I’ve loaded up the data page with more than 20 new public opinion surveys. That’s the good news.
The bad news is that these are all surveys fielded in Iraq by D3 Systems and KA Research Limited. So I expect them to be loaded with fabricated data. Ultimately, I think the main value of these datasets will be that they will help people to refine their skills at detecting fabricated survey data.
I’ve now posted everything I have by D3/KA.
There is more D3/KA Iraq data out there but organizations like ABC News and Langer Research Associates are hiding what they have. There have been some developments on this front which I will start sharing in a new regular feature of this web site which I will call Secret Data Sunday.
Another poll in two versions which seem to be slightly different on the data page.
Hi. I got kind of wrapped up with submitting a paper to a journal last Friday and didn’t mention to dump any data but I’ve just make up for this by putting a Kirkuk poll on the data page.
I’ve posted one more Iraq poll on the conflict data page.
More importantly, after help from Andrew Wheeler I can now extract codebooks from SPSS datasets. So I’ve gone back and posted proper codebooks for all the datasets on the site.
I’ve put another Iraq poll up on the data page. There seems to have been some coding errors with the first version of the dataset so there is a second, correct version which I have also posted.
In the process of doing this it suddenly dawned on me that I need to do still more for people who don’t have SPSS. I’ve been posting the poll questions which you can’t see in the csv files. But you also need to see the possible answers or else you can’t interpret the numerical answers you see for most questions.
The problem is that I see no easy way to pull these lists of answers from SPSS because, so far as I can tell, I would need to go through question by question, cutting and pasting each time. There has to be an easier way.
Can anyone out there help me on this?
Hi everybody. I’m on the road this week and next and realize that I’m lacking certain capabilities on my laptop that I need to do more data postings. So please give me a little extra time and I’ll do a big dump when I get back.
I managed to get another dataset up on the conflict data page today.
More importantly, I realized that the csv files I’ve been posting for the people who don’t have SPSS were actually not very useful for….errr… people who don’t have SPSS. The reason is that the questions are lost when I do the conversion.
So I’m now posting pdf files with the missing questions. Today this seems to have added about an hour to my dumping activities but after some practice I should be able to move much faster.
Meanwhile, the feeling of despair grow. At the current rate I’ll have a long, grey beard before I finish posting all the data. I’ve gotta look into getting a grant.