Data files

Data file configurations: The NAES08-Online data are provided in the following configurations:

  • All interviewing waves — full file; compact file (excludes verbatim data)
  • Individual waves — full file; compact file (excludes verbatim data)

File formats: Both SPSS (.sav) and plain text tab-delimited (.txt) formats are provided; the downloads are .zip archives:

  • The SPSS files are compatible with version 7.5 and higher.
  • The plain text tab-delimited files can be imported into a range of data analysis applications. Cases are delimited by line feed characters (Unix LF), and within cases, variables are delimited by tab characters. Variable names are displayed in the first row of the file. The files are saved in UTF-8 encoding.

File layouts: In the data files, each case/row represents a unique respondent, and each column corresponds with a variable for a particular wave.

Data columns are named based on the corresponding variable in the variable catalog, plus a suffix that indicates the interviewing wave:

  • _1 — Wave 1 (pre-primary)
  • _2 — Wave 2 (primaries)
  • _3 — Wave 3 (spring and summer 2008)
  • _4 — Wave 4 (general election)
  • _5 — Wave 5 (post-election)
  • _a — Wave A (core panel profile)
  • _b — Wave B (public affairs profile)

Variable RKEY, which uniquely identifies respondents across interviewing waves, has no suffix.

In some cases, a single variable for a wave is represented by multiple columns, for example, variables that are multiple-answer questions and those that record verbatim answers (more information).

Custom configurations: The data files can be truncated or merged to create custom configurations of multiple waves. For example, to create a data file containing Waves 2, 3, and 5, there are two options:

  • Option 1: Truncate the all-waves data file so that it contains only variables for Waves 2, 3, and 5 and respondents interviewed for these waves. As needed, retain the data for profile Waves A and/or B.
  • Option 2: Merge the individual data files for Waves 2, 3, and 5 using variable RKEY. Note that as a convenience, the individual data files repeat data for profile Waves A and B; to properly represent these variables in a custom dataset, use a merge procedure that collates repeated data (e.g., the Update procedure in SPSS).

Sample weights: Sample weights are provided to correct for known sources of error in the unweighted sample. Weights are optional and can be added to the data files of interest by matching on variable RKEY (more information).

Contents: Data for all main NAES interviewing waves (Waves 1 – 5) plus data for the profile waves (Waves A and B); includes all respondents interviewed for at least one main NAES wave (n = 28,985)

Contents: Data for Wave 1 plus corresponding data for the profile waves (A and B); includes respondents interviewed for Wave 1 (n = 19,190)

Contents: Data for Wave 2 plus corresponding data for the profile waves (A and B); includes respondents interviewed for Wave 2 (n = 17,747)

Contents: Data for Wave 3 plus corresponding data for the profile waves (A and B); includes respondents interviewed for Wave 3 (n = 20,052)

Contents: Data for Wave 4 plus corresponding data for the profile waves (A and B); includes respondents interviewed for Wave 4 (n = 19,241)

Contents: Data for Wave 5 plus corresponding data for the profile waves (A and B); includes respondents interviewed for Wave 5 (n = 19,234)

Contents: Sample weights for all NAES respondents (n = 28,985)

Separate weight variables are provided for each combination of NAES Wave(s) 1 – 5. The suffix of the weight variable name indicates the wave combination for which the weight should be used; for example:

  • WT_1 is the weight for Wave 1 analyzed individually.
  • WT_12 is the weight for Waves 1 and 2 analyzed together.
  • WT_12345 is the weight for Waves 1 – 5 analyzed together.