Abstract

The purpose of this study was to investigate the effects of personal characteristics and theory of planned behavior (TPB) constructs on the intention to participate in a restaurant health promotion program. In total, 830 adults residing in Seoul were sampled by a multi-stage cluster and random sampling design. Data were collected from a structured self-administered questionnaire, which covered variables concerning demographics, health status and TPB constructs including attitude, subjective norm and perceived behavioral control. A path analysis combining personal characteristics and TPB constructs was used to investigate determinants of the customers' intention. Positive and negative attitudes, subjective norms and perceived behavioral control directly affected the intention to participate. Demographics and health status both directly and indirectly affected the intention to participate. This study identifies personal characteristics and TPB constructs that are important to planning and implementing a restaurant health promotion program.

INTRODUCTION

An annual report on Korean household incomes and expenditures indicated that the percentage of money spent on groceries, which accounted for 29.0% of total household expenditures of urban populations in 1995, decreased to 26.4% in 2005, whereas spending on dining out, which accounted for 9.1%, increased to 12.2% (Korea National Statistical Office, 2006). The frequency of dining out also increased; 45.7% of metropolitan residents, 47% of residents of smaller cities and 32.5% of rural residents dined out at least once a day in 2005 (Ministry of Health and Welfare and Korean Health Industry Development Institute, 2006).

Increased reliance on the food service industry reflects the increased influence of restaurant cuisine on health. Compared with the Korean dietary reference intakes, restaurant meals in Seoul had large portion size and excessive calories, fat and sodium (Joung, 2007), which were risk factors for many chronic diseases (WHO and FAO, 2003). For these reasons, public health concerns underscore the need for a program to promote healthy restaurant menus. Nutrients labeling can be a good technique for restaurant health promotion programs providing consumers with the information to select healthy food by themselves.

A restaurant health promotion program aims to bring about behavioral changes for healthy eating. Since dining out is a highly complex activity, it is necessary to identify factors that affect dining-out behavior in order to plan and implement effective programs to promote healthy dining-out practices.

The theory of planned behavior (TPB) (Ajzen, 1988) provided the framework for this study. The TPB is an expectancy-value model which can be used to understand the intention of health behaviors. The TPB (Ajzen, 2001) has suggested that interventions to change behavior can be developed by analyzing beliefs which are related to behavior. The TPB has been employed in many studies to analyze behavior in relation to healthy meals or healthy foods. These include the analyses of factors affecting the consumption of a low-fat diet (Armitage and Conner, 1999), sugar-restricted foods by college students (Masalu and Åstrøm, 2003), sugar-free products by youth (Messina et al., 2004), fruits and vegetables (Kellar and Abraham, 2005) and the changes in dietary behavior by diabetics (Blue, 2007).

However, factors influencing the intention to participate in a restaurant health promotion program were not adequately investigated. Therefore, this study intends to analyze the factors that influence customers’ intention to participate in such a program by applying a path analysis combining personal characteristics and the TPB, and to suggest implications for planning and implementing a restaurant health promotion program.

THEORETICAL BACKGROUND AND MODEL FOR ANALYSIS

The TPB (Ajzen, 1985, 1991, 2001) posits that an individual subjectively considers outcomes before making decisions. According to the TPB, the best single predictor of one's behavior is the intention to perform that behavior. Intention is in turn a function of the person's attitude (i.e. perceptions of whether a behavior will yield a positive or negative outcome), the subjective norm (i.e. perceptions of whether relevant others think one should or should not perform a behavior) and perceived behavioral control (i.e. perception of the ease or difficulty of carrying out a behavior). Therefore, perceptions of more favorable outcomes, stronger subjective norms and greater perceived control over the behavior indicate a higher probability that the behavior will be performed.

The TPB has been used to explain a broad range of behaviors; however, such applications have frequently involved modifications and revisions. Some studies have used models with new variables (Raats et al., 1995; Ouellette and Wood, 1998; Armitage and Conner, 1999; Payne et al., 2004) or substitute variables (Conner et al., 2003) that enhance the prediction of behavioral intentions and behaviors. Other studies have used models with different internal structures of the TPB, that is, those that consider interaction among the variables (Conner and McMillan, 1999) or that analyze attitudes (Ito et al., 1998) or subjective norms (Povey et al., 2000a; Baker et al., 2003) multi-dimensionally.

