The Prevalence of Students' Guesswork in Multiple-Choice, Matching Items, and True-False Test Formats: Implications for Academic Performance in Tertiary Institutions in Tanzania

Abstract

This study examines the prevalence of guessing impact on performance in multiple-choice (MC), matching items (MI), and true-false (TF) test forms among Tanzania's university students. Since educational tests increasingly rely on objective test forms, the impact of guessing on test scores and student performance is worth understanding. In a correlational research design, data were collected from 121 participants from various programs in three institutions of higher learning in the Dodoma region. Data were collected using questionnaires designed to quantify students' guessing behaviour and perceived effects on examination performance. Data analysis using the Statistical Package for the Social Sciences (SPSS) revealed that 76% of students indicated guessing during examinations, and only 24% said they never did. The analysis also uncovered a statistically significant weak negative correlation (rs = -0.281, p = 0.002) between students' confidence in guessing and their belief about its impact on examination marks. This means that students who are more confident in their guessing strategies are less likely to view guessing as detrimental to their performance. The findings also indicate that guessing can threaten the validity of test outcomes such that correct answers do not accurately represent students' actual knowledge. It therefore implies that in this research, instructors should employ tactics that minimise guesswork behaviours.