Date of this Version
Published in The Future of Testing, edited by Barbara S. Plake & Joseph C. Witt (Hillsdale, NJ: Lawrence Erlbaum Associates, 1986).
This chapter is an attempt to outline where interest testing may be or should be in the near future: What changes will be seen in the development or revision of inventories, what new areas of application will occur, and what technical, social, and professional problems need resolution to get to a more desirable future.
This sounds like a rational task. I have been asked to describe a desirable future by canoeing through the rapids of psychometric fashions , disgruntled test takers, passive publishers, worried professionals and their righteous associations, and future islands of unpredictable theory. To make this task easier, the sponsor cautioned me to rely on empirical data, not daydreams.
Fortunately, I can recognize an impossible task without the aid of consultants. For several reasons, it appears helpful to redefine the task. Earlier opinions by distinguished pioneers in interest measurement have occasionally been off the mark. For example, Kuder (1954) suggested that occupational titles made poor items and that activity items would be the wave of the future . News item: Occupational items continue to be useful and popular in most inventories. And inventories that use only activity items usually include occupational titles disguised as "Be an accountant," or "Be a counselor." Developers apparently get tired of looking for good items by following a restrictive rule. At an earlier time, Strong (1943) and others dismissed a person's vocational aspiration as a weak index of the occupation a person would actually enter because this index did not have a substantial correlation with a person's measured interests. However, Dolliver (1969) started a cottage industry of research by demonstrating that aspirations and measured interests have about equal predictive validity. Later, we learned that the use of interest inventories and aspirations in tandem produced very substantial predictions.
These events and the work of futurists imply that it is helpful to see future developments not only as the continuation of current trends but also as developments that will be shaped by economic, social, technological, and theoretical forces that we cannot always anticipate or control. Consequently, I attempt to relate current developments to future developments, but my forecasts will surely be deflected by unanticipated events. I also try to distinguish long-term trends that I believe are desirable and helpful and those that may be undesirable and not helpful.
My reservations about this task have considerable empirical support. I have multiple conflicts of interest. I am the author of two interest inventories that have been the object of close scrutiny for 10 years. I am familiar with the evidence and issues about inventory biases, development, effects and usage, but my beliefs about these matters have received only mixed reviews (Gottfredson & Holland, 1978; Tittle & Zytowski, 1978) . The most constructive outcome of this experience for me has been to perceive interest inventories in the context of usefulness, validity and reliability- and about in that order.