Assessing general and work values among university students in Hong Kong

Abstract

 This study had two aims. The first was to examine the reliability and validity of the Chinese version of Schwartz Values Survey(SVS-58) and Super’s Work Values Inventory (WVI-45) for measuring general and work values of Hong Kong Chinese university students. The second aim was to explore the relationship between ‘general’ and ‘work’ values.

The validity and reliability of the two measures were checked using a sample of 211 university students from various subject majors in five universities in Hong Kong. Data collection was facilitated through the help of staff in the Student Affairs Office and in academic departments. Respondents’ participation was voluntary, and their anonymity was ensured. The composite questionnaire used to obtain information contained the Chinese versions of SVS-58 and WVI-45,plus the Rosenberg Self-Esteem Scale and the Marlowe-Crown Social Desirability Scale. Participants also provided basic demographic information. The same methodology and instruments (SVS with one item deleted and WVI with 3 items removed) were adopted later in the main study involving a larger sample—451 students from seven universities.

Results of the pilot study showed that SVS-58was acceptable as a reliable and valid instrument with this population. The reliability of the Hedonism subscale was enhanced by deleting an item. Some of the WVI subscales were found to be unreliable, but the reliabilities increased when 3 problematic items were removed. The construct validity of WVI-42was confirmed by Confirmatory Factor Analysis in the main study sample.

In terms of relationships between ‘general’ and ‘work’ values, Roe and Ester (1999) had proposed three possible scenarios: (1) work values and general values are discrete constructs, or (2) work values are a sub-set of general values (as represented in a model developed by Ros, Schwartz, and Surkiss, 1999), or (3) general values are a subset of work values. Due to the lack of well-established theory for studying possibility (3), this study focused on the first two.

The main study with a sample of 451 students showed that both scenarios (1) and (2) were to some extent supported by findings from Multidimensional Scaling, Exploratory Factor Analysis, Pearson Correlation Analysis, and Canonical Correlation Analysis. This was in spite of the unexpected results from Hierarchical Cluster Analysis and Structural Equation Modelling. The findings here can therefore be used to suggest a combination of scenarios (1) and (2) as an explanation of the relationship between ‘general’ and ‘work’ values, with work values being both a distinct construct to some extent but closely integrated with general values. Clearly, these findings cannot rule out the possibility of scenario (3). The findings indicate that classifications of general and work values proposed by Roe and Ester (1999) seem not to be mutually exclusive. The findings also provide empirical support for the model proposed by Ros, Schwartz and Surkiss (1999) ―with a theoretical link between general and work values. The implications for career counseling point to a need to include assessments of both general and work values when planning individualized comprehensive career guidance for university clients.