Food Control 22 (2011) 862e868. Contents lists available at ScienceDirect. Food Control. journal homepage:

Transcription

1 Food Control 22 (2011) 862e868 Contents lists available at ScienceDirect Food Control journal homepage: The impact of inspection reports on consumer behavior: A pilot study Jinkyung Choi *, Douglas C. Nelson 1, Barbara Almanza 2 Purdue University, Hospitality and Tourism Management, 700 West State street, West Lafayette, IN 47906, United States article info abstract Article history: Received 8 March 2010 Received in revised form 15 October 2010 Accepted 2 November 2010 Keywords: Health inspection report Information Food safety Protection motivation theory Consumer behavior This pilot study evaluated the impact of inspection score information on consumer behavior. Protection motivation theory was used to investigate consumer s behavioral intentions resulting from food safety information provided in the questionnaire. The questionnaire was created to measure consumers perceived likelihood of changing behavior based on protection motivation theory after reading restaurant food safety information in the form of inspection reports. The pilot showed that the methodology used to collect the data was sound. While the number of respondents was too small to show significance, several important trends were noted. One of the more important finding was that the affordable cost of selecting an alternative restaurant rose with the number of violations. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Food safety practices have always been sensitive topics for food service providers and consumers. An outbreak at a single restaurant in Pennsylvania resulted in 601 patrons contracted Hepatitis A leading to 124 hospitalizations and 3 deaths (Wheeler, Vogt, Armstrong, & Vaughan, 2005). Foodborne illness outbreaks can certainly damage a company s image and lead to decreasing sales. Grover and Dausch (2000) estimated that the average foodborne outbreak costs an operation $100,000 and a 30 percent reduction in sales. There have been many studies on the impact of health inspection scores (Almanza, Ismail, & Mills, 2002; Almanza, Nelson, & Lee, 2003), employees perception of food safety and training (Martin, Knabel, & Mendenhall, 1999; Pragle, Harding, & Mack, 2007; Robert, Binkley, Nelson, & Almanza, 2005) and consumers perception of food safety (Houghton, Kleef, Rowe, & Frewer, 2006; Knight, Worosz, & Todd, 2007). However, specific research targeting the impact of inspection scores on consumers behavior is lacking. Consumer perceptions of food safety are especially critical for restaurant managers and owners; perceptions of poor sanitation might lead to consumers choosing a safer restaurant resulting in a loss of revenue. Previous research has found that 70 percent of respondents would no longer buy food from food service * Corresponding author. Fax: þ addresses: choi108@purdue.edu (J. Choi), nelsond@purdue.edu (D.C. Nelson), almanzab@purdue.edu (B. Almanza). 1 Tel.: þ ; fax: þ Tel.: þ ; fax: þ establishment where they had concerns about hygiene (FSA, 2004). A study conducted by Knight et al. (2007) found that people who perceived that a restaurant was not at all committed to food safety were less likely to choose the restaurant when eating out. A study conducted by Henson, Majowicz, and Masakure (2006) found that cleanliness was the most important determinant for consumers perceptions of restaurant food safety. Health inspection scores by health inspectors are a reflection of restaurant cleanliness and presumably represent the safety of eating at the restaurant. To fully understand the impact of inspection reports, it is critical to measure consumer perception of the seriousness of inspection violations. If sanitation is important to a consumer, this perception may affect behavior. This paper reports the results of a pilot study that tested a scenario type questionnaire to determine the impact of inspection reports on consumers selection of restaurants. In order to investigate consumer behavior, the study adopted protection motivation theory (PMT) in relation to fear appeals (Rogers, 1975). A fear appeal is a communication about a threat to an individual s well-being that can change attitudes and behavior (Milne, Sheeran, & Orbell, 2000). This pilot study investigated the likelihood that inspection scores will generate a fear appeal and motivate consumers to modify their behavior. PMT has previously been adopted for studies on reducing alcohol consumption (Rogers, 1983), the effects of antismoking advertisements (Pechmann, Zhao, Goldberg, & Reibling, 1993) and AIDS-related health behavior (Van der Velde & Van der Plight, 1991). This study uses this theory to explain how available information such as health inspection scores or reports influence consumer behavior when selecting a safe place to eat. PMT is /$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi: /j.foodcont

2 J. Choi et al. / Food Control 22 (2011) 862e a relatively new approach for determining consumer perceptions of food safety in restaurants. 