- Longer (at least ten sessions), more comprehensive interventions, although requiring much more time from the client and the counselor, will have roughly twice the beneficial effects of briefer interventions (Baker & Popowicz, 1983; Oliver & Spokane, 1988; Phillips, Friedlander, Kost, Specter-man, 8c Robbins, 1988).
- Long-term effects of career interventions have been demonstrated for up to six months after treatment, with little loss of potency (Azrin, Philip, Thienes-Hontos, & Besalel, 1981; R. A. Myers, 1971; Johnson, Johnson, & Yates, 1981), although "booster" treatments have not been tried. R. A. Myers (1971) suggested that the follow-up period most appropriate for a career intervention is the time at which the next career decision must be faced, but assigned no time interval to that next point.
- Individual counseling, although much more costly than other approaches, is the most efficient intervention in terms of amount of gain per hour of effort (Oliver & Spokane, 1988).
- Inventories and self-assessment devices are still the most sought-after and effective intervention strategies in the counselor's armamentarium, with computer-assisted interpretations enhancing their effects with written information and media presentations (Fretz, 1981; Kirschner, 1988; Phillips et al., 1988; Wiggins, 1987).
- Job-seeking self-presentation skills can be effectively learned using role plays, role models, and group discussions (R. A. Myers, 1986).
- Typically, one can expect an average of one standard deviation of gain following a career intervention on measures of career maturity, decisional status, appropriateness of choice, and information-seeking behavior (Krumboltz et al., 1979; Spokane & Oliver, 1983).
- On the average, clients will seriously consider one additional career option following a career intervention (Fretz & Leong, 1982).
- The teaching of cognitively oriented decision strategies to intuitive deciders (Harren, 1980) is the only intervention that has been found to produce consistently ineffective and/or harmful outcomes (Holland et al., 1981).
- Individuals who are low in self-esteem and social skills, who are high in indecision and low in vocational identity, or who have unrealistic aspirations will have better outcomes in individual counseling than in self-guided or group interventions (R. A. Myers, 1986).
- Special groups (e.g., gifted or minority individuals) will have better outcomes with more structured as opposed to more vague and diffuse career interventions (Savickas, in press).
Evaluating the Full Range of Career Interventions
A volume edited by Krumboltz and Hamel (1982) contained an informative set of articles on career program evaluation. An effective evaluation-accountability model, according to Krumboltz (1974), must accomplish several outcomes: (1) define the counselor's role and the goals of counseling; (2) chart observable behavior as the outcome; (3) state costs; (4) promote professional effectiveness rather than assigning blame; (5) tolerate failures; (6) seek input from all concerned parties; and (7) be open to periodic revision.
Special Considerations in Evaluating Individual Interventions
The collection of individual client data and ratings by the counselor may be sufficient for a superficial evaluation of individual career interventions. Follow-up data on graduation rates, employment stability, use of counseling services (Spokane, 1979), and the ability to make an improved career decision (R. A. Myers, 1971) are all important aspects of such an evaluation. In addition, the cost effectiveness in terms of counselor time, testing expenses, and resource center usage, and the utilization rates among the population in question should be considered. In other words, how many clients actually use the available counselor time? Some estimates suggest that a small proportion of a given population (usually 20 percent) will utilize most clinical time (generally 80 percent).
Finally, process measures, especially of the working alliance, and/or counselor evaluation may be important in evaluating intervention effectiveness.
Special Considerations in Evaluating Group Interventions
There is some disagreement among evaluation specialists about the unit of analysis in group interventions. Whereas studies of group interventions have generally presumed that each individual within the group should be considered as a separate subject, many specialists suggest that the group as a whole should be aggregated and treated as a single subject. When the group is the unit of analysis the degrees of freedom and power are vastly reduced, which means that many more groups must be studied than are typically required in career intervention research and evaluation. As an example, two counseling groups with ten participants each would be treated as only two subjects, rather than twenty. In either case, multiple groups should be evaluated in any careful evaluation of group interventions. The success of the group as a whole can be as important an outcome as the behavior of the individuals within the group.
Also, in evaluating group interventions, it is important to consider whether participants have received individual counseling during intake or at some previous time, and whether the group is viewed as a supplement to individual intervention or as the only intervention. If clients are receiving several intervention modalities simultaneously, it may be difficult to establish the contribution of each. Dropout rates are generally high in group interventions. Ongoing groups occasionally allow temporary membership so that individuals may enter and leave freely. It should also be remembered that the availability of the group is frequently a more salient determinant of referral for intervention than client diagnostic status. Also, as indicated, some individuals do not benefit from group interventions.
Finally, group process factors (e.g., cohesion and leader style) may be important indicators of the success of a career group. Some groups may take on a very different tone than others, and the tone may be more responsible for derived benefits than any particular treatment content. The level of structure (structured versus unstructured) in the group should also be assessed, and an attempt should be made to determine whether the groups being evaluated were actually similar in tone and process.
Special Considerations in Evaluating Classroom Interventions
Since classroom interventions involve large groups, multiple classes should always be evaluated. Tone and process questions may also be important. In addition, classroom interventions last longer than any other interventions (Oliver & Spokane, 1988) and thus more susceptible to the influences of time (T. D. Cook & Campbell, 1979). Also, the possibility that subjects received additional assistance is quite large; the interim measurement of outcomes may reduce this problem somewhat. If multiple instructors or guest lecturers are employed, their contributions should be assessed separately.
Special Considerations in Evaluating Computer-Assisted Interventions
The evidence suggests that the length of time a client spends on the computer is an important determinant of the benefit derived. In addition, some individuals will derive more benefit from computer interventions than others, making individual differences a factor in evaluations. As with group interventions, computer interventions are sometimes used as standalone treatments and at other times as adjuncts to a comprehensive intervention package, which makes their evaluation more difficult.
The price of hardware, software, and annual maintenance fees must be figured into the costs of computer-assisted interventions. The initial costs of purchase, however, may be offset over time if a large number of clients use the system; computers are also decreasing in price, as are the microcomputers required to operate them. Evaluation of these systems has frequently been conducted in terms of user satisfaction rather than client outcomes, but more rigorous studies are now beginning to appear.
