Printer Friendly

Using a problem-solving strategy to teach adaptability skills to individuals with mental retardation.

To ensure success in competitive or supported employment for persons with disabilities, concentrated efforts have been made to identify work skills considered critical for job tenure. These skills have been referred to as survival skills (Rusch, Schutz, & Agran, 1982) and include both vocational and social behaviors. Salzberg, Agran, and Lignugaris/Kraft (1986) indicated that production/work-related skills (e.g., working continuously, getting to work on time), task-related social skills (e.g., offering to help co-workers, asking for assistance when needed), and personal social skills (e.g., using social amenities, expressing appreciation to co-workers) were behaviors that contributed to entry-level employment success. Conversely, Salzberg, Lignugaris/Kraft, and McCuller (1988) reported that deficits in social skills (e.g., interacting with others) and various job responsibilities (e.g., not attending to tasks, late for work) were often reasons for job loss by persons with disabilities.

One critical skill area that has been overlooked in the survival skill literature involves the ability to adapt to changing work environments and to solve problems independently (Agran, Martin, & Mithaug, 1987). Mithaug, Martin, and Agran (1987) suggested that supported employment programs must use instructional methods to teach consumers to solve their own problems and determine appropriate responses to problem situations. Mithaug et al. referred to such problem-solving skills as adaptability skills. That is, workers need to learn to adapt to on-the-job changes in order to maintain a satisfactory level of performance (Martin, Mithaug, Agran, & Husch, 1990).

A potentially successful method of teaching problem solving to persons with disabilities is modified generation of alternatives (Foxx & Faw, 1990; Foxx, Martella, & Marchand-Martella, 1989; Martella, Agran, & Marchand-Martella, 1992; Martella, Marchand-Martella, Agran, & Allen, in press). This approach involves teaching individuals a strategy to solve problems and generate content-validated alternative solutions (Foxx & Faw, 1990). A content-validated solution is a solution considered to be appropriate by persons who are important in a trainee's environment (e.g., employers, job coaches). Rather than teaching an individual to generate multiple alternatives without placing an emphasis on the content of each solution, the modified approach involves teaching an individual to generate an initial solution, as well as to provide an alternative to solutions provided by others. In this approach, the content of both initial and alternative solutions is emphasized.

The purpose of this investigation was to assess the effects of a problem-solving skills program in which participants with mental retardation were taught to solve problems that may exist in the work place. In addition, generalization assessments were conducted. These assessments provided a measure of the participants' ability to solve problem situations verbally. Finally, the maintenance of skill performance was assessed up to 8 weeks after the end of the last phase of training.


Participants and Settings

Three individuals with mild mental retardation participated in this investigation. They were involved in counseling services at an area mental health facility; two of the three participants (i.e., John and Karen) were also enrolled in a high school special education class. Steve was selected by counselors and John and Karen were selected by a special education teacher who indicated that the participants could benefit from such training. None of the participants had received previous problem-solving training. Participant information is presented in Table 1.


All baseline, training, and generalization assessments were conducted in a 2.44 m by 2.44 m room at the mental health center. This room contained a couch and two chairs.


Twenty-four problem-solving situations and responses were developed in the following ways. First, a survey was distributed to three consumers of mental health services and four counselors at the mental health facility. This survey asked consumers and counselors to list actual problem situations experienced by consumers at work. The situations represented three categories of problems: social problems (e.g., trouble asking for help), work problems (e.g., unsatisfied with job), and personal problems (e.g., poor hygiene). There were eight situations developed per category (four for training and four for generalization assessments). Second, the lists from the consumers and counselors were combined and the situations were listed in a question format (e.g., "A co-worker does not feel comfortable asking for help when she does not understand instructions. What should your co-worker do?"). Third, the questions were distributed to the consumers and counselors, and they were asked to provide an appropriate response for participants to emit when faced with the each problem situation on the job. Therefore, there were multiple responses that were considered acceptable due to the heterogeneity of responses. Finally, after the consumers and counselors provided their responses, the solutions to problems were divided into four criterion components (see Dependent variable and measurement), which were to be used in scoring the participants' responses. These criterion components were established by noting the common structure of the consumers and counselors' responses [i.e., most responses referred to a goal statement (When will the problem be solved?, When), a person with whom to talk (Who would you talk to?, Who), the work environment (Where would you talk to (name)?, Where), and a verbal statement that would be appropriate (What would you say?, What)].