Considering the unique characteristics of participation in a restaurant health promotion program, we modified attitudes of the TPB model. The difficulty of changing eating habits (i.e. of altering preferred tastes, meals or restaurants) can interfere with the customers’ intention to participate in a restaurant health promotion program. Therefore, rather than analyzing attitudes as a single factor, we conducted separate analyses of the effects of both positive and negative attitudes on the intention to participate. We used subjective norms and perceived behavioral control in accordance with their roles in the TPB because participation in a restaurant health promotion program not only can be influenced by the advice of relevant others such as family members, friends and colleagues but also is subject to uncontrollable factors such as accessibility to restaurants that serve healthy meals, affordable price of meals and capacity to negotiate with companions about preferences.

In addition, there are studies reporting that demographics and health status may influence the likelihood of choosing a healthy diet. Fogli-Cawley et al. found that male and obesity had negative effects on eating a healthy diet (Fogli-Cawley et al., 2007). Dynesen et al. and Ball et al. reported that lower education and single marital status had negative effects on eating a healthy diet (Dynesen et al., 2003;,Ball et al., 2004). Blue found that diabetes-related cognitions were positively related to a healthy diet, and Andrykowski et al. reported that a cancer diagnosis was positively associated with health behaviors (Andrykowski et al., 2005; Blue, 2007). On the basis of the hypothesis that these factors could influence the intention to participate, we planned to use a model (Fig. 1) that combined measurements of demographics and health status with TPB constructs.

Fig. 1:

Path analysis model combining personal characteristics and the TPB.

Fig. 1:

Path analysis model combining personal characteristics and the TPB.

METHODS

Sample selection and data collection

This study was conducted to investigate the factors affecting the intention to participate in a restaurant health promotion program among Korean adults. The sample was selected among Seoul residents aged 19 and above by using a multi-stage cluster and random sampling design. Two districts of Seoul were chosen at random, and then one elementary school, one middle school and one high school were selected randomly in each district. The parents of students in one homeroom class per grade within the selected schools were sampled. One university in Seoul was selected randomly; two lectures were selected randomly, and the students who attended those lectures and their parents were sampled. Out of 1000 questionnaires distributed, 830 questionnaires were used for final analysis, excluding unreturned and incomplete questionnaires. A cross-sectional survey, using self-administered questionnaires, was conducted for these samples. Data were collected from December 2007 to January 2008.

Measures

The questionnaire consisted of two parts. One part contained demographics and health status. The other part included the TPB variables.

Demographics included gender, education and marital status. Education was measured in years of education. Marital status was assessed as married, unmarried or separated/divorced/bereaved. In the analysis, marital status was classified into two categories: ‘with a spouse’ and ‘without a spouse’ including unmarried. The body mass index (BMI) and chronic disease status served as measures of health status. BMI was calculated on the basis of self-reported height and weight. Chronic disease status was measured by self-reports of having been diagnosed by a medical doctor for diabetes, hypertension, hypercholesterolemia, arteriosclerosis, liver disease, cardiovascular disease, osteoporosis or any type of cancer.

The questionnaires for the TPB were developed from in-depth interviews with 20 adult residents of Seoul conducted by trained members of the research team. Items that were designated as necessary by 30% or more of these subjects were selected.

Positive and negative attitudes toward the behavior, subjective norms and perceived behavioral control were treated as four intermediate variables. The intermediate variables were measured using items to identify beliefs and to evaluate beliefs. This study used six items to measure the beliefs of positive attitudes toward the behavior, four items for negative attitudes toward the behavior, five items for subjective norms and five items for perceived behavioral control. Each item was measured on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Five-point Likert scales were also used for measurement in the evaluation section. Evaluation of attitude ranged from 1 (very bad) to 5 (very good), evaluation of subjective norms ranged from 1 (strongly do not want to follow) to 5 (strongly want to follow) and evaluation of perceived behavioral control ranged from 1 (very difficult) to 5 (very easy). After the data were collected, we performed a factor analysis and measured Cronbach's alpha to verify the validity and reliability of the questionnaires. The results of the factor analysis indicated that the original four constructs composed of items to measure beliefs were valid, with eigenvalues higher than 1 (Table 1). However, an item about restaurant ambiance was classified as a different factor and was deleted from the five items within perceived behavioral control. Cronbach's alpha ranged from 0.71 for beliefs about perceived behavioral control to 0.90 for beliefs of positive attitudes (Table 1).

