The devil is in the data

@Inglish_teecha popped up in my Twitter feed with the following tweet:

@GuardianTeach Trouble is, with all the time spent managing the data, there's no time to prepare the teaching and learning.

Instantly my blood was up. I'm a very big fan of data in schools, and I have come across this sort of attitude before. Then I remembered one of the golden rules of schools: context is everything.

My philosophy regarding tracking student progress and target setting is simple. I believe that at the end of a lesson all kids should know something that they did not know at the start of the lesson. For more practical subjects, you can talk about being able to "do" something new instead if you like. That's my idea of learning, and if we hold on to that idea and make it happen through teaching, then we're getting it about right.

In order to know whether all the kids now know something they did not before we have to do some form of assessment. Black and Williams ideas on "assessment for learning" are now commonplace in teacher training (though perhaps not being implemented in classrooms) and if we regularly check student understanding as we go along, doing summative assessments at the end of term should be a doddle since we have plenty of evidence to base those assessments upon.

The next step is to use these assessment data against meaningful targets so as to identify which students are falling behind and need support, or equally to identify able students who need greater challenge. Classroom teachers should be doing this as well as middle and senior leaders.

As I say, simple, and frankly essential in any decent school.

Which is where the context comes in. Context is everything. A decent school with good leadership knows how to do data right. Do it wrong, and it can easily lead to a situation where data gathering gets in the way of the core business, which in schools is of course teaching and learning. It's fair comment for @Inglish_teecha to claim that in their context, in their school, they feel it is going wrong.

If leaders then turn this data on the teachers to try and beat them up over their performance, then we have a problem. Plus if the targets are not meaningful, because they are set too low, or are too "aspirational", or are simply made up, then the whole thing falls apart, and the devil is in the data.

So I apologise, @Inglisha_teecha, whoever you are. I was angered when I first saw your tweet, but it helped remind me that actually successfully tracking student progress is a tricky business, and I bet there are more schools getting it badly wrong than there are getting it right.

2012 preview: part I

Now my broken finger is mostly fixed, I'm starting again with a look ahead to what is thundering towards us down the educational tracks in 2012. For my money, there are three main bodies to watch with some big issues next year. I'm going to work through them in a series of posts.

In at number three, it's the teaching unions.

The two main battlegrounds are fairly obvious: pensions and academies. Reading about the pensions "deal" announced to parliament by Danny Alexander, you could be forgiven for thinking that the negotiations were finished and that everyone is happy. For most of the teaching unions, this could not be further from the truth. ATL are the only major teaching union who appear to have signed the "heads of agreement" proposed by the government, whilst the big two, the NUT and NASUWT, have most definitely not. Interestingly, the union for headteachers, the NAHT, has given little away on whether they have signed the agreement or not, but are consulting members on the government's current offer in the new year (click the links to see each union's statement on their current position).

This seems to have been given very little coverage in the media, possibly because it's not much of a story until the union executives announce what they are going to do next. It would seem that the government has still not provided any evidence to show that the current teacher pension scheme is unsustainable since it was revised in 2006 to make it affordable for the nation. I predict that this will be a key sticking point, as it would seem that the government don't really have an argument against this, and I don't think either party is going to budge. I think that teachers are unlikely to support further strikes leading up to the GCSE/SATS season, but action short of a strike could well be on the cards in early 2012.

More trouble is likely to come along in September 2012 when more schools convert over to academy status, plus a new crop of free schools will open up. The teaching profession as a whole is still very suspicious of academies, and even more so of free schools. Many union members will find themselves teaching at a newly converted academy in September and could well make quite a lot of noise about it. Free schools so far seem to be almost union-free zones (odd that), but I'm guessing they will work their way up the unions' hit lists, especially once the pensions dispute is finally settled.


c

(Raspberry) Pi in the sky

"I wish I were Dave Grohl" I declared to my poker buddies, mostly just to fill the time having once more crashed out early and handed over yet another of my hard-earned tenners to our head of maths. None of them responded, so I continued to sit and watch in quiet awe of the Foo Fighters frontman executing textbook rock-star poses in front of thousands of delighted fans at some festival on MTV.

