Abstract:

In the quest for knowledge-based economies, many countries especially developing countries face challenges with low throughput rates within their higher education systems. South Africa, like many others, aims to build a knowledge-based economy but contends with high dropout (at-risk students) rates among first-year students and prolonged completion times for degrees. Early identification of at-risk students is crucial in addressing this issue. While previous studies have focused on indicators such as class attendance and summative assessment marks, the evolving landscape of education, particularly with the advent of the "New Normal," prompts a reconsideration of methodologies. While most administrative reports tend to put pressure on staff performances (quality of lecture notes, punctuality in class, use of technology, teaching style, assessment and flexibility of assessment, timing of assessment among many others) as a means to improve throughput, this study builds from a previous case study attempting to link multiple and early course and teacher evaluation to student performances as a potential determinant of at-risk students. This study monitored the variation of multiple teacher evaluations within the semester by students in different agricultural courses and its relationship to student performances (class averages) as a means to identify potential at-risk students. The study was carried out at the Department of Agriculture in a rural-based University, using students from five different courses, Animal Health (CS1), Ruminant Nutrition (CS2), Pig and Poultry Production (CS3), Applied Pig and Poultry Production (CS4) and Principles of Animal Production (CS5). Three evaluations (early (E1), mid (E2) and late (E3)) were done online using Moodle. Lecturers employed mostly fundamental principles, problem-solving, and critical thinking concepts for teaching using hybrid method. After each evaluation, lecturers had the opportunity to go through student feedback to improve their teaching. Each lecturer was scored on lecturer's knowledge of course (LKM), Lecturer organization (LO), Lecturers pace (LP), Lecturers ability to explain (LEXP), Lecturers availability (LAV), Lecturer encouraging creating thinking (LCT) and Lecturer overall rating (LOR) from strongly disagree to strongly agree. Chai-square of SPSS was used to compare differences among and between the normal data generated. The results generally varied (P<= 0.05) from one module to another. Most of the parameters scored observed a slight decrease (P<= 0.05) in students' percentage scores from E1 to E3 (CS3 and CS1) while CS4 showed the worst drop in percentage scores from E1 to E3, especially on the following paraments (LOR, LP and LEXP). However, CS1 feedback did not change but CS4 scores generally improved. The class average test marks (Test-1, Test-2 and Test-3) hosted during each period of evaluation also slightly decreased but for CS2 that was higher at E3. This study showed that students' responses vary with time and may be influenced by their class performances (test marks) despite the lecturer's several mitigation strategies. The study also extrapolated that, if students (those dropping class average marks) can be identified early at E1 and E2, it can be a major step towards the identification of at-risk students and the development of a better mitigation programme that can potentially reduce throughput.

Keywords:

Teacher evaluation, at-risk, Agriculture.