Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Bio-based and Applied Economics BAE © 2024 Author(s). Open access article published, except where otherwise noted, by Firenze University Press under CC-BY-4.0 License for content and CC0 1.0 Universal for metadata. Firenze University Press | www.fupress.com/bae Citation: Sogari, G., Wongprawmas, R., Andreani, G., Lefebvre, M., Pel- legrini, N., Gómez, M.I., Mora, C., & Menozzi, D. (2024). Intention and behavior toward eating whole grain pasta on a college dining campus: Theory of Planned Behavior and message fram- ing. Bio-based and Applied Economics 13(3): 301-316. doi: 10.36253/bae-15456 Received: November 28, 2023 Accepted: March 15, 2024 Published: October 16, 2024 Data Availability Statement: All rel- evant data are within the paper and its Supporting Information files. Competing Interests: The Author(s) declare(s) no conflict of interest. Editor: Francesco Pagliacci, Valentina Raimondi, Luca Salvatici ORCID GS: 0000-0002-2561-571X RW: 0000-0002-4897-7414 GA: 0000-0001-8660-7732 NP: 0000-0002-9178-5274 MIG: 0000-0002-3591-5249 CM: 0000-0003-3180-5496 DM: 0000-0002-5241-1587 Intention and behavior toward eating whole grain pasta on a college dining campus: Theory of Planned Behavior and message framing Giovanni Sogari1, Rungsaran Wongprawmas1, Giulia Andreani1, Michele Lefebvre2, Nicoletta Pellegrini3, Miguel I. Gómez4, Cristina Mora1, Davide Menozzi1,* 1 Department of Food and Drug, University of Parma, Parco Area delle Scienze 47/A, 43124 Parma, Italy 2 Director, Nutrition Management, Cornell University, Ithaca, NY 14850, USA 3 Department of Agricultural, Food, Environmental and Animal Sciences, University of Udine, Via Sondrio 9 2/A, 33100 Udine, Italy 4 Charles H. Dyson School of Applied Economics and Management, Cornell University, Ithaca, NY 14850, USA * Corresponding author. E-mail: davide.menozzi@unipr.it Abstract. The consumption of whole grains has several health benefits, however, most US consumers – including young adults – do not meet the recommended consump- tion intake. To understand the underlying factors affecting the intention and consump- tion of whole grain pasta, a survey based on the Theory of Planned Behavior (TPB) was developed and administered to US college students. For four weeks, participants (n = 325) either did not receive any information (control) or received weekly messages on the health benefits of whole grain pasta (e.g., high fiber and niacin contents) in the forms of gain- (treatment 1) or loss-framed (treatment 2) information. Variables of the TPB mod- el and consumers’ perceptions were investigated both at Time 1, when the first message was received (week 0), and at Time 2, one month after the intervention (week 4). Results from the two moments were compared. We found that the TPB measures and perceived usefulness were not influenced by the treatment group; however, the gain-framed mes- sage engendered greater message engagement than the loss-framed one. Finally, results from the structural equation model showed that attitude, subjective norms, and per- ceived behavioral control were positively associated with the intention to consume whole grain pasta, and the intention was a strong determinant of participants’ behavior. Based on our results, implications and suggestions for future studies are discussed. Keywords: dietary fiber, message framing, gain-framed, loss-framed, message engage- ment. JEL Codes: I12, D91, D83. 1. INTRODUCTION Substantial socio-environmental changes from adolescence to college can be challenging for many young adults (Christoph, Ellison, & Meador, 2016). https://doi.org/10.36253/bae-15456 https://creativecommons.org/licenses/by/4.0/legalcode https://creativecommons.org/publicdomain/zero/1.0/legalcode http://www.fupress.com/bae https://doi.org/10.36253/bae-15456 https://orcid.org/0000-0002-2561-571X https://orcid.org/0000-0002-4897-7414 https://orcid.org/0000-0001-8660-7732 https://orcid.org/0000-0002-9178-5274 https://orcid.org/0000-0002-3591-5249 https://orcid.org/0000-0003-3180-5496 https://orcid.org/0000-0002-5241-1587 mailto:davide.menozzi@unipr.it 302 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. In a situation in which young adults are now faced with making their own dietary choices, this transition is often associated with unhealthy eating habits (Quick, Wall, Larson, Haines, & Neumark-Sztainer, 2013; Stok, Renner, Clarys, & Deliens, 2018), which can contribute to overweight and obesity and other diet-related dis- eases (Kann et al., 2018; World Health Organization, 2014). Therefore, campus dining programs are working to change the perception of nutrition and healthy eat- ing within their food eateries (Franchini, Biasini, Rosi, & Scazzina, 2023). From new and innovative design strategies and different approaches to healthy menus to the inclusion of more produce, many campus dining programs have tested and used health principles and guidelines to nudge customers’ decisions (Andreani, Sogari, Wongprawmas, Menozzi, & Mora, 2023). One example comes from the US-based Menus of Change program. Menus of Change, founded in 2012 by the CIA and Harvard School of Public Health, is an initiative to achieve healthy and sustainable menus, with the tagline “The Business of Healthy, Sustainable, and Delicious Food Choices”. Menus of Change University Research Collaborative (MCURC) was established with working groups of scholars and campus dining leaders interested in using college and university dining as a platform to establish and accelerate efforts to move campus diners towards healthy menus. Healthy eating habits should include high consump- tion of food considered to be healthy, such as fruit, veg- etables, and other high-fiber options, such as whole grains 1 and legumes (U.S. Department of Health and Human Services and U.S. Department of Agriculture & US Department of Agriculture, 2015). Among healthy food choices, whole grain intake is a pivotal aspect to be considered in weight management and overall health of young adulthood, which helps in overweight and obesity prevention (Quick et al., 2013). Grains, including whole grains, are staple foods in many countries of the world (European Commission, 2019) and can be consumed as single foods (e.g., rice, oatmeal), or included as an ingredient in many food products (e.g., breads, cereals, crackers, and pasta) (U.S. Department of Health and Human Services and U.S. Department of Agriculture & US Department of Agri- culture, 2015). Evidence showed that higher consump- tion of whole grains and dietary fiber is inversely asso- 1 “Grains and grain products made from the entire grain seed, usually called the kernel, which consists of the bran, germ, and endosperm. If the kernel has been cracked, crushed, or flaked, it must retain the same relative proportions of bran, germ, and endosperm as the original grain in order to be called whole grain. Many, but not all, whole grains are also sources of dietary fiber.” (U.S. Department of Health and Human Services and U.S. Department of Agriculture, 2015, pag. 96). ciated with the risk of obesity and weight gain (Maki et al., 2019; Slavin, 2005), type 2 diabetes mellitus and cardiovascular disease (Ye, Chacko, Chou, Kugizaki, & Liu, 2012). Because of the health benefits linked to die- tary fiber (see Jones & Engleson, 2010 for a more com- prehensive review), governmental institutions and nutri- tional experts have developed nutrition education and health promotion campaigns to recommend the inclu- sion of whole grains in the diet (Jones & Engleson, 2010; Marquart, Wiemer, Jones, & Jacob, 2003; Shepherd et al., 2012). For instance, the 2015–2020 Dietary Guide- lines for Americans suggests that a healthy eating pat- tern should include grains, at least half of which should be from whole grains (U.S. Department of Health and Human Services and U.S. Department of Agriculture & US Department of Agriculture, 2015). Previous reserach (e.g., Wongprawmas et al., 2021) indicates that the availability of whole grain options at comparable prices to conventional ones could be ben- eficial for students since it may mitigate consumption barriers such as availability and price (Meynier, Chan- son-Rollé, & Riou, 2020). Moreover, another barrier to consuming whole grain products is consumers’ negative perceptions of their sensory attributes (i.e., taste and tex- ture) (Bisanz & Krogstrand, 2007; Dammann, Hauge, Rosen, Schroeder, & Marquart, 2013). Despite the relevance whole grains have in a healthy diet, limited research (Ugunesh, Siau, Mohd Sanip, & Koo, 2023; Weingarten & Hartmann, 2023) has investi- gated the links between consumer attitudes, intention, and behavior to consume whole grain foods, especially among young adults. Therefore, we tested the Theory of Planned Behavior (TPB) (Ajzen, 1991) – which is an expectancy-value model of behavior change – to meas- ure the variables influencing the consumption of whole grain pasta. The TPB model postulates that behavioral