Why we need the SATs (pt1)

Hello everyone, I hope you're all enjoying the joys of slowly reverting back to normal.  Whatever that is.  I'm going to dive straight in this month because I have a feeling it's going to be a long one.  So, without further ado, let's have a look at:

Assessment this Summer

Back in January of this year, the UK Prime Minister, Boris Johnson, made the decision to suspend all assessments in schools.  That's what a lot of people are saying but it's not actually what was said.  What he actually said was that "exams in summer 2021 could not go ahead as planned" [1].  This is very different from children not being assessed at all.  It was also exclusively in relation to Secondary schools and colleges.  A quick ctrl+f reveals that the term Primary only shows up once in the entire address and that is in relation to the earlier closures and move to remote teaching.  In the Ofqual guidance, Primary schools do not feature at all.  

Now, you could very easily argue that this is because the Key Stage 1 and 2 tests (the SATs) have been officially cancelled, and you'd be spot on.  However, some up-to-date guidance would be nice, no?  After all, the most recent guidance for Primary schools is from last year and merely states that reporting progress is not required and nor should schools attempt comparative information about attainment [2].  So, for the second year in a row, Primary children are entering Secondary school with a fairly woolly idea of where they are educationally speaking.

But Carl, I hear you say, without national testing, schools are free to assess in whatever way they see fit.  Teacher assessment will generate grades and that's a good thing, right?


Here's the thing.  I'm going to start with an anecdote and then seamlessly segway into my issue with relying solely on teacher assessment.  I'll subtitle them so you can skip straight to the meat if you're short of time.

When I were a lad...

Hop into my way-back machine and travel with me to the previous millennium.  The year is 1997.  The Spice Girls dominated the charts; Titanic and Men In Black were in the cinema (remember the cinema?); Buffy the Vampire Slayer and South Park had only just come out on TV and your humble blogger was a happy-go-lucky Year 9 pupil (that's around 15-years-old for anyone not familiar with the English education system).

I was in the last term of mandatory History classes (I dropped them as soon as I could, probably due to a dull curriculum and a series of terrible teachers) and I was awaiting the results of the test that would inform our final grade.  I was thrilled to be told that I had scored the second-highest in the class (second place would be a regular feature in my life).  I was even more thrilled when the person who scored the highest mark openly admitted to cheating; going so far as to show the notes he had smuggled in and used.  So, technically, I had scored the best in the class.  Go me.  I felt pretty good about myself.

Until my teacher announced to the whole class that I didn't deserve that accolade.  In fact, he went on to say, given my lack of enthusiasm for the subject, I didn't deserve to even be in the top 10.  I felt pretty bad about this but at least I had that grade.  I could hold on to that.

Alas, my end-of-year report would reflect the teacher's opinion of my ability and not my actual ability.  Since it was not a terminal exam, the teacher was under no pressure to include any actual grades and so my final History report was concerned not with how much History I had learned but more with what kind of impression I had left on the teacher.

Teacher Bias is Real

Make no mistake about it.  I don't care how good a teacher you are, heck, I don't care how good a person you are, you are going to hold biases.  They may be unconscious, they may be positive, they may be justifiable but they are still biased and assessment should not be based on bias.  

Let's have a look at what I mean...

Current research has identified nine different types of bias [3], all of which can affect a teacher's decision when it comes to grading.  Before we go through them, I just want to make it clear that I am not trash-talking teachers.  Remember, I am a teacher and I am just as guilty of some of these as the next person.  Also, note that these are unconscious biases; no-one (I hope) is deliberately applying them.  My aim here is to highlight the benefit of externally-graded, normalised assessment over simply asking a highly invested and emotional human being for their opinion.

Gender Bias

Pretty straightforward, this comes down to 'girl subjects' and 'boy subjects'.  The idea that girls are better at English and Art, while boys' brains are better suited to maths and science.  I can hear you scoffing.  But it happens.  It is an inherent bias that does exist and to pretend it doesn't is naive.  A recent study in the Journal of Economics found that:

the gender gap in math performance is substantially affected by teachers’ implicit stereotypes. Girls, especially those with lower initial skills, are lagging behind when assigned to teachers with stronger math-male and literature-female implicit associations. [4]

 This was in 2019.  Some people still hold the view that female brains are not as good with numbers.  The same report also suggested that having a teacher of the same gender helps improve performance (so boys learn better from men and girls learn better from women).  Just over 75% of UK primary teachers are female [5], so maybe the girls' chances of negative bias when are evened out?  But, if that is the case (and I'm not convinced it is), then where does that leave the boys? A lack of bias isn't about making sure that one group is less affected than the other, it's about eliminating a difference altogether.

Okay, okay, so we know that gender bias may be a problem.  Why don't we just account for it and maybe be a little less harsh on the potential victims?  Then we fall into the 'I think they could have done better' trap and award a grade that is not based on actual attainment and is therefore invalid and useless to the next school, to the parents, and to the child.

Now, I'm not suggesting that national assessments are devoid of gender bias, but they are subject to rigorous testing and reviews such as Differential Item Function analyses that greatly reduce the chances of it.

Affinity Bias

This is where race, personality and shared histories come into play.  Again, I'm not saying that teachers are racist (although a report in 2020 stated that "both teachers and nonteachers hold pro-White explicit and implicit racial biases" [6]).  Based on my own personal experiences, teaching is rather a colour-blind profession... however, it is not just race that affects a person's opinion.

We've all taught children who we dislike.  Be honest. There are those kids who you cannot wait to see the back of.  The children who you are secretly (or maybe not-so-secretly) hoping will 'get their comeuppance in Secondary school; those kids who 'need to be taken down a peg or two.'  We're human.  We have foibles.  The thing is, when you're in charge of affecting a person's future, you shouldn't have the opportunity to even risk this sort of thinking.

