« Back to Results
Hyatt Regency Atlanta, Hanover B
Hosted By:
Labor and Employment Relations Association
Improving Selection of Job Applicants: Harnessing Resume, Interview and Recommender Signals for K12 Teaching
Paper Session
Friday, Jan. 4, 2019 2:30 PM - 4:30 PM
- Chair: Barbara Biasi, Yale University
Using Machine Learning to Translate Pre-Hire Work History into Predictors of Performance and Retention
Abstract
Work history information reflected in resumes and job application forms is commonly used to screen job applicants; however, there is little consensus as to how to systematically translate information about one's past into predictors of future work outcomes. In this paper, we apply machine learning techniques using job application form data (including previous job descriptions and stated reasons for changing jobs) to develop measures of work experience relevance, tenure history, and history of involuntary turnover, history of avoiding bad jobs, and history of seeking better jobs. We empirically examine our model on a longitudinal sample of 16,071 applicants for teaching positions in the Minneapolis Public School district and predict subsequent work outcomes including student evaluations, expert observations of performance, value-added to student test scores, voluntary turnover, and involuntary turnover. We find that work experience relevance and a history of seeking better jobs are linked to positive work outcomes, whereas a history of avoiding bad jobs was associated with negative outcomes. We also quantify the extent to which our model can improve the quality of selection process relative to conventional methods of assessing work history, while lowering the risk of adverse impact.Making the Cut: The Effectiveness of Teacher Screening and Hiring in the Los Angeles Unified School District
Abstract
Despite evidence that many schools and districts have considerable discretion when hiring teachers and the existence of an extensive literature on teacher quality, little is known about how best to hire teachers. This is, in part, because predicting teacher quality using readily-observable teacher characteristics has proven difficult and there is very little evidence linking information collected during the teacher hiring process to teachers' outcomes once they are hired. We contribute to this literature using data from a recently-adopted teacher screening system in the Los Angeles Unified School District (LAUSD) that allows applicant records to be linked to student and teacher-level data for those teachers who are subsequently employed in the district. We find that performance during screening, and especially performance on specific screening assessments, is significantly predictive of applicants' eventual employment in LAUSD and teachers' later contributions to student achievement, evaluation outcomes, and attendance, but not to teacher mobility or retention. However, applicants' performance on individual components of the screening process are differentially predictive of different teacher outcomes, highlighting potential trade-offs faced by districts during screening. In addition, we find suggestive evidence across time and between districts that the shift to the new teacher screening system improved hiring outcomes in LAUSD relative to other similar districts and schools.Direct from the Source: To What Extent Do Ratings of Teacher Applicants By Professional References Predict the Likelihood of Being Hired and Job Performance?
Abstract
The composition of the teacher workforce is an important determinant of student outcomes, but most academic research has focused on interventions designed to improve incumbent teachers (through professional development or financial incentives, for example). Far less attention has focused on how to improve the front-end of the teacher pipeline, and in particular, the teacher hiring process. We use data from the Spokane Public School system and a novel survey of the professional references of teacher applicants to predict the extent to which the ratings by the professional references predict the likelihood of being hired, various measures of job performance, including value added, and retention.Discussant(s)
Jen Brown
,
University of Utah
Peter Cappelli
,
University of Pennsylvania
JEL Classifications
- M1 - Business Administration