Why is Grohl one of my idols? He's a rock legend, was part of arguably one of the most important bands of the 90s, and then went on to various projects creating some really great music along the way. According to Google search he is also the most famous "Dave" in the world and the "nicest guy in rock".

Back in the real world, another legend has a new project. David Braben co-wrote the classic game Elite way back in the early 80s and has now designed the Raspberry Pi, a cheap and cheerful computer-on-a-USB-stick that will cost about 15 quid.

The Raspberry Pi is an attempt to lower the barrier of entry for kids to become programmers. Eben Upton, who founded the project, says in the Observer:

"While a lot of homes have a computer these days, kids aren't encouraged to start messing around with programming languages on these family machines. No one wants their home PC going into meltdown. A useful analogy might be that you wouldn't let your children take the family car apart, but you might be happy to let them loose on a bike – so the Raspberry Pi is the computer equivalent of that cycle.

You can program it using scripting languages like Python, or compiled languages like C and C++. You could write a game of a similar level to Angry Birds or Quake 3...

We need to convince the government that programming should be more prominent in the curriculum."

This will sit nicely with Google chairman Dr Eric Schmidt, who was in town this week singing a similar tune:

It's not widely known, but the world's first office computer was built in 1951 by Lyons' chain of tea shops. Yet today, none of the world's leading exponents in these fields are from the UK...

I was flabbergasted to learn that today computer science isn't even taught as standard in UK schools... Your IT curriculum focuses on teaching how to use software, but gives no insight into how it's made.

It seems to me that these commentators are making the following three points:

1. The UK needs to produce more computer programmers
2. Cheap hardware will encourage kids to "tinker" and learn to program
3. Computer science should be part of the National Curriculum

I'm not sure any of these make sense, and I'll treat them one at a time.


1. The UK needs to produce more computer programmers

We are actually producing a lot of programmers. In fact, according to UCAS, computer science was the fifth largest subject in the UK university system in 2010:

Rank	Subject	Acceptances	% change	Applicants	% change
1	Law	16,931	-3.4%	94,231	3.1%
3	Psychology	16,063	4.9%	93,845	16.2%
5	Computer science	11,605	1.3%	55,259	11.2%
8	English studies	10,092	-2.8%	60,882	5.7%
11	History	8,616	1.4%	50,215	6.8%
18	Mathematics	7,276	5.2%	40,972	8.9%
42	Physics	3,657	2.9%	20,372	11.9%

Sources: Acceptances - "Final end of year figures 2010" Applicants - "Applicants and applications received by 30th June - complete tables" both from UCAS

So we have plenty of computer science undergraduates at university and lots of competition for places with nearly 5 applicants to every place available. Do we really need more than 11k per year?


2. Cheap hardware will encourage kids to "tinker" and learn to program

The Raspberry Pi is clearly an amazing development, and I love it. I love what the project is doing and what they are trying to achieve - but is this a silver bullet that will inspire a generation of coders in the same way that the BBC Micro did for David Braben and so many others in the 80s?

As a kid having your very own machine to play with, break, and fix for hours on end is clearly better than sharing the family PC. However, in three years of doing computer science at university I didn't meet anyone who had caused their computer to "meltdown" through their coding efforts.

What's more important to my mind is not the hardware, but the software. The Raspberry Pi's processor can't run Windows, instead the FAQ page suggests that only Ubuntu or Debian Linux will be supported at first. Using a free OS means the cost barrier is kept low for our wannabe geeks, but whilst Linux distributions such as Ubuntu are much more stable these days, they are still more complicated to install and maintain than, say, Windows or Mac OS. This may put kids and/or schools off before they have even started.