Table 1:

Results of the factor analysis and Cronbach's alpha (n = 830)

Construct Modal belief items Factor analysis
 
Alpha coefficient 
  Factor loading Eigenvalue  
Behavioral beliefs ‘If I choose food with nutrition labels (calories, fat, sodium), it will ∼’    
Positive help me control my weight 0.78 4.18 0.90 
help me reduce my intake of saturated or trans fats 0.83   
help me reduce my intake of sodium 0.81   
help me prevent overeating 0.76   
help me improve my overall health 0.79   
help me prevent food-related chronic diseases 0.78   
Negative decrease the chance of enjoying my favorite tastes 0.64 2.14 0.75 
decrease the chance of enjoying my favorite foods 0.72   
be inconvenient to find restaurants that use the nutrition-labeling program 0.78   
be inconvenient to decipher the labeled information on food 0.74   
Normative beliefs ‘∼ will support my choice of nutrition-labeled food.’    
 Family members (parents, spouse, children, brothers or sisters) 0.67 3.23 0.87 
Friends 0.80   
Colleagues or neighbors 0.81   
My companion for the current meal 0.78   
Acquaintances working in the food industry (chef, cook, server) 0.52   
Control beliefs ‘I can ∼’    
 go to eat the nutrition-labeled food whenever I want to do so 0.70 2.04 0.71 
encourage myself to seek out nutrition-labeled food no matter what 0.72   
afford the nutrition-labeled food with my financial ability 0.74   
persuade my dining-out companion to eat the nutrition-labeled food when he/she hesitates to do so 0.55   
Construct Modal belief items Factor analysis
 
Alpha coefficient 
  Factor loading Eigenvalue  
Behavioral beliefs ‘If I choose food with nutrition labels (calories, fat, sodium), it will ∼’    
Positive help me control my weight 0.78 4.18 0.90 
help me reduce my intake of saturated or trans fats 0.83   
help me reduce my intake of sodium 0.81   
help me prevent overeating 0.76   
help me improve my overall health 0.79   
help me prevent food-related chronic diseases 0.78   
Negative decrease the chance of enjoying my favorite tastes 0.64 2.14 0.75 
decrease the chance of enjoying my favorite foods 0.72   
be inconvenient to find restaurants that use the nutrition-labeling program 0.78   
be inconvenient to decipher the labeled information on food 0.74   
Normative beliefs ‘∼ will support my choice of nutrition-labeled food.’    
 Family members (parents, spouse, children, brothers or sisters) 0.67 3.23 0.87 
Friends 0.80   
Colleagues or neighbors 0.81   
My companion for the current meal 0.78   
Acquaintances working in the food industry (chef, cook, server) 0.52   
Control beliefs ‘I can ∼’    
 go to eat the nutrition-labeled food whenever I want to do so 0.70 2.04 0.71 
encourage myself to seek out nutrition-labeled food no matter what 0.72   
afford the nutrition-labeled food with my financial ability 0.74   
persuade my dining-out companion to eat the nutrition-labeled food when he/she hesitates to do so 0.55   

Intention to participate in a restaurant health promotion program was treated as a dependent variable. Intention was measured with eight items (Table 2) using a five-point Likert scale (Cronbach's alpha = 0.89).

Table 2:

Items about intention to participate (n = 830)

Intention Alpha coefficient 
‘If nutrition-labeling restaurants open, within a month I will ∼’ 0.89 
 dine out in those restaurants 
 try to dine out in those restaurants 
 plan to dine out in those restaurants 
 prepare to dine out in those restaurants 
‘If nutrition-labeling restaurants open, within a year I will ∼’ 
 dine out in those restaurants 
 try to dine out in those restaurants 
 plan to dine out in those restaurants 
 prepare to dine out in those restaurants 
Intention Alpha coefficient 
‘If nutrition-labeling restaurants open, within a month I will ∼’ 0.89 
 dine out in those restaurants 
 try to dine out in those restaurants 
 plan to dine out in those restaurants 
 prepare to dine out in those restaurants 
‘If nutrition-labeling restaurants open, within a year I will ∼’ 
 dine out in those restaurants 
 try to dine out in those restaurants 
 plan to dine out in those restaurants 
 prepare to dine out in those restaurants 

Data analysis

Descriptive statistics of the demographics and health variables and TPB constructs of the subjects were conducted. A path analysis was used to analyze the factors affecting intention to participate in a restaurant health promotion program by using AMOS Version 5.0.

