Determinants of good academic performance among university students in Ethiopia: a cross-sectional study

Study design, setting, and period

This is an institution-based cross-sectional study conducted among Hawassa University students from December 1 – 28, 2020. The University is one of the oldest public and residential national universities found in Hawassa city, Sidama Region. It is located 278 km south of Addis Ababa on the Trans-African Highway for Cairo-Cape Town. By the year 2020/21, the university has enrolled 21,579 students: 7955 Females and 13,624 Males. In general, there is one Institute of Technology and 10 colleges that offer 81 undergraduate, 108 Master’s, and 16 Ph.D. programs.

Sample size and eligibility criteria

The sample size was calculated using Open Epi version 3.03 statistical software using percent of controls exposed (58%), odds ratio 0.63 [19], 80% power, and 95% confidence interval. By considering a 5 % non-response rate the final sample size was 659. All students who undergo their education in the selected departments and are available at the time of data collection were included in the study. Non-regular students, mentally and physically incompetent, and those who were not willing to fill out the questionnaire were excluded.

Sampling procedure

The study was conducted among Regular Hawassa University students. A multi-stage sampling technique was applied to select study participants. The simple random sampling (SRS) technique was used to select representative colleges and departments. Students were stratified based on their batch/academic year. The sample size was distributed using probability proportional to size (PPS). Thereafter, SRS was applied to pick the required sample size from the predetermined sampling frame.

Variables

Academic performance was the dependent variable. Independent variables include sociodemographic characteristics (age, gender, residence, parents’ education, family size, and faculty), individual factors (study hours, working after school, English language proficiency, sleeping hour, missing class, and entrance exam score), lifestyle and behavioral factors (substance use, breakfast, attending night club, and physical activity), and family and psychosocial variables (parents’ occupation, weight loss, and parent’s involvement).

Data collection tool and quality control

The data was collected using a structured, self-administered questionnaire. Four data collectors and two supervisors participated in data collection. The questionnaire was prepared by reviewing similar published articles [2, 7, 20]. It was translated from English to the local language, Amharic, and then back to English by an independent translator to keep the consistency of the tool. Pre-testing was done on 5 % of the samples (33 students) at Dilla University and necessary adjustments were considered following the result (i.e., ethnicity, income). The principal investigators trained data collectors and supervisors about the objective and procedure of the study. The data were daily checked for completeness, consistency, and clarity.

Measurement

Academic performance

Students who scored a cumulative GPA of 2.75 and above were categorized as “Good”, whereas those with a cumulative GPA of below 2.75 were categorized as “Poor” [7].

Habits

Participants who smoke at least one cigarette per day will be evaluated as smokers, and those who use more than one drink per day (any type of alcohol) will be considered alcohol consumers. Similarly, those who consume at least four glasses of tea and three cups of coffee per day will be accepted as those consuming tea and coffee, respectively [21].

Sugar intake

Excessive if individuals took 12 or more teaspoons of table sugar daily, moderate if 6 to 12 teaspoons; and restricted use if less than 6 teaspoons [22].

Extracurricular activities

Participation in school-based activities, i.e., sports, arts, and academic clubs [23].

Data management and analysis

Data were cleaned and entered into Epi-Data version 4.6 and SPSS statistical package version 25 was applied to perform all the statistical analysis. Cross-tabulation of variables was computed and the Chi-Square (X2) test was used to analyze the variables. Pearson Chi-Square test was reported for variables that fulfill the assumption of the X2 test. Whereas Fisher’s Exact Test was reported for variables having an expected count of less than five. Bivariable and multivariable logistic regression analysis were performed to identify independent predictors of academic performance. Variables with a p-value of ≤0.25 in the bivariable logistic regression were included in the final model. Descriptive statistics were used to describe the characteristics of participants. Adjusted odds ratios (AORs) with 95% confidence intervals were used to interpret the strength of association, and the Hosmer-Lemeshow goodness-of-fit was used to check for model fitness. A two-tailed p-value of ≤0.05 was considered to declare statistically significant.

Source