered that underlining did not improve overall
retention, but recall of the specific underlined sentences
was increased in the experimental group compared to the
control group. Therefore, underlining might be used with a
particularly important word or phrase to increase retention
of the isolated message.

In other studies investigating underlining, Idstein and
Jenkins (1972) found no significant differences in the
recall of information when pieces of text were underlined or
when the same passages were read repeatedly by a group of
college students. Furthermore, there were no significant
differences in retention when the reader actively underlined
the topic sentences from the text or when the experimenter
underlined the main ideas (Rickards and August, 1975).

These researchers found no differences in recall of either

the main or secondary points in the text.

29

Pictures Another factor affecting comprehension of
a text is the use of pictures. Illustrations and diagrams
have been shown repeatedly to have a positive effect on the
retention of information . In one study (Tajika et al.,
1988), drawings or photographs pertaining to a passage
(called pictorial organizers) were shown to a group of fifth
graders before they read the passage. A second group of
similar students only read the passage. Recall of
information was tested immediately after reading and one
week after reading the text. The results showed that
students given the drawing or photographs retained
significantly more information both immediately and after
one week compared to their classmates not shown the
pictures.

Mayer (1989) reviewed several studies where diagrams
were used to highlight the major objects and actions of a
piece of text. He found the following similarities among
study results: Students who used diagrams recalled more
conceptual information and scored lower on verbatim
retention than students not exposed to diagrams. Students
exposed to diagrams were more likely to answer problem
solving questions correctly. Diagrams were most useful if
placed before or within a lesson, but not after the relevant
piece of text. Finally, diagrams are most useful to
students of low aptitude or with low prior knowledge.

Irwin and Davis (1980) developed a checklist (Appendix

y~-..

II“

 

.
'qun.

_
..._
.
-~-
_\

§-...q
a..._
-

n.

- v-
... ‘

 

O

“.~

\‘VN.
.

‘\

 

 

 

 

30

C) of 36 items as an alternative or supplement to using
readability formulas. The list estimates the readability of
the text with each item being rated on a scale of one to
five. The items are divided into two main categories:
understandability and learnability. Learnability is further
divided into organization, motivation and reinforcement.
The researchers report that these factors are important in
retention. However, in this report there were no statements
about validity or reliability of the checklist and all
scores were subjective.
The problem of unreadable health-related materials

Several studies have indicated that low-income, low-
literate individuals in need of health and nutritional
instruction are the people least likely to understand the
message in written health-care material. At the Public
Health Services Hospital in Norfolk, Va., the Cloze
procedure was used to test comprehension of patients. This
test showed that approximately 50% could not read at a fifth
grade level (Doak and Doak, 1980). Written materials
covering subjects such as diabetes, family planning, infant
care, prescriptions and clinical procedures were found to
have few instructions below an eighth grade reading level.
Amazingly, the surgical consent form at the same
institution, had a required reading ability of 16th grade

(college senior level).

31

In a similar study, Doak et al. (1985) examined a
sample of 291 sets of health instructions for reading
difficulty level. These researchers found fewer than 10% of
the materials with a seventh grade or lower difficulty
level, leaving them to conclude that virtually none of the
instructions would be understood by individuals with low-
literacy skills.

In another study, Samora et al. (1961) tested 125 low-
income, hospital patients for the correct definitions of 50
common health-related words. Less than 79% of this
population had a high school education. The researchers
found that younger patients could identify more words than
could individuals over 65 years of age. Four words (vomit,
relieve, appointment, constipated) were recognized by 90% of
the test population, whereas, four other words (tendon,
terminal, malignant, nerve) could be defined by only 25% of
the patients despite the syllabic similarities between the
two groups of words.

Lanese and Thrush (1963) have used the Dale-Chall
formula to estimate the readability of materials written for
diabetic patients. The researchers investigated the
effectiveness of diabetic materials in the form of
pamphlets, booklets or reprints distributed by clinics,
hospitals and health agents. These researchers sampled the
literature of 21 teaching hospitals across the country and

computed the median grade level and found it to be slightly

 

.-

.nv ‘

. .

no no.
.—

C. a...

s
on .7".

-"vv.

inc. ..

"'.|

Van... .

-~u-_

 

..
" v.-

~ ‘.
‘1‘..

1..“

'(1

32
above the ninth grade in reading difficulty. Yet half of
the diabetic population in 1962 had not completed nine years
of education. They concluded that the reading levels of the
material were beyond the reading levels of the patients,
therefore, comprehension and recall would decrease and
motivation for acquiring additional information would be
reduced.

Problems with the readability of health-related
materials are as much a concern today as in the past.
Recently, Meade et al. (1989) estimated the reading
difficulty of 49 materials on smoking education using a
computer software program developed by the University of
Minnesota. The program is able to compute grade levels of
reading text by using the following formulas: Dale-Chall,
SMOG, Gunning-Fog, Fry and Flesch. The researchers compared
the readability of materials (median=10th grade) to the
reading levels of smoking patients at a primary care clinic
(median WRAT score=6th grade). These researchers concluded
that a serious disparity existed between the reading level
of the population and the skill level required to comprehend
smoking materials.

In another study, Freimuth (1979) used the SMOG test to
measure reading grade level of health-related articles of
popular magazines and pamphlets written for patients. The
researchers found that the articles ranged in difficulty

from a ninth to a fifteenth grade (third year college)

own.

bu¢o

..- I!
.o.‘

..‘ _‘

I"

'l!

‘-

'v

33
reading level and the pamphlets ranged from an eleventh to a
fourteenth grade (second year college) reading level.
Freimuth concluded that patients who do not have at least
two years of high school might have difficulty in
understanding the written health material, adding that the
technical nature of the subject of health might increase the

number of polysyllabic words and inevitably increase the

SMOG score.
Teacher-produced materials According to Klare

(1974), readability formulas do not indicate how to write
readably. Therefore, teachers often take guesses, rely on
past experience or listen to feedback to estimate how to
write appropriate material for a particular audience. Rules
have been developed by researchers to guide educators when
writing for ease of readability, but caution must be taken
so structure is not emphasized at the risk of content. When
preparing materials for low-literate individuals sentences
should be kept as short as possible. However, more
importantly, the vocabulary load should be kept to a minimum
because it is the greater obstacle to readability (Lanese
and Thrush, 1963). The writer can control this factor by
using a vocabulary list of familiar words such as the Dale
list or Thorndike-Lorge List (Thorndike and Lorge, 1944).
Doak et a1. (1985) recommend the use of visuals for

three reasons. Simple illustrations can decrease the amount

34
of necessary reading in the text, emphasize the important
instructions and provide motivation. These researchers
report that both simple line drawings and photographs might
be useful to the low-literate reader. However, photographs
were found to be more effective when the reader has control
over his viewing time, whereas, simple line drawings are a
greater asset when time is controlled by the teacher,
television or filmstrip.

Researchers have outlined several strategies for
designing written materials for the adult with low-literacy
skills (Sticht, 1975; Klare, 1976; Laubach and Koschnick,
1977; US DHHS, 1982; Doak et al., 1985; Fry, 1988; Nitzke,

1989).

To improve the readability of the text:

* Use simple language. Avoid jargon, technical
vocabulary, polysyllabic words, abbreviations and
long explanations.

* Use circles or underline for emphasis as opposed
to using italics or bold print (Doak et al.,
1985). However, overuse of circles and lines
decreases the attention value (Laubach and
Koschnick, 1977).

* Emphasize the desired action.
* Use the active voice (Klare, 1976).
* Use words instead of symbols, such as "add"

instead of "+" (Doak et al., 1985).
* Summarize the major points (US DHHS, 1982).

* When appropriate use headers to give readers clues
to the message.

35

* Write instructions in the sequence they are to be
performed.
* Present the minimum information essential for the

adult’s needs (Sticht, 1975).

* Keep sentence length as short as possible (Klare,
1976; US DHHS, 1982).

* Avoid using words beginning with pre-, dis- or
multi— (Fry, 1988).
* Choose a large type size. Doak et a1. (1985)

recommend 12-14 point.

* Use the same type style throughout the text. A
Roman style with serifs is recommended to help the
letters blend together into words.

* Use lower case letters instead of capitals because
the shapes of lower case letters are easier to
recognize (Laubach and Koschnick, 1977).

* Use personal pronouns rather than impersonal.

* Terms presumed to be new should be defined when
used for the first time (Fry, 1988).

* Keep the subject as near the verb as possible
(Nitzke, 1989).

To improve the illustrations:

* Use simple line drawings or photographs (Doak et
al., 1985).
* Show drawings or photographs from the reader’s

point of view (Sticht, 1975).

* Make drawings lifelike; avoid cartoon figures.

* When drawing organs of the body, draw as much of
the body as possible for easy identification of
the organ (Doak et al., 1985).

* Place visuals near the corresponding text. Avoid
the use of arrows to portray the message (Sticht,
1975; US DHHS, 1982).

* Illustrate only the correct behavior.

36

* To draw attention to major points, use circles,
arrows, or a little color.

* Too many colors are distracting. Keep the variety
to two or three colors per page. Nitzke (1986)
recommends using bright colors.

* Step by step instructions using pictures should be
on one page or on opposing pages (Sticht, 1975).

To simplify page format and organization:

* Use black print on a white background rather than
white on black. The latter combination reduces
reading speed by 10% (Doak et al., 1985).

* Organize information into categories to improve
recall (Doak et al., 1985).

* Use titles and subtitles to clearly define the
organization of the text, but keep headers simple
(US DHHS, 1982).

* Use plenty of white space to decrease eye
distractions (Nitzke, 1989).

Before using any text or illustrations with clients,
the materials should be pretested with a sample population.
According to the U.S. Department of Health and Human
Services (1982), the purpose of pretesting is to find out if
the message attracts attention, is understood, is personally
relevant to the audience, is believable or is offensive.
Doak et a1. (1985) add that pretesting addresses two
factors: comprehension and interest. They caution,

however, that self-administered pretests carry little

validity when used with a low—literate audience.

37

Learner-produced materials A major cause of failure to
achieve the desired goals of nutrition education is the poor
comprehension of instruction (Doak and Doak, 1980).

However, strategies can be used to eliminate the gaps
between instruction and comprehension. One of these
strategies is the Language Experience Approach (LEA).

During the 1960’s, LEA was designed to teach reading to
children and adults by using the student’s language, words,
sentences and stories as the basis for content (Pickering,
1989). According to Heller (1988), standard LEA includes
introduction of a story topic by the teacher or student that
relates to the previous experience of the student. The
learner contributes information about the topic as the
teacher transcribes the text on the chalkboard. Then the
teacher reads the new passage to the student, repeats the
passage and reads the passage with the student until the
student is able to read his/her creation on his/her own.
Kennedy and Roeder (1975) suggest that students’ statements
might be tape recorded for later transcription. The
recording method allows students to talk spontaneously at a
normal rate without waiting for the transcriber to write
down each word.

The advantages of this technique are that it produces
text from the reader’s own experience, combining the oral
and written language of the student (Stahl and Miller, 1989)

and it improves the learners’ self-concept and motivation

38
(Aderman et al., 1987). Harman and Edelsky (1989) report
that language approaches are not as smothering to poor
readers because the culture and language of all people are
welcomed into the classroom. The teacher learns to share
the control over the material with the students (Shannon,
1989), empowering the student to take control over the
learning process. Kazemek (1984) reports that adult
education is more ethical when people are able to express
themselves. The educator must ask people about their
interests, knowledge and beliefs, not simply make
assumptions about what needs to be taught.

Since the 1960’s, LEA has been used to teach a wide
variety of topics. Children have learned to analyze
television commercials by restating the main ideas and
advertising claims in their own words (Allen et al., 1988).
Doak et al. (1985) support the use of this approach to teach
hospitalized patients of low literacy levels. These
researchers claim that three factors interact to support
comprehension: logic, language and experience. The
patients must find that the language used to explain the
health professional’s instructions is close to their own
vernacular so there are fewer barriers to learning.

Nitzke (1987) used an adaptation of the LEA to revise a
nutrition pamphlet, and at the same time, increased the
clarity and personal relevance of the writing. The target

audience consisted of low-income, low-literate, white,

39
central-city mothers of children under twelve years of age.
During the needs assessment phase of the project, the
mothers expressed concerns over the cost of food and the
adequacy of their children’s diets. Nitzke addressed the
latter issue using two pamphlets on the subject of healthy
snacks.

One pamphlet was produced by mothers using the
adaptation of the LEA. These women participated in a two
hour nutrition education program where they were presented
with snack information through a series of lectures,
discussions and demonstrations. After each series there was
an LEA experience where participants restated what they had
learned. Statements were transcribed, organized and edited
by the educators. A second pamphlet was produced by
nutrition educators using a standard method. The pamphlets
had the same snack information and layout, but different
words and both pamphlets were measured at a fifth grade
readability level by several readability tests.

The results showed that the pamphlet written by the
target audience was more difficult to read, but its message
was at least as well understood and more personally relevant
to readers than the pamphlet written by nutrition educators.
Nitzke suggests that readers are less likely to reject
information expressed in peer language.

Twenty-six women outside the original test group were

asked to evaluate the pamphlet. The pamphlet written by the

40
target audience was rated as more neighborly, helpful,
sensible and informative than the nutrition-educator
version. Furthermore, the target audience pamphlet was
rated as appropriate for a racially mixed group of mothers,

even though the producers were all white.

41

Nutrition Education in Pregnancy

It is often difficult for pregnant women to provide
adequate nutrition to the fetus while maintaining an
appropriate weight, especially if the woman is at risk for
nutritional problems (Trouba et al., 1991). Risk factors
include: age (less than 2 years post—menarche or over 35
years), high parity, poor obstetric history, faddist food
habits, substance abuse, therapeutic diets, high or low
maternal weight before pregnancy, anemia, glucosuria and
inadequate or excess weight gain during pregnancy (Trouba et
al., 1991). Nutrition education programs attempt to address
these risk factors before and during pregnancy to improve
dietary intake, health practices and weight gain to
influence the outcome of pregnancy. This portion of the
review of literature will describe nutrition education
programs with a focus on the Expanded Foods and Nutrition
Education Program (EFNEP).

Kafatos et al. (1989) investigated whether prenatal
nutrition counseling in rural Greece was associated with
improvements in diet and weight gain by analyzing dietary
habits, nutrient intake, hemoglobin (Hb) levels, and serum
vitamin concentrations in both intervention and control
groups. The intervention consisted of home visits by nurses
every two weeks to 300 village women. Instruction was given

on the basics of nutrition during pregnancy and techniques

' .
.Av

.U. i

‘9"...

h
01..-.
“D.-

“n...

An

(..-V‘

~
‘v-Q -
‘u.

‘

....~‘

- a.
N~ ~\
5“”

 

'(I

42
for improving the diet. Women were encouraged to prepare
and preserve locally grown foods. Dietary habits were
assessed by using a 24 hour recall and a food weighing
inventory at the beginning of the study and every four weeks
until delivery.

The results showed that in the intervention group,
kilocalories increased significantly throughout pregnancy
(p<0.0001), protein, fat and carbohydrate intake increased
significantly (p<0.001) and mean weight gain increased at 32
and 37 weeks (p<0.05) when compared to the control group.
The researchers concluded that nutrition education
intervention can have a positive effect on maternal
gestational weight gain.

Researchers have found that nutrition education
programs that focus on an increase in knowledge do not
necessarily have a positive change on attitudes or behaviors
(Kaplowitz and Olson, 1983). Kaplowitz and Olson provided
pamphlets on breast-feeding to literate, low-income,
pregnant women and assessed knowledge on the subject before
and after reading the material. Knowledge of breast-feeding
significantly increased , but no change was found in
positive attitude or behavior towards the practice. These
researchers suggest that to be successful, future programs
should address individual attitudes and experiences.

Brush et el. (1986) support the view that educators

must not only teach nutrition knowledge, but tie the

43
knowledge to the experiences of the learner. These
researchers improved post-test scores of adults by
encouraging learners to relate new knowledge to their own
experiences. The Language Experience Approach is one
technique which can be used to tie knowledge to experiences
by allowing the learner to express his/her own views in

his/her own language.

The Expanded Food and Nutrition Education Program
Established in 1968, the Expanded Food and Nutrition

Education Program (EFNEP) is a national program funded by
the U.S.D.A. and state legislatures. Within each state,
EFNEP is administered through the Cooperative Extension
Service of the state land grant university and is
implemented at the county level by paraprofessional
nutrition instructors under the direction of the County
Extension Home Economist. Each state is responsible for
designating geographic areas in which the program will
operate. Coordinators of the program provide training and
develop or obtain educational resources for the instructors.
County Extension Home Economists provide training to
paraprofessionals who teach low-income homemakers within the
county.

Michigan EFNEP is administered from Michigan State
University and serves 15 counties statewide. The goal of

EFNEP is to help families, especially those families with

44
young children, living at or near the poverty level to
acquire the necessary skills and knowledge to achieve a
nutritionally balanced diet (Michigan Expanded Food and
Nutrition Education Program, 1990). During 1990, the
program reached 4,730 homemakers with 9,155 children in 15
Michigan counties. A total of 91% of these homemakers had
annual incomes below $13,000 and 47% were from minority
groups (Michigan Expanded Food and Nutrition Education
Program, 1990).

Adult participants in EFNEP are taught lessons in basic
nutrition, food purchasing, food storage, sanitation and
management of available resources. Families are informed
how to apply for food stamps and how to obtain donated food.
Homemakers are taught in small groups or individually by
paraprofessionals who have been trained by the County
Extension Home Economists and State Extension Specialists.
The curriculum consists of "Eating Right is Basic" a series
of handouts and flipcharts used with all adults, "Eating
Right ..... for Two" used with pregnant homemakers, "Feeding
Your Infant" and "Feeding Your Baby Solid Food" used with
mothers of young children. All of these materials were
developed in Michigan in 1985 with the assistance of a
national advisory committee and revised in 1992. They are
currently used in 47 states, Guam, Canada and the Virgin
Islands. Although the nutritional information in these

materials was thoroughly evaluated for appropriate content,

45
no documentation exists as to the reading grade levels of
the materials.

In addition to the Michigan EFNEP curriculum, other
states have developed materials for use with pregnant
clients. A group of educators at the Pennsylvania State
University EFNEP program developed a special curriculum
designed for use with pregnant teens. "Eating for a Better
Start" addresses adolescent lifestyle characteristics which
might adversely affect the outcome of pregnancy: skipping
meals, alcohol and substance use or abuse, fad dieting and
nutritionally poor food choices (Strycher et al., 1991).
Only the Michigan materials, "Eating Right...for Two," will

be used in this research.

46

Nutrition in Pregnancy

Adequate nutrition during pregnancy is essential for
optimum health for the mother and the fetus. Poor nutrition
during pregnancy has been associated with an increased risk
of prematurity, infant mortality, pregnancy-induced
hypertension (PIH) and anemia (Endres et al., 1981). This
portion of the literature review focuses on the major
nutrition-related areas which might affect pregnancy
outcome: low birth weight and premature birth, maternal
weight and weight gain, smoking, drug use, nutrient intake,

cravings, aversions and pica.

Low birth weight and premature birth In the U.S. in the

early 1900’s, the infant mortality rate was 100 infant
deaths in 1000 live births with two thirds of those deaths
occurring in the post-neonatal period from 28 days to one
year of life (McCormick, 1985). Many of those deaths were
blamed on environmental factors, therefore, research focused
on prevention of infectious diseases. Since 1950, infant
mortality rates in the U.S. have declined to 10.1 deaths per
1,000 live births (National Center for Health Statistics,
1990). Due to the increased prevention of infectious
diseases and improved nutrition most infant deaths now occur
earlier, in the neonatal period from 0 to 28 days of life.

The majority of deaths occur to low-birth weight (LBW)

47
babies who are at increased risk of infant mortality and
morbidity compared to their normal weight counterparts
(Abrams, 1991). There are two major categories of LBW
infants, preterm (those babies born at a gestational period
of g 37 weeks) and small for gestational age (SGA, those
babies born at a weight 5 25009).

Researchers in the area of LBW have repeatedly found
significant correlations between several factors and the
risk of having a SGA or premature baby: low maternal weight
gain (Mitchell and Lerner, 1989; Doyle et al., 1982; Mora et
al., 1979), low maternal prepregnancy weight (Abrams 1991;
O’Sullivan et al., 1965), high parity (Kleinman and Kessel,
1987), short intervals between pregnancy (Abrams et al.,
1989), incompetent cervix (Abrams, 1991), smoking ten or
more cigarettes per day (Haworth et al., 1980; Gormican et
al., 1980; Meyer et al., 1976; Beal, 1971; Mulcahy et al.,
1970), use of illicit drugs (Abrams et al., 1989), history
of loss during pregnancy (McCormick, 1985; Meyer et al.,
1976); unmarried status (Kleinman and Kessel, 1987); less
than 12 years education (Kleinman and Kessel, 1987); low
calorie ($1500 kcal/day) intake by the mother (Doyle et al.,
1982; Lechtig et al., 1975), adolescent pregnancy (Brennan
et al., 1983; McCormick, 1985), non-white race (Abrams et
al., 1989; Kleinman and Kessel, 1987; Brennan et al., 1983;
McCormick, 1985), and parity (Brown et al., 1986).

Many of these variables which influence the outcome of

48

pregnancy were established by analyzing research data using
simple regression. At a Boston city hospital, O’Sullivan et
al. (1965) asked 5,883 pregnant women about their ages,
races, usual weights, numbers of previous pregnancies and
the numbers of previous births in excess of nine pounds.
These characteristics were compared to data of their present
pregnancies. When the other factors were removed, racial
characteristics, age, parity, and blood sugar had no
significant effect, whereas, self-reported usual weight had
a significant effect (p<0.01) on the birth weight of the
infant. However, the income, race or education levels of
these women were never reported or analyzed.

Only the studies by Haworth et al. (1980), Abrams et
al. (1989), Abrams (1991) and Brown et al. (1986),
specifically targeted a low-income population although a
high prevalence of LBW babies are born to women in poor
communities and this problem has become a major health
concern (Lechtig et al., 1975). Brown et al. (1986)
compared clinical data of low-income women with the birth
weights of their infants. Multiple regression analyses
showed that gestational length, weight gain during
,pregnancy, prepregnancy weight, cigarette smoking and parity
Eire all individually correlated with infant birth weight.
Hkawever, the influence of gestational length on birth weight
deacreased as prepregnancy weight increased only when the

mczther was < 150% of her ideal body weight.

49

In two similar research projects, Abrams et al. (1989)
and Abrams (1991) attempted to separate the risk factors of
prematurity from the risk factors of SGA among low-income
populations. Data of maternal characteristics and birth
outcomes from 2228 women who participated in the University
of California, San Diego Prenatal Nutrition Project between
1978 and 1988 were examined. Multivariate analyses of the
data showed that significant risk factors for SGA births
included: cigarette smoking, low rate of maternal weight
gain, black or asian ethnicity, pregravid underweight (< 90%
of the 1959 Metropolitan Life Standards), primiparity, and
low maternal height. With the exception of black ethnicity
and low maternal weight gain, the risk factors for premature
birth were different: incompetent cervix, use of illicit
drugs, exposure to trauma, surgery or anesthesia during
pregnancy, maternal age less than 16 years and short
interpregnancy intervals.

Lieberman et al. (1987) focused on the problem of race
and premature birth. Primarily concerned with the fact that
twice as many black babies die in the first year of life
compared to white babies, these researchers collected data
through medical records and personal interviews of 14,458
‘MOmen at Boston Hospital for Women between 1977 and 1980.
The results showed that 9.2% of black infants were born
prematurely, whereas, only 4.9% of white infants were born

at: less than 37 weeks gestation. The black mothers were

50
often more anemic, underweight, younger, single, less
educated and on welfare compared to white mothers. However,
when socioeconomic status (SES) risk factors were taken into
account, race was no longer a significant predictor of

premature birth.

Maternal weight and weight gain The Institute of Medicine

(1990) recommends that a woman should gain weight according
to her prepregnancy Body Mass Index (BMI). Women with a low
BMI ($19.8) should gain 28 to 40 lbs, with a normal BMI
(19.8 to 26.0) should gain 25 to 35 lbs, and with a high BMI
(>26.0) should gain 15 to 25 lbs during pregnancy. Less
than half of the mass gained goes to the fetus, placenta and
amniotic fluid, but the majority is added to reproductive
tissues, fluid, blood and body fat stores (Worthington—
Roberts and Williams, 1989). The adequacy of weight gain
can only be determined when the length of gestation is known
(Institute of Medicine, 1990). The length of gestation is
often estimated by calculating the length of time since the
last menstrual period, but this method is only as accurate
as the memory of the mother.

Ideally, there is little weight gain in the first weeks
<3f pregnancy, two to four lbs (0.9 to 1.8 kg) by the end of
tlae first trimester and about one lb (0.45 kg) each week
thaereafter (Hytten and Leitch, 1971). According to

Wc>rthington-Roberts and Williams (1989), steady growth in

51
which most of the gain is in lean body tissue is desired. A
sudden gain in weight might signal fluid retention, a sign
of pregnancy-induced hypertension (PIH).3 Low pregnancy
weight gain can increase the risk of poor pregnancy outcome.
The lowest incidence of PIH, LBW and perinatal death are
associated with gaining 0.45 kg per week during the last 20
weeks of pregnancy (Hytten and Leitch, 1971). Those women
at greatest risk for low weight gain (<6.8 kg or 15 lbs) are
unmarried, black or Hispanic, cigarette smokers or women
with low levels of education (Institute of Medicine, 1990).
These women might need prenatal counseling to lower the risk
of inadequate weight gain.

In a study by Mitchell and Lerner (1989), regression
was used to determine if a relationship existed between
prenatal weight gain and pregnancy outcome. The researchers
obtained data on 1081 middle-class urban women who had
volunteered for a previous study between 1976 and 1984. The
women were matched for age, parity, occupation, height, race
and smoking habits in an attempt to control for factors that
have been correlated with negative pregnancy outcomes. A
total of 362 pairs of normal weight and underweight women
were formed. The results showed that no relationship

existed between the initial weight of the mother and the

 

3 Pregnancy-induced hypertension is defined as a 20 to 30 mm
Hg rise in systolic blood pressure or a 10 to 15 mm Hg rise in
diastolic blood pressure measured on two separate occasions at
least six hours apart. This condition is characterized also by
edema and proteinuria.

52
birth weight of the infant. However, birth weight was shown
to have increased significantly with maternal weight gain
(p<.0001). For every kilogram of weight gained during
pregnancy by an underweight mother, 12.85 g were gained by
the infant and for each kilogram gained by a normal weight
mother, the weight of the offspring increased by 8.59 g.

In a similar study of the relationship of maternal
weight gain and infant birth weight, Brown et al. (1986)
obtained data on low—income women collected by the Maternal
and Infant Care (MIC) Project in Cleveland and Minneapolis.
Data records from 1,342 women enrolled in the project
between 1979 and 1982 were randomly selected. Underweight
was defined as g 95%, normal weight as 95% to 115%, over
weight as 115% to 150 % and obesity as 2 150% of ideal body
weight from the 1960 Metropolitan Life Insurance Tables. As
in the studies of women at higher income levels, infant
birth weight was found to be significantly correlated with
prenatal weight gain (r=.20, p< 0.001).

In another study of the effect of maternal mass and
weight gain on birth weight, Abrams et al. (1986) collected
data through interviews and medical records of 2946 live
births with deliveries after 37 weeks gestation at a
University Hospital in San Francisco, California between
1980 and 1983. Through multiple regression, the researchers
found that both pregravid body mass and weight gain during

LDregnancy significantly influenced infant birth weights, but

 

53
only for moderately overweight, normal weight and
underweight women (those women with prepregnancy weights
(5135% of ideal body weight). Body mass was estimated using
Quetelet’s index (kg/cm?) and the researchers found that for
every unit increase in maternal prepregnancy body mass,
there was an increase of 15.9g in body weight of the infant
(p<0.00001). Furthermore, for every kg increase in maternal
weight there was a 25.9g, 28.3g or 17.8g increase in body
weight of infants of underweight, normal weight and
moderately overweight women, respectively (p<0.00001). No
significant birth weight differences were found associated
with an increase in body mass or weight among obese pregnant

women .

Smoking Hundreds of studies have demonstrated that smoking
during pregnancy increases the risk of low birth weight
babies (52500 grams) and perinatal mortality (Langford et
al., 1983; Meyer et al., 1976). On the average, smokers’
babies are 150 to 250g lighter than non-smokers’ babies and
this decrease in birth weight is independent of other
factors that have been correlated with LBW such as maternal
weight, age, race and socioeconomic status (Meyer et al.,
1976). However, the results of recent studies conflict as
to whether low birth weight is due to decreased dietary
intake among smokers. Some researchers have reported

(decreased energy or nutrient intake among pregnant smokers

54
(Smithells et al., 1977; Rogozinski et al., 1983), whereas,
other studies have shown an increase in energy intake among
pregnant women who smoke when compared to pregnant non-
smokers (Beal, 1971; Haworth et al., 1980).

In a study to determine whether the effect of maternal
cigarette smoking on fetal low birth weight could be due to
decreased dietary intake in smokers compared to non-smokers,
Haworth et al. (1980) assessed the nutrient intake of 302
smoking and 234 non-smoking women. Dietary intake was
assessed by calculating the average daily nutrient intake
from a diet history and by analyzing 24-hour recalls of
women in the last trimester of pregnancy.

The results of the study showed that smokers had
significantly smaller infants than the non-smokers, although
pregnancy weight gain did not significantly differ between
the two groups. Furthermore, the analyses of the diet
records and recalls indicated that the energy intake of
smokers was not less than that of non-smokers. Leaving the
researchers to conclude that fetal growth retardation due to
smoking is not caused by a mother’s decreased food intake.

Other researchers are looking beyond diet to explain
the relationship between low birth weight infants and
maternal smoking. Naeye (1980), Naeye (1981) and Meyer et
al. (1976) report that cigarette smoking increases fetal and
neonatal mortality by increasing the frequency of: abruptio

E>lacentae (the premature detachment of the placenta),

55
placenta previa (the implantation of the placenta in the
lower part of the uterus so that it blocks the cervix
opening) and major congenital malformations.

In one study Naeye (1980) analyzed data from the
Collaborative Perinatal Project of the National Institute of
Neurological and Communicative Disorders and Stroke which
included information gathered on 53,518 pregnant women
between 1959 and 1966. The investigation looked at the
possibility that cessation of smoking might decrease the
frequency of smoking-related disorders. The results showed
that mothers who quit smoking after conception had a 23%
lower frequency of abruptio placenta and a 33% lower
frequency of placenta previa than women who continued to
smoke. The greatest difference between groups was found
when comparing smokers to non-smokers over thirty years of
age.

Using data from the same study, Naeye (1981) reported
that one disorder, placenta previa, was found to be directly
related to the number of years the mother smoked regardless
of how long before the pregnancy she had quit. The
frequency of the three disorders did not decrease in infants
of women smokers as weight gain increased, therefore, the
researcher concluded that increasing food intake of pregnant
women who smoke might not protect the fetus.

In a related study, Meyer et al. (1976) investigated

the relation of the amount of smoking related to increased

56
fetal and infant mortality by analyzing data from the
Ontario Perinatal Mortality Study. Regression analyses
showed that placental complications increased consistently
with smoking level. For smokers who consumed less than one
pack per day, placenta previa rates increased by 25% and
abruptio placenta rates increased by 23% over the control
group. Furthermore, rates for smokers of greater than one
pack per day increased 92% for placenta previa and 86% for
abruptio placenta. Meyer et al. report that these
complications carry high risks of perinatal mortality and
account for 1/3 to 1/2 of the perinatal deaths attributed to
maternal smoking. Due to the high health risks associated
with smoking, the Institute of Medicine (1990) has
recommended that public health programs for pregnant women
target populations which tend to have a high prevalence of
smokers such as groups of women with low levels of
education. The Institute suggests that women who smoke be
encouraged to quit, but if a woman continues to smoke during
pregnancy, she might benefit from multivitamin/mineral

supplementation.

Alcohol use According to the Institute of Medicine (1990),
a review of scientific literature indicated that maternal

use of alcohol did not affect the weight gain of the mother
during pregnancy. However, maternal ethanol ingestion can

interfere with transport across the placenta resulting in

57
fetal alcohol syndrome (FAS) (Worthington-Roberts and
Williams, 1989). This syndrome was characterized by Jones
et al. (1973) as abnormalities of the eyes, nose, heart and
central nervous system, growth retardation, small head
circumference and mental retardation. Moreover, excessive
intake of alcohol by pregnant women has been reported to be
associated with an increased risk of spontaneous abortion,
abruptio placenta, LBW and prenatal mortality (Worthington-
Roberts and Williams, 1989). The infants who do survive
fetal exposure to alcohol often experience symptoms of
withdrawal such as irritability and hyperactivity in their
first few weeks of life (Worthington-Roberts and Williams,
1989).

The Council of Scientific Affairs (1983) reported that
studies have consistently found a positive correlation
between excessive maternal alcohol intake and the risk of
fetal abnormalities, regardless of race or income level of
the woman. The effects of moderate alcohol intake are not
clear and, therefore, no safe level of alcohol consumption
during pregnancy has been determined. The Institute of
Medicine (1990) reports that mothers who consume alcohol
during pregnancymight have a low intake of protein,
essential fats, vitamins and minerals. Furthermore, they
might have difficulty metabolizing calcium, amino acids and
some vitamins and might have decreased nutrient utilization

due to liver impairment. There is no convincing evidence

58
that nutritional supplementation can compete with the
adverse effects of alcohol, therefore, the Institute of
Medicine recommends that women limit or eliminate alcohol

intake during pregnancy.

