Assessment vicious cycle – How susceptible are primary teachers to subconscious stereotyping?
Research seems to show that teacher assessments can be shaped by assumptions regarding gender, family income and other areas – Tammy Campbell considers the scale of the problem, and asks what can be done to tackle it…
Are primary teachers biased in their judgements of the children they teach? Do they perceive some pupils to be more able than others, because of their background or characteristics? And do they assess one child as attaining more highly than another, simply based on the labels those children have been given? I’m sure most would say ‘no’ – yet while they almost certainly wouldn’t do these things consciously or deliberately, my research suggests that many teachers are prone to these biases. They happen because of unconscious stereotyping and cognitive shortcuts; processes that everybody – unintentionally – draws upon to some degree.
Stereotyping and biased judgements
Stereotypes are uniform judgements of a group of people that assume similarity among that group. For years, studies have looked at the ways that stereotypes are formed and how they may influence assessments in the classroom. For example, research has suggested that pupils from low-income backgrounds may be stereotyped as less able than children from wealthier families, regardless of their actual capacity or performance as individuals. It is this kind of bias that’s the focus of my work. I’m using data for around 5,000 Y2 children and their teachers in England, who are taking part in the Millennium Cohort Study. I look at disparities between teachers’ evaluations of pupils’ ability and attainment, and those same pupils’ performance on independent cognitive tests. Teachers in my study were asked to rate children in their class across a number of areas, including maths and reading ability. The same pupils then took standardised cognitive tests in maths and reading at home, supervised by survey staff. The teachers were not made aware of the children’s scores in these home tests, while survey staff knew nothing of the children’s schooling, or of their teachers’ opinions of them. When I looked at the teachers’ ratings of different groups of children, I found biases in judgements that seem to indicate stereotyping. For example, low-income children who got the same scores as higher-income pupils on the reading cognitive test tended to be rated at lower levels by their teachers. Similarly, pupils with any label of SEN were, on average, rated lower than equivalently-scoring peers with no attributed SEN. Black Caribbean children were also underrated, compared with white children.
Teacher judgements of pupils’ reading and maths ability additionally seemed to be based on gender stereotypes. Girls were evaluated more highly at reading than boys with the same score on the reading test, while boys were perceived to be at a higher standard in maths. These cognitive biases are crucial, because teachers are, of course, such key people in primary school children’s lives and education. For one thing, much primary attainment is teacher-assessed, meaning that any biased judgement can affect the levels awarded to pupils and their subsequent progress and opportunities. For another, the ‘Pygmalion effect’ means that the opinions and expectations a teacher holds of a child can influence everyday classroom interactions, and the pupil’s overall educational experience.
Where do these stereotypes come from?
Along with previous studies, my research makes it clear that there are biases and disparities in judgements of primary children, indicating that stereotyping takes place. What is less clear is how these stereotypes are formed. One possibility is that the way in which attainment is reported and monitored creates a vicious cycle. Schools and teachers are required to set targets based on pupils’ characteristics, and national statistics repeatedly describe gaps according to background factors such as income level, SEN, gender and ethnicity. This may unconsciously feed into a biased judgement that, for example, all low-income children are less able. Alongside this, previous studies have described situations where schools are not ‘allowed’ to allocate children’s achieved levels according to their actual performance. Instead, they are pressured to alter grades to fit in with pre-existing configurations, based on pupil ‘type’. This may also contribute to stereotyping by emphasising differentiated expectations. Alternative explanations for stereotyping have also been offered. One study, for example, suggests that the average performance of pupils in a group may be over-generalised to all children in the group. If, for example, some low-income children perform at a lower level on average than higher-income children, unconscious cognitive shortcuts may lead to teachers believing that all low-income children are less capable – with the result that more able low-income children may not be identified. Lastly, it’s possible that certain high profile government policies have had unintended effects that contribute to stereotyping. The Pupil Premium, for example, allocates funding to pupils from low-income families. It does so very ostentatiously and explicitly, while potentially conveying the message that these children are all, as a group, inherently less academic. This may strengthen a stereotype that says less should be expected from low-income pupils.
How can we tackle stereotyping?
It seems there might be a role for teachers, policymakers and researchers to play in addressing stereotyping and bias in our primary schools. Firstly, policymakers and politicians should consider the possible consequences of their initiatives and monitoring strategies. The implementation and effects of characteristics-based target setting and provision need to be evaluated, balancing the positive against the negative.
Secondly, while teachers may have limited immediate powers with regards to these types of processes, there is still room for change and development at the school- and teacher-level. Studies suggest that individuals can tackle their own tendencies to stereotype – initially, by becoming aware that these cognitive shortcuts may be taking place. Armed with this awareness, teachers can then challenge their own subconscious, while schools can make sure that biased assessments of pupils are avoided. The best ways of doing this, however, are not yet certain. This is partly because there’s a lack of recognition in government and policymaking regarding the prevalence and impact of human psychological processes throughout education. Information has not been shared publicly – or even collected nationally – regarding the means by which some schools might already be confronting and preventing stereotyping.
Can you help?
As a researcher, I feel that there are at least two essential steps to addressing and lessening stereotyping and bias in primary schools. First, we need to collect and collate information and opinions from teachers on how stereotypes may be formed. What are the other possible causes of the kinds of biased judgements that my study has shown? Can teachers point towards reasons other than, or in addition to, those I have suggested above? Secondly, it is probable that, at a local level, certain schools and teachers are already consciously working on initiatives and practices intended to counteract stereotyping. If this information is pooled, alongside further investigation into what can work, stereotyping could begin to be eliminated throughout primary education.
I’d like to end this piece with a request. If you are a teacher or work in a primary school, and if you have ideas about what causes or could eliminate stereotyping, please get in touch. I hope to take forward my work on biases to look properly at how they can be overcome.
Only with the input of the core people within education will I be able to do this in a grounded and useful way.
Tammy Campbell is a research student in quantitative social science at the Institute of Education