It's not a deal-breaker, especially if the project develops a distro tailored for the Pi, or even starts shipping units with this pre-installed. But one did not have to set up a BBC Micro in this way - it had an OS and a nice accessible programming language (BBC BASIC) ready to go straight out the box, which helps explain why this computer led to so many people "tinkering" with coding. Users purchased Beebs to do other things - be it business, education, or games - then started playing around with creating their own inventions. Why would anyone other than an established geek buy a Raspberry Pi? I certainly don't anticipate many school network managers queuing around the block to purchase machines that don't fit the standard model of a PC, no matter how cheap they come (they'll just get nicked for a start).


3. Computer science should be part of the National Curriculum

This bit really baffles me. It just does not make sense to force every child through programming on the national curriculum. Imagine the reaction if my man Dave Grohl proclaimed: "It's an outrage that the country who gave us Jimmy Page, Eric Clapton and Brain May does not teach every kid how to play the electric guitar as standard"?

Don't get me wrong, I think there are huge problems with the ICT curriculum, and agree that merely teaching kids to use Office is not what the subject should be.

However, coding on the curriculum is never going to happen for the simple reason that we don't have the teachers to deliver it. I have not had the opportunity to meet vast numbers of fellow ICT teachers, but the majority that I do know come from backgrounds other than computer science and have little or no programming experience.

Even if we did magic up extra teachers, the kids would hate it. Programming is often a slow, tedious process with endless hours of debugging and battling with incomprehensible error messages before getting to the point where it "works". Whilst environments such as Scratch are already used in some schools to ease kids into developing simple code, for me the gap between this and "real" programming is too big and it would be a disservice to students to have them believe otherwise.

Adding programming to the curriculum would not even necessarily lead to more computer scientists. Look at the numbers studying maths and sciences at university - both compulsory to age 16, but have far fewer undergrad students than computer science.

Finally, it seems terribly old-fashioned to think that the way to get kids into something is to ask teachers to light the spark. Most programmers, like most guitar players, started in their bedrooms with no formal tuition and took it from there. I don't see any reason to change this, especially with the dawn of the Internet and open-source software kids can "learn on the job" and contribute to real projects without ever having to get dressed.

We could certainly encourage more students in terms of after school clubs and extra support, but this is crucially different to crowbarring coding into an already crammed curriculum, which would not even serve the purpose Google and others want. Sorry Eric.


c

Image from Galaxy fm on Flikr

Lies, damned lies, and physics: part II

... but wait, there's more!

In "Has physics become cool again?" a Beeb science columnist tries to claim that physics is the most in demand of the sciences. Quote (next to another photo of Prof Cox):

"The total number of students entered for physics A-level has increased by 6.1%, from 30,976 in 2010 to 32,860 in 2011. Applications for physics courses at university are also up by more than 17% on last year and astronomy is up by a whopping 40%.

...

Higher tuition fees are likely to be a major factor in students opting for courses that are more likely to get them a job.

But that doesn't fully explain why physics and astronomy in particular are so much more in demand than other science based subjects."

Physics? Really?

Let's look at the numbers again shall we, starting with A-level students as I've got that data to hand from researching my last post:

Subject	Entries 2009	Entries 2010	Entries 2011	Increase 2010-11
Biology	55485	57854	62041	7.24%
Chemistry	42491	44051	48082	9.15%
Physics	29436	30976	32860	6.08%

Source: JCQ web site

So A-level students much prefer biology to physics by almost 2:1 in 2011.

Nevermind. To be fair, the quote is more about university courses than A-levels, so let's turn our attention to what UCAS can tell us about what courses new science students started last autumn:

Subject	Accepted applicants	Percentage change
Astronomy	155	n/a
Biology	4997	7.4%
Chemistry	4316	10.8%
Physics	3657	2.9%

Source: "Final end of year figures 2010" UCAS

This is similar to the A-level data: once again biology comes out on top, and the biggest growth is in chemistry undergraduates.

The only potential support for the line "physics and astronomy in particular are so much more in demand than other science based subjects" comes from what UCAS shows sixth formers have been applying to study in the 2010/11 application cycle:

Subject	Applicants	Percentage change
Astronomy	1223	40.3%
Biology	30066	8.3%
Chemistry	24939	9.5%
Physics	23924	17.4%

Source: "Applicants and applications received by 30 June 2011" UCAS

The percentage changes match the numbers quoted in the Beeb article and show that in terms of increase in demand, physics and astronomy have seen the biggest surges. However, once again the most in demand science is in fact biology with nearly 7,000 more applicants than physics.