intention is the central determinant of behavior. Pre- vious systematic reviews have demonstrated that the TPB and similar psycho-social theories (e.g. the Theory of Reasoned Action, TRA) can serve as reliable tools for predicting sustainable (e.g., Biasini et al., 2021) and health-promoting behaviours (e.g., McEachan, Conner, Taylor, & Lawton, 2011), including healthy eating behav- iours (e.g., McDermott et al., 2015). These reviews have shown that, in general, attitude towards the behaviour is the most significant predictor of intention, and intention is the most significant predictor of behaviour (McDer- mott et al., 2015; Biasini et al., 2021). Biasini et al. (2021) observed a wide range of explained variance in inten- tion (7–87%) and/or behaviour (3–81%) across different applied models and study designs. As suggested by these authors, longitudinal studies can provide a prospective 303 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Intention and behavior toward eating whole grain pasta on a college dining campus prediction analysing the causal relationship between dependent and independent variables, which would be otherwise precluded in cross-sectional investigations (McEachan et al., 2011; Biasini et al., 2021). Based on these considerations, first, the model we tested hypothesizes that the intention to include whole grain pasta in the diet is influenced by the atti- tude (a person’s favorable or unfavorable evaluation of the behavior), the subjective norms (what other people think one should do), and the perceived behavioral con- trol (the perceived ease or difficulty of performing the behavior). Second, we hypothesized that the prospective behavior (actually eating whole grain pasta), measured after four weeks (Time 2), is determined by the inten- tion and perceived behavioral control. Figure 1 shows the theoretical framework. In addition, past studies sug- gest that whole grain food consumption could be pro- moted by using positive information about its health benefits presented at the point of consumption. One study by Sogari et al. (2019) found that a psychologi- cal health benefit (i.e., vitamin benefits reduce fatigue) related to whole grain foods significantly increased the number of individuals preferring whole grain vs. regular pasta. Another study by Weingarten and Hart- mann (2023) showed that repeated exposure to positive information about the health benefits of whole grain increased attitudes and led to higher intentions to con- sume such products. Therefore, the use of health claims and messages to encourage the consumption of whole grain pasta over regular pasta is one communication strategy that could support the shift toward a healthy eating pattern. Based on this evidence, it is relevant to understand the effectiveness of different communi- cation strategies on the attitude towards whole grain options in terms of the framing effect, i.e. decisions are influenced by the way the outcomes are presented (Dol- gopolova, Li, Pirhonen, & Roosen, 2022). Meta-analysis results have recently indicated that product attributes framed as gains have a higher effect on attitudes and intentions than product attributes framed as losses (Dolgopolova et al., 2022). Other researches have indi- cated that encouraging positive behaviors by evoking loss aversion is not necessarily a guiding principle when it comes to health benefits (e.g., Gallagher & Upde- graff, 2012). Dolgopolova et al. (2022) have suggested that loss-framed messages are mainly effective when it comes to decisions involving significant risk, and that food choices are not associated with an immediate high level of risk. Thus, a secondary aim of our study is to understand whether providing information on the health benefits of whole grains, under two different framing conditions (gain vs. loss-framed), would influ- ence the TPB measures as well as other variables (e.g., perceived usefulness of eating whole grain pasta). 2. METHODS 2.1. Data Collection and the Sample Data collection was carried out across several dining halls at Cornell University, Ithaca, NY, US in spring 2019. An online questionnaire was distributed using the Qual- trics LLC platform (Provo, US), and included attitudinal and motivational items derived from the TPB framework, as well as questions on overall eating habits. Some sur- vey sections, including the message frames, were revised to improve the clarity of their meaning and reduce the total survey length to approximately 12 minutes. The entire survey was pre-tested with 50 students and Fac- ulty staff members. The data collection took place dur- ing dinner time in front of the pasta station in a dining setting (Time 1). A final sample of 499 college students (female 53.6%, mean age 18.8y), all pasta consumers, par- ticipated in this study. Participants mostly had a healthy weight range (Body Mass Index between 18.5 and 24.9), were mainly omnivores with a slightly high proportion of flexitarian and vegan or vegetarian, and only 10% had dietary or healthy restrictions. Table 1 shows the full set of socio-demographics of the participants. One month after Time 1 (Time 2), a follow-up ques- tionnaire was sent via email to all the participants in order to evaluate whether any changes in their attitudi- nal variables occurred and to assess the reported con- sumption behavior of eating whole grain pasta over the last month. Most of the participants returned the elec- tronic questionnaire on the day they received it, and few of them completed it in the following days. A final sam- ple of 325 respondents returned the questionnaire. The full survey flow (Time 1 and Time 2) is shown in Fig- ure A1 in the Appendix. The two surveys at the two time Figure 1. Theoretical framework of the Theory of Planned Behavior (TPB) in Time 1 (main survey in week 0) and Time 2 (follow-up survey after 4 weeks). 304 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. points were linked through the student ID number. Fol- lowing the completion of the study, participants received a monetary compensation of $5. The study was approved by the Institutional Review Board (IRB) of the Office of Research Integrity and Assurance of Cornell University (Protocol Number: 1810008359). 2.2. Measures The main survey (Time 1) consisted of three sec- tions. The first section included the message or framing treatment (control, gain-framed, and loss-framed mes- sages) – details are reported in section 2.3. In the two treatment groups, the participants were asked to care- fully read the information provided. The second section was structured to measure the various components of the TPB (Ajzen, 1991) and other factors in relation to the participant’s behavior of including whole grain pasta in the diet over the next month (for details see Table A1). The TPB survey items and the health claims were based on a review of the existing literature (Fishbein & Ajzen, 2011) followed by a revision by two nutrition experts as well as three experts in social sciences. Finally, the third section of the survey included socio-demographic data (i.e., participants’ age, gender, and Body Mass Index2), self-perception of overall health, physical exercise, eating behavior, and dietary/healthy restrictions. For the TPB section, all measures were assessed using a 7-point scale, from strongly disagree (1) to strongly agree (7). Two items measured the Perceived Behavioral Control (PBC), which is related to the con- trol of performing the behavior. Three items assessed the Subjective Norms (SN), which is an individual’s per- ception of social pressure on the way a person should or should not demonstrate a specific behavior. Attitude towards the behavior (ATT), which represent the degree of a favorable or unfavorable evaluation of a specific behavior, was based on two items about the likelihood that consuming whole grain pasta would result in per- sonal beliefs (i.e., tasty, easy). Behavioral Intention (INT) is the willingness of an individual to perform a specific behavior and it was measured using three items. The factors of the TPB model have prior determi- nants: ATT is guided by behavioral beliefs about the likely consequences of performing the behavior, SN is driven by the normative beliefs about the opinions/ expectations of important others, and PBC is influenced 2 The body mass index, abbreviated as BMI, is a measure of a person’s weight relative to height that correlates well with body fat (Eurostat, 2017). A person is considered underweight if they have a BMI below 18.5, normal weight between 18.5-24.9, and overweight if they have a BMI greater than or equal to 25. by the control beliefs about barriers and facilitators to perform that behavior (Fishbein & Ajzen, 2011). All these beliefs (n=12) were measured using a 7-point Lik- ert scale from strongly disagree (1) to strongly agree (7). In addition, we asked about the perceived usefulness of whole grain pasta, which measured subjects’ percep- tions of performance and