Twelve of the 24 problem-solving situations were used for training (four per category) and 12 (six similar and six dissimilar to training situations) were used for generalization assessments (two per category for similar and dissimilar situations) (see Table 2). Each training and generalization situation was printed on a 10. 16 cm by 15.24 cm index card. A 12.70 cm by 20.32 cm cue card that listed the four problem-solving components was given to each participant during the training with cue card phase.


Data Collection

Data were collected four to five times per week in sessions that lasted from 15 to 20 min. Data were collected on participant solutions to problem situations, and these solutions were compared to solutions provided by the mental health consumers and counselors.

Dependent variable and measurement. The dependent variable was participant responses to problem situations. The dependent measure was the percentage of correct problem-solving criterion components present in a response to a problem situation. Problem-solving criterion components referred to specific response elements that were required in each solution in order to be scored as correct (i.e., When, Who, Where, and What) (see Table 3). The percentages of criterion components completed correctly for baseline and training sessions and generalization assessments were obtained by dividing the number of correct criterion component responses by the total number of criterion components possible and multiplying by 100.


Observer Training and Scoring

One rater and one observer received training in scoring responses to problem situations. Training was conducted by the first author and consisted of demonstrations of scoring appropriate and inappropriate responses using the criterion components. Following this training, the rater and observer were presented with four problem situations with at least three different response solutions. The rater and observer received feedback after scoring each solution to the problem situations. Training ended when the rater and observer achieved 100% agreement with the first author on three consecutive problem solutions for each of the four problem situations.

Experimental Design

To assess the effects of the intervention, a multiple baseline design across three participants was used (Barlow & Hersen, 1984). Participants were exposed to baseline and training conditions. The training condition involved a training with cue phase and training without cue phase. The criterion for each phase of training was three consecutive sessions at 100% of the criterion components completed correctly. In addition, generalization assessments were conducted throughout the investigation.

Baseline. The trainer presented the participant with 12 problem situations per session (e.g., "A co-worker has a hard time making friends at work"), followed by instructional cues (e.g., "What should your co-worker do?"). The participant was told to listen to every problem situation and think of an initial and an alternative solution. The alternative solutions needed to include different responses to the Who and Where components. Thus, the participant generated 12 initial solutions and 12 alternative solutions. The trainer did not solve any of the problem situations or provide any feedback to the participant.

Training with cue. All conditions were the same as in baseline with the following additions. The participant was given a cue card which listed the criterion components. He or she was instructed to refer to the cue card when formulating a response to a problem situation. If a correct response was given, the trainer stated, "good answer, you included all of the components."

If a response was incorrect, the trainer indicated to the participant which component(s) was incorrect or omitted and prompted the correct use of the component(s) (e.g., "You left out the Where component. Tell me where you would talk to name"). The trainer provided the correct response(s) only if the participant was unable.

Training without cue. All training conditions remained the same except that the cue card was removed.

Generalization assessments. Generalization assessments were conducted at least once during or after baseline, training with cue, and training without cue, and 2 and 8 weeks (maintenance) after the training without cue phase ended. These assessments included either similar or dissimilar situations to those used in training. Similar situations required the same response as the parallel training situation (e.g., training = "Your coworker has trouble interacting with those in authority"; similar = "Your friend does not get along with his boss"). Dissimilar situations required a different response than one performed in a training situation (e.g., training = "Your coworker has a hard time making friends at work"; dissimilar = "Your friend states that a co-worker often uses profanity on the job and this upsets her"). During these assessments, 12 situations (six similar and six dissimilar) were presented to each participant by the observer. Following the presentation of each problem situation, an instructional cue ("What should your friend/co-worker do?") was stated by the observer. No feedback was given during or after these situations.

