Had a comment a while back on analyzing survey data…hmm…that is a quandary as most surveys are done on line (see Survey monkey, among others).

If you want to reach a large audience (because your population from which you sampled is large), you will probably use an on-line survey. The on-line survey companies will tabulate the data for you. Can’t guarantee that the tabulations you get will be what you want, or will tell you want you want to know. Typically (in my experience), you can get an Excel file which can be imported into a soft ware program and you can run your own analyses, separate from the on line analyses.

However…if your sample is your population and often times it is, you may not use an on-line system because your target audience is small, 30 or so. That being the case, a mail survey or a face-to-face survey will work–sometimes paper and pencil is still best.  Regardless of what form you use (and Dillman is my favorite guide), you will want to know something about your target audience. So the first thing you do is to compute the demographic statistics. Demographic statistics are the frequency and the percents of your different variables as well as the measures of central tendency (mean, median, mode) and distribution (range, standard deviation, and dispersion). These statistics may may include the demographic variables. You really want to know how many of each option you have. Most journals want to know that information as well as results of your other questions (which hopefully will tell you if you are making a difference with your program).

Then if you have a way to compare the participants in your target audience you will want to do those comparisons. Rule of thumb, it takes approximately 30 cases (participants in your target audience) to have a meaningful result. I think it is important to remember the difference between parametric and non-parametric statistics. It is rare that you will know the parameters of your target audience when it comes to descriptive statistics. I also think it is important to keep in mind that computing a mean on data that are NOT interval probably doesn’t make sense (after all, what does a mean of 3.5 actually tell you on a 4 point Likert scale?). This is just a quick review. I suggest you look in my archives for more detail–search on statistics or analysis; you will find a lot of relevant posts.

Analysis can be fun… 🙂

–molly.