In light of recent changes in fast food in-take, this study sought to determine the correlation between the frequency of fast food consumption (FFFC) and age, gender, proximity of fast food restaurants (FFR), exercise and healthy eating habits. Using Google Document Application and Microsoft Word documents, the survey was disseminated via email to 83 participants who volunteered from various first-world countries. Distribution of dependent variable (FFFC) was mildly skewed to the left and assumption for homoscedasticity was violated for the predictor variable, healthy eating habits. No transformations were attempted as this study used multiple regression analyses, which is rather insensitively robust to moderate violations. Results indicated that only gender and age were significant predictors for the above-stated modal. The reasons for the failure of the remaining predictor variables to attain significance were then discussed, alongside possible implications of the results found and future research directions.
I'm Lovin' It:
Gender & Age Significantly Related to Increased Fast Food Consumption
In 2008, a survey organized by the BBC found that '45% in the UK agreed with the statement "I like the taste of fast food too much to give it up", while 44% of Americans said they would be unable to give up their burgers, pizzas and chicken wings.' (BBC News, 2008). The Independent had also reported that Britons eat an average of 7m worth of fast food a day (Watson-Smyth, 1999). Hence, it would seem that first-world countries like the UK's eating habits have undergone a huge change towards convenient food consumption, especially eating out at fast food restaurants (FFR) (Nestle & Woteki, 1995).
One possible reason could be due to gender and age for it was found that in the average week, older women have healthier food habits than did younger women, and women in general consume less fast food than men (Cohen, Felix & Brownell, 1990; French, Harnack & Jeffery, 2000; Paeratakul, Ferdinand, Champagne, Ryan & Bray, 2003; Pereira, Kartashov, Ebbeling, Van Horn, Slattery, Jacobs, Ludwig, 2005). Pereira et al. (2005) researched into the link between reported fast-food habits and weight gain as well as change in insulin resistance over a 15-year period and found that there was a significant difference between men and women's fast food consumption (FFC), with women reportedly having less FFC. Meanwhile, Paratakul et al. (2003) conducted a survey with over 16,000 participants and found that FFC decreases with age.
Another factor that influences the frequency of fast food consumption (FFFC) can be due to the proximity of a FFR to one's residence or workplace. Burns & Inglis (2007) found that in Melbourne, Australia, participants who live within 8-10minutes of a car ride from a FFR are more likely to have a higher FFFC than those who live within the same time frame but from a supermarket (which in the study, was an indication of healthy eating). Another project also found that children who live in neighbourhoods with easy access to a FFR (either by walking or driving) have a significantly higher FFFC than those who do not (Grier, Mensinger, Huang, Kumanyika, & Stettler, 2007). In addition, participants who live further away from a convenience or drug store had higher fruit, juice and low-fat vegetable consumption but those who live nearer to a FFR had increased high-fat vegetable, fruit and juice consumption (Jago, Baranowski, Baranowski, Cullen, & Thompson, 2007). However, one study found no significant association between the proximity of FFRs with FFFC, highlighting that different countries have different densities of FFR, hence participants have enough access such that physical access was not a limiting factor governing FFR patronage (Jeffery, Baxter, McGuire & Linde, 2006).
Exercise habits may also significantly affect the FFFC. In a survey of 7th grade students of Asian-American and Hispanic descent, it was found that lower frequency of exercise was correlated with a higher FFFC, and that this significant associations still persist even after confounding variables such as English language usage and socioeconomic status were controlled for (Unger, Reynolds, Shakib, Spruijt-Metz, Sun, & Johnson, 2004).
The final factor that this model will look into is healthy eating habits. In a survey of 658 African-Americans in North Carolina, it was found that a higher FFFC was positively correlated with higher total and saturated fat intakes yet negatively correlated with vegetable intake (Satia, Galanko & Siega-Riz, 2004). A higher FFFC was also found to positively associate with higher intake of total energy and percent fat but negatively associated with fibre intake (French, Jeffery & Oliphant, 1994).
Though many previous studies had evaluated each of the above factors, none was placed in a single model to correlate with FFFC. This paper thus hypothesizes that using a model of the above-mentioned predictor variables (in the average week), men will consume more fast food than women, and younger adults will also have a higher frequency of FFC. It also hypothesized that participants who had a shorter walking distance to a FFR, exercised less frequently and had a lower frequency of healthy food consumption will have a higher FFFC.
A cross-sectional survey design was employed, with gender, age, frequency of physical activity, healthy eating habits and proximity of fast food restaurants as factors affecting the frequency of fast food consumption.
83 participants recruited from families and friends of first-world countries participated in this survey. There were 45 males and 39 females, and all ranged between the ages of 18 to 60 (M = 29.59, SD = 12.91). There was one missing datum for age.
The survey was devised from prior research into possible factors that influences the frequency of fast food consumption.
These were measured using two items that explicitly requested for participants' age and gender at the beginning of the survey (refer to Appendix B).