RESULTS

Descriptive findings

In terms of demographics, 64.5% of respondents were female, 65.2% were those with a spouse and the mean of years of education of the respondents was 14.48 (SD = 2.56). With regard to health status, the mean BMI score of the respondents was 22.13 (SD = 2.40), with the score ranging from 15.23 to 31.64, and the average number of chronic diseases was 0.94 (SD = 1.46), with the numbers ranging from 0 to 8 (Table 3).

Table 3:

Descriptive findings of study variables (n = 830)

Variable Frequency Mean SD 
Demographics 
 Gender     
  Male 295 35.5   
  Female 535 64.5   
 Education (years)a   14.48 2.56 
 Marital status     
  With a spouse 541 65.2   
  Without a spouse 289 34.8   
Health status 
 BMI (kg/m2  22.13 2.40 
 Number of chronic diseases   0.94 1.46 
TPBb 
 Positive attitudes   16.01 4.20 
 Negative attitudes   7.15 3.34 
 Subjective norms   14.09 4.78 
 Perceived behavioral control   13.47 5.41 
 Intention   3.79 0.78 
Variable Frequency Mean SD 
Demographics 
 Gender     
  Male 295 35.5   
  Female 535 64.5   
 Education (years)a   14.48 2.56 
 Marital status     
  With a spouse 541 65.2   
  Without a spouse 289 34.8   
Health status 
 BMI (kg/m2  22.13 2.40 
 Number of chronic diseases   0.94 1.46 
TPBb 
 Positive attitudes   16.01 4.20 
 Negative attitudes   7.15 3.34 
 Subjective norms   14.09 4.78 
 Perceived behavioral control   13.47 5.41 
 Intention   3.79 0.78 

aIn education, elementary school was calculated as 6, middle school as 9, high school as 12, junior college as 14, university as 16, master graduate as 18, PhD graduate as 21.

bThe scores of positive attitudes, negative attitudes, subjective norms and perceived behavioral control are between 1 (least positive) and 25 (most positive), and the score of intention is between 1 (least positive) and 5 (most positive).

In terms of TPB constructs, the mean score of positive attitudes was 16.01 (SD = 4.20) out of a maximum possible score of 25. Respondents’ mean score of negative attitudes was 7.15 (SD = 3.34) out of a maximum possible score of 25. The average score of subjective norms was 14.09 (SD = 4.78) out of a maximum possible score of 25. Respondents’ average score of perceived behavioral control was 13.47 (SD = 5.41) out of a maximum possible score of 25. The mean score of intention was 3.79 (SD = 0.78) out of a maximum possible score of 5.

Variables affecting the intention to participate in a restaurant health promotion program

Table 4 revealed the correlation matrix among the variables. The correlations between intention and four proximal TPB constructs such as positive attitudes, negative attitudes, subjective norms and perceived behavioral control were 0.398, −0.194, 0.343 and 0.383, respectively (p < 0.001). Table 4 also showed that female gender was positively associated with intention, positive attitudes and subjective norms. Education was positively related to intention, positive attitudes and perceived behavioral control, and negatively related to negative attitudes. Marital status (without a spouse) was negatively associated with positive attitudes and subjective norms. BMI was positively related to positive attitudes and negative attitudes and negatively related to perceived behavioral control. Chronic disease was positively associated with intention, positive attitudes, perceived behavioral control and BMI.