Caffeine intake The Institute of Medicine (1990) reports
that there is not conclusive evidence to show that caffeine
causes birth defects, but there is some evidence to suggest
that the drug might be associated with decreased infant
birth weight. At this time, the Institute of Medicine makes
no official recommendation pertaining to caffeine intake

during pregnancy.

Drug use Among the general pregnant American public,
marijuana use is estimated at 10 to 27% and among inner city
pregnant women cocaine use is about 18% (Institute of
Medicine, 1990). According to Archer et al. (1983) cocaine
use is associated with increased blood pressure and should
not be confused with PIH. Cocaine-induced hypertension
might lead to abruptio placenta.

A study by Chasnoff et al. (1985) compared pregnancy
outcomes of 23 cocaine users with patients receiving
methadone and patients free of drugs. First trimester
miscarriages in the three groups were 38%, 16% and 0%
respectively. Four cocaine users experienced onset of labor

with abruptio placenta following an injection. The

59
Institute of Medicine (1990) recommends that highest
priority be given to efforts to stop drug abuse because of
the adverse effects on mother and fetus. Diet counseling
and vitamin/mineral supplementation are recommended in
addition to drug counseling due to the high incidence of

nutritional deficiencies that accompany drug use.

Nutrient intakes of pregpant women According to Brennan et

al. (1983), the groups most at risk for poor pregnancy
outcome are the groups with an increased risk for poor
nutrition: very young, low—income, non-white, high-parity,
low-literate mothers living in the inner city. Beal (1971)
reports that nutrition during pregnancy is most important
for the woman who has been poorly nourished throughout her
life and, therefore, has decreased reserves of nutrients to
support the fetus. These women often continue to consume a
less-than-optimum diet throughout pregnancy and, therefore,
need extra attention from health care officials (Snowman,
1979). Beal (1971) recorded the nutritional histories of
149 white, middle-income, pregnant and non-pregnant American
women between 1946 and 1966. These women were found to be
well nourished before pregnancy, consuming an average of
1900 kcal, 65 g protein and 0.8 g calcium per day. The
researchers found no consistent dietary patterns among women
who gave birth before 37 weeks gestation compared to women

who gave birth to full-term infants. Furthermore, the

60
change in kilocalories from one pregnancy to the next among
multiparous women did not result in a comparable change in
birth weight of the infant.

A study of an entirely different population, black and
white, low-income pregnant women showed that the diet of
these women differed considerably from the well-nourished
population in Beal’s study. Sixty-eight urban mothers in
unskilled occupations and incomes of $5,000 (1970 income) or
less annually were asked to respond to questions on personal
dietary habits and to give a 24 hour food recall to a home-
interviewer. The recall was given verbally to the
interviewer to bypass any problems the subject might have
had with literacy. The results showed that no typical
dietary pattern existed among these low-income mothers,
however, a few common practices were discovered. Vitamin
and mineral supplement usage was high (96%), but half of the
supplement users (53%) practiced some form of pica (the
ingestion of non-food items). In this case, pica was not
found to be significantly associated with altered dietary
intakes, lowered hemoglobin or hematocrit levels, pregnancy
complications or LBW, possibly due to the high prevalence of
vitamin and mineral supplements. Intakes of energy, calcium
and iron were found to be at low levels for both black and
white women compared to the 1974 RDA’s.

Similar results were found in the Ten State Nutrition

Survey (TSNS) and in a study by Smithells et al. (1977).

61
Data collected in the TSNS showed that diets of pregnant
women in the U.S. are below the RDA for energy, iron,
calcium, vitamin A and protein (Brennan et al., 1983). In
the study by Smithells et al. (1977), the researchers
calculated the mean daily dietary intake of 195 British
women ages 18 to 55 years in their first trimester of
pregnancy. Volunteers were interviewed by a doctor and
filled out a questionnaire on employment, family income and
personal smoking habits. The women kept a five-to-seven day
dietary record by weighing the food before eating it and
noting any episodes of nausea or vomiting. When compared to
the British RDAs for pregnant women, 80% of the women in the
study had low intakes of energy and iron and 100% had low
cholecalciferol intakes. In general, nutrient intakes
tended to decrease with social class, severity of vomiting
and age, but not with the number of cigarettes smoked. The
researchers expressed specific concerns over the decrease in
nutrient intake with social class, noting that due to the
required literacy skills to keep accurate dietary records,
women from the lowest social classes are often under
represented in this type of research.

In 1983, Brennan et a1. collected and weighed duplicate
portions of food consumed in 24 hours by a group of black
pregnant women of low SES. The 22 female volunteers ranged
in age from 16 to 38 years. Through interviews it was

revealed that half of the subjects consumed three meals

62
daily, whereas, the other 50% ate only two meals per day
plus snacks. The consumption of fruits and vegetables was
low with 14 out of 22 women eating less than three servings
per day. Laboratory analyses on the duplicate portions and
calculations of the data from these analyses were used to
assess the actual nutrient intake of the women. The results
were compared to the 1980 RDAs for pregnant women. The
researchers reported that the average energy intake was low,
but a wide range existed from 772 to 2,723 kcal per day.
Less than half of these women met the RDA for magnesium,
zinc or folacin. However, intakes of phosphorous, vitamin
A, riboflavin and niacin met or exceeded the RDA.

In a third study, Doyle et al. (1982) assessed a seven
day weighed dietary record of 76 low SES women in the third
trimester of pregnancy. The researchers compared the data
of those women who delivered SGA babies to the data of those
women who delivered normal weight-for-age babies and they
compare also the nutrient intake levels of all subjects to
the British RDAs. All mothers were 17 to 29 years of age,
non-smokers, primigravida or previously had SGA babies and
were literate. Once again, due to the need for literate
mothers to keep dietary records, the illiterate and low—
literate individuals were not included in the study.

The results showed that mean energy intakes were lower
for mothers who delivered LBW babies compared to mothers who

gave birth to normal weight-for-date babies (p<0.001).

 

63
Secondly, the intake of almost all nutrients was lower for
mothers of LBW infants, but only the levels of total fat and
pyridoxine were significantly lower. When comparing the
maternal intakes to the British RDAs, many of the results
are similar to the results of studies by Brennan et al.
(1983) and Smithells et al. (1977). Doyle et al. (1982)
found low intakes of energy, iron, and calcium like the
other researchers, but they found also low intakes of
vitamin D and folic acid among mothers who delivered LBW
babies compared to mothers who delivered normal weight
babies.

Recommended Dietary Allowances (RDAs) are based on
evidence from metabolic studies. Requirements are estimated
to avoid deficiencies and maintain balance, then
recommendations are adjusted upward to cover individual
differences in the population due to variations in
digestion, absorption and utilization (Worthington—Roberts

and Williams, 1989).

64

 

 

 

 

 

 

 

 

 

 

 

Table 4: Recommended Dietary Allowances for Pregnant
Women1
Nutrient Allowance Nutrient Allowance
f_0_r 152.1;
Pregnancy Pregpancy
protein 60 g folacin 400 ug
vitamin A 800 ug RE vitamin B-12 2.2 ug
vitamin D 10 ug calcium 1200 mg
vitamin E 10 mg TE phosphorous 1200 mg
vitamin K 65 ug magnesium 300 mg
vitamin C 70 mg iron 30 mg
thiamin 1.5 mg zinc 15 mg
riboflavin 1.6 mg iodine 175 ug
niacin 17 mg NE selenium 65 ug
vitamin B-6 2.2 mg

 

 

 

 

 

 

1. The RDA’s for pregnancy are adjusted upward when studies
show an increased need for a nutrient during pregnancy
compared to the needs of non-pregnant women.

Energy intake An increased requirement for energy

during pregnancy occurs for several reasons: new tissues
are being produced, there is an increase in maternal BMR and
the workload of the mother has increased due to the
increased weight (Blackburn and Calloway, 1976).
Worthington-Roberts and Williams (1989) have reported that
the estimated energy cost of pregnancy is 85,000 kcal or 300
kcal per day, although this increased need might be

partially offset by decreased physical activity. Recent

65
studies on the effect of food restriction in pregnant rats
fed half of their normal caloric intake resulted in
significantly decreased body weight in the fetuses
(Worthington-Roberts and Williams, 1989). Therefore, the
fetus does not completely parasitize the mother in its need
for energy and a lack of maternal energy could mean an
increased mortality risk for the fetus. The Institute of
Medicine (1990) reports that energy supplements would be
beneficial for pregnant women whose average daily intake is
g 1900 kcal.

In humans, due to the laws of ethics, it is difficult
to separate the effects of energy deficiency from those
effects of protein deficiency. However, maternal protein-
calorie malnutrition has been shown to decrease the ability
of the mother to conceive, interfere with the normal growth
of the placenta and fetus, and cause malformations,
resorption or death in the fetus (Worthington-Roberts and
Williams, 1989).

Protein intake Protein is required to build new cells
in both the mother and the fetus, therefore, the requirement
is increased during pregnancy. Schuster et al. (1981) found
that among low-income, black and white, pregnant women the
mean protein intake was 106% of the 1980 RDA. The Institute
of Medicine (1990) recommends that protein requirements
during pregnancy be met from the diet and no supplementation

such as high protein powders are recommended.

66

Vitamins

Vitamin A intake No supplement of vitamin A is
recommended during pregnancy, although the requirement is
increased (Institute of Medicine, 1990). Pregnant women
should be advised to avoid an excess (2 25,000 IU per day)
of this vitamin, because it can be easily transported across
the placenta by simple diffusion and has been reported to
cause fetal malformations (Worthington-Roberts and Williams,
1989).

According to Shah and Rajalakshmi (1984), relative
vitamin A intake decreases with income level. These
researchers compared the vitamin A status of newborns in
relation to gestational age, birth weight and maternal
vitamin A status among high and low income groups. Infants
born to mothers from the high income group had significantly
higher cord serum vitamin A levels than cord serum levels
among babies from the low income group. Mothers of
premature infants had lower levels of serum vitamin A
compared to mothers of full—term infants, indicating a
correlation between maternal vitamin A status and

prematurity in poorly nourished individuals.

Vipamin D ippake Like vitamin A, vitamin D easily
diffuses across the placental membrane and, therefore, an

excess intake can be toxic causing infantile hypercalcemia

67
(Worthington-Roberts and Williams, 1989). -However, this
vitamin positively affects calcium balance during pregnancy,
therefore, those at risk (complete vegetarians, women with a
low intake of vitamin D fortified milk and women with
minimal sunlight exposure) should supplement their diets

with 10 ug per day (Institute of Medicine, 1990).

Vitamin E intake The main role of vitamin E in the
body is to prevent oxidation of unsaturated fatty acids in
the cell membrane. Supplementation during pregnancy is not
recommended as this vitamin is found in commonly eaten foods
(Institute of Medicine, 1990). Deficiency of this vitamin
is rare and has not been linked to any decrease in
reproductivity or fertility (Worthington-Roberts and

Williams, 1989).

Thiamin, riboflavin and niagin intake These three

water-soluble B-vitamins are all part of reactions to
produce energy and their requirement in the body is
dependent upon caloric intake. During pregnancy, the need
for these vitamins is increased due to the increase in
dietary calories (Worthington-Roberts and Williams, 1989).
Heller et a1. (1974) found no significant correlation
between riboflavin status in 651 women (assessed by means of
an erythrocyte glutathione reductase activation test) and

pregnancy outcome. The Institute of Medicine (1990) reports

68
that there is no basis for recommending supplementation of
these three B vitamins for the general population of

pregnant women .

Vitamin C intake Low plasma levels of vitamin C are
associated with the premature rupture of membranes and pre—
eclampsia. For the pregnant woman, extra daily intake of
this vitamin is recommended, therefore, the RDA is raised to
70 mg per day. This level can easily be obtained with the
American diet and no recommendation for supplementation has
been made (Institute of Medicine, 1990). However, excessive
maternal intake of vitamin C during pregnancy has been
positively correlated with an increased requirement of this
vitamin by the infant after birth, therefore, excessive

supplementation should be avoided.

Folic acid intake The deficiency of folate in the diet
is one of the most common vitamin deficiencies seen,
especially among lower socio—economic groups (Subar et al.,
1989), women and African Americans (Blocker et al., 1989).
Deficiency, leading to abnormal cell division, might be
caused by decreased dietary intake, alcohol consumption,
smoking or use of specific drugs such as anticonvulsants
(Bailey, 1990). During pregnancy, a woman’s need for folate
doubles due to the expanding red cell mass and continuous

fetal growth. Therefore, the RDA for folate during

69
pregnancy is increased from 180 ug to 400 ug per day. A
decrease in folate utilization or a failure to increase the
level of folate intake might result in megaloblastic anemia.
Most often occurring in the third trimester, this type of
anemia causes the mother’s heart, liver and spleen to become
enlarged, threatening the health of the mother and the life
of the fetus (Willis, 1984). Pregnant women consuming a low
level of folate place themselves at risk for poor pregnancy
outcomes, including: prematurity, LBW, abruptio placenta,
spontaneous abortion, stillbirths (Blocker et al., 1989),
and infant neural tube defects (Vergel et al., 1990; Bower
and Stanley, 1989; Milunsky et al., 1989).

In an animal study involving folic acid levels in ggppp
albifrons monkeys, Blocker et al. (1989) compared dietary
folic acid intake during pregnancy with folate levels in the
mother and child postpartum. Maternal, infant and non-
pregnant hematologic indices, blood and liver folate
concentrations and urinary formimino glutamine acid
excretion levels all varied with dietary folate intake and
pregnancy status. The results of the study suggest that
since fetal tissues do not have preferential access to
maternal folate stores when those stores are compromised,
the fetus is highly dependent on the folate status of the
mother.

In a human study, the mean daily folate intakes of 195

women in the first trimester of pregnancy were assessed by

70
weighed dietary records of five to seven days. Like many
essential nutrients, Rogozinski et al. (1983) found that the
lowest intakes of folate during pregnancy are associated
with the lowest social classes, maternal age below 20 years
and vomiting three or more days per week. The researchers
found also that folate intake was greater in women who did
not smoke compared to women who smoked ten or more
cigarettes per day (p<0.05).

At a prenatal clinic in Florida, Bailey et al. (1980)
determined folacin and iron status of 269 low income women.
These researchers found that serum folacin levels were low
in 45% (<6ng/ml) and deficient in 15% (<3ng/ml) of the
subjects. At the same time, serum iron was normal in 96% of
the subjects. These data suggest that folacin deficiency is
more prevalent than iron deficiency in low income
populations.

Due to the increased requirement of folic acid during
pregnancy, supplementation is often prescribed, despite the
ongoing controversy over this practice. One of the critics
of folic acid supplementation is B.M. Hibbard of the
University of Wales. Hibbard (1988) believes that folic
acid supplementation for pregnant women is valuable only in
selected individuals and that most women in Western society
can get their necessary nutrients from a good mixed diet.

He reports that folic acid intake is hindered by ignorance,

inertia or a lifetime of bad habits, therefore, healthy

71
eating campaigns and not supplementation should be stressed
during pregnancy.

Contrary to the view of Hibbard, Horn (1988) argues
that most women are not able to meet the demands of
expanding red cell mass and fetal growth in pregnancy
through diet alone. Folate circulation in tissues is
reduced and many pregnant women become anemic. She adds
that supplements of 100 to 300 ug per day of folic acid are
effective in maintaining the folate state of the healthy
pregnant woman.

National committees differ in their support of folic
acid supplementation during pregnancy. The Institute of
Medicine (1990) sides with Hibbard, recommending no routine
supplementation of folate. However, the Institute vaguely
recommends a 300 ug per day supplement for cases where
doubts exist about folate adequacy in the diet. The U.S.
Centers for Disease Control recommend that all women of
childbearing age increase their intake of folic acid to .4
mg daily, whereas, women with a history of neural tube
defects increase their supplementation to ten times that

amount to 4 mg folic acid per day.

Vitamin B-6 intake Schuster et al. (1981) report that
a significant portion of low income pregnant women are in
poor vitamin B-6 status. Maternal B-6 deficiency during

pregnancy can result in infants born with lower Apgar scores

72
compared to the scores of infants born to mothers with
adequate levels of this vitamin. Study participants
consisted of 127 low income, black and white, adult and
adolescent females. Vitamin B-6 intake data, collected by
means of a 24-hour recall, showed that 57% of subjects
consumed 53%, 77% of subjects consumed 567% and 88% of
subjects consumed less than the 1980 RDA for vitamin B-6 for
pregnant women. At the same time, the mean energy intake of
the women was 84% of the RDA and the mean protein intake was
106% of the RDA. Apgar scores of infants at one minute
after birth, but not birth weights, were significantly
higher (p<0.05) for infants whose mothers’ diets were B-6
adequate at their first prenatal clinic visit compared to
scores of infants whose mothers had low vitamin B—6 diets.
The Institute of Medicine (1990), however, has found no
convincing evidence that vitamin B-6 supplementation is
needed in pregnancy and makes no recommendation for

supplementation at this time.

Vitamin B-12 intake A deficiency in vitamin B-12
produces the same anemia associated with a lack of folic
acid: megaloblastic anemia. Besides anemia, severe
deficiency can cause irreparable damage to the nervous
system (Worthington-Roberts and Williams, 1989). Although,
severe B-12 deficiency during pregnancy is rare, some

researchers recommend a supplement (Willis, 1984). The

73
Institute of Medicine (1990) recommends a vitamin B-12

supplement of 2.0 ug/day for complete vegetarians.

Minerals

Iron intake A 30% increase in iron intake is required
for the production of hemoglobin in the fetus and placenta
and an increased maternal blood volume during pregnancy
(Dawson and McGanity, 1987). Willis (1984) reports that the
iron requirement during pregnancy cannot be met by the
average American diet and although absorption during
pregnancy increases, stores of the mineral consistently
decrease (Dawson and McGanity, 1987). Furthermore, the
Council of Foods and Nutrition of the American Medical
Association has reported that many women enter pregnancy
with inadequate iron stores (Ashe et al., 1979), and,
therefore, the Council recommends supplementation of 30 to
60 mg per day. The Institute of Medicine (1990) recommends
also 30 mg of ferrous iron per day for women in the second
and third trimesters of pregnancy.

The diet of the mother might affect the absorption of
iron in her body and, therefore, the availability of iron to
her fetus. Absorption is decreased with the consumption of
tannic acid found in tea, EDTA, calcium phosphate, phosvitin
found in egg yolk, phytates and soy products, whereas,

absorption is increased with the intake of ascorbic acid,

74
meat, poultry or fish (Raper et al., 1984).

A study by Dawson and McGanity (1987) investigated the
usefulness of iron supplementation during pregnancy by
determining whether supplementation of iron could maintain
satisfactory hematologic values and maternal iron stores
throughout pregnancy, delivery and up to three months post
partum. Forty—two white and black women less than 16 weeks
pregnant received multivitamin supplements either with 65 mg
iron or without the iron. After 16 weeks of
supplementation, those women without the iron supplement had
significantly lower serum ferritin levels (0.05) and 43% of
the non-iron group failed to maintain hemoglobin levels at
11.0 g/dl. These researchers concluded that use of iron
supplementation is necessary to maintain iron stores and

prevents the development of iron-deficiency anemia during

pregnancy.

Celcium intake The fetus acquires most of its calcium
in the last trimester of pregnancy when the majority of bone
and teeth formation takes place. If calcium is not present
or present at low levels, the mineral is removed from the
mother’s bones to be used for the fetus. Belizan et al.
(1988) suggest that a lack of calcium intake during
pregnancy might be also associated with an increased risk of
PIH. The RDA for pregnant women is set at 1200 mg per day

or 400 mg greater than for non-pregnant individuals.

75

To ascertain the effectiveness of supplements, Ashe et
al. (1979) determined the retention of calcium and iron in
women on supplemented and unsupplemented diets. Six 7-day
metabolic studies were carried out on ten pregnant white
women ages 19 to 29 years. The women ate self-selected
diets in which they weighed, recorded and collected
duplicate food samples for laboratory analysis. Medications
were noted and all waste collected. The results showed no
significant differences in calcium balance, but
significantly greater retention in iron among supplemented
compared to unsupplemented women. The researchers concluded
that only iron supplementation can be substantiated.
However, this study did not include young, low income or
other high risk groups of pregnant women. The Institute of
Medicine (1990) recommends calcium supplementation (600 mg)
for women under age 25 whose daily dietary intake of calcium
is less than 600 mg. Researchers have consistently found
low calcium intakes among low-income pregnant women (Brennan

et al., 1983; Snowman, 1979).

Zinc and Copper intake The recommended intake for zinc
is set at 15 mg per day during pregnancy. Most women are
only able to obtain 9 to 11 mg through their daily diets and
it is possible that iron supplementation exacerbates the
situation by decreasing absorption (Worthington-Roberts and

Williams, 1989). Lack of zinc and copper in the prenatal

76
diet have been shown to have teratogenic effects on the
fetuses of animals. In rats, severe copper deficiency
increased the risk of fetal resorption and stillbirths
(Taper et al., 1985).

In humans, Taper et al. (1985) evaluated zinc and
copper utilization from the self-selected diets of 24
pregnant women in their second trimester of pregnancy.

After conducting two 7-day metabolic studies in which all
food, drink, fecal and urine samples were analyzed for the
minerals, the researchers found that the RDA for zinc or the
suggested intake for copper were not met unless supplements
were used. According to Taper et al. intakes above 81mg
resulted in positive retention, but due to individual
variation, the RDA of 15 mg zinc would not be excessive.
Furthermore, adequate copper retention was not possible
without supplementation. The Institute of Medicine (1990)
does not recommend supplementation of these minerals for the
general population, but supplements of 15 mg zinc and 2 mg
copper are recommended for women who are taking therapeutic

doses of iron (> 30 mg/day) for anemia.

Supplementation Several researchers have found low
intakes of nutrients when assessing the diets of pregnant
women. The most commonly found deficiencies include iron,
calcium and folate, therefore, these nutrients are most

often prescribed by doctors for supplementation of the

 

 

(D

(1'

 

 

77
pregnant woman’s diet. However, for the general population
the Institute of Medicine (1990) recommends supplementation
of iron only. The American Dietetics Association (ADA)
takes the position that supplement usage might be indicated
for women who are pregnant or breastfeeding, especially
supplementation of iron and calcium (ADA, 1987). The
Institute of Medicine (1990) recommends vitamin/mineral
supplementation for women who might be at risk: women
carrying multiple fetus, cigarette smokers, alcohol users

and drug abusers. The supplement should contain the

following:
30 mg iron 15 mg zinc 2 mg copper
250 mg calcium 2 mg vitamin B-6 300 ug folate
50 mg vitamin C 5 ug vitamin D

In poor communities where food might be in short
supply, supplementation might be better accomplished by
adding extra food, with its much needed calories, to the
diet as opposed to distributing vitamin/mineral supplements
to pregnant women. Lechtig et al. (1975) studied the
effects of food supplementation of pregnant women on the
birth weight of the infant in four rural villages of
Guatamala. Compared to the control group, the researchers
found that the experimental group had an increased caloric
intake (with or without increased protein intake) and
experienced an increase in the birth weight of their babies
(29 g birth weight per 10,000 supplemented calories). This

increase was unchanged after controlling for initial

 

78
maternal diet, height, head circumference, parity,
gestational age, duration of disease during pregnancy and
socio-economic status.

In the U.S. low income women are eligible to
participate in the Women, Infants and Children (WIC) program
to obtain food coupons to improve their prenatal diets. A
study by Endres et al. (1981) showed that pregnant
participants in the WIC program increased their consumption
of protein, calcium, iron, vitamin A, vitamin C, thiamin,
riboflavin, niacin and folacin compared to low income, non-

participating, pregnant women.

Cravings and aversions One way in which women meet the
increased need for calories and nutrients during pregnancy
is to change their dietary behavior. These changes might be
affected by cultural beliefs, food availability, food costs
and food preferences which in turn might influence cravings
and aversions for certain foods.

Tierson et al. (1985) conducted a study of the
influence of maternal dietary cravings and aversions during
pregnancy on the frequency of consumption of specific
dietary items. Cravings and aversions were defined as
having extreme likes or dislikes for an item for at least 24
hours. The researchers interviewed 400 white, well-
nourished women from Albany, New York at the 16th, 20th,

30th and 38th weeks of pregnancy and at eight weeks post

79
partum. During the interviews, the women were asked to
report any episodes of cravings, aversions, nausea or
vomiting since the previous interview.

The results showed that aversions tended to appear
before cravings and last for a longer period of time.
Cravings of at least one item were reported by 67% of the
women and 85% of the subjects experienced one or more
aversions. Commonly reported cravings included sweet foods:
ice cream, chocolate, cookies, candy, fruits and Italian
food. Aversions, on the other hand, included more bitter
flavors: fish, meat, coffee and alcohol. Most reasons
given for overindulging or avoiding a food item pertained to
personal feelings of nausea, vomiting or food preference as

opposed to advice from a physician or concern for the fetus.

Pipe The most common pica practices (the ingestion of
non-food substances) involve eating dirt, clay, starch and
ice, but even large toxic compounds have been consumed
during pregnancy. Worthington-Roberts and Williams (1989)
report that pregnant women have been known to consume burnt
matches, hair, gravel, charcoal, soot, cigarette ashes,
mothballs, antacids, milk of magnesia, baking soda, coffee
grounds and tire inner tubes.

Researchers and health care workers are concerned about
pica for several reasons. First of all, non—food substances

might replace nutritional foods, starch is particularly high

 

80
in calories and might lead to excessive weight gain,
substances such as charcoal, air freshener and mothballs are
toxic if consumed, the chemical ingredients in some
substances might interfere with the absorption of some
minerals (Willis, 1984) and parasitic infections might
result from the ingestion of contaminated clay or soil
(Worthington-Roberts and Williams, 1989). The reasons that
pica occurs are unknown, however, researchers have suggested
that it might be associated with iron deficiency (Willis,
1984).

In the study of aversions and cravings by Tierson et
al. (1985), there were no reports of pica. The lack of
occurrences of this type of craving might be due to the fact
that the population for the study were white women from New
York, whereas, pica is more prevalent among southern blacks
(Willis, 1984). Snowman (1979) interviewed 68 low-income,
black and white pregnant women from Syracuse, N.Y.
concerning their pica practices. He found a 53% rate of
pica practices among the expectant mothers: 30 consumed
ice, 8-seeds, 6—laundry starch, 1-clay, 1-baby powder.
However, this researcher found no association between pica
and altered dietary intakes, low hemoglobin or hematocrit
levels, complications during pregnancy or LBW. He adds that
this particular population had a high (96%) rate of
vitamin/mineral supplementation which might have masked any

adverse effects of pica.

81

Horner et al. (1991) reviewed scientific literature
from 1950 to 1990 on the prevalence of pica in the U.S. The
researchers found that after a decline in the prevalence of
pica between 1950 and 1970, the numbers have leveled off
with no further decline in the past twenty years.
Approximately 20% of the women in high risk groups practice
some form of pica. Those women at greatest risk include
black, southern, rural women from families with a history of
practicing pica during pregnancy. The researchers found
that the data suggesting that pica results in iron
deficiency anemia are not definitive, but pica was
consistently associated with an increase in maternal and
perinatal mortality. Horner et al. suggest that counselors
of pregnant women address the problem of pica in any

nutritional assessment of clients.

 

82

III. Materials and Methods
A. Approval to Conduct This Study

Approval to conduct the preliminary research phase of
this study was granted by the Michigan State University
Committee on Research Involving Human Subjects (UCHRIS) in
July, 1992 prior to the collection of any preliminary data
(Appendix D). Approval from the UCHRIS for the data
collection phase of this research was sought and granted in
July of 1993 prior to the administration of all surveys

(Appendix D).

B. Research Procedures

This research plan has been divided into three phases:
the preliminary research phase, the material development
phase and the data collection phase (Figure 1). Each phase
is described separately.

Preliminapy Research Phase

The reading grade level equivalents of females of
child-bearing age (18-40 years) in a sample of the Michigan
EFNEP population were determined by using the Slosson Oral
Reading Test (SORT) (Slosson, 1963). The readability of
Michigan EFNEP’s written nutrition materials for pregnancy
(Eating Right...for Two, flipchart and handouts) were
determined by using the Flesch (Flesch, 1948), Fog (Gunning,
1950), SMOG (McLaughlin, 1969) and Fry (Fry, 1969)

readability formulas. The median client SORT scores and the

 

83

reading grade level equivalency of the existing written

materials were compared to determine differences.

Material Development Phase To obtain information
needed to develop a new flipchart, an oral question-answer
format was used with four separate groups of EFNEP clients
(Appendix E). Sessions were facilitated by Ann Niewhenhuis,
county Extension home economist in Kalamazoo County.

Clients were asked to restate nutrition information
contained in "Eating Right...for Two" using their own words.
All sessions were tape-recorded and transcribed by the
principal investigator (Karen Schmitz). A second researcher
(Susan Miller) double-checked all transcriptions. The
questions were written by the principal investigator and
designed to cover material in the "Eating Right...for Two"
flipchart. The written EFNEP flipchart for pregnant women,
"Eating Right...for Two," was then rewritten using the
clients’ own language (after the statements were checked for
accuracy by the principal investigator and Dr. Jenny Bond)
to result in a new flipchart (Client-Adapted Language

flipchart).

84

Preliminagy Research Phase

 

Assessment of Assessment of
Reading Level of < - > Readability of
EFNEP Clients Written Materials
(Eating Right...for Two)

Comparison of Reading
Level and Readability
of Clients and Materials

Material Development Phase
Collection of Client

Wording of Nutrition Content
(Client-Adapted Language--
Question/Answer Sessions)

Development
of New Flipchart
(Client-Adapted Language)

Assessment of
Readability of
Client-Adapted Language Flipchart

Development of Surveys: Knowledge, Materials
Acceptability and Comparison Surveys

Pilot Testing of Surveys

Data Collection Phase --Random assignment of clients to
groups:

 

Group 1 ----- > Knowledge Pretest > Treatment: Eating
Right...for Two Flipchart >Knowledge
Posttest >Eating Right...for Two
Handouts >Materials Acceptability Survey—-
-—-> Treatment:Client-Adapted Language
Flipchart > Comparison Survey

 

 

 

 

OR

 

Group 2 ----- > Knowledge Pretest > Treatment: Client-
Adapted Language Flipchart—————>Knowledge
Posttest >Eating Right...for Two
Handouts > Materials
Acceptability Survey > Treatment:
Eating Right...for Two Flipchart
Comparison Survey

 

 

 

 

>

Figure 1. Research Procedures

85

The Client-Adapted Language flipchart was written as
close as possible to the median reading grade level of EFNEP
clients in Kalamazoo County as measured by the SORT. The
mean SORT score could not be calculated because all scores
below nine are given a number score, whereas, all scores of
nine or above are grouped together into one classification
called "high school". This new flipchart was assessed for
readability in the same manner as the "Eating Right...for
Two" flipchart.

During focus group discussions of "Eating Right...for
Two" nutrition information, the subject of pica came up
repeatedly. Many women admitted to practicing pica and
expressed a need for more information on the subject.
Information about pica offered by clients, but not contained
in the "Eating Right...for Two" materials, was not used in
the development of the Client-Adapted Language flipchart.
However, a handout on pica was produced by the principal
investigator using the clients’ own language (Appendix P).
The handout was reviewed by Dr. Carolyn Lackey of North
Carolina State University, an expert on pica. After review,
suggestions were incorporated into the handout and the final
copy was given to the EFNEP (Michigan State University) to
use. The new information on pica was not included in the
Client-Adapted Language materials because it might have
affected the acceptability or interest of the materials

measured by the Material Acceptability and Materials

 

— ._;r—--u-— _, A.“ , . ‘..- ‘ -

 

86
Comparison Surveys.

Three surveys were developed by the investigator:
Knowledge Survey, Material Acceptability Survey and
Materials Comparison Survey (Appendices G, J, L). All
surveys were reviewed by experts in nutrition and pilot
tested with EFNEP clients.

In addition, a readability scorecard was developed to
assess characteristics of readability in written text that
are not measured by readability formulas (Development of
Readability Scorecard Section). The written EFNEP materials
were assessed for ten specific selected factors that affect
readability4 (Appendix H): percent white space, leading
size, number of abbreviations, type size (10 point, 12
point, 14 point, etc.), font type (Roman serif, Helvetica,
etc.), letter type (capital vs. lower case), emphasis on
action (positive vs. negative), use of voice (passive vs.

active), use of symbols and number of headers used.

Data Collection Phase A knowledge survey was
administered (pretest/posttest) to determine which of two
flipcharts produced the greater increase in knowledge of
nutrition during pregnancy among a sample of EFNEP women

(n=331). Instructors were trained to use both flipcharts

 

“In this research project, readability will be defined as
having two parts: 1) the reading grade level of a text, and 2)
other characteristics that affect how easy or difficult a text is
to read (Appendix H).

