# The most important data handling skill?

Discussion in 'Mathematics' started by MathsHOD, Mar 3, 2012.

1. ### MathsHOD

My class sat the EdExcel unit 1 exam (higher tier) on Friday. This is the one that is basically data handling with a few things thrown in to make it up to 30% of the full GCSE.

Throughout the course I've made a big play that all of the data skills on their own are basically useless unless you can interpret the outcomes in context. We've done loads of comparing 2 lots of data - whether that be stem and leaf plots; box and whisker; cumulative frequency curves; pie charts; histograms (or combinations of any 2 of these) right back to having 2 sets of raw data and making decisions about which graphs/calculations are most appropriate to allow us to compare.

Of course I've done this not because it is likely to be in the exam but because it's what data handling is all about and they'll need to know it in a range of other subjects and beyond school in many cases.

I am however surprised that this skill wasn't tested in the exam at all. I know that exams need to vary and not be predictable but there is almost boundless ways that they can test this skill that it couldn't be described as predictable.

It's not a complaint about the paper - which, although I've not worked through it fully yet, looks fine - just a more general comment.

The question that stands out as being out of place is the one where you are given a completed histogram other than the fact that the y axis (relative frequency) scale is missing. You are told that there were x people in the 10-20 category and told to find an expression in terms of x for the total number of people. Just struck me as unnecessary and the examiner showing off rather than trying to get information about what students did and didn't know.

2. ### AnonymousNew commenter

I am a tutor so have not seen this paper. I agree that data interpretation is crucial - I often ask children why a mean is not very helpful by itself. You only know if you are above or below average. This builds up to the quartiles, CF diagrams and histograms. I used to be a scientist so have spent loads of time collecting data - we used to do 100 experiments just to get real accuracy with a data point. I think data is fascinating - and useful for other areas of education. I had a pupil who wanted to get into the army so was using CF curves to emphasise they would only take the top x% or those who got under a certain time.
Comparing data, using it in the real world - fascinating stuff.
On a separate point - I have had two tutees who have been given data homework as part of their geography / science lessons. The geography one was to do with Spearmans Rank Coefficent testing. This was for a pupil all set to get a Grade D. He had no idea what it meant nor how to interpret the results.
The second pupil (in year 8) was asked to draw a line graph (according to the homework) comparing the diameter of a planet with the weight of a person on that planet. They were then told to draw a line of best fit - was supposed to be a scattergram but again this pupil had not covered scattergrams.
I don't work in a secondary school but wondered if departments talked to each other about the mathematical aspects of their subjects?

3. ### PiranhaStar commenter

In my experience, interpretation in GCSE seems to be little more than "positive correlation" or a comparison of some kind of average and some measure of spread. Mostly the exam is doing calculations/drawing graphs with the odd reverse calculation of the kind you mention and a couple of probability calculations. I don't think A-level is much better. To many students, Statistics is boring but easier than a lot of the Pure.