English title: Investigating the dimensionality of the Schwartz scale of human values: Evidence from the European Social Survey of 2002 for Greece and Slovenia

Author(s): Anastasia Charalampi -

Language: English

Type: Thesis / dissertation

Year: 2014

Abstract

Purpose: To investigate the dimensionality, reliability and construct (factorial) validity of the Short Form of the Human Values Scale as proposed by Schwartz (1992). Method: The Greek and the Slovenian European Social Survey data of 2002 (Round 1) were used. First, the samples of both countries were split randomly into two halves. For the data of the first split-half sample in both cases, item analysis was carried out to examine the distributional properties of the scale and decide on the items to be included in the analysis. For the construct validity of the scales, Exploratory Factor Analysis (principal axis factoring with promax rotation) was adopted. The structure was validated in both cases by carrying out Confirmatory Factor Analysis (maximum likelihood) on the data of the second split-half sample. Results: In both cases, the Exploratory Factor Analysis resulted in a three-factor solution. Three subscales were constructed based on the defining items of the respective factors. Reliability coefficients and internal consistencies of the three subscales showed that the third subscale was not reliable. Confirmatory Factor Analysis indicated poor fit for three of the models considered but better fit for the model defined by the following two first-order correlated factors based on 14 and 12 items for Greece and Slovenia, respectively: Openness to change/Self-enhancement and Self-transcendence/Conservation. This solution provided two subscales that were both reliable and valid. Conclusions: Our results indicated that a two factor solution was both reliable and valid. This finding does not confirm the dimensionality of the Schwartz Human Values Scale as proposed in the literature. The implications of our results suggest that further research and analysis is necessary for each country and each round of the European Social Survey.

Type/degree:

Awarding institution: Panteion University of Political and Social Sciences

Number of pages: 61