Fornito, Matthew C.2014-05-282014-05-282014-05-27vt_gsexam:32127969_Fornito_Thesis.pdfhttp://hdl.handle.net/10919/48126Cross-cultural research requires tests of measurement invariance to determine if different populations have equivalent latent constructs. A psychometric assessment of the Agricultural Production Perspectives Scale (APPS) was necessary to determine the validity of the scale constructed and whether data from multiple countries could be compared. Using 918 farmers sampled from Kenya, Uganda, Mali, and Lesotho, I conducted exploratory factor analysis and scale reliability tests to determine whether the item loadings and factors were equivalent across populations. No factor structure could be obtained across country or agroecological populations. The data were reanalyzed within each agroecology to determine localized factor structures. Results indicate that a market driven factor and agrarian driven factor tend to emerge across multiple agroecologies suggesting some emergence of latent variables. Recommendations for scale revisions are included to increase reliability and measurement invariance.ETDIn CopyrightOIREDSANREMAgroecologiesSmallholdersFarmersKenyaUgandaLesothoMaliExploratory Factor Analysis (EFA)Measurement Invariance (ME/I)Thesis