2019 NAEPSDP Conference

Introduction/Theoretical Framework/Review of Literature
In 1980, Borich argued for using Kaufman’s (1972) concept of needs as a gap between present and desired states as the foundation for identifying training needs. Borich (1980) adapted Kaufman’s work to define a training needs “as a discrepancy between an educational goal and trainee performance in relation to this goal” (p. 39). Borich suggested using competency statements as the foundation of needs assessment instruments. Participants could then rate the relevance of each competency to their job functions and to what level they had attained each competency, ultimately allowing for the calculation of what Borich called mean weighted discrepancy scores (MWDS). The MWDS of all assessed competencies could be rank-ordered to determine priority training needs. The Borich model has often been applied within agricultural and extension education (e.g. Conner, Dev, & Krause, 2018).
Limitations of the Borich (1980) model exist despite its popularity within the literature. The Borich model relies upon the calculation and use of means based on single Likert-type items. Though generally accepted during the time period during which Borich developed his approach, contemporary scholars (e.g. Bishop & Herron, 2015; Boone & Boone, 2012) have repeatedly raised concerns about the use of parametric statistics for handling single Likert-type items. More practically, the Borich method is not complicated but data analysis is time intensive; researchers need to painstakingly enter dozens of formulas into Excel, or use custom calculators, such as the one developed by McKim and Saucier (2011) for use in Excel. A better method to assess organizational training needs is desirable.

Research/Guiding Questions
An exploratory study was conducted to evaluate the appropriateness of two different data analysis approaches for use with self-report data, the traditional Borich method and a newly proposed method. Specifically, the objectives were to identify the priority training needs using each data analysis technique, and to compare the priority training needs resulting from the application of each approach.

Methods and/or Data Sources
This study used existing data from a professional development assessment of Utah State University Extension agents. Data collection occurred in April 2019 using Qualtrics, and the sample size consisted of 87 faculty (n = 87). Data were gathered from a closed-ended questionnaire consisting of Extension competency items found in the literature. The response options followed the Borich (1980) framework. On a 5-point Likert-type scale, respondents were asked to rate (a) their perception towards the importance of each competency i.e. perceived importance (I), and (b) their perception of their ability to perform the competency i.e. perceived ability (A).             
The traditional method of calculating mean weighted discrepancy scores described by Borich (1980) was followed. Then, the proposed Rank-Order approach was applied to the same data set. The same survey instrument was used for either method.

In brief, the proposed contemporary Rank-Order approach to identify priority competencies for professional development follows three steps. The first step requires the researcher to perform the Wilcoxon Signed-Rank Test to determine number of occurrences when: (a) respondents’ perceived ability (A) exceeds a competency item’s perceived importance (I), (b) A equals I, or (c) I exceeds A. The second step is to convert the number of occurrences within each category into a percentage. The third step is to assign a relative weight to each category. Weights were selected based on three assumptions: (a) professional development training is necessary when A < I, (b) training is not required when A > I, and (c) professional capacity is sufficient or at equilibrium when A = I. The use of weights allows a weighted Rank-Sum Score (RSS) to be calculated for each item. The RSS is a standardized ranking of professional development needs that ranges between -100 to 100, with lower scores indicated a greater priority for professional development.