effectiveness gains from eating whole grain pasta (e.g., stay in shape, improve work per- formance) by using three 7-point Likert scaled items. Two factors were also used to evaluate the quality of the messages provided in the two information condi- tions. The first factor was the consumer evaluation of the message (Hung & Verbeke, 2019), which was based on five items with a 7-point Likert scale, to measure several characteristics of the health claim, including familiarity, understandability, credibility, interest, and importance. The second factor was the argument quality of the mes- sage (Bhattacherjee & Sanford, 2006), which was used to measure whether the information provided was help- ful, valuable, informative, and persuasive, by using four 7-point Likert scaled items. Four weeks after the initial survey (Time 2), partici- pants’ behavior was also assessed by using two measures of reported behavior using a 7-point scale (Fishbein & Ajzen, 2011). In the first item, respondents were asked to indicate how frequently they consume whole grain pasta, on aver- age, ranging from ‘never’ to ‘almost every day’. In the sec- ond item, participants were asked whether they had includ- ed whole grain pasta in their diet at least once over the past month. In addition, attitude, intention, and perceived usefulness were measured again in Time 2 using the same items as in Time 1. Note that all canteens on the Campus offer whole grain dishes daily; therefore, product availabil- ity is not a barrier for the participants. 2.3. Intervention with health messages At the beginning of the study, participants were ran- domly assigned to either a no-information group (control, n=100) or one of the two treatment groups, namely gain- framed (n=202) or loss-framed (n=197) messages. Stu- dents in the gain or loss-framed treatment received four messages about whole grain pasta health benefits. The health benefits were adapted by authorized health claims. In the US, a food-related health claim3 must be approved by public authorities (i.e. the Food and Drug 3 “Health claim means any claim made on the label or in labeling of a food, including a dietary supplement, that expressly or by implication, including “third party” references, written statements (e.g., a brand name including a term such as “heart”), symbols (e.g., a heart symbol), or vignettes, characterizes the relationship of any substance to a disease or health-related condition.” (Food and Drug Administration, 2023). 305 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Intention and behavior toward eating whole grain pasta on a college dining campus Administration, FDA) and must be supported by a significant body of research showing the relation- ship between the food/constituent and a health effect in humans. Based on this context, four specific health claims related to whole grains were considered (Table A2). Moreover, following previous works (see Deliens et al., 2016 for a systematic review) a media-based approach was used to communicate such expected healthy ben- efits. In our study, we decided to use health claims in the form of messages considering both general benefits of whole grain foods (e.g., fibers have positive effects on weight management) and more specific ones (e.g., the relationship between fibers and gut health or bowel function) (EFSA Panel on Dietetic Products, Nutrition, and Allergies (NDA), 2010). For the two treatment groups, we decided to convey identical information but differently framed in terms of gains or losses associated with an expected outcome (Dolgopolova et al., 2022). A gain-framed message might take the form of ‘‘If you perform the advocated action, desirable outcome X will be obtained’’, whereas a loss- framed message might be “If you do not perform the advocated action, desirable outcome X will be avoided” (O’Keefe & Jensen, 2008). The rationale is that one type of framing may be more effective than another at pro- moting health behavioral change (Gallagher & Upde- graff, 2012). Participants in the treatment groups read a similar health message that differentiated for details of either the benefits of including whole grain pasta (gain- framed), or the health dangers of not including whole grain pasta (loss-framed). In addition, participants in the two treatment groups received four emails (one per week) that includ- ed a different health claim message, still considering the same framework group (gain-framed or loss-framed) and were blinded to the other intervention. Thanks to the online platform used to send out per- sonalized emails (mailchimp.com), we were able to elec- Table 1. Socio-demographic characteristics, lifestyle variables, and health-related factors reported for the total sample and by the groups at Time 1. Variables All Information treatments p-value Control Gained-frame Loss-frame N 499 100 202 197 % 100 20.0 40.5 39.5 Age1 (mean, sd) 18.8 (1.16) 18.6 (1.13) 18.9 (1.16) 18.8 (1.17) 0.267 Gender2 0.451 Male 44.4 41.0 47.5 42.9 Female 53.6 59.0 49.5 55.1 Others or prefer not to answer 2.0 0.0 3.0 2.0 BMI1 22.9 (5.79) 22.0 (5.00) 23.1 (5.77) 23.1 (6.16) 0.267 Eating behavior2 0.357 Omnivore 80.1 79.0 83.2 77.6 Vegetarian 6.6 6.0 4.0 9.7 Vegan 3.2 3.0 3.0 3.6 Flexitarian 8.8 11.0 7.9 8.7 Others 1.2 1.0 2.0 0.5 Dietary/Healthy restrictions2 0.461 Yes 10.4 10.0 8.9 12.2 No 87.8 90.0 88.6 85.7 Prefer not to answer 1.8 0.0 2.5 2.0 Self-perception of overall health3 5.0 (4.0-6.0) 6.0 (4.2-6.0) 5.0 (4.0-6.0) 5.0 (4.0-6.0) 0.145 Physical excercise3 4.0 (3.0-5.0) 4.0 (3.0-5.0) 4.0 (3.0-5.0) 4.0 (3.0-5.0) 0.255 Note: Data are presented as the mean (SDs) for continuous variables, as number (%) for nominal variables, and as the median (IQRs) for categorical variables. SDs = standard deviations. IQRs = Interquartile ranges. BMI: Body Mass Index. N = 498 for age, gen- der, eating behavior, dietary/healthy restrictions, self-perception of overall health; N=481 for BMI; and N=495 for physical exercise. 1ANOVA. 2Pearson chi-square. 3Kruskal–Wallis Test. Self-perception of overall health: How healthy do you consider yourself? (from very bad = 1 to very well= 7) Physical exercise: How often do you usually engage in physical exercise (30 minutes of exercise)? (from never = 1 to more than 3 times per week = 5. They can choose “I do not want to answer”). http://mailchimp.com 306 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. tronically assess whether the recipient opened the email with the health claim message. For those who did not open it, a reminder was sent the following day. How- ever, we cannot be sure whether the participants actu- ally read the text incorporated in the email. The infor- mation sent via email was different every week to avoid the boredom of reading the same message and the risk of dropping out of the study. The messages were sent to participants in a random order. In this way, the subjects were exposed to all four types of claims (see Table A2) in order to have a broader knowledge of the several benefi- cial roles of eating whole grain food. 2.4. Data Analysis Descriptive statistics were used to report the per- centages, median, means, and standard deviations. One- way ANOVA, Pearson Chi-square, and Kruskal-Wallis tests for independent samples were performed in order to determine the existence of significant differences between the control and treatment groups regarding the socio-demographic data, lifestyle variables, and health- related factors. The internal consistency, validity, and reliability of ATT, SN, PBC, INT, and Perceived Usefulness (PU) fac- tors were tested using Cronbach’s alpha, factor loadings (λ), and composite reliability (CR), respectively, and con- sidering all participants at each time point (Time 1 and Time 2). Discriminant validity was tested by comparing the square root of the AVE of each construct with the inter-construct correlation (Bagozzi & Yi, 2012). Then, the internal consistency was assessed for each factor at each time point in all groups. Almost all of Cronbach’s alphas of each factor at each time point were above the accept- able threshold (α > .60) (van Griethuijsen et al., 2015). Eleven composite variables were created by averaging the items within each factor (Table 2). Details of the internal consistency of each factor of the TPB model and other variables in Time 1 and Time 2 are presented in Table A3. One-way ANOVA tests were used to analyze the impacts of different health claim messages as well as the effects of providing information under two differ- ent framing scenarios (gain vs. loss-framed) on the TPB measures. Repeated measures ANOVA was used to examine the interaction of time and information treatments on attitude, intentions, and perceived usefulness at base- line (week 0) and week 4. The results indicated that there were no different effects between the control and the framings nor differences among health claim messages. Therefore, the following Structural Equation Mod- elling (SEM) model analysis was performed on the