The problem with appeal statistics

So a tweet today from the Refugee Council said:


My first thoughts were that doesn’t add up. A quarter of all decisions wrong first time?
This would mean that 90% of all decisions would have had to have gone to appeal. Surely we are a) granting more than 10% and b) appeals aren’t 100% of all refusals.

So I went to the Office of National Statistics and found the numbers released today.

Looking at Asylum data tables immigration statistics January to March 2015 volume 1 – as_01 we have the following numbers for 2014:

Total initial decisions 19,936
Total grants 8,096
Total refusals 11,840

Straight away we can see that 40% of cases are granted.

Then looking at Asylum data tables immigration statistics January to March 2015 volume 4 – as_14 we get the appeals information

Total appeals determined 6,130
Appeals allowed 1,744
Appeals dismissed 4,047

So the 28% figure has been found from 1,744 in 6,130.

However there were 19,936 decisions made in total in 2014, which results in 8.7%* of all applications being wrong first time.

This isn’t the first time I’ve seen groups take the percentage of successful appeals and then used them as a figure of number of times someone got it wrong, the Disability Living Allowance Assessments springs to mind with Tweets and Facebook posts.

* An issue with this number is that it’s comparing the number of initial application in 2014 with the number of appeals in 2014, it’s highly likely that some of the appeals made in 2014 were based on decisions years earlier, however that data doesn’t appear to be available.

Leave a Reply

Your email address will not be published.

Human test: Enter Ryan backwards