Are the effects of diet on academic performance mediated by stress?

Abstract

This cross-sectional study aims to investigate the mediatory effects of stress on the relationship between diet and academic performance via a survey comprising of a validated Short Form Food Frequency Questionnaire, the Perceived Stress Scale and the Pittsburgh Sleep Quality Index. Utilising the PROCESS macro for mediation analysis in SPSS, it is predicted that stress will be a partial mediator with poor diet causing increased levels of stress and subsequently lower academic performance. The relationships between stress and gender, and sleep quality and academic performance will also be explored. The aims of this research are to expand the limited literature set among this population group and raise awareness to help promote a more holistic wellbeing approach across academic institutions

Literature Review

In the UK, there were 2.34 million university students in 2017, 18% higher than in 2000 ("HE Student Data | HESA," 2013). This trend is reflected globally with an increasing number of students enrolled in college or university level education, with 50% of the UK young adult population comprised of university students. But more importantly, the future success of graduates as a measure of quality of life, health, income and career are all considerably influenced by their academic achievements at university (Ross & Wu, 1995). Consequently, the elements associated with greater academic performance are of interest to both academic institutions and their students.

The impact of diet on academic performance has been a growing area of research in other population groups, with studies showing dietary intake does affect academic achievement in both children and adolescents (Burrows, Goldman, Pursey, & Lim, 2016), with regular consumption of breakfast, increased fruit and vegetable intake and lower frequency of junk food all being associated with higher levels of academic achievement. It is theorised the link between academic performance and diet relates to nutritional components such as folate, iron and omega 3 which are vital to the development and function of the brain, with increased frequency of fruits and vegetables (nutrient rich foods) and lower frequency of junk foods (nutrient poor foods) explaining the higher levels of essential micronutrients available for cognitive development and function (Gómez-Pinilla, 2008).

While factors such as excess alcohol consumption and sleep deprivation have been shown to have detrimental effects on academic achievement in university students (Singleton & Wolfson, 2009), other lifestyle variables such as diet and nutrition have been neglected somewhat within this population group. This should be of concern considering prior research has shown unhealthy diets are prevalent amongst university students; (Tanton, Dodd, Woodfield, & Mabhala, 2015) found that just 19% of British university students showed favourable healthy eating behaviours, with almost a third being categorised within the riskiest eating clusters. In a systematic review of the efficacy of interventions aimed at university and college students, (Plotnikoff et al., 2015) found that most nutritional interventions have been ineffective. Further insights into the association between diet and academic performance may assist in developing effective methods to improve the eating behaviours of university and college students, in addition to helping motivate students.

A recent study by (Michels, Man, Vinck, & Verbeyst, 2019) also found that dietary deterioration amongst university students occurred during the stressful examination period, supporting the theory that stress stimulates unhealthy dietary choices leaning towards fatty and sweet foods. (Mikolajczyk, El Ansari, & Maxwell, 2009) also found perceived stress to be linked to more frequent consumption of sweet and fatty foods, with depressive symptoms associated with less frequent consumption of fruit and vegetables. However, these results were only present in female students indicating further research into the links between diet and stress amongst the university population group are needed.

Research Aims

This study will be conducted to examine how the relationship between academic exam performance and diet is mediated by stress, if at all (see Appendix K). Expanding the literature within this population group using a validated measure of diet has potential implications to assist schools, universities and other academic institutions in promoting a more holistic approach to wellbeing, in addition to helping to raise awareness for students regarding the importance of their wellbeing and how it can affect their education. Furthermore, a greater emphasis on stress management strategies during key points in the academic year such as examination schedules may prevent deterioration of diets.

Hypotheses

1. It is first hypothesised that the relationship between diet and academic performance is partially mediated by stress; with poorer diets causing increased stress and subsequently lower academic performance.
2. It is predicted that females will experience more stress than males.
3. It is hypothesised that sleep quality will significantly affect academic performance.

Research Methodology

Participants

Participants will comprise of undergraduate Psychology students from the University of Leicester receiving course credit for their co-operation. Students will be sitting semester one exams in early January and reflect on their experience over the past month, in addition to providing their exam results upon release in early February.

This convenience sample will include students from both year one and year two as this will (1) provide an insight into the effects of MCQ multiple choice based exams sat in year one versus essay based exam questions faced in year two, (2) the differences in levels of stress and subsequent management strategies between relatively new year one students facing their first university examinations compared with more experienced year two students, and (3) the differences of lifestyle habits including diet, sleep and time spent studying between undergraduates in year one and two.

Design

The cross-sectional study will have a non-experimental design based on a selfadministered online questionnaire. Participants stress levels will be measured by the Perceived Stress Scale, the sleep quality variable will be assessed using the Pittsburgh Sleep Quality Index, and the Short Form Food Frequency Questionnaire (SFFFQ) will provide a Diet Quality Score variable via the supplementary Excel spreadsheet, with academic performance to be determined by an average of the examination scores submitted by participants. Participant study habits will be assessed using a 5-point Likert scale (see Appendix G).

Materials

Materials to be used will include a questionnaire consisting of closed questions which incorporates the Perceived Stress Scale (see Appendix A), Short Form Food Frequency Questionnaire (see Appendix B) and supplementary Excel spreadsheet (see Appendix C), and the Pittsburgh Sleep Quality Index (see Appendix F). A consent form (see Appendix D) and participant information sheet (see Appendix J) will also be provided.

The application and submission of this data will be completed via the Leicester EPR system and JISC online survey platform. IBM's SPSS 25 will be used for data analysis, with the PROCESS macro (see Appendix H) for mediation analysis installed (Hayes, 2017).

The Perceived Stress Scale is the most widely used psychological instrument for measuring the perception of stress with evidence for validity shown in studies demonstrating higher PSS scores in failure to quit smoking, failure among diabetics to control blood sugar levels, as well as greater vulnerability to stressful life-event-elicited depressive symptoms (Cohen, 1994). Furthermore, prior research investigating eating behaviours among college and university students has also opted to use the same instrument (Daigle Leblanc & Villalon, 2008), as well as the measure being validated amongst UK university students with no significant gender bias evidenced (Denovan, Dagnall, Dhingra, & Grogan, 2017).

The Short Form Dietary Questionnaire is a non-quantitative short food frequency questionnaire which includes 25 food items and focuses on fruit, vegetables, fibre-rich foods, high fat and high-sugar foods, meat, meat products and fish. The tool has been validated against both an extensive Food Frequency Questionnaire (FFQ) and a 24-hour diet recall showing efficacy in ranking people according to diet quality (Cleghorn et al., 2016), as well as being used in a recent study assessing diet quality in non-alcoholic fatty liver disease patients (Bredin et al., 2019). There have also been variations of an SFFQ being effectively used in other nutritional studies such as assessment of legume intake in Scottish women (Papadaki et al., 2007). A supplementary Excel spreadsheet which calculates Diet Quality Score from the Short Form Dietary Questionnaire data will also be used (see Appendix C).

The Pittsburgh Sleep Quality Index (PSQI) is a self-report questionnaire assessing sleep quality over a 1-month period and is appropriate for use in this study with a similar timeframe. The PSQI has shown good internal consistency, test-retest reproducibility, and adequate construct and criterion validity (Popević et al., 2018).

The PROCESS Macro is an observed variable OLS and logistic regression path analysis modelling tool for SPSS and SAS. It is widely used through the social, business, and health sciences for estimating direct and indirect effects in single and multiple mediator models (Hayes, 2017). Version 3.4.0 released on August 12^th 2019 will be used in conjunction with IBM's SPSS 25.

Procedure

Ethical approval will need to be granted to the principal investigator before the study commences by the University of Leicester Psychology Research Ethics Committee (PREC). Written consent to utilise the SFFFQ in this study was granted by Leeds University (see Appendix E). Participants will be provided a consent form, be assured of anonymity, and be informed that they are able to withdraw up to the point of data submission by closing the browser window. Participants will access the questionnaire online via the Leicester EPR system and complete it on the JISC online survey website in exchange for course credit with there being no time limit for completion. Participants will be debriefed in full at the end.

Plans for Data Analysis

The Perceived Stress Score will be obtained by reversing responses to the four positively stated questions and then summing across all scale items. The Diet Quality Score will be calculated via the supplementary Excel spreadsheet after collected data from the SFFQ is entered. The exam scores will be provided by participants in the questionnaire with a mean exam score calculated afterwards. The Sleep Score will be calculated by summing the seven derived component scores from the Pittsburgh Sleep Quality Index to produce a global score ranging from 0 to 21.

Descriptive statistics and cross-tabulation will first be conducted to obtain sex, age and year of study demographical information. A linear regression between exam scores (DV) and diet quality score (IV) will be carried out to determine if there is significance of beta weight and ANOVA. The linear regression will be run again with the second predictor variable of Perceived Stress Score (IV) added. If the previous significant beta changes to insignificant, and the Perceived Stress Score has a significant beta weight, then the PSS will be considered a mediator for exam scores in relation to diet. The covariates of time spent studying and sleep quality will be controlled for within the model.

An independent samples T-test will be used to compare the differences between year one and year two students in lifestyle factors such as sleep quality, diet quality and time spent studying, in addition to testing the hypotheses that females experience more stress than males. All scale measures will be treated as interval data. The significance level accepted for the study will be 0.05.

