Sunday, 26 June 2016

Data Analyst Anyone?

In this age of Multi-Academy Trusts, five zillion data collection points an academic year, and measuring the "progress" of pupils with summative assessment every time they enter the school building, we are generating loads of data. And that's not counting the data we generate on a day to day basis as teachers by completing a register, filing behaviour incidents, etc etc.

Whats the point of all this data? Well, in most schools I've been in, I would suggest there's not much point. This is not down to the desire to see normal teachers die under a workload of data recording by SLT, rather I would put it down to ignorance. Schools feel they have to record everything to cover all the bases, most importantly that when the dreaded OFSTED show up they can produce some nice case studies of pupils with their full back story, progress history, what they had for lunch etc. There is also the knowledge that some kids, whilst they may not have made "enough" progress academically, have made progress in terms of their behaviour, ability to stay in lesson etc, schools know this is in their data somewhere, but because it tends to be spread out over different systems it is difficult to show this using the data recorded. The majority of data goes to waste because people, genuinely, don't know what to do with it to turn it into some sort of meaningful information which can better the learning experience for the kids.



So are we stuck in this mess? No! Lots of tools are out there for teachers to use to have a look at their data. However, all of these require time to learn to use effectively, then we can begin the process of analysing data to produce information. So at very least investigative data analysis should be a TLR if not a full time job, and it will take time to produce results. It also helps if the person doing the analysis is up to speed with data and statistical analysis methods (I'm a science teacher so I'm going to take the biased view point that a science graduate would be best, but I realise that's a pretty narrow view).

So what would a school data analysts job look like? I would suggest that their time would be split between producing reports for progress etc on a regular basis, highlighting for staff anomalies in the data and disparities (to lead to higher data reliability), but they also need at least 25% of their time devoted to "blue sky" research. By this I mean they are free to investigate a hypothesis and questions, for example can we detect kids at risk of exclusion in KS4 early from their KS2/3 data and thereby setup an "early warning" system for future pupils? Obviously the data analyst would need some fairly serious computing resourcing if they're analysing data over the past 10 years, but in the age of cloud computing on demand that should be doable.

To conclude. I reckon it would be a good job! I'd quite enjoy it, I also reckon it would allow schools to turn piles of languishing data into information that has an impact on pupils education.

Useful links


As my old university tutor (Dr Paul Kenton) used to say "the plural of data is not information" (or similar)





No comments:

Post a Comment