Maintenance. The 2- and 8-week generalization assessments were conducted to assess maintenance of each participant's problem-solving abilities. These assessments were conducted in the same manner as previously described.

Interobserver Agreement

All baseline and training sessions were scored by the trainer. The rater independently scored the training sessions from audiotapes. The trainer and rater were provided a scoring matrix for each situation. This scoring matrix included all of the problem situations and a sample of correct responses to the criterion components. The generalization assessments were scored by the observer and the rater. During these assessments, the participant's responses were audiotaped and scored by the rater. Interobserver agreement was calculated during 24% of the baseline and training sessions and 42% of the generalization assessments.

Interobserver agreement was calculated by dividing agreements of criterion components correct/incorrect by agreements plus disagreements of criterion components correct/incorrect and multiplying by 100 (Hall, 1983).

The range of mean interobserver agreement scores across participants and sessions during baseline, training with cue, and training without cue was 79% to 99% (average = 93%) between the trainer and rater. The mean interobserver agreement scores across all participants and generalization assessments ranged from 81% to 98% (average = 95%) between the observer and rater.


Baseline and Training

Figure 1 shows the percentages of criterion components completed correctly by the participants across initial and alternative responses. During baseline, John averaged 20.3% and 16.0% for initial and alternative responses, respectively. Karen averaged 16.5% for initial responses and .3% for alternative responses. Steve's averages for initial and alternative responses were 3.5% and .5%, respectively. During the training with cue phase, participant averages for initial responses increased to 99.5% for John, 96.0% for Karen, and 84.7% for Steve. Also, averages for alternative responses increased to 98.8%, 95.6%, and 82.2% for John, Karen, and Steve, respectively. All participants achieved 100% for both initial and alternative responses during the training without cue phase.

Generalization Assessments

As shown in Figure 1, all participant percentages increased across the generalization assessments for similar and dissimilar situations from the baseline condition to those measured during the training without cue phase. All participant responses generalized to similar and dissimilar situations in the generalization assessments. For example, John's percentages for similar and dissimilar situations during the baseline condition were not any greater than 21%. These percentages increased during the training with cue phase to 57% and 66% for similar and dissimilar situations, respectively. Karen's percentages also increased from a baseline mean of 19% for similar and dissimilar situations to a mean of 67% and 71% for similar and dissimilar situations, respectively, during the training with cue phase. Finally, Steve's percentages increased for similar situations from an average of 4% during baseline to 100% for the training with cue phase. His percentages for dissimilar situations also increased from an average of 2% during baseline to 100% during the training with cue phase. All participants achieved 100% of components correct for both similar and dissimilar situations during the training without cue phase.


As shown in Figure 1, percentages fluctuated for all participants throughout the 8 weeks of maintenance. During the generalization assessments, the percentages maintained for all participants at the 2-week generalization assessment. John's percentages also maintained at 100% for both similar and dissimilar situations at the 8-week generalization assessment. For Karen, a 5% decrease was measured for similar situations and a 9% decrease was recorded for dissimilar situations from the 2-week to the 8-week generalization assessment. Steve's percentages also decreased at the 8-week generalization assessment by 8% for similar situations and 17% for dissimilar situations.


The training results indicated that the program not only increased the verbal problem-solving skills of all participants when they used cue cards, but their skills maintained when the cue cards were eliminated. These findings suggest that the participants had learned and were using the problem-solving strategy. The generalization assessments further suggest the effectiveness of training since verbal responding improved across novel problem situations. In all, the findings suggest that persons with mental retardation can be taught to use a problem-solving strategy to identify appropriate responses to problematical situations.

