Development and Validation of a Structural Model of Work Values

Several attempts have been made to build instruments for measuring work values. nfortunately, they tended to lack theoretical support (i.e., empirically driven instruments: Manhardt, 1972), were incomplete (i.e., outdated instruments: Mirels and Garrett, 1971) or had an ambiguous conceptual/operational definition of work values (overlap between work values and other concepts such as organizational values: Van Dyne, Graham and Dienesch, 1994). Driven by the fit paradigm, whereby the congruence between the needs and values of the individual and the inducements provided by the organization should lead to positive individual and organizational outcomes, several studies have sought to define measures of values that would apply equally to individuals and organizations. Although no such instrument to measure work values currently exists, the value model developed by Schwartz can be used to build one. This conceptual framework is promising because it is the most widely tested and thoroughly validated model of general values. It was initiated by Schwartz and Bilsky (1987, 1990), further refined by Schwartz (1992, 1994, 1999) and validated on tens of samples across five continents by Schwartz and his colleagues. According to our conceptual model of values inspired by Schwartz (1992), work values, a particular category of general values related specifically to the work domain, have the same structure as the values defined by Schwartz, along two orthogonal and bipolar axes, clustering in ten distinct motivational domains. The structural relations between work values follow the circular logic of a two-dimensional space projection of the distances between their inter-correlations. According to the Smallest Space Analysis (Guttman, 1968), values that are close to each other on the map have positive strong correlations, while those placed further apart, i.e. nearer the opposite poles of each axis, have strong negative correlations. The sign and strength of the correlations between work values determine their concentric ordering around the centre of the map. This order signals their compatibility in the work value system of an individual, which affects, in turn, the individual’s behavioural choices in work situations. The adaptation of Schwartz’ inventory of values (IV) to the working world underpins our 42-value inventory of work values (IWV). In order to test the six structural hypotheses of this work value model, we collected data from a sample of 174 university professors from two major French language universities in Quebec. Respondents completed an online questionnaire in which they rated the importance of 42 values on a nine-point scale ranging from -1 (opposite of my values) to 7 (fundamental to me). Multidimensional analyses using the Guttman-Lingoes technique confirmed that work values have a circumplex structure on two quasi-orthogonal axes opposing, on the first axis, values related to self-enhancement to those related to self-transcendence and, on the second axis, values related to openness to change to those that focus on conservation. Furthermore, the data matrix fits a ten-factor structure of motivational types clustered, according to the theory, in the four sectors delineated by the two quasi-orthogonal axes, thus confirming the structural hypotheses proposed in this study. In particular, the correlational analysis at the axial level indicates that the “self-enhancement” pole is negatively correlated with the “self-transcendence” pole (-0.64 significant at p There are several limitations that affect the interpretation and generalizability of these results. First, our sample consists solely of professors, which, although considered by Schwartz and Bilsky (1987) as carriers of the characteristic values of a culture, are not representative of the entire working population. Nevertheless, the matched samples strategy allows comparisons between the results of this study and those of Schwartz et al. Second, successive MDS analyses prompted us to progressively eliminate a number of variables initially selected for testing from Schwartz’s list of general values, because of their placement on the MDS map. We attributed the misplacements to methodological explanations related to translation and adaptation of the items, yet structural inconsistencies might also have surfaced. Lastly, this model should be further validated with more diversified samples, and the issue of construct validity should be tested on a nomological network.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.