total sample without separating groups according to the framings. A SEM approach was used to test the theo- retical framework presented in Figure 1. SEM allows the specification of a model with both latent (e.g., attitude towards including whole grain pasta in the diet) and observed variables (e.g., the questionnaire items) (Kline, 2016). The latent variables, namely the abstract phenom- ena that cannot be directly measured by the researcher, have been analyzed using confirmatory factor analy- sis (Byrne, 2010). Confirmatory Factor Analysis (CFA), often referred to as the measurement model, is used when the researcher has some knowledge of the underly- ing latent variable structure or wishes to evaluate a prio- ri hypotheses driven by theory. In our case, to improve the overall goodness-of-fit of the model, we decided to apply the latent variable structure for all TPB variables but PBC, for which we used the observed averaged varia- ble. The goodness-of-fit of the models was assessed using χ2 and their degrees of freedom (df), Tucker-Lewis Index (TLI), comparative fit index (CFI), root mean square error of approximation (RMSEA) with a 90% confidence interval, and the standardized root mean square residual (SRMR). Statistical analysis was performed using SPSS v.28.0 and AMOS v.27.0 statistical software (IBM Corpo- ration, Armonk, NY, USA). 3. RESULTS 3.1. Descriptive statistics Table 2 shows the descriptive statistics of the latent and observable variables: the factor loadings of the vari- ables items (λ) above 0.50, CR values above 0.70, Cron- bach’s α above 0.70 with the only exception of PBC (0.62), and AVE values above 0.50 show strong reliabil- ity, and convergent validity of all factors in the meas- urement model. The results demonstrate a moderately positive consumer attitude toward including whole grain pasta in their diet (mean score: 4.75). Nevertheless, sub- jective norms did not show to greatly influence consum- ers (3.57) whereas they reported relatively strong control over the behavior (5.49). Again, consumers exhibited a moderately positive intention to include whole grain pasta in their diet (4.23). In general, participants report- ed consuming whole grain pasta occasionally (4.63). As shown in Table 3, the squared root of the AVE of each construct was greater than the Spearman’s rank- order correlation (ρ) between the constructs, which also indicates the discriminant validity of the model. We also tested the effects of information (gain vs. loss-framed) on the TPB constructs and other variables in Time 1 and Time 2 (see details in Appendix Table 307 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Intention and behavior toward eating whole grain pasta on a college dining campus A3). No significant differences between control, gain- and loss-framed groups were found for the TPB meas- ures and PU, neither in Time 1 nor Time 2. Regarding how participants evaluate the type of message and the quality of the argument, significant differences were found between the gain- and loss-framed condition. The gain-framed message was found to slightly but signifi- cantly engender greater message engagement in terms of overall evaluation (M= 4.86) and quality of the message (M=4.77) than the loss-framed message (overall evalua- tion: M=4.16, and quality of the message: M = 3.96). Interestingly, the results of repeated measures ANO- VA (Table 4) suggested that time (Time 1 vs. Time 2) had a positive impact on perceived usefulness (p < 0.001), intention (p < 0.001) and attitude (p = 0.006). Neverthe- less, there was no significant effect of the interaction of time and treatments (framing) for perceived useful- ness (Wilks lambda = 0.99, F =2.41, p = 0.092), intention (Wilks lambda = 0.99, F =1.10, p = 0.334) and attitude (Wilks lambda = 0.99, F =0.42, p = 0.659). The explana- tion for this finding could be that the request to fill out a follow-up questionnaire in the control group might have positively affected the perceived usefulness of and inten- tion to consume whole grain pasta in Time 2. 3.2. Effect of beliefs The correlations (ρ) between behavioral, normative, and control beliefs with their relative constructs (atti- tudes, subjective norms, and PBC, respectively), inten- tion to eat whole grain pasta over the next month, and behavior are reported in Table 5. Intermediate correlation levels (ρ = 0.40–0.70) are reported for the association of normative beliefs with subjective norms and behavioral beliefs with attitude to Table 2. Mean values (standard deviation, SD) of single items and TPB constructs, factor loadings (λ), composite reliability (CR), average variance extracted (AVE) and Cronbach’s α of the total sample (N=499) and follow-up (N=325). N Mean (SD) λ CR AVE α Time 1 Attitude (Including whole grain pasta in my diet over the next month will be) 499 4.75 (1.48) 0.74 0.59 0.70 Difficult/Easy 499 4.98 (1.67) 0.59 Not tasty/Tasty 499 4.51 (1.72) 0.92 Subjective norm 499 3.57 (1.41) 0.92 0.79 0.90 Most people who are important to me think that I should include whole grain pasta in my diet over the next month 499 3.69 (1.54) 0.95 Most people who influence my decisions think that I should include whole grain pasta in my diet over the next month 499 3.61 (1.45) 0.93 It is expected that I should include whole grain pasta in my diet over the next month 499 3.41 (1.65) 0.78 Perceived behavioral control 499 5.49 (1.13) 0.84 0.72 0.62 I believe that including whole grain pasta in my diet over the next month is possible 499 5.43 (1.33) 0.85 The decision to include whole grain pasta in my diet over the next month will be only up to me 499 5.56 (1.34) 0.85 Intention 499 4.23 (1.55) 0.91 0.77 0.91 I intend to include whole grain pasta in my diet over the next month 499 4.40 (1.64) 0.89 I will try in anyway to include whole grain pasta in my diet over the next month 499 4.25 (1.68) 0.84 I will definitely include whole grain pasta in my diet over the next month 499 4.03 (1.74) 0.89 Follow Up (Time 2) Behavior 325 4.63 (1.71) 0.77 0.62 0.76 In the past month, how often have you included a meal with whole grain pasta in your diet? 325 3.84 (1.70) 0.83 I have included whole grain pasta in my diet at least once in the past month 325 5.42 (2.10) 0.75 Table 3. Spearman’s rank-order correlations (ρ) between the TPB constructs including the squared root of the AVE of each construct (reported in bold). ATT SN PBC INT BEH ATT 0.77 0.22*** 0.30*** 0.45*** 0.32*** SN 0.89 n.s. 0.58*** 0.31*** PBC 0.85 0.25*** 0.16** INT 0.88 0.55*** BEH 0.79 Note: ATT = attitudes; SN = subjective norms; PBC = perceived behavioral control; INT = Intentions; BEH = behavior; *** indicates significance at p<0.001, ** significant at p<0.01, ns=not significant 308 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. eating whole grain pasta (ρ = 0.40). In particular, par- ents’ and friends’/partners’ opinions are the two norma- tive beliefs that primarily affect subjective norms and intention. Regarding behavioral beliefs, the two most rel- evant beliefs associated with eating whole grain pasta are a long-term investment for the individual and less diet- related diseases. Control beliefs are negatively associated with PBC, in particular, for the higher costs of whole grain pasta and the perceived lack of availability in the dining halls. These represent the main barriers that decrease the perceived ability of respondents to perform the behavior. Finally, the link between control beliefs and intention has positive values, although it is almost non-significant. The effect of the beliefs on behavior is less relevant (ρ ≤ 0.30) and significant only for normative and behav- ioral beliefs. 3.3. Structural equation model results The results of the SEM analysis with standardized path coefficients and R2 are reported in Figure 2, while the unstandardized coefficients and standard errors are reported in Table A4. The SEM analysis was performed on the entire sample because framing had no effect on the TPB measures. The results show that there is a satis- factory fit between the hypothesized model and the data (χ2 (df) = 112.61 (37); CFI = 0.975; TLI = 0.955; RMSEA (90% C.I.) = 0.064 (0.051–0.078)). Overall, the TPB mod- el explains 53.2% of the variance for the intention to consume whole grain pasta over the next month (meas- ured in Time 1), and 44.5% of the variance in the self- reported behavior measured in Time 2. Attitude, subjec- tive norms, and perceived behavioral control are signifi- cant predictors of the intention to consume whole grain pasta over the coming month. Specifically, subjective norms (β = 0.50, p < 0.001) and attitude (β = 0.36, p < 0.001) have a greater influence on the intention than the PBC (β = 0.16, p < 0.001). The intention is also a strong determinant of the behavior to consume whole grain pasta (β = 0.68, p < 0.001), measured after four weeks (self-reported behavior). 