Fast Food Consumption
This was assessed with 6 items, so as to ensure specificity in the types of fast food consumed from a FFR in an average week, for some participants may at times consumed only a side order. The first item measured the frequency of consuming a side order i.e. fries or onion rings, followed by the frequency of consuming soda. The next 2 items measured the frequency of consuming various types of meats. To measure the type of bread consumed from a FFR, an item was reverse coded by asking the participants "How often do you eat wholemeal or seeded bread during the average week?" Unless requested, most FFR's meals would provide white bread (Paeratakul et al., 2003). The last item measures the frequency of consuming pizzas in a FFR.
The location of the nearest FFR was measured in a single item and required participants to round off to the nearest minute (for time taken). Behavioural Variables
Physical activity was measured using three items, each measuring the frequency of various levels of exercise (light, moderate and vigorous). Time taken during each physical activity was also measured, using a scale ranging from 'Not Applicable' to 'more than 90 minutes'.
Healthy Eating Habits
Healthy food consumption in the average week was measured using four items, each testing various categories of food. The first item measured the frequency of dairy i.e. milk or yogurt consumed, followed by the frequency of lean meats consumed. The last two measured the frequency of fruits and vegetables consumed.
All items for Fast Food Consumption, Physical Activity and Healthy Eating Habits were measured on a scale from 'Never' to 'Daily' and these were then add up to operationalise the specific variable.
Participants were recruited using a simple random sampling, and the surveys disseminated via email, either through Google Document Application or Microsoft Word. Participants fill up the surveys online and/or email the surveys back to the experimenters. They were also given the option of filling it in hard copy, in the presence of the experimenter. Informed consent and the subsequent debrief was given in soft copy to the participants.
Participants were given informed consent, whereby they were told that their participation was voluntary, and were therefore allowed to withdraw from the survey or have their data removed at any given time. They were also assured of confidentiality and anonymity (Google Document Application provides anonymity). During debrief, participants were also given local helplines (participants were drawn from various countries) and were told that should they feel aggravated or upset by the survey, they can seek help from the numbers provided.
Tests for normality was first carried out for FFFC (skewness = 0.640, kurtosis = 0.543) and the Figures 1 & 1.2 reflected the moderate positive skewedness of the distribution. Assumption for Homoscedasticity was also tested and the results shown in Table 1. There was homogeneity of variance for all except healthy eating habits. Research have shown that regression is robustly insensitive to small and moderate violations of homoscedasticity, as well as moderately skewed distributions (Fox, 2005; Lindquist, 1953), hence no transformation was performed.
This study hypothesized that a model consisting of the predictor variables gender, age, proximity, exercise and healthy eating habits would significantly predict FFFC, whereby males, younger adults, and participants who had a shorter walking distance to a FFR, exercised less frequently and had a lower frequency of healthy food consumption will have a higher FFFC. The model was found to be statistically significant together with only two predictor variables - gender and age.
The results for the significant variables were reflective of previous experiments (Blanck, Yaroch, Atienza, Yi, Zhang & Msse, 2009; Pereira et al, 2005). This could be due to females being more concerned with body image and older participants having health-related concerns due to aging. After all, fast food have been significantly related to obesity which may hamper one's body image and also health-related illnesses (Caltabiano & Sarafino, 2002; Kenny & Adams, 1994). Proximity was not a significant predictor and this was replicated by Jeffery et al. (2005). This probably indicated that the proximity of a FFR may be relatively homogenous across first-world countries.
However, exercise and healthy eating habits' non-significant results differed from previous studies although healthy eating habits was significantly correlated to FFFC (French et al., 1994; Satia et al., 2004; Unger, et al., 2004). This could be due to the collinearity and these two variables were actually measuring the same thing. They were thus not significant in the multiple regression (Howell, 2007). In light of these predictors as well as the low adjusted R2, it may be possible that the variables were poorly measured or that important variables to this model have been excluded. Future studies should seek to find a better model for the prediction of FFFC.
Correlation does not equate causation, and though indicative of a significant relationship, these predictor variables do not necessarily cause a higher/lower FFFC. Nevertheless, governments can use this study to help minimise FFFC by targeting the relevant groups i.e. males and younger adults in national school curriculum. This will therefore encourage younger adults (especially those in schools) to switch to healthier food alternatives. This study can also be helpful to companies dealing in healthier food, for they can switch their marketing strategies towards those who have higher FFFC. Future studies should also look into the reasons that made these predictors significant, so as to better deal with the problem of an increasing FFFC.
As with surveys, participants may have provided answers that were socially desirable. It was found that socially desirable answers (e.g. amount of exercise) were often over-reported while undesirable ones (e.g. FFFC) were under-reported (Foddy, 1993). Some participants completed this survey online and as such may be subjected to effects of opinion distortion, whereby other people's opinion may have influenced their answer (De Vaus, 2002). Lastly, there may be a mode effect, in which different mode of administration (face-to-face & internet) may affect the way people respond (Dillman, 2000). This was unavoidable because some participants wanted to have their survey conducted in one mode instead of the other.
In conclusion, this study showed that males were significantly more likely to have a higher FFFC than females, and that younger participants were also significantly likely to have a higher FFFC than older participants. It however did not find any significance for lower FFFC in participants who had a shorter walking distance to a FFR, who exercised more frequently and had a higher frequency of health food consumption did not significantly.
