How can AI help us better understand FE recruitment challenges?

A new study by SchoolDash for Gatsby is using Large Language Models (LLMs) to explore trends in FE recruitment.

This new analysis by SchoolDash tracked job adverts from the Association of Colleges (AoC) job board over a 28-month period, collecting data on over 36,000 vacancies. The analysis then used ChatGPT to summarise the terms and conditions for roles, and spot when the same jobs were being advertised repeatedly.

Key findings

• The most advertised teaching roles were in construction, engineering, and health, with maths being the single most frequently mentioned subject. Teaching jobs tended to focus on courses at Levels 1, 2, or 3.
• Most roles were permanent and full-time. Part-time teaching jobs often specified smaller time commitments (e.g., 0.5, 0.6, or 0.8 of a full-time position), while part-time support roles tended to require more hours.
• Annual leave time varied a lot but typically ranged between 30 and 40 days.
• Salaries depended on the role. Support staff were usually offered average salaries of £20,000–£25,000 per year, teaching staff £30,000–£35,000, and leadership roles £40,000–£45,000 (this category encompasses a wide range of roles).
• Around 40% of adverts were likely reposts for the same vacancy, suggesting these roles are not being filled immediately.

There are significant advantages to using LLMs to explore recruitment trends as FE colleges often lack access to national data on recruitment activity, which limits colleges’ ability to adapt strategies and policymakers’ capacity to offer support. This research is beginning to fill this gap and is part of ongoing work to understand FE recruitment and retention.

For a full analysis by Timo Hannay, Founder of SchoolDash, visit: SchoolDash - Blog (November 2024)