2. Literature review PMT was originally developed to provide conceptual clarity in the area of fear appeals and to bridge the gap between research on fear appeals and research on attitude change (Milne et al., 2000). Protection motivation is an intention to perform a health behavior and it is measured by threat appraisal and coping appraisal. According to a study by Boer and Seydel (1996), threat appraisal and coping appraisal are behavioral options that diminish or reduce threats. PMT originally proposed that the intention to protect oneself depends upon four factors: 1) the perceived severity of a threatened event; 2) the perceived probability of the occurrence, or vulnerability; 3) the efficacy of the recommended preventive behavior; and 4) the perceived self-efficacy (Rogers, 1983). Boer and Seydel (1996) explained that in PMT the outcome of the threat appraisal is the estimation of the likelihood of contracting a disease and the seriousness of that disease. They defined coping appraisal as a combination of self-efficacy and response efficacy. Belief in following recommendations to protect oneself is response efficacy. The ability to execute the recommended courses of action successfully is self-efficacy (Boer & Seydel, 1996). The purpose of PMT research is usually to persuade people to follow the communicator s recommendations; therefore intentions indicate the effectiveness of the attempted persuasion (Floyd, Prentice-Dunn, & Rogers, 2000) and intentions are viewed as a precursor to behavior (Ajzen & Fishbein, 1980). Recent research included behavior in the PMT construct as the outcome variable (Floyd et al., 2000; Milne et al., 2000; Norman, Boer, & Seydel, 2005; Tulloch, Reida, & D Angeloa, 2009). PMT variables were all found to be significantly associated with concurrent behavior. The association between intention and concurrent behavior was the strongest and most consistent association found in the metaanalysis (Milne et al., 2000). In addition, PMT was used to predict exercise intentions and behaviors in the year following hospitalization for coronary artery disease (Tulloch et al., 2009). In their study, the PMT variables accounted for moderate percentages of variance for exercise intentions and behaviors at 6 months posthospitalization. The PMT model has provided a good fit for predicting shortterm exercise intentions and behaviors. PMT may be a useful framework for understanding actions in the short-term when recommended action and consequences are presented together (Tulloch et al., 2009). Rogers s model showed protection motivation was a mediating variable between responses to information and coping modes such as action or inhibition action, a single act, repeated acts, multiple acts and repeated multiple acts (Rogers, 1983). To investigate the impacts of the level of protection motivation on a consumer behavior, this study developed the following conceptual model based on the PMT. PMT has an advantage over the health belief model (Becker, 1974), the theory of reasoned action (Ajzen & Fishbein, 1980; Fishbein & Ajzen, 1975), and the theory of planned behavior (Ajzen, 1988, 1991) in that it has provided more consistent results in experimental tests. Health Belief Model and PMT are better models than Theory of Reasoned Action and Subjective Expected Utility theories to assess individual s response efficacy: the Theory of Reasoned Action and Subjective Expected Utility theories lack elements of response efficacy (Floyd et al., 2000). In previous studies, researchers have used communications designed to manipulate PMT variables and then measure the effects of the communication on PMT variables (Pechmann et al., 1993; Rogers, 1983; Van der Velde & Van der Plight, 1991). This study manipulated the PMT variables and health inspection scores and measured the effects on PMT responses. Previously food safety researchers have concluded that college students aged 20e29 years old engage in unsafe practices, including risky food handling and food consumption (Li-cohen & Bruhn, 2002; Morrone & Rathbun, 2003). Ford and Goode (1994) suggested that individuals in this group would change their health behaviors if the related health issue was perceived as relevant or of concern to them or their peers. In order to change their attitude toward health issues, it was found that students strongly agreed that they needed more information on health issues (Ford & Goode, 1994). This study attempted to examine if college students were influenced by a restaurant inspection report when selecting a restaurant to dine. 3. Methodology A questionnaire was created to measure consumers perceived likelihood of changing behavior after reading restaurant food safety information in the form of inspection reports. The questionnaire addressed several components: the perceived risk of the violations identified on the inspection reports, consumers likelihood of modifying their behavior based on the inspection reports, their perception of the effectiveness of their behavior modification, and general demographics. Prior to measuring consumers protection motivation, it was necessary to determine if inspection reports would impact any of the 6 constructs in the conceptual model. These constructs are vulnerability, severity, benefits, response efficacy, self-efficacy, and costs (see Fig. 1). A scenario approach was used to evaluate consumers perceptions of threat, severity, and vulnerability. Four different scenarios of the inspection reports ( 0 violations, 6 violations, 8 violations, and 10 violations) were presented to find out if changing the numbers of violations would trigger different protection motivation responses. The violations used in this study were taken from the inspection form used by the Tippecanoe Health Department, Indiana. They use the critical and non-critical violations system to inspect foodservice establishments. Critical violations refer to violations that need to be corrected immediately otherwise they pose an immediate threat to food safety. Non-critical violations refer to violations that need to be corrected but do not pose an immediate threat to food safety. Examples of critical violations include items like Employees not washing hands and Food not being held at the proper temperature. Non-critical violations include items like No shatter resistant covers on light bulbs. To get the maximum response with the minimum number of health inspection violations, and to simplify the task for respondents, only critical violations, which appeared on actual health inspection reports, were used in the scenarios. The selected critical violations were based on the most common violations found by the Tippecanoe Health Department. Because of the fact that a failing