Adapting the MMPI for Use in Assessing Late Adolescents in Trinidad and Tobago

Article excerpt

Trinidad and Tobago are islands situated in the southernmost point of the archipelago of the West Indies. Together they form one country, which achieved its independence from British rule in August, 1962, and became a republic in August, 1976. The population now is a little more than one million, mainly of African and East Indian origins, resulting in two distinct subcultures despite some acculturation. It should also be noted that the United States is exerting increasing influence by exporting its culture through movies, television, and radio programs. Professionals in the mental health field now practicing in these islands are primarily graduates of programs in Canada, the United Kingdom, and the United States.

To determine the effectiveness of the Minnesota Multiphasic Personality Inventory (MMPI), one of the most widely employed assessment instruments in the United States, for use in Trinidad and Tobago, it was important to assess its readability and discriminative power in this cultural setting. The most amenable population for this initial investigation was normal and abnormal late adolescents (Phillips, 1984).

In discussing the transplant validity of the MMPI in cultures differing from the United States, Butcher and Pancheri (1976) concluded: "The importance of adequate item translation cannot be over-emphasized" (pp. 259-260). Many of the adaptations of the MMPI to other countries covered in their survey involved not only translations to a new language, but also alterations in item content to fit social customs, religious orientations, or environmental circumstances peculiar to the new locale. In addition, efforts to maintain a comparable level of the vernacular in the new language--one that characterizes the MMPI statements in their original form-have often required that translators of the test employ idiomatic constructions not usual in more formal literary translations. Thus, even though English is the common language throughout the Republic of Trinidad and Tobago, special steps had to be taken to adapt the MMPI for practical application to clinical problems on these islands.


Item modifications. In order to maximize the idiomatic acceptability of the test items to subjects in this investigation, a pilot study was carried out with a sample of 23 young residents of Trinidad to identify potentially troublesome items in the MMPI. A special answer sheet was developed for use with Form R of the original MMPI. In addition to the usual response alternatives of "true" and "false," a third response was permitted, "don't understand" (DU). A criterion of 20% or higher of DU responses in the pilot group was established to identify items that were difficult for readers to comprehend. As a result, ten items needing restatement in a more idiomatic form were identified. Based on suggestions from both mental health professionals and students enrolled in area schools, these items were slightly rewritten to incorporate phraseology more suitable to local usage. Table 1 lists the original idioms that appear in these ten items together with the modifications introduced to make each more understandable. (It is noteworthy that all but two of these items--246 and 482--were either altered or dropped from the MMPI test booklet during the course of restandardizing the original MMPI and developing a test booklet for the MMPI-2 [Dahlstrom, 1993].) For use in the main investigation, the modified answer sheet format was retained and the ten reworded items were appended to the short form (399-item version) of Form R. Thus, the total number of items used in the main testing survey was 409.

Research samples. Subjects ranging in age from 15 to 20 years were recruited from three sources: several high schools in Trinidad and Tobago, adolescent patients in treatment for diverse emotional problems, and juvenile offenders incarcerated for various delinquent acts. Table 2 shows the numbers of subjects of African and East Indian background in the normal samples with valid test results, together with the numbers of each gender from the two deviant populations. …