Date of Award


Document Type


Degree Name

Doctor of Philosophy


School of Education


Counseling Psychology, Ph.D.

First Advisor

Wilfred G. A. Futcher

Second Advisor

Donna J. Habenicht

Third Advisor

Marion J. Merchant


Problem. There is a need in Portugal for valid psychological instruments. The Temperament Inventory (Blitchington & Cruise, 1979) seemed adequate to help meet this need. The present study attempted to validate the TI with the Portuguese population in mainland Portugal and to obtain norms on the TI for this population.

Methodology. A factor analytical research design was used to determine whether the items selected to measure a particular temperament in the original American version did, in fact, measure that temperament when translated into Portuguese. The 80 original items, together with 10 new items designed to measure the phlegmatic and choleric factors, were translated into Portuguese. After a pilot study, 88 of these items were retained for the major study. A total of1,346 respondents included 499 males, 818 females, and 29 of unidentified gender. These subjects represented the northern, central, and southern areas of Portugal.

Tucker's coefficient of congruence was used to compare the four factors of the new Portuguese version with the four factors of the original English version of the TI.

Results. Factor analysis of the Portuguese instrument yielded four factors which were clearly equivalent to the factors of the American version, that is choleric, sanguine, melancholic, and phlegmatic factors. However, 19 items did not load on any of these factors and were therefore dropped from the instrument. The final set of 56 items included 55 of the original 80 items of the American TI and one item which had been added by the researcher.

Item analysis of each of these four scales indicated that each item had an adequate biserial correlation with the total scale of which it was a part. Reliability coefficients (KR-20) for these scales ranged from.7889 to.7964. As a result, a set of 69 items constituted the final instrument.

The results indicate that the Portuguese version of the TI has a well-established reliability and both content and construct validity.

Percentile rank norms were developed for each of the factors. A matter of concern was themarked negative skewedness of each of the four distributions.

Conclusion. The Portuguese version of the TI is a valid and useful psychological device to assess the temperament of the people in mainland Portugal.

Subject Area

Temperament--Testing, Psychological tests--Portugal

Files over 3MB may be slow to open. For best results, right-click and select "save as..."