score under the critical and non-critical violation system is based on the judgment of the health inspectors and is not a set number or level, it was first necessary to standardize the number of critical violations in the scenarios that would elicit differences in responses. Under the traditional 100 point demerit system described in older federal food codes, numeric scores could be transformed into letter grades in the following analogy: 90e100 points equals A, 89e80 points equals a B, 79e70 points equals a C, and below 70 points equals a failing grade. Major (or critical) violations under the older system were worth 4 to 5 points, so the following critical violation numbers were selected for this study to represent similar levels: no violations would represent an A grade, 6 violations would represent a B grade, 8 violations a C, and 10 violations a failing grade. A panel study was conducted prior

3 864 J. Choi et al. / Food Control 22 (2011) 862e868 Severity Vulnerability Benefits Self-efficacy Protection Motivation Behavioral Intention Response-efficacy Cost Fig. 1. A Conceptual PMT framework for consumer s intention to change behavior. to this pilot study to find if number of violations triggers consumer s protection motivation. The panel study utilized 0, 6, 8, and 10 violations (see Table 1). Collected data were analyzed by descriptive analysis to refine and validate the multi-item scales of each construct for the further analyses. The GameseHowell procedure was used for post-hoc comparisons to compensate for unequal sample sizes and the unequal variance in the inspection scores. Table 2 presents the description of measurement items used in this study. These measurement items were developed through previous studies and consultation with food service experts who have conducted research into consumer behavior. Based on current literature, the measurement items were divided into two groups: threat appraisal and coping appraisal measures Threat appraisal measures Threat appraisal consists of perceived severity and perceived vulnerability. Threat appraisal is a key element to the protection motivation theory; if there is no perceived vulnerability or severe consequence, then there is not likely to be a change in behavior. In other words, if a person does not believe that food safety violations are likely to lead to a severe illness, then those violations are not likely to figure into the restaurant selection process. To assess perceived severity, the questions were adopted from items used by Plotnikoff and Hagginbotham (1998) and Tulloch et al., (2009) Table 1 Description of the violations used in the study. such as: If you contract foodborne illness how seriously do you think it will affect you? rated on a scale from one (very mild discomfort) to seven (life threatening). Perceived vulnerability was assessed by asking questions related to how likely the violations will lead to an illness. Another important aspect to threat appraisal is perceived benefits of engaging in the activity: such as eating in a restaurant with food safety violations. If the benefits outweigh Table 2 Description of measurement items. Construct Vulnerability Severity Benefits Measurement items If I eat at the Restaurant X one time in a week, I am likely to contract foodborne illness. If I eat at the Restaurant X several times in a week, I am likely to contract foodborne illness. If I contract foodborne illness, it would be seriously affecting me. If I contract forborne illness, medication/medical attention would be required. I would still eat at the Restaurant X if the quality of the food is very good. I would still eat at the Restaurant X if the restaurant is my family s/friends favorite restaurant. I would still eat at the Restaurant X if the price of the food is very reasonable. I would still eat at the Restaurant X if I was near the restaurant. I would still eat at the Restaurant X if the restaurant offers high quality service. Number of Description of the violations violations 6 violations Shell eggs in the walk-in refrigerator were found to be at 46 F. A person with an infection was handling food. Dishes and equipment were not sanitized after washing. A bottle of yellow liquid was found in the kitchen with no label and a bottle of orange liquid was found at the front register with no label. An employee used the same gloves to handle a sanitizer soaked towel and then to prepare food. A dishwasher handled clean equipment without properly washing his hands after touching dirty dishes. 8 violations Previous 6 violations plus: Employees were using the same sink to wash their hands and prepare food. Raw chicken was stored over cooked foods in a walk-in refrigerator. 10 violations Previous 8 violations plus: The shelves in the walk-in refrigerator were soiled with food and debris. Containers used to store cabbage and onions were not made of food safe material. Response efficacy Self-efficacy Protection motivation Behavioral intention If I choose a restaurant with a higher inspection score, it will greatly reduce my chances of contracting foodborne illness. Reviewing inspection scores before selecting a place to dine will help me make a wise selection. Understanding the different violation will help me to identify which is a safe restaurant to dine at. I understand inspection reports well enough to gauge the risk associated with eating at selected restaurants. I understand inspection reports well enough to gauge the risk associated with eating at selected restaurants. I want to dine at a restaurant that received high inspection scores. I want to avoid any chances of contracting foodborne illnesses. I want to know current inspection score of other restaurants in the area when selecting a restaurant. I want to do a visual inspection of the public area of the restaurant (dining area, restroom, etc.) to assess food safety. I would not eat at the Restaurant X.