 

87
(Training of Data Collectors), but used only one flipchart
with each client before the administration of the posttest.
After taking the posttest, clients were given four EFNEP
handouts: Iron, Folic Acid, Protein/Calcium and Discomforts
during Pregnancy (Appendix 0). The clients were instructed
to read the handouts on their own before the next session.
During the next lesson (a different day) clients were asked
to complete the Materials Acceptability Survey to assess the
usability/acceptability of the flipchart they used and the
handouts they read. The clients were allowed to view the
flipchart and handouts while completing this survey. Next,
instructors used the other flipchart, "Eating Right...for
Two" or Client-Adapted Language flipchart, with the same
clients. Both flipcharts were available for review by the
clients when completing the Materials Comparison Survey to
compare the two flipcharts. The instructors read all
questions and response choices on the Knowledge Survey,
Materials Acceptability Survey and Materials Comparison
Survey to minimize the effect of potential poor reading

skills.

C. Sample Selection/Assignment

Women participating in EFNEP cannot be assumed to be
representative of women in the general population who
qualify for EFNEP, but do not participate, because

participants have elected to join the program. Therefore,

 

88
the target population has been limited to women enrolled in
EFNEP. The results of this study are generalizable to all
Michigan women who participate in EFNEP. In the preliminary
phase of the research, all women who were enrolled in EFNEP
in Kalamazoo County in the month of August, 1992 and who
spoke English as a first language were asked to participate
in the SORT to measure reading levels.

In the data collection phase of the research in which
the two flipcharts were compared and the handouts were
assessed for acceptability, all women who were native
English-speakers and who were newly enrolled in EFNEP
beginning October 1st, 1993 in Kalamazoo, Ingham, Oakland,
Kent or Genesee Counties were invited to participate. The
investigator contacted the county Extension home economists
in all counties to set up meetings with instructors to
explain the project. The instructors asked their clients to
participate. It was necessary to have only newly enrolled
clients as subjects in this research to avoid the influence
of knowledge gained from other EFNEP lessons. Recruitment
ended when at least 331 subjects agreed to complete at least
the pretest/posttest portion of the research. This number
of clients (331) was determined to be needed for a sample
that is representative of a Michigan EFNEP population which
is an average of 1700 participants at any one time (Fitz-
Gibbon and Morris, 1987). Although this sample selection

does not involve random selection from all Michigan

89
counties, the self-selected participants were randomly
assigned to experimental or control groups. Furthermore,
the total sample size was large enough to generalize results
to the total target population.

Instructors in Kalamazoo County indicated an interest
in this evaluation project in the summer of 1992 when they
were involved in the preliminary research phase. Meetings
with the county Extension home economist and instructors in
Ingham County and other counties were arranged in the fall
of 1993 to train instructors to use both flipcharts.
Instructors were told that one flipchart would be used to
teach clients prior to the administration of the knowledge
posttest and Materials Acceptability Survey, and the
administration of the Materials Comparison Survey would
occur after use of the other flipchart. Figure 2 indicates
which flipchart was used first. Each client was assigned an
envelope containing the surveys to be completed. The
investigator rolled a die and assigned each envelope, and
therefore, each client to a group. The "Eating Right...for
Two" flipchart was to be used first with clients randomly
assigned the numbers one, three or five by the roll of a die
and the Client-Adapted Language flipchart was to be used
first with clients assigned the die numbers two, four or
six. All clients were assigned to one of the two groups by

the roll of a die.

90

Client Die No. FlipchartZGroups

1 5 EFNEP flipchart

2 6 Client-Adapted flipchart

3 3 EFNEP flipchart

4 4 Client-Adapted flipchart
Figure 2. Example of Assignment of EFNEP Clients

to Research Groups

This random assignment of clients to groups distributed the
effects of individual differences between subjects to the
two groups. Therefore, the change in knowledge scores
between the pretest and posttest of the two groups could be
compared because the groups were equivalent.

This design provides information about change in
knowledge within and between groups, while controlling
several threats to validity. The effects of history,
maturation, attrition and regression were assumed to be
distributed equally between the groups because random

assignment of subjects to groups was used.

D. Research Sites

Reading level of a sample of the target population was
tested by program instructors in Kalamazoo County under the
direction of the county Extension home economist. The SORT
was administered by the instructors during a regularly

scheduled lesson (Training of Data Collectors). To compare

91
the two flipcharts and assess the usability of the handouts
in the data collection phase, data were collected by
instructors in the client’s home or regular meeting place in
five counties: Kalamazoo, Oakland, Kent, Ingham and Genesee

(Data Collection, Materials Assessment).

E. Program Description

A description of the Expanded Food and Nutrition
Education Program (EFNEP) is outlined in the "Review of
Literature" section. The primary curriculum used by the
program is "Eating Right is Basic" (Michigan State
University Cooperative Extension Service, 1992). For
pregnant women enrolled in the program, this curriculum is
supplemented with materials about nutrient needs during
pregnancy, "Eating Right...for Two" (Michigan State
University Cooperative Extension Service, 1992). This
supplemental lesson consists of a flipchart and six handouts
(Appendix 0). The flipchart covers general health and
nutrition issues in pregnancy, whereas, the handouts cover
specific topics: assessment of a sample diet for adequacy
during pregnancy, iron, folic acid, calcium and protein
information, discomforts during pregnancy, and a weight gain
grid. The weight gain grid and the assessment of a sample
diet were not analyzed for readability because of the lack

of written text. In the usual EFNEP instruction, the

92
instructor reads the flipchart to the client as the client
looks on, but the handouts are generally left with the
client to read unassisted by the instructor. Both "Eating
Right is Basic" and "Eating Right...for Two" were originally
developed by the Michigan EFNEP in 1985 and revised in 1992.
These materials are currently used in 47 states, Guam,

Canada and the Virgin Islands.

F. Selection of the Reading Levels Test

The reading levels of the population were assessed by
use of the SORT (Appendix B). This test was given
individually to the women in their homes by their
instructors to minimize embarrassment or anxiety associated
with limited reading skills.

Readability of text has been measured manually using
the SMOG and Fry tests, and by computer using the Flesch and
Fog formulas (Review of Literature). These tests are widely
used, so scores are meaningful to many professionals who
might benefit by reviewing the data. The scores produced by
these tests correlate well with SORT reading level scores.
One limitation of the tests is that they are only useful if
the reading levels of the population are known. However, in
this research the reading levels of a sample of the target
population were measured and, therefore, this limitation

does not apply.

93

G. Instrument Development

In this evaluation, the following variables were
assessed using instruments developed by the investigator:
knowledge of nutrition in pregnancy [Pregnancy Survey I
(pretest) and Pregnancy Survey II (posttest), Appendix G],
characteristics of readability other than reading grade
levels of the two sets of written materials (Readability
Scorecard, Appendix H), usability/acceptability of the
written materials (Material Acceptability Survey, Appendix J

and Materials Comparison Survey, Appendix L).

1. Development of the Knowledge Survey
(pretest/posttest)

A Knowledge Survey (Appendices F and G) was developed
to measure knowledge change (pretest/posttest) for the EFNEP
objectives developed by EFNEP program planners for the
content area "pregnancy". This instrument was measured for
readability and was found to be written at the median
reading grade level of the sample of EFNEP clients,
according to the results of the Flesch formula (Table 15).
Therefore, it was not necessary to rewrite the survey at a
lower reading level. The Knowledge Survey was administered
individually to women in their homes, before and after the
nutrition lessons were provided, by the teacher with whom
they were working. A limitation of this instrument is the
potential variation in administration and assistance

(prompting) provided by instructors. Consistent training of

 

 

94

instructors helped to minimize this limitation.

The following steps were used to construct the
instrument:
a. Competencies of the EFNEP curriculum "Eating
Right...for Two" were obtained to use for the

development of questions. The competencies are:

i) To describe the importance of slow and steady
weight gain during pregnancy.
ii) To identify the changes in the number of

servings needed daily from the milk group
during pregnancy.

iii) To determine if a meal plan includes the
recommended number of servings from the food
groups for a pregnant woman.

iv) To identify iron, folic acid, calcium and
protein as nutrients that are needed in
additional amounts during pregnancy.

b. Two questions were developed for each competency.

c. The instrument underwent expert review to
determine content validity by Dr. Jenny Bond, Dr.
Anne Murphy, Dr. Wanda Chenoweth, Dr. Sandra
Andrews, Dr. Sharon Hoerr, and Dr. Mary Noel
because these women have expertise in one or more
of the following areas: familiarity with EFNEP,
knowledge of nutrition in pregnancy, data
analysis, or survey development.

d. The survey was assessed for reading grade level

equivalency using the Flesch, Fog, Fry and SMOG

tests (Table 15).

95

e. The instrument was pilot-tested with 5 EFNEP
clients in Ingham County and 7 EFNEP-eligible
women in Kalamazoo County (Pilot Test).

f. An item analysis was conducted to assess
difficulty, discrimination and reliability.

9. Based on the results of the expert review, the
readability tests, the pilot tests and the item

analysis, the instrument was revised (Appendix G).

2. Development of the Readability Scorecard

Characteristics of readability of written materials:
number of abbreviations, font type (Roman serif, Helvetica,
etc.), type size (12 pt., 14 pt., etc.), use of voice
(passive vs. active), emphasis on action (positive vs.
negative), percent white space (in margins and around
headers), size of leadings,'type of letters (capital vs.
lower case), use of symbols and number of headers used were
selected for use in the scorecard by the principal
investigator, after extensive review of the literature on
characteristics of readability (Review of Literature) and
after consulting other nutritionists with experience using
readability: Dr. Jenny Bond, Char McKay of EFNEP and Dr.

Anne Murphy.

 

SLeading (pronounced with a short e) is the size of the space
between lines of text.

96

Characteristics of Readability: Definitions and

Recommendations

Table 5:

fl Characteristic

Definition or
Example

 

9:13:11
Recommendation1

 

1) Use of voice

Number of sentences
using passive voice
per total number of
sentences; expressed
as percent;

Example of passive:
The ball was thrown
by the boy.

Example of active:
The boy threw the
ball.

Use active
voice

 

2) Type size

Size of type: 8
point, 10 point, etc.

Use a type size
2 12 points

 

+ 3) Font type

Roman serif or no
serifs

Use Roman serif
type

 

4)
Abbreviations

Number of abbreviated
words per total
number of words;
expressed as percent

Do not
abbreviate
words

 

5) Emphasis on
action

Number of sentences
emphasizing negative
action per total
number of sentences;
expressed as percent;
Example of negative
emphasis: Do not eat
candy for a snack.
Example of positive
emphasis: Eat fruit
instead of candy for
a snack.

Emphasize the
positive, as
opposed to the
negative,
action

 

6) Type of

letters

 

 

Capital or lower case
letters

 

Use lower case
except for the
first letter of
the sentence h

 

 

 

1 Recommendations have been reported by experts in education

and nutrition education: Sticht, 1975; Klare, 1976; Laubach and
Koschnick, 1977; US DHHS, 1982; Doak et al., 1985; Fry, 1988;
Nitzke, 1989.

97

 

Table 5 (continued)

7) Symbols

Number of symbols per
total number of
words; expressed as
percent;

Examples of symbols:
/ or \ or @ or +.

Avoid using

symbols such as
’+’, write out
’plus’ instead

 

8) Headers

Number of headers per

Use of headers

 

and
headers

total number of increases
paragraphs; expressed. readability; no
as percent specific
recommendation
9) White space: Any area of the text IReader

in margins
around

that does not contain
written words
(pictures are
considered white
space) and is not
part of the leading.

perceives white
space as making
a text easier
to read; no
specific
recommendation

 

10) Leading

 

Size of the space
between lines; This
text is written with
2 point leading.

 

 

Use 3 2 point

98
Characteristics were chosen for the scorecard based on three
criteria:

i) Research has shown that the characteristic affects
readability or is perceived by the reader to
affect readability (Review of Literature).

ii) Assessment of the characteristic is not
subjective.

iii) The characteristic is not assessed by the Fog,

Flesch, Fry or SMOG readability formulas.

Ten characteristics were defined, compared to research
recommendations (Table 5) and assessed according to
procedures outlined in Table 6. All assessments were
checked by a second researcher (Susan Miller).

The assessment procedures for estimating percent white
space were outlined after consultation with Dr. Darcy Greene
of the College of Journalism, an expert in layout of text.
Dr. Greene stated that there is no formula for determining
white space, but in journalism two aspects of white space
are considered when assessing the readability of text:
white space in the margins and around the headers, and white
space between lines of text (leading). Therefore, these two
aspects of white space are assessed by the Readability

Scorecard (Table 6).

99

Table 6: Characteristics or Readability: Assessment

 

Characteristic Description of Assessment1

1) Use of voice The number of times the passive voice
(PV) was used, rather than active
voice, were counted. Total number of
sentences were counted. Use of voice
= PV/total number of sentences.

 

2) Type size The size of type of the main body of
the text was compared to known type
sizes printed in Laubach and
Koschnick, 1977.

 

 

3) Font type The font type of the main body of the
text was compared to Roman serif type.
The type was determined to be either
Roman serif or not Roman serif.

4) Abbreviations The number of abbreviations (AB) in
the text were counted. Contractions of
words were not counted as
abbreviations. The total number of
words were counted. Percent of AB =
(total number of AB/total number of
words) x 100

 

5) Emphasis on The number of times negative action

action (NA) was emphasized, as opposed to
positive action, in a command or
suggestion were counted. Any sentence
telling the reader what pep to do was
counted. Percent NA = The total
number of NA/total number of
sentences.

 

6) Letter type The main body of the text was
examined. It was noted that the text
was written in either capital letters
or lower case letters (excluding the
first letter of each sentence).

7) Symbols The total number of symbols (NS) used
in the main body of the text were
counted. Some examples include "+",
"%" and "&". Percent symbols
=(NS/total number of words) x 100

 

 

 

 

 

1 Total number of words, total number of sentences, total
number of paragraphs and number of times passive voice is used can
be determined using the computer program, Grammatik III.

Table 6 (continued)

100

 

8) Headers

 

The number of headers (H) used in the
text were counted. The title was
counted if it appeared on a written
page of text being analyzed. The
number of paragraphs (P) in the text
were counted. Percent headers = (H/P)
x 100

Note: Computer programs count a new
paragraph any time there is a break in
the leading. For accuracy, you might
want to count paragraphs by hand.

 

9) White space:
Margins
and area
around headers

Boxes were drawn around blocks of text
and around headers. Boxes were drawn
as close to the text as possible, but
boxes were always in a rectangle
shape. Any time there was a space
greater than the leading of the main
body of the text, a box was finished
and new box was started. The area of
the page was calculated (length x
width). The area of the boxes were
calculated and totaled. The percent
white space = [1 - (total area of
text/area of page)] x 100.

 

 

10) Leading

The leading of the main body of the
text was compared to known leading

 

 

a

sizes (Laubach and Koschnick, 1977).

 

101

Thirty-four handouts on nutrition written at different
reading grade levels were assessed using the descriptions of
characteristics in Table 6. The handouts were written text
that discussed various nutrition-related issues and were
distributed by different organizations (i.e. EFNEP, American
Heart Association, March of Dimes). The results of the
assessment of these texts were used to develop the
scorecard. The texts ranged in reading grade level from 5
to 12. The mean and range (or other description) of
measurements for the characteristics in the 34 texts are

displayed in Table 7.

Table 7: Mean Scores and Range of Scores, or Descriptions
of Characteristics of Readability Used for
Development of the Readability Scorecard

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Characteristic Mean Range Description
1)Use of voice 9.2:9.0% 0 - 50%
2) Type size 11 point 8 - 18 pt
3) Font type --------------- 23 Roman serif
11 without
serifs
4)Abbreviation 1.312.2% 0 - 8%
“5) Emphasis on 1.8 i3.6 0 - 12%
action
6) Letter type ---------------- 34 lower case
0 capital
7) Symbols 1.011.6% 0 — 7%
8) Headers 56132% 11 - 100%
9) White 58118% 30 - 89%
space: margins
10) Leading 1.3 points 1 - 2 pts
“F: ==

 

 

 

 

 

 

102
For characteristics that could be measured, the mean
and standard deviation of each characteristic were
calculated. For characteristics that had set
recommendations from research, could be measured and the
measurements expressed as percentages (use of voice,
emphasis on action, abbreviations and symbols), points on

the scorecard were given as follows:

2 points = The measurement met the
recommendation.

1 point = The measurement was between (or
equal to) the mean and the
recommendation.

0 points = The measurement was above the mean.

For characteristics without specific recommendations from
research, but which can be measured and expressed as
percentages (headers, white space in margins and around

headers), points on the scorecard were given as follows:

2 points = The measurement was equal to the
mean or above the mean.

1 point = The measurement is within one
standard deviation below the mean.

0 points = The measurement is below one

standard
deviation below the mean.
For characteristics that had set recommendations from
research and could be directly compared to a standard (type
size, font type, letter type and leading), points on the

scorecard were given as follows:

2 points = The characteristic met the
recommendation.

103

0 points = The characteristic did not meet the
recommendation.

The scorecard was pilot tested for usability with five
graduate and undergraduate students. Corrections were made
when directions were found to be unclear. The final version
of the Readability Scorecard is in Appendix H. The
Readability Scorecard needs to undergo further testing for
reliability and validity to design a meaningful scoring
system. For the purpose of this research project, however,
scores were assigned to one of three categories for

interpretation:

If the text scored:

17-20 The text contains many characteristics that aid
readability.
13-16 The text contains some characteristics of

readability, but might need to be reviewed to
further increase readability.

<12 The text contains few characteristics of
readability. It should be reviewed and improved
to increase readability.
3. Development of Materials Acceptability and
Materials Comparison Surveys
The opinions of EFNEP clients on the different
flipcharts and the "Eating Right...for Two" handouts were
obtained using two surveys, the Materials Acceptability
Survey (Appendix J) and the Materials Comparison Survey
(Appendix L). The Materials Acceptability Survey was given

to the client after the Knowledge posttest and applied only

104
to the acceptability/usability of the flipchart used in the
knowledge assessment portion of the investigation and the
"Eating Right...for Two" handouts. Then the Materials
Comparison Survey was administered after use of the other
flipchart. Questions regarding acceptability, interest,
appropriateness of language used, and related
characteristics of the materials were developed by the
researcher and pilot-tested for usability with 5 EFNEP
clients in Ingham County and 7 EFNEP-eligible women in

Kalamazoo County (Data Collection).

H. Training of the Data Collectors

1. For Reading Level Assessment--During July, 1992 the
investigator met with the instructors from EFNEP in
Kalamazoo County. They were given a set of instructions
(Appendix M) tape recorder, cassette tapes, ten lists of
twenty words (Appendix B) and a letter asking for client

consent to participate (Appendix N). Training consisted of:

a. practice using the tape recorders
b. reading through the instructions and

c. practicing the SORT on other instructors.

The instructors were asked to explain to their clients
that the researcher was trying to improve the EFNEP

materials, but to rewrite the materials she needed some idea

105
of the reading level of the clients. Clients were told that
they would not be identified. After reading the letter
asking for consent and obtaining consent from the client to
participate, instructors would ask the clients to read the
words from list one through list ten into the tape recorder.
Clients were told to stop when they could not correctly
pronounce five consecutive words. After testing the
clients, tapes were turned in to the county Extension home
economist and mailed to the principal investigator. All
reading grade level scores were determined by the principal
investigator using the tape recordings and directions on the
SORT (Appendix B). All scores were verified by a second
researcher, Susan Miller. When discrepancies in scoring
occurred, the tapes in question were reviewed by the
researchers and errors were corrected.

2. For Materials Assessment--Instructor training for
teaching the two versions of materials began in the fall of
1993. Instructors were told that they would teach and test
their own clients to avoid EFNEP client anxiety that might
result from an outsider conducting these procedures.
Training consisted of:

a. The principal investigator reading the
directions for instructors’ procedures
(Appendix M).

b. The principal investigator and instructors

reviewing the curriculum by reading through

106
the text of each curriculum.

c. The principal investigator demonstrating
procedures for teaching the materials
following the procedures outlined in Appendix
M.

d. The instructors teaching the curriculum to
each other.

e. The principal investigator and instructors
reviewing the Knowledge Survey directions,
questions, and answer choices by reading
through the survey (Appendix G).

f. The principal investigator and instructors
reviewing the Materials Acceptability and
Comparison Surveys questions, answer choices
and directions by reading through the surveys

(Appendices J and L).

In the intervention, the messages on the flipchart were read
to the client page by page with the client looking on. The
"Eating Right...for Two" handouts were not read to clients,

but were left with the clients to read on their own.

I. Data Collection
1. Reading Level Assessment—-During August, 1992, EFNEP
instructors tested their clients for reading level using

SORT. Each female EFNEP client in Kalamazoo County between

107
the ages of 18 and 40 years who spoke English as her first
language was asked to participate. Following informed
consent, words were read by the client into a tape recorder
according to the procedures (Training of Data Collectors).
All data were recorded and mailed to the principal
investigator for analysis.

2. Pilot testing of the Knowledge Survey-—After the
assessment of reading levels was completed, arrangements for
the pilot test began. The county Extension home economists
in Ingham County (Margaret Bucklin) and Kalamazoo County
(Delores Hayden) were contacted about using a group of
enrolled clients or EFNEP-eligible women to evaluate the
wording of the surveys: Knowledge and Material
Acceptability Surveys. The wording on the Materials
Comparison Survey is similar to the wording on the Materials
Acceptability Survey, so there was no need to pilot test the
Materials Comparison Survey separately. One group of 5
women in Lansing and 7 women in Kalamazoo County were
located and the women agreed to participate. The procedures
were as follows:

a. The principal investigator distributed surveys and

pencils, welcomed the participants and made

introductions.
b. The procedures were explained to the participants
by the principal investigator: "I’ve developed 3

surveys to be used with women like yourselves

108
after they have had a lesson in nutrition during
pregnancy. Now I need to get your opinions on the
wording of questions and answers in order to
improve the surveys. Although you might or might
not have used materials on nutrition in pregnancy,
you will be asked to complete one survey and look
at the question and answer choices on all 3
surveys. No names should be written on the
surveys and any responses will not affect
participation in EFNEP. I would appreciate any
comments you can give me."
To begin the participants were asked to look at
Pregnancy Survey I (Knowledge Survey). The
principal investigator read the directions, each
question and each answer choice. The client
selected the answer that she thought was the most
correct. The investigator told the clients that
she realized that they might not have had the
lesson that goes with the survey questions, but
they should try to circle the correct answer.
The clients took the Knowledge Survey.
There was a break for food.
Each question in the Knowledge Survey was reviewed
by the principal investigator and the clients.
Clients were asked for general responses to survey

items and then asked specific questions:

i)

ii)

iii)

iv)

109

Are there enough categories for question 4?

Did anyone find the question confusing?

Is question 5 clear or is it confusing the

way it is written?

Any problems with the directions?

Questions 1 through 8:

1)

2)

3)

4)

5)

Is it clear that you need to total
servings of milk + cheese + yogurt? Is
the word "serving" confusing? When I say
a serving of milk, do you think about a
glass of milk? a slice of cheese?

By "daily meal plan" I mean all the food
you eat in a day including snacks,
breakfast, lunch and dinner. Is there a
better way to say that?

Is there too much information in
question 3? Any confusing words?
Problems?

Do you think I should use the word
uterus or would it be better to use
womb? Should I use getting larger or
growing?

What do you think the word "nutrients"
means? Is it clear that I mean the
nutrient that stays at the same level as

before pregnancy?

6)

7)

8)

110
Is the "s" in parentheses confusing? Is
the response "a and b only" confusing?
All of the above?
Any problems with #7? I want to know the
best word or words that go in the blank.
Is question 8 too long? Too much

information to follow?

Material Acceptability Survey: The principal

investigator read the directions, each question

and each response aloud. After reading each

survey question, she asked the participants for

comment S :

1)

2)

4)

Are the directions clear? Would you use
the word "materials" or "flipchart" and
handouts? What do you call this
nutrition program?

Are the agree/disagree response choices
acceptable or is there a better way to
ask for your opinion?

What do you think of the word
"insulting"? Should the question read:
"talked down to?"

Any problems with #3? Should the
question read: "neighborly" or
"friendly"? "conversational"? "familiar

words"?

111

5) Would you use "nutrition information" or
"nutrition recommendations for pregnant
women" or "nutrition needs?"

6) In your opinion, is "easy to read"
different than "easy to understand?"

7) Is the word "useful" clear? Would you
say "worthwhile reading" or

"informative" or "helpful"?

8) What do you call your nutrition
instructor?
9) Any problems with 10 or 11?

10) Do you understand the word "appearance"?
Would you say "arrangement"? Would you
use the term "blank space" or "white
space"?

11) Would you say "print"? "type"?
"letters"?

12) Are there other answer choices you would
select for the question about the

pictures?

Materials Comparison Survey: The principal
investigator read the directions and the response
choices aloud, then asked the clients for their

opinions on the:

112

1) Directions
2) Response choices
i. Construct validity was assessed for the Knowledge

Survey by calculating a discrimination index of
items completed by the twelve pilot test
participants (Table 14). Items with a low score
(< .33) would have been revised or omitted, but no
items scored below .33. Items were assessed for
difficulty by calculating the percent of people
who answered each question correctly (difficulty
index, Table 14). Easy items (>80%) and difficult
items (>17%) were reviewed by the principal
investigator. Only one item scored below 20%
(item #2 = 17%). Although the item was difficult,
it was not removed or revised because it met a
teaching objective for EFNEP and this research
project. Other indicators of difficulty are the
reading grade level of the questions and the
reaction of the participants in the pilot test to
the questions. Reliability of the knowledge test
questions was assessed by using the KR-20 formula

(Fitz-Gibbon and Morris, 1987).

3. Materials assessment--The following steps were used
in obtaining data for the materials assessment.

a. During March and July of 1992 meetings with EFNEP

113
clients were held at a church in Kalamazoo to
collect information for the development of the
Client-Adapted Language flipchart (Research
Procedures). Questions were asked by Ann
Nieuwenhuis of EFNEP (county Extension home
economist) because she had developed a comfortable
relationship with the clients after leading them
in a six week parenting class.
Instructor training for teaching the two versions
of materials began in the fall of 1993. For
details on instructor training, see Training of
the Data Collectors.
Instructors pretested clients’ nutrition knowledge
according to directions on the Knowledge Survey.
The Knowledge Survey, when given as a pretest,
included demographics: ethnic background, age,
education level and pregnancy status (Appendix G).
Instructors used one flipchart with the clients
according to the results of the random assignment
(Existing EFNEP or Client-Adapted Language
materials).
Instructors administered the Knowledge posttest to
clients according to the directions on the
Knowledge Survey (Appendix G). "Eating
Right...for Two" handouts were left with the

clients to read on their own before the next

 

114

lesson.

On a different day, instructors administered the
Materials Acceptability Survey to the clients
(Appendix J).

Instructors used the other flipchart with the

clients.

Instructors administered the Materials Comparison

Survey to the clients (Appendix L).

115

IV. Results and Discussion

A. Preliminary Research Phase
1. Client Reading Level: During August and

September of 1992, all clients enrolled in EFNEP in
Kalamazoo County, Michigan (n=38) were measured for reading
levels using the SORT (Appendix B). All clients were tested
individually by their regular EFNEP instructors. Each test
took less than 10 minutes and was tape-recorded. The tapes
were given to the Kalamazoo county Extension home economist
and mailed to the principal investigator who scored all the
tests. Scores were double-checked by a second researcher,
Susan Miller. The SORT scores are reported as grade levels
below ninth grade. Anyone reading at the ninth grade level
or above is grouped into a single category called high
school. The results of the SORT are presented in Table 8.
The median SORT score, the range of SORT scores, the percent
of subjects reading at and below the high school level, and
the mean SORT score of subjects reading below the high
school level are reported (See Table 9).

Estimates of the percentage of illiterate American
adults as defined by reading below the sixth grade level
range from 10 to 13% (Harman, 1987). In this study, the
results of the SORT showed only 7.8% of a sample of EFNEP
clients read below the sixth grade level, despite the fact
that poverty is a risk factor for illiteracy (Nitzke, 1989)

and 91% of Michigan EFNEP homemakers have annual incomes

116
below $13,000 (Michigan Expanded Food and Nutrition
Education Program, 1990). The Michigan EFNEP might not
attract poor readers who have had bad experiences with

previous learning situations.

Table 8: Preliminary Research Results: The SORT Data

Grade Level Score Number of Clients[(%)
1.6 (2.6%)
(2.6%)
(2.6
(5.3%)
(2.6 )
(2.6%) ll
)
)

 

 

.h

\oooqmwmtoqmwwpwww

 

 

 

 

 

 

 

 

 

 

 

 

m (n a: m (n a: q ~q \J O ~4 ox m in
+4 F’ N to DJ u) N .e +4 F‘ H +4 K) F‘ H i4
H
o
m
°\O

 

9 11 (28.9%)

N

 

 

 

 

Total 38

117

Table 9: Preliminary Research Results: Summary of SORT
Data

Median Score

 

 

 

Range of Scores 1.6 to > 9 (high
school)

Number of subjects with scores n = 27, (71%)

below the ninth grade reading level

Number of subjects with scores n = 11, (29%)

above the ninth grade reading level

Mean score of subjects reading 7.4 i 1.6 s.d.
below the ninth grade_reading level

 

2. Correlation Coefficients Correlation coefficients were
determined for SORT scores of the clients’ reading below the
high school level (dependent variable) and the independent
variables:

a) Clients’ education level

b) clients’ age
The SORT scores are restricted to those scores below the
high school level because only individuals reading below the
ninth grade level are given a number score. These data are

displayed in scattergrams labeled Figure 3 and Figure 4.

118

Client Education Level

 

 

ex on '5 t3 3 a
0 L l l l
2—
CD
8
o ..l
m 4
E-‘
m .
(D
m 6~ o 0
H o
t:
O
2:: t
O 0 : o o
8‘ O O ‘
; .
I

 

 

10

Figure 3: Client Education Level vs. Client SORT Scores

 

 

 

 

10
: 8 . I c

3‘ ° . : .
U)
1‘5) 0
o 6-
a:
[—
fig .
a: 4-
E
.2
L) 2-

0 I I I i

2 s a s a 3
Client Age

Figure 4: Client Age vs. Client SORT Scores

119

The correlation coefficient for the client education
level and SORT score was -.1901 and not statistically
significant at the p<.05 level. This result agrees with the
results of a 1956 study by the National Tuberculosis Society
which showed little correlation between self—reported
education levels of adults and actual reading ability
(Mohammed, 1964). In addition, Ruud et al. (1993)
interviewed a group of 152 men and women between the ages of
16 and 60. Individuals reported their highest level of
education completed ranged from 6th to 12th grade, but their
measured reading levels ranged from 3rd to 8th grade.

The correlation between age and SORT score was -.5807,
a number which is significant at the p<.001 level. This
result shows a relationship between lower reading levels and
increasing age. In 1961, Samora et al. tested 125 low-
income hospital patients for comprehension of health—related
words and found that patients of high school age could
identify more words than older individuals. After finishing
formal schooling, individuals from this population might not

read on a regular basis, allowing their skills to decrease.

3. The Readability of EFNEP’s Written Materials The

results of using four formulas to measure the readability of
Michigan EFNEP’s written materials for nutrition during

pregnancy are shown in Table 10. The Flesch and Fog scores

120
were determined by the principal investigator using the
computer software program, Grammatik III (Grammatik III,
1989). The Fry and SMOG scores were determined by hand by
two separate investigators. Results indicated that scores
ranged from a Fog score of 12th grade on the "Iron" handout
to a Fry score of 6th grade on the "Folic Acid" and

"Calcium/Protein" handouts (Table 10).

Table 10: The Reading Grade Levels of EFNEP’s Written
Materials for Nutrition During Pregnancy

 

 

 

 

 

 

 

 

 

 

 

 

I Material Flesch Fog SMOG Epy
FIEating Right...for Two 8 10 10 7
Flipchart
Discomforts During 7 8 8 7
Pregnancy
Special Nutrient Needs: 8 8 8 6
Folic Acid
Special Nutrient Needs: 7 12 7 7
Iron H
Special Nutrient Needs: 7 10 8 6
Calcium/Protein

 

At the 50 to 75% comprehension level (Fry and Flesch
scores), the reading grade level scores of EFNEP’s written
materials for nutrition during pregnancy tended to be very
near the median reading grade level of the EFNEP clients.
The SMOG scores (90% comprehension) were near the median

SORT scores for the handouts, but were higher for the

121
flipchart. In many cases, the Fog scores were higher than
the Fry and Flesch scores, but this is typical because the
Fog, similar to the SMOG which also gives higher reading
grade levels, considers any polysyllabic word as difficult
instead of using a specific word list (Laubach and
Koschnick, 1977).

The median SORT score of 8.3 for reading levels of
EFNEP clients is higher than the Fry, Flesch and SMOG scores
for the handouts. Using the SMOG readability formula as a
predictor of reader comprehension, based on the results of
the SORT over half of the EFNEP clients sampled could be
expected to comprehend 90% of the material written in the
handouts. However, an EFNEP client with a SORT score of 8.3
or above could probably comprehend only 75% (Flesch score of
8) of the written material in the flipchart. At the 90%
comprehension level, a reader would need a SORT score at the
high school level when reading the flipchart. Only twenty—
nine percent of the EFNEP clients sampled read at this

level.

B. Material Development Phase
1. Client-Adapted Langpage Flipchart During March and July

of 1992, four focus groups of 8 to 12 women each were used
to collect restated nutrition information contained in
EFNEP’s "Eating Right...for Two" flipchart and handouts.