4. DISCUSSION AND CONCLUSIONS Understanding how the behavior towards the inclu- sion of whole grain products is formed becomes a cru- cial stage to develop efficient healthy food choice strate- gies. In our study, the TPB model provides a significant explanation for the variance of the intention to consume whole grain pasta over the next month (R2=0.53), as well as the (self-reported) prospective behavior (R2=0.45). Thus, our results of the TPB model show that when individuals have strong attitudes, subjective norms, and perceived behavioral control toward eating whole grain pasta, their intention to eat this product increases, and this higher motivation would be strongly associated with Table 4. Results of repeated measures ANOVA. Variables Times Wilks lambda F Partial eta squared p-valueTime 1 Time 2 M SD M SD ATT (N = 325) 5.20 1.46 5.52 1.34 0.95 7.73 0.05 0.006 PU (N = 325) 4.35 1.10 4.84 1.12 0.86 51.99 0.14 <0.001 INT (N = 325) 4.18 1.55 4.41 1.49 0.96 13.70 0.04 <0.001 Note: ATT = Attitude; PU = Perceived usefulness; INT = Intentions; M = Means; SD = Standard Deviation. Table 5. Spearman’s rank order correlations (ρ) between beliefs and their respective direct measure (attitude, subjective norm, and per- ceived behavioral control – PBC), intention, and behavior. Beliefs ρ Sig. ρ Sig. ρ Sig. Control beliefs PBC Intention Behavior ConBel1 0.11 ** 0.11 ** 0.03 ns ConBel2 -0.28 *** 0.08 * -0.03 ns ConBel3 -0.28 *** 0.20 *** 0.06 ns ConBel4 -0.09 ** 0.08 * 0.06 ns Behavioral beliefs Attitude Intention Behavior BehBel1 0.40 *** 0.40 *** 0.24 *** BehBel2 0.43 *** 0.38 *** 0.18 ** BehBel3 0.45 *** 0.42 *** 0.22 *** Normative beliefs Subjective norms Intention Behavior NorBel1 0.62 *** 0.41 *** 0.14 ** NorBel2 0.66 *** 0.41 *** 0.18 ** NorBel3 0.45 *** 0.32 *** 0.08 ns NorBel4 0.51 *** 0.35 *** 0.09 * NorBel5 0.54 *** 0.33 *** 0.17 ** 309 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Intention and behavior toward eating whole grain pasta on a college dining campus the actual behavior. Similar results were found in other studies with regard to healthy dieting; for instance, in studies conducted by Hagger et al. (2006), the applied models explained 69% (Hagger & Chatzisarantis, 2006) and 56% (Hagger, Chatzisarantis, & Harris, 2006) of the intention, with relatively high variability in the explained behavior (66% and 32%, respectively). In line with previous studies (Biasini, Rosi, Scazzina, & Menoz- zi, 2023; Sogari et al., 2022), the intention well predicts young adults’ behavior. In particular, subjective norms (i.e., the perceived social influence) affected the intention more than the attitude and PBC (Li, Long, Laubayeva, Cai, & Zhu, 2020). Usually, adolescents or young adults are more influenced by social and peers than other age groups, and this may explain why subjective norms have a stronger influence on intention in the TPB model (Bar- beris, Gugliandolo, Costa, & Cannavò, 2022; Friedman et al., 2022). In our case, the effect of behaviors of other students in the canteen (the social context) might affect the participant’s motivation to comply. Providing health messages at the point of consump- tion could, however, steer consumer decisions and be an effective method of delivering strategies to increase healthy eating. A message can be framed either to promote the advantages of consuming a particular food (gain-framed) or to stress the negative outcomes of not consuming that particular food (loss-framed) (Gallagher & Updegraff, 2012). The success of various message-framing strategies is usually assessed by measuring consumer behaviors, inten- tions, or attitudes (Dolgopolova et al., 2022). Our findings show no effect of frame condition on the TPB measures in Time 1. This is in line with a review by Gallagher & Updegraff (2012) that showed no significant effect of framing on attitudes and inten- tions. Moreover, our results align with recent find- ings by Weingarten and Hartmann (Weingarten & Hartmann, 2023), who found that participants did not change their behavior toward whole grain consumption directly after receiving the first messages on the health benefits. Ottersen et al., (2022) conducted a study with Norwegian consumers to test whether daily mobile phone text message reminders about animal welfare, and the environmental and health consequences of meat would reduce people’s meat consumption. They showed that meat consumption did not change. Therefore, sim- ply reminding consumers about these issues may not be enough without further interventions, as eating and dietary habits are strongly entrenched behaviors that are primarily controlled by autonomic processes. Our study is one of the few to assess the self-report- ed prospective behavior change (after four weeks of intervention) as a measure of message framing persua- siveness (Gallagher & Updegraff, 2012). As suggested by Meynier et al., (2020) information provision will more likely lead to a behavioral change if the information is provided on more than one occasion. For instance, Weingarten and Hartmann (2023) found that provid- ing information over time about the health benefits of whole grain consumption contributed to increasing the positive attitude and behavioral intentions to consume such products. However, in Time 2, we found no impact of the informative message (health information) on attitude, intention, and the reported behavior of eating whole grain pasta. This could be also due to the weekly Figure 2. Results of the TPB model in Time 1 (n=499) and in Time 2 (n=325). Notes: *** indicates a significant difference at p < 0.001, n.s. = not significant. Goodness-of-fit statistics: χ2 (df) = 112.61 (37); CFI = 0.975; TLI = 0.955, RMSEA (90% C.I.) = 0.064 (0.051-0.078). 310 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. information treatment (once per week), rather than a more intense exposure (daily messages for 14 days, as in the case of Weingarten and Hartmann (2023)). Anoth- er possible reason could be that information messages might have a short-lived effect on participants rather than other types of messages. For instance, Carfora et al., (2019) showed that participants exposed to emotional messages experienced a more enduring and long-lasting effect than information-type messages. The specific characteristics of the sample (young adults with a healthy status) may be one reason why the health claim message did not have an impact in chang- ing the perception towards whole grain. Past studies (e.g., Rothman & Updegraff, 2011) suggest that gain- framed and loss-framed messages may be amplified when the message is of high personal relevance, which might not be our case. Another possible reason for the lack of impact from the message is that it did not spe- cifically target consumers’ relevant beliefs (Fishbein & Ajzen, 2011; Weingarten & Hartmann, 2023). In our study, we found that the opinions of important others (e.g., parents, friends, and partners) were the strongest normative beliefs influencing the subjective norms (de Leeuw, Valois, Ajzen, & Schmidt, 2015); whereas the two most important behavioral beliefs relating to eat- ing whole grain pasta were a personal long-term invest- ment and the possibility of having fewer diet-related ill- nesses. Hence, the messages and interventions should target changing these key beliefs in order to lead to the desired changes. However, gain-framed messages were evaluated in terms of “Consumer evaluation” and “Argument qual- ity” better than loss-framed ones. The positive message about the health consequences associated with eating whole grain pasta was considered to be more appropri- ate, helpful, valuable, and persuasive. Thus, in line with the literature (Dolgopolova et al., 2022; Gallagher & Updegraff, 2012; Rothman, Bartels, Wlaschin, & Salovey, 2006), our results confirm the higher appropriateness of gain-framed health messages when encouraging behavior with ‘little risk’ compared to loss-framed messages (more persuasive with a ‘significant risky’ behavior to perform). Several limitations of our study occur. The first limitation is that we collected data only from a single University in the US, with a limited targeted popula- tion. Therefore, based also on the characteristics of this convenience sample (students enrolled in a US college), generalization of the findings to the broader popula- tion may be limited. Second, this study used self-report measures about the behavior of eating whole grain pas- ta which may be subject to response biases or limited memory. Third, although we focused our analysis on the individuals who actually opened the emailed messages, we cannot be sure whether the messages were truly read by the participants. Despite these limitations, we believe that our work will serve as a stimulus for further inves- tigation on how to better develop communication strat- egies for the health benefits of whole grain products. Future research could explore different types of mes- sages in terms of content and formats, as well as evaluate the results after a longer exposure. If concentrating on young adults, further studies