4 J. Choi et al. / Food Control 22 (2011) 862e the potential consequences, then the food safety violation may again not figure in to the restaurant selection process. While it is not possible to anticipate every potential benefit from dining in a particular restaurant, a series of 6 question were developed (see Table 2) based on previous studies and expert opinions. Participants were asked to rate their agreement with the statements from strongly disagree to strongly agree Coping appraisal measures Self-efficacy is an individual s belief in his/her ability to initiate actions recommended to prevent harmful events. As far as this study is concerned, it relates to the individual s ability and willingness to collect information (health inspection reports) understands the information, and develop alternatives to eliminate or mitigate the threat (prevention of foodborne illness). To assess selfefficacy, participants were asked to rate their level of agreement with the three self-efficacy statements in Table 2. These statements were selected from a much longer list because they best measured the self-efficacy construct as it relates to restaurant selection and food safety. Response efficacy is an individual s certainty of the outcome when the proposed action was followed: what are the odds of a person becoming ill if that person does not eat at the restaurant with numerous food safety violations. As with self-efficacy, a long list of statements was developed from the literature and input from industry experts. That list was narrowed down to the three questions shown in Table 2. The focus of those questions respondents belief that health inspection reports are a good source of information on dining safely. In a study measuring the role of cost and response efficacy in persuasiveness of health recommendations (Cismaru, Nagpal, & Krishnamurthy, 2009), cost was defined as the sum of all monetary and non-monetary barriers (Janz & Becker, 1984). Previous study (Cismaru et al., 2009) found that response efficacy and cost jointly affect persuasion following a multiplicative relationship as suggested by Rogers (1975) and Prochaska, Norcross, and Diclemente (1994). There are numerous costs associated with restaurant selection ranging from time to collect and analyze inspection reports to travel time and expense. Many of the costs are almost impossible to accurately measure theoretically without an unrealistically long scenario which locates all other alternatives and their potential costs. It was determined this would make an already lengthy survey so long that the response rate would be too low to get any meaningful information about selection behavior. In addition, with the number of alternative dining choices, additional costs associated with modifying behavior were not likely to have been significant. Therefore, cost modifying behavior was not included in this study Protection motivation Rogers s model showed that protection motivation is a mediating variable between responses to information and coping modes such as action or inhibition action, a single act, repeated acts, multiple acts and repeated multiple acts (Rogers, 1983). In the present study, protection motivation adopts the approach-avoidance motivation concept. Approach motivations are directed towards desirable outcomes whereas avoidance motivations are directed away from undesirable outcomes. Therefore, protection motivation was assessed by asking questions such as I want to avoid any chances of contracting foodborne illnesses in a range from 1 (strongly disagree) to 7 (strongly agree); see Table 2 for the exact wording of protection motivation statements. Table 3 Profile of the Respondents (n ¼ 73). Characteristics Frequency Percentage (%) a Gender Female % Male % Age Under 20 years % 20e % 30e % 40e % 50 years and above 1 1.4% Education level Some High School 0 0.0% High School Diploma or GED 0 0.0% Some College % Graduate Technical or Trade School 2 2.7% Associate s Degree 6 8.2% Bachelor s Degree % Graduate Degree % Marital status Married 6 8.2% Single/divorced 3 4.1% Single/widowed 0 0.0% Single/never been married % Annual income $28,000 or less % $28,001e$58, % $58,001e$88, % $88,001e$118, % $118,001e$148, % Over $148, % a Percentage does not add to 100 because of rounding Behavioral intention measures Behavioral intention was conceptualized as a substitute indicator of actual behavior (Fishbein & Ajzen, 1975). Previous studies found that behavioral intention was affected by perceived vulnerability, severity, self-efficacy, and response efficacy to the message (Prentice- Dunn, McMath, & Cramer, 2009; Umeh, 2004). Behavioral intention Table 4 Health characteristics of the respondents. Health characteristics Frequency (n ¼ 73) % Health condition in general Very poor 0 0 Somewhat poor 0 0 Poor Neither poor nor good Good Very good Excellent Do you have health issues that increase your risk of contracting foodborne illness? Yes No I don t know Have you ever had a foodborne illness? Yes No I don t know Have you ever had a foodborne illness that was caused by a restaurant? Yes No I don t know

