The use of progress and attainment data is ubiquitous in schools, and informs a plethora of decisions made by school leaders - which is hardly surprising in a climate that sees schools judged and compared according to their Progress 8 score.

The use of data has been cited as a workload issue and, after a government-commissioned review by the independent teacher workload group in 2016, Ofsted took the decision to stop examining internal data relating to progress and attainment.

But could the detrimental impact of data also extend to the learning of our pupils? Arguably, data may - with its far-reaching sway on myriad aspects of the education students receive - have a far more pernicious influence than a neatly colour-coded spreadsheet might at first suggest.

By “data”, I am referring to assessment-linked or graded attainment data, which, while no longer a feature of inspection, continues to pervade school life. Assessment data can evidently be useful for gauging learning and for reflecting on gaps in understanding and learning needs. Unfortunately, though, it can also be used to place children into categories that determine the educational experience they receive.

For instance, it is commonly used for setting or streaming, with only a small minority of secondary schools eschewing attainment-based grouping. This means that, in many schools, certain students will be differentiated into the bottom or lowest sets in Year 7 based on their Year 6 Sats results or after tests undertaken right at the start of the year.

Before they have even had a chance to begin their secondary school learning in a subject, they are classified as “lower ability”. And, in some cases, this data will then follow them and influence the school experience they receive for the next five years.

This is because, once children are in a low or bottom set, they are - unfortunately - less likely to be exposed to higher-level ideas and teachers’ expectations are likely to be lower. They are seen as low ability, creating a self-fulfilling prophecy of lower achievement.

Teacher expectations

The Pygmalion effect, developed by psychologist Robert Rosenthal and school principal Lenore Jacobson, in their 1968 study of teacher expectations and pupil development, suggests that the higher the teachers’ expectations of students, the more likely students are to do well. This is a finding that has been supported in multiple subsequent studies.

Setting students into attainment-based groupings is very likely to affect our expectations of the group, even if teachers are not consciously aware of the way their teaching changes. For instance, research by education professor Christine Rubie-Davies has shown that implicit differences in teachers’ expectations of students’ abilities at a whole-class level are inferred by students and can affect their achievement.

The detrimental impact of setting is supported by a wealth of attainment-linked research, including an extensive meta-analysis by the Education Endowment Foundation in 2018. According to the EEF, disadvantaged pupils, in particular, tend to make less progress in streamed ability groups. When we consider that these same disadvantaged pupils are, as a result of contextual factors, more likely to achieve lower results in their Sats or Year 7 tests in the first place, and therefore to be placed in lower sets by schools that stream on entry, the problem becomes glaringly obvious. Existing disadvantage or underachievement is entrenched by data-based setting.

Compounding an already dire data scenario in the form of setting is schools’ use of data-linked flight paths. Flight paths are a form of internal data progress journey that students are expected to follow, which are often based on data relating to students’ prior attainment. In some schools, students are told their projected grades for the end of Year 11 in Year 7, despite having barely begun the secondary school curriculum. Or they may be asked to stick their target grades on the front of their exercise books as a regular reminder of their current level and the level they are aiming for. At parents’ evenings, students might expect to be told their expected progress against a preset flight path.

In some classrooms, students may even be given differentiated learning objectives based on data, exacerbating existing inequalities and compounding learning gaps by providing a different learning experience for students.

Similarly, teachers may refer to existing data or students’ target grades when giving feedback or setting a task, in an attempt to personalise learning - which may inadvertently hold students back from aiming for the top grades. Again, the Pygmalion effect can help to explain why such use of data to inform our preconceptions of students’ abilities may be fundamentally unhelpful - and even detrimental to the progress of many students. Students who have achieved lower grades earlier in their school career have the odds stacked against them, as teachers’ expectations and their own will mean they are unlikely to significantly exceed their target.

Sharing data on prior attainment with students and using this to shape their learning is not only ethically questionable but is also ineffective. Being labelled as “high prior attaining” might have a positive impact on a select few, but the use of predicted and target grades calculated by prior attainment arguably has a cost far too high on the rest of the cohort to justify their extensive implementation.

Consider the costs

Education professor John Hattie notes that, even when an educational practice might have a positive effect in some circumstances, we must consider the costs as well as the benefits of interventions. What is the cost to most of our students if we use data to emphasise differences in pre-existing ability?

We need to put ourselves in the position of those children for whom failure is embedded in their mindset when they see a string of low targets and predicted grades at the beginning of September, and reflect on whether we can justify this practice in light of the cost.

Even withholding the grades and numbers themselves, giving students grades such as “meeting expected progress”, “above” or “under” at report checkpoints is likely to be damaging, as this means we are still, as teachers, having to evaluate students’ current achievement based only on their prior grades.

Even though, in 2019, Ofsted chief inspector Amanda Spielman spoke to the National Governance Association about her misgivings in relation to flight paths - calling for us to focus on progress in terms of each individual’s learning of the curriculum rather than whether they are hitting a set target grade - the message still hasn’t filtered through to many schools in the practices they use to report data and progress to students and parents.

If we base our aspirations for children on their previous attainment, we inevitably restrict what our students will achieve. Essentially, even if we tell students that they have the power to improve their grades, we risk teaching them to have a fixed mindset and a rigid viewpoint of their own ability if we have already shown them that they are in a certain category based on their target grade.

The use of attainment data to shape targets and learning may also be fundamentally damaging to self-confidence for lower-attaining students and detrimental to students’ academic self-concept.

Academic self-concept is a term coined by renowned psychotherapist Carl Rogers, which refers to students’ own perceptions of how competent or clever they are in relation to others. Research suggests that academic self-concept develops as early as preschool or key stage 1, and is developed by their experiences throughout their school life.

A 2009 study found that high-school pupils with a more positive perception of their own academic abilities were more likely to expect to go to university, even after controlling for ability and students’ characteristics. And multiple research studies have indicated a correlation between academic self-concept and performance.

A progress-driven system means that we have, in some cases, become laser-focused on ensuring every student attains the grade that their previous data might imply they are meant to achieve. Yet this way of framing education is quintessentially counterproductive, as it compounds academic and social inequalities.

A child who has performed poorly in Year 6 will be predicted to do less well than many of their peers in Year 11. But that Year 6 child may, for their primary school years, have been exposed to a raft of contextual and social factors influencing their lower attainment.

When we use this form of data-based prediction to judge students, therefore, we embed differential factors further and cement social disadvantage.

Although there is a clear link between prior attainment and future grades, we should be striving to close the gap rather than widen it further. By regularly reminding students of their data-driven target grade, we harm the chances of those students who most need pushing to reach the top and escape the constraints of their prior attainment.

While there is no avoiding the fact that our government compares the efficacy of schools in terms of the attainment progress students make, in the form of Progress 8, there is no need for prior data to shape so explicitly and fundamentally the learning experience that students receive.

Rather than focusing only on analysing patterns and trends on a numerical level, we must ensure we take time to consider the potentially harmful impact of data-driven policies on students both on an educational and an interpersonal level.

Megan Mansworth is an English teacher and PhD student. She tweets @meganmansworth

This article originally appeared in the 7 May 2021 issue under the headline “Why we must divert from flight paths”