Extension program evaluations often present opportunities to analyze data in multiple ways. This article suggests that program evaluations can involve more sophisticated data analysis approaches than are often used. On the basis of a hypothetical program scenario and corresponding data set, two approaches to testing for evidence of program impact are compared. These approaches are (a) a bivariate approach involving contingency table analysis (chi-square, Kendall's tau tests) and (b) a multivariate approach involving logistic regression. Both approaches address the primary evaluation questions, but the multivariate approach introduces additional variables, allowing for a more comprehensive understanding of program dynamics. Multivariate approaches can enhance insights about programs and increase opportunities for dissemination of research results.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 4.0 License.
Braverman, M. T. (2016). Better Crunching: Recommendations for Multivariate Data Analysis Approaches for Program Impact Evaluations. The Journal of Extension, 54(3), Article 25. https://doi.org/10.34068/joe.54.03.25