A Brave New World………of Data!

Data has the power to change the world……and it is! I am a huge fan of DATA and in particular BIG DATA. For those of you unfamiliar with the phrase BIG data, it simply means the collection and interpretation of data streams traditionally too large to process in traditional ways. Our world is full of data, about us, about our habits, our likes and dislikes and our health but those utilising this data are mainly private businesses, who do so to maximise their business activities. Nicky Morgan, the Education Secretary, and the DfE recently released a White Paper with many dramatic changes to the education system but one thing it made clear was it wanted the education system to make better use of data and to improve teaching through research based development, but I will come back to this in a moment.

Firstly let me tell you what big data can do. I was first introduced to the term when I read about Carolyn McGregor a doctor of computer science, who having worked in data based decision support for big business, turned her attentions to research and healthcare. In 1999 she was asked to go to see a neonatal intensive care unit in Toronto to see if her expertise could improve care for the infants. She was pregnant at the time and her own child was born premature and subsequently died. Seeing all the monitors around the unit bleeping and flashing, recording endless streams of data she wondered what happened to it. The doctors explained that nurses recorded the data once an hour on paper but the data streams themselves simply spooled out of memory, so none of this data was recorded or analysed. Over the next decade she was instrumental in the foundation of the Artemis Project, named after a greek goddess who protected babies and pregnant women. The Artemis Project began recording these data streams for all babies in the unit and soon started spotting useful trends. About 20% of babies in the unit caught infections in the and around 20% of those subsequently died due to the infection. The data could see that the babies, normally erratic baseline heart rate, would become more steady and less erratic between 12 and 24 hours before infection presented itself in the child. Doctors could now see the infection happening in the infants heart rate, long before any outward signs and could administer medication earlier. This data  has helped save many babies lives and the project is now working with many more hospitals across Canada helping to reduce child mortality. That is the POWER of BIG DATA…….it has the power to change the world!

Screen Shot 2016-03-26 at 08.04.40

So the Education Secretary telling teachers, in her white paper, to employ more data and research in our continued professional development (CPD) must be a good thing, and believe me I applaud this suggestion however in this brave new world there are pitfalls and dangers along the way. Collecting data is a skill, it needs to be done in a robust way which reduces contamination and increases depth and breadth of possible results. Also having data, no matter how robust, is only part of the challenge. Drawing conclusions from the data collected can be a hard task and one even “experts” get wrong. In the same week as the white paper was released the DfE also released a document linking poor attendance with poor academic achievement. In their press release Nick Gibb, Schools Minister, even went as far as to say that “this new research is further evidence that missing school for even a day can mean a child is less likely to achieve good grades, which can have a damaging effect on their life chances”. Now this assertion is WRONG and a DANGEROUS one. It also explains my point of why more research based decisions may not be the promised land after all.

So why do I say he is wrong and why am I concerned about this push for more research. Well this report is fairly robust in it’s collection of data, there are a few contaminating factors not addressed but let accept the data as correct. This report clearly shows a correlation between time spent in school and achievement at the end of KS2 and KS4 and that is as far as any good researcher would go. However the report suggests strongly in its wording that their is a consequential link between these two, a causal link between attendance and attainment. It also implies, and the MP’s assert strongly, that the lower attainment is a symptom of lower attendance. Every teacher believes that to learn a student needs to be in school, the more we see our students the more we can teach them which is why this report is so clever and dangerous. It takes a belief commonly held and then looks for data to prove it, some may even say it’s release and timing may also have more political reasons than educational. If we had irrefutable proof that more hours in school equals higher achievement we could justify say, longer school days, or fining parents for absence and because most teachers would easily believe higher attendance equals higher attainment they would not question it, especially when given figures such as this report headlines. However this, like so much research I have seen in recent years, is flawed. For one lets actually look at the data, for example at KS2 a student who has no sessions (half days) absent has a 51.5% chance of getting level 5 or above at the end of KS2, whereas if they have missed between 1 and 4 sessions they only have a 38.3% chance. Now we teachers are an amazing breed, we can inspire and impart knowledge at a phenomenal rate BUT do you really believe that missing one afternoon when you are 10 is going to reduce your chances of getting a particular grade by 13%. As a lover of research I would say there is a probable link between attendance and attainment but I would see the lack of attendance as a symptom of lower engagement in their education and thus lower attainment. This research should pose the question how do we engage young people and parents in their learning. Forcing students to attend school and fining parents will not increase engagement and making the school day longer will not increase results for all. Research can be dangerous and is leading many, like Alice chasing her rabbit, down holes that won’t lead to improvements. Ministers and dare I say it many teachers, make flawed decisions based upon their flawed understanding of what the data shows. Some may believe that ministers knowingly use flawed interpretations of data knowing few will question decisions made off the back of “robust” research.