could also consider testing the information across multiple dining halls to evalu- ate whether results are consistent across different cities. Finally, partnerships between nutrition, social scientists, and culinary professionals could support the develop- ment of relevant and useful information materials about whole grains consumption benefits. FUNDING This study is part of a wider project called “CON- SUMEHealth. Using consumer science to improve healthy eating habits” and has received funding from the European Union’s Horizon 2020 research and Inno- vation programme under the Marie Sklodowska-Curie grant agreement No 749514. ACKNOWLEDGEMENTS The authors are grateful to all the staff of the Cor- nell Dining venues for their collaboration and help in carrying out this study. Additionally, we want to express our gratitude to the students Mary Gaffney, Martina Cirelli, and Andrew Peterson who helped us in the data collection. REFERENCES Ajzen, I. (1991). The theory of planned behavior. Organi- zational Behavior and Human Decision Process- es, 50(2), 179–211. https://doi.org/10.1016/0749- 5978(91)90020-T Andreani, G., Sogari, G., Wongprawmas, R., Menozzi, D., & Mora, C. (2023). Indulgent or informative logos? Effects on university students’ intention to purchase healthy and sustainable food. International Journal of Gastronomy and Food Science, 33, 100774. https://doi. org/https://doi.org/10.1016/j.ijgfs.2023.100774 Bagozzi, R. P., & Yi, Y. (2012). Specification, evalua- tion, and interpretation of structural equation mod- https://doi.org/10.1016/0749-5978(91)90020-T https://doi.org/10.1016/0749-5978(91)90020-T https://doi.org/https https://doi.org/https http://doi.org/10.1016/j.ijgfs.2023.100774 311 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Intention and behavior toward eating whole grain pasta on a college dining campus els. Journal of the Academy of Marketing Science, 40, 8–34. https://doi.org/10.1007/s11747-011-0278-x Barberis, N., Gugliandolo, M. C., Costa, S., & Cannavò, M. (2022). Healthy and binge eating behaviours: the motivational processes underlying peer pressure. Psy- chology, Health & Medicine, 27(5), 1144–1153. htt- ps://doi.org/10.1080/13548506.2021.1903054 Bhattacherjee, A., & Sanford, C. (2006). Influence Pro- cesses for Information Technology Acceptance: An Elaboration Likelihood Model. MIS Quarterly, 30(4), 805–825. https://doi.org/10.2307/25148755 Biasini, B., Rosi, A., Giopp, F., Turgut, R., Scazzina, F., & Menozzi, D. (2021). Understanding, promot- ing and predicting sustainable diets: A systematic review. Trends in Food Science & Technology, 111, 191–207. https://doi.org/https://doi.org/10.1016/j. tifs.2021.02.062 Biasini, B., Rosi, A., Scazzina, F., & Menozzi, D. (2023). Predicting the Adoption of a Sustainable Diet in Adults: A Cross-Sectional Study in Italy. Nutrients, Vol. 15. https://doi.org/10.3390/nu15122784 Bisanz, K. J., & Krogstrand, K. L. S. (2007). Consump- tion & Attitudes about Whole Grain Foods of UNL Students Who Dine in a Campus Cafeteria. Retrieved from https://api.semanticscholar.org/Corpu- sID:70589297 Byrne, B. M. (2010). Structural equation modeling with AMOS: Basic concepts, applications and programming (2nd ed.). New York, US: Routledge/Taylor & Francis Group. Carfora, V., Bertolotti, M., & Catellani, P. (2019). Infor- mational and emotional daily messages to reduce red and processed meat consumption. Appetite, 141, 104331. https://doi.org/10.1016/j.appet.2019.104331 Christoph, M. J., Ellison, B. D., & Meador, E. N. (2016). The Influence of Nutrition Label Placement on Awareness and Use among College Students in a Dining Hall Setting. Journal of the Academy of Nutri- tion and Dietetics, 116(9), 1395–1405. https://doi. org/10.1016/j.jand.2016.05.003 Dammann, K. W., Hauge, D., Rosen, R. A., Schroeder, N., & Marquart, L. (2013). 6 - Consumption and consumer challenges of wholegrain foods. In J. A. Delcour & K. B. T.-F.-R. and W. F. Poutanen (Eds.), Woodhead Publishing Series in Food Science, Technol- ogy and Nutrition (pp. 120–149). https://doi.org/htt- ps://doi.org/10.1533/9780857095787.1.120 de Leeuw, A., Valois, P., Ajzen, I., & Schmidt, P. (2015). Using the theory of planned behavior to identify key beliefs underlying pro-environmental behavior in high-school students: Implications for educational interventions. Journal of Environmental Psychology, 42, 128–138. https://doi.org/https://doi.org/10.1016/j. jenvp.2015.03.005 Deliens, T., Van Crombruggen, R., Verbruggen, S., De Bourdeaudhuij, I., Deforche, B., & Clarys, P. (2016). Dietary interventions among university students: A systematic review. Appetite, 105, 14–26. https://doi. org/10.1016/j.appet.2016.05.003 Dolgopolova, I., Li, B., Pirhonen, H., & Roosen, J. (2022). The effect of attribute framing on consumers’ atti- tudes and intentions toward food: A Meta-analysis. Bio-Based and Applied Economics, 10(4), 253–264. https://doi.org/10.36253/bae-11511 EFSA Panel on Dietetic Products, Nutrition, and Aller- gies (NDA). (2010). Scientific Opinion on Dietary Reference Values for carbohydrates and dietary fibre. EFSA Journal, 8(3), 1462. https://doi.org/10.2903/j. efsa.2010.1462 European Commission. (2019). Whole grain. Retrieved from Health Promotion and Disease Prevention web- site: https://ec.europa.eu/jrc/en/health-knowledge- gateway/promotion-prevention/nutrition/whole-grain Fishbein, M., & Ajzen, I. (2011). Predicting and chang- ing behavior: The reasoned action approach. In Pre- dicting and Changing Behavior: The Reasoned Action Approach. https://doi.org/10.4324/9780203838020 Food and Drug Administration. 21CFR101.14. , (2023). Franchini, C., Biasini, B., Rosi, A., & Scazzina, F. (2023). Best practices for making the university campus a supportive environment for healthy and sustain- able diets. Current Opinion in Environmental Science & Health, 32, 100436. https://doi.org/https://doi. org/10.1016/j.coesh.2022.100436 Friedman, V. J., Wright, C. J. C., Molenaar, A., McCaffrey, T., Brennan, L., & Lim, M. S. C. (2022). The Use of Social Media as a Persuasive Platform to Facilitate Nutrition and Health Behavior Change in Young Adults: Web-Based Conversation Study. Journal of Medical Internet Research, 24(5), e28063. https://doi. org/10.2196/28063 Gallagher, K. M., & Updegraff, J. A. (2012). Health mes- sage framing effects on attitudes, intentions, and behavior: a meta-analytic review. Annals of Behavio- ral Medicine : A Publication of the Society of Behavio- ral Medicine, 43(1), 101–116. https://doi.org/10.1007/ s12160-011-9308-7 Hagger, M. S., & Chatzisarantis, N. L. D. (2006). Self- identity and the theory of planned behaviour: Between- and within-participants analyses. British Journal of Social Psychology, 45(4), 731–757. https:// doi.org/https://doi.org/10.1348/014466605X85654 Hagger, M. S., Chatzisarantis, N. L. D., & Harris, J. (2006). The Process by Which Relative Autonomous https://doi.org/10.1007/s11747-011-0278-x https://doi.org/10.1080/13548506.2021.1903054 https://doi.org/10.1080/13548506.2021.1903054 https://doi.org/10.2307/25148755 https://doi.org/https http://doi.org/10.1016/j.tifs.2021.02.062 http://doi.org/10.1016/j.tifs.2021.02.062 https://doi.org/10.3390/nu15122784 https://api.semanticscholar.org/CorpusID https://api.semanticscholar.org/CorpusID https://doi.org/10.1016/j.appet.2019.104331 https://doi.org/10.1016/j.jand.2016.05.003 https://doi.org/10.1016/j.jand.2016.05.003 https://doi.org/https https://doi.org/https http://doi.org/10.1533/9780857095787.1.120 https://doi.org/https http://doi.org/10.1016/j.jenvp.2015.03.005 http://doi.org/10.1016/j.jenvp.2015.03.005 https://doi.org/10.1016/j.appet.2016.05.003 https://doi.org/10.1016/j.appet.2016.05.003 https://doi.org/10.36253/bae-11511 https://doi.org/10.2903/j.efsa.2010.1462 https://doi.org/10.2903/j.efsa.2010.1462 https://ec.europa.eu/jrc/en/health-knowledge-gateway/promotion-prevention/nutrition/whole-grain https://ec.europa.eu/jrc/en/health-knowledge-gateway/promotion-prevention/nutrition/whole-grain https://doi.org/10.4324/9780203838020 https://doi.org/https http://doi.org/10.1016/j.coesh.2022.100436 http://doi.org/10.1016/j.coesh.2022.100436 https://doi.org/10.2196/28063 https://doi.org/10.2196/28063 https://doi.org/10.1007/s12160-011-9308-7 https://doi.org/10.1007/s12160-011-9308-7 https://doi.org/https https://doi.org/https http://doi.org/10.1348/014466605X85654 312 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. Motivation Affects Intentional Behavior: Compar- ing Effects Across Dieting and Exercise Behaviors. Motivation and Emotion, 30(4), 306–320. https://doi. org/10.1007/s11031-006-9046-5 Hung, Y., & Verbeke, W. (2019). Consumer evalua- tion, use and health relevance of health claims in the European Union. Food Quality and Prefer- ence, 74, 88–99. https://doi.org/10.1016/J.FOOD- QUAL.2019.01.002 Jones, J. M., & Engleson, J. (2010). Whole Grains: Ben- efits and Challenges. Annual Review of Food Science and Technology, 1(1), 19–40. https://doi.org/10.1146/ annurev.food.112408.132746 Kann, L., McManu, T., Harris, W. A., Shanklin, S. L., Flint, K. H., Queen, B., … Ethier, K. A. (2018). Youth Risk Behavior Surveillance – United States, 2017 MMWR Surveill Summ 2018;67(No. SS-8): https://doi. org/http://dx.doi.org/10.15585/mmwr.ss6708a1 Kline, R. B. (2016). Principles and practice of structural equation modeling, 4th ed. In Principles and practice of structural equation modeling, 4th ed. New York, NY, US: Guilford Press. Li, L., Long, X., Laubayeva, A., Cai, X., & Zhu, B. (2020). Behavioral intention of environmentally friendly agricultural food: the role of policy, perceived value, subjective norm. Environmental Science and Pol- lution Research, 27(15), 18949–18961. https://doi. org/10.1007/s11356-020-08261-x Maki, C. K., Palacios, M. O., Koecher, K., Sawicki, M. C., Livingston, A. K., Bell, M., … McKeown, M. N. (2019). The Relationship between Whole Grain Intake and Body Weight: Results of Meta-Analyses of Obser- vational Studies and Randomized Controlled Trials. Nutrients , Vol. 11. https://doi.org/10.3390/nu11061245 Marquart, L., Wiemer, K. L., Jones, J. M., & Jacob, B. (2003). Whole grain health claims in the USA and other efforts to increase whole-grain consumption. Proceedings of the Nutrition Society, 62(1), 151–160. https://doi.org/DOI: 10.1079/PNS2003242 McDermott, M. S., Oliver, M., Simnadis, T., Beck, E. J., Coltman, T., Iverson, D., … Sharma, R. (2015). The Theory of Planned Behaviour and dietary patterns: A systematic review and meta-analysis. Preventive Medicine, 81, 150–156. https://doi.org/10.1016/j. ypmed.2015.08.020 McEachan, R. R. C., Conner, M., Taylor, N. J., & Lawton, R. J. (2011). Prospective prediction of health-related behaviours with the Theory of Planned Behaviour: a meta-analysis. Health Psychology Review, 5(2), 97–144. https://doi.org/10.1080/17437199.2010.521684 Meynier, A., Chanson-Rollé, A., & Riou, E. (2020). Main Factors Influencing Whole Grain Consumption in Children and Adults – A Narrative Review. Nutrients, 12(8), 2217. https://doi.org/10.3390/nu12082217 O’Keefe, D. J., & Jensen, J. D. (2008). Do Loss-Framed Persuasive Messages Engender Greater Message Pro- cessing Than Do Gain-Framed Messages? A Meta- Analytic Review. Communication Studies, 59(1), 51–67. https://doi.org/10.1080/10510970701849388 Ottersen, I. S., Benningstad, N. C. G., & Kunst, J. R. (2022). Daily reminders about the animal-welfare, environ- mental and health consequences of meat and their main and moderated effects on meat consumption. Cleaner and Responsible Consumption, 5, 100068. htt- ps://doi.org/https://doi.org/10.1016/j.clrc.2022.100068 Quick, V., Wall, M., Larson, N., Haines, J., & Neumark- Sztainer, D. (2013). Personal, behavioral and socio- environmental predictors of overweight incidence in young adults: 10-yr longitudinal findings. Internation- al Journal of Behavioral Nutrition and Physical Activ- ity, 10(1), 37. https://doi.org/10.1186/1479-5868-10-37 Rothman, A. J., Bartels, R. D., Wlaschin, J., & Salovey, P. (2006). The Strategic Use of Gain- and Loss-Framed Messages to Promote Healthy Behavior: How The- ory Can Inform Practice. Journal of Communica- tion, 56(s1), S202–S220. https://doi.org/https://doi. org/10.1111/j.1460-2466.2006.00290.x Rothman, A. J., & Updegraff, J. A. (2011). Specifying when and how gain-and loss-framed messages motivate healthy behavior: an integrated approach. In G. Keren (Ed.), Perspectives on framing. Psychology Press. Shepherd, R., Dean, M., Lampila, P., Arvola, A., Saba, A., Vassallo, M., … Lähteenmäki, L. (2012). Commu- nicating the benefits of wholegrain and functional grain products to European consumers. Trends in Food Science & Technology, 25(2), 63–69. https://doi. org/https://doi.org/10.1016/j.tifs.2012.01.002 Slavin, J. L. (2005). Dietary fiber and body weight. Nutri- tion (Burbank, Los Angeles County, Calif.), 21(3), 411–418. https://doi.org/10.1016/j.nut.2004.08.018 Sogari, G., Li, J., Lefebvre, M., Menozzi, D., Pellegrini, N., Cirelli, M., … Mora, C. (2019). The influ- ence of health messages in nudging consumption of whole grain pasta. Nutrients, 11(12). https://doi. org/10.3390/nu11122993 Sogari, G., Li, J., Wang, Q., Lefebvre, M., Huang, S., Mora, C., & Gómez, M. I. (2022). Toward a reduced meat diet: University North American students’ acceptance of a blended meat-mushroom burger. Meat Science, 187, 108745. https://doi.org/https://doi.org/10.1016/j. meatsci.2022.108745 Stok, F. M., Renner, B., Clarys, P., & Deliens, T. (2018). Understanding Eating Behavior during the Transition from Adolescence to Young Adulthood : A Literature. https://doi.org/10.1007/s11031-006-9046-5 https://doi.org/10.1007/s11031-006-9046-5 https://doi.org/10.1016/J.FOODQUAL.2019.01.002 https://doi.org/10.1016/J.FOODQUAL.2019.01.002 https://doi.org/10.1146/annurev.food.112408.132746 https://doi.org/10.1146/annurev.food.112408.132746 https://doi.org/http https://doi.org/http http://dx.doi.org/10.15585/mmwr.ss6708a1 https://doi.org/10.1007/s11356-020-08261-x https://doi.org/10.1007/s11356-020-08261-x https://doi.org/10.3390/nu11061245 https://doi.org/DOI https://doi.org/10.1016/j.ypmed.2015.08.020 https://doi.org/10.1016/j.ypmed.2015.08.020 https://doi.org/10.1080/17437199.2010.521684 https://doi.org/10.3390/nu12082217 https://doi.org/10.1080/10510970701849388 https://doi.org/https https://doi.org/https http://doi.org/10.1016/j.clrc.2022.100068 https://doi.org/10.1186/1479-5868-10-37 https://doi.org/https http://doi.org/10.1111/j.1460-2466.2006.00290.x http://doi.org/10.1111/j.1460-2466.2006.00290.x https://doi.org/https https://doi.org/https http://doi.org/10.1016/j.tifs.2012.01.002 https://doi.org/10.1016/j.nut.2004.08.018 https://doi.org/10.3390/nu11122993 https://doi.org/10.3390/nu11122993 https://doi.org/https http://doi.org/10.1016/j.meatsci.2022.108745 http://doi.org/10.1016/j.meatsci.2022.108745 313 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Intention and behavior toward eating whole grain pasta on a college dining campus Nutrients, 10(667), 1–16. https://doi.org/10.3390/ nu10060667 U.S. Department of Health and Human Services and U.S. Department of Agriculture, & US Department of Agriculture. (2015). 2015–2020 Dietary Guidelines for Americans. Retrieved April 4, 2019, from http:// health.gov/dietaryguidelines/2015/guidelines/ Ugunesh, D., Siau, C. S., Mohd Sanip, M. N. A. Bin, & Koo, H. C. (2023). Perceived barriers towards whole grain consumption among the Malaysian adult popu- lation: findings from a theory-based qualitative study. British Food Journal, 125(3), 1130–1147. https://doi. org/10.1108/BFJ-10-2021-1152 van Griethuijsen, R. A. L. F., van Eijck, M. W., Haste, H., den Brok, P. J., Skinner, N. C., Mansour, N., … Bou- Jaoude, S. (2015). Global Patterns in Students’ Views of Science and Interest in Science. Research in Science Education, 45(4), 581–603. https://doi.org/10.1007/ s11165-014-9438-6 Weingarten, N., & Hartmann, M. (2023). Fifty shades of grain – Increasing whole grain consumption through daily messages. Appetite, 187, 106608. https://doi.org/ https://doi.org/10.1016/j.appet.2023.106608 Wongprawmas, R., Sogari, G., Menozzi, D., Pellegrini, N., Lefebvre, M., Gómez, M. I., & Mora, C. (2021). Determinants of US University Students’ Willingness to Include Whole Grain Pasta in Their Diet. Interna- tional Journal of Environmental Research and Public Health, 18(6). https://doi.org/10.3390/ijerph18063173 World Health Organization. (2014). Adolescents: health risks and solutions. Retrieved from Fact sheet N°345 website: http://www.searo.who.int/thailand/ factsheets/fs0027/en/ Ye, E. Q., Chacko, S. A., Chou, E. L., Kugizaki, M., & Liu, S. (2012). Greater whole-grain intake is associated with lower risk of type 2 diabetes, cardiovascular dis- ease, and weight gain. Journal of Nutrition, 142(7), 1304–1313. https://doi.org/10.3945/jn.111.155325 APPENDICES Figure A1. Survey flow. https://doi.org/10.3390/nu10060667 https://doi.org/10.3390/nu10060667 http://health.gov/dietaryguidelines/2015/guidelines/ http://health.gov/dietaryguidelines/2015/guidelines/ https://doi.org/10.1108/BFJ-10-2021-1152 https://doi.org/10.1108/BFJ-10-2021-1152 https://doi.org/10.1007/s11165-014-9438-6 https://doi.org/10.1007/s11165-014-9438-6 https://doi.org/https https://doi.org/https http://doi.org/10.1016/j.appet.2023.106608 https://doi.org/10.3390/ijerph18063173 http://www.searo.who.int/thailand/factsheets/fs0027/en/ http://www.searo.who.int/thailand/factsheets/fs0027/en/ https://doi.org/10.3945/jn.111.155325 314 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. Table A1. Constructs and Items. Codes Constructs and items Behavioral beliefs (1 = strongly disagree; 7 = strongly agree) (time 1) BehBel1 If I include whole grain pasta in my diet over the next month I believe I will live a better quality of life in my old age BehBel2 If I include whole grain pasta in my diet over the next month I believe I will have made a long-term investment for myself BehBel3 If I include whole grain pasta in my