The tradrat and survself scores are the two variables that are used to build the Cultural Map.
These factors are the result of running a factorial over a set of variables (using the longitudinal file variable names). The procedure should be run ONLY over a time-series data-set (ideally: merged WVS+EVS (European Values Study) data-sets).
- A008 Feeling of happiness
- A165 Most people can be trusted
- E018 Future changes: Greater respect for authority
- E025 Political action: signing a petition
- F063 How important is God in your life
- F118 Justifiable: homosexuality
- F120 Justifiable: abortion
- G006 How proud of nationality
- Y002 Post-Materialist index 4-item
- Y003 Autonomy Index
The specific syntax in SPSS is as follows:
MISSING VALUES a008 a165 e018 e025 f063 f118 f120 g006 y002 (-9 to -1).
MISSING VALUES y003 ().
FACTOR /VARIABLES a008 a165 e018 e025 f063 f118 f120 g006 y002 y003
/ANALYSIS a008 a165 e018 e025 f063 f118 f120 g006 y002 y003
/PRINT INITIAL EXTRACTION ROTATION
/FORMAT SORT BLANK(.3)
/CRITERIA FACTORS(2) ITERATE(25)
RENAME VARIABLES (fac1_1=tradrat5) (fac2_1=survself).
COMPUTE TradAgg = 1.61 * TradRat5 - .1 .
COMPUTE SurvSAgg = 1.81 * SurvSelf + .038 .
The first lines remove the MISSING VALUES definition so that SPSS doesn't skip the cases. The two factors in the FACTOR command are renamed as tradrat5 and survself. Finally, survself is inverted.
There are some important considerations to make here:
- First, the FACTOR is run over the whole aggregate of all waves (both EVS and WVS) and consequently you may get different results if the command is run over a single wave file or the WVS longitudinal file alone. This is a limitation inherent to the way these factors are computed.
- Second, any cases without answers to the "ingredient variables" may be skipped. As a consequence, some countries, will not have scores because of missing variables in their official questionnaire.
The TradAgg and SurvSAgg variables can be found in the longitudinal file: http://wvs/WVSDocumentationWVL.jsp
After building the two variables, you can find the mean by country using the command:
MEANS TABLES=TradAgg SurvSAgg BY S025 /CELLS MEAN.
This will give you the scores that can be used to draw the Cultural Map.