All four groups consisted of EFNEP-eligible pregnant women

122
from Kalamazoo County who had just completed a parenting
class, including nutrition lessons. The women were asked a
set of questions (Appendix E) and encouraged to discuss
answers. Originally, the questions were based on material
in EFNEP’s "Eating Right...for Two“ flipchart and handouts,
but questions were added to cover the subject of "pica" when
the discussions turned in that direction. Pica was common
among the women in the focus groups, but few women expressed
any concern that their practices might be harmful. .
Therefore, a handout was developed by the investigator on
the subject of pica (Appendix P). The handout was reviewed
by Dr. Carolyn Lackey of North Carolina State University, an
expert on pica. It was edited and given to the Michigan
State University EFNEP office.

The sessions were tape-recorded by the principal
investigator. These recordings of the answers to questions
on EFNEP’s materials for nutrition during pregnancy were
edited for incorrect statements by the principal
investigator and Dr. Jenny Bond, and used by the principal
investigator to write the Client-Adapted Language flipchart.
The EFNEP flipchart and the Client—Adapted Language
flipchart have the same layout, number of pages and
pictures, only the words have been changed (Table 11).

In Table 11 and in the discussion of the wording, the
original "Eating Right...for Two" flipchart is referred to

as "Version X" and the Client-Adapted Language flipchart is

123
referred to as "Version Y". In some cases when there were
no suitable substitute words offered by the participants,
the original wording was maintained. Exact quotes from the
participants used in version "Y" are in brackets [] instead
of quotation marks, so they will not be confused with
quotation marks already present in the "Eating Right...for

TWO" text .

124

Table 11: Wording of the Two Flipcharts

 

 

I?
Page1 version Wording "
2M X Now that you are pregnant you may

hear that you are "eating for two."
That’s right--you are eating for you
and your baby.

 

'Y Now that you are pregnant [you’re

eating for two. That’s the way you
should look at it.] That’s right——
[the baby is feeding off] the food

 

you eat.
4M X Even if you feel that you are I
overweight, it is important that you

gain weight during your pregnancy.

 

Y [Even if you feel "fat", you need to
gain weight for the baby’s sake] when
you are pregnant.

 

6M X Adequate weight gain during pregnancy
increases the chance of having a q
healthy baby.

 

Y [You need to gain weight during
pregnancy. If you do, you have a
better chance of having a healthy,
normal weight baby.]

 

8M X How much weight should you gain
during pregnancy? Everyone is
different so there are no rules.
Generally, though, women who gain 22-
30 pounds have a better chance of
having a healthy baby.

 

Y How much weight should you gain
during pregnancy? [It depends on the
person, so talk to your doctor. The
doctor will tell you what you should
weigh and what you should gain. Most
women should gain 22 to 30 pounds
throughout the whole pregnancy.]

 

 

 

 

 

 

1The pages of both flipcharts were numbered from 2M to 40M
because the original "Eating Right...for Two" was numbered in that
fashion.

125

Table 11 (continued)

 

10M

X

You probably will not gain-more than
2-4 pounds during the first three
months of your pregnancy.

 

In the first three months of
pregnancy [you won’t gain much
weight. It will probably not be more
than 2 to 4 pounds because you have a
long way to go! ]

 

12M

After three months, a good pattern to
follow is to gain about 1 pound per
week until your baby is born. You
won’t hit 1 pound "right on the nose"
every week. However, the important
thing is a slow and steady weight
gain.

 

After three months, [you should gain
gradually through the next six
months.] This gradual weight gain is
called slow and steady weight gain.
[Sometimes you might gain a pound a
week. Sometimes you might gain a
pound and a half.]

 

 

14M

The baby grows as you gain weight.
The baby begins about the size of a
grain of sand and grows to be about 7
pounds or more before it is born.

In addition, the mother’s uterus, or
womb, grows to make room for the
growing baby. Breasts also get
bigger in preparation for breast-
feeding. The amount of blood
increases in the body to help get
food and oxygen to the baby.

 

 

 

 

When you gain weight in pregnancy,
[the weight can go all over. Your
bottom, hips, thighs, face and even
your feet can get bigger. Your
breasts get bigger because of the
milk. The uterus gets bigger to make
room for the baby. The baby is
growing.] Blood is increasing to
help get food and oxygen to the baby.

 

 

126

Table 11 (continued)

 

16M

X

During pregnancy a lot of changes
take place inside your body. Weight
gain is important for these changes
to occur. These changes help to
support the growth and development of
the baby.

 

 

[Pregnancy affects every part of your
body. Weight gain is important for
both the baby and the mom.) These
changes help to support the baby who
is growing and developing.

 

18M

Dieting during pregnancy is harmful.
There is plenty of time to diet after
the baby is born.

 

[Don’t go on a diet during pregnancy.
It’s not a good idea because you can
rob the baby of certain nutrients.
When you diet you tend not to eat
what your supposed to. Just eat the
proper foods and watch your weight.]
You want to gain the amount of weight
that your doctor recommends.

 

20M

What you eat during pregnancy is
important not only because it
provides nutrients for the baby, but
also because it helps you look and
feel good. Eating a variety of
nutritious foods makes pregnancy and
delivery easier.

 

What you eat during pregnancy is
important. [The baby gets nutrition
from your body and you need nutrients
for your own body. Eat 3 meals a
day. You need your strength so you
won’t feel all weak and stuff.]

 

22M

 

 

 

How can you be sure you’re getting
enough of the right nutrients?

By planning your meals to include a
variety of foods.

 

 

Table 11

127

(continued)

 

Y

How can you be sure you’re getting
enough of the right nutrients?

By [eating foods from different food
groups, like vegetables, fruits,
meat, bread and cereal.] Make
[snacks that can be nutritious, too.
Take finger foods like grapes,
muffins or juice boxes with you when
you go places.]

 

 

24M

How do nutrients get from the food
you eat to the baby?

The nutrients in foods that you eat
go directly to the baby through the
placenta. Therefore, it is important
to eat a variety of nutritious foods
daily so that the baby’s and your
nutritional needs are met.

 

How do nutrients get from the food
you eat to the baby?

The nutrients go [through the little
thing in the womb called the
placenta. Food goes through the
placenta to the baby. Eat healthy so
both baby and mom can get the
nutrients they need.]

 

26M

During pregnancy, you and your baby
need extra nutrients. This means
that, in addition to your regular
nutrient needs, you need to eat extra
amounts of foods high in protein,
calcium, iron and folic acid.

 

During pregnancy, you and your baby
need extra nutrients. [You need to
eat more foods with iron, folic acid,
protein and calcium.]

 

28M

Your baby needs calcium for healthy
bones and teeth. To get all the
calcium needed, increase milk and
cheese servings to 3 or more per day
during pregnancy.

 

 

 

 

During pregnancy, you need more
[calcium for baby’s bones and teeth.
Eat] at least 3 servings or [more
cheese, yogurt, ice cream, milk,
cottage cheese and tofu] everyday.

 

 

Table 11

128

(continued)

 

[ 30M

X

Two servings of meat or dried beans
provide needed protein. Protein is
necessary for every part of the
baby’s growing body, especially brain
development.

 

[You need 2 servings] daily [of meat,
eggs, beans, or peanut butter for
protein.] Protein is necessary for
every part of the baby’s growing
body, especially brain development.

 

32M

 

In addition to calcium and protein,
the need for two other nutrients
increases greatly during pregnancy.
One of these nutrients is iron. Iron
is needed to form blood, which
carries oxygen. Because both the
mother’s body and the baby’s body
need oxygen, blood volume increases
during pregnancy. As blood volume
increases, the need for iron also
increases.

 

[You need iron to build more blood]
during pregnancy, [so you don't get
anemic (low blood). Blood carries
oxygen to the mother and baby.] To
get more iron you need to eat more
[foods like liver, broccoli, meat and
fish.]

 

34M

The other nutrient that should be
increased during pregnancy is folic
acid. Folic acid is a B vitamin that
is needed for healthy blood.

Pregnant women need twice as much
folic acid as non-pregnant women.

 

 

 

 

The other nutrient that you should
increase during pregnancy is folic
acid. Pregnant women need twice as
much folic acid as non-pregnant
women. [To get more folic acid you
need to eat more stuff like orange
juice, oranges, spinach and
grapefruit.]

 

 

Table 11

129

(continued)

 

36M

X

Vitamin and mineral supplements
prescribed by your doctor during
pregnancy are added "insurance."

They don’t take the place of eating a
variety of foods; they only add to
it.

   

 

 

  
 

Vitamin and mineral supplements [help
you make up for whatever you don’t
get otherwise.] They don’t take the
place of eating a variety of foods;
they only add to it. [Some women who
are pregnant don’t take vitamin and
mineral supplements because they say
they are too big or they can’t get
them. But its a good idea to get
vitamin and mineral supplements from
your doctor in case you’re not
getting the vitamin and minerals you
need.]

 

I 38M

There are some things that can harm
your growing baby. While you are
pregnant, it is best to avoid:
* Smoking
* Drinking alcoholic beverages
* Drinking or eating foods with
caffeine, like coffee, tea,
cola, and chocolate, and
* taking non-prescription
medicines

 

 

Many women [are surprised just how
many things they have to watch out
for when they’re pregnant,

* like cough medicine], or other
medicines, [that you’ve taken
all your life.]

* [Smoking harms the baby. When a
woman smokes, the baby smokes
with her. Ask people not to
smoke in your house.]

* Also, [drinking and taking drugs
may harm the baby.]

* Even drinking or eating foods
with caffeine, like coffee, tea,
cola, and chocolate

   
  
  

 

40M

 

 

 

 
 

Now you know how to eat for two. You
know how to choose foods for a
healthier baby and a healthier you!

 

130
Table 11 (continued)

 

Y [Be smart! You’re eating for two
now, so eat healthy.]

 

131

2. The Readability gf the Client-Adapted Langpage Flipchart

Four formulas (Fog, Fry, SMOG and Flesch) were used to
assess the readability of the Client-Adapted Language
flipchart. The results are compared to the readability of
EFNEP’s "Eating Right...for Two" flipchart (Table 12). The
Client-Adapted Language flipchart is easier to read
according to all four formulas. An individual reading at a
sixth grade level could comprehend 50% of the material in
the Client-Adapted Language flipchart according to the Fry
formula, and an eighth grade level reader could comprehend
90% of the_material according to the SMOG readability
formula. An eighth grade reading level was the median
reading grade level of EFNEP clients according to results of
the SORT, therefore, the Client—Adapted Language materials

should be easy-to read for most women in EFNEP.

Table 12: The Reading Grade Level Equivalents of the
Client-Adapted Language Flipcharts

 

 

 

 

 

 

 

 

 

Material Flesch Fog SMOG F
Client-Adapted Language 7 8 8
Flipchart
"Eating Right...for Two" 8 10 10 7
Flipchart

 

132

3. The Readability Scorecard Used with phe Flipcharts and

131.1%

Educators cannot assume that a client can read text
just because her reading levels match the readability of the
written material (Pichert and Elam, 1985). Many other
factors affect comprehension, including the factors assessed
by the "Readability Scorecard (Appendix H)." The
"Readability Scorecard" was used with both flipcharts
(Client-Adapted Language and "Eating Right...for Two") and
the four handouts (Iron, Folic Acid, Protein/Calcium and
Discomforts during Pregnancy) to estimate the number of
characteristics present that might affect the readability of
written text, but are not measured by reading grade
formulas. The results of this assessment appear in Table
13. .All assessments were double-checked by other

researchers, Paula Montgomery and Ann Shanks.

133

Table 13: Using the Readability Scorecard: Assessment of
Characteristics of Readability of Written
Materials

Characteristic Eating Client- Iron
Right Adapted
for Lang.
Two

 

Type size 2 2 2
W .2 12 pt=2
< 12 pt=0

 

 

Font Type 2 2 2 2 2 2
Roman serif=2
All others=0

 

Abbreviations 2 2 2 2 2 2
0%=2
5 2%=1
>2%=0

 

Use of Voice 0 1 2 2 1 2
0% passive=2
510% passive=1
>10% passive=0

 

Emphasis on 2 1 2 2 2 1
action

0% negative=2
52% neg.=1

 

 

Letters 2 2 2 2 0 2
Lower case=2
Capital=0

 

White Space 2 2 2 2 2 2
258%=2
40-57%=1
<40%=0

_:=

 

Leading 2 2 2 0 2 0
32 pt.=2
<2 pt.=0

 

 

Symbols 2 2 1 2 2 2
0% = 2
5 1%=l

 

Headers 0 0 2 2 2 2
2 56%=2
24-55%=1
<24%=O

 

 

 

 

 

 

 

Mu)

 

134

Inteppreting the Score:

 

If the text scored:

17-20 The text contains many characteristics that aid
readability.
13-16 The text contains some characteristics of

readability, but might need to be reviewed to
further increase readability.

<12 The text contains few characteristics of
readability. It should be reviewed and improved
to increase readability.

All handouts scored in the 17 to 20 range, but both
flipcharts scored lower (16 points). The Readability
Scorecard is intended to be a guide to help detect possible
problems, so both flipcharts were analyzed to identify which
characteristics of readability might need improvement and
which are better left unchanged. For example; the EFNEP
flipchart used passive voice 5 times when indicating that
specific vitamins and minerals are needed by the body.

These statements could be reworded in the active voice to
increase readability. The Client-Adapted Language flipchart
contained one sentence emphasizing negative rather than
positive action. This sentence was included in the
flipchart in an effort to use as many direct quotes from
participants as possible. However, this sentence could be
restated more positively to increase readability. Both
flipcharts scored no points for "use of headers," but the
flipchart contains two sentences per page, on average, and

is used along with discussion rather than simply read. Use

135
of headers could interrupt the flow of the conversation.
For these reasons, headers seem inappropriate in the

flipchart.

4. Pilot Test Results: Wording of Surveys On July 28,

1993 five women who were currently enrolled in EFNEP in
Ingham County met in Lansing, Michigan to pilot test the
Knowledge and Material Acceptability Surveys. A second
pilot test was done in September of 1993 with seven EFNEP—
eligible women from Kalamazoo County. The women ranged in
age from 22 to 33 years and were either pregnant or had
children. All of the women had at least an eighth grade
education. The procedures were followed according to the
outline in the Materials and Methods section.

During the discussions, some of the women stated that
they approved of the use of the word "instructor" as opposed
to "teacher", although they usually called the instructor by
name. They preferred the term "homemaker" to "client" or
"EFNEP participant" when referring to themselves. The terms
"flipchart" and "handouts" were easily understood because
they were commonly used by EFNEP instructors and homemakers.
The women made the following specific comments which led to
changes in the final instruments (Appendices F, I and K =
pilot instruments, Appendices G, J and L = final

instruments).

136
a) Knowledge Survey:
Page 2, Question 4: What is your race?
African American/Black
Indian (American)/Alaskan Native
Asian/Pacific Islander
White

other what?

 

I do not wish to answer

 

Are you of Hispanic origin? yes no

The women stated that this question was too confusing
as written. They preferred "Hispanic" to be listed as a
race just like on the EFNEP surveys. Although Hispanic is
not technically a race, it is thought of as a race by women
in the program. Women of Hispanic origin will look for the
word Hispanic and might not check the "white" or the "other"

categories.

Page 3, Question 2: A daily meal plan for a pregnant

 

woman should include 3 to 5 servings of

The women stated that this question should read "A
daily meal plan including snacks" because several women did
not think to include snacks when totaling their daily

servings.

137
Page 4, Question 4: Weight gain in pregnancy is
important because:
a) the baby is growing
b) the mother’s breasts are getting larger
c) the mother’s uterus is getting larger
d) a and b only

e) all of the above

The term "getting larger" was preferred to the term
"growing" when referring to the breasts and uterus because
the first term was perceived as simpler and straight to the
point. There were no problems with the answer choices "a

and b only" and "all of the above."

Page 4, Question 5: Which nutrients are not needed in

extra amounts during pregnancy?

This question was confusing and had to be explained.
The words "not needed in extra amounts" were interpreted as
simply "not needed." They preferred: The need for calcium,
protein and folic acid:
a) increasesduring pregnancy
b) decreases during pregnancy

c) does not change from non-pregnancy to pregnancy

138

Page 4, Question 6: The need for which nutrient(s)

increases during pregnancy:

The "s" in parentheses was interpreted correctly as

meaning that the answer might be plural or singular.

Page 5, Question 8: Betty is pregnant. Yesterday she

ate the following:

fruits: 2 servings

milk and cheese: 3 servings

meat, poultry, fish, eggs and beans : 4 servings

breads and cereals: 4 servings

vegetables: 3 servings

Betty needs to eat more

a)
b)
c)
d)

e)

 

fruits

vegetables

meat, poultry, fish, eggs and beans
breads and cereals

milk

The women stated that there were no problems with the

length or other aspects of this question.

b) Material Acceptabilitv Survey:

The instructions on completing the survey by circling

one answer choice for each question were interpreted

correctly by the pilot test participants. The women

139
understood when to choose one answer and when more than one
answer could be selected from the possible choices. They
liked the scale of "strongly agree", "agree", "disagree" and
"strongly disagree." They stated that they liked the choice
of "I have no opinion" being available and the five answer
choices were about the right number of choices to select
from. The following comments were made about speCific

questions:

Page 1, Question 2: The way the flipchart is written

is insulting to me.

The word "insulting" was thought to be a little strong
for a flipchart, but better than "childish" or "talked down
to." As long as the participants in the study would be
allowed to disagree with the statement, the women thought
that the term "insulting" would be better than any

alternatives for this question.

Page 1, Question 3: The flipchart uses words similar

to the words I use.

The women stated that they liked this question. The
phrase "words that are similar to the words I use" was
perceived as simple and better than "conversational" or

"friendly."

140
Page 2, Questions 4 and 9: 4) The flipchart helped me
learn about nutrition information and 9) The flipchart

answered questions I had about nutrition information.

The women commented that they liked having two
questions on nutrition information instead of one because,
they said, they often learn about nutrition information

before they even have enough knowledge to form a question.

Page 2, Question 5: The flipchart is easy-to—read.

Unlike the women in EFNEP in Kalamazoo County, these
women in Ingham County stated that they never read the
flipchart. Therefore, this question seemed to be
irrelevant. (However, this question was left in the survey
to allow for the differences between counties. There was an

answer choice, "I did not read the flipchart").

Page 2, Question 8: I would be able to understand the

flipchart without an instructor helping me.

This question was eliminated prior to pilot-testing
because the EFNEP homemaker always has an instructor at the

lesson to explain the flipchart.

141
Page 3, Question 10: The number of pages in the flipchart
made it:
The phrase "number of pages" was preferred to the term

"length."

Page 3, Question 11: The appearance of the flipchart
(the pages with the writing) are:

The women repeated that they don’t see the pages with
the writing, so this question is irrelevant like question 5.
(Again, this question will be left in the survey to allow
for differences between counties. The women might answer

that they did not see the pages of text).

Page 3, Question 12: The size of the print used in the
flipchart is:
The term "print" was preferred to the term "type" or

"letters."

Page 3, Question 13: The pictures in the flipchart
are:

One woman suggested adding the choice "held my
attention", but otherwise the women agreed that there were
enough choices. The women added that too many choices would

be confusing.

142
c) Materials Comparison Survey: The women agreed that the
directions were clear and the answer choices were easy-to-
understand. The wording on this survey is similar to the
wording on the Material Acceptability Survey, so comments on
the Material Acceptability Survey apply to the Materials

Comparison Survey, also.

5. Pilot Test Results: Item Analyses

An item analysis of the eight items on the Knowledge
Survey was conducted using the responses of the twelve
participants in the pilot test. The mean knowledge score

was 2.75 (34.4%).

a) Discrimination and Difficulty Indices for the Knowledge
Survey

During pilot-testing, five EFNEP homemakers and the
seven EFNEP-eligible women completed the knowledge survey,
although none of them had had the lesson on nutrition during
pregnancy, "Eating Right...for Two.". Table 14 contains the
results of pilot-testing: the number of correct responses
and the item difficulty and the discrimination of each
question.

The difficulty of the items ranged from 17% to 50%,
with a mean of 35%. Item #2 was difficult (17% answered
correctly) for most women because it required knowledge of

specific numbers of daily servings recommended for the food

143
groups. Although this item was difficult, it was not
removed or revised because it met the objectives of EFNEP

and of this research. No items scored above 80% (easy).

Table 14: Number of Correct Responses, Difficulty and
Discrimination of Knowledge Survey Items

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Qeestion No. Correct Difficulty; Discrimination
Responses (%)
(Out of 12)
ll 1 4 33 .67
u 2 3 25 .67
3 2 17 .67
fl 4 6 50 .67
(I 5 5 42 .33
II 6 5 42 1.00
n 7 4 33 .33
u 8 3 _ 25 .33

 

 

The discrimination ranged from .33 to 1.00, with a mean
of .42. The ability of the items to discriminate between
people who performed well on the test and people who
performed poorly on the test was acceptable for all survey

items.

144
b) Assessing Reliability of the Survey Questions
The Kuder-Richardson (KR-20) formula was used to assess
the reliability of the Knowledge Survey questions (Fitz-
Gibbon and Morris, 1987). This formula measures the
internal consistency of the survey items. The value of the
score lies between zero and one, and the higher the value,

the stronger the inter-correlation of test items.

KR20 = Number of test items x Population variance of an item
Number of items - 1 Population variance of total

Analysis of the pilot test data, showed that the Knowledge
Survey had a reliability score of .874, indicating a strong
inter-correlation of test items and, therefore, a highly

reliable survey.

6. Readability of Surveys Four formulas were used to
measure the readability of the pilot forms of the Knowledge,
Material Acceptability and Materials Comparison Surveys.
Different formulas were used because the results indicate
different levels of comprehension at the different grade
level scores. The Flesch (75% comprehension) and Fog Scores
were determined using the Grammatik III computer program
(Reference Software, Inc., 1989), whereas, the Fry (50%
comprehension) and SMOG (90% comprehension) scores were

determined by hand by two separate investigators (Fry, 1968;

145
McLaughlin, 1969). The readability scores of the pilot

instruments can be found in Table 15.

Table 15: The Readability of the Pilot Survey Instruments

 

 

 

 

 

 

 

Pilot Instrument Flesch Fog SMOG EEijl
Knowledge Survey 6 6 7 5
Material Acceptability 8 10 9 7

Survey
Materials Comparison 7 7 8 7
Survey __

 

 

 

 

 

 

 

The scores ranged from a Fry score of 5th grade (50%
comprehension) on the pilot form of the Knowledge Survey to
a Fog score of 10th grade on the pilot form of the Material
Acceptability Survey. However, like SMOG scores (90%
comprehension), Fog scores tend to be higher than Fry or
Flesch scores. The Flesch formula (75% comprehension),
which has been validated on adult education materials,
indicated that the pilot form of the Knowledge Survey is
written at the 6th grade level; the pilot form of the
Material Acceptability Survey is written at the 8th grade
level and the pilot form of the Materials Comparison Survey
is written at the 7th grade level. The median reading grade
level of the EFNEP population was determined to be 8.3 by

the SORT. All 3 surveys are written below this level

146
according to the Flesch formula, so the readability was not
changed in developing the final instrument. To control for
different reading levels among clients, all questions and

answer choices were read by the instructor.

C. Data Collection Phase
1. Pre nanc Surve I and II Pretest-Posttest
Response frequencies, mean scores and range of scores
were reported for responses to the Knowledge Survey, pretest
and posttest. The differences in Knowledge Survey pretest
and posttest change (posttest minus pretest) scores were

compared between and within groups 1 and 2.

a. The frequency of responses on the pretest and
posttest for each question and each response choice have

been calculated and presented as percentages in Table 16.

Response frequencies for the correct answer choices
increased for all questions from the pretest to the
posttest, regardless of flipchart used. However, response
frequencies of two incorrect answer choices from two

separate questions increased from pretest to posttest.

147

Table 16: Response Frequencies on the Knowledge Survey

 

 

 

 

 

 

 

 

 

 

 

 

 

    
 

 

 

 

 

 

 

(Pretest/Posttest)
Question Response (%)1
=—_——7 _=_
a b c d e No or >1
answer
1.3/ - 7.9/ *27.1/ 62.9/ ------ 0.8/
1 0.4 4.6 55.0 39.2 0.8
*28.3/ 13.8/ 19.2/ 15.8/ 15.8/ 7.1/
2 43.3 5.4 13.8 17.1 14.2 6.3
16.7/ 24.2/ 11.3/ *47.1/ ------ 0.8/ fl
3 5.8 7.9 6.3 76.7 3.3
35.0/ 3.8/ 2.9/ 5.4/ *52.1/ 0.8/
4 22.5 1.3 1.7 7.1 67.1 0.4
*70.8/ 9.2/ 5.8/ 12.5/ ------ 1.7/
“5 86.3 4.6 2.5 5.8 0.8
22.1/ 5.0/ 7.1/ *58.3/ 6.3/ 1.3/
6 5.0 5.0 1.3 85.0 2.9 0.8
19.6/ *27.9/ 14.6/ 36.3/ ------ 1.7/
7 18.3 40.4 11.7 27.9 1.7 ||
28.8/ 8.3/ 6.7/ *23.3/ 27.5/ 5.4/
8 21.7 8.3 3.3_ 39.2 21.3 6.3
T e correct response is indicated by *. Posttest

 

responses were obtained after only one reading of the
flipchart.

148

For one of those questions, question 2, "A daily meal
plan, including snacks, for a pregnant woman should include
3 to 5 servings of "

a) vegetables

b) fruits

c) breads and cereals

d) milk and cheese

e) meat, poultry, fish, eggs and beans

more women circled choice "d" than the correct answer "a."
When the results of the pretest and posttest are divided by
the flipchart used, for flipchart X (the original EFNEP
flipchart) there was a 6.3% increase of the number of women
who chose the incorrect answer "d" on the posttest compared
to the pretest, whereas, for flipchart Y (the Client-Adapted
Language flipchart), there was a 3.1% decrease in the number
of women who selected choice "d." Clearly, the Client-
Adapted Language flipchart was more effective in teaching
the correct number of recommended servings from the food
groups.

For the other question (question 4), "Weight gain in

pregnancy is important because:"

149

a) the baby is growing

b) the mother’s breasts are getting larger

c) the mother’s uterus is getting larger

d) a and b only

e) all of the above
more women chose the incorrect answer choice "d" than the
correct answer "e." Again there was a difference in which
flipchart was used. Among those women who used the original
EFNEP flipchart, 3.6% more women chose the incorrect answer
"d" on the posttest compared to the pretest. Among the
women who used the Client-Adapted Language flipchart there
was no difference in response frequencies for the selection
of choice "d" between the two tests. Although the results
of the pilot test did not indicate that the answer choice "a
and b only" might be confusing, the results of the actual
study suggest that confusion might have been a problem.

When response frequencies for the correct answer
choices were divided by the flipchart used, participants who
used the Client-Adapted Language flipchart selected the
correct answer choice on the posttest for all eight
questions, more often than participants who used the
original EFNEP materials (Table 17). Increases in
frequencies of correct answer selection ranged from 5.2% for
question four on the importance of weight gain in pregnancy
to 28.2% for question eight where the client is given a

sample daily diet and she needs to select which food group

150

needs to be increased in number of servings.

Table 17: Response Frequencies for Correct Posttest Answers

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Question Group 1: Eating Group 2: Difference
Right...for Two Client- (Group 2
% Adapted minus Groupl)
(n=112) Languaqe(%) %
(n=128)
1 40.2 68.0 +27.8
2 32.1 53.1 +21.0
3 68.8 83.6 +14.8
4 64.3 69.5 +5.2
5 76.8 94.5 +17.7
6 76.8 92.2 +15.4
7 26.8 52.3 +25.5
8 24.1 52.3 +28.2
Mean 51.2 70.7 +19.5

 

b. The mean pretest, posttest and change scores
(posttest minus pretest) for each group, along with the
standard error, and range of scores were calculated and
reported (Table 18). An unpaired t—test was used to
determine if there is a significant difference between the
means of posttest scores and change scores for group 1 vs.

group 2 (Table 18).

151

Table 18: T-Test Results, Mean Scores and Range of Scores

for the Knowledge Survey

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1. indicates Significant difference

scores between groups
2. indicates significant difference

scores between groups

(p<.001) mean posttest

(p<.001) mean change

The mean pretest scores between Group 1 and Group 2

were not significantly different, but the mean posttest

scores and mean change scores were significantly different

(p<.001) between groups who used the two different

flipcharts. Within both groups the difference between

pretest and posttest scores were signifiicantly different

(p<.001). These results indicate that the two groups

fl =
Scores Group 1: Eating Group 2: Client-
Right for Two .Adapted Language
(n=112) (n=128)
Pretest:
mean 3.5 i 0.1 3 4 5 O 1
range 0 to 8 0 to 8
Posttest:
mean 4.1 i 0.21 5 7 i 0 21 ||
range 0 to 8 0 to 8 "
Change:
mean 0.8 i 0.22 2.4 i 0.22
range -4 to 6 -2 to 8

152
started with approximately the same amount of knowledge of
the specific subject area tested. However, the group that
used the Client-Adapted Language flipchart increased their
knowledge of nutrition in pregnancy to a greater extent than
the group that used the original EFNEP flipchart as
indicated by the significant differences in posttest and

change (posttest minus pretest) scores.

c. Demographic information for study participants is
reported in Tables 19-24. Women from five Michigan counties
(Oakland, Ingham, Kalamazoo, Kent and Genesee) participated
in the data collection phase of this study. Participant
numbers per county varied with the number of newly enrolled
clients during the study period, the number of instructors
in each county available for data collection, the motivation
of the county Extension home economist to encourage
participation of instructors and the feasibility to complete
the testing within each county. The total number and

percent of participants from each county appear in Table 19.

153

Table 19: Number of Participants per County

 

 

 

 

 

 

 

============u=================
County1 Eating Client-Adapted Tepel
Right...for Two Langpage
Oakland 32 10 42 (17.5%)
Ingham 16 18 34 (14.2%)
Kalamazoo 24 43 67 (27.9%)
Kent 31 35 66 (27.5%)
Genesee 9 22 31 (12.9%)
Total 112 (46.7%) 128 (53.3%) 240 (100%)

 

 

 

 

 

 

 

1. Instructors were given pacfiets of surveys randomly

assigned to either flipchart group, resulting in unequal
numbers of participants per county assigned to each group.

Mean posttest and mean change scores were determined
for different groups within each demographic variable (age,
pregnancy status, number of babies, race/origin and highest
education level achieved) and by flipchart used. The
results are entered into Tables 20-24. Correlation
coefficients were determined to detect if a relationship
existed between mean posttest scores or mean change scores

and each demographic characteristic.

Age

The original target population for this study was adult
women of childbearing age (18-40 years) who participated in
EFNEP. A few instructors, however, collected data from

fifteen women under 18 years and two women over 40 years of

154
age. There was no significant correlation between age of
participants and posttest or change scores, regardless of
the flipchart used. Therefore, all data were included in the
analyses. Mean posttest scores, mean change scores, and
standard errors of the mean determined for all age groups
are shown in Table 20. The mean age of study participants

was 25.1 years and the median was 24.0 years of age.

155

Scores) by Age of EFNEP Clients

Table 20: Mean Knowledge Survey Scores (Posttest and Change

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age (years) Subjects Group 1: Group 2:
(n=group 1, % Eating Right.. Client-Adapted
group 2) (n=240) for Two Language
Mean Posttest Mean Posttest
Scores(i s.e.) Scores(i s.e.)
<18 (n=14,1) 15 3.2 i 0.2 6.0
(6.3%)
18-21 62 4 1 i 0 3 5 6 i 0 4
(n=33,29) (25.8%)
22-25 59 4 6 i 0 3 6 2 5 0 3
(n=24,35) (24.6%)
26-29 32 3 6 i 0 7 5 9 i 0 4
(n=14,18) (13.3%)
30-33 34 4 6 i 0 6 4 9 i 0 4
(n=13,21) (14.2%)
34-37 16 4.5 i 0.5 6 4 i 0 5
(n=5,11) (6.7%)
38-42 14 4.5 i 1.0 4 9 1 0 5
(n=4,10) (5.8%)
Mean Change Mean Change
Scores(is.e.) Scores(is.e.)
<18 15 -.15 i 0.3 6.0
(n=14,1) (6.3%)
18-21 62 1.2 i 0.4 2 4 i 0 4
(n=33,29) (25.8%)
22-25 59 1 1 i 0 4 2 6 i 0 4
(n=24,35) (24.6%)
26-29 32 0 3 i 0 7 2 7 5 0 5
(n=14,18) (13.3%)
30-33 34 1 l i 0 4 1 3 i 0 4
(n=13,21) (14.2%)
34-37 16 0 4 i 1 2 2 5 i 0 4
(n=5111) (6.7%)
38-42 14 0 8 i 0 5 2 1 i 0 4
(n=4,10) (5.8%)

 

 

156

Pregpancy status
At the time of the study, 27.1% of the participants

reported that they were pregnant and 72.1% that they were
not pregnant (Table 21). Two participants (0.8%) were
unsure of their pregnancy status. There was no significant
correlation between pregnancy status and posttest scores or
change scores, regardless of the flipchart used. Although
Klare (1976) reported that motivation can sometimes override
the effect of readability on comprehension, increased
motivation to learn about nutrition in pregnancy for
pregnant women compared to non—pregnant women did not appear

to be a factor in this study.