diet over the next month I believe I will have less diet-related diseases in my life Normative beliefs (1 = strongly disagree; 7 = strongly agree) (time 1) NorBel1 My parents think I should include whole grain pasta in my diet over the next month NorBel2 My friends/partner think I should include whole grain pasta in my diet over the next month NorBel3 Nutritionists think I should include whole grain pasta in my diet over the next month NorBel4 My doctor thinks I should include whole grain pasta in my diet over the next month NorBel5 Chefs think I should include whole grain pasta in my diet over the next month Control beliefs (1 = strongly disagree; 7 = strongly agree) (time 1) ConBel1 The limited advertising from the dining halls/restaurants I usually go does not encourage me to include whole grain pasta in my diet over the next month ConBel2 The higher costs of whole grain pasta stops me from including this product in my diet over the next month ConBel3 The lack of availability in the dining halls I usually go stops me from including whole grain pasta in my diet over the next month ConBel4 The limited information from public authorities about whole grain benefits does not encourage me to include whole grain pasta in my diet over the next month Attitude towards the behavior (time 1 and time 2) For me, including whole grain pasta in my diet over the next month (7-point scale) ATT1 Difficult - Easy ATT2 Not tasty - Tasty Subjective norm (1 = strongly disagree; 7 = strongly agree) (time 1) SN1 Most people who are important to me think that I should include whole grain pasta in my diet over the next month SN2 Most people who influence my decisions think that I should include whole grain pasta in my diet over the next month SN3 It is expected that I should include whole grain pasta in my diet over the next month Perceived behavioral control (1 = strongly disagree; 7 = strongly agree) (time 1) PBC1 I believe that including whole grain pasta in my diet over the next month is possible PBC2 The decision to include whole grain pasta in my diet over the next month will be only up to me Behavioral Intention (1 = strongly disagree; 7 = strongly agree) (time 1 and time 2) INT1 I intend to include whole grain pasta in my diet over the next month INT2 I will try in anyway to include whole grain pasta in my diet over the next month INT3 I will definitely include whole grain pasta in my diet over the next month Behavior (after one month) (7-point scale) (time 2) Beh1 In the past month, how often have you included a meal with whole grain pasta in your diet? Never - Almost always Beh2 I have included whole grain pasta in my diet at least once in the past month. False-True Consumer evaluation of the claim (1 = strongly disagree; 7 = strongly agree) (time 1) ConsEval1 I am familiar with the health claim I just read ConsEval2 I understand this health claim ConsEval3 This health claim is credible ConsEval4 This health claim is interesting ConsEval5 This health claim is important Argument quality (1 = strongly disagree; 7 = strongly agree) (time 1) ArgQua1 The information provided about whole grain pasta is informative ArgQua2 The information provided about whole grain pasta is helpful ArgQua3 The information provided about whole grain pasta is valuable ArgQua4 The information provided about whole grain pasta is persuasive (Continued) 315 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Intention and behavior toward eating whole grain pasta on a college dining campus Codes Constructs and items Perceived Usefulness (1 = strongly disagree; 7 = strongly agree) (time 1 and time 2) PercUse1 Including whole grain pasta in my diet will help me to stay in shape (e.g., maintaining my body weight). PercUse2 Including whole grain pasta in my diet will improve my work performance (e.g., make my working/studying life more productive). PercUse3 Including whole grain pasta in my diet will make my diet more balanced and healthy (e.g., right amount of fiber intake). Table A2. In italic the messages shown to participants. Message Health benefits of eating whole grain Gain framed message (Gfm) Loss-framed message (Lfm) 1 Better chance of success in maintaining your body weight (BW) If you include whole grain pasta in your diet, you might have a better chance of success in maintaining your body weight. If you do not include whole grain pasta in your diet, you might not have a better chance of success in maintaining your body weight. 2 Its fiber content will contribute to your normal bowel function (BF) If you include whole grain pasta in your diet, its fiber content will contribute to your normal bowel function. If you do not include whole grain pasta in your diet, a lack of fiber content will not contribute to normal bowel function. 3 Niacin content (vitamin B3) will contribute to the reduction of tiredness and fatigue (T&F) If you include whole grain pasta in your diet, its niacin content (vitamin B3) will contribute to the reduction of tiredness and fatigue. If you do not include whole grain pasta in your diet, a lack of niacin (Vitamin B3) will not contribute to the reduction of tiredness and fatigue. 4 Its fiber content will promote your healthy gut (HG) If you include whole grain pasta in your diet, its fiber content will promote your healthy gut. If you do not include whole grain pasta in your diet, a lack of fiber content will not promote your gut health. Four different types of health messages were developed, based on the latest scientific opinion on the substantiation of health claims related to (1) whole grain (EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA), 2010), (2) wheat bran fibre and increase in faecal bulk (EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA), 2010a), and (3) niacin and reduction of tiredness and fatigue (EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA), 2010a). Table A3. Internal consistency of TPB constructs and other variables in Time 1 and Time 2. Variable N. of Items Control Gain Frame Loss Frame p-value a N Cronbach’s alpha M SD N Cronbach’s alpha M SD N Cronbach’s alpha M SD Time 1 ATT 2 100 0.628 4.750 1.319 202 0.717 4.849 1.566 197 0.702 4.655 1.473 0.426 Time 2 ATT 2 77 0.598 4.773 1.344 134 0.748 4.787 1.516 114 0.756 4.956 1.351 0.572 Time 1 PU 3 100 0.762 4.443 1.062 202 0.825 4.315 1.141 197 0.847 4.201 1.221 0.225 Time 2 PU 3 77 0.837 4.714 1.016 134 0.861 4.925 1.157 114 0.824 4.818 1.140 0.408 Time 1 SN 3 100 0.857 3.443 1.311 202 0.905 3.705 1.370 197 0.897 3.504 1.497 0.216 Time 1 PBC 2 100 0.559 5.505 1.067 202 0.616 5.505 1.116 197 0.643 5.472 1.185 0.951 Time 1 INT 3 100 0.902 4.120 1.496 202 0.918 4.297 1.611 197 0.897 4.191 1.517 0.613 Time 2 INT 3 77 0.933 4.416 1.369 134 0.931 4.368 1.572 114 0.907 4.450 1.479 0.910 Time 2 Bahavior 2 77 0.732 4.727 1.572 134 0.760 4.493 1.746 114 0.768 4.676 1.803 0.566 Time 1 ConsEval 5 - - - - 202 0.622 4.857 0.886 197 0.742 4.154 1.127 <0.001 Time 1 ArgQua 4 - - - - 202 0.859 4.774 1.132 197 0.908 3.956 1.415 <0.001 Note: TPB = Theory of Planned Behavior; ATT = Attitude; PU = Perceived usefulness; SN = Subjective Norms; PBC = Perceived Behavioral Control; INT = Intention; ConsEval = Consumer evaluation of the claim; ArgQua = Argument Quality. M = Means; SD = Standard Devia- tion. a Comparison between groups using ANOVA tests. Table A1. (Continued). 316 Bio-based and Applied Economics 13(3): 301-316, 2024 | e-ISSN 2280-6172 | DOI: 10.36253/bae-15456 Giovanni Sogari et al. Table A4. TPB Model: unstandardized beta coefficients, standard errors (S.E.), p-values, in Time 1 (n=499) and in Time 2 (n=325). Predictors Path coefficients Beta S.E. p Predictors of Behavioral Intention (in Time 1) ATT 0.565 0.076 <0.001 PBC 0.220 0.051 <0.001 SN 0.604 0.055 <0.001 Predictors of Behavior (in Time 2) INT 0.612 0.057 <0.001 PBC 0.077 0.067 0.250 Note: ATT: attitude towards the behavior; SN: subjective norms; PBC: perceived behavioral control; INT: behavioral intention. Guns, Germs and Climate: Food Security and Food Systems in a Risky World Francesco Pagliacci1,*, Valentina Raimondi2, Luca Salvatici3,* The implications of the Russian invasion of Ukraine for African economies: A CGE analysis for Ethiopia Amsalu Woldie Yalew1,2,*, Victor Nechifor3, Emanuele Ferrari3 Participation of farmers in market value chains: A tailored Antràs and Chor positioning indicator Tulia Gattone Adapting to climate change: what really drives the choices of the producers of Geographical Indications? Francesco Pagliacci1, Dana Salpina1,2,* Exploring preferences for contractual terms in a scenario of ecological transition for the agri-food sector: a latent class approach Stefano Ciliberti, Angelo Frascarelli*, Gaetano Martino, Andrea Marchini Intention and behavior toward eating whole grain pasta on a college dining campus: Theory of Planned Behavior and message framing Giovanni Sogari1, Rungsaran Wongprawmas1, Giulia Andreani1, Michele Lefebvre2, Nicoletta Pellegrini3, Miguel I. Gómez4, Cristina Mora1, Davide Menozzi1,*