Commonly cited figures on teacher attrition in Australia are not reliable according to a new research paper. It found there is no robust evidence to support claims that 30–50% of Australian teachers leave teaching within their first five years. In fact, it says, the teacher attrition rate in Australia is unknown.
The paper found that comments on early career teacher attrition tend to gravitate to one percentage figure with little consideration of important differences such as those between the primary and secondary teacher workforce, or between metropolitan, regional, and rural schools.
Attrition in recent Australian literature is generally portrayed as high, negative, and specifically related to the context of employment—lack of support, burn out —which ignores a number of other relevant contexts and issues such as short-term contracts and lack of ongoing positions, personal issues such as illness or family concerns, or choosing to leave to pursue an alternative career. [p. 2]
The paper examined the sources of figures on teacher attrition cited in academic studies and the media over the last 17 years and found little evidence for them. For example, a commonly cited figure is that up to 25% of teachers leave within their first five years. However, this figure appears to be based on only anecdotal evidence submitted by one state department to a review of teacher education in 2003. Other higher attrition rates commonly cited are similarly not based on any robust evidence.
…there is little actual evidence in Australia to confirm the high figures often quoted for the attrition of early career teachers in active employment. [p. 10]
There is some more substantial evidence for recently cited lower rates of teacher attrition. The NSW Department of Education regularly publishes figures on newly appointed teachers resigning within their first five years and these have been consistently about 8% over the past five years. However, the paper notes that these figures relate to teachers in permanent positions and do not include temporary teachers who make up about 16% of the NSW teaching force.
One of the factors that may contribute to over-estimates of teacher attrition is that teachers move from one school sector to another. As the paper says “…attrition at the school or sector level is not the same as attrition from teaching” [p. 9]. Analysis of the most recent Staff in Australian Schools (SiAS) survey in the paper showed that 18% of primary and 27% of secondary early career teachers were teaching in a different sector to the one in which they started. Teachers also move between states and from single-employer schools to others.
These data highlight the difficulty inherent in obtaining accurate attrition data for teachers in Australia. The SiAS data demonstrate that early career teacher attrition figures from a single employer do not represent attrition from teaching: a proportion will be teachers moving to other employers.
It would be disingenuous to claim that any single figure could refer to teachers leaving the profession because data are available which show that teachers move between states and sectors, and also leave and return… [p. 9]
Another factor confounding the data is that teachers often have career interruptions. For example, analysis of the SiAS data showed that about 9% of female teachers in their first five years of teaching had interrupted their career and half of these had done so for three years or more. The data also show that nearly half of all teachers who resign from teaching for personal or family reasons later return to teaching.
The paper says a national system of data collection to monitor actual attrition rates over time and for important subsets of the teaching workforce is an important and overdue innovation. It says location, gender, school socioeconomic status, sector, and schooling level are all likely to affect attrition rates in different ways. Data at this level have the potential to assist with the better targeting of policies to reduce attrition.
Obtaining accurate information on teacher attrition will require a national teacher registration or identification system. The paper notes that recent work for the Australian Department of Education and the Australian Institute for Teaching and School Leadership (AISTL) has recommended the further development of a national teacher workforce dataset, including a unique teacher identifier.
The paper is published in the Australian Journal of Education .
Trevor Cobbold