Tests used to select people who make good doctors could work for teachers too

Great at the job: can we test for that? Primary teacher via Shutterstock/oliveromg

Television documentaries such as Tough Young Teachers and Educating Yorkshire have shown how a teacher’s “soft skills” are key for success in the classroom – both for novices and veterans. Audiences followed teachers who struggled, coped, and sometimes failed at their job, not primarily because of how well they knew their subject, but because of the personal attributes they brought with them into the classroom.

The department for education’s current Troops to Teachers programme, where former service personnel are encouraged to train as teachers, is built on the notion that “non-cognitive attributes” such as interpersonal skills and resilience are necessary for effective teaching.

One way of trying to hit upon which applicants to teacher training courses might have these skills is to emulate the way doctors are selected for medical training. Our research is showing that the kinds of tests used to assess whether a person would make a good doctor can work for teacher training too.

At the moment, teacher application processes do try and assess whether potential teachers have the non-cognitive attributes associated with successful teaching, usually through interviews and personality tests. But current methods may not be good enough.

Although the “soft skills” of novice teachers may improve with training and experience, teaching effectiveness varies within cohorts of new teachers, and effectiveness remains reasonably stable over time. Research at the US Calder Center has found that the most and least effective new teachers tend to stay in the same position relative to their peers even after several years of teaching.

A rigorous process that includes assessment of “suitability to teach” is one of the selection criteria listed by the UK’s National College for Teaching & Leadership. But most of the available non-cognitive tests are based on personality models that are only weakly related to teaching effectiveness.

The medical way

But assessing whether a person is suitable to teach is a challenging task, and has only received limited attention from researchers. In sharp contrast, the selection process for medical training has been the subject of a lot of research, and enjoys a robust evidence base that may be useful for developing selection processes for teacher training.

In selection processes for potential doctors, applicants are tested for their interpersonal skills. This is important because the quality of doctor-patient relationships has been found to influence patients’ adherence to medical treatments.

For entry into most medical schools in the UK, applicants complete a situational judgement test (SJT) that assesses non-cognitive aptitudes such as empathy, integrity, teamwork, and resilience. Scores from SJTs are often better predictors of success in medical training and subsequent practice as a doctor than scores on cognitive tests.

These tests use scenarios to assess implicit traits – the spontaneous evaluation of a situation – in a workplace setting. The tests present applicants with a situation accompanied by possible responses to choose from. For example:

You are reviewing a routine drug chart for a patient with arthritis during an overnight shift. You notice that your consultant has inappropriately prescribed a drug daily instead of weekly. Rank in order the following actions: (a) Tell the patient, (b) phone the consultant at home…

The tests measure judgement and “general domain knowledge”, or the internal rules people have about how to behave in a situation. The rules originate from personal dispositions and past interactions with parents and peers.

Research using SJTs for selection into medical training show that the tests are reliable and valid predictors of aptitude for being a doctor. Longitudinal research carried out over a number of years shows that non-cognitive attributes measured using these tests predict subsequent work performance nearly a decade after initial entry into training. For example, they could predict whether a person would get higher ratings from a supervisor for technical and interpersonal skills.

Designing a test for teachers

Our own research conducted over the last two years has focused on developing tests to identify the non-cognitive attributes of prospective teachers. Guided by a thorough job analysis, our work has identified three relevant competencies necessary for being a teacher. These were “empathy and communication”, “organisation and planning” and “resilience and adaptability”.

We have developed our own pilot SJT consisting of more than 30 scenarios generated from critical incidents that teachers might face in the classroom. Experienced teachers helped us to design the scenarios.

Early results show our pilot test is statistically reliable and positively correlated with interview criteria. This means that the test may be useful for initial screening of personal attributes. Most candidates agreed that the test was clearly relevant to those applying for teacher training. Initial results demonstrate the feasibility of the approach and provide a robust rationale for extending the work on a wider platform.

Other tests in the pipeline

The research has now been extended to Australia, Canada, and Finland, with a number of teacher-training providers and education authorities involved in development of locally-relevant content.

Other researchers in education and applied psychology are addressing the selection of prospective teachers. In Australia, John Hattie and his team at the University of Melbourne are designing selection instruments based on Five Factor personality models. The results from their instrument can also be used to help prospective teachers identify areas that might need further work.

The advantages of using tests like SJTs as part of the selection process for teacher training include more accurate predictions of who will be successful, cost efficiencies and more reliability in application decisions. There is clearly more work needed, but a close look at approaches used in other fields, like medicine, may help improve the selection of prospective teachers.