Applying the Theory of Planned Behavior to Predict Dairy Product Consumption by Older Adults Article 2022

ABSTRACT
Objective: The purpose of this study was to explain intention
to consume dairy products and consumption of dairy prod-
ucts by older adults using the Theory of Planned Behavior
(TPB). The factors examined were attitudes, subjective
norms, and perceived behavioral control.
Design: A cross-sectional questionnaire was administered.
Setting: Community centers with congregate dining pro-
grams, group classes, and recreational events for older adults.
Subjects: One hundred and sixty-two older adults (mean age
75 years) completed the questionnaire. Subjects were mostly
women (76%) and white (65%), with about half having less
than a high school education or completing high school.
Variables Measured: Variables based on the TPB were
assessed through questionnaire items that were constructed
to form scales measuring attitudes, subjective norms, per-
ceived behavioral control, and intention to consume dairy
products. Dairy product consumption was measured using a
food frequency questionnaire.
Analysis: Regression analyses were used to determine the
association between the scales for the 3 variables proposed in
the TPB and intention to consume and consumption of
dairy products; the α level was set at .05 to determine the
statistical significance of results.
Results: Attitudes toward eating dairy products and per-
ceived behavioral control contributed to the model for pre-
dicting intention, whereas subjective norms did not. Atti-
tudes toward eating dairy products were slightly more
important than perceived behavioral control in predicting
intention. In turn, intention was strongly related to dairy
product consumption, and perceived behavioral control was
independently associated with dairy product consumption.
Conclusions and Implications: These results suggest the util-
ity of the TPB in explaining dairy product consumption for
older adults. Nutrition education should focus on improving
attitudes and removing barriers to consumption of dairy
products for older adults.
KEY WORDS: dairy product consumption, older adults,
Theory of Planned Behavior, questionnaire research
(J Nutr Educ Behav. 2003;35:294-301.)
INTRODUCTION
Based on National Health and Nutrition Examination Sur-
vey (NHANES) data (1988-1994), it was estimated that 10%
of adults aged 50 years and older had osteoporosis as
measured by low total femur bone mineral density (age
adjusted to the year 2000 standard population). 1 A meta-
analysis regarding the effect of calcium on bone density and
fractures in postmenopausal women found that calcium was
more effective than placebo in reducing rates of bone loss,
with a trend toward a reduction in vertebral fractures. 2
Adequate calcium intake may be a simple and inexpensive
strategy to prevent osteoporosis, which is a major public
health problem. A recent review of national dietary intake
data showed that calcium intake was about half of the rec-
ommended level for women in their 50s to 80s, whereas
intake for men was slightly higher but still less than recom-
mended. 3 Dairy foods contribute about 73% of the calcium
in the US per capita food supply. 4 Based on 24-hour recall
data, intake of dairy products for adults 50 years and over was
reported to be 1.3 servings per day for women and 1.0 to 1.1
servings per day for men. 5 Because of the positive health
benefits derived from an adequate intake of dairy products,
and therefore calcium, an increase in consumption of dairy
products by older adults is recommended.
294
R E S E A R C H A RT I C L E
Applying the Theory of Planned Behavior to Predict Dairy Product
Consumption by Older Adults
K Y U N G WO N K I M , P H D; 1 M A R L A R E I C K S , P H D, RD; 2 S A R A S J O B E R G , BS 2
1 Department of Food Science and Nutrition, Seoul Women’s University, Seoul, Korea; 2 Department of Food Science
and Nutrition, University of Minnesota, St. Paul, Minnesota
Funding for this project was obtained from the Minnesota Agricultural Experiment
Station (MIN-054-026).
Address for correspondence: Marla Reicks, PhD, RD, Department of Food Science
and Nutrition, 1334 Eckles Avenue, St. Paul, MN 55108;Tel: [phone]; Fax: (612)
625-5272; E-mail: [e-mail].
