In this publication (in press) at Frontiers in Psychology – Cultural Psychology, we challenge the current state of the art in culture-level value research. We likewise suggest a re-conceptualization of cultural level values along an orthogonal structure defined by dimensions of Alteration/Preservation and Amenability/Dominance.
A new empirical approach to intercultural comparisons of value preferences based on Schwartz’s theory is an empirically-driven article (uses two rounds of the European Social Survey) that highlights shortcomings of the available approach of arriving at cultural level value preferences from individual level value preferences, the so-called averaging approach. The fictitious middle individual on which the averaging approach is based is, we argue, an improper empirical reproduction at the cultural level of the true value profiles of individuals of a country. As in scale construction, where one must demonstrate that scale-items reliably pertain to one common latent factor, so is the case in constructing a culture-level construct from individual-level observations – one needs to show that there is sufficient homogeneity between value profiles of people in a country before averaging over them to arrive at a culture-level concept. This, however, is not the case in practice. Moreover, we also know from past research that in some cases there are negative correlations between individual-level observations that are otherwise disregarded in the averaging approach. We propose the distribution approach as an alternative.
This method facilitates via an unfolding technique a direct reproduction at the cultural level of the individual level values. Observed value profiles of individual members of a country are compared against theoretical relations of value compatibility-incompatibility (circumplex value model unfolded as ideal value profiles, Table 1), comparisons which subsequently serve to classify each case into one of eleven value classes, 10 as theorized by Shalom Schwartz and 1 as non-classified.
After a value class is assigned to each individual, we calculate frequencies of value classes in each country which are then transformed into rank-orders. Based on the rank-transfored distribution of value profiles we then perform Principal Component Analysis and extract as substantially meaningful two components – two ways in which value profiles of individuals organize collectively at the cultural level (Figure 1).
Finally, using these two newly found dimensions we can predict each country’s cultural-level value preferences from indices of societal challenges (education, religiosity, ethnic fractionalization, etc.) (Figure 2).