References

Bredin, C., Naimimohasses, S., Norris, S., Wright, C., Hancock, N., Hart, K., & Moore, J. B. (2019). Development and relative validation of a short food frequency questionnaire for assessing dietary intakes of non-alcoholic fatty liver disease patients. European Journal of Nutrition, 10.1007/s00394-019-01926–5. https://doi.org/10.1007/s00394-019-01926-5s

Burrows, T., Goldman, S., Pursey, K., & Lim, R. (2016). Is there an association between dietary intake and academic achievement: a systematic review. Journal of Human Nutrition and Dietetics, 30(2), 117–140. https://doi.org/10.1111/jhn.12407

Cohen, S. (1994). PERCEIVED STRESS SCALE. Retrieved from http://www.mindgarden.com/documents/PerceivedStressScale.pdf

Cleghorn, C. L., Harrison, R. A., Ransley, J. K., Wilkinson, S., Thomas, J., & Cade, J. E. (2016). Can a dietary quality score derived from a short-form FFQ assess dietary quality in UK adult population surveys? Public Health Nutrition, 19(16), 2915–2923. https://doi.org/10.1017/s1368980016001099

Daigle Leblanc, D., & Villalon, L. (2008). Perceived stress and its influence on the eating behaviours of students at the University of Moncton, Moncton Campus. Canadian Journal of Dietetic Practice and Research : A Publication of Dietitians of Canada = Revue Canadienne de La Pratique et de La Recherche En Dietetique : Une Publication Des Dietetistes Du Canada, 69(3), 133–140. https://doi.org/10.3148/69.3.2008.133

Denovan, A., Dagnall, N., Dhingra, K., & Grogan, S. (2017). Evaluating the Perceived Stress Scale among UK university students: implications for stress measurement and management. Studies in Higher Education, 44(1), 120–133. https://doi.org/10.1080/03075079.2017.1340445

Gómez-Pinilla, F. (2008). Brain foods: the effects of nutrients on brain function. Nature Reviews Neuroscience, 9(7), 568–578. https://doi.org/10.1038/nrn2421

Hayes, A. F. (2017). Introduction to Mediation, Moderation, and Conditional Process Analysis, Second Edition : a Regression-Based Approach. Guilford Publications.

HE Student Data | HESA. (2013). Retrieved December 12, 2019, from Hesa.ac.uk website: https://www.hesa.ac.uk/data-and-analysis/students

Michels, N., Man, T., Vinck, B., & Verbeyst, L. (2019). Dietary changes and its psychosocial moderators during the university examination period. European Journal of Nutrition. https://doi.org/10.1007/s00394-019-01906-9

Mikolajczyk, R. T., El Ansari, W., & Maxwell, A. E. (2009). Food consumption frequency and perceived stress and depressive symptoms among students in three European countries. Nutrition Journal, 8(1). https://doi.org/10.1186/1475-2891-8-31

Papadaki, A., Scottà, J., Correspondence, A., & Papadaki. (2007). The British Dietetic Association Ltd. J Hum Nutr Diet, 20, 467. Retrieved from https://www.nutritools.org/pdf/tool-papers/T134%20Papadaki_et_al-2007Journal_of_Human_Nutrition_and_Dietetics%20(1).pdf

Popević, M. B., Milovanović, A. P. S., Milovanović, S., Nagorni-Obradović, L., Nešić, D., & Velaga, M. (2018). Reliability and Validity of the Pittsburgh Sleep Quality Index-Serbian Translation. Evaluation & the Health Professions, 41(1), 67–81. https://doi.org/10.1177/0163278716678906

Ross, C. E., & Wu, C. (1995). The Links Between Education and Health. American Sociological Review, 60(5), 719. https://doi.org/10.2307/2096319

Singleton, R. A., & Wolfson, A. R. (2009). Alcohol Consumption, Sleep, and Academic Performance Among College Students. Journal of Studies on Alcohol and Drugs, 70(3), 355–363. https://doi.org/10.15288/jsad.2009.70.355

Tanton, J., Dodd, L. J., Woodfield, L., & Mabhala, M. (2015). Eating Behaviours of British University Students: A Cluster Analysis on a Neglected Issue. Advances in Preventive Medicine, 2015, 1–8. https://doi.org/10.1155/2015/639239

Yaribeygi, H., Panahi, Y., Sahraei, H., Johnston, T. P., & Sahebkar, A. (2017). The impact of stress on body function: A review. EXCLI Journal, 16, 1057–1072. https://doi.org/10.17179/excli2017-480

Appendices

Appendix A – Perceived Stress Scale (PSS)

Appendix B Short Form Food Frequency Questionnaire

Appendix C Supplementary Excel spreadsheet to calculate DQS

Appendix F Pittsburgh Sleep Quality Index

Appendix G Time spent studying Likert Scale

Appendix H PROCESS Macro for SPSS

Appendix J Participant Information Sheet

Appendix K – Mediation model