©2003 SOCIETY FOR NUTRITION EDUCATION

Journal of Nutrition Education and Behavior Volume 35 Number 6 November • December 2003 295
A better understanding of important psychosocial vari-
ables that influence dietary behavior is needed to develop
effective interventions involving an increase in dairy prod-
uct consumption by older adults. Use of an appropriate the-
oretical framework provides structure to the identification of
factors influencing dairy product consumption.The Theory
of Planned Behavior (TPB) has recently been used to iden-
tify important factors influencing dairy product intake by
young women eligible for the WIC program, 6 healthful eat-
ing by adolescents, 7 and fruit and vegetable consumption. 8
Based on a meta-analysis involving 185 independent studies,
the TPB accounted for 27% and 38% of the variance in
behavior and intention, respectively, 9 which was comparable
to the predictive ability of other theories that have been used
in relation to dietary behaviors. 10
The TPB includes 3 constructs that explain intention to
perform health behaviors (Figure). 11,12 They include attitudes
toward the behavior, subjective norms, and perceived behav-
ioral control. Attitudes are determined by beliefs about the
likelihood of outcomes and their importance. Subjective
norms are determined by what others think the individual
should do and the individual’s motivation to comply. Per-
ceived behavioral control is determined by control beliefs
that can facilitate or inhibit the behavior, such as internal
factors (skills, abilities) and external factors (opportunities,
barriers). Both intention and perceived behavioral control
have direct influence on behavior.
Inadequate consumption of dairy products may be related
to low calcium intake for older adults. In the present study,
the TPB was used to explain intention to consume as well as
consumption of dairy products among older adults by exam-
ining factors including attitudes, subjective norms, and per-
ceived behavioral control. Addressing important factors that
predict consumption of dairy products by older adults may
contribute to improved effectiveness of educational efforts
targeted to this population.
METHODS
Study Design and Subjects
This study used a cross-sectional questionnaire design.
Initial individual interviews using open-ended questions
were conducted with 33 older adults to obtain information
necessary to develop the questionnaire. Subjects complet-
ing the questionnaire were recruited from 7 community
centers in the Minneapolis-St. Paul metropolitan area.
Those who participated in initial interviews were not asked
to complete the questionnaire. Investigators recruited 197
older adults (aged 65 years or older). Some provided
incomplete information (n = 20); another 15 individuals
did not meet the age criteria.The final sample included 162
older adults.
Measures Used in the Questionnaire
The questionnaire was developed according to the steps sug-
gested by Montano et al 12 and Ajzen and Madden 13 based on
Behavioral beliefs
(eg, Eating dairy products
regularly helps me have a
balanced diet.)
Attitude toward
consuming dairy
products
Subjective norm
(social influence)
Intention to
consume dairy
products
Dairy product
consumption
Normative beliefs
(eg, My doctor thinks I
should drink/eat dairy
products.)


Motivation to comply
(eg, I want to comply with
doctor.)
Perceived control
over consuming
dairy products
Control beliefs
(eg, It is difficult to keep
dairy products always avail-
able in my home.)
Figure. Proposed relationship between variables in this study based on the Theory of Planned Behavior.

the information obtained from initial interviews with 33
older adults.
Attitudes toward consuming dairy products. During
the initial individual interviews, older adults (n = 33)
responded to open-ended questions regarding the benefits
and advantages or disadvantages related to eating dairy
products. After counting frequencies and reviewing the
responses, 19 items were selected to measure behavioral
beliefs. These items included nutritional benefits (such as
strong bones, balanced meals, vitamins, and minerals), prac-
tical reasons for eating dairy foods (such as taste, going well
with other foods, providing a snack), and disadvantages of
eating dairy foods (such as make me feel sick, high in fat or
cholesterol).
Each item was designed to be measured on a 5-point
scale (very unlikely to very likely) to indicate the strength of
these beliefs. Outcome evaluation, which is the other com-
ponent comprising attitudes in the TPB, was not considered
in this study because prior experience indicated that most
people similarly evaluate the outcome of behavior. 14 Atti-
tudes toward consuming dairy products were defined as the
summated score of the 19 behavioral beliefs.The Cronbach
coefficient α15 was .82 and was considered acceptable for the
attitudes scale.
