A little learning is a dangrous thing

You’re probably getting sick of hearing about the GCSE paper in ICT that my colleagues and I tried last week. But no one’s yet mentioned a particular question which I think is the very worst I’ve ever seen in an exam.

It’s in a section on spreadsheets. You get a picture of a mock-up screen from a generic spreadsheet package and there are various questions about what you can see. One of the cells contains the number “4”, and, for one mark, candidates are asked to “identify the cell format.”

Of course, that’s an impossible question. I guess the examiners are looking for “number”, but we don’t know anything about this made-up spreadsheet. Who’s to say the format isn’t “integer” or “real”? Or something more complicated, like “short” or “double”?

If this were Excel, the format would most likely be “general” – but then it could be “fraction”, “special” or “custom”, all of which would, by default, appear identical.  It probably wouldn’t be “number”: that defaults to two decimal places, so unless they’d deliberately customised the format the figure would show up as “4.00”.

There is one clue: the cell in question appears to be involved in a nearby calculation, suggesting that it is a genuine numeric value. That would mean the cell format isn’t “text”, “image” or anything like that.

But then that might be a red herring. Perhaps the formula is picking up the value 4 from somewhere else; or perhaps there’s no formula at all and someone’s simply typed in the result. Without inspecting the cells, there’s no way to tell.

All right, I’m being deliberately obtuse. I realise that, for a GCSE question worth one mark, candidates aren’t expected to worry about all these vexing possibilities. They’re simply expected to reach a plausible interpretation of what they see and move on.

But that’s precisely my objection. This is a question that rewards kids for making assumptions about data types and functions. That’s a terrible lesson to be teaching. It’s diametrically opposite to what I would call an education in IT.

It’s particularly troubling because this isn’t some obscure issue that only matters to the tiny proportion of candidates who go on to more advanced qualifications and careers in computing. Of course, it’s a bad start for them too, but after their first few all-night bug-hunts they’ll learn not to make such assumptions.

What I’m really worried about is the countless students who don’t take their computing education any further than GCSE. They’re going to emerge into the world of work thinking they have a solid grounding in office applications – while in reality they’ve been taught a naïveté that can only trip them up. I’m sure they’ll develop a suitable scepticism after working with spreadsheets in earnest for a while, but they’ll have to gain that experience the hard way, by making potentially costly mistakes in a real business environment. What was the point of the GCSE again?

Perhaps I’m getting a little too hung up on this one issue. It’s a terrible question, but it’s far from the only dubious part of the GCSE paper. Almost every page contains something unrealistic, vague and/or simply idiotic.

But I keep coming back to this particular question because it has a unique distinction that I don’t believe I’ve seen anywhere before: learning to answer it correctly actually makes you worse at the subject. That’s pretty special.