5 866 J. Choi et al. / Food Control 22 (2011) 862e868 Table 5 Multiple comparisons of PM variables among four different scenarios for inspection violations using ANOVA (n ¼ 73). Dependent variable No Violation 6 Violations 8 Violations 10 Violations Reliability Vulnerability (2 items) b a a a 0.93 Severity (2 items) Benefits (5 items) b a a a 0.93 Response efficacy (3 items) Self-efficacy (2 items) Protection Motivation (4 items) Note: Means for same dependent variable with same subscripts are not significantly different based on the GameseHowell test at the 0.05 level. was assessed by asking questions such as I would not eat there in a range from one (strongly disagree) to seven (strongly agree): see Table 2 for the exact wording of behavior intention statements. 4. Results The research investigator visited three classes in the Hospitality and Tourism Management Department (Sanitation and Health in Foodservice, Lodging, and Tourism, Lodging Management, and Hospitality and Facilities Management) with approximately 200 students registered for the three classes. Students were encouraged to visit the web link to participate in the survey with an opportunity to win a prize. Survey participants were recruited after class was dismissed. Of the 200 students, 109 responses were collected. Due to incomplete responses, however, only 73 responses were included in the analyses. It seemed that the participants were discouraged by the number of questions. In addition, 3 were eliminated due to the fact that they completed the survey in less than 5 min (two respondents completed the survey in less than 1 min) suggesting that these respondents only wished to participate to be eligible to win the prize and did not read the questions before selecting their responses. As shown in Table 3, all of the respondents were college students with ages between 20 and 29 years old. Approximately 66% were female and 61% responded they have less than $28,000 in annual income. Fifty-six percent of the respondents thought foodborne illness from a restaurant might cause moderate or severe discomfort and 65% of the respondents assumed the illness would last 1 to 4 days. More than 88% of the respondents believed that their health condition was good or excellent and about 86% answered that they did not have any health issues that could increase their risk of contracting a foodborne illness. In addition, about 55% of the respondents had experienced a foodborne illness and about 46% of the respondents thought they had contracted a foodborne illness from a restaurant (see Table 4). Chi-square statistic showed there was a significant difference between previous experience with foodborne illness cause by a restaurant and protection motivation (c 2 (1, N ¼ 63) ¼ 7.78, p ¼ 0.005). Participants who had previously experienced a foodborne illness were more likely to exhibit protection motivation. In this study, the No Violation scenario was used as the control against the other scenarios to evaluate whether violations would generate different scores for the variables used to evaluate protection motivation. Significance was found between no Table 6 PM variables on protection motivation. (n ¼ 59). Predictor B SE b t Constant *** Response efficacy ** Self-efficacy ** Note: F ¼ 12.30***, R 2 ¼ 0.31, Adjusted R 2 ¼ **p < ***p < violations and all other scenarios for two of the variables: vulnerability and benefits (see Table 5). Self-efficacy, response efficacy, cost, and severity did not show any significance among the scenarios. No significance was found between the 6, 8, and 10 violation scenarios for any of the variables. While no significance was found, for some of the variables trends were noted in this pilot study. The vulnerability scores increased slightly as the number of violations increased. Additionally, the response efficacy showed a similar trend. A larger sample is therefore needed to investigate the true significance of both of those trends. All PM variables were regressed on protection motivation. Significant effects were found between protection motivation and response efficacy (Beta ¼ 0.28, t ¼ 3.09) and self-efficacy (Beta ¼ 0.22, t ¼ 2.90). To investigate if PM has direct effects on behavioral intention, the PM variable was regressed on behavioral intention. Table 6 shows that response efficacy and self-efficacy were significant predictors of behavioral intention regardless of number of violations. Protection motivation was significantly related to behavioral intention (F(1, 57) ¼ 18.70, p < 0.001, R 2 ¼ 0.25, adjusted R 2 ¼ 0.24). The results suggest that the higher an individual s protection motivation the more likely the individual will modify behavior based on inspection scores. 