It goes the other way as well.  I've taught children who have such a similar personality to me that we have been able to discuss YouTube videos and recommend books for each other.  There are children in our classes whom we genuinely like.  Are you honestly saying that, given the chance, you wouldn't bump their grade up a little?  Maybe give them the benefit of the doubt on a grade boundary?  We shouldn't.  But we are human.  So we shouldn't be given the opportunity.

Fundamental Attribution Bias

This is a weird one.  Basically, it's what my Year 9 History teacher did to me.  When someone achieves more highly than we think they should, we label them as lucky. It was a one-off.  They did well on the day but they wouldn't be able to repeat it.  That sort of thing.  At its worst, we take all the achievement of the child and claim any success for ourselves - they only did well because they had a good teacher.  How many of us have been guilty of that?  Be honest.

Even with externally-graded assessments, this bias can rear its ugly head ('so-and-so doesn't deserve that score') but when the grades are normalised against a national cohort, it's difficult to argue with them.  

As with all of these biases, there is a positive version as well.  In this case, it would be blaming the test questions, or the day, or the glare of the sun... any and all excuses (referred to as Construct Irrelevant Variance in the trade) can be used to make allowances for a bad grade and possibly justify an artificial uplift.  Did you catch the problem adjective there? Artificial.  It's not real.  It shouldn't happen.

Beauty Bias

Yes, you read that right. Yes, it is a thing.  No, it shouldn't be.  Sean Talamas, a post-doc researcher, said:

The impact of the attractiveness halo effect on perceptions of academic performance in the classroom is concerning as this has shown to influence students’ future performance. [7]

 This 'attractiveness halo' can be anything that makes a person more attractive to us - not, I hasten to add, necessarily in a romantic way!  Things like a nice smile, a 'can-do' attitude, a tendency to laugh at our jokes or make pleasant comments in class.  These all contribute to it and they can create a beauty bias, encouraging a more favourable result despite any evidence to the contrary.  The paper suggested that girls are more likely to be in receipt of beauty bias than boys, regardless of the gender of the teacher.

And again, the opposite is true.  A person deemed less attractive is in danger, according to the research, of negative bias.  I would go a step further and argue that, particularly for Primary teachers, it is not just physical appearance that can affect us but also psychological and environmental aspects.  How many of you reading this have ever marked more favourably purely because it is a wonder the child even made it to school that day?  Or maybe given an extra half mark because a child is having difficulties at home?  I know I have.  People are emotional creatures and we shouldn't be allowed to influence terminal grades.

We're at the tipping point now, so if you're still here, well done!

This post is taking a long time because I'm having to do a lot more research than I usually would (and I'm falling down a lot of research-rabbit holes in the process).  So I'm going to add a 'Part 1' to the title and continue with the final five biases and my closing argument next time.  Hopefully in a couple of weeks... maybe next month. It depends on how organised I can be!

Thanks for reading this far, I'm sure I've said some things that you disagree with - the title probably.  Please feel free to debate me, or convince me that I'm wrong.  My MA is in Educational Assessment so I am, I guess, inevitably biased towards a system of assessment.  I feel like I should also point out that I never used to be on Team SATs.  That's the result of the MA and the very wise teachings of Dr Mary Richards and her team!  So if you have any questions that I find too difficult to answer, that's where I'll be heading!

I guess I'll close off this week by reiterating that I am very passionate about education and teaching in general.  I think Primary education is the most important phase of anyone's life.  The current system isn't perfect and it is certainly not fair but I think that cancelling the SATs and removing the chance to truly take stock of the effect of the pandemic and remote teaching, whether successful or not, is a misstep.  For the first time in pretty much ever, we had a chance to force the End of Key Stage tests into becoming what they were always supposed to be: an honest evaluation of Primary education, free from league tables and floor targets and everything else that was only ever intended to be a tool for analysis.

Subscribe to the Blog to make sure that you don't miss part two and follow me on Twitter to continue the discussion.  I have a new website coming out soon (I've been saying that for a while but it really, truly is coming soon, like before the end of the month), so look out for that as well.  Until next time, look out for each other and try to keep smiling!

Carl Headley-Morris

Email me!            Tweet me!            Visit me!

References for this post:

[1] https://www.gov.uk/government/speeches/prime-ministers-address-to-the-nation-4-january-2021


[3]Atewologun, D., Cornish, T. and Tresh, F., 2018. Unconscious bias training: An assessment of the evidence for effectiveness. Equality and Human Rights Commission Research Report Series.

[4] The Quarterly Journal of Economics, Volume 134, Issue 3, August 2019, Pages 1163–1224, https://doi.org/10.1093/qje/qjz008

[5] https://www.ethnicity-facts-figures.service.gov.uk/workforce-and-business/workforce-diversity/school-teacher-workforce/latest#:~:text=By%20ethnicity%20and%20gender,-Location%3A%20England%20Time&text=and%20gender%20Summary-,The%20data%20shows%20that%3A,and%2085.7%25%20of%20female%20teachers

[6]   Starck, J. G. et al. (2020) ‘Teachers Are People Too: Examining the Racial Bias of Teachers Compared to Other American Adults’, Educational Researcher, 49(4), pp. 273–284. doi: 10.3102/0013189X20912758.

[7] Blinded by Beauty: Attractiveness Bias and Accurate Perceptions of Academic Performance
Sean N. Talamas ,Kenneth I. Mavor,David I. Perrett
Published: February 17, 2016https://doi.org/10.1371/journal.pone.0148284