Conference Agenda

Our paper addresses two interrelated issues concerning the measurement properties and the implied conceptual interpretation of value scores. The first issue is the structural stability of the human value instrument developed by Schwartz. Over the years several studies have tested the measurement invariance of Schwartz value measures across countries and have shown some problems with the cross-cultural invariance. We extend these previous analyses and test both the relative structural features across time as well as across countries within each wave. We compare different strategies that have been suggested previously, using confirmatory factor analyses, alignment and multidimensional scaling with Procrustes rotation. The second and interrelated issue is the possibility of a general response bias factor within the value instrument. A number of authors over the years have argued that response biases such as a tendency to rate all values as more or less important need to be controlled. Such responses tendencies by default should impact structural properties and therefore, this may be tractable via traditional psychometric techniques. Beyond just being a pure methodological artefact, some evidence suggests that these response tendencies may capture substantive variance rather than representing a pure response factor in participants responding to values measures. We explore these contrasting perspectives using both cross-sectional and longitudinal data at the nation- as well as individual-level and link these patterns back to the measurement invariance properties of the instrument.

Social change among the East European countries has been transforming their cultures for three decades yet only the European Social Survey provides a systematic and repeated measurement of personal values. In this paper, we focused on the change of basic values in East European countries during the period of 2002-2018 as compared to the dynamics in the other parts of Europe. Given that the age-period-cohort analysis is futile, we fitted a series of the descriptive mixed effects models predicting the four higher order values with the year of survey, cohort, their interactions, and a number of controls. Overall, over time, East European population increasingly turned to the Personal Focused in expense of the Social Focused values. We observed decrease in Self-Transcendence as well as increase in Self-Enhancement values within East Europe, whereas outside this region, Self-Transcendence values have been increasing and Self-Enhancement values decreasing. Further results demonstrated that, first, value change as well as between-cohort differences were more pronounced in East Europe. Second, different cohorts have been converging in their values. In some countries (e.g., Russia), the convergence as well as the overall societal value change was due to the older cohorts whereas outside East Europe, younger cohorts led the value change. Third, cross-country similarity of younger cohorts’ values was the highest, and this similarity has been increasing throughout the period of study. Finally, we explored possible moderators of this change (membership in EU, income inequality, wealth) and speculate about possible contribution of the period and cohorts’ aging effects.

Human values from Schwartz’ seminal theory (Schwartz, 1992, 2012) have been used to describe differences between individuals and provide insight in their relationship with attitudes and behaviors. Values are found to be universal and although value priorities are different across countries (Schwartz & Bardi, 2001), similar value structures have been found across countries. Recent, the availability of panel data encouraged research in value change over time (Leijen et al. 2022). In the current study, using the European Social Survey (ESS), we investigate value change across five generations in 29 countries over 20 years (2002-2020). We expect these changes will not be similar across countries over time as contextual differences may affect individuals’ value development.

In Schwartz’ values theory, values are located on a theoretical circumplex with opposing and converging values. On a more general level, the values describe the dichotomy of the importance of group/individual, and change/conservation. Although scores on specific values are insightful, when analyzing value change over time it can be more insightful to look at all values simultaneously: As values are related, change in one value change also affects other values.

To estimate value change over time in the cross-sectional ESS data, following Beugelsdijk et al. (2022), we use archetype analysis (Cutler & Breiman, 1994; Eugster & Leisch, 2009). This approach allows us to describe all individuals in the data using a few distinct observations at the boundary of the datacloud. These few extreme observations are called archetypes and represent a specific combination of value scores. All other observations in the data can be represented as a linear combination of these archetypes. Archetypes are stable, making them suitable for monitoring change across groups and over time.

Using the human values in the ESS, we find 3 archetypes: Growth-focus, Self-focus, and Social-focus. Each archetype clearly shows the oppositions and compatibilities as in Schwartz’ theory. The Growth archetype considers benevolence and universalism of high importance, and power and achievement of low importance. The Self-focus archetype considers achievement and power as very important and universalism, benevolence, and tradition as unimportant. The Social-focus archetype considers security and tradition very important, and stimulation and hedonism unimportant.

We focus on value change between generations, across countries. Following Norris and Inglehart (2019), we distinguish 5 generations: Silent-generation, Baby-Boomers, Generation-X, Millennials, and Generation-Z. As expected in adults, changes in values due to time in the 20 years of the ESS are small but present. However, both between generations and between countries substantive changes in value priorities are found. Over generations, from Silent to X there is a strong increase in Self-focus and a decrease in Social-focus in people holding for all countries. Prevalence of Growth-focus is primarily explained by the political and economic situation in the country: The wealthier and more democratic a country in people's youth, the higher the prevalence of growth-focused individuals. Implications of our findings for both academics and practitioners will be discussed.