What interests me is why are Pearson/Edexcel and the Beeb so keen to talk up a resurgence of interest in physics when the A-level numbers show that in real terms less students are studying physics now than back in 2000, and biology and chemistry remain more popular at undergraduate level too?

The big ol' cynic in me thinks Pearson/Edexcel have a clear interest in getting more students to take physics exams and buy pricey textbooks by persuading them that physics is the cool subject de jour. The Beeb may want to reinforce their public service credentials by demonstrating that their programming with ex-pop-star-turned-professor Cox is turning kids on to reading physics.

More frustrating is that all this fuss over TV and physics distracts from the key issue that science subjects have not been popular over the last ten years and very little has been done to fix this. A TV series showing some pretty pictures of the sky doesn't really cut it.

New Labour and now the Coalition have done precious little to encourage students into STEM subjects and then into industry. The best the latest government has delivered is to spare STEM subjects from the HE funding bomb that it's detonated under arts and humanities subjects.

In a century which is likely to be defined by a desperate rush to move from oil-based energy products to something else, physicists are going to be of vital important. I just wish that the Powers That Be worried less about talking up physics and started taking action to support it and other science subjects instead.


c

Lies, damned lies, and physics

Do you know this man?

According to Ziggy Liaquat from Pearson/Edexcel (experts in making huge bundles of cash from exam fees and related text books), the "Brian Cox effect" has led to a "boom" in students studying maths and physics at A-level, as reported by the Telegraph, Guardian, Metro...

All very exciting. The Guardian (link above) reports:

"Entries for A-level maths are up by just over 40% over five years, while the number of entries for physics has risen by 19.6% and chemistry is up 19.4%"

On the face of it, great news for everyone. Science/maths subjects are crucial in pretty much everything we do in the 21st century. However, looking at the actual numbers, this analysis starts running into problems:

Subject	Entries 2006	Entries 2011	Increase
Chemistry	40,064	48,082	20.01%
Maths	55,982	82,995	48.25%
Physics	27,368	32,860	20.07%
All subjects	805,698	867,317	7.65%

Source: Joint Council for Qualifications (JCQ) website

The first obvious issue is that my numbers just don't match the Guardian's. I will be writing to them to ask what their source is. My guess is they are either just focusing on Edexcel and not all exam boards, or maybe they are focusing on only England - the JCQ numbers above are for "All UK candidates".

The Telegraph article (link above) have used the same source as me:

"According to data published on Thursday, the number of sixth-formers studying physics alone has soared by almost a fifth in just five years to around 32,860"

so I am not completely mad.

Ignoring the difference in exact numbers, there is still a problem. More students are taking A-levels than 5 years ago. Therefore, pro-rata, the increases in science and maths are not as big as they may seem.

It is also important that they have chosen to talk about the last 5 years, because 2006 was the low point in a particularly alarming dip in the number of students studying physics A-level in particular.

To show what I mean, let's go back a bit further to the year 2000:

Subject	Entries 2000	Entries 2011	Increase
Chemistry	40,856	48,082	17.69%
Maths	67,036	82,995	23.81%
Physics	32,059	32,860	2.5%
All subjects	771,809	867,317	12.37%

Source: Year 2000 numbers are included in brackets in the 2001 report from the JCQ website - see page 10 for A-level Physics

We can see that compared to just over 10 years ago, the "boom" in students studying maths is actually much smaller, even more so when we factor in that nearly 96,000 more A-levels were sat in 2011 than 2000.

The physics number is not a typo. The "Brain Cox effect" is that all of 801 more students took A-level physics in 2011 than in 2000. Again, when we consider that so many more A-levels are being sat now, we have actually gone backwards in terms of the proportion of A-levels being taken in physics over the last 11 years.