5. Conclusions and discussion This study provided some useful preliminary data on behavioral intention of consumers when selecting a restaurant using inspection reports. The pilot showed that the methodology used to collect the data was sound. While the small sample size made it difficult to establish significance for most variables, significance was found for several. While the sample size made it difficult to confirm significance, several trends were noticed. Most of the trends were as expected, but there were a few surprises. As the number of violations increased so did the respondent s perceived vulnerability and response efficacy. Logically, as the number of violations increased so did the participants fear that they were likely to contract a foodborne illness. Equally as logical was the finding that when the number of violations increased so did the response efficacy score. If the likelihood of contacting a foodborne illness by eating at a restaurant increases, then the individual s belief that eating at another restaurant will prevent illness will also increase. There were several variables that did not change with the number of violations; those variables were severity, benefits, and self-efficacy. While vulnerability increased, the severity did not. It appears that the respondent did not think that even though they are more likely to get sick, they are not likely to experience more severe symptoms based on additional violations. Vulnerability and benefits were expected to be inversely related. As the likelihood of becoming ill increases, then the required benefits necessary to make the participant engage in the risky behaviordeating in the restaurantdshould have increased as well. Between zero violations and 10 violations the benefit variable behaved as expected, the perceived benefit of eating at a restaurant with zero violation was greater than that for eating at a restaurant with 10 violations. This

6 J. Choi et al. / Food Control 22 (2011) 862e pattern was not observed between 6, 8, and 8 violation. The reasons why this was not true could not be explained by the small sample. It could be that the difference between the numbers of violations was not significant enough to elicit the expected behavior. Future studies should look a different violation spreads to determine the true impact of the number of violations on this variable. The self-efficacy variable behaved as expected. A person either has the ability to modify behaviordeat in the restaurantdor not. The number of violation may increase a person s willingness to change, but not that person s ability. The protection motivation theory predicts that the variables described above will impact a person s protection motivation. This study found that to be true for only a couple of variables: response efficacy and self-efficacy. Clearly if a person has the ability to change behavior to prevent illness and if that changed behavior is likely to be effective, then that person is more likely to engage in protective behavior. In addition to response efficacy and self-efficacy, respondents who had experienced foodborne illness were more likely to engage in protection motivation behaviors. However, they did not rate their vulnerability or severity of the illness higher as was expected. Severity and vulnerability should have had similar effects on protection motivation, but they did not. Again, the sample size was too small to determine the reason for this unexpected result. One likely cause was once again the spread of the number of violations: the lack of a significant impact of the number of violation on vulnerability between 6, 8, or 10 violations very likely affected vulnerability s impact on protection motivation. The restaurant inspection information triggered respondents response efficacy and self-efficacy which influenced protection motivation of the respondents. It is clear that coping responses were more likely to induce protection motivation than threat responses. Therefore, if the goal of the communication is to trigger protection motivation information related to coping responses must be included. Health inspection reports provide information which relates primarily to threat responses. To provide greater protection to the general public, when inspection reports are published they should include coping responses information. For example, the violation: Employees in the kitchen were not washing their hands between handling raw shrimp and ready to eat foods. could be amended to read: Employees in the kitchen were not washing their hands between handling raw shrimp and ready to eat foods. Choosing a restaurant where employees regularly wash their hands will reduce your risk of contacting a foodborne illness. In addition, words used in the report should be carefully selected to prevent misinterpretation of the results. The results of the study showed that protection motivation was a strong predictor of behavioral intention. Individuals with higher protection motivation are more likely to select a restaurant which they feel will keep them safe from contracting foodborne illness. It is evident that protection motivation induced by a restaurant health inspection report directly influences consumers behavioral intention to dine at the restaurant. Restaurant health inspection