Table 4:

Correlation matrix of personal characteristics and TPB constructs (n = 830)

  
Intention          
Positive attitudes 0.398***         
Negative attitudes −0.194*** −0.011        
Subjective norms 0.343*** −0.007 0.011       
Perceived behavioral control 0.383*** 0.017 0.014 0.006      
Gender (female) 0.469*** 0.256*** −0.059 0.202*** 0.038     
Education 0.154*** 0.130*** −0.107** 0.027 0.118*** 0.015    
Marital status (without a spouse) −0.061 −0.121*** 0.026 −0.154*** 0.016 0.020 −0.027   
BMI −0.009 0.088* 0.118*** 0.038 −0.142*** −0.031 0.029 −0.001  
10 Number of chronic diseases 0.168*** 0.157*** −0.020 −0.026 0.152*** 0.048 −0.008 0.007 0.104** 
  
Intention          
Positive attitudes 0.398***         
Negative attitudes −0.194*** −0.011        
Subjective norms 0.343*** −0.007 0.011       
Perceived behavioral control 0.383*** 0.017 0.014 0.006      
Gender (female) 0.469*** 0.256*** −0.059 0.202*** 0.038     
Education 0.154*** 0.130*** −0.107** 0.027 0.118*** 0.015    
Marital status (without a spouse) −0.061 −0.121*** 0.026 −0.154*** 0.016 0.020 −0.027   
BMI −0.009 0.088* 0.118*** 0.038 −0.142*** −0.031 0.029 −0.001  
10 Number of chronic diseases 0.168*** 0.157*** −0.020 −0.026 0.152*** 0.048 −0.008 0.007 0.104** 

In gender, the control group is male; in marital status, the control group is with a spouse.

*p < 0.05.

**p < 0.01.

***p < 0.001.

To investigate the factors affecting the intention to participate in a restaurant health promotion program, a path analysis combining personal characteristics and the TPB constructs was used. A χ2 goodness-of-fit test was used to assess the appropriateness of the model. The larger the probability that is associated with χ2 computed for a model, the better the model fits the data (Pedhazur, 1982). The χ2 for this model in Fig. 2 was 33.53, df = 26, p = 0.147, indicating congruence between the model and the data. The root mean square error of approximation (RMSEA) is less affected by sample size than the χ2. The RMSEA ranges from 0 to ∞, with fit values less than 0.05 indicating close fit and values less than 0.10 indicating reasonable fit (Kim, 2007). The RMSEA for this model in Fig. 2 was 0.019, supporting the appropriateness of the model.

Fig. 2:

Results of path analysis model. In gender, the control group is male; in marital status, the control group is with a spouse. #p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 2:

Results of path analysis model. In gender, the control group is male; in marital status, the control group is with a spouse. #p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.

The direct, indirect and total effects of variables that affect the intention to participate in a restaurant health promotion program are summarized as follows (Fig. 2, Table 5). Gender (female) (direct effect 0.307, indirect effect 0.133), education (direct effect 0.043, indirect effect 0.099) and chronic disease (direct effect 0.057, indirect effect 0.101) had both direct and indirect effects on the intention to participate. Marital status (without a spouse) (−0.082) and BMI (−0.056) had indirect effects on the intention to participate. Positive attitudes (0.297), negative attitudes (−0.175), subjective norms (0.283) and perceived behavioral control (0.354) had direct effects on the intention to participate.

Table 5:

Direct, indirect and total effects on the intention to participate (n = 830)

Independent variable Direct effect Indirect effect Total effect 
Demographics 
 Gender (female) 0.307 0.133 0.440 
 Education 0.043 0.099 0.142 
 Marital status (without a spouse)  –0.082 –0.082 
Health status 
 BMI  –0.056 –0.056 
 Number of chronic diseases 0.057 0.101 0.158 
TPB 
 Positive attitudes 0.297  0.297 
 Negative attitudes –0.175  –0.175 
 Subjective norms 0.283  0.283 
 Perceived behavioral control 0.354  0.354 
Independent variable Direct effect Indirect effect Total effect 
Demographics 
 Gender (female) 0.307 0.133 0.440 
 Education 0.043 0.099 0.142 
 Marital status (without a spouse)  –0.082 –0.082 
Health status 
 BMI  –0.056 –0.056 
 Number of chronic diseases 0.057 0.101 0.158 
TPB 
 Positive attitudes 0.297  0.297 
 Negative attitudes –0.175  –0.175 
 Subjective norms 0.283  0.283 
 Perceived behavioral control 0.354  0.354 

An examination of the total effects indicates that with regard to demographics, female respondents and those with more education reported a stronger intention to participate, whereas those without a spouse reported a weaker intention. In terms of health status, respondents who were obese reported a weaker intention, whereas those with chronic diseases reported a stronger intention. In terms of the TPB variables, respondents who expressed more favorable attitudes toward healthy diets, perceived greater influence from relevant others and perceived greater control over their diet reported a stronger intention, whereas those who expressed more unfavorable attitudes reported a weaker intention.

