TQB: Teacher Quality Bulletin

An easy fix? How technology can provide teacher demand data

See all posts

Schools and districts around the country are now making their final push to find strong teachers for positions that remain open. Unfilled teaching positions are stressful–for principals juggling the loss of capacity, for the other teachers in the school who must pick up the slack, but especially for impacted students and families. Yet most states lack good data about patterns in vacancies and demand for teachers, making it difficult to anticipate needs and plan for hiring. Further, given the localized nature of hiring, states often don't have a picture of the workforce needs across their state and worry about their capacity to undertake new, labor-intensive data collection efforts.

A recent CALDER working paper on job posting data by Dan Goldhaber, Grace Falken, and Roddy Theobold offers a promising, relatively simple option for how states can better understand staffing at the district- and school-level. Utilizing an automated tool that regularly pulled job postings from nearly 250 Washington school districts, the study's authors examined job posting data to understand variations in postings across subject areas and school characteristics and to explore this data as an indicator of patterns in teacher demand. The tool pulled data at both the district and school level twice weekly over the period of one year.

Consistent with past research on teacher turnover and hiring, the study found that certain subjects and school types have more vacancy postings than others. Specifically, special education and English Language Learner teaching positions and schools with high percentages of students of color (defined in the paper as including students who are Black, Hispanic, and American Indian/Alaska Native) had greater numbers of postings per full-time-equivalent (FTE) position.

Perhaps of greatest interest to state leaders, when the authors looked at actual school and district hires, they found that the posting data provided reliable information about teacher hiring patterns and needs. Put simply, the results from the job-posting analysis largely lined up with actual new hire data, especially at the district level.

This suggests that states and districts can look at job posting data to better understand teacher demand and hiring challenges. While the automated tool used by the researchers isn't ready to be distributed widely, the tool's creators are exploring the possibility of disseminating it down the road once they've validated it in other states. In the interim, it could also be recreated by a savvy staff with knowledge of Python coding, at an estimated cost in the tens of thousands of dollars (likely less expensive than other means of gathering this data from districts).

Increasingly, the field is recognizing the need for timely, accurate teacher workforce data to drive more strategic decision-making around teacher staffing. As the authors conclude, "Our job scraping method appears to be a low-cost strategy…states could employ to understand their staffing needs better and more quickly." The idea of having this kind of resource in the future could be welcome news to those currently feeling the stress of filling teacher vacancies before the new school year begins.