processes are designed to communicate with health inspectors and restaurant managers in terms of food safety practices. Therefore, health inspectors need to work closely with restaurant managers to educate them rather than punish them with violations. From this study there is clearly a basis to say that the number of violations will impact variables which impacts a person s protection motivation behavior which in turn impact their behavioral intentions. Therefore, it is evident that foodborne illness outbreaks can damage the reputation of a business and result in decreasing sales. While it is evident that the number of violations will impact behavioral intentions via protection motivation behavior, several limitations with this study impact its ability to predict the exact number of violation required to cause that impact or which statements are most likely to have an impact. One area that should be evaluated in future study is the number of violations for each scenario and the way they are worded. While trends were noted in the data, it appears that there was not enough differentiation between the numbers of violations. One possible explanation is that respondents did not see a significant difference between the unsanitary condition of a restaurant with 8 violations and one with 10. It is recommended that future study should use a wider range in the number of violations such as 3, and 9 or 4 and 10 in the different scenarios. Another area that should be modified is the description associated with each violation. It may be that some of the respondents did not understand the significance of each violation. Previous study recommended that educational intervention would influence college students food safety attitude (Yarrow, Remig, & Higgins, 2009). For example, after educational intervention, respondents mean rating of the statement, If I follow safe food handling practices, my chances of sickness would decrease, increased significantly (Yarrow et al., 2009). However, the results of this study suggested that the respondents did not consider foodborne illness a relevant issue to them. Therefore, providing more information about foodborne illness relevant to one s social or shared interest group would increase one s awareness of the importance of restaurant inspection reports. Additional versions of each scenario should be added to future studies that include different descriptions of the violations. The following are examples of 3 versions of violation scenarios. 1) Employees in the kitchen were not washing their hands between handling raw shrimp and ready to eat foods. 2) Employees in the kitchen were not washing their hands between handling raw shrimp and ready to eat foods. Employees shall clean their hands during food preparation to prevent cross contamination. 3) Employees in the kitchen were not washing their hands between handling raw shrimp and ready to eat foods. Handling potentially hazardous food without washing hands can increase chances of contracting foodborne illnesses. 4) Employees in the kitchen were not washing their hands between handling raw shrimp and ready to eat foods. Choosing a restaurant where employees regularly wash their hands will reduce your risk of contacting a foodborne illness. While this study did not have enough respondents to be analyzed appropriately for significance, it still yielded interesting information for future studies. The results of this study showed that response efficacy and self-efficacy were significant predictors of protection motivation. In addition, protection motivation was a significant predictor of behavioral intention. However, other PM variables remain insignificant predictors of protection motivation. In addition, this was self-reported data and may not reflect the actual behavior of the respondents. Even with those limitations, it appears that protection motivation theory can be a valuable tool to determine the impact of food safety violations on consumer behavior. References Ajzen, I. (1988). Attitudes, personality, and behavior. Milton Keynes, UK: Open University Press. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179e211. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting behavior. Englewood Cliff, NJ: Prentice Hall. Almanza, B. A., Ismail, J., & Mills, J. E. (2002). The impact of publishing foodservice inspection scores. Journal of Foodservice Business Research, 5(4).