DISCUSSION

By using a path analysis, this study investigated the direct and indirect effects of personal characteristics, as well as the direct effects of the TPB constructs, on the intention to participate in a restaurant health promotion program. The model that combined personal characteristics and the TPB constructs was assessed as adequate (χ2 = 33.53, df = 26, p = 0.147) for predicting the intention to participate in a restaurant health promotion program.

The major finding in the model was the considerable direct effects of the TPB constructs on the intention to participate. Positive attitudes had a significant effect on the intention (p < 0.001, Fig. 2) to participate positively. This is consistent with the findings of Povey et al., Andrykowski et al. and Blue (Povey et al., 2000b;,Andrykowski et al., 2005;,Blue, 2007). Negative attitudes negatively affected the intention (p < 0.001, Fig. 2) to participate. This result is similar to that obtained by Conner et al. in an analysis of the negative effects of ambivalent attitudes on intentions to participate (Conner et al., 2003). Subjective norms were associated with the intention (p < 0.001, Fig. 2) to participate. A significant effect of subjective norms was also found by Baker et al. in explaining the effects of parents and friends on the intentions of teenagers to eat a healthy diet (Baker et al., 2003). Perceived behavioral control had a significant effect on the intention (p < 0.001, Fig. 2) to participate. This result is consistent with the findings of Sparks et al., Povey et al., Conner et al. and Blue (2007), but inconsistent with that of Messina et al. (Sparks et al., 1997;,Povey et al., 2000b; Conner et al., 2002, 2003; Messina et al., 2004; Blue, 2007).

Another noteworthy finding in the path analysis was the indirect and direct effects of personal characteristics. Marital status and BMI had no direct effect on intention, exerting all of their influence through TPB constructs which had direct effects on intention. Education and chronic disease had stronger indirect effects on intention through TPB constructs than direct effects on intention. Being female had a strong indirect effect on intention through TPB constructs, even though its effect was weaker than a direct effect on intention. Among the demographic indicators, female gender and education had positive effects on the intention to participate in a restaurant health promotion program. Several studies have identified negative effects of male gender (Fogli-Cawley et al., 2007) and lower education (Dynesen et al., 2003) on the intention to eat a healthy diet; this is congruent with our results. Being without a spouse had a negative effect on the intention to participate, which was similar to the results of Dynesen et al. and Ball et al. (Dynesen et al., 2003;,Ball et al., 2004). Among the health indicators, the presence of chronic diseases had a positive effect on the intention to participate in a restaurant health promotion program, which was similar to the result of Blue (Blue, 2007). However, a higher BMI had a negative effect on the intention to participate, which was consistent with the finding of Fogli-Cawley et al. (Fogli-Cawley et al., 2007).

The results of this study suggest that by increasing positive attitudes, subjective norms and perceived behavioral control while decreasing negative attitudes, it is possible to increase the intention to participate. The results also suggest that consideration should be given to the finding that females, those with higher education and those with chronic disease, have a stronger intention to participate, whereas those without a spouse and those who are obese have a weaker intention to participate.

On the basis of these grounds, several implications for the implementation of a restaurant health promotion program can be suggested as follows. First, in order to improve individuals’ positive attitudes toward a health promotion program, social education and public advertisements are necessary to inform people that participation in a restaurant health promotion program can improve health. Second, in order to reduce individuals’ negative attitudes toward a restaurant health promotion program, meals with nutritional information should be provided with wide availability, good taste and a good variety. Third, programs which include roles of relevant others such as family members, friends and colleagues to encourage individuals to participate in a restaurant health promotion program must be developed. Fourth, in order to mitigate the obstacles of perceived behavioral control which might discourage participation in a health promotion program, meals with nutritional information should be priced affordably and social environments that garner widespread acceptance of going to restaurants for healthy food should be created. Also, menus that cater to individuals who go to nutrition-labeling restaurants alone should be developed and restaurants providing nutrition-labeled meals should be strategically located where they are readily accessible. Fifth, in order for a restaurant health promotion program to be successfully implemented, program planning should target groups that are identified as most likely to participate: females, those with higher education and those with chronic disease. Sixth, programs need to be developed promoting positive attitudes, subjective norms and perceived behavioral control, and reducing negative attitudes in groups that are characterized by more negative intentions to participate: males, those with lower education, those without a spouse and those who are obese. Various promotional activities, including community events and in-restaurant promotions providing target consumers with information that will help create awareness of healthy menu choices, can help decrease the negative intentions of groups that are reluctant to participate. Program planners can utilize the Internet by establishing a website in order to widely and efficiently inform consumers about the positive effects of the restaurant health promotion program. The website can also provide information about the restaurants which adopt restaurant health promotion programs and particular menus of each restaurant.

