Introduction to the Special Issue on Interest Measurement
Chope, Robert C., Healy, Charles C., Measurement and Evaluation in Counseling and Development
The five articles chosen for this special issue on interest measurement represent themes that we felt were worthy of consideration. In our original call for articles, we asked that topics include material on interests and leisure, interests of lower socioeconomic status clients, interests and multiculturalism, validation of interest measurement, online interest assessment, interest assessment and constructivism, interest assessment over the life span, and responsibilities for consequential validity. We were not disappointed; many fine papers were contributed and those that were eventually chosen addressed a majority of these topics. The articles illustrate different emphases in interest assessment, including validity and logical positivism, postmodernism, contextualism, and meaning making in career choice.
The utility of interest inventory data is naturally dependent on evidence of its validity based on empirical research along with clinical use of the inventory. The long-term predictability of interest measures has been enormously important to consumers and counselors alike and, of course, will continue to be. Harrington's (2006) work has been included in this special issue because it addresses the long-term predictive validity of the interest inventory portion of the Harrington-O'Shea Career Decision-Making System (CDM; 1980, 1992, 2003) that was developed in 1975. Long-term follow-up studies (Campbell, 1966; Strong, 1955; Zytowski, 1976) of the predictive nature of interest inventories have been far too rare in the literature on interest inventories, and both Harrington and Rottinghaus and Zytowski (2006) advocate for more longitudinal investigations. Harrington's work, using a longitudinal hit rate approach summarized by Hood and Johnson (1997), provides counselors with information about whether or not hit rates for the CDM significantly exceed chance expectancies. With his results in hand, Harrington advocates for future longitudinal studies that begin in the middle schools and incorporate more extensive ethnic and multicultural diversity in sampling.
Substantial correlation of one scale with that of a second inventory assessing the same construct is another indicator of validity. The surprising findings of Savickas and Taber (2006) that some clients may hold different high RIASEC (Realistic, Investigative, Artistic, Social, Enterprising, Conventional) theme scores on different inventories underscore the need to be cautious in interpreting interest scores and remind clinicians to use multiple data points in assessing interests. The RIASEC cognitive scheme is used widely, both for categorizing interests and educational majors and occupations and for connecting them. The RIASEC scheme is also easily understood, and substantial research supports its utility for making broad, rough distinctions among a person's interests and among incumbents of different majors and occupations. Many counselors, including Savickas and Taber, believe it is helpful for clients to understand and remember the RIASEC scheme as an initial sorting device. The Savickas and Taber findings point out, directly, that there is need for more understanding of the basis for particular RIASEC estimates and, indirectly, that there is a need for information about the likely stability of people's profiles, not just their individual RIASEC scores.
In analyzing empirical evidence and deciding on interpretive statements to test out based on that evidence, counselors will want to remember the differences among inventories in their methods of rating and score computation as well as their item format. Whereas some use a single item for a client to endorse or not on two or more points, others, like the Kuder Career Search (KCS; Zytowski, 2004), use forced-choice triads. Furthermore, inventories like the Self-Directed Search (Holland, Fritsche, & Powell, 1994) use theme scores that are the sum of a client's raw scores on its subscales; theme scores for others, like the Strong Interest Inventory (Harmon, Hansen, Borgen, & Hammer, 1994), are norm-referenced standardized scores, and still others like the KCS theme scores are criterion-referenced scores. …