As part of the IRCHSS funded final wave of the 2002-2007 election study, a full 2002-2007 data set has been prepared for general distribution. This has been carried out by Jane Suiter under the direction of the Principal Investigator, Michael Marsh.
The combined data set is constructed by including all variables in any of the five waves, giving a standard variable name so that where a question is item is repeated across more than one wave it can be identified easily.
In addition, there are two data sets. The first is a 'long' data set, which takes the variable names in the codebook and presents the data wave by wave. This enables a very simple analysis of any question by wave, and allows the same analysis to be carried out on one or more waves with the same set of commands.
The second data set is a 'wide' one - this takes the respondent as the unit of analysis, and contains all responses across all available waves for each respondent. The variables names here are as in the codebook, but with a suffix indicating wave year 02, 03, 04, 06 and 07. This data set allows analysis of responses by the same respondent across different waves, such as how did a person's vote in 2002 change (or not) in 2007.
Anyone is free to download and explore this data. All questions and suggestions for improvement should be addressed to Jane Suiter: .
|Long datafile in SPSS file format||Download|
|Long datafile in STATA 10 file format||Download|
|Wide datafile in SPSS file format||Download|
|Wide datafile in STATA 10 file format||Download|