Table 21:

of EFNEP Clients

   

157

Mean Knowledge Survey Scores by Pregnancy Status

 

 

 

 

  

 

 

 

 

 

 

 

 

 

Pregnancy Subjects Group 1: Group 2:
Status1 % Eating Right.. Client—Adapted
(n=group 1, (n=240) for Two Langpage
group 2)
Mean Posttest Mean Posttest
Scores(is.e.) Scores (5s.e.)
Pregnant 65 4.1 5 0.3 5.9 5 0.3
(n=38,27) (27.3%)
Not Pregnant 173 4 1 5 0 3 5 8 5 0 2
(n=74,99) (72.1%)
Mean Change Mean Change
Scores (5s.e.) Scores (5s.e.)
Pregnant 65 0.8 5 0.3 2.4 5 0.4 ;
(n=38,27) (27.3%)
Not Pregnant 173 0.8 5 0.2 2 4 5,0 2 ;
(n=74,99) (72.1%) i

1. Pregnancy status 1no1cates status at time 0
all women had previously been pregnant.

intervention,

Number of babies

Most women in this study had given birth previously;

only 12.1% of participants were about to be first-time

mothers.

given birth once,

Approximately one-third (30.4%) of the women had

35.4% twice and 20.4% three or more times.

Two individuals (0.8%) refused to report this information.

See Table 22 for data on number of babies by flipchart used.

There was no significant correlation between number of

babies born to a mother and posttest or change score

achieved.

Like pregnancy status,

increased motivation to

158
learn about nutrition in pregnancy by women who had less
experience at giving birth did not appear to be a factor in
this study. Research by Recht and Leslie (1988) showed that
previous knowledge can greatly increase the amount of
information recalled from a passage and can even compensate
for poor reading ability. However, in this study previous
knowledge of pregnancy among women experienced at giving

birth did not appear to effect nutrition knowledge.

Table 22: Mean Knowledge Survey Scores by Number of Babies
Born to the EFNEP Client

 

 

 

 

 

 

 

 

 

 

 

 

 

 

No. of Babies Subjects Group 1: Group 2:
(n=group 1, (%) Eating Right.. Client-Adapted
group 2) (n=240) for Two Langpage
Mean Posttest Mean Posttest
Scores (is.e.) Scores (5s.e.)
none (n=17,12) 29 3.8 5 0.4 5.7 5 0.3
(12.1%)
one (n=35,38) 73 4.3 5 0.3 5.8 5 0.3
(30.4%)
two (n=32,53) 85 3.9 5 0.3 5.8 5 0.2
(35.4%)
three or more 49 4.0 5 0.4 5.5 5 0.4
(n=26,23) (20.4%)
Mean Change Mean Change
Scores (is.e.) Scores (58.e.)
[none (n=17,12) 29 0.2 5 0.3 ' i 3.0 5 0.5
(12.1%)
one (n=35,38) 73 1.1 5 0.3 2.3 5 0.4
(30.4%)
two (n=32,53) 85 1.0 5 0.4 2.3 5 0.2
(35.4%)
three or more 49 0.4 5 0.4 2.5 5 0.5
(n=26,23) j20.4%)

 

 

 

159

 

Overall the majority of study participants classified
themselves as either African American (26.7%), White, not
Hispanic (68.3%), or Hispanic (7.9%). When responding to
the question of race and ethnicity, three individuals (1.3%)
classified themselves as Native American and two (0.8%) as
Asian/Pacific Islander. All participants responded to this
question (Table 23). This representation of different
races/ethnicities is very close to the actual EFNEP
representation. In 1991, Michigan EFNEP served 5430
clients: 51% white, 38% African American, 8% Hispanic, 2%
Asian/Pacific and 1% Native American or Alaskan. In this
study, there was no significant correlation between race and
either posttest or change scores, regardless of flipchart
used. This result implies that the flipchart material is
not culturally or racially biased toward or against
Caucasians, Hispanics or African Americans. However, the
groups of Native Americans (n=3), and Asians/Pacific
Islanders (n=2) were too small to make the same
generalization. Change scores were higher for Hispanics
using the Client-Adapted Language flipchart, partially due
to the fact that two individuals achieved perfect scores on
the posttest after poor pretest showings.

Harman and Edelsky (1989) report that using LEA is

160
advantageous to learning because it welcomes all cultures
and languages into the classroom. However, in this study no
one particular race or ethnic group increased their
nutrition knowledge to a significantly greater extent than
other groups using the Client—Adapted Language flipchart.
The fact that only native English speakers were accepted as
participants in this study probably had a great impact on

these results.

161

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 23: Mean Knowledge Survey Scores by Race or Origin of
EFNEP Clients
RaceZOrigin Squecps Group 1: Group 2:
(n=group 1, (%) Eating Right.. Clienp-Adapped
group 2) (n=240) fer Two Language
Mean Posttest Mean Posttest
Scores (5s.e.) Scores (5s.e.)
African 52 3.2 5 0.4 5.4 5 0.4
American (21.7%)
(n=23, 29)
Native American 3 4.0 3 05 1 0
(n=1, 2) (1.3%)
Asian/Pacific 2 7.0 3.0
Islander (0.8%)
(n=1, 1)
White (not 164 4 3 5 0 2 S 9 5 0 2
Hispanic, (68.3%)
n=83, 81)
Hispanic 19 3 3 5 0 5 5 9 5 0 4
Mean Change Mean Change
Scores (5s.e.) Scores (5s.e)
African 52 0.2 5 0.4 2.3 5 0.4
American (21.7%)
(n=23, 29)
Native American 3 -3.0 2 0 5 2 0
(n=1, 2) (1.3%)
.Asian/Pacific 2 2.0 2.0 fl
Islander (0.8%)
(n=1,1)
White (not 164 1 0 5 0 2 2 3 5 0 2
Hispanic, (68.3%)
n=83,81)
Hispanic 19 00509 30507
(n=4,15) (7.9%)

 

 

 

 

 

 

162

Highest level of education achieved

All women who participated in this study had completed
at least the seventh grade level of formal schooling.
Sixty-eight percent had graduated from high school, with
eight of those women (3.3% of participants) graduating from
college, also. See Table 24 for data on education by
flipchart used. The mean highest level of education
achieved was 11.8 grade and the median was nearly the same
at 12 grades completed. Therefore, half of the study
participants had graduated from high school and half had
not. Two individuals (0.8%) refused to answer the question
on education level. There was no significant correlation
between highest level of education achieved and posttest or
change scores. This lack of significance is probably due to
the relatively high level of education among women in this
study. Only 1.8% of participants had less than a tenth

grade education.

Table 24:

163

Education of EFNEP Clients

Mean Knowledge Survey Scores by Level of

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(n=3,5)

 

 

 

 

Higheep Level Subjects Group 1: Group 2:
of Education (%) Eating Right.. Client—Adapted
(n=group 1, (n=240) for Two Language
group 2)
Mean Posttest Mean Posttest
Scores (5s.e.) Scores (5s.e.)
Grades 7-9 10 3.5 5 0.5 5.0 5 0.7
(n=2,8) (4.2%)
Grades 10-11 64 3 1 5 0 3 5.7 5 0.4
(n=35,29) (26.7%)
High School 103 4 4 5 0 3 6.0 5
Graduate/GED (42.9%)
(n=51,52)
Some College/ 53 5 1 5 0 3 6.0 +
Professional (22.1%)
School
(n=20,33)
College 8 5 7 5 1 3 4.4 +
Graduate (3.3%)
(n=3,5)
Mean Change Mean Change
Scores (5s.e.) Scores (5s. e.
Grades 7-9 10 0.0 5 1.0 1.5 5
(n=2,8) (4.2%)
Grades 10-11 64 0 3 5 0 4 3.1 +
(n=35,29) (26.7%)
High School 103 1 3 5 O 3 2.5 5 0.3
Graduate/GED (42.9%)
(n=51,52)
Some College/ 53 0 4 5 0 3 1 7 5 0 1
Professional (22.1%)
School
(n=20,33)
College 8 1 0 5 1 0 2 0 5 0 9
Graduate (3.3%)

 

164
2) Material Acceptability Survey

Responses to the Materials Acceptability Survey
questions (Appendix J) have been tallied and frequencies of
responses have been entered into Table 25 (questions 1-8),
Table 27 (questions 9-11), Table 28 (question 12) and Table
31 (questions 15-19). Significant differences cannot be
reported because the data are frequency data.

When frequency scores were combined into one category
for "agree" plus "strongly agree" and one category for
"disagree" plus "strongly disagree," group 2 scores were
subtracted from group 1 scores to show differences in

agreement (Table 26, questions 1-8).

165

Table 25: Frequency of Responses (Material Acceptability
Questions 1-8)

 

Survey,

Response Choices

 

 

 

 

 

 

 

 

Questionz Strongly Agree Dis- Strongly
Gpoup1 Agree (%) agree Dis-
Group 1 (%) (%) agree
(n=67) (%)
Group 2
(n=61)
Easy to
Understand
Group 1 38.8 46.3 1.5 3.0
Group 2 41.0 55.7 3.3 0.0
Insulting
Group 1 3.0 11.9 46.3 22.4
Group 2 3.3 11.5 42.6 34.4
Uses
Similar
Words
Group 1 14.9 59.7 16.4 3.0
Group 2 24.6 59.0 11.5 1.6
Informative
Group 1 19.4 65.7 10.4 0.0
Group 2 36.1 60.7 3.3 0.0
Easy to
Read
Group 1 20.9 29.9 1.5 3.0 44.72
Group 2 19.7 23.0 3.3 0.0 54.1
Interesting
Group 1 11.9 50.7 28.4 1.5 7.5
Group 2 26.2 45.9 11.5 3.3 13.1
Useful
Group 1 19.4 59.7 7.5 0.0 13.4
Group 2 32.8 59.0 3.3 1.6 3.5

 

 

 

 

 

 

Table 25 (continued)

166

 

 

2.

 

 

 

 

 

 

 

 

Answered my
Questions
Group 1 11.9 44.8 10.4 0.0 32.9
Group 2 24.6 54.1 8.2 0.0 13.1
used the EFNEP flipchart

1. Group 1=C11ents who

Group 2=Clients who used the Client-Adapted Language

Many participants
preferred to have

declined to read the flipchart, but
the instructor read it to them.

167

Table 26: Differences in Frequency of Responses
(Group 2 minus Group 1)

Combined Response Choices

 

 

 

 

 

 

 

 

 

 

Question Strongly Strongly Disagree
Agree + Agree + Disagree

Easy-to-Understand +11.6 —1.2
Insulting -0.1 +8.3

Uses Similar Words +9.0 —6.3
Informative +11.7 -7.1
Easy-to-Read -8.1 -1.2
Interesting +9.5 -15.1
Useful +12.7 -2.6
Answered My +22.0 —2.2

Questions

 

 

 

 

 

Therefore, a negative score in this table would
indicate a higher percentage of women who used the "Eating
Right...for Two" flipchart agreed or disagreed with the
statement and a positive score would indicate a higher
percentage of women who used the Client—Adapted Language
flipchart agreed or disagreed with the statement.

Results of the Material Acceptability Survey showed
that 11.6% more clients agreed that the Client—Adapted
Language flipchart was easy to understand compared to the
percentage of clients who thought that the original EFNEP
flipchart was easy to understand. This "ease of
understanding" might refer to how easy the flipchart is to
understand when the words are read to the client because

this result is opposite to the result of the question on how

168
easy the flipchart is to read. It is possible that the
instructors’ familiarity with the "Eating Right...for Two"
flipchart might have played a role in how easy a flipchart
was to understand. Although the instructors were asked to
read the flipcharts, they might have paraphrased some of the
text because they had previously taught the same lesson many
times.

When responses to the statement, "The flipchart is easy
to understand" were considered by the education level of
participants, most clients from every educational level
agreed or strongly agreed with the statement. As would be
expected, more clients (4.8%) with a high school education
or less disagreed with the statement than clients with some
college or professional school (2.9%). Responses grouped by
age showed more agreement among individuals over 25 years of
age (97.9% agreed or strongly agreed) that the flipchart was
easy to understand compared to participants 25 years of age
or less (84.6% agreed or strongly agreed). When responses
were further divided by education level or age egg by
flipchart used, there was little difference in client
agreement of the understandability of the flipchart between
"Eating Right...for Two" and the Client-Adapted Language
flipchart at each education level or age group.

More clients (8.1%) agreed that the original flipchart
was easy-to-read compared to the percentage of clients that

thought the Client-Adapted Language flipchart was easy to

169
read. Although 50.8% of group 1 and 42.7% of group two
thought that the flipcharts were easy to read, 44.7% of
group one and 54.1% of group two never read the flipchart.
The reasons for not reading the flipchart were not stated by
the participants, however, instructors in some counties do
not routinely let clients read the flipchart and might have
discouraged clients from doing so during this project. The
clients who chose to read the flipchart might have
volunteered because they are better readers, therefore, the
percentage of women who reported the flipchart was easy to
read might be inflated compared to the opinion of Michigan
EFNEP women in general.

When responses were divided by education level, similar
numbers of individuals who had completed 5 12 years of
school and individuals who had completed >12 years of school
agreed with the statement, "The flipchart is easy to read."
This result is not surprising because according to the
results of the SORT in the preliminary phase of this
research, education level achieved and reading levels are
not correlated.

When responses were further divided by flipchart used,
more women with less than a high school education (11.6%)
agreed that "Eating Right...for Two" was easy to read
compared with 3.9% of women who thought that the Client-
adapted Language flipchart was easy to read. These numbers

were reversed for women with a post-high school education.

170
Less highly educated women (5.5%) agreed that "Eating
Right...for Two" was easy to read compared to the number
that agreed that the Client-Adapted Language flipchart was
easy to read (7.8%).

Whether or not a flipchart is easy-to-read might seem
irrelevant because the flipchart is often read to the
client. However, the process of using client language could
be repeated to make handouts or other materials for clients
to read. In this case, developers should be aware that
using client language does not necessarily produce text that
is easier for the client to read. Both readability formulas
and the Readability Scorecard might be used to assess the
readability of new written text.

Nitzke (1987) developed a nutrition pamphlet using a
process similar to the one used to develop the Client-
Adapted Language flipchart. She found that the newly
developed pamphlet was actually mepe difficult to read than
the original pamphlet, but the readers found the new
pamphlet understandable and more personally relevant than
the original pamphlet.

Less than 16% of the clients in this study thought that
either flipchart was insulting; however, more clients (8.3%
more) disagreed that the Client-Adapted Language flipchart
was written in insulting language, despite the fact that
several clients reported not liking the words "thing" and

"stuff" (Appendix Q). Shepard et al. (1994) reported that

171
consumers prefer informal writing. When 28 adults (25% with
college degrees) evaluated nutrition education materials,
the least formal writing style was rated most preferred
compared to more formal or scientific styles. The clients
who commented that the words "thing" or "stuff" were
insulting in the Client-Adapted Language flipchart were all
young (5 26 years of age) and had less than a high school
education. The older, more educated women tended to comment
on the flipchart as a whole and not specific words.

More clients (9% more) agreed that the Client-Adapted
Language flipchart used words similar to the words they
used. However, both flipcharts scored high in this area
with >74% of clients agreeing that either flipchart used
words similar to their own, despite the fact that only the
Client-Adapted Language flipchart used exact quotes from an
EFNEP-eligible population. The original EFNEP flipchart
language was pilot-tested with clients; therefore, scores
would be expected to be high in this area. When responses
to this question were divided into categories by age,
education level or race there was little difference between
groups.

When responses to this statement were further divided
by age and by flipchart used, there was little difference
between responses of younger compared to older women.
However, more African American women agreed that the Client-

Adapted Language flipchart used words similar to their own

172
(41.3%) compared to "Eating Right...for Two" (34.8%).
Different results were seen among white women who agreed
that "Eating Right...for Two" contained similar language to
their own (47.0%) compared to the Client-Adapted Language
flipchart (40.7%). The Client-Adapted Language flipchart
was written using quotes from both African American and
white women. Although the data shows differences in
opinions among racial groups, many women (>30%) from each
race chose either flipchart. It would be difficult to
justify producing race or ethnic-specific materials for
native english speakers based on this data.

Furthermore, the Client-Adapted Language flipchart was
viewed by a higher percentage of clients as being
informative (11.7% more agreement), interesting (9.5% more
agreement), useful (12.7% more agreement) and answered more
client questions on nutrition (22% more agreement). Over
15% more clients disagreed that EFNEP’s "Eating Right...for
Two" was interesting compared to the percentage of women who
disagreed that the Client-Adapted Language flipchart was
interesting. When responses to the question, "The flipchart
helped me learn about nutrition information" were divided by
pregnancy status, non-pregnant women found the flipchart
they used more informative (95.4% vs. 85.3%), more
interesting (76.3% vs. 69.7%) and more useful to pregnant
women (95.1% vs. 85.3%). Pregnant women might be more

discriminating than non-pregnant women in choosing or

173
evaluating educational materials.

When responses were divided by education level and
flipchart used, a greater number of the less educated women
(5 12th grade) agreed that "Eating Right...for Two" was
informative (37.5%) and useful (36.6%) compared to women who
used the Client-Adapted Language flipchart (28.9% agreed
informative, 28.9% agreed useful). The opposite result was
found among women with a post-high school education where
more women agreed that the Client-Adapted Language flipchart
was informative (52.7%) and useful (52.1%) compared to
"Eating Right...for Two" (46.4% and 50.6%, respectively).

Both flipcharts had 20 pages of written text and 18
point type. Of the clients who responded to questions 9, 10
and 11 on the Material Acceptability Survey (Table 26), most
respondents thought both flipcharts had the correct number
of pages (68.2%, Group one and 73.8%, Group two), was "O.K."
in appearance (65.1%, Group one and 66.7%, Group two), and
had about the right size print (69.5%, Group one and 80.3%,

Group two).

Table 27:

174

Frequency of Responses (Material Acceptability

Survey, Questions 9-11—-Length, Appearance and
Print Size)

Response Choices

 

 

 

 

 

 

 

 

 

 

 

 

lfMGroup 1=Clients
Group 2=Clients
flipchart

0.0
who used the EFNEP fl‘b

 

 

QuestionZGroup1 too leng too ehort about p_
(%) (%) right opinion
(%) (%)
9) No. of Pages
Group 1 (n=66) 18.2 4.5
Group 2 (n=61) 14.8 1.6
too too much
crowded space
10) Appearance
Group 1 (n=63) 0.0 6.3
Group 2 (n=57) 1.8 3.5
too too 1
large
11) Print Size
Group 1 (n=59) 6.8 0.0
Group 2 (n=61) 4.9

who used the Client-Adapted Language

Over 14% of the clients responded that 20 pages were

too many, whereas,

was too few.

less than 5% thought that number of pages

increasing the readability of the flipchart.

thought this size was about right (>69%),

The type size (18 point) is a large size type,
Most clients

less than 7%

thought this size might be too large, and no one thought

that the print was too small.

Not surprisingly,

all

responses that the print size was too large were from women

less than 30 years of age.

Researchers have reported that

175
12 point or larger types are easier to read (Doak et al.,
1985) and consumer groups evaluating nutrition education
materials preferred large (12 to 16 point) types, also
(Shepard et al., 1994).

The appearance of the flipchart did vary slightly
between the two versions, with the Client-adapted Language
flipchart having more words (708 words compared to 617 words
in "Eating Right...for Two." However, only 1.8% of the
women who used the Client-Adapted Language flipchart thought
that it looked too crowded, whereas, 6.3% of the women who
used the "Eating Right...for Two" version thought that the
latter flipchart had too much blank space. The majority of
women (>65%) thought the appearance of both flipcharts was
O.K. Similar results were reported by Shepard et al. (1994)
who found consumers prefer "a good deal of white space"
because it helps you organize the material and too much text
is intimidating. These researchers report that the ideal

amount of white space for maximum readability is unknown.

Pictures

The next question addressed the pictures which were the
same for both flipcharts. When asked to select an answer
from several choices to the closed-ended question, "The
pictures in the flipchart are ," most of the
responses were positive (Table 28). Many of the women

thought the pictures helped them understand the nutrition

176
information (52.3% of respondents), were interesting (32.8%
of respondents) and held their attention (25.0% of
respondents). Fewer women thought the pictures could have
been left out (11.7% of respondents) and even fewer viewed
the pictures as insulting (5.5% of respondents). Although
few women selected the negative responses to this close-
ended question, some negative responses about the pictures
that were written as comments to other questions (e.g. ugly,
outdated, etc.) were not among the answer choices for this
question. More women might have viewed the pictures
negatively, but could not find the appropriate answer choice
to express their feelings.

There were some interesting results when responses to
the questions about the pictures were divided by education '
level. A greater percentage of women with less than a tenth
grade education thought the pictures were interesting
(44.4%) and understandable (66.7%) compared to women with a
high school education (28.6% and 41.7%) or post high school
education (25.7% and 51.4%). Also, more of the less
educated women found the pictures insulting (11.1%) compared
to women with a post high school education (0%). In this
case, the less educated women seemed to be more opinionated,
regardless of whether their opinions were positive or
negative. According to Shepard et al. (1994), consumers
prefer realistic drawings, paintings or color photographs in

nutrition education materials. These researchers report

177
that common foods should be used in pictures compared to
exotic or expensive foods, so consumers understand that good

nutrition is affordable.

Table 28: Frequency of Responses (Material Acceptability
Survey, Question 12——Pictures)

 

 

 

 

 

 

 

 

 

 

 

 

Characteristics of Pictures Fregeency of Responses
(Question 12) (EFNEP Materials)1 (%)
Interesting (n=42) 32.8

I Helped Me Understand Information 52.3

(n=67)

Insulting (n=7) 5.5 I
Could Have Been Left Out (n=15) 11.7 J
Held My Attention (n=32) 25.0 n

I 1. Pictures are the same on both f 1pcharts.

These results seem to contradict the result of the
written comments about the pictures (Table 29). In the
written comments, women responded with nearly twice as many
negative comments as positive comments. Some of the
negative responses referred to the pictures as outdated,
insulting, unrealistic, old "crud", ugly, boring and in need
of improvement (60% of responses, 20% of women). It appears
that either question 12 did not address the problem or that
the pictures were not perceived as negatively as the written
comments would lead the reader to believe. The women who

did not like the pictures might have been more motivated to

178
write out their comments than the women who were satisfied

with the present pictures.

Written Responses: Material Acceptability Survey

The principal investigator reviewed the written
responses to questions 2, 10, 12 and 13 in the Material
Acceptability Survey and has summarized them in narrative
form. Although over 14% of clients who used either
flipchart agreed that the flipchart was insulting, very few
clients (<8%) responded with written comments to the
question, "Which part of the flipchart was insulting?" Some
comments were very general, simply stating that the writing
(one respondent who used "Eating Right...for Two") or
pictures (two respondents who used either flipchart) were
insulting. Of the clients who used the Client-adapted
Language flipchart, three women (4.9%) commented that the
words "thing" or "stuff" sounded unprofessional when used in
a flipchart.

Sixteen of 128 women (12.5%) responded to question 10
"The appearance of the flipchart could be improved by

." Most responses (87.5%) addressed

 

concerns over the quality of the pictures and not the
writing. Fourteen women stated that the pictures needed to
be improved, updated or made more realistic. One woman

noted that a prescription bottle in one picture was dated

179
1985. Other comments included that the flipchart was O.K.
as is (one response) or that the flipchart needed more color

(one response).

Table 29: Written Responses About Flipchart Pictures1

-

 

 

 

 

 

 

 

 

 

 

 

Comment Number of Percent of
Responses Responses
(n=43)
Outdated/Old crud 3%
Simple 3%
Helpful/Useful 3%
Easy to look at 8%
O.K. .3%
Ugly 6%
Childish/Insulting 5%
Necessary
Boring 8%

 

Not realistic

Colorful

 

 

Educational/Informative

 

Self-explanatory

 

Didn’t catch attention

 

HHI—‘HNWI—‘l—‘Qle-‘wwm
OOOOI—‘NOOU‘II—‘mowwm
(D
c\°

Upside down ‘ _
1. The pictures are the same for both flipcharts.

 

 

 

 

 

 

180
Thirty three percent of 128 women responded to question

12, "The flipchart pictures are

 

." There were approximately twice as many negative as

 

positive comments. Comments were grouped by subject and
appear in Table 29. The most common negative responses to
the pictures were that the pictures were outdated (18.6% of
responses) or childish (16.3% of responses). However, 18.6%
of responses did classify the pictures as O.K. as is. One
woman wrote that the pictures were "the best that could be
expected." Another woman wrote that the pictures were
upside down. However, the pictures would have been right
side up if she had used the flipchart as it was intended.

Illustrations and diagrams that successfully portray
the message in the text have repeatedly been shown to have a
positive affect on retention of information (Tajika et al.,
1988). Mayer (1989) found that diagrams within a lesson
helped students answer problem-solving questions and were
most useful to students of low aptitude or with low prior
knowledge. Doak et al. (1985) recommend simple
illustrations to decrease reading time for poor readers,
emphasize important points and provide motivation. Simple
drawings were found to be effective when the instructor
controls viewing time, but photographs were found to be more
effective than drawings when the reader has the control over
her viewing time (Doak et al., 1985). Therefore, the

Michigan EFNEP participants might benefit from new

181
illustrations or diagrams in the "Eating Right...for Two"
flipchart and photographs to accompany handouts left with
the client to aid in understanding the specific objective

being taught.

Table 30: Written Responses About Improving the Flipcharts

 

 

 

 

 

 

 

 

 

 

Comment EFNEP Flipchart Client-Adapted
No. Responses Language
(%) M1212.
No. Responses
(%)
Make more interesting 2 (3.0%) 1 (1.6%)
Like as is 4 (6.0%) 6 (9.8%)
Don’t use flipchart 1 (1.5%) 0 (0.0%)
Make easier 3 (4.5%) 0 (0.0%) i
Make harder/more 2 (3.0%) 2 (3.3%)
detailed
Change wording 1 (1.5%) 0 (0.0%)
Add more pictures 0 (0.0%) 1 (1.6%)
Comments didn’t 4 (6.0%) 1 (1.6%)
answer question
Improve/update 10 (14.9%) 10 (16.4%)
_pictures

 

 

 

 

 

Of the women who wrote comments on the Material
Acceptability Survey, more women (37.5%) responded to
question 13, "How could the flipchart be improved?" than to
any other question. Of the women who used "Eating
Right...for Two," 27 women (40.3%) wrote comments with ten

of those (37% of comments) commenting that the pictures

182
should be updated, made more realistic or otherwise
improved. Likewise, of the women who used the Client—
Adapted Language flipchart, 21 women (34.4%) made written
comments of which ten (47.6% of comments) responded that the
pictures should be updated, made more realistic or otherwise
improved. As expected, the response frequencies were
similar with either flipchart because the pictures were
exactly the same.

The comments from question 13, ''How can the flipchart
be improved?," were grouped into categories by subject and
appear in Table 30. When participants were asked this
question, the number of responses that pertained to the
pictures exceeded the number of responses to question 12,
which asked specifically about the pictures. After reading
question 12, the women might have been thinking about the
pictures when answering question 13 about how to improve the
flipchart. In all (responses from question 12 and 13), there
were 63 comments about the pictures with 46 (73.0%) of those
comments being negative (i.e. old, outdated, boring, not
realistic).

Over 14% of the responses indicated that the women
liked the present flipchart as it is, but twice as many
(28.5%) liked the Client-Adapted Language flipchart. Two
users of either flipchart asked to make the flipchart more
difficult, and three users of the "Eating Right...for Two"

flipchart requested that it be made easier. One woman who

183
used "Eating Right...for Two" stated that no flipchart
should be used. Other comments included making the "Eating
Right for Two" flipchart more interesting (n=2, 7.4% of
comments) or changing the wording (n=1, 3.7% of comments).
Client-Adapted Language flipchart users thought the
flipchart could be more interesting (n=1, 4.8% of comments)
and that more pictures could be added (n=1, 4.8% of

comments).

Materials Acceptability Survey: Handout Acceptability
Of the 128 participants provided with the handouts to

read on their own time, about 62% reported that they had
read at least one of the four: Iron (55.4%), Folic Acid
(50.7%), Protein/Calcium (56.2%) and Discomforts During
Pregnancy (48.4%). A majority of the women who read the
handouts thought that they were easy to read (>89%) and easy
to understand (>89%) (Table 30). However, it is quite
possible that the better readers actually read the handouts
and the poorer, less interested readers did not. The
handouts have 7th or 8th grade reading levels according to
the Flesch formula, making them easy to read for about half
of the EFNEP population which has a median reading skill
level of 8.3. In addition to an appropriate reading grade
level, all four handouts scored in the highest range (17-20
points) on the Readability Scorecard (Table 13), suggesting

that the handouts contain many characteristics of

184
readability.

Many of the clients thought that all four of the
handouts were useful (>80%) and the handouts on iron, folic
acid and protein/calcium were nutritionally informative
(70.8 to 76.9%). The handout, Discomforts During Pregnancy,
scored lower regarding nutrition information (43.5%),
probably due to the fact that it focuses on problems during
pregnancy that are not specifically nutrition-related. The
Folic Acid handout was reported to be the most interesting
(73.8%), and most useful (86.2%), followed by Discomforts
During Pregnancy in which 64.5% of respondents classified it

as interesting and 85.5% as useful.

Table 31: Frequency of Responses (Material Acceptability
Survey, Questions 15-19--Handout Acceptability)

"Eating Right...for Two" Handouts

 

 

 

 

 

 

 

Question Iron Folic Acid Proteinz Discomforts
(n=71) (n=65) Calcium (n=62)
(n=72)

15. Easy to 94.4% 89.2% 90.3% 95.2%
Understand

16. Easy to 91.5% 89.2% 90.3% 93.5%

Read

17. 74.6% 76.9% 70.8% 43.5%

Informative

18. 56.3% 73.8% 51.4% 64.5%

Interesting

19. Useful 83.1% 86.2% 80.6% 85.5%

 

 

 

 

 

 

 

3) Materials Comparison Survey

185

Data about material

preferences have been grouped into three categories

according to the three possible responses ("Flipchart X",

"Flipchart Y" and "About the same"; Appendix I) to each

survey question (questions 1-7).

for each item are presented in Table 32.

The frequency of responses

 

 

 

 

 

 

 

 

Table 32: Frequency of Responses (Comparison Survey)
Question Flip X Flip Y About the
same
1) Easier to Understand 27.2% 30.9% 41.9%
(n=136)
2) Uses More Words That 35.0% 29.2% 35.8%
I Would Use (n=137)
3) More Useful (n=137) 24.8% 19.0% 56.2%
4) Easier to 26.0% 36.0% 38.0%
Read(n=50)1
5) More Interesting 27.4% 18.5% 54.0% "
(n=138)
6) More Helpful (n=137) 30.7% 31.4% 38.0%
7) Women in EFNEP Would 36.5% 28.5% 35.0%
Like Better (n=137)

 

 

 

 

 

read the flipchart.

[1 ,5 A fi-f .
1. Part1c1pants were given the option to read the fIichart
or have it read to them.

 

Most participants chose not to

When EFNEP clients were given the two flipcharts

("Eating Right...for Two" and Client-Adapted Language) to

compare using the Materials Comparison Survey, many of the

women responded that the two flipcharts were about the same

186
in the areas of usefulness (56.2%) and interest (54.0%).
These results might indicate that both flipcharts are
equally useful rather than equally not useful, because in
the Material Acceptability Survey 79.1% of women who used
"Eating Right...for Two" agreed that it was useful and 91.8%
of women who used the Client-adapted Language flipchart
agreed that this flipchart was useful. A majority of women
agreed that the flipcharts were interesting, as well as
useful. Nearly 63% of women who used "Eating Right...for
Two" agreed that it was interesting and 72.1% of Client-
Adapted Language flipchart users agreed that this flipchart
was interesting, also. I

The Client-Adapted Language flipchart was thought to be
easier to understand by more women (30.9%) compared to
"Eating Right...for Two" (27.2%), but again more women
thought the two flipcharts were about the same in the area
of understandability (41.9%). The Material Acceptability
Survey showed that both flipcharts are easy to understand
according to 85.1% of women who used "Eating Right...for
Two" and 96.7% of women who used the Client-Adapted Language
flipchart.

Fewer women thought that the Client-Adapted Language
flipchart (29.2%) contained language similar to their own
compared to "Eating Right...for Two" (35.0%). About a third
of the participants (35.8%) thought that both flipcharts

contained similar language, despite the fact that the

187
Client-Adapted Language flipchart contained direct quotes
from an EFNEP-eligible population. However, the "Eating
Right...for Two" flipchart was pilot-tested with EFNEP
clients to improve language understandability. Comparing
these results to the results of the Material Acceptability
Survey, the Material Acceptability Survey showed high
percentages of clients thought that either flipchart used
language similar to their own (74.6% for "Eating Right...for
Two" and 83.6% for Client-Adapted Language).

More women perceived the Client-Adapted Language
flipchart as easier to read (36.0%) compared to "Eating
Right...for Two" (26.0%), yet 38% of the clients thought
there was little difference in readability between the two
flipcharts. The flipcharts are written at different reading
grade levels, with "Eating Right...for Two" being one grade
higher (Flesch score = 8th grade, Fry score = 7th grade)
according to the Flesch and Fry formulas. The majority of
participants in the study (60.9%) did not read the
flipchart, so this data might not be useful in evaluating
readability of the flipcharts for the EFNEP population. It
is possible that this large group of women might have
contained poorer readers who might have been more sensitive
to differences in the readability of the flipcharts. When
given handouts to read about 62% of participants read at
least one of four handouts. Although many women did read

the handouts, Michigan EFNEP should be aware of the large

188
segment of their population that might not-benefit from
reading material left with the client.

