Well, Graham’s comments notwithstanding… and also bear in mind that I can’t find copies of any of the previous surveys to look at, so I don’t know if this is in part duplicated. Also, it generally takes quite a few hours of consideration to get really good questions, and I don’t have too long, apparently. So these are just an outline of the procedure.
At the initial stage generally the questions to ask (and this would work both ways around, for either mag, and probably should be done that way) would be the ones that will help to categorise the readership. From this you should get a better idea of why people subscribe to one of the mags, but not the other one. So it’s not necessarily all questions for all readers, although the first part certainly is.
I think initially you might want to get people to describe themselves in terms of what they do, and the way to do this is to take account of the fact that people may fit into different categories in differing degrees. And this inevitably has to go into sections with sub-sections. Fortunately it works with tick boxes, and although it is very much based on a Lickert analysis, it’s slightly modified because it rather assumes the attitude part. Also please note that it is not asking people what they like!
So across the top of the grid we get a continuum with five categories, which says something like:
It’s a primary activity – I do a lot of this – I do very little of this – I don’t do this at all – I wouldn’t dream of doing this
And down the side there will be a list of activities: (you may be able to think of more; this is just representative)
I construct replica models
– from kits
– from raw materials
I use engineering processes and machines
for repairing and rebuilding machinery
– workshop
– other
I carry out experimental engineering and processes
– with a view to making models
– with a view to making and improving engineering machinery
It’s important to ask specifically whether they are subscribers to one or both mags too – but I suspect that happens anyway. From this information you can glean quite a lot about the way people approach their engineering hobby, whatever it is.
Another thing I’d consider might be to list some recent articles from both mags, only mixed up, and see which mag people think that they should be in. You could also include a few future articles in this as well, if that’s possible. Or even make a few up, especially if you can find some titles that might be ambiguous. I certainly found a few, just looking at the few MEs that I have. The thing here is that you shouldn’t allow an indecision point (no don’t care option):
So across the top it will say
Should definitely only be in ME – would prefer this in ME – would prefer this in MEW – should definitely only be in MEW
And then some articles like
How to drill square holes (this was ME 4285)
Musings on Drawings, Dimensions, tolerances and marking out
A Simple Spark Erosion Machine
…etc. On the face of it, the answers to a lot of this might seem blindingly obvious, but if you correlate this with the answers to the previous questions, it might well get rather more interesting.
Unfortunately I don’t have time to develop this fully today, but the general idea is that no one section on its own gives you definitive answers – you have to get sneaky, and not necessarily make it plain what you are trying to get out of people. Not because you are trying to trick them, but simply because they may not have thought through what it would be that you need in order to satisfy what they are actually doing, and might need. It’s a subtle shift of emphasis, really.
The other thing that I would prefer to do, especially on a mass survey, is to trial a few of these correlated questions first, and generally refine the way that the final survey was built. As with all surveys, you have to ask yourself some careful questions about what you really need in the way of answers before you commit to a final version.
Oh, and for everybody else, the reason that we generally go for tick-boxes if we can is that these are so much easier to analyse than written comments are, that you almost wouldn’t believe it. I did a mass survey about fifteen years ago with two comment boxes, and it took about 8 weeks to analyse and break these down, compared to a couple of days for the tick-box answers. Since then I have given a lot more thought to other ways of extracting the same information without the comments!
Anyway, I hope that this is of a little help – even if it’s more in the future than the present.