Subjective norms. Thirteen different significant others
or information sources were listed by older adults during the
initial individual interviews as having an influence over dairy
product consumption.These were categorized into 7 groups:
family members (daughters/sons, sisters/brothers), spouse,
friends, doctor, cooks at senior centers, television programs,
and newspapers/magazines. Items for normative beliefs were
rated on a 5-point scale (very unlikely to very likely).The cor-
responding motivation to comply with each significant other
or source of information was measured on a 5-point scale
(not at all to very much).There was also a response category
for subjects to check if these significant others did not apply
to the subjects, for example, if the person did not have a spouse
or never watched television.This response category was coded
as a neutral point.The subjective norms variable was defined
as the summated score of the product of each normative belief
and motivation to comply. The Cronbach coefficient α was
.81 for this scale and was considered acceptable.
Perceived behavioral control over consuming dairy
products. During the initial interviews, subjects identi-
fied factors or situations that made it difficult to consume
dairy products. Similar responses were grouped together,
resulting in 21 items to measure control beliefs. These
included perceived confidence in eating dairy foods (such as
for snacks, milk-based desserts, low-fat products, eating with
meals), confidence in eating dairy foods in several different
situations (such as when eating out, whenever you want to),
and availability issues and access to dairy foods.These items
were rated on 5-point scales (disagree a lot to agree a lot or
very difficult to very easy). Perceived behavioral control over
consuming dairy products was defined as the summated
score of control beliefs.The Cronbach coefficient α for this
scale was .90 and was also considered acceptable.
Intention. Two items were used to measure intention to
consume dairy products.The item,“How likely is it that you
will eat dairy products regularly (2-3 servings a day) for the
next month?” was rated on a 5-point scale (very unlikely to
very likely).The other item was “How many servings of dairy
products do you plan to eat for the next month?” and had end
points from never/rarely to more than 3 servings a day. Inten-
tion was defined as the summated score of the 2 items.
Dairy product consumption. Dairy product consump-
tion was measured using food items in Block’s Food Fre-
quency Questionnaire. 16 Ten food items were used, includ-
ing milk, milk on cereal, cheese, and yogurt.To measure dairy
product consumption more accurately, milk or cheese used
in cooking foods was also included. Subjects were asked to
indicate how often and how much they usually ate for each
food item. Dairy product consumption was quantified as the
number of servings consumed per day.
A preliminary draft of the questionnaire was reviewed by
3 university professionals and pilot tested with 5 older adults
to check understanding of items. Based on the pilot test,
minor changes were made in format and wording.The study
was approved by the Human Subjects Protection Commit-
tee of the University of Minnesota’s Institutional Review
Board.
Data Collection
Data were collected after meals at congregate dining centers
or after classes or meetings in community centers between
November 2001 and May 2002. The food frequency ques-
tionnaire items regarding dairy products were completed
prior to completion of the measures related to the 3 variables
proposed in the TPB. In most cases, the food frequency items
were administered by investigators through personal inter-
views, followed by subjects responding to the remaining
items without assistance. However, some subjects (about 10%)
needed assistance because they had difficulty either reading or
marking the questionnaire. It took 20 to 30 minutes for most
of the older adults to complete the questionnaire.
Statistical Analysis
Data were analyzed using the Statistical Analysis System
(SAS, Version 6.12, Cary, NC). Descriptive statistics were
used to examine demographic characteristics and study vari-
ables. Student’s t-tests were done to determine differences in
dairy product consumption according to demographic vari-
able categories. Correlation analysis was used to examine the
simple association between study variables.To investigate the
association between the scales for the 3 variables proposed in
the TPB and intention to consume and consumption of
296 Kim et al/THEORY OF PLANNED BEHAVIOR TO PREDICT DAIRY CONSUMPTION BY OLDER ADULTS