Files in this item

FilesDescriptionFormat

application/pdf

application/pdfBUKENYA-DISSERTATION-2017.pdf (3MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Development and validation of a general nutrition knowledge questionnaire for adults in Uganda
Author(s):Bukenya, Richard
Director of Research:Andrade, Juan E.
Doctoral Committee Chair(s):Chapman-Novakofski, Karen M.
Doctoral Committee Member(s):Grigsby-Toussaint, Diana S.; Muyonga, John H.
Department / Program:Nutritional Sciences
Discipline:Nutritional Sciences
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Nutrition knowledge questionnaire adults
Abstract:Knowledge is an essential element in several theoretical frameworks aimed at explaining human behavior and nutrition practices among individuals, and in institutions and communities. Head teachers are key change agents in a school environment that most likely would influence any nutrition intervention. Basic nutrition knowledge may enable them to improve school nutrition practices. In general, nutrition knowledge has been poorly evaluated in Uganda and there is no data on nutrition knowledge of school stakeholders such as head teachers and teachers. This dissertation study aimed at developing a general nutrition knowledge questionnaire (GNKQ) to obtain valid and reliable data using psychometric measures from head teachers in elementary schools in Uganda. The first draft of GNKQ comprised of five knowledge dimensions and 133 items evaluating basic nutrition knowledge on Expert recommendations (16 items), Food groups (70 items), Selecting food (10 items), Relationship of nutrition and disease (23 items) and Food fortification (14 items). The draft GNKQ was reviewed twice by a panel of five experts. Face validation took place in between expert reviews and involved independent reviews and three focal groups with 15 head teachers and 12 health workers from Kampala. Experts finally agreed (Content Validity Index = 0.97, and Gwet’s AC1 = 0.96) with revisions that items in the questionnaire were relevant to evaluate general nutrition knowledge of adults in Uganda. In the initial pilot testing of the survey instrument, 40 head teachers from schools in Kampala along with 40 nutrition students and 37 engineering students from Makerere University were recruited. The participants filled the questionnaire twice within two weeks. Results showed that the GNKQ had acceptable internal consistency (Cronbach α = 0.95), test-retest reliability (r = 0.89), and concurrent validity, in which the nutrition knowledge scores of nutrition and engineering students obtained using the instrument were significantly different (67 ± 5 vs. 39 ± 11; p < 0.001). Only the domain on Expert recommendation had unreliable data (α = 0.51, test-retest, r = 0.55). Results from initial pilot were used to review items in the study. The final draft comprised of 137 items to evaluate basic nutrition knowledge on Expert recommendations (16), Food groups (67), Selecting food (10), and Relationship of nutrition and disease (22) and one on Food fortification (22). The pilot was followed up with a larger sample of head teachers (n = 255) who filled out the drafted GNKQ. One hundred and thirty-six head teachers completed their surveys the second week of the retest. The overall internal consistency was α = 0.89 and 0.92 at time one and two, respectively on 94 items. Results from test-retest reliability indicated that two domains, Expert recommendations (Intraclass Correlation Coefficient; ICC = 0.64) and Selecting food (ICC =0.41) were not acceptable (r < 0.7 and ICC < 0.7), and therefore, their items were removed from the proceeding analyses. The other domains had acceptable test-retest reliability: Food groups (ICC = 0.9), Relationship of nutrition and disease (ICC = 0.91), and Food fortification (ICC = 0.95). To ascertain external validity, the GNKQ was applied to a sample of agricultural extension agents (n = 80) living in the Kiboga and Kyankwanzi districts. The questionnaire (85 items) had adequate internal consistency (Cronbach α = 0.93), while 52 items were removed from analyses because they had poor item discrimination (r < 0.2). With exception of Selecting foods, all other nutrition knowledge domains had adequate internal consistencies (Cronbach α > 0.7). The ICC for the whole questionnaire was 0.9. All the remaining nutrition knowledge domains had adequate test-retest reliability (ICC > 0.7). Finally, the Diffusion of Innovation framework was applied in an effort to demonstrate that nutrition knowledge of head teachers would influence adoption (awareness and implementation) of the Guidelines for School Feeding and Nutrition Intervention Program of Uganda (GSFNIP). Results from correlations and multiple linear regression analyses revealed that head teachers’ basic nutrition knowledge was related to awareness (r = 0.2, p < 0.01; β = 0.158, p = 0.006), but not to the degree of implementation (r = -0.04, p > 0.05; β = -0.032, p = 0.621) of the GSFNIP. Factors associated with the school environment such as being a private school, involvement of parents, higher number of qualified teachers, and more school materials predicted a higher degree of implementation of the GSFNIP. These studies show that data collected using the GNKQ has acceptable content, face, concurrent, predictive, and external validity, and reliability. This is the first of its kind general nutrition knowledge questionnaire for adult population in Uganda. This tool can be used to evaluate general nutrition knowledge of head teachers and extension workers. Future studies should continue the validation of this survey instrument in other adult populations.
Issue Date:2017-07-13
Type:Text
URI:http://hdl.handle.net/2142/99103
Rights Information:Copyright 2017 Richard Bukenya
Date Available in IDEALS:2018-03-02
Date Deposited:2017-08


This item appears in the following Collection(s)

Item Statistics