Research School Network: What we need to know about cognitive bias As we think about assigning grades to year 11, we need to be aware of the biases that might be in play and what to do about it..


What we need to know about cognitive bias

As we think about assigning grades to year 11, we need to be aware of the biases that might be in play and what to do about it..

by Durrington Research School
on the

The world is a complex place. Our minds are constantly bombarded with more information than they could ever handle. As a survival mechanism we developed a series of shortcuts and systems to handle information. Most of these shortcuts are helpful, sometimes these shortcuts lead to errors in thinking which effect decisions. We call these errors cognitive biases, and we all have them. Yes, even you I’m afraid.

With the announcement that the exam season will be replaced with Teacher Assessed Grades (TAGs) I thought it might be prudent to look at some of the key cognitive biases and fallacies that we might all fall foul of and how they might contribute to the grading decisions.

1. The anchoring effect
.

This effect is caused by the first piece of data you experience on a subject, in this case a student. The data initially creates an anchor in you mind and all your decisions of their worth (in this case their grade) are adjustments from that point. The biggest risk of this causing an issue in TAGs is due to target grades. Say we have two students, A + B. Both students have a mock grade of a 5 sat in November. Student A has a target grade of a 7 but Student B has a target grade of a 5. The anchoring effect describes the fact that subconsciously you are more likely to increase the TAG of Student A than B, even if it turns out Student B scored a few more marks. This is because the target grade of Student A is anchoring your expectations that they should be higher. The rebuttal to this is that if they have a higher target grade then they must be better naturally. This is not as simple as it seems. Target grades are not always as reliable as schools tend to believe. You can learn more about some of the issues with target grades here.

2. Confirmation bias. 

This well-known bias describes how we subconsciously select and prioritise information that confirms our personal opinion. Instead of being the enlightened objective evaluator we like to think we are, we are in fact fickle. The most likely effect of this on TAGs is that if Students A is better’ than B in your opinion you are more likely to find the best features in their work. This has huge implications for essay marking, where there is a high degree of subjectivity.

3. Authority bias.

This is simply the idea that people in authority have more sway over the group’s opinion than the others. In the case of TAGs this might come from the department team being less likely to argue the case against the head of departments grading decisions. Leading to their personal subconscious biases affecting the TAGs to a greater degree.

4. Recency bias. 

This is the idea that you prioritise recent events over historic events. For TAGs this could be a stronger weighting being placed on their final assessments than earlier ones. This might be a valid decision, depending on the context. There is a concern that conditions in the summer might not be as robust as earlier in the year, so we should try to justify our weightings and not make assumptions that newer is always better.

There are literally hundreds of biases. I have just picked a few. After reading this you might feel like the whole thing is pointless and it is impossible to come to a fair conclusion. When overcoming cognitive bias, we have one superpower at our disposal, each other. When assigning TAGs aggregation is your friend. Having multiple people share their opinions increases the chance that a single person’s bias will sway the decision. Being aware of our biases also helps to stop them. Just pausing and running through the evidence for each decision can dispel those knee jerk instincts. When it comes to assessment, we can also take steps to reduce biases. Anonymous marking and rigorous moderation will help.

At the end of the day, we must come to a decision for each student. This is not going to be easy . We will just have to do our best, make it as fair as possible and hope it is the last time. I’m sure the one thing to come out of this process for many teachers will be a renewed appreciation for the challenge’s exam boards face yearly. Good luck.

Adam Robbins

More from the Durrington Research School

Show all news

This website collects a number of cookies from its users for improving your overall experience of the site.Read more