One contributing factor to the effectiveness of the training program was the use of a modified generation of alternatives training approach. This approach enabled participants to generate two responses per situation and to receive feedback from the trainer. Thus, the participants had increased opportunities to respond and were able to rehearse a strategy for each problem situation.

The modified generation of alternatives training approach can be used by rehabilitation professionals to help individuals succeed in supported employment. For example, an individual in an employment situation may have difficulty in a particular aspect of his or her job. The rehabilitation counselor could then teach the employee an appropriate method(s) of responding to the situation in order to avoid future difficulties. This training program could consist of an employee being presented with the problem(s) by the counselor and modifying the employee's response(s) accordingly. Therefore, the modified generation of alternatives training program can be used to solve existing problems in more specific circumstances (i.e., counseling an employee, handling a problem at a certain job site, or modifying an existing job preparation program).

In addition to the demonstrated acquisition of skills, several findings are worth noting during the generalization assessments. First, the participants' performance improved for situations that were similar to those used in training. Therefore, participants were responding to situations that were similar to training situations under different environmental conditions (i.e., observer versus trainer, no feedback versus feedback). Second, the participants' responding improved when they were exposed to situations that were dissimilar to those used in training. Therefore, the findings suggest that participants learned to generate new solutions to novel problem situations. The generation of new responses is important since the goal of problem-solving training is to teach individuals to generate solutions to situations that they have never encountered (Hale & Holt, 1986). Third, another important finding was the maintenance of skill performance. All participants maintained performance above baseline generalization assessment levels for up to 8 weeks following the training without cue generalization assessment.

Although there were a number of positive findings in this investigation, two caveats and areas for future research exist. First, since only verbal behavior was assessed, the extent to which modifying the participants' verbal behavior impacted their actual job performance or employability is not known. However, based on the findings of other researchers (i.e., Foxx & Faw, 1990; Foxx, Marchand-Martella, Martella, Braunling-McMorrow, & McMorrow, 1988; Foxx et al., 1989; Martella, Agran, & Marchand-Martella, 1992; Martella, Marchand-Martella, Agran, & Allen, in press), it is likely that changing verbal behavior with the modified generation of alternatives approach to problem solving had an impact on participants' actual performance. Although data were not obtained on the participants' execution of the problem solving strategy at work, given the strong and immediate effect language may have on behavior (Israel, 1978; Watson & Tharp, 1989), it seems likely that the training could have impacted the participants' work performance. Second, since the generalization assessments were conducted by a familiar individual, it is not known the extent to which the participants' performance was predictive of how they would respond in novel locations with unfamiliar individuals. Clearly, further research on the use of this strategy by persons with mental retardation across a variety of novel stimuli is needed.

In summary, the present investigation suggests that individuals with mental retardation can be taught a problem-solving strategy to help them respond to problem situations and improve skills related to job success. More importantly, the participants' use of the problem-solving strategy improved during generalization assessments to situations that were either similar or dissimilar to those used in training. In addition to programming for generalization, the training program was successful in teaching a skill that could be maintained up to 8 weeks after training ended. The ability to identify a problem situation and respond appropriately to it represents a critical variable in ensuring job success. To promote success, it is essential that employment preparation programs include such instruction.

Authors' Notes

We wish to thank Bear River Mental Health Services, Logan, Utah for allowing us to conduct this investigation at their facility. Additionally, we would like to thank Tami Pyfer and Michelle Meyerhoffer for their assistance in the program.

This project was supported by a National Institute for Disability Rehabilitation Research field-initiated grant. However, the opinions expressed herein do not necessarily reflect the position or policy of the U.S. Department of Education.

