Development of a food knowledge test for first-year students at a University of Technology in the Western Cape, South Africa

Abstract

Objective: To develop a valid and reliable test to determine the food knowledge of first-year students at a university of technology (UOT) in the Western Cape, South Africa. Design: Two preliminary food knowledge tests were developed covering the content domains, namely fruit and vegetables and fats and oils, as these topics attend to the areas of concern in the dietary intake of young adults. Both tests consisted of multiple-choice questions and incomplete statements compiled following the test item construction rules. The items of both tests were evaluated by experts in the field of food science and nutrition to ensure item content and face validity. Both tests were independently administered to two sample groups represented by knowledgeable students (having food-orientated subjects as part of their course) and less knowledgeable students (not having food-orientated subjects as part of their course syllabus) at a UOT in the Western Cape, SA for the item analysis and test construct validity and reliability determinations. The second preliminary test incorporating three response alternatives was developed as the number of items retained after the item analysis of the first preliminary test incorporating four alternatives was less than the envisaged number of about 20 items. Results: After the item analysis of the first preliminary test (n = 72 items) only 10 and 13 items were respectively retained after two administrations to two sample groups. The second preliminary test (n = 135 items) completed by knowledgeable (n=119) and less-knowledgeable (n=91) student sample groups delivered 74 items after its item analysis across the two content domains, namely fruit and vegetables (n=49) and fats and oils (n=25). The test was found valid, with a significant difference (p < 0.001) in the knowledge scores between the two sample groups (Mann-Whitney test, z = 9.74) and highly reliable (KR20 and Cronbach's alpha= 0.934). Conclusions: The test being a valid and reliable assessment tool can be used to determine the food knowledge of first-year students at a UOT in the Western Cape, SA, across the two content domains to establish if guidance and possible teaching is necessary to equip them with basic food knowledge to support them in their food provision.