The results of the Material Acceptability Survey showed
50.8% of women who read "Eating Right...for Two" agreed that
it was easy to read and 42.7% of women who read the Client-
Adapted Language flipchart agreed that this flipchart was
easy to read. However, similar to the Materials Comparison
Survey results, >60% of participants did not read either
flipchart.

Nearly equal groups of women selected the Client-
Adapted Language flipchart or "Eating Right...for Two" or
"About the same" to answer the question on which flipchart
would be more helpful to women in EFNEP. However, more
women (36.5%) thought that EFNEP clients would like "Eating
Right...for Two" better than the Client-Adapted Language
flipchart (28.5%). Over one-third of the women (35.0%)
thought that women in EFNEP would like either flipchart
about the same.

In addition, data have been grouped into 6 categories
according to the responses to the survey questions
("Flipchart X", "Flipchart Y" and "About the same" and to
which set of materials the client used first (EFNEP or

Client-Adapted Language materials, See Table 33).

189

Table 33: Responses to Comparison Survey Questions,
Reported by the Order of Use of the Materials

 

 

 

 

 

 

 

QuestionZGroup1 Flip X(%) Flip Y
(Eating (%) the
Right...for (Client- same
Two) Adapted
Language) (%)
1) Easier to understand
Group 1 (n=70) 25.7 32.9 41.4
Group 2 (n=66) 28.8 28.8 42.4
2) Uses more words that
I would use
Group 1 (n=71) 38.0 29.6 32.4
Group 2 (n=66) 31.8 28.8 39.4
3) More useful
Group 1 (n=70) 24.3 15.7 60.0
Group 2 (n=67) 25.4 22.4 52.2
4) Easier to read
Group 1 (n=26) 33.3 33.3 33.3
Group 2 (n=24) 17.4 39.1 43.5
5) More interesting
Group 1 (n=71) 21.3 14.8 63.9
Group 2 (n=53) 33.3 22.2 44.4
6) More helpful
Group 1 (n=71) 22.9 34.3 42.9
Group 2 (n=66) 38.8 28.4 32.8
7) Women in EFNEP would
like better
Group 1 (n=71) 31.0 33.8 35.2
Group 2 (n=66) 42.4 22.7

 

 

 

 

 

1. Group 1=CI1ents wfio used the EFNEP fIipchart first

Group 2=Clients who used the Client-Adapted Language
flipchart first

190

Which flipchart the client used first (either "Eating
Right...for Two" or the Client-Adapted Language flipchart),
did not make a difference as to which flipchart the client
selected as easier to understand, using more similar words
to her own, more useful, easier to read, more interesting,
more helpful or better for women in EFNEP (Table 33). The
average frequencies of clients who (numbers are not

significantly different at p<.05 level):

used flipchart X first and chose X = 28.1 5 0.3%
used flipchart Y first and chose Y = 27.5 5 0.3%
used flipchart X first and chose Y = 27.8 5 0.7%
used flipchart Y first and chose X = 31.1 5 0.7%

Written comments from the Comparison Survey (question
8, "Do you see any other differences between Flipchart X and
Flipchart Y?") have been summarized and reported. Of 138
people who completed the survey, 45 made written comments
(32.6%). Fourteen people (31.1% of comments) simply stated
that the wording was different. One woman reported that the
two flipcharts were the same. Other comments are divided by

flipchart and are listed in Table 34.

191

Table 34: Written Responses about Differences Between

 

 

 

 

 

 

 

 

 

 

 

Flipcharts
Comment Flipchart X(%) Flipchart Y(%)
Eating Right... Client-Adapted
for Two Language
I(Easier to Understand 2.2 2.2
More for adults 6.7 0.0
Better for young 0.0 15.5
mothers
Less professional 0.0 4.4
More information/ 2.2 13.3
longer/addressed
specific subjects
Feels better/more 2.2 4.4
personable
Shorter/less 6.7 0.0
information
Better grammar 2.2 0.0
Talked down to/childish 0.0 4.4

 

 

 

 

Most written comments on the Materials Comparison
Survey addressed two subjects: the appropriate audience for
the flipchart and which flipchart had more information.
Three women (6.7% of comments) thought flipchart X ("Eating
Right...for Two") was more appropriate for adults and 15.5%
of the women who made written comments thought younger
mothers or teenage mothers would prefer flipchart Y (Client-
Adapted Language flipchart). The women who thought the
flipchart was appropriate for younger, teenage women were
all over 25. Possibly, they thought the Client-Adapted

Language Flipchart was less professional or more childish

192
and, therefore, more suitable to younger or less educated
women.

Many women (13.3% of comments) thought that the Client-
Adapted Language Flipchart contained more information on
subjects such as nutrition, weight gain and smoking. One
woman thought "Eating Right...for Two" had more information
on pregnancy in general, whereas, three women (6.7% of
comments) thought that this flipchart had less information
compared to the Client-Adapted Language Flipchart. In
reality, both flipcharts were 20 pages in length, but
"Eating Right...for Two" had fewer words (617) compared to
708 words in the Client-Adapted Language Flipchart.
Overall, more comments were positive (68.4% of comments on
"Eating Right...for Two" and 80.0% of comments on the
Client-Adapted Language Flipchart) than negative (31.6% of
comments on "Eating Right...for Two" and 20.0% of comments
on the Client—Adapted Language Flipchart) for either
flipchart.

In addition to summarizing written comments, mean
overall scores (numbers between zero and 100 given by
clients to rate the flipcharts) for both flipcharts have
been calculated (question 9). Women gave each flipchart a
score after using both flipcharts and comparing them. The
only criteria given on which to base the scores were that
"zero=terrible" and "100=wonderful". The mean score given

to "Eating Right...for Two" was 67.1 5 2.3 (S.E.M.) and the

193
Client-Adapted Language Flipchart was 70.1 5 2.1 (S.E.M.).
A paired T-test indicated that these two scores are not
significantly different (p<.05). For both flipcharts the
median score was 75 and the range of scores given was 9 to
100. Many women (31.3%) assigned both flipcharts identical
overall scores. Frequency of scores given are listed in

Table 35.

Table 35: Frequency of Overall Flipchart Scores

 

 

 

 

 

 

 

Range of Eating Right...for Client—Adapped H
Scores _Iue (n=112) Language (n=128)
(I 0-25 13 (11.5%) 10 (7.8%) H
II 26-50 38 (33.9%) 33 (25.8%) H
II 51-75 17 (15.2%) 23 (18.0%) "
76—100 44 (39.3%) 62 (48.4%)

 

 

Correlation coefficients were determined to compare
Comparison Survey overall scores to Knowledge Survey
posttest scores and Knowledge Survey change scores. The
data are presented in scattergrams (Figures 5, 6, 7 and 8).
Client scores for either flipchart X or flipchart Y are
positively correlated (p<.01) to Knowledge Survey posttest
scores, but not to Knowledge Survey change scores. The

correlation coefficients are:

Flipchart X and posttest scores r=.2155
Flipchart X and change scores r=-.0588
Flipchart Y and posttest scores r=.2164

Flipchart Y and change scores r=.1789

194
The EFNEP clients who were more knowledgeable about
nutrition during pregnancy (higher posttest scores) gave the
flipcharts higher overall scores compared to clients who
were less knowledgeable (lower posttest scores), regardless
of the flipchart used. However, the clients who greatly
increased their nutrition knowledge (higher change scores)
did not necessarily rate the flipchart higher than the

clients who increased their knowledge to a lesser extent

(lower change scores).

 

125

100- ....

75- - .

..for Two "

Overall score
"Eating Right.

 

 

 

.1

Posttest Score

Figure 5. Comparison Survey Overall Scores for "Eating
Right...for Two" vs. Knowledge Survey Posttest

Scores

195

 

 

 

 

125
t
.2
U
:2. 1m— 0 ° : ° . ° °
m 0 O O 0
%$ , , . . .
a 75— o : o
3 ' = '
13 50‘ - : ° : 2 ' '
”a O .
"‘ a
O-c °
0
m<' 25_ .
:=‘g , ° ,
8 o . .
95 - -
C):
0

T i
n In

0—
7.5"
10

a
Posttest Score

Figure 6. Comparison Survey Overall Scores for the Client-

Adapted Language Flipchart vs. Knowledge Posttest
Scores

 

 

 

 

125
= 100‘ ' °
0 . .
3 n -
F O
H o o o o c o o
:3 75- ° I . °
*3 o
22.—g, -
0.
8a 50—1 0 o o o o o 0
::°0 ° . .
a £1
8"?
> o o c o
05 . . .
. o o 9
O i i
V1 O in

10

Change Scores

Figure 7. Comparison Survey Overall Scores for "Eating

Right...for Two" vs. Knowledge Survey Change
Scores

196

 

 

 

 

125
E
"5
.e‘ loo—I o o o c O 0 0
E: . .
:g 0 ' 0 O C
a '75—: c o o o o
3 O O
'8 50" ' '
H
e 8
0'0
O
U) <| 25_,
= a
e 5 . .
9'5 - -
O =
0 I
.., 5 n , e
' Change Scores

Figure 8. Comparison Survey Overall Scores for the Client-

Adapted Language Flipchart vs. Knowledge Survey
Change Scores

V. Benefits, Strengths and Limitations of the Study

Benefits

From the results of this study, EFNEP program planners
and staff have become aware of the average reading level of
a sample of its population and have received detailed
information about the readability of its written materials
in the area of maternal nutrition. The organization
administrators can use this information when making changes
in existing materials or purchasing new materials in the
future. Procedures followed in the material development
phase, could be followed to revise other EFNEP lessons.
Michigan EFNEP has been provided with all materials produced
during this project for use in teaching pregnant women about
nutrition. Furthermore, Michigan EFNEP clients who are
pregnant will have access to acceptable, informative written
materials at an appropriate reading grade level. Finally,
information gathered in the study will be useful to other
nutrition educators who are interested in assessing the
readability of nutrition education materials or in producing
written materials at the appropriate literacy level for

particular populations.

197

198

Strengths and Limitations

According to Fitz-Gibbon and Morris (1987), the
critical characteristic of an evaluation is that it provides
the best possible information that could have been collected
under the circumstances at the time of the research. In
this study, several steps were taken to increase the
strengths of the research design and increase the validity

and usefulness of the results.

A. The Research Design

The greatest strength of this study lies in the
research design. Participants were selected from five
Michigan counties and randomly assigned to one of two
treatment groups (Research Procedures). Although the sample
selection does not involve random selection from all
Michigan counties, all new clients in the selected counties
were asked to participate and treatment assignment was
random. This random assignment of clients to groups
distributed the effects of individual differences between
the two groups. Therefore, the change in knowledge scores
between the pretest and posttest of the two groups could be
compared because the groups were equivalent. This design
provides information about change in knowledge within and
between groups, while controlling several threats to
validity. The effects of history, maturation, attrition,

and regression were assumed to be distributed equally

199
between the groups because random assignment of subjects to
groups was used.

Other threats to validity were addressed by this
research design. Client anxiety during testing can
interfere with obtaining valid results. Therefore, in the
Preliminary Phase of this research, the SORT was given to
women in their homes by their instructors to minimize
embarrassment associated with limited reading skills. Also,
in the Data Collection Phase, all surveys were administered
individually by the EFNEP instructor in the client’s home or
regular meeting place to decrease anxiety. However, it was
impossible to control for ringing telephones, noise from
children, or other interruptions which could have decreased
the attention clients gave to the surveys. Response bias, a
bias due to the desire to please (Henerson et al., 1987)
could have played a role in the results of the Materials
Acceptability Survey. Greater than 70% of all clients
agreed that the flipchart they used first was easy to
understand, informative and useful, regardless of which
flipchart they were given. Although respondents were told
that responses would not affect their standing in EFNEP,
individuals could have responded favorably in an effort to
please the instructor or these results could reflect true
quality and appropriateness of materials.

The issue of the validity of the test instruments was

addressed, also. According to Morris et al. (1987) mistakes

200
made when scoring a test decrease the measure’s validity.
Therefore, steps were taken to decrease scoring errors. All
scores were entered into a computer by the primary
investigator and checked independently, by a second
researcher to minimize discrepancies in scoring. All scores
were entered in number form, eliminating subjective scoring.
All instructors were trained in specific procedures for data
collection to minimize variation in procedures or assistance
to clients.

Ideally, the results of this study would be
generalizable to the Michigan EFNEP population. The total
sample size of the Data Collection Phase was intended to be
331 participants, large enough to generalize results to the
Michigan EFNEP population as a whole. However, the length
of time estimated to process each client was underestimated
by the investigator and EFNEP, resulting in 240
participants. However, the results of this study are
generalizable to the EFNEP population in the five

participating counties in southern Michigan.

B. Test Instruments

Appropriate evaluation requires valid and reliable test
instruments. Assessments of validity and reliability help
to determine the amount of faith people can have in the
results of an instrument (Morris et al., 1987). For the

Preliminary Research Phase, a standardized instrument was

201
used (the SORT) that had been previously tested for validity
and reliability (Appendix B). The scores obtained from the
SORT test have been shown to correlate well with scores of
the reading grade level of text. Therefore, text with a
reading grade level of 6 would be appropriate for clients
with a SORT score of 6. Limitations of the SORT include
variations in administration and assistance of instructors.
Instructors were trained in the administration of the SORT
to minimize possible variations.

For the Data Collection Phase, three instruments were
developed by the investigator (Knowledge, Material
Acceptability and Materials Comparison Surveys) and tested
for validity. The Knowledge Survey was tested for
reliability, also.

Reliability can increase the usefulness of an
instrument because it determines whether the instrument will
yield consistent results. Reliability of an instrument
pertains to the specific audience for which the instrument
is tested. An instrument might prove reliable for one
group, but not for another; therefore, there is a need to
test the internal consistency of an instrument with the
specific population with which it will be used (Fitz-Gibbon
and Morris, 1987). In this research, internal consistency
was determined by using a KR-20 formula on the results of
the pilot test. For this assessment, the value of the score

lies between zero and one with the higher score indicating a

202
higher correlation of test items. The Knowledge Survey had
a score of .874, indicating a strong intercorrelation of
test items.

Reliability is necessary, but alone it is not
sufficient to justify the use of an instrument (Talmage and
Rasher, 1981). Unless an instrument is valid for a specific
purpose, a high reliability coefficient has little value.
According to Morris et al., 1987, judgements of validity
determine if the instrument assesses what is intended to be
measured. In order to be valid, an instrument must match
the depth and scope of its intended topic, must be
reasonable and understandable to the target audience and
must bear a measurable relationship to the characteristic it
is intended to assess (Talmage and Rasher, 1981). All of
these issues were considered in the development of test
instruments for this research.

Construct validity, the extent to which you can be sure
the instrument represents the subject of interest (Morris et
al., 1987) was assessed for the Knowledge Survey by
calculating a discrimination index of results of the pilot
test. Items with a low score (<.33) would have been revised
or omitted, but no items scored below .33. Difficulty of
the items was assessed by calculating the percent of people
who answered each question correctly (difficulty index).
Only one item scored below 20% (difficult), but it was not

removed or revised because it met the EFNEP objectives.

203

Content validity, the extent to which an instrument
measures specific behaviors (Morris et al., 1987) was
assessed through expert review. For the Knowledge Survey,
competencies of the EFNEP curriculum "Eating Right...for
Two" were obtained and used for the development of
questions. Two questions were developed for each
competency. All three surveys underwent expert review by
six individuals with expertise in one or more of the
following areas: familiarity with EFNEP, knowledge of
nutrition in pregnancy, knowledge of data analysis or
experience with survey development. In addition, all
surveys were measured for readability and found to have
Flesch scores 5 8.3, the median reading level of a sample of

the Michigan EFNEP population.

C. Other Materials Developed by the Primary Investigator

The strengths of the materials produced by the primary
investigator in the Materials Development Phase of the study
(Client-Adapted Language flipchart, Readability Scorecard
and Pica handout) included: participation of members of the
target population in the wording of the Client-Adapted
language flipchart, assessment of the materials for
readability and expert review of content. Actual quotes
from EFNEP-eligible pregnant women were used to develop the
Client-Adapted Language flipchart. Nitzke (1987) found this

method beneficial in developing a snack pamphlet for low-

204
income, low-literate mothers. She reported that the target
audience found the pamphlet more personally relevant and
readers were less likely to reject the information expressed
in peer language than the same information expressed in the
language of nutrition educators. In this study, the
flipchart developed using peer language significantly
increased maternal knowledge of nutrition in pregnancy.

In other studies, a serious disparity has been found to
exist between the reading level of the target population and
the skill required to read health-related materials
(Freimuth, 1979; Meade et al., 1989). Therefore, in this
study the Client-Adapted Language flipchart and Pica handout
were written at or below the median reading level of a
sample of the target population (5 8.3 Flesch score) and
other characteristics of readability not measured by
formulas were assessed using the readability scorecard.

The Readability Scorecard, Client-Adapted Language
flipchart and Pica handout underwent pilot-testing and/or
expert review by individuals with expertise in the subject
area. The Readability Scorecard was pilot-tested with
graduate students to assess clarity of instructions, then
underwent expert review by individuals with experience in
readability: Dr. Jenny Bond, Dr. Anne Murphy, Ms. Char McKay
and Ms. Pat Hammerschmidt. The Client-Adapted Language
flipchart was produced and the wording reviewed for accuracy

by Dr. Jenny Bond. The flipchart was not pilot-tested

205
because the research committee and the investigator felt
that the major part of the study was testing the flipchart.
However, pilot-testing might have been beneficial in
detecting specific words that could have been offensive to
the target audience.

In addition, a handout on pica was develOped as an
outgrowth of the Client-Adapted Language flipchart after
EFNEP clients expressed an interest in the topic. The
handout was written by the primary investigator and reviewed
by Dr. Carolyn Lackey of North Carolina State University, an

expert on the subject of pica.

RE COMMENDAT I ONS

Based on the results of this research, the primary
investigator has several suggestions for the development and
testing of nutrition education materials, training data

collectors and future research.

A. Working within the EFNEP

This study was conducted with the assistance of several
individuals at the Expanded Food and Nutrition Education
Program at Michigan State University. This assistance was
invaluable. However, the primary investigator did not have
direct and regular contact with EFNEP workers in the
counties during the planning of the project. Many questions
arose during the planning of the data collection phase (i.e.
Do clients ever read the flipchart? or How many clients can
one instructor see in two months?) that could have been best
answered by individuals working in the counties.
Representatives from the counties would be an asset in
planning the data collection because they could highlight
differences between counties and give researchers an idea of
what is a realistic timeline. In the future, researchers
working with EFNEP clients might benefit from having county
Extension home economists and instructors serve as an
advising committee to be involved in the initial and on-

going decision making process.

206

207
B. Pilot-Testing of Materials

The surveys used in this project were pilot-tested for
understandability of the wording used, but the Client-
Adapted Language flipchart was not pilot-tested. The
research committee and the investigator felt that the
testing of the flipchart was part of the data collection
phase of the project. However, pilot-testing could have
provided valuable information on whether the message in the
material attracted attention, was understood, was personally
relevant, was believable or offensive (USDHHS, 1982).
Researchers have reported that developers of nutrition
education materials should seek audience feedback on
prototypes even when prototypes were developed with the help
of the audience (Shepard et al., 1994). In the process any
errors in translation of reported preferences can be
corrected.

Several written comments (3.5%) were returned
complaining about the insulting nature of specific words.
These words could have given women using the flipchart a
different perception of the writing than the same flipchart
with the words changed to more acceptable wording. In the
future, all materials developed by the researcher should be
pilot-tested with the target audience. Pilot-testing only

strengthens the actual research project.

208

C. Instructor Training

When working with an organization such as EFNEP,
confidence in the validity of the data is only as good as
the confidence in the individuals collecting the data. In
this case, EFNEP instructors were trained by the
investigator in data collection procedures to minimize
variations that could threaten validity. During training
questions arose in some counties as to why instructors or
clients must read the flipchart to the client if they
ordinarily paraphrased it. Other instructors expressed
frustration because they thought the goal of the research
project was to replace the original flipchart with the
Client-Adapted Language flipchart and they had not been
consulted about the development of the new material. Still
other instructors complained about some of the wording as
not sounding professional. The researcher explained that
the idea was to test the process of using the clients’ own
words to make materials in the future and not to replace the
existing flipchart or force instructors to read flipcharts
in the future. The instructors were willing to follow the
research procedures for data collection, but steps could
have been taken to minimize the initial frustrations.
Listed are some ideas for decreasing those frustrations in
future research:

1) The material chosen to be tested should be one

(i.e. a handout) that the client normally reads

2)

3)

4)

209
anyway.
Instructors need to be involved in the development
of materials by helping with collecting quotes or
pilot-testing or playing some other role in the
process.
Instructors need to be motivated to use materials
that they might think are less than professional
if the materials increase client knowledge of
nutrition. County Extension home economists need
to be consulted during project planning about the
details of the project and encouraged to motivate
their instructors.
Results of studies such as this one need to be
shared with EFNEP program directors, county
Extension home economists and instructors working
with materials in the field to convince everyone
involved of the benefits of the process of using

client input to produce materials.

SUMMARY

The Client-Adapted Language flipchart consisted of 20
written pages of text on nutrition in pregnancy developed
using direct quotes from EFNEP-eligible women in Kalamazoo
County. When EFNEP women used this flipchart, their
knowledge of nutrition in pregnancy increased to a greater
extent than women using the original EFNEP "Eating
Right...for Two" flipchart as indicated by significant
differences in Knowledge Survey posttest and change scores.
The age, race, highest level of education achieved,
pregnancy status or number of babies delivered by the women
seemed to have no bearing on Knowledge Survey outcomes.
These results show promise for the future development of
written materials for EFNEP clients of all ages, races and
education levels using input from EFENP clients.

However, for EFNEP women to use written materials to
increase their knowledge, the materials must be acceptable.
Most study participants agreed that either flipchart was
easy to understand, informative, easy to read, interesting,
useful and answered their nutrition questions. The Client-
Adapted Language flipchart scored slightly higher than the
original "Eating Right...for Two" flipchart in all of these
areas, and lower in the area of easy-to—read. Several women
commented (3.5%) that the words "thing" and "stuff" used in

the Client-Adapted Language flipchart were insulting or not

210

211
professional. In the future, problem words that might be
perceived negatively should be weeded out during pilot-
testing.

When the two flipcharts were compared many women
thought that the flipcharts were about the same in the areas
of understandability, usefulness, interest and helpfulness.
More women chose "Eating Right...for Two" as the flipchart
that women in EFNEP would like better than chose the Client-
Adapted Language flipchart. The reasons for this choice as
stated in the written comments included, "Eating Right...for
Two is for more educated women" and "The Client-Adapted
Language flipchart used slang and stuff." There were many
positive written comments on the Client-Adapted Language
flipchart, also. Women stated that it helped them
understand the importance of weight gain, had more
information, was more personable and was easier to
understand. In the final survey question when women were
asked to rate each flipchart from one to one-hundred, there
was no significant difference between the two ratings.

The process of using quotes from members of a
population to produce written materials for that population
has a major advantage of producing understandable materials
that can increase knowledge. The challenge now is to make
everyday spoken language as acceptable as polished, more
professional language that might not be as understandable,

regardless of reading grade level. Some issues to be

212

addressed when repeating this process are:

a)

b)

to pilot-test all materials, looking for specific
words that might be perceived as insulting.

to motivate instructors to portray the materials
in a positive light for the benefit of the client,
regardless of personal desires for "professional"

materials.

APPENDICES

APPENDIX A

Fry Graph

213

Avevcoewdmur In“

toe- In no no III In I): In I40 nu III I” use no not ill In I]. too tlzo
2300

 

zoo ace
'07 “BI
'l) II)
n: In
II! III
too ".10
9| 9|
: a: I:
= .. II
9: 7‘ ll
- 67 u
3 s: n
f- 59 so
E, so as
E 5? s:
g so so
3 a! cl
'5’ A: as
5 a] I)
g 42 ‘2
i’ no ‘0
‘ )0 u
21 II
3‘ J.
as J:
:J J:
)0 :o
25 2:
2 20

‘03.”: H6 '20 In :20 I)? I)! no In no u: use too we nu In In '00 I029

DIRECTIONS Randomly select 3 one hundred word passages from a book or an article. Plot
average number of syllables and average number of sentences per 100 words
on graph to determine the grade level of the material. Choose more passages
per book if great variability Is observed and conclude that the book has un-
even readability. Few books will fall in gray area but when they do grade level
scores are invalid. '

Count proper nouns. numerals and initializations as words. Count a syllable

for each symbol. For example. ‘l945' is l word and 4 syllables and ‘lRA‘ is
l word and 3 syllables.

 

 

EXAMPLE: SYLLABLES SENTENCES
lst Hundred Words 124 6.6
2nd Hundred Words MI 5.5
3rd Hundred Words 158 6.8
AVERAGE l4! 6.3

READABILITY 7th GRADE (see dot plotted on graph)
Reproduction Permitted. No Copyright.

APPENDIX B

SORT Test

2114

To find a child's me for reading. count
the total number of words he was able to pronounce
correctly in all lists and add the words below the
starting list for which he automatically receives
credit. To obtain the Reading Level. look up the
value of this raw score in Table 1 below. A Simple
way to determine the Reading Level is to take half
the raw score. For example. if the raw score were
46.halfof'this number would be 23 and the Reading
Level would be 2.3 or the 3rd month of 2nd grade.

 

 

 

 

 

 

 

 

 

SCORE GRADE scone GRADE SCORE GRADE scan GRADE'
O—i 0.0 25-27 ; 1.3 52-53 2.5 78-79 3.9
2-3 0.2 29-29 1.8 58-55 2.7 80-8l 8.0
8-5 0.2 30-3l 1.5 55-57 2.8 82—83 8.1
5-7 0.3 32-33 1.5 58-59 2.9 88-85 8.2
8-9 0.8 38-35 (.7 50-5: 3.0 85-87 8.3
lO-Il 0.5 35-37 |.8 52-53 3.: 88-89 8.8
(2-l3 0.5 38-39 1.9 58-55. 3.2 90-9l 8.5

-Iu-Is 0.7 80-8) 2.0 55-57 3.3 92-93 8.5
:5-17 0.8 82-83 2.) 58-59 3.8 98-95 8.7
l8-l9 0.9 88-85 2.2 70-7l 3.5 95-97 8.8

20-2l l.o 85-87 2.3 72-73 .3.5 98-99 8.9
22-23 1.: 88-89 2.8 78-75 3.7 I00-i0i 5.0
28-25 :.2 50-51 2.5 75-77 3.8 l02-i03 5.:
scoRE GRADE SCORE GRADE SCORE GRADE?
, ID8-Ios 5.2 :Iao-Ial 5.5 155-:57 7.3
; i06-lO7 5.3 :132-I33 5.5 lSB-lSQ 7.9
. |08-|09 5.8 I38-I35 5.7~ 'lSO-lai 8.0
' ilO-ili 5.5 l36-i37 5.8 I62-IG3 3.!
li2-II3 5.5 l38—i39 5.9 I58-I55 8.2
II8-II5 5.7 I8o-I8I 7.0 I55-l57 8.3
il6-ll7 5.8 182-I83 7.: l68-I69 8.8
||8-|i9 5.9 188-185 7.2 l70-i7l 'a.5
l20-l2i 5.0 185-I87 7.3 172-173 8.6
122-123 5.: ~188-I89 7.8 178-I75 8.7
128-I25 5.2 lSO-i5) 7.5 l76-l77 3.3
l26-i27 5.3 l52-l53 7.5 l78-i79 5.9
l28-l29 5.8 158-155 7.7 (so-zoo H.S.

 

 

 

 

215

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

List P (to) List 1 («n . List 2 (so)
1. sec 1. with 1, game
2. look 2. friends 2. hide
3. mother 3. came 3. grass L1“ 3 (to) U" 4 (too) List 5 (tan
4. little 4, horse 4, across 1 sale 1 harness 1. cushion
" . ° 5 . around 2 against 2 pncc 2 generally
5 ° here . 5 ride 1) 1d 3 smash 3 flakes 3 extended
6- can 6' under 6 ° rea 8'“ '4 reward 4 silence. 4 custom
'i . want 7- W85 7- field 5 evening 5 develop 5 tailor
3 . come 8, what 8, large 6 stream 6 promptly 6 haze
9. one ‘9. bump 9. better 3 32:? Z, 323:; :1 fig“
- 8'" Y
10- baby 10' I’ve 1° ° suddenly 9 grove 9 forehead .9 terrace
11. three 11. very 11. happen to desire 10 distant. to applause
12, run 12. puppy 12, farmer 1! ocean 11 anger 11 jungle
13- jump 13- dark - 13- "V“ 1?. 32'3" ii 13““im if $2233.”
14. $101.11!" 14- first 14- lunch 14 timid 14 spcccltlPea cos 1:1 marriage
15,15 15. Wish 15. sheep 15 perform 15 region is profitable
15. up 15; basket 16. hope 16 destrpy 16 slumber 16' define
1". make 17- food 1". “rest if. 332?.“ if. £131.11 if. “if-‘9’"
19. help 19. lull 19. heavy 211 understood 20 dainty 20 .merchant
20 play 20. along 20. station
P =====l —_.._- _______5
E i. #5
List 6 it“) List 7 (160) List 8 (in) "1:1! School (299) SCORE
1 ‘ installed 1 administer 1 prairies 1 traverse
2 importance 2 “311.10? 2 evident '2 afiable , 1.1:: P
3 medicine 3 , enmonment 3 nucleus 3 compressible List 1
4 rebellion 4 counterfeit 4 antique 4 excruciating u“ z
s infected 5 931515 . 5 twilight S pandemoniurn 1
6 ’“PWSiblc 6 ”‘1‘”um 6 memorandum 6 scrupulous ,. L 't 3
7 .liquid 7 approxunate 7 whimsical 7 primordial 1.1:: ‘
3 tremendous 3 50°13” 3 propqnioncl 8 chastisement ' 1.1:: s
9 customary 9 architecture 9 intangible 9 sojourn - 1.
In malicious 111 malignant 10 fomulgted 10 panorama “t 6
‘1 spectacular U pensive . 11 articulate (11 facsimile u" 7
12 inventory 1‘3 standardize 12 deprecate 12 auspicious Lista'
1:1 yearning 13‘ exhausted 13 remarkably 13 contraband List 11. 5
1-1. Imaginary l4 reminiscence 14 contrasting 14 envisage . O
15 consequently 15 intricate 15 irrelevance 15 futility '
16 excellence 16 contemporary 16 supplement [6 cnamoured :"
1.? dungeon 17 attentively l7 inducement 11 cry a n
3' detained 18 compassgonatc 8 nonchalant u: ecipher (131.1"... , if
19 abundant 1" complexion 19 exuberant 19 inadequacy correct o. '4,
L2“ compliments 2° °°““"“°le 20 grotesque 20 simultaneous including the
no rd s~ b c i o v
=—== ======i ’t'rufli I‘V‘Ll'

 

 

 

 

 

 

 

 

APPENDIX C

Irwin/Davis Checklist

216

Readability checklist

 

This checklist is designed to help you evaluate the readability at your classroom texts
It can best be used it you rate your text while you are thinking at a spec-tic class. Be sure
to compare the textbook to a fictional ideal rather than to another text. Yom goal is to
iind out what, aspects oi the text are or are not less than ideal. Finally. consider
supplementary workbooks as part at the textbook and rate them together. Have tun!

Rate the questions below using the tollowing rating system:
5 - Excellent
4 - Good
3 - Adequate
2 - Poor
1 - Unacceptable ,
NA - Not applicable

Further comments may be written in the space provided.
Textbook title:

Publisher:
Copyright date:
Understandability

 

 

 

 

A. _ Are the assumptions about students‘ vocabulary knowledge appropriate?

 

 

8. Are the assumptions about students' prior knowledge at this content area
appropriate?

C. Are the assumptions about students' general experiential backgrounds
appropriate?

D. __ Does the teacher's manual provide the teacher with ways to develop and review
the students' conceptual and experiential backgrounds?

E. _ Are new concepts explicitly linked to the students' prior knowledge or to their
experiential backgrounds?

F. __ Does the text introduce abstract concepts by accommnying them with many

concrete examples?
G. _. Does the text introduce new concepts one at a time with a sufficient number oi

examples tor each one?

Are detlnitions understandable and at a lower level oi abstraction than the
concept being detined?

l is the level at sentence complexity apprOpriate tor the students?

J. _. Are the main ideas at paragraphs. chapters. and subsections clearly stated?
K. __ Does the text avoid irrelevant details?

L.

_ Does the text explicitly state important complex relationships (e.g.. causality.
conditionality. etc.) rather than always expecting the reader to inter them trom
the context?

M. _ Does the teacher's manual provide lists oi accessible resources containing
alternative readings tor the very poor or very advanced readers?

 

 

APPENDIX D

Approval to Conduct Study

217

1. Approval to Conduct Study: The Preliminary Research
Phase

MICHIGAN STATE UNIVERSITY

 

OFFICE OF VICE PRESIDENT FOR RESEARCH EAST LANSING 0 MICHIGAN 0 48824-1046
AND DEAN OF THE GRADUATE SCHOOL

July 23, 1992

Karen Schmitz
236 Food Science

RE: DEVELOPING NUTRITION EDUCATION MATERIALS FOR LOW-LITERATE
AUDIENCES- DETERMINING EFNEP PARTICIPANTS’ READING LEVEL, IRB
#92-355 '

Dear Ms. Schmitz:

The above project is exempt from full UCRIHS review. The proposed
research protocol has been reviewed by a member of the UCRIHS
committee. The rights and welfare of human subjects appear to be
protected and you have approval to conduct the research.

