|
?:comment
|
-
Factor: Sample size adjustment
(en)
-
VARIABLE NOTES: D1013
This variable reports the ratio of the average sample size to each election study sample. Note that this factor is calculated on the basis of the samples appearing in the CSES data files (i.e. does not incorporate booster samples, panel respondents who did not participate in the CSES wave of multi-wave studies, etc.). Further, this factor treats elections, and not political systems, as the unit of analysis. Analysts wishing to compare across-countries, instead of a cross-election studies, should adjust this weight accordingly.
The resulting factor is then used to create the derivative "Dataset Weights" in variables D1014_1 through D1014_3.
To follow is the STATA code used to create variable D1013:
gen n=1
gen tot_obs = _N /*Number of observations*/
gen estudies = _S /*Number of election studies*/
gen mean_res = tot_obs/estudies
gen n_cases = .
foreach x of local elec {
su n if D1004=="`x'"
replace n_cases = r(sum) if D1004=="`x'"
}
replace D1013 = mean_res / n_cases
drop n-Ncases
The STATA code to create the derivative variables in the CSES dataset was run on the original, unrounded version of the original weight variables (D1010_1-D1010_3). Thereafter the derivative variables were rounded to four decimal places at maximum (i.e. 1.1234) using STATA.
It is due to this rounding that the mean value of derivative weight variable D1013 for the full dataset is close to, but not necessarily exactly equal to, 1.0000.
(en)
-
VARIABLE NOTES: D1013
This variable reports the ratio of the average sample size to each election study sample. Note that this factor is calculated on the basis of the samples appearing in the CSES data files (i.e. does not incorporate booster samples, panel respondents who did not participate in the CSES wave of multi-wave studies, etc.). Further, this factor treats elections, and not political systems, as the unit of analysis. Analysts wishing to compare across-countries, instead of a cross-election studies, should adjust this weight accordingly.
The resulting factor is then used to create the derivative "Dataset Weights" in variables D1014_1 through D1014_3.
To follow is the STATA code used to create variable D1013:
gen n=1
gen tot_obs = _N /*Number of observations*/
gen estudies = _S /*Number of election studies*/
gen mean_res = tot_obs/estudies
gen n_cases = .
foreach x of local elec {
su n if D1004=="`x'"
replace n_cases = r(sum) if D1004=="`x'"
}
replace D1013 = mean_res / n_cases
drop n-Ncases
The STATA code to create the derivative variables in the CSES dataset was run on the original, unrounded version of the original weight variables (D1010_1-D1010_3). Thereafter the derivative variables were rounded to four decimal places at maximum (i.e. 1.1234) using STATA.
It is due to this rounding that the mean value of derivative weight variable D1013 for the full dataset is close to, but not necessarily exactly equal to, 1.0000.
(de)
|
|
?:groupDescription
|
-
CSES is a collaborative program of research among election study teams from around the world. The research program as well as the questionnaire and the study design are developed by an international committee of leading social researchers. Participating countries include a common module of survey questions in their post-election studies. The data contain information on individual choice behavior, demographics, the constituencies, the national election results and the political systems. The studies are then merged into a single, free, public dataset for use in comparative study and cross-level analysis.
(en)
-
Die CSES ist eine weltweite Kooperation von Wahlforschern. Das Forschungsprogramm, der Fragebogen und das Studiendesign werden durch ein internationales Planungskomitee führender Sozialforscher erstellt. Teilnehmende Länder implementieren einen einheitlichen Fragenkomplex in ihre Nachwahlstudien. Die erhoben Daten umfassen das individuelle Wahlverhalten, politische Einstellungen und sozio-demographische Merkmale sowie Angaben zu den Wahlbezirken, den nationalen Wahlergebnissen und dem nationalen politischen System. Die einzelnen Länderstudien werden in einen gemeinsamen Datensatz integriert und stehen der Wissenschaft für vergleichende und länderübergreifende Untersuchungen frei zur Verfügung.
(de)
|
![[This page as RDF]](https://data.gesis.org/gesiskg/static/rdf-icon.gif){kind=icon}