Accusations of unfair algorithm


characteristics of algorithm

COVID 19 has thrown everything up in the air. The government choose to cancel school exams. 

To ensure education system didn't grind to a halt. Government set about developing an algorithm to calculate what students would have attained had students taken their exams. 

Two fundamentals were used to estimate grades.

  1. Student ability and how well their school or college performed.
  2. In previous year exams.

Results were produced by combining the ranking of pupils. Plus the grades expected by the school previous years attainment.

There were other minor adjustments, but those were the shaping factors.

Government objective was to prevent exam inflation. By ensuring estimated results mirrored recent years.

The problem with that. Talented outlier student studying at a low-achieving school or an improving school. Are more than likely to receive an unfair result. 

Ofqual warned government computer says no bias.

In March education secretary, Gavin Williamson. Requested head of Ofqual the regulator “should ensure, as far as is possible. That qualification standards are maintained and the distribution of grades. Follows a similar profile to that in previous years”.

In a letter to Sally Collier, Ofqual chief regulator. He says students should be issued with calculated results. Based on schools’ assessments of their ability. In the relevant subjects, supplemented by other evidence.

As instructed Ofqual began to test a range of algorithms. Simulating predictions of results for 2019 exams. And comparing the outcomes to the actual results.

In April Ofqual consulted schools, unions, teachers, and employers. 

May consultation responses were published. Some raised concerns grade calculation may lead to negative outcomes.

In June teachers submitted their student exam assessments. It became evident the algorithm was flawed. In June Sir John Coles, a former director general at the Department for Education (DfE). Warned 75% accuracy at best.

In July, Parliaments education select committee raises concerns of students being systematic disadvantaged. The exam body choose for whatever reason to ignore the warning. Nick Gibb, schools minister. Holds a meeting with Ofqual and senior officials at the DfE, to examine the concerns raised. Ofqual resists calls to publish their model. But admits to having tested 12 different statistical models. 

August exam grades are announced.

A downgrade of 39% effecting poorer students the most. 

Independent schools appear to have been less affected. 

The inevitable outcome a government U-turn a week later. A good thing provided government takes the opportunity to learn from their mistakes.


The following are few guidelines when you are writing an algorithm

Correctness. (Check each line)

Runtime analysis. (Analyse the running time. Particularly important when using large data sets)

Implementation. (One, learn how an algorithm works and two how to use an algorithm)

You may not want to know how an algorithm works, but you really need to know how the algorithm solves the problem. This is particularly important when you work with nonstandard and hard problems. 

Essential properties of an algorithm:

  • Finiteness
  • Definiteness
  • Effectiveness
  • Input
  • Output

In other words, while writing an algorithm, you should keep in mind first two points. That is, your algorithm finishes in finite number of steps. And every step deterministically leads to following one.

About the author 

Christopher Bird

Building your own Power App, BI solution, or automated workflow can be a mind-blowing experience. It can also be a nightmare. Particularly when you begin with a blank screen. My advice, get professional help as and when you need it. That's what successful people do.

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}
Subscribe to get the latest updates