An Analysis of Organizational Behavioral Interventions in Terms of the Three-contingency Model of Performance Management
by Jason T. Otto, Greene Valley Developmental Center		and	Richard W. Malott, Western Michigan University DickMalott.com

The three-contingency model of performance management applies to people who have language skills (see Malott, 1993; 1998). The model suggests that (1) people often fail to act in their best interest because of ineffective natural contingencies, as much as because of effective competing contingencies, (2) effective performance-management contingencies often involve behavior-contingent outcomes too delayed to reinforce or punish that behavior, and (3) the underlying behavioral processes involve inferred direct-acting reinforcement or punishment contingencies.

This article reports the application of the three-contingency model to an analysis of OBM studies published in The Journal of Applied Behavior Analysis (JABA), volumes 21 through 30. These studies involved performance-management (PM) contingencies with outcomes that were too delayed to reinforce or punish the managed behavior.

In-depth Analysis

Johnson, Welsh, Miller, and Altus (1991) added a PM contingency to increase the amount of household tasks completed in a student-housing cooperative. Performance management was needed because the natural avoidance contingency in Figure 1 had an outcome that was too small to support the student’s behavior. For example, failing to start cleaning a kitchen within one hour is not significant; the outcome is just one more hour of a messy kitchen. However, the cumulative outcome of failing to clean the kitchen one hour, and then another hour, and another, etc. can be a significantly less sanitary, less functional kitchen that affects the quality of life of all involved.

Figure 1. Johnson, et al., (1991): Ineffective natural contingency.

The second contingency is the effective, indirect-acting PM contingency, managed by the house managers. The before condition is I will lose the opportunity for a rent reduction at the end of the week. Completing the task results in the after condition, I will not lose the opportunity for a rent reduction. This outcome is both sizable and probable, in contrast to ineffective contingencies that have outcomes that are too small or too improbable to control behavior. The house managers used the sizeable and probable cost of rent in an analog to the avoidance of the loss of a reinforcer. For example, if reading and studying the housing-cooperative manual were not done by the weekly deadline, the opportunity for a rent reduction would be lost ( Altus, Welsh, Miller, and Merrill, 1993; see Figure 2). Contrary to the common analysis, the three-contingency model suggests that PM contingencies that increase or maintain behavior must be analogs to avoidance and must have a deadline because, without a deadline, infinite procrastination is likely.

Figure 2. Altus, et al., (1993): Indirect-acting performance-management contingency.

The rent reduction occurs at the end of the month, a long delay after cleaning or studying. Therefore, the delayed rent reduction cannot affect task completion in the same way as a direct-acting contingency with its immediate, reinforcing outcome. The PM contingency is an analog to an avoidance contingency because the contingency has much the same effect as a direct-acting avoidance contingency with an immediate, reinforcing outcome, but the underlying behavioral mechanism is not the same.

The third contingency of this model is the inferred, direct-acting contingency. When the PM contingency is indirect-acting, the underlying behavioral mechanism must still involve some sort of direct-acting contingency immediately reinforcing the behavior. For example, the before condition might be I have aversive thoughts of losing the rent reduction. And beginning and completing tasks immediately reduces aversive thoughts of losing the rent reduction, a direct-acting escape contingency (Figure 3).

Figure 3. Johnson, et al. (1991): Inferred, direct-acting contingency.

Once-Over-Lightly Analyses

The following is a less detailed review of the other JABA OBM studies, with a brief discussion of how the three-contingency model might clarify the underlying behavioral mechanisms. Several PM studies involved money in what the model suggests were analogs to avoidance-of-loss contingencies to increase and maintain household tasks (Altus, Welsh, & Miller, 1991; Altus, et al., 1993; Johnson, et al., 1991), increase membership recruitment (Herndon, & Mikulas, 1996), increase sales (George, & Hopkins, 1989), and increase seatbelt buckling at work (Rogers, Rogers, Bailey, Runkle, & Moore, 1988). Several studies used performance feedback in analogs-to-avoidance contingencies to increase desired bank teller tasks (Crowell, Anderson, Abel, & Sergio, 1988), increase teaching and token reinforcement skills (Harchik, Sherman, Sheldon, & Strouse, 1992), increase and maintain the teacher’s accuracy of behavioral teaching tasks (Ingham & Greer, 1992; Witt, Noell, LaFleur, & Mortenson, 1996), increase latex glove use (DeVries, Burnette, & Redmon, 1991; Babcock, Sulzer-Azaroff, Sanderson, & Scibak, 1992), and increase both the feedback from head nurses to assistants and prompted voids by assistants (Burgio, Engel, Hawkins, McCormick, Sheve, & Jones, 1990).

In all cases, the employees seem to have been told rules describing the delayed-outcome contingencies, though not described as avoidance contingencies. In all cases, there were either explicit or implicit deadlines before which the managed behavior had to occur for the avoidance of the loss of the opportunity to receive a financial reinforcer or positive feedback or avoidance of the presentation of corrective feedback. In all cases, the natural contingencies that might have supported the desired performance appear to have been ineffective because the outcomes of each individual response were either too improbable, too small, or both and were therefore ineffective.

To decrease theft, one study used what the three-contingency model suggests is an analog to punishment (Carter, Holmstrom, Simpanen, & Melin, 1988). In such cases of analogs to punishment and penalty, the model also suggests the underlying behavioral mechanisms are based on inferred, direct-acting punishment contingencies involving immediate thoughts of sizable and probable, though delayed, aversive outcomes.

For a more detailed analysis of the application of the three-contingency model to OBM please see Malott (1993), Malott, Malott, & Shimamune (1993), and Malott, Shimamune, & Malott (1993).

References

Altus, D. E, Welsh, T. M., & Miller, L. K. (1991). A technology for program maintenance: Programming key researcher behaviors in a student housing cooperative. Journal of Applied Behavior Analysis, 24, 667-675.

Altus, D. E., Welsh, T. M., Miller, L. K., & Merrill, M. H. (1993). Efficacy and maintenance of an education program for a consumer cooperative. Journal of Applied Behavior Analysis, 26, 403-404.

Babcock, R. A., Sulