Methodological Problems of Measuring Values and Value-Oriented Human Behavior

Abstract

Objectives. To review studies and consider the advantages and limitations of the current scientific practices for measuring values and value-associated behavior. Background. The methodological problems of measuring values and a value-oriented value-associated behavior reflect a number of difficult questions relevant for modern psychology: 1) the possibility of constructing predictive models of human behavior based on self-report data (survey methods); 2) problems with replication; 3) the use of mathematical tools for data analysis that are adequate to the specifics of the subject of social research. Key contradictions in research on values are connected with the topic of the conditions in which values influence behavior. Another important question is about the suitability of the survey method and, in particular, the ranking for measuring values, taking into account the likelihood that they are nontransitive, poorly aware and context-dependent. Russian-speaking readers are familiar with a very limited set of studies and translated methodological tools nowadays. It makes difficult to study values, taking into account the latest data on the theoretical validity of value models and value-associated behavior and on the reliability and validity of the corresponding diagnostic instruments. Methodology. The paper is a scientific review which includes a comparative analysis of the advantages and limitations of the current scientific practices for measuring values and value-associated behavior. Conclusions. It is shown that the currently widespread questionnaires for the measurement of values are based on the theoretical models, not all of which find empirical confirmation of suitability for use. Sh. Schwartz’s Survey is characterized by fairly good validity and reliability, including cross-cultural validity, but it has drawbacks common to all self-reporting survey methods. In particular, it does not allow to confidently predict the manifestation of the values in behavior. Supplementing survey data with self-reports on behavior, or reconstruction of past experience gives a slightly more complete picture, however such studies are performed in a correlation design, and therefore they do not allow one to draw causal conclusions and build reliable predictive models. The prospect for the study of values is the construction of complex models which include personal and situational variables, and the development of research procedures that measure values both in everyday life situations and in a complex multicultural context. Progress in this area will be impossible without a methodological reflection of the properties of values — transitivity, degree of awareness, universality/specificity. These are ideas which underlie the design features of research procedures and the choice of a data analysis method.