Blanck, H. M., Yaroch, A. L., Atienza, A. A., Yi, S. L., Zhang, J. & Msse, L. C. (2009). Factors influencing lunchtime food choices among working Americans. Health Education & Behavior, 36(2), 289-301.
British Broadcasting Corporation. (2008, 2nd January). UK and US 'keenest on fast food'. Retrieved 2nd January, 2010 from: http://news.bbc.co.uk/1/hi/health/7165990.stm
Burns, C. M. & Inglis, A. D. (2007). Measuring food access in Melbourne: Access to healthy and fast foods by car, bus and foot in an urban municipality in Melbourne. Health & Place, 13(4), 877-885.
Caltabiano, M. L. & Sarafino, E. P. (2002). Health Psychology: Biopsychosocial Interactions, An Australian Perspective. Singapore: Wiley.
De Vaus, D. (2002). Surveys in Social Research, 5th ed. Australia: Routledge.
Dillman, N. K. (2000). Mail and Internet Surveys: The Total Design Method. New York: Wiley.
Cohen, R. Y., Brownell, K. D. & Felix, M. R. (1990). Age and sex differences in health habits and beliefs of schoolchildren. Health Psychology, 9(2), 208-224.
Foddy, W. H. (1996). The in-depth testing of survey questions: A critical appraisal of methods. Quality & Quantity, 30(4), 361-370.
Fox, J. (2005). Regression Diagnostics: Quantitative Applications in the Social Sciences, 13th ed. Newbury Park, CA: Sage.
French, S. A., Harnack, L. & Jeffery, R. W. (2000). Fast food restaurant use among women in the Pound of Prevention study: Dietary, behavioral and demographic correlates. International Journal of Obesity And Related Metabolic Disorders, 24(10), 1353-1359.
French S. A., Jeffery R. W. & Oliphant J. A. (1994). Facility access and self-reward as methods to promote physical activity among healthy sedentary adults. American Journal of Health Promotion, 8, 257-62.
Grier, S. A., Mensinger, J., Huang, S. .H., Kumanyika, S. K. & Stettler, N. (2007). Fast-food marketing and children's fast-food consumption: Exploring parents' influences in an ethnically diverse sample. Journal of Public Policy & Marketing, 26(2), 221-235.
Howell, D. C. (2007). Statistical Methods for Psychology, 6th ed. USA: Thomson Wadsworth.
Jago, R., Baranowski, T., Baranowski, J. C., Cullen, K. W. & Thompson, D. (2007). Distance to food stores & adolescent male fruit and vegetable consumption: Mediation effects. The International Journal of Behavioral Nutrition & Physical Activity, 4, 35. doi:10.1186/1479 5868-4-35.
Jeffery, R. W., Baxter, J., McGuire, M. & Linde, J. (2006). Are fast food restaurants an environmental risk factor for obesity? The International Journal of Behavioral Nutrition & Physical Activity, 3, 2. doi:10.1186/1479-5868-3-2.
Kenny, D. & Adams, R. (1994). The relationship between eating attitudes, body mass index, age, and gender in Australian university students. Australian Psychologist, 29(2), 128-134.
Lindquist, E. F. (1953). Design and Analysis of Experiments in Psychology and Education. Boston: Houghton Mifflin Co.
Nestle, M. & Woteki, C. E. (1995). Tends in american dietary patterns: Rsearch issues and policy implementations. In F. Bronner (Ed), Nutrition And Health: Topics and Controversies, pp. 1 44. USA: CRC Press, Inc.
Paeratakul, S., Ferdinand, D. P., Champagne, C. M., Ryan, D. H. & Bray, G. A. (2003). Fast-food consumption among US adults and children: Dietary and nutrient intake profile. Journal of the American Dietetic Association, 103(10), 1332-1338.
Pereira, M. A., Kartashov, A. I., Ebbeling, C. B., Van Horn, L., Slattery, M. L., Jacobs Jr, D. R. & Ludwig, D. S. (2005). Fast-food habits, weight gain, and insulin resistance (the CARDIA study): 15-year prospective analysis. Lancet, 365(9453), 36-42.
Satia, J. A., Galanko, J. A. & Siega-Riz, A. M. (2004). Eating at fast-food restaurants is associated with dietary intake, demographic, psychosocial and behavioural factors among African Americans in North Carolina. Public Health Nutrition, 7(8), 1089-1096.
Unger, J. B., Reynolds, K., Shakib, S., Spruijt-Metz, D., Sun, P. & Johnson, C. A. (2004). Acculturation, physical activity, and fast-food consumption among Asian-American and Hispanic adolescents. Journal of Community Health: The Publication for Health Promotion and Disease Prevention, 29(6), 467-481.
Watson-Smyth, K. (1999, May 13). Britons eating pounds 7m of fast food every day. Retrieved 2nd January, 2010 from: http://www.independent.co.uk/news/britons-eating-pounds-7m-of-fast food-every-day-1093273.html