Does the institute of physics understand basic stats?

While it would seem a strange way to present the information keep in mind that this could be simply poor conveyance by the journalist
 
Just to caveat this, I do understand the difference between median and average, but the article days almost half are under the average which implies a normal distribution (and you wouldn't expect anything much other than a bell curve in this case anyway)
 
Some bad stats from either the IoP or the BBC, but I think the message is still sound.

I see nothing implying a normal distribution. It probably is, but there's nothing in the article to support that.

What they are saying is that coed state schools are below average, so single sex or independent schools are the ones doing best. The coed state schools, therefore, need to learn from what the others are doing. If they did then, of course, the average would go up.
 
I guess it's just making the point at how badly the state funded schools do. If the averages take into account private, and single gender schools and then measure the state-funded mixed schools against that, then that's probably where the 49% comes from.

Makes a bit more sense than just saying that nearly half are below average.
 
Last edited:
Probably just the usual media failing to understand statistics.

Co-ed state schools are doing a worse job than single sex and private. At everything.
 
ahar said:
Just to caveat this, I do understand the difference between median and average, but the article days almost half are under the average which implies a normal distribution (and you wouldn't expect anything much other than a bell curve in this case anyway)
Click to expand...

I'm sorry but that post suggests you don't understand it either. First the median is an average. I think you meant to say mean. Also you can't draw the conclusion that the distribution is approximately normal. It may be the case that some few schools are trying and achieving a balance but then a lot aren't, meaning a skewed distribution
 
frying_pan_cat said:
First the median is an average.
Click to expand...
No, the median is the midpoint value, the term average usually refers to the mean (what everyone would understand to be the average - adding all values, dividing by the number of measurements). It's confusing to use the terms median and average interchangeably. (Not that I'm suggesting this doesn't happen!)
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom