Abstract
M.A.
The present study was conducted in order to construct a personality inventory that is
cross-culturally valid for the South African population with at least a Grade 12 level
of education. The Basic Traits Inventory was developed on the basis of the five-factor
model of personality, which measures personality in terms of five broad domains,
most commonly known as: Neuroticism, Extraversion, Openness to Experience,
Agreeableness, and Conscientiousness (McCrae & John, 1992). The five factors of
Basic Traits Inventory bear the same name as those in the five-factor model, and are
each divided into five facets, except Neuroticism, which has four facets. The factors
and their facets were defined after a study of the literature, and the items of the
inventory were generated in order to represent these definitions. After performing an
item-sorting procedure, some items were revised or removed. The final format of the
Basic Traits Inventory consisted of a single list of 173 items, which were all grouped
according to their respective facets, and presented together for each factor.
Three comparison groups were formed in order to evaluate the psychometric
properties of the Basic Traits Inventory, namely: Race (Black and White), gender, and
language (English, Afrikaans, and Indigenous African). The sample consisted of 1510
participants, made up mostly of second year psychology students and their parents,
with a mean age of 37.52 (range = 17 to 75). There were 566 men and 914 women (30
unspecified) in the sample. The language groups were designated as follows:
Afrikaans (n = 558); Indigenous African (n = 73); and English (n = 822). The race
groups compared in the present study consisted of 1121 White participants and 114
Black participants.
The internal consistency reliability coefficients were calculated for the factors and
facets of the Basic Traits Inventory. The reliability coefficients of each of the five
factors in the total group were all satisfactory: Extraversion (á = 0.89); Neuroticism (á
= 0.94); Openness to Experience (á = 0.90); Agreeableness (á = 0.88); and
Conscientiousness (á = 0.94). The five factors also demonstrated satisfactory
reliability in each of the comparison groups. The facets Liveliness,
Straightforwardness, and Modesty each had low reliability coefficients across all
comparison groups. Openness to Values demonstrated low reliability in all groups
except for the Black group and Indigenous African language group. Excitementseeking
and Prosocial tendencies had low reliability coefficients in the Black group
and Indigenous African language group. These facets will have to undergo revision in
future versions of the Basic Traits Inventory.
In a factor analysis, five factors were extracted using a maximum-likelihood factor
analysis with Promax rotation (k = 4) for each of the comparison groups. The resultant
factor structures were compared by rotating the structures of the comparison groups to
a target structure, and calculating Tucker’s phi for each of the factors and facets in the
structure. The White group, women, and English language group were designated as
the respective target groups for the race, gender, and language groups.
The factor structures of all comparison groups demonstrated satisfactory fit with the
theoretical structure of the Basic Traits Inventory, with exception of the Black group
and Indigenous African language group. Positive affectivity was the only facet that
did not fit the theoretical structure of the Basic Traits Inventory for all comparison
groups. In the Black group, the five factors that were extracted in the factor analysis
were identifiable as the five factors expected from the theory. There was some overlap
between Openness to Experience and Agreeableness, but most of the facets had
primary salient loadings on their posited factors. The eight facets that did not perform
as expected were: Positive affectivity, Anxiety, Order, Openness to Actions,
Openness to Values, Modesty, and Prosocial tendencies.
In the Indigenous African language group, there was a clear overlap between the
Agreeableness and Openness to Experience factors. Agreeableness was the only
factor that did not emerge clearly for the Indigenous African language group. Most of
the Agreeableness facets loaded on the Openness to Experience factor, or on a fifth
factor that could not be identified. This factor consisted of the following facets:
Positive affectivity, Tendermindedness, Compliance, Openness to Values, and
Depression.
In the factor congruence analyses, for the race groups Agreeableness was the only
factor that had a congruence coefficient less than 0.90. Six facets had congruence
coefficients below 0.90, namely Positive affectivity, Openness to Actions, Openness
to Values, Modesty, Tendermindedness, and Prosocial tendencies. For the gender
group, all facets and factors had congruence coefficients of above 0.90, suggesting
that the constructs measured by the Basic Traits Inventory manifest in the same way
for men as they do for women. For the language groups, a comparison between factor
structures of the English and Afrikaans language groups demonstrated good structural
agreement, with Prosocial tendencies and Modesty the only facets with congruence
coefficients less than 0.90. There was poor agreement between the factor structures of
the Indigenous African and English groups, as well as between the Indigenous
African and Afrikaans groups. In both comparisons, Conscientiousness was the only
factor with all its facets having a congruence coefficient of above 0.90.
The results indicated a relatively stable five-factor structure for the Basic Traits
Inventory. However, due to the small size of the Black group (n = 114) and
Indigenous African Language group (n = 73), it was recommended that more research
be done with larger samples in order to verify the structure of the Basic Traits
Inventory (Comrey & Lee, 1992). Suggestions for future research included replication
of the analyses conducted with a larger, more representative sample, translation of the
Basic Traits Inventory into Afrikaans and Indigenous African languages, the
development of a shortened version of the Basic Traits Inventory, as well as studies
investigating the convergent and divergent validity of the inventory. In conclusion, on
the basis of the encouraging results, it seems that it would be worthwhile to continue
the development of the Basic Traits Inventory.