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 UCRIi-IS approval one month
prior to July 22, 1993.

Any changes in procedures involving human subjects must be reviewed
by 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. If I can be
of any future help, please do not hesitate to let me know.

Sincerely,

 
  

  

  

David E. Wrigh@.D., Chair

University Conn -e on Research Involving
Human Subjects (UCRIHS)

218

2. Approval to Conduct Study: The Data Collection Phase

July 26, 1993

TO: ' Ms. Karen Schmitz
236 Food Science Building

RE: IRB #: 93~313
TITLE: AN EVALUATION OF THE READABILITY AND
ACCEPTABILITY OF MATERNAL NUTRITION
EDUCATION MATERIALS FOR EFNEP CLIENTS

REVISION REQUESTED: N/A
CATEGORY: 1-A, l-B
APPROVAL DATE: 07/19/1993

The University Committee on Research Involving Human Subjects’ (U CRIHS) review of this project
is complete. I am pleased to advise that the rights and welfare of the human subjects appear to be
adequately protected and methods to obtain informed consent are apprOpriate. Therefore, the
UCRIHS approved this project including any revision listed above.

Renewal: UCRIHS approval is valid for one calendar year, beginning with the approval
date shown above. Investigators planning to continue a project beyond one year
must use the enclosed form to seek updated certification. There is a maximum of
four such expedited renewals possible. Investigators wishing to continue a project
beyond that time need to submit it again for complete review.

Revisions: UCRIHS must review any changes in procedures involving human subjects, prior
' to initiation of the change. If this is done at the time of renewal, please use the
enclosed form. To revise an approved protocol at any other time during the year,
send your written request to the UCRIHS Chair, requesting revised approval and
referencing the project’s IRB # and title. Include in your request a description of
the change and any revised instruments, consent forms or advertisements that are
applicable. the year, please outline the proposed revisions in a letter to the
Committee.

Problems]

Changes: Should either Of the following arise during the course of the work. investigators
must notify UCRII-IS promptly: (1) problems (unexpected side effects, complaints,
etc) involving human subjects or (2) changes in the research environment or new
information indicating greater risk to the human subjects than existed when the
protocol was previously reviewed and approved.

If we can be of any future help, please do not hesitate to contact us at (517) 355-2180 or FAX (517)

336-1171.

David E. Wright, Ph.D.
UCRIHS Chair

Sincerely,

  

APPENDIX E

Questions to Obtain Client Language

219

Questions to Obtain Client Language

The following questions were asked of three separate groups

of EFNEP eligible pregnant women who had just finished a

parenting class which covered the information in "Eating

Right...for Two" on nutrition during pregnancy. The purpose

of the question/answer sessions was to obtain original

language for the Client-Adapted Language materials used in

this research.

1.

Your friend Diane is three months pregnant. She
asks you what foods she should eat while she is
pregnant. What would you tell her?
Diane was normal weight for her height before
pregnancy. Why should Diane be eating more food
now than she ate when she was not pregnant?
Is it important to gain weight during pregnancy?
Why?
What if Diane was overweight before pregnancy and
she says she doesn't want to gain weight while
she's pregnant.

a. What advice would you give her?

b. What if she said she wanted to diet while

pregnant (reduce her weight)?

What if Diane says, "I don’t want to gain weight
so I can have a small baby." What would you say

to her?

10.

11.

12.

13.

14.

220
How much weight should a woman gain during
pregnancy?
a. In the first 3 months?
b. In the last 6 months?
What does "slow and steady" weight gain mean to
you?
When a woman gains weight during pregnancy, where
does the weight go?
a. Why does the amount of blood in the woman’s
body need to increase during pregnancy?
b. Why do breasts grow larger?
c. Why does the uterus get larger?
What does "eat a variety of foods" mean to you?
Why should a pregnant woman eat a variety of
foods?
How do nutrients get from the mother’s mouth to
the baby?
a. What nutrients do you need to eat more of
during pregnancy?
b. What foods contain calcium? Why do you

need increased calcium? When?

c. iron?

d. folic acid?

e. protein?

a. Should pregnant women take a vitamin/mineral

supplement? Why or why not?

15.

16.

17.

18.

221

b. Are you taking a vitamin/mineral supplement?
Why or why not?

a. Why shouldn't a pregnant woman smoke?

b. What else harms the baby and should be
avoided by pregnant women?

a. Have you had any cravings for certain foods
or even things that aren’t foods?

b. DO you have friends or relatives that ate
strange things while they were pregnant?

Have you heard any myths about pregnancy from

friends or relatives, such as pregnant women

shouldn’t cross the railroad tracks?

Is there anything else that you think pregnant

women should know about nutrition or healthy

habits?

APPENDIX F

Knowledge Survey (First Pilot)

222

Knowledge Survey

Knowledge Survey: This survey will be called Pregnancy

Survey I when given as a pretest and Pregnancy Survey II

when given as a posttest.

Pregnancy Survey I (Pilot)

 

DIRECTIONS for instructors to read to homemakers: Please
answer the first 4 questions by putting an "X" on the
correct line. You can choose not to answer any item. Do

not put your name on this form.

1. How old are you now? years
I do not wish to answer

2. Are you pregnant now?

yes

I10

I do not know/I do not wish to answer

survey 1, page 1

223

How many babies have you had?

What

none
one
two
three or more

I do not wish to answer

is your race?

African American/Black

Indian (American)/Alaskan Native
Asian/Pacific Islander

White

other what?

 

I do not wish to answer

Are you of Hispanic origin? yes no

What was the highest level of education you completed?

grade
OR
high school graduate or GED
some college or professional school

college graduate

survey 1, page 2

224
DIRECTIONS for instructors to read to the homemaker: I will
read the questions and each answer choice. You need to
choose the best answer. Circle only one choice for each
question. Hg flipcharts or handouts should be used while

taking the survey.

1. Ann is pregnant. How many servings of milk or cheese

or yogurt should she have each day?

a) l
b) 2
c) 3
d) 4
2. A daily meal plan for a pregnant woman should include 3

to 5 servings of

 

a) vegetables
b) fruits
c) breads and cereals
d) milk and cheese
e) meat, poultry, fish, eggs and beans
3. Slow and steady weight gain in pregnancy means:
a) gaining about 2 pounds per week throughout pregnancy
b) gaining less than 25 pounds by the end of pregnancy
c) gaining 5 pounds per month throughout pregnancy
d) gaining about 1 pound per week, after the first 3

months of pregnancy

225
Weight gain in pregnancy is important because:
a) the baby is growing
b) the mother's breasts are getting larger
c) the mother’s uterus is getting larger
d) a and b only

e) all of the above

The need for calcium:

a) increases during pregnancy

b) decreases during pregnancy

c) does not change from non-pregnancy to pregnancy

d) Calcium is not needed during pregnancy.

The need for which nutrient(s) increase(s) during
pregnancy: '

a) iron

b) folic acid

c) protein

d) all of the above

e) a and b only

survey 1, page 4

Betty needs to eat more

226

Tina is pregnant. She needs to increase the number of
servings of from 2 to 3 each day.
a) fruit

b) milk and cheese
c) breads and cereals

d) all of the above

Betty is pregnant. Yesterday she ate the following:
fruits: 2 servings

milk and cheese: 3 servings

meat, poultry, fish, eggs and beans : 4 servings
breads and cereals: 4 servings

vegetables: 3 servings

 

a) fruits

b) vegetables

c) meat, poultry, fish, eggs and beans
d) breads and cereals

e)'milk

survey 1, page 5

APPENDIX G

Knowledge Survey (Final Instrument)

227

Pre anc Surve I Final Instrument

Pre anc Surve I
DIRECTIONS for EFNEP Instructors: Please read each question
and all the answers listed to the homemaker. Try to use the
same tone of voice for all responses so you don’t give away
the correct response. Ask the homemaker to select the best
answer without your assistance. If she does not answer
right away, read the question and answers again. She gag
choose not to answer any item. Please answer the first 4
questions by putting an "X" on the correct line. Do not put

the homemaker’s name on this form.

228
DIRECTIONS for instructors to read to the homemakers:
Please answer the first 4 questions by putting an "X" on the
correct line. You gag choose not to answer any item. Do

not put your name on this form.

1. How old are you now? years
I do not wish to answer

2. Are you pregnant now?

yes

no

____ I do not know/I do not wish to answer
3. How many babies have you had?
none

one

two

three or more

I do not wish to answer

survey 1, page 1

229
What is your race or origin?
African American/Black
Indian (American)/Alaskan Native
Asian/Pacific Islander
Hispanic
White (not Hispanic)

other what?

 

I do not wish to answer

What was the highest level of education you completed?
grade
OR

high school graduate or GED

some college or professional school

college graduate

survey 1, page 2

230

DIRECTIONS for instructors to read to the homemaker: I will

read the questions and each answer choice. You need to

choose the best answer. Circle only one choice for each

question. N9 flipcharts or handouts should be used while

taking the survey.

1. Ann is pregnant. How many servings of milk or cheese

or yogurt should she have each day?

a) 1
b) 2
c) 3
d) 4
2. A daily meal plan, including snacks, for a pregnant

woman should include 3 to 5 servings of

 

a) vegetables
b) fruits
c) breads and cereals
d) milk and cheese
e) meat, poultry, fish, eggs and beans
3. Slow and steady weight gain in pregnancy means:
a) gaining about 2 pounds per week throughout pregnancy
b) gaining less than 25 pounds by the end of pregnancy
c) gaining 5 pounds per month throughout pregnancy
d) gaining about 1 pound per week, after the first 3

months of pregnancy

231
4. Weight gain in pregnancy is important because:
a) the baby is growing
b) the mother's breasts are getting larger
c) the mother’s uterus is getting larger
d) a and b only

e) all of the above

5. Which nutrients are n9; needed in extra amounts during
pregnancy?
a) protein
b) folic acid
c) calcium

d) The need for each of these nutrients does increase

 

during pregnancy.

6. The need for which nutrient(s) increase(s) during
pregnancy:
a) iron
b) folic acid
c) protein
d) all of the above

e) a and b only

survey 1, page 4

232

7. Tina is pregnant. She needs to increase the number of
servings of from 2 to 3 each day.
a) fruit

b) milk and cheese
c) breads and cereals

d) all of the above

8. Betty is pregnant. Yesterday she ate the following:

fruits: 2 servings

milk and cheese: 3 servings

meat, poultry, fish, eggs and beans : 4 servings

breads and cereals: 4 servings
vegetables: 3 servings

Betty needs to eat more

 

a) fruits

b) vegetables

c) meat, poultry, fish, eggs and beans
d) breads and cereals

e) milk

survey 1,

page 5

APPENDIX H

Characteristics of Readability: Scorecard

233

CHARACTERISTICS OF READABILITY: SCORECARD

According to Pickert and Elam (1985), use of
readability formulas should be supplemented by other means
of judging the quality of printed text. This checklist is
designed to help assess many of the characteristics of
readability that are not necessarily taken into account by
formulas that measure reading grade level. The checklist
compares written text to recommendations made by experts in
education and nutrition education: Sticht, 1975; Klare,
1976; Laubach and Koschnick, 1977; US DHHS, 1982; Doak et

al., 1985; Fry 1988; Nitzke, 1989.

Directions: Assess the following ten characteristics of a
written text (or of three randomly selected pages for long
texts) using the instructions for assessment outlined on the

following page.

234

Characteristics or Readability: Assessment

Characteristic Description of Assessment "

 

1)

Use of voice

The number of times the passive voice
(PV) was used, rather than active
voice, were counted. Total number of
sentences were counted. Use of voice =
PV/total number of sentences.

   

 

2)

Type size

The size of type of the main body of
the text was compared to known type
sizes printed in Laubach and Koschnick,
1977.

  

 

3)

Font type

 
 
 

The font type of the main body of the
text was compared to Roman serif type.
The type was determined to be either
Roman serif or not Roman serif.

 

4)

Abbreviations

The number of abbreviations (AB) in the
text were counted. Contractions of
words were not counted as
abbreviations. The total number of
words were counted. Percent of AB =
(total number of AB/total number of
words) x 100

 

 

5)

Emphasis on
action

‘telling the reader what not to do was

The number of times negative action
(NA) was emphasized, as opposed to
positive action, in a command or
suggestion were counted. Any sentence

 

counted. Percent NA = The total number
of NA/total number of sentences.

 

6)

 

Letter type

 

The main body of the text was examined.
It was noted that the text was written
in either capital letters or lower case
letters (excluding the first letter of

each sentence).

  

 

. Table (continued)

235

 

7) Symbols

The total number of symbols (NS) used
in the main body of the text were
counted. Some examples include "+", "%"
and "&". Percent symbols =(NS/total
number of words) x 100

 

8) Headers

The number of headers (H) used in the "
text were counted. The title was
counted if it appeared on a written
page of text being analyzed. The
number of paragraphs (P) in the text
were counted. Percent headers = (H/P)
x 100

Note: Computer programs count a new
paragraph any time there is a break in
the leading. For accuracy, you might
want to count paragraphs by hand.

 

 

9) White space:
Margins and
area around
headers

Boxes were drawn around blocks of text
and around headers. Boxes were drawn
as close to the text as possible, but
boxes were always in a rectangle shape.
Any time there was a space greater than
the leading of the main body of the
text, a box was finished and new box
was started. The area of the page was
calculated (length x width). The area
of the boxes were calculated and
totaled. The percent white space = [1
- (total area of text/area of page)] x
100.

 

 

 

10) Leading

 

The leading of the main body of the
text was compared to known leading
sizes (Laubach and KoschnickL1977).

 

236

 

 

Characteristic Score This text
1) Type Size 3 12 point = 2 points
< 12 point = 0 points

3)

4)

7)

Font Type Roman serif = 2 points

 

All others = 0 points

Abbreviations
0% words abbreviated = 2 points

 

% words abbreviated = 1 point

5.
> 2% words = 0 points

Use of voice 0% passive 2 points

 

1 point
0 points

5 10% passive
> 10% passive

Emphasis on action 0% negative = 2 points

 

g 2% negative = 1 point
> 2% negative = 0 points

Letters Lower case (except
first letter of sentence) =

 

2 points

Capital letters (or any
other combination of capital
and lower case letters) = 0 points

White space: margins and headers
3 58% = 2 points

 

 

 

40-57% = 1 point
<40% = 0 points
Leading 3 2 points = 2 point
< 2 points = 0 points
Symbols 0% symbols used = 2 points
5 1% symbols = 1 point
> 1% symbols = 0 points

237

10) Headers 2 56% headers = 2 points
24-55% headers = 1 point
< 24% = 0 points

 

If the text scored:

17-20 The text contains many characteristics that aid
readability.
13-16 The text contains some characteristics of

readability, but might need to be reviewed to
further increase readability.

<12 The text contains few characteristics of
readability. It should be reviewed and improved
to increase readability.

APPENDIX I

Material Acceptability Survey (First Pilot)

238

Material Acceptability Survey (Pilot)

DIRECTIONS for instructors to read to the homemaker: I need
to ask you some questions about the pregnancy nutrition
materials we have just used. There are no right or wrong
answers. Your opinions will help us improve the EFNEP
nutrition materials used with pregnant women. Please select
only one response for each question (questions 1 to 14).

You might write your answers or ask me to write them for
you. You can choose £9; to answer any of these items if you
desire. You might look at the flipchart and handouts when

answering the questions.

How much do you agree with these statements:
1. The flipchart is easy for me to understand.
a. strongly agree b. agree c. disagree

d. strongly disagree e. I have no opinion
2. The way the flipchart is written is insulting to me.
a. strongly agree b. agree c. disagree

d. strongly disagree e. I have no opinion

If you answered yes, which part was insulting?

 

 

survey 2, page 1

3.

239

The flipchart uses words that are similar to words I

use .

strongly agree b. agree c. disagree

strongly disagree e. I have no opinion

The flipchart helped me learn about nutrition
information.
strongly agree b. agree c. disagree

strongly disagree e. I have no opinion

The flipchart is easy to read.
strongly agree b. agree c. disagree
strongly disagree e. I did not read the flipchart

I have no opinion
The flipchart is interesting.
strongly agree b. agree c. disagree

strongly disagree e. I have no opinion

The flipchart would be useful to pregnant homemakers in

EFNEP.
strongly agree b. agree c. disagree
strongly disagree e. I have no opinion

survey 2, page 2

240

 

 

 

8. I would be able to understand the flipchart without an
instructor helping me.
a. strongly agree b. agree 0. disagree
d. strongly disagree e. I have no opinion
9. The flipchart answered the questions I had about
nutrition during pregnancy.
a. strongly agree b. agree c. disagree
d. strongly disagree e. I didn’t have any questions
10. The number of pages in the flipchart made it:
a. too long b. too short c. about right
d. I have no opinion
11. The appearance of the flipchart (the pages with the
writing on them):
a. is too crowded b. has too much blank space
on the pages
c. is OK d. I have no opinion
The appearance could be improved by
12. The size of the print used in the flipchart is:

a. too large b. too small c. about right

d. I have no opinion

survey 2, page 3

241
For question 13, circle any or all responses you agree with:

13. The pictures in the flipchart:

a. are interesting b. helped me to understand the
information
c. are insulting d. could have been left out

In my opinion, the flipchart pictures are

 

 

 

 

14. How could the flipchart be improved?

 

 

 

 

For questions 15 to 20 circle any or all responses you agree

with:

15. Your instructor left four pregnancy handouts with you
to read. Circle the handouts that you read:
a. Iron b. Folic Acid c. Protein/Calcium
d. Discomforts during Pregnancy

e. I didn’t read any handouts (Skip questions 16 to 20).

survey 2, page 4

16.

l7.

l8.

19.

20.

242
Which handouts were easy to understand? (Circle any or
no responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts were easy to read? (Circle any or no
responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts helped you learn about nutrition
information? (Circle any or no responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts were interesting? (Circle any or no
responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts would be useful to pregnant women in

EFNEP? (Circle any or no responses).

a. Iron b. Folic Acid c. Protein/Calcium
d. Discomforts during Pregnancy

Note: Have you circled all of the responses that you

agree with in questions 15 to 20? Thank you for
completing this survey!

APPENDIX J

Materials Acceptability Survey (Final Instrument)

243

Material Acceptability Survey (Final Instrument)

DIRECTIONS for the instructor: Please administer this
survey to the homemaker after you have used the first set of
pregnancy materials and afte; she has completed the
Pregnancy Survey II. You should read the directions aloud
first. Then read aloud each question and all possible
answer choices. The homemaker (or yourself if she prefers)

should circle only one answer for each question.

IMPORTANT: Circle the set of materials you have just used
with the homemaker.
The EFNEP materials or The new client materials

(Eating Right... for Two) "Flipchart Y"
"Flipchart X"

244

Material Acceptability Survey (Final Instrument)

DIRECTIONS for instructors to read to the homemaker: I need
to ask you some questions about the pregnancy nutrition
materials we have just used. There are no right or wrong
answers. Your opinions will help us improve the EFNEP
nutrition materials used with pregnant women. Please select
only one response for each question (questions 1 to 14).

You might write your answers or ask me to write them for
you. You can choose £9; to answer any of these items if you
desire. You may look at the flipchart and handouts when

answering the questions.

How much do you agree with these statements:
1. The flipchart is easy for me to understand.
a. strongly agree b. agree c. disagree

d. strongly disagree e. I have no opinion
2. The way the flipchart is written is insulting to me.
a. strongly agree b. agree c. disagree

d. strongly disagree e. I have no opinion

If you answered yes, which part was insulting?

 

 

survey 2, page 1

3.

245

The flipchart uses words that are similar to words I

use .

a.

d.

strongly agree b. agree c. disagree

strongly disagree e. I have no opinion

The flipchart helped me learn about nutrition
information.
strongly agree b. agree c. disagree

strongly disagree e. I have no opinion

If you read the flipchart answer question 5, otherwise go to

que

stion 6.

The flipchart is easy to read.

strongly agree b. agree c. disagree

strongly disagree e. I did not read the flipchart

I have no opinion

The flipchart is interesting.
strongly agree b. agree c. disagree

strongly disagree e. I have no opinion

The flipchart would be useful to pregnant homemakers in
EFNEP.

strongly agree b. agree c. disagree

strongly disagree e. I have no opinion

survey 2, page 2

246
8. The flipchart answered the questions I had about
nutrition during pregnancy.
a. strongly agree b. agree c. disagree

d. strongly disagree e. I didn’t have any questions

9. The number of pages in the flipchart made it:
a. too long b. too short c. about right

d. I have no opinion

11. The appearance of the flipchart (the pages with the
writing on them):

a. is too crowded b. has too much blank space
on the pages

c. is OK d. I have no opinion

e. I did not see the pages with the writing on them.

The appearance of the flipchart could be improved by

 

 

 

12. The size of the print used in the flipchart is:
a. too large b. too small c. about right

d. I have no opinion

survey 2, page 3

247
For question 13, circle any or all responses you agree with:

13. The pictures in the flipchart:

a. are interesting b. helped me to understand the
information
c. are insulting d. could have been left out

e. held my attention

In my opinion, the flipchart pictures are

 

 

 

 

14. How could the flipchart be improved?

 

 

 

 

For questions 15 to 20 circle any or all responses you agree
with:
15. Your instructor left four pregnancy handouts with you
to read. Circle the handouts that you read:
a. Iron b. Folic Acid c. Protein/Calcium
d. Discomforts during Pregnancy
e. I didn’t read any handouts (Skip questions 16 to 20).

survey 2, page 4

l6.

l7.

l8.

19.

20.

248
Which handouts were easy to understand? (Circle any or
no responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts were easy to read? (Circle any or no
responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts helped you learn about nutrition
information? (Circle any or no responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts were interesting? (Circle any or no
responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

Which handouts would be useful to pregnant women in
EFNEP? (Circle any or no responses).
a. Iron b. Folic Acid c. Protein/Calcium

d. Discomforts during Pregnancy

APPENDIX K

Materials Comparison Survey (First Pilot)

249
Co arisen Surve Pilot
DIRECTIONS for the instructor to read to the homemaker: The
purpose of this survey is to ask you to compare the two sets
of materials. I will read all questions and answer choices.
Please select one answer choice for each question (questions
1 to 7). You may write your answers or ask me to write them
for you. There are no correct answers. These are opinion
questions. You may look at either or both sets of nutrition
materials (flipcharts and handouts) to help you answer
questions if needed. Questions 1 to 8 apply to the

flipcharts only.

1. In your opinion, which flipchart do you think is easier

to understand?

a. Flipchart X b. Flipchart Y c. About the same

2. Which flipchart uses words more similar to the words
you use?

a. Flipchart X b. Flipchart Y c. About the same

3. Which flipchart do you think contains the more useful
information?

a. Flipchart X b. Flipchart Y c. About the same

250
Which flipchart is easier to read?
Flipchart X b. Flipchart Y c. About the same

d. I didn't read them

Which flipchart do you think is more interesting?
Flipchart X b. Flipchart Y c. About the same

d. Neither are interesting

Which flipchart do you think would be more helpful to
women in EFNEP?

Flipchart X b. Flipchart Y c. About the same

Which flipchart do you think women in EFNEP would like
better?

Flipchart X b. Flipchart Y c. About the same

Do you see any other differences between "Flipchart X"

and "Flipchart Y"?

 

 

 

 

9.

251
Give each flipchart an overall score. The score can be
any number between 0 (terrible) and 100 (wonderful):
a. Which score would you give flipchart X?

b. Which score would you give flipchart Y?

Thank you for taking this survey!

APPENDIX L

Materials Comparison Survey (Final Instrument)

252
Com arison Surve Final Instrument
DIRECTIONS for the Instructor: Administer this Comparison
Survey after the homemaker has used pppp sets of materials.
Both sets of materials (flipcharts and handouts) should be
available for the homemaker to look at while answering the
survey questions. This survey should be the lap; survey
given to the homemaker. She should have completed the
Pregnancy Survey I and II, and the Material Acceptability
Survey before completing this survey. Please read the
directions to the homemaker first. Then read all questions
and answer choices to the homemaker. The homemaker (or you,
if the homemaker prefers) should circle only one answer for

each question.

IMPORTANT: Circle the set of materials you used first with
this client.
1. The original EFNEP materials (Flipchart X) or

2. The new Client-Adapted Language materials (Flipchart Y)

253

Co arison Surve Final Instrument

DIRECTIONS for the instructor to read to the client: The
purpose of this survey is to ask you to compare the two sets
of materials. I will read all questions and answer choices.
Please select one answer choice for each question (questions
1 to 7). You may write your answers or ask me to write them
for you. There are no correct answers. These are opinion
questions. You may look at either or both sets of nutrition
materials (flipcharts and handouts) to help you answer
questions if needed. Questions 1 to 8 apply to the

flipcharts only.

1. In your opinion, which flipchart do you think is easier

to understand?

a. Flipchart X b. Flipchart Y c. About the same

2. Which flipchart uses words more similar to the words
you use?

a. Flipchart X b. Flipchart Y c. About the same

3. Which flipchart do you think contains the more useful
information?

a. Flipchart X b. Flipchart Y c. About the same

254

If you read the flipchart answer question 4. If your

instructor read the flipchart to you skip question 4. Go to

question 5.

4. Which flipchart is easier to read?

a. Flipchart X b. Flipchart Y c. About the same

d. I didn’t read them
5. Which flipchart do you think is more interesting?
a. Flipchart X b. Flipchart Y c. About the same
d. Neither are interesting

6. Which flipchart do you think would be more helpful to
women in EFNEP?

a. Flipchart X b. Flipchart Y c. About the same

7. Which flipchart do you think women in EFNEP would like
better?

a. Flipchart X b. Flipchart Y c. About the same

8. Do you see any other differences between "Flipchart X"

and "Flipchart Y"?

 

 

 

 

255
Give each flipchart an overall score. The score can be
any number between 0 (terrible) and 100 (wonderful):
a. Which score would you give flipchart X?

b. Which score would you give flipchart Y?

Thank you for taking this survey!

APPENDIX M

Directions for Instructors

256
Instructions for SORT Testing
Instructions for the Paraprofessional:
Thank you so much for offering to help us with our
research to try to make better materials for your clients.

Please follow these steps with your clients:

1. Explain to your clients that we need their help. If we
are to improve the materials we need some idea of the
reading level of the clients.

2. Ask the client if she would be willing to read 10 lists
of words into a tape recorder. We do not need to know
her name. Her score will be kept confidential. Let
her know that some words are difficult and she
shouldn't feel bad if she doesn’t know them.

3. If she is willing to participate, have her state into
the tape recorder that she is willing to participate.
Next, have her read the words in order from list one to
ten. Please do not help her read the words. We need
to know the true reading level without help from
others. If she misses five in a row and does not want
to continue, she may stop.

4. To avoid confusion, only test people who speak English

as their first language.

Thank you for your time and trouble,

Karen Schmitz

257

Instructions for Data Collection:

Thank you for your help with this project to improve

EFNEP materials. Please follow the steps below with each

newly-enrolled, English-speaking client:

Da

1.

Da

1.

1 about 1 hour

Read the letter of consent to the client. Ask her to
participate.

If she agrees, give her the pretest. The directions
are on the test.

Teach her the nutrition lesson using the appropriate
flipchart, either X or Y.

Give her the posttest. The directions are on the test.
Leave the 4 handouts with her (Iron, Folic Acid,
Protein/Calcium and Discomforts during Pregnancy) and

ask her to read them before your next meeting (Day 2).

2 about 1 hour

Give her the "Acceptability Survey." Make sure she has
her handouts and the flipchart she used last time in
front of her.

Read through the other flipchart (either X or Y) with
the client (the flipchart she did pp; use on Day 1).
Give her the "Comparison Survey." Make sure she has
ppph flipcharts in front of her.

Put all 4 surveys in the envelope. Put your initials
on the back of the envelope and turn it in to the

County Extension Home Economist.

APPENDIX N

Consent Forms

258
Please read to the homemaker:
Dear homemaker,

We hope that you can help us out with making new
materials for EFNEP. In order to improve the materials we
need to have an idea of the reading levels of the .
homemakers. Your participation is voluntary and we do not
need to know your name. If you are willing to help, we ask
that you read ten lists of twenty words into a tape
recorder. The process should take 20 to 30 minutes. Some
of the words are difficult, so don’t feel badly if you can't
make it to list ten. You may quit when you don’t know 5
words in a row.

The instructor will also ask you if it is O.K. to give
us information on your race, education and age. This
information is optional. You may say no if you don't want

to answer any or all of the questions.

Thank you for your time and trouble,

Karen Schmitz

259

Please read to the homemaker:

Dear homemaker,

We hope that you can help us out with making new
materials for EFNEP. In order to improve the materials we
need to know how well different materials help you learn
about nutrition and your opinion of the different materials.
Your participation is voluntary and we do not need to know
your name. Your answers to questions will not affect your
participation in EFNEP. If you are willing to help us, you
will use two different sets of nutrition materials in your
next two EFNEP lessons. You will be asked to answer several
questions, by circling a response, before and after each
lesson. The process should take about two one-hour
sessions.

Your instructor will also ask you if it is o.k. to give
us information on your age, race, education, and pregnancy
status. This information is optional, you may say no if you

don’t want to answer any or all of the questions.

Thank you for your time and trouble,

Karen Schmitz

APPENDIX O

EFNEP Handouts

260

§Pfi©m WEEK}? NEEDQ

DURING?) PRECGNANCQ?
FOLIO acm

Folic acid (which is also ealled folate and folacin) is a B vitamin. It is used to form red
blood cells and is used for growth of new cells. Your need for folate doubles during

pregnancy-

Foods High In Folic Acid
Bean sprouts Chickpeas (garbonzo beans) Wheat germ
Liver Fortified cereals . Greens
Asparagus Kidney beans ‘ Winter squash
Broccoli Lima beans Navy beans
Spinach ' Beets Pinto beans
Cauliflower Cantaloupe Black-eyed peas
Tofu ‘ Whole grains

NOTE: Cooking (heat) destroys some of the folate: raw vegetables provide more folic acid
than cooked ones.

References: Basic Nutrition Facts. MSU Cooperative Extension Service and Michigan
Department of Public Health.

Understanding Nutrition: Whitney. Hamilton. Rolfes. West Publishing
1990.

 

261

§P3©IUAE MERIT“? NEEED§ DOME?)

PREONANCY
IRON

Iron in red blood cells carries oxygen throughout the body. Some of your
iron supply will be “donated" to the baby and some iron will be lost at birth
(with blood loss). 60 it is important to eat foods high in iron every day to
meet your increased need for iron.

Liver ' Prunes / j uice

Beef Kale. chard

Pork Black-eyed peas

Dried beans (navy, soy. kidney) Whole wheat bread

Broccoli Fish, seafood, shelfish

Dina beans Enriched cereals V
(hot and cold)

Split peas . Raisins

Green peas Molasses (blackstrap)

Enriched bread, rice & macaroni Spinach

NOTE: Vitamin C increases iron absorption. so try to eat fruit. fruit juice or vegetables
along with iron-rich foods.

 

 

262

’ @PEC‘EW NUTRIENT NEEDS
@UE’EIIRT® EDI—RECONMCOY * '
CALCIUM
Extra calcium is needed for your baby's developing bones and teeth.

Try to have three servings from the milk/ cheese group daily. This group includes
foods like:

buttermilk evaporated milk
nonfat milk yogurt

lowfat ' cheese

skim milk ice cream

 

Use the One Serving Is: Milk and Cheese handout to
compare amounts of fat and calories in calcium-rich foods.

 

Some people are lactose intolerant and have "trouble" digesting milk/ products. For
these people to Obtain enough calcium they can drink special milk (Lact—aid). eat
dairy foods with low levels of lactose (yogurt. cheddar cheese, swiss cheese). or eat
non-dairy foods with calcium such as:

dried beans tofu
broccoli spinach
canned sardines. mackerel greens

or salmon with bones oysters
shrimp

‘\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\‘\\\\\\\\\\\
\\\\\\\\\\\\\\\\\\\\\ \\\\\\ ‘\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

PROTEIN

PROTEIN is needed for growth and other functions.

Choose two servings of meat. poultry, fish or eggs or dried beans and
peas every day. The three servings of milk or cheese that is recommended during
pregnancy will also help to meet your protein needs.

263

EDI®©©M§©RE§ DURIN@ PRECQNMCOY?

You may have heard about these discomforts of pregnancy - "morning sickness.“
heartburn. and constipation.

It's true that one or more of these might affect you at some time during your pregnancy.
They occur because of normal changes during pregnancy. However. you may iind that
you aren’t affected by any of these discomfortsl

Here are a few suggestions to prevent or decrease these problems.

MORNING 8mm

0 Eat crackers. dry toast or cereal 0 Drink liquids between meals rather than
in the morning before getting out with them.
of bed.

0 Haveasnackbeforebed - cheese or Cl Sleep inawellventilated room.
a hard cooked egg are good choices. 0 Avoid foods that do not ”agree" with you

0 Eat small. frequent meals. such as fried or spicy foods.

Morning sickness usually passes by the fourth month of pregnancy. Some women
continue to have morning sickness throughout their entire pregnancy and others are not
nauseous or sick at all. If morning sickness is affecting how much you are eating (or
‘keeping down') be sure to tell your doctor.