Screen Shot 2016-03-26 at 08.05.02My school is ahead of the data initiative, we have been completing research projects as teachers for some time now and many gains have been made with student engagement and attainment but none the less it can still mislead people. Lets take a similar thread to the DfE’s “attendance equals attainment” one above and look at revision sessions. We have always allowed our students study leave and put on revision sessions for them within this period. These have been optional and statistically those that attend these sessions get better overall results than those who do not. There is correlation between attendance at these sessions and attainment in exams, this is not in doubt. However I have heard a number of people assert that attendance at these sessions increases the students results based upon that data. This however is simply not proven in fact. Take my class from last year as a microcosm of this theory. Of 12 students 3 attended the optional revision session I put on and those 3 students all got A* in the exam, while no-one else gained a grade above a C. If we concentrated on headline figures of C grades of those who did not attend only 3 got a C so I could claim that my 1 hour revision session increased their chances of gaining a C grade by 67%, data backs this up. Many in the light of statistics may accept this. Data also backs up my assertion that the ONLY way to get a A grade is to attend my revision session. Less will believe this even though it is the same statement. Both of these assertions are not what the data really shows. The data really shows that those students who are already engaged with their learning and already working tirelessly at home doing revision and practice papers, are also those students who will likely attend optional sessions. Are those A*’s a symptom of my session? Of course not, they are a symptom of those students hard work for 2 years of GCSE study. Their attendance at my sessions is a symptom of their engagement in their education and future, which by the way is what also causes them to get A*’s. Those students who got less than favourable grades would, in my mind without doubt, still have got unfavourable grades even if they had attended my session. I like to think my sessions are worth while and help the students but to assert they were the difference between C’s and A*’s is ludicrous. This year our school has made these sessions compulsory which I for one think is a great move, a chance to refocus all of my students and give them last minute advice a day before the exam is a chance I am glad to have and  keen to fully utilise. It is a good decision, but not one backed by data. No data can show this approach will increase results because academic success is more complex than any one single factor. It is equally possible that forcing all students to attend will reduce these sessions effectiveness for those who are engaged, as more time may now be spent managing the behaviour of those who are not. I often hear teachers say “studies show attendance at extra curriculum clubs” increases students chances of exam success. Again this is not true, the best that can be said is there is a correlation between the two. It is highly likely that attendance at clubs is a symptom of an engaged student, and thus of course those that attend clubs get better grades, as they are the most dedicated students. To suggest attendance at extra curriculum clubs has a causal effect on a students results is simply not backed up by rigorous data. Many will see the flaw in these two arguments but these are the basis of the DfE’s argument about attendance. If i told you in 1 hour I moved students from a C to a A* in their future exams you would feel uneasy, you may question the data. The government say missing one afternoon at school when your 10 reduces your chances of a grade 5 by 13% and no-one delves deeper. As much as data can lead us to enlightenment it can also be used to blind.

So a brave new unstoppable world is coming and data is driving it forward. As I said at the beginning big data has the power to change the world. Imagine the possibilities! I recorded my heart rate for a month and published a typical week in the life of a teachers heart. I learnt loads about myself and the effect teaching has on me, imagine if every teacher was monitored, or every year eleven. Imagine if we recorded every students heart rate, sleep patterns, activity, state of mind and ultimately results. Imagine the patterns we may spot, imagine the ways in which we might help students to learn. Grab this new found opportunity with both hands, do research, collect data, question everything and learn so much. Data offers endless horizons to explore but watch out for dangers along the way. For me, I am excited by the prospects it holds and am keen to get involved, but for all you other keen researchers out there remember this.

Never try to prove a theory, instead collect data, as much as you can, and then see if any patterns emerge. From these patterns shape a theory to fit the data and then most importantly work your very hardest to disprove your own theory. If you can’t you may be on to something great!

 

Advertisements