Estimating a Performance Standards Adjustment Model
for Workforce Programs that Provides Timely Feedback
and Uses Data from Only One State

Upjohn Institute Working Paper 09-144

Timothy J. Bartik, Senior Economist
W.E. Upjohn Institute for Employment Research

Randall W. Eberts, President
W.E. Upjohn Institute for Employment Research

Ken Kline
W.E. Upjohn Institute for Employment Research

July 2004
(revised January 28, 2009)

JEL Classification Codes: J64, J68

The purpose of this paper is to describe a methodology for adjusting performance standards for workforce programs offered by local workforce areas (LWAs). By performance standards adjustment, we mean a model that uses a statistical approach to attempt to better measure the relative performance of different local workforce areas in providing workforce system customers with “value added” in terms of the system’s desired outcomes. Our paper’s approach has four distinguishing features. First, the performance standards are based on the common measures proposed by the U.S. Department of Labor, which include short- and longer-term employment outcomes. Second, the model is estimated using data from only one state, which allows each state greater flexibility in adapting the adjustment model to the state’s needs and available data. Third, the model is estimated using data on individual customers, which offers some estimation advantages, particularly when data from only one state is available. Fourth, since some of the common measures are not available until long after the program year is completed, we include real-time predictions of the current performance of the LWA and an assessment of whether or not it will meet its performance standards when the common measure data is eventually available. This more timely feedback on performance provides administrators the opportunity to better manage their operations and offer services that best meet the needs of their customers.

Authors’ note: This paper formed the basis for a performance reporting system that was developed for the State of Michigan. The system is referred to as the “Value-Added Performance Improvement System” (VAPIS) and has been implemented for more than a year. The system adjusts the U.S. Department of Labor’s common measures for WIA workforce programs for factors that are beyond the control of local administrators, such as the characteristics of program participants and local labor market conditions. The common measures include three labor market outcomes: entered employment, job retention, and earnings levels. By making these adjustments, the common measures more closely approximate the value added that the workforce programs contribute to the labor market outcomes of participants. VAPIS also provides a short-term forecasting component that assists local workforce administrators in understanding the likelihood of that their current participants will find and retain jobs. Because of the long lag in reporting common measures, local administrators have little systematic knowledge of their performance. VAPIS tries to fill that gap.

Full text | Institute Home Page | Back to Working Papers