Requests for reprints, a list of the training and generalization situations, or the scoring matrix should be sent to Nancy E. Marchand-Martella, Ph.D., Department of Curriculum and Instruction, School of Education, University of Montana, Missoula, Montana 59812-1054


Agran, M., Martin, J. E., & Mithaug, D. E. (1987). Transitional assessments for students with mental retardation. Diagnostique, 12, 173-184. Barlow, D.H., & Hersen, M. (1984). Single case experimental designs: Strategies for studying behavior change (2nd ed.). New York: Pergamon Press. Foxx, R. M., & Faw, G. D. (1990). Problem-solving skills training for psychiatric inpatients: An analysis of generalization. Behavioral Residential Treatment, 5, 159-176. Foxx, R. M., Marchand-Martella, N. E., Martella, R. C., Braunling-McMorrow, D., & McMorrow, M. J. (1988). Teaching a problem-solving strategy to closed head-injured adults. Behavioral Residential Treatment, 3, 193-210. Foxx, R. M., Martella, R. C., & Marchand-Martella, N. E. (1989). The acquisition, maintenance, and generalization of problem-solving skills by closed head-injured adults. Behavior Therapy, 20, 61-76. Hale, A., & Holt, A. S. (1986). Behavioural science. In J. Ridley (Ed.), Safety at Work (2nd ed. pp. 229-318). London: Butterworths. Hall, R. V. (1983). Behavior modification: The measurement of behavior (rev. ed.), Austin, TX: Pro-Ed. Israel, A. C. (1978). Some thoughts on correspondence between saying and doing. Journal of Applied Behavior Analysis, 11, 271-276. Martella, R. C., Agran, M., & Marchand-Martella, N. E. (1992). Problem solving to prevent accidents in supported employment. Journal of Applied Behavior Analysis, 25, 637-645. Martella, R. C., Marchand-Martella, N. E., Agran, M., & Allen, S. (in press). Using a problem-solving strategy to prevent work-related accidents due to unsafe worker behavior. British Columbia Journal of Special Education. Martin, J. E., Mithaug, D. E., Agran, M., & Husch, J. V. (1990). Consumer-centered transition and supported employment. In J. L. Matson (Ed.), Handbook of behavior modification with the mentally retarded (2nd ed.) (pp. 357-389). New York: Plenum Press. Mithaug, D. E., Martin, J. E., & Agran, M. (1987). Adaptability instruction: The goal of transitional programming. Exceptional Children, 53, 500-505. Rusch, F. R., Schutz, R. P., & Agran, M. (1982). Validating entry-level survival skills for service occupations: implications for curriculum development. The Journal of the Association of the Severely Handicapped, 8, 32-41. Salzberg, C. L., Agran, M., & Lignugaris/Kraft, B. (1986). Behaviors that contribute to entry-level employment: A profile of five jobs. Applied Research in Mental Retardation, 7, 299-314. Salzberg, C. L., Lignugaris/Kraft, B., & McCuller, G. L. (1988). Reasons for job loss: A review of employment termination studies of mentally retarded workers. Research in Developmental Disabilities, 9, 153-170. Watson, D. L., & Tharp, R. G. (1989). Self-directed behavior: Self modification for personnel adjustment. Pacific Grove, CA: Brooks/Cole. Received: July 1991 Revision: December 1991 Acceptance: January 1992
COPYRIGHT 1993 National Rehabilitation Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Agran, Martin
Publication:The Journal of Rehabilitation
Date:Jul 1, 1993
Previous Article:Computer training for the young adult patient with chronic mental illness.
Next Article:Head Injury: A Family Matter.

Related Articles
Programming for vocational competence in sheltered workshops.
Emerging rehabilitation needs of adults with developmental disabilities.
Rehabilitation services for people with mental retardation and psychiatric disabilities: dilemmas and solutions for public policy.
Current trends in mental health care for persons with mental retardation.
Identifying psychiatric disorders in persons with mental retardation: a model illustrated by depression in Down syndrome.
Impact of person-centered later life planning training program for older adults with mental retardation.
Equal access to employment opportunities for people with mental retardation: an obligation of society.
Career development for adolescents and young adults with mental retardation.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters