Personality tests

An experiment was conducted to demonstrate the effect of situational constraint
as a moderator of the predictive validity of trait constructs
and of the cross-situational consistency of behavior. Subjects were
administered the extraversion scale of the Eysenck Personality Inventory,
and their social behavior in a waiting-room situation and in a role-played
job interview was observed under conditions of either low (neutral condition)
or high (forced-introversion condition) situational constraint.
The hypothesis that, under high constraint, restriction of range on the
dependent variables would attenuate validity and consistency correlations
was only partially confirmed. The strongest finding was that judges'
ratings of subjects' talkativeness, overall exhibited behavior, and
inferred dispositional extraversion yielded significantly higher correlations
for the more subjective and broader measures than for the more
objective, narrow ones. The utility of these types of data and their
place in the consistency-specificity debate are discussed.

The purpose of this study was to examine the relationship between the personality measurements of college students according to the Myers-Briggs Type Indicator (MBTI) and the Personal Profile System (PPS). Both instruments were administered to 265 students in two community colleges and two universities in Central and South Florida. A frequency distribution was computed for the respondents according to MBTI type and according to PPS scale. Each individual PPS scale was analyzed against the four separate MPTI dimensions, and against the sixteen MBTI types. A chi-square, a lambda (asymmetric), and an uncertainty coefficient (asymmetric) statistical test were employed in the analysis of the data. The .05 level of significance was used in testing the hypotheses. The following conclusions were based on the analysis of data: (1) In every case the chi-square test indicated that a statistically significant relationship existed between the measurements of the two instruments. (2) Both the lambda and the uncertainty coefficient tests indicated that knowledge of a person's measurement on one test improved the ability to predict the person's measurement on the other test. (3) Each Myers-Briggs type related to a specific Personal Profile System scale over 60% of the time, and six of the sixteen MBTI types matched with a particular PPS scale over 90% of the time. (4) The high correlation of similar measurements of the two tests tends to reinforce the validity of both instruments. (5) Since the scales and dimensions of the two instruments did not correlate consistently in all measurements, the tests themselves cannot be used interchangeably. In summary, this study provided data which indicated that a statistically significant relationship exists between the measurements of the Myers-Briggs Type Indicator and the Personal Profile System, as noted within the limitations of the study. Possibilities for future research were outlined.