|
The output from statistical packages often contains a baffling amount of information. Some of this will of course be important, but some of it will be of little or no value or interest to you (such as corrections to analyses for failures to meet assumptions when the assumptions have been met, tests of trend when the IV is qualitative). Thus the same output can contain material that is useful, tangential or esoteric and as a student it can be hard to work out which is which. I hope that the book and also what I have to say here helps. Do not feel alone with this problem, though – many people face it, some of us on a daily basis!
There are two golden rules to live by:
Rule one is to get help when you need it. Your tutor should be your first port of call, but if you are lucky enough to be working in a university or college, lurking somewhere there is likely to be a departmental statistics wizard, and if you are really lucky s/he may even be paid in part to provide help to students. If so, try to find out who it is, where their cave is and if and when they are available to help. Ask your tutor about whether there is someone like this in your department and if it is appropriate for you to go to see them.
Rule two is to make sure that you understand the output you include in the RESULTS. There is no point in including some feature of the output in the RESULTS if you haven’t the foggiest idea what it means. Believe me, it can be very tempting to do this but it is almost always a mistake – a dead giveaway that you do not know what you are doing. My rule of thumb here is that it is better to analyse your data in a slightly clumsy or less efficient way that you are comfortable with and understand than to perform a more purist and sophisticated analysis, but not know what it means. (This is not, however, the same thing as saying that you should not attempt to master new, more sophisticated methods of data analysis as you develop as a student.)
Like all rules these are tempting to break. Like all rules breaking them often ends in tears …
|