HEARTBURN

0 Eat small frequent meals. 0 Don't lie down right after eating.
0 Avoid foods that give you problems - 0 Wear loose—fitting clothing.
maybe spicy or fatty foods. 0 Eat slowly. chew food well.
Ci Avoid large meals. 0 Do not drink liquids with your meals
OONSTIPATION
0 Get some ocercise - maybe a daily 0 Include plenty of fiber in your diet
walk by eating fruits. vegetables. and whole
0 Drink plenty Of liquids - especially grains

water. milk. and juices.

DO NOT take any drugs or medications without
first asking your doctor about them. This
includes laxatives. antacids and aspirin.

APPENDIX P

Pica Handout (Pilot and Final Forms)

 

 

264
Pica (Pilot form)
Many pregnant women have cravings. Some cravings for food

won’t harm the baby. These may include:

Chocolate Ice cream
Pizza Pickles
Tacos Watermelon

What foods do you crave?

 

 

 

 

Some pregnant women crave stuff that could harm the baby.
Avoid doing these things:
Eating baking soda, toilet paper, toothpaste,
soap, cornstarch
OR
Smelling gasoline
OR
Biting matches
OR

Sipping pine sol

Do you do any of these things? Do you do other things that
you think might harm the baby? If you think the answer
might be yes, stop the behavior and talk to your doctor.

Your baby’s health depends on you!

 

 

265
Pica

Many pregnant women have cravings. Some women crave chocolate, ice

cream, pizza, pickles or watermelon.

Other pregnant women crave things that aren’t really foods. Some women

report eating:

baking soda toilet paper facial tissue
toothpaste soap ashes
cornstarch clay laundry starch

OR smelling gasoline or pine sol

OR biting matches

Have you had a craving to eat or smell any of these things or other things

that aren’t really foods? List some here:

 

 

 

 

Eating or smelling things that are not real foods could harm you and your
baby. Some things like cornstarch and laundry starch can make you gain
extra weight. Other things like gasoline and pine sol are poisons.

Please tell your doctor or nutritionist about any items you have listed.

 

APPENDIX Q

Comments of EFNEP Clients

266

Materials Acceptability Survey Comments
Note: Comments have been transcribed word for word from
surveys. No corrections in grammar, spelling or accuracy of
information were made.

Question #2: If you answered yes (to the flipchart being
insulting), which part was insulting?"

Client identification Responses:
number:

Flipchart X:

006 "Made me feel dumb"

074 "The writing is not
professional."

079 "Pictures"

Flipchart Y:

039 "24M - the little thing"

045 "The way they used short
grammar in a few sentences"

072 "It's written for people who
have hardly any education."

097 "Stuff and thing"

101 "Don’t like the word stuff."

105 "Too hard"

114 "It’s not for and adult."

Question #10: The appearance of the flipchart could be
improved by:

Flipchart X:
010 "Pictures outdated"
019 "Being updated"

036 "Updating your pictures"

 

O38

O44
O66
095

114

115

149

Flipchart Y:
001

007
info"

039
105

116

183

267

"Updated pictures and
information"

"Like it how it is"
"Get real pictures"
"Pictures more realistic"

"Make it look like its for a
adult"

"Pictures are outdated."
"Update pictures. Page 35M

prescription bottle is dated
1985."

"Better pictures"

"More updated pictures and

"Improving the pictures"
"More color"

"Pictures could be improved.
They look generic."

"Pictures could be updated and

realistic."

Question #12: In my opinion, the flipchart pictures are:

Flipchart X or Y:

001

005

006

007

010

"Outdated"
"Simple"

"Help me pay attention to
instructions"

"Type expected in a health
type class"

"Outdated"

 

268

013 "Easy to look at?

014 "Very basic"

019 "Good, okay"

020 "The pictures are ok"

021 "Ugly"

024 "Insulting"

026 "Very useful"

029 "Alright, it is better than
just a book with no pictures."

032 "Fine"

036 "Needed for people who have
problems comprehending what is
read"

038 "Out of date"

039 "Insulting. They look like

they should be used for a
little Child."

042 "0k"

045 "They didn’t catch my
attention."

046 "Hard to see because they are
upside down"

066 "OK for younger moms, but not
experienced moms"

069 "Childish"
072 "Needing to be updated"
074 "Sorta ugly, could be

realistic or real pictures??"

075 "Need to be updated in colors
and figures. It looks like
60’s material."

079 "Boring"

080

095

111

113

116

122

146

149

158

159

160

162

168

169

182

183

184

Question #13:

Flipchart X:

006

010

019

269

"Simple, really"

"More realistic"
"Helpful"

"Too much like comics"
"A little outdated"
"Like the pictures O.K."
"Old crud"

"They're eye catching,
colorful"

"Not real enough. Using real
pictures of people and food
items makes it more inviting
for people to listen and learn
more."

"As good as can be expected
for the space."

"Very educational and
informative"

"Not real"
"Self-explanatory"
"O.K."

"Colorful"

"Dated"

"It seemed childish because of
the cartoons."

How could the flipchart be improved?

"I hate being read to"
"More realistic pictures"

"By being updated"

270

020 "It could be more interesting"

021 "Better pictures"

024 "Select better pictures. Make
them more normal looking."

036 "Just update the pictures"

038 "Updated!"

039 "Its better if its not read
word for word."

044 "Like it how it is"

046 "Turn the pictures around"

051 "Last week was nicer. Why
change?"

066 "Use real pictures."

079 "I wouldn’t have a flipchart
unless used with illiterate
people."

087 "I don’t think it needs

' improvement."

095 "Make it more interesting"

108 "Made more easy"

110 "Not so tough"

111 "More detail"

112 "Charts"

115 "Update pictures"

144 "It good as it is"

146 "Be a little more realistic

looking like real people
rather than cartoonist"

149 "Make it shorter, smaller.
New pictures"

150 "The wording"

271

154 "I think it is alright the way
you have it"

158 "By using real pictures of
people and food and changing
page 37 from may to will. I
do not like any written
material about pregnancy and
drugs and alcohol stating that
they MAY only harm the baby.
Because they DO harm babies."

168 "It fine by my opinion"

Flipchart Y:

001 "Update pictures"

005 "Okay"

007 "Statistics would be more
int."

008 "Looks fine"

013 "It’s fine"

039 "Update the pictures."

043 "Make the pictures more
adult."

045 "Better to catch peoples
attention"

047 "Explain good, well-worded"

069 "Realistic pictures"

070 "It is just OK. Clear, good,
nice picture or diagram"

072 "Update pictures"

105 "More pictures"

116 "Improve pictures"

152 "Wanted a little more

information about weight gain
in pregnancy"

272

160 "No improvement needed."

162 "Real pictures"

167 "No cartoons for pictures"
169 "Real pictures"

182 "Give exact information. How

many servings is needed for
veggies isn’t mentioned and
why is weight gain in
pregnancy important."

183 "Overall O.K., want real
pictures of babies progress."

273
Materials Comparison Survey Comments

Note: Comments have been transcribed word for word from
surveys. No corrections in spelling, grammar or accuracy of
information were made.

Question #8: "Do you see any other differences between
flipchart X and flipchart Y?"

Client identification Response:
number:
001 "X was easier to understand.

The words made sense."

006 "Did not like Y wording as
much."

009 "Y has more information than X
does."

012 "They’re both the same to me."

013 "Do not like stuff"

014 "Flipchart X seems to be

geared more for an adult.
Flipchart Y seems to be geared
for teenagers."

016 "Y is better because I think
that more young mothers will
understand it better."

021 "Wording--don't make us feel
stupid by using word stuff."

023 "Y is easier to understand."

024 "Word stuff is dumb."

029 "Wording was different. Chart
Y would be easier for teens
and some ethnic groups. I

believe chart Y would be the
best if some of the wording
were changed (such as stuff).
It helped that they list foods
in the chart."

274

032 "Just a little different
words."
036 "Yes, X is for a higher

educated group. Y is for your
younger (teenage) group.

039 "Yes, X is for a higher level
and Y is for more common. Y
is real good if they change
words like stuff and little
thing and the pictures are
REAL BAD! Pictures look like
they came from the seventies"

045 "X used better grammar."

046 "Stuff word needs to go."

047 "Flipchart X words things
better than Y."

050 "Word stuff"

066 "X is for the more normal

homemaker, Y is for younger,
less experienced girl"

067 "X is taught on a more
mature/adult level and Y is
for the younger teenager."

070 "They define what you need to
know and you understand the
first after you read the
flipchart easier"

073 "Y is not as professional."
077 "X seemed shorter."
089 "Sometimes hard to understand

or remember. I’m not sure how
well I done."

095 "The words in flipchart Y are
not professional."

096 "Y is too long. Some of the
details are not necessary."

275

099 "The wording in_flipchart Y is
insulting."

110 "Flipchart X is not so tough."

112 "Flipchart Y has more
information."

114 "Flipchart X is less reading."

115 "Y was more personable, but

you could leave out the
'stuff’ and ’little thing.'"

116 "Y assumes you already know
some things."

119 "The Y version may appeal more
to the young."

121 "Wording is better in X.
Sounds better"

153 "Flipchart X talks about more
during pregnancy and flipchart
Y is more about nutrition."

154 "Flipchart Y helps you more to
understand the importance of
weight gain"

156 "Y used slang and stuff.
Talked down to the
individual."

158 "The use of words like stuff
and thing I don’t care for.
Page 37 talks about drugs and
alcohol MAY cause problems —
that is wrong."

162 r "Y made me feel talked down
to. X used proper english"

163 "Language - Y is childish -
poor use of words ’thing' and
’stuff' - they are not

informative words."

164 "Slang in Y"

276

182 "X answered questions exactly
and how many servings was
needed and also importance of
gaining weight."

183 "X felt better and more like
what homemaker feels
comfortable with"

184 "Y = eliminate thing and
stuff. Y = better for
explaining smoking. Otherwise
they are the same."

 

LI ST OF REFERENCES

 

LIST OF REFERENCES

Abrams, B. and R. Laros. Prepregnancy weight, weight gain
and birth weight. American Journal of Obstetrics and
Gynecology 154(3):503-509, 1986.

Abrams, B. and V. Newman. Small-for-gestational-age birth:
maternal predictors and comparison with risk factors of
spontaneous preterm delivery in the same cohort. American
Journal of Obstetrics and Gynecology 164:785-790, 1991.

Abrams, B., V. Newman, T. Key and J. Parker. Maternal
weight gain and preterm delivery. Obstetrics and Gynecology
74:577-583, 1989.

Acker, D., B. Sachs, K. Tracey and W. Wise. Abruptio
placentae associated with cocaine use. American Journal of
Obstetrics and Gynecology 146:220-221, 1983.

Aderman, B., S. Nitzke, S. Pingree and J. Voichick.
Readers’ responses to language experience approach
materials. Adult Literacy and Basic Education 2(1):13-22,
1987.

Allen, B., J. Wright and L. Laminack. Using Language
experience to alert pupils’ critical thinking skills. The
Reading Teacher 41(9):904-910, 1988.

American Dietetic Association. Recommendations concerning
supplement usage: ADA statement. Journal of the American
Dietetic Association 87:1342-1343, 1987.

Ashe, J., F. Schofield and M. Gram. The retention of
calcium, iron, phosphorous and magnesium during pregnancy:
the adequacy of prenatal diets with and without
supplementation. American Journal of Clinical Nutrition
32:286-291, 1979.

Bailey, L. Folate status assessment. Journal of Nutrition
120:1508-1511, 1990.

Bailey, L., C. Mahan and D. Dimperio. Folacin and iron

status in low-income pregnant women. The American Journal
of Clinical Nutrition 33:1997-2001, 1980.

277

 

278

Beal, V. Nutritional studies during pregnancy. Journal of
the American Dietetic Association 58:321-326, 1971.

Belizan, J., J. Villar and J. Repke. The relationship
between calcium intake and pregnancy-induced hypertension.
American Journal of Obstetrics and Gynecology 158:898-902,
1988.

Blackburn, M. and D. Calloway. Energy expenditure and
consumption of mature, pregnant and lactating women.
Journal of the American Dietetic Association 69:29-36, 1976.

Block, G., C. Cox, J. Madans, G. Schreiber, L. Licitra and
N. Melia. Vitamin supplement use, by demographic
characteristics. American Journal of Epidemiology
127(2):297-309, 1988.

Blocker, D., L. Ausman, C. Meadows and S. Thenan.
Experimental maternal and neonatal folate status
relationships in non-human primates. American Journal of
Clinical Nutrition 50:120-128, 1989.

Bower, C. and F. Stanley. Dietary folate as a risk factor
for neural—tube defects: evidence from a case-control study
in western Australia. Medical Journal of Australia 150:613-
617, 1989.

Brennan, R., M. Kohrs, J. Nordstrom, J. Sauvage and R.
Shank. Nutrient intake of low-income pregnant women:
laboratory analysis of foods consumed. Journal of the
American Dietetic Association 83:546-550, 1983.

Brown, J., K. Berdan, P. Splett, M. Robinson and L. Harris.
Prenatal weight gains related to the birth of healthy-sized
infants to low-income women. Journal of the American
Dietetics Association 86:1679-1683, 1986.

Brush, K., D. Woolcott and G. Kawash. Evaluation of an
affective-based adult nutrition education program. Journal
of Nutrition Education 18(6):258-264, 1986.

Byrd—Bredbenner, C., R. Martin, C. Lewis and B. Shannon.
The effectiveness of two methods of delivering nutrition
information to the general public. Journal of Nutrition
Education 20(2):63-69, 1988.

Campbell, D. and J. Stanley. Experimental and Quasi-
Experimental Designs for Research. American Education
Research Association, Chicago, IL, 1963.

279

CEBU Study Team. Underlying and proximate determinants of
child health: The CEBU longitudinal health and nutrition
study. American Journal of Epidemiology 133:185-201, 1991.

Chanarin, D. Further observations on the relation between
iron and folate status in pregnancy. British Medical
Journal 2:81—84, 1971.

Chasnoff, I. Cocaine use in pregnancy. New England Journal
of Medicine 313:616, 1985.

Council on Scientific Affairs. Fetal effects of maternal
alcohol use. Journal of the American Medical Association
249:2517-2521, 1983.

Dale, E. A comparison of two word lists. Educational
Research Bulletin 10:484-489, 1931.

Dale E. and J. Chall. A formula for predicting readability.
Educational Research Bulletin 27:11-20, 1948.

Dawson, E. and W. McGanity. Protection of maternal iron
stores in pregnancy. The Journal of Reproductive Medicine
32(6)(Supplement):478-487, 1987.

Doak, L. and C. Doak. Patient comprehension profiles:
recent findings and strategies. Patient Counseling and
Health Education 2(3):101-106, 1980.

Doak, C., L. Doak, and J. Root. Teaching Patients with Low
Literacy Skills. J.B. Lippincott Co., Philadelphia, PA.,
1985.

Doyle, W., M. Crawford, B. Laurance and P. Drury. Dietary
survey during pregnancy in a low socioeconomic group. Human
Nutrition Applied Nutrition 36A:95-106, 1982.

Dwyer, D., E. Krall and K. Coleman. The problem of memory
in nutritional epidemiology research. Journal of the
American Dietetic Association 87(11):1509-1512, 1987.

Endres, J., M. Sawicki and J. Casper. Dietary assessment of
pregnant women in a supplemental food program. Journal of
the American Dietetic Association 70:121-126, 1981.

Farkas, C, P. Glenday, P. O'Connor and J. Schmeltzer. An
evaluation of the readability of prenatal health education
materials. Canadian Journal of Public Health 78:374-378,
1987.

 

280

Fitz-Gibbon, C. and Morris, L. How to Design a Program
Evaluation. Sage Publications, Inc. Newbury Park, Ca,
1987.

Flesch, R. A new readability yardstick. Journal of Applied
Psychology 32(3):221—233, 1948.

Freimuth, V. Assessing the readability of health education
messages. Public Health Reports 94(6):568-570, 1979.

Fry, E. Judging readability of books. Teacher Education
5:34-39, 1964.

Fry, E. A readability formula that saves time. Journal of
Reading 513-516, 575-579, April 1968.

Fry, E. Writeability: The principles of writing for
increased comprehension. lg Readability: Its Past, Present
and Future. International Reading Association, Newark, DE.,
1988.

George, J. Language experience approach. Lifelong
Learning: An Omnibus of Practice and Research 7(1):29-30,
1983.

Gormican, A., J. Valentine and E. Satter. Relationships of
maternal weight gain, prepregnancy weight and infant birth
weight. Journal of the American Dietetics Association
77:662-667, 1980.

Gray, W. Standardized Oral Reading Paragraphs. Bobbs-
Merril Company, Indianapolis, IN, 1960.

Gunning, Robert. The Technique of Clear Writing. New York,
New York, 1968.

Harman, D. Illiteracy: A National Dilemma, Cambridge Book,
Co., New York, 1987.

Harman, S. and C. Edelsky. The risk of whole language
literacy: alienation and connection. Language Arts
66(4):392—406, 1989. '

Haste, F., O. Brook, H. Anderson, J. Bland, A. Shaw, J.
Griffin and J. Peacock. Nutrient intakes during pregnancy:
observations on the influence of smoking and social class.
American Journal of Clinical Nutrition 51:29-36, 1990.

Haworth, J., J. Ellestad-Sayed, J. King and L. Dilling.
Fetal growth retardation in cigarette-smoking mothers is not
due to decreased maternal food intake. American Journal of
Obstetrics and Gynecology 137:719-723, 1980.

 

281

 

Heller, L., R. Salkeld and W. Korner. Riboflavin staus in
pregnancy. American Journal of Clinical Nutrition 27:1225-
1230, 1974.

Heller, M. Comprehending and composing through language
experience. The Reading Teacher 42(2):130-135, 1988.

Henerson, M.E., L.L. Morris and C.T. Fitz-Gibbon. How to
Measure Attitudes. Center for the Study of Evaluation,
University of California, Los Angeles, 1987.

Hibbard, B. Iron and folate supplements during pregnancy:
supplementation is valuable only in selected patients.
British Medical Journal 297:1324-1325, 1988.

Horn, E. Iron and folate supplements during pregnancy:
supplementing everyone treats those at risk and is cost-
effective. British Medical Journal 297:1325-1326, 1988.

Horner, R., C. Lackey, K. Kolasa and K. Warren. Pica
practices of pregnant women. Journal of the American
Dietetic Association 91:34-38, 1991.

Hunter, C. and D. Harman. Adult Illiteracy in the United
States, A Report to the Ford Foundation. McGraw Hill Book
Co., New York, NY., 1979.

Hytten, F. and I. Leitch. The Physiology of Human
Pregnancy, 2nd edition. Blackwell Scientific Publications,
Oxford, England, 1971.

Idstein, P. and J. Jenkins. Underlining versus repetitive
reading. The Journal of Educational Research 65(7):321-323,
1972.

Institute of Medicine, subcommittee on Nutritional Status
and Weight Gain During Pregnancy. -Nutrition During
Pregnancy. National Academy of Sciences. National Academy
Press, Washington, D.C., 1990.

Irwin, J. and C. Davis. Assessing readability: the
checklist approach. Journal of Reading 124-129, 1980.

Jacobson, M., C. Kirkland and R. Selden. An examination of
the McCall-Crabbs Standard Test Lessons in Reading. Journal
of Reading 224-230, 1978.

Johnson, L. Effects of underlining textbook sentences on
passage and sentence retention. Reading Research and
Instruction 28(1):18—32, 1988.

282

Jones, K. Pattern of malformation in offspring of chronic
alcoholic mothers. Lancet 1:1267, 1973.

Kafatos, A., I. Vlachonikolis and C. Codrington. Nutrition
during pregnancy: the effects of an educational
intervention program in Greece. American Journal of
Clinical Nutrition 50:970—979, 1989.

Kazemek, F. An examination of the Adult Performance Level
Project and its effects upon adult literacy education in the
U.S. Lifelong Learning 9(2):24-28, 1985.

Kazemek, F. Adult literacy education: an ethical endeavor.
Adult Literacy and Basic Education 8(2):1984.

Kennedy, K. and S. Roeder. Using Language Experience with
Adults: A Guide for Teachers. New Readers Press, Syracuse,
New York, 1975.

Kirsch, I. and J. Guthrie. The concept and measurement of
functional literacy. Reading Research Quarterly 13(4):486-
507, 1977.

Klare, G. Assessing readability. Reading Research
Quarterly 62-102, 1974—1975.

Klare, G. A second look at the validity of readability
formulas. Journal of Reading Behavior 8:129-152, 1976.

Klare, G. The formative years, 1p Readability: Its Past,
Present and Future, International Reading Association,
Newark, Delaware, 1988.

Klein, L. Premature birth and maternal prenatal anemia.
American Journal of Obstetrics and Gynecology 83:588-590,
1962.

Kleinman, J. and S. Kessel. Racial differences in low birth
weight: trends and risk factors. The New England Journal
of Medicine 317:749-753, 1987.

Lanese, R. and R, Thrush. Measuring readability of health
education literature. Journal of the American Dietetics
Association 42:214-217, 1963.

Langford, E., E. Thompson and S. Tripp. Smoking and health
education during pregnancy: evaluation of a program for
women in prenatal classes. Canadian Journal of Public
Health 74:285-289, 1983.

 

 

283

Laubach, R. and K. Koschnick. Using Readability: Formulas
for Easy Adult Materials. New Readers Press, Syracuse, New
York, 1977.

Lechtig, A., J. Habicht, H. Delgado, R. Klein, C. Yarbrough
and R. Martorell. Effect of food supplementation during
pregnancy on birth weight. Pediatrics 56(4):508-520, 1975.

Lieberman, E., K. Ryan, R. Monson and S. Schoenbaum. Risk
factors accounting for racial differences in the rate of
premature birth. The New England Journal of Medicine
317:743-748, 1987.

Lorge, I. Predicting reading difficulty of selections for
children. The Elementary English Review 16:229-233, 1939.

Lorge, I. The Lorge and Flesch readability formulae: a
correction. School and Society 67(1730):141-143, 1948.

Mayer, R. Models for Understanding. Review of Educational
Research 59(1):43-64, 1989.

McCall, W. and L. Crabbs. Standard Test Lessons in Reading.
Bureau-of Publications, Teachers College, Columbia
University, New York, 1925; 1950 edition, 1950; 1961
edition, 1961.

McCormick, M. The contribution of low birth weight to
infant mortality and childhood morbidity. The New England
Journal of Medicine 312:82-90, 1985.

McLaughlin, G. SMOG grading, a new readability formula.
Journal of Reading 639-646, May 1969.

Meade, C. and J. Byrd. Patient literacy and the readability
of smoking education literature. American Journal of Public
Health 79(2):204-206, 1989.

Meyer, M. How does maternal smoking affect birth weight and
maternal weight gain? Evidence from the Ontarion Perinatal
Mortality Study. American Journal of Obstetrics and
Gynecology 131:888-893, 1978.

Meyer, M., B. Jonas and J. Tonascia. Perinatal events
associated with maternal smoking during pregnancy. American
Journal of Epidemiology 103(5):464-476, 1976.

Michigan Expanded Food and Nutrition Education Program.
Michigan State University, Cooperative Extension Service,
1990.

 

284

Milunsky, A., J. Hershel, S. Jick, C. Bruell, D. McLaughlin,
K. Rothman and W. Willett. Multivitamin/folic acid
supplementation in early pregnancy reduces the prevalence of
neural tube defects. Journal of the American Medical
Society 262:2847-2852, 1989.

Mitchell, M. and E. Lerner. Weight gain and pregnancy
outcome in underweight and normal weight women. Journal of
the American Dietetic Association 89:634-638, 641, 1989.

Mohammed, M. Patients’ understanding of written health
information. Nursing Research 13(2):100-108, 1964.

Mora, J., B. deparedes, M. Wagner, L. denavarro, J. Suescun,
N. Christiansen and M. Herrera. Nutritional supplementation
and the outcome of pregnancy 1. Birth weight. American
Journal of Clinical Nutrition 32:455-462, 1979.

Morris, L.L., C.T. Fitz-Gibbon and E. Lindheim. How to
Measure Performance and Use Tests. Center for the Study of
Evaluation, University of California, Los Angeles, 1987.

Moscovitch, L. and B. Cooper. Folate content of diets in
pregnancy: comparison of diets collected at home and diets
prepared form dietary records. American Journal of Clinical
Nutrition 26:707-714, 1973.

Mulcahy, R., J. Murphy and F. Martin. Placental changes and
maternal weight in smoking and nonsmoking mothers. American
Journal of Obstetrics and Gynecology 106:703-704, 1970.

Naeye, R. Nutritional/nonnutritional interactions that
affect the outcome of pregnancy. American Journal of
Clinical Nutrition 34:727-731, 1981.

Naeye, R. Abruptio placentae and placenta previa:
frequency, perinatal mortality, and cigarette smoking.
Obstetrics and Gynecology 55(6):701-704, 1980.

Naeye, R. Causes of fetal and neonatal mortality by race in
a selected U.S. population. American Journal of Public
Health 69(9):857-861, 1979.

National Center for Health Statistics. 1990 Health U.S. 1989
DHHS Publication no. (PHS) 90-1232 NCHS PHS USDHHS,
Hyattsville, MD 291pp.

285

NRC, National Research Council. Recommended Dietary
Allowances, 10th ed. Report of the Subcommittee on the Tenth
Edition of the RDA’s, Food and Nutrition Board, Commission
of Life Sciences. National Academy Press, Washington D.C.
1989.

Neggers, Y., G. Cutter, R. Acton, J. Alvarez, J. Bonner, R.
Goldenberg, R. Go and J. Roseman. A positive association
between maternal serum zinc concentration and birth weight.
American Journal of Clinical Nutrition 51:678-684, 1990.

Nitzke, S. Developing materials for low-income, low-
literacy audiences. Journal of Nutrition Education 18:226B-
C, 1986.

Nitzke, S. Reaching low-literate adults with printed
nutrition materials. Journal of the American Dietetic
Association 87(suppl):73-77, 1987.

Nitzke, S. Improving the effectiveness of nutrition
education materials for low literacy clients. Nutrition
Today 17—23, September/October 1989.

O'Sullivan, J., S. Gellis, B. Tenney and C. Mahan. Aspects
of birth weight and its influencing variables. American
Journal of Obstetrics and Gynecology 92:1023-1029, 1965.

Philipps, C. and N. Johnson. The impact of quality of diet
and other factors on birth weight of infants. American
Journal of Clinical Nutrition 30:215-225, 1977.

Pichert, J. and P. Elam. Readability formulas may mislead
you. Patient Education and Counseling 7:181-191, 1985.

Pickering, C. Whole language: a new signal for expanding
literacy. Reading Improvement 26(2):144-149, 1989.

Picone, T., L. Allen, M. Schramm and P. Olsen. Pregnancy
outcome in North American women 1. Effects of diet,
cigarette smoking and physiological stress on maternal
weight gain. American Journal of Clinical Nutrition
36:1205-1213, 1982.

Que Software, Inc., Rightwriter Software Program, Version
5.0, Prentice Hall Computer Publishing, 1992.

Raper, N., J. Rosenthal and C. Woteki. Estimates of
available iron in diets of individuals one year old and
older in the Nationwide Food Consumption Survey. Journal of
the American Dietetic Association 84:783-787, 1984.

286

Recht, D. and L. Leslie. Effect of prior knowledge on good
and poor reader’ memory of text. Journal of Educational
Psychology 80(1):16-20, 1988.

Reference Software, Inc., Grammatik III Software Program,
Wang laboratories, San Francisco, Ca, 1989.

Rickards, J. and G. August. Generative underlining
strategies in prose recall. Journal of Educational
Psychology 67(6):860-865, 1975.

Rogozinski, H., C. Ankers, D. Lennon, J. Wild, C. Schorah,
S. Sheppards and R. Smithells. Folate nutrition in early
pregnancy. Human Nutrition Applied Nutrition 37A:357-364,
1883.

Rudd, J., N. Betts and J. Dirkx. Developing written
information for adults with low literacy skills. Journal of
Nutrition Education 25:11-16, 1993.

Samora, J., L. Saunders and R. Larson. Medical vocabulary
knowledge among hospital patients. Journal of Health and
Human Behavior 2(2):83-92, 1961.

Schuster, K., L. Bailey and C. Mahan. Vitamin B—6 staus of
low-income adolescent and adult pregnant women and the
condition of their infants at birth. American Journal of
Clinical Nutrition 34:1731-1735, 1981.

Scientific Research Association Achievement Series, ed.
Naslund, R., L. Thorpe and D. Lefevre, Science Research
Associates, Inc., Chicago, IL, 1971.

Shah, R. and R. Rajalakshmi. Vitamin A status of the
newborn in relation to gestational age, body weight, and
maternal nutritional status. American Journal of Clinical
Nutrition 40:794-800, 1984.

Shannon, P. The struggle for control of literacy lessons.
Language Arts 66(6):625-634, 1989.

Shepherd, 8., L. Sims, C. Davis, A. Shaw and F. Cronin.
Panel versus novice focus groups: reactions to content and
design features of print materials. Journal of Nutrition
Education 26:10-14, 1994.

Slosson, R. Slosson Oral Reading Test. Slosson Educational
Publications, New York, 1963.

Smith, S. and B. Alford. Literate and semi-literate
audiences: tips for effective teaching. Journal of
Nutrition Education 20:238B-238C, 1988.

287

Smithells, R., C. Ankers, M. Carver, D. Lennon, C. Schorah
and S. Sheppard. Maternal nutrition in early pregnancy.
British Journal of Nutrition 38:497-506, 1977.

Snowman, M. Nutrition component in a comprehensive child
development program. Journal of the American Dietetic
Association 74:124-129, 1979.

Spache, G. A new readability formula for primary-grade
reading materials. Elementary School Journal 53:410-413,
1953.

Stahl, S. and P. Miller. Whole language and language
experience approaches for beginning reading: a quantitative
research synthesis. Review of Educational Research
59(1):87-116, 1989.

Stevens, K. Readability Formulae and McCall-Crabbs Standard
Test Lessons in Reading. The Reading Teacher 413-415,
January 1980.

Sticht, T. Reading for Working: A Functional Literacy
Anthology. Human Resources Research Organization,
Alexandria, VA., 1975.

Strychar, 1., C. Achterberg, and L. Sullivan. EFNEP for
pregnant teens. Journal of Nutrition Education 23:40C-40D,
1991.

Subar, A., G. Block and L. James. Folate intake and food
sources in the U.S. population. American Journal of
Clinical Nutrition 50:508-516, 1989.

Suitor, C., J. Gardner and M. Feldstein. Characteristics of
diet among a culturally diverse group of low-income pregnant
women. Journal of the American Dietetic Association 90:543-
549, 1990.

Tajika, H., A. Taniguchi, K. Yamamoto and R. Mayer. Effects
of pictorial organizers on passage retention. Contemporary
Education Psychology 13:133-139. 1988.

Talmage, H. and S.P. Rasher. Validity and reliability
issues in measurement instrumentation. J. of Nutrition
Education 13:83—85, 1981.

Taper, L., J. Oliva and S. Ritchey. Zinc and copper
retention during pregnancy: the adequacy of prenatal diets
with and without dietary supplementation. American Journal
of Clinical Nutrition 41:1184—1192, 1985.

288

Taylor, A., J. Skelton and R. Czajkowski. Do patients
understand patient-education brochures? Nursing and Health
Care 305-310, 1982.

Taylor, B. Good and poor readers’ recall of familiar and
unfamiliar text. Journal of Reading Behavior 11(4):375-380,
1979.

Taylor, W. Cloze procedure: a new tool for measuring
readability. Journalism Quarterly 415-433, Fall 1953.

Thorndike, E. and I. Lorge. The Teacher’s Word Book of
30,000 Words. New York Teachers’ College, Columbia
University, New York, 1944.

Tierson, F., C. Olsen and E. Hook. Influence of cravings
and aversions on diet in pregnancy. Ecology of Food and
Nutrition 17:117-129, 1985.

Trouba, P., N. Okereke and P. Splett. Summary document of
nutrition intervention in prenatal care. Journal of the
American Dietetic Association S21-S26, 1991.

USDHHS. Pretesting in Health Communications: Methods,
Examples and Resources for Improving Health Messages and
Materials. NIH Publications, no. 83-1493, Bethseda, MD.,
National Cancer Institute, 1982.

Van Raaij, J., C. Schonk, S. Vermaat-Miedema, M. Peek and J.
Hautvast. Body fat mass and basal metabolic rate in Dutch
women before, during and after pregnancy: a reappraisal of
energy cost of pregnancy. American Journal of Clinical
Nutrition 49:765-772, 1989.

Vaughan, J. Interpreting Readability Assessments. Journal
of Reading 635-639, May 1976.

Vergel, R., L. Sanchey, P. Heredero, P. Rodriguez and A.
Martinez. Primary prevention of neural tube defects with
folic acid supplementation: Cuban experience. Prenatal
Diagnosis 10:149-152, 1990.

Worthington-Roberts, B. and S. Williams. Nutrition in
Pregnancy and Lactation. Times Mirror Publishing, Boston,
MA., 1989.

Zakaluk, B. and S. Samuels. Toward a new approach to
predicting text comprehensibility. Ip Readability: Its
Past, Present and Future. International Reading
Association, Newark, DE., 1988.