7 868 J. Choi et al. / Food Control 22 (2011) 862e868 Almanza, B. A., Nelson, D. C., & Lee, M. L. (2003). Food service health inspectors opinions on the reporting of inspections in the media. Journal of Environmental Health, 65(10), 9e14. Becker, M. H. (Ed.). (1974). The health belief model and personal health behavior. Health Education Monographs, Vol. 2 (pp. 324e473), 4. Boer, H., & Seydel, E. R. (1996). Protection motivation theory. In M. Connor, & P. Norman (Eds.), Predicting health behavior. Buckingham: Open University Press. Cismaru, M., Nagpal, A., & Krishnamurthy, P. (2009). The role of cost and responseefficacy in persuasiveness of health recommendations. Journal of Health Psychology, 14(1), 135e141. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Reading, MA: Addison-Wisely. Floyd, D. L., Prentice-Dunn, S., & Rogers, R. W. (2000). A meta-analysis of research on protection motivation theory. Journal of Applied Social Psychology, 30(2), 407e429. Food Standards Agency (FSA). (2004). Consumer attitudes to food survey London: FSA. Ford, D. S., & Goode, C. R. (1994). African American college students health behaviors and perceptions of related health issues. Journal of American College Health, 42(5), 206e211. Grover, S. F., & Dausch, J. G. (2000). Hepatitis A in the foodservice industry. Food Management, February, 80e88. Henson, S., Majowicz, S., Masakure, O., Sockett, P., Jones, A., Hart, R., et al. (2006). Consumer assessment of the safety of restaurants: the role of inspection notices and other information cues. Journal of Food Safety, 26, 275e301. Houghton, J. R., Kleef, E. V., Rowe, G., & Frewer, L. J. (2006). Consumer perception of the effectiveness of food risk management practices: a cross-cultural study. Health, Risk & Society, 8(2), 165e183. Janz, N. K., & Becker, M. H. (1984). The health belief model: A decade later. Health Education Quarterly, 11, 1e47. Knight, A. J., Worosz, M. R., & Todd, E. C. D. (2007). Serving food safety: consumer perception of food safety at restaurants. International Journal of Contemporary Hospitality Management, 19(6), 476e484. Li-Cohen, A. E., & Bruhn, C. M. (2002). Safety of consumer handling of fresh produce from the time of purchase to the plate: a comprehensive consumer survey. Journal of Food Protection, 65(8), 1287e1296. Martin, K. E., Knabel, S., & Mendenhall, V. (1999). A model train-the-train program for HACCP-based food safety training in the retail/food service industry: an evaluation. Journal of Extension, 37(3). Retrieved: December 28, 2008, from. Milne, S., Sheeran, P., & Orbell, S. (2000). Prediction and intervention in healthrelated behavior: a meta-analytic review of protection motivation theory. Journal of Applied Social Psychology, 30(1), 106e143. Morrone, M., & Rathbun, A. (2003). Health education and food safety behavior in the university setting. Journal of Environmental Health, 65(7), 9e15. Norman, P., Boer, H., & Seydel, E. (2005). Protection motivation theory. In M. Connor, & P. Norman (Eds.), Theory of planned behavior and health behavior (pp. 81e126). England: Open University Press. Pechmann, C., Zhao, G., Goldberg, M. E., & Reibling, E. T. (1993). What to convey in antismoking advertisements for adolescents: the use of protection motivation theory to identify effective message themes. Journal of Marketing, 67, 1e18. Plotnikoff, R., & Hagginbotham, N. (1998). Protection motivation theory and the prediction of exercise and low-fat diet behaviors among Australian cardiac patients. Psychology and Health, 13, 411e429. Pragle, A. S., Harding, A. K., & Mack, J. C. (2007). Food workers perspectives on handwashing behaviors and barriers in the restaurant environment. Journal of Environmental Health, 69(10), 27e32. Prentice-Dunn, S., McMath, B. F., & Cramer, R. J. (2009). Protection motivation theory and stages of change in sun protective behavior. Journal of Health Psychology, 14(2), 297e305. Prochaska, J. O., Norcross, J. C., & Diclemente, C. C. (1994). Changing for good: A revolutionary six-stage program for overcoming bad habits and moving your life positively forward. New York: Quill Harper Collins Publishers. Robert, F., Binkley, M., Nelson, D., & Almanza, B. A. (2005). Transfer of training efficacy in U.S. food safety accreditation. Journal of Culinary Science & Technology, 4(2/3). Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change. Journal of Psychology, 91, 93e114. Rogers, R. W. (1983). Cognitive and physiological processes in fear appeals and attitude change: a revised theory of protection motivation. In J. Cacioppo, & R. Petty (Eds.), Social Psychophysiology. New York: Guilford Press. Tulloch, H., Reida, R., D Angeloa, M. S., Plotnkoff, R. C., Morrina, L., Beatona, L., et al. (2009). Predicting short and long-term exercise intentions and behavior in patients with coronary artery disease: a test of protection motivation theory. Psychology and Health, 24(3), 255e269. Umeh, K. (2004). Cognitive appraisals, maladaptive coping, and past behavior in protection motivation. Psychology and Health, 19(6), 719e735. Van der Velde, F. W., & Van der Plight, J. (1991). AIDS-related health behavior: coping, protection, motivation, and previous behavior. Journal of Behavioral Medicine, 14, 429e451. Wheeler, C., Vogt, T. M., Armstrong, G. L., Vaughan, G., Weltman, A., Nainan, O. V., et al. (2005). An outbreak of Hepatitis A associated with green onions. New England Journal of Medicine, 353, 870e897. Yarrow, L., Remig, V. M., & Higgins, M. M. (2009). Food safety educational intervention positively influences college students food safety attitudes, beliefs, knowledge, and self-reported practices. Journal of Environmental Health, 71(6), 30e35.