Photo by widdowquinn on Flikr

Teachers' pets

Today it was music. And citizenship. At the same time.

Last month it was RE. For Michael Gove it's history, and maybe geography too, though no-one is too sure about how the latter slipped in. Carol Vordaman has even popped back up waffling on about hers.

Yes, everyone has their pet subject, and everyone thinks that there's is the most important, the most undervalued, the most essential for curing all of society's myriad problems. Now that the mystical Ebacc has landed from the sky, everyone also thinks that their pet subject should be included as a compulsory subject.

As a computer scientist, mine is ICT. Not many people seem too fussed about ICT (except parents, but they don't matter nearly as much as Carol - she was on the telly!). It's clearly an important subject for the 21st century, and kids have got a much better chance of getting totally minted by coming up with a cunning e-business than they are becoming a professional footballer or even the next Carol.

But I seriously hope that ICT does not become one of the chosen few included in the Ebacc. Why?

1. Beyond being able to read, write and count properly, everything else we should or should not be teaching kids is very much in the eye of the beholder. I haven't made up my mind on foreign languages, and I'm even starting to wonder about science. I definitely think teaching kids citizenship is a pointless waste of time. I'm sure many citizenship teachers would say the same about ICT, and I'd agree with them. Why?

2. ICT courses are terrible. Some more than others, but none are currently worth making compulsory. The very nature of the subject means most courses are out of date in an instant. Writing a letter in Word is an important skill, but one most kids can pick up no problem when they get out into the world of work. Unless we get some courses that includes stuff like how to set up a web server with a MySQL database it's really not worth it.

3. The worst possible way to get the average teenager to do something is to tell them that they have to do it. (Myself, I have never managed to grow out of it. Give me an order, and I will probably say no. It's a reflex. I just don't like it.) We should really let them choose the majority of the subjects that they study for a variety of obvious reasons, not least of all that they are much more likely to pay attention when I try to teach them.

So the music and citizenship teachers can keep it. I don't want my subject to be any part of this Ebacc malarky - it will only end in tears anyway.

c

Riots: we deserve this. We brought it on ourselves

It's time to wake up. We sleepwalked into this situation, and we only have ourselves to blame.

I rarely agree with David Cameron, but when he talked about a "broken society", this is what he meant. An article in the Independent by Camila Batmanghelidjh has been doing the rounds on Twitter, with lots of people posting with it that the author is "someone who knows what she is talking about".

Why are we surprised and impressed by being able to read something written by someone with a clue? She doesn't actually say anything that groundbreaking... Lots of us know - surely, lots of us do actually know - what is going on in our society, and that it is broken. We know that successive governments have deliberately neglected the poorest people in this country. It was also surely obvious that things could go wrong in this way if that neglect continued. We didn't need Batmanghelidjh to tell us.

The hard part is who is to blame. I have just read a blog post by Sunny Hundal that makes the point I was about to make: that by sitting at home, tweeting and blogging, we are not part of the solution, so we are part of the problem. We brought this on ourselves, and we need to get out into the "community" and work to support those who need it. Not out of altruism, but out of what we have seen now: if we don't do this, our cities will burn.

We have to take the blame for trying to kid ourselves that we can either ignore the people struggling in society, or rack up huge national debt based on make-believe financial systems to try and pay for it. If we want to maintain our society, our quality of life, our safety, we have to pay for it through taxation. We also need to make sure the banks and multinationals start paying their taxes too.

We must also realise that we are also to blame for voting in these governments that have conned us that all this will work. I like Ken Livingstone less the more I learn about him, but he got it spot on when he called our top politicians a "cabinet of millionaires" - but we picked those people! Until we realise that professional politicians picked out from the social elites who have no experience of the real world are not the way forward, this will continue to happen again and again for generations to come.

We need to build better communities. We need to elect proper people of substance as our politicians instead of the hairdos we've got at the moment.

The important thing is that we now do this. We must not simply talk or type about it. We have to blame ourselves, take responsibility as individuals and a society, and start to become part of the solution.