This study is limited in its generalizability because the sample is restricted to adults who reside in a specific geographic area, i.e. the city of Seoul, Korea. Further research using larger sample sizes over wider geographic areas is needed to increase the generalizability. Longitudinal research is needed to assess the effectiveness of programs that are developed on the basis of the results of this study. This study also could not investigate the determinants of actual behavior because a restaurant health promotion program of the sort that we suggest was not yet implemented. However, behavioral theory posits that intention represents the best predictor of future behavior. Research is needed to validate intervention strategies that are designed to facilitate movement from intention to actual behavior after restaurant health promotion programs are implemented. Further research including various demographic characteristics such as income, occupation and frequency of eating out is also needed to uncover the detailed factors affecting the intention of consumers.

FUNDING

This work was supported by the Management Center for Health Promotion, Ministry of Health and Welfare, Republic of Korea (05-74).

REFERENCES

Ajzen
I.
Kuhl
J.
Beckmann
J.
From intentions to actions: a theory of planned behaviour
Action Control: From Cognition to Behaviour
 , 
1985
New York
Springer
Ajzen
I.
Attitudes, Personality, and Behavior
 , 
1988
Chicago
Dorsey Press
Ajzen
I.
The theory of planned behaviour
Organizational Behavior and Human Decision Processes
 , 
1991
, vol. 
50
 (pg. 
179
-
211
)
Ajzen
I.
Nature and operation of attitudes
Annual Review of Psychology
 , 
2001
, vol. 
52
 (pg. 
27
-
58
)
Andrykowski
M. A.
Beacham
A. O.
Schmidt
J. E.
Harper
F. W. K.
Application of the theory of planned behaviour to understand intentions to engage in physical and psychosocial health behaviours after cancer diagnosis
Psycho-Oncology
 , 
2005
, vol. 
15
 (pg. 
759
-
771
)
Armitage
C. J.
Conner
M.
Distinguishing perceptions of control from self-efficacy: predicting consumption of a low fat diet using the theory of planned behaviour
Journal of Applied Social Psychology
 , 
1999
, vol. 
29
 (pg. 
72
-
90
)
Baker
C. W.
Little
T. D.
Brownell
K. D.
Predicting adolescent eating and activity behaviours: the role of social norms and personal agency
Health Psychology
 , 
2003
, vol. 
22
 (pg. 
189
-
198
)
Ball
K.
Mishra
G. D.
Thane
C. W.
Hodge
A.
How well do Australian women comply with dietary guidelines?
Public Health Nutrition
 , 
2004
, vol. 
7
 (pg. 
443
-
452
)
Blue
C. L.
Does the theory of planned behaviour identify diabetes-related cognitions for intention to be physically active and eat a healthy diet?
Public Health Nursing
 , 
2007
, vol. 
24
 (pg. 
141
-
150
)
Conner
M.
McMillan
B.
Interaction effects in the theory of planned behaviour: studying cannabis use
British Journal of Social Psychology
 , 
1999
, vol. 
38
 (pg. 
195
-
222
)
Conner
M.
Norman
P.
Bell
R.
The theory of planned behaviour and healthy eating
Health Psychology
 , 
2002
, vol. 
21
 (pg. 
194
-
201
)
Conner
M.
Povey
R.
Sparks
P.
James
R.
Shepherd
R.
Moderating role of attitudinal ambivalence within the theory of planned behaviour
British Journal of Social Psychology
 , 
2003
, vol. 
42
 (pg. 
75
-
94
)
Dynesen
A. W.
Haraldsdottir
J.
Holm
L.
Astrup
A.
Sociodemographic differences in dietary habits described by food frequency questions: results from Denmark
European Journal of Clinical Nutrition
 , 
2003
, vol. 
57
 (pg. 
1586
-
1597
)
Fogli-Cawley
J. J.
Troy
L. M.
Dwyer
J. T.
Meigs
J. B.
Saltzman
E. S.
Jacques
P. F.
, et al.  . 
The 2005 dietary guidelines for Americans and insulin resistance in the Framingham offspring cohort
Diabetes Care
 , 
2007
, vol. 
30
 (pg. 
817
-
822
)
Ito
T. A.
Larsen
J. T.
Smith
N. K.
Cacioppo
J. T.
Negative information weighs more heavily on the brain: the negativity bias in evaluative categorizations
Journal of Personality and Social Psychology
 , 
1998
, vol. 
75
 (pg. 
887
-
900
)
Joung
H.
The Need and Capacity Assessment for Developing Health Promotion Programs in the Food Service Industry
 , 
2007
Seoul, Korea
Seoul National University and Management Center for Health Promotion
Kellar
I.
Abraham
C.
Randomized controlled trial of a brief research-based intervention promoting fruit and vegetable consumption
British Journal of Health Psychology
 , 
2005
, vol. 
10
 (pg. 
543
-
558
)
Kim
K. S.
An Analysis of Structural Equation Model.
 , 
2007
Seoul, Korea
Hannare Publishing Company
Korea National Statistical Office
Annual Report on the Household Income and Expenditure Survey
 , 
2006
Seoul, Korea
Korea National Statistical Office
Masalu
R. M.
Åstrøm
A. N.
The use of the theory of planned behaviour to explore beliefs about sugar restriction
American Journal of Health Behavior
 , 
2003
, vol. 
27
 (pg. 
15
-
24
)
Messina
F.
Saba
A.
Vollono
C.
Leclercq
C.
Piccinelli
R.
Beliefs and attitudes towards the consumption of sugar-free products in a sample of Italian adolescents
European Journal of Clinical Nutrition
 , 
2004
, vol. 
58
 (pg. 
420
-
428
)
Ministry of Health and Welfare and Korean Health Industry Development Institute
The Third National Health & Nutrition Examination Survey (KNHANES III), 2005 Nutrition Survey (I)
 , 
2006
Seoul, Korea
Ministry of Health and Welfare
Ouellette
J. A.
Wood
W.
Habit and intention in everyday life: the multiple processes by which past behaviour predicts future behaviour
Psychological Bulletin
 , 
1998
, vol. 
124
 (pg. 
54
-
74
)
Payne
N.
Jones
F.
Harris
P. R.
The role of perceived need within the theory of planned behaviour: a comparison of exercise and health eating
British Journal of Health Psychology
 , 
2004
, vol. 
9
 (pg. 
489
-
504
)
Pedhazur
E. J.
Multiple Regression in Behavioral Research.
 , 
1982
New York
Holt, Rein-hart and Winston
Povey
R.
Conner
M.
Sparks
P.
James
R.
Shepherd
R.
The theory of planned behaviour and healthy eating: examining additive and moderating effects of social influence variables
Psychology and Health
 , 
2000
, vol. 
14
 (pg. 
991
-
1006
)
Povey
R.
Conner
M.
Sparks
P.
James
R.
Shepherd
R.
Application of the theory of planned behaviour to two dietary behaviours: roles of perceived control and self-efficacy
British Journal of Health Psychology
 , 
2000
, vol. 
5
 (pg. 
121
-
139
)
Raats
M. M.
Shepherd
R.
Sparks
P.
Including moral dimensions of choice within the structure of the theory of planned behaviour
Journal of Applied Social Psychology
 , 
1995
, vol. 
25
 (pg. 
484
-
494
)
Sparks
P.
Guthrie
C. A.
Shepherd
R.
The dimensional structure of the perceived behavioural control construct
Journal of Applied Social Psychology
 , 
1997
, vol. 
27
 (pg. 
418
-
438
)
WHO and FAO
Diet, Nutrition and the Prevention of Chronic Disease. WHO Technical Report Series 916
 , 
2003
Geneva
World Health Organization