Options
Validation of the Global Personality Inventory in Singapore
Author
Lim, Li Ai
Supervisor
Tay-Koay, Siew Luan
Abstract
This research study was aimed at validating the use of the Global Personality Inventory (GPI) in the Singapore context. The GPI was based on the Five Factor model of personality and was developed to measure personality attributes and competencies underlying job performance. It was designed for use in recruitment, succession management and people development in organizational settings across cultural barriers.
A stratified sample of 30 individuals at the officer and managerial levels with various educational levels and working in the public and private sector in Singapore was administered the Global Personality Inventory (GPI) and Neuroticism Extraversion Openness Personality Inventory - Revised (NEO PI-R) over a period of 2 months.
The validity of the GPI scales was investigated by conducting construct validation analyses. The validity of the instrument was confirmed by the high internal consistency of the scales; high correlation between similar GPI and NEO PI-R scales; high correlation between scales within each GPI Performance factor; high correlations between related GPI scales under different Performance factors, and differences in mean scale scores between groups varying in educational level, working in public versus private sector as well as salary range and number of years of working experience.
The results showed, firstly, that most of the GPI scales had high internal consistencies and were highly correlated with similar scales from the NEO PI-R. However, a few scales such as the Work Focus, Openness and Risk-taking scales were quite general and seemed to cover a wide array of different constructs. Secondly, there were also highly significant inter-correlations between GPI scales grouped according to the performance factor. Thirdly, there were highly significant correlations between GPI scales that were based on the same underlying attribute, were causally or indirectly related as well as logically incompatible. Fourthly, GPI scales grouped according to the 'Big Five' factors correlated highly with corresponding domain and facet scales of the NEO PI-R. This was, however, not the case for the GPI and NEO PI-R Extroversion and Openness factors and scales. Fifthly, groups of individuals differing in educational levels and salary ranges showed significant differences on how they scored on the scales.
In general, the study validated the constructs used in the GPI to some extent. More comprehensive and sophisticated research is required to determine the content, concurrent and predictive validity of the instrument in Singapore.
A stratified sample of 30 individuals at the officer and managerial levels with various educational levels and working in the public and private sector in Singapore was administered the Global Personality Inventory (GPI) and Neuroticism Extraversion Openness Personality Inventory - Revised (NEO PI-R) over a period of 2 months.
The validity of the GPI scales was investigated by conducting construct validation analyses. The validity of the instrument was confirmed by the high internal consistency of the scales; high correlation between similar GPI and NEO PI-R scales; high correlation between scales within each GPI Performance factor; high correlations between related GPI scales under different Performance factors, and differences in mean scale scores between groups varying in educational level, working in public versus private sector as well as salary range and number of years of working experience.
The results showed, firstly, that most of the GPI scales had high internal consistencies and were highly correlated with similar scales from the NEO PI-R. However, a few scales such as the Work Focus, Openness and Risk-taking scales were quite general and seemed to cover a wide array of different constructs. Secondly, there were also highly significant inter-correlations between GPI scales grouped according to the performance factor. Thirdly, there were highly significant correlations between GPI scales that were based on the same underlying attribute, were causally or indirectly related as well as logically incompatible. Fourthly, GPI scales grouped according to the 'Big Five' factors correlated highly with corresponding domain and facet scales of the NEO PI-R. This was, however, not the case for the GPI and NEO PI-R Extroversion and Openness factors and scales. Fifthly, groups of individuals differing in educational levels and salary ranges showed significant differences on how they scored on the scales.
In general, the study validated the constructs used in the GPI to some extent. More comprehensive and sophisticated research is required to determine the content, concurrent and predictive validity of the instrument in Singapore.
Date Issued
1999
Call Number
BF698 Lim
Date Submitted
1999