WOW! Stupid Survey!

My survey might finally be posted to the internet where people can go out and fill it in and I might get the results. It only took three different attempts and finally writing the survey and capture code by myself.

Initially I was going to use an open source product called Lime Survey that Dr. Pfaffman has set up for us. This cumbersome piece of software looked like I might be able to use it until 2 hours later I finally made it to the part of my survey with a semantic differential. For those that don't know, that is a question with a word on both ends and you select the choice that is closest to your feeling for example,

I think this blog post is:
Brilliant 1 2 3 4 5 Stupid
A good source of information 1 2 3 4 5 A bad source of information

etc...

Anyway, Lime Survey won't do that. We have this group of people in OIT that are dedicated to statistics. SPSS makes a program called MRInterview. Since SPSS is the leading statistics software package one would think that their survey software would be superb. NOPE! It was slightly less cumbersome than Lime Survey, but I was left at the point where I e-mailed back and said: "My survey is now ready for you to finalize so I can use it. Please let me know what else needs to be done."

Apparently MRInterview doesn't do S.D.s either. I thought this was a standard survey item type, but I must be wrong about that. They need to apply a template that was written by one of the people that use the software... that's in limbo at the moment.

SO.... I wrote my own form in PHP which is SIMPLE and laid it out how I want it. Then I wrote code to capture the results which ended up being easier than I anticipated. The only problem I encountered was skipping a page when someone indicates that have never participated in a webcast. Then I had to generate a userid for each person. I was going to have a file with a number in it, but I used the date function to generate the number instead.

Of course the HOURS I spent with Lime Survey and MRInterview were all complete wastes of time and one would think that with all these people out there doing research and putting surveys online that there would be a simple solution that would include at least the general survey type questions that are dicussed in most overviews of quantative research!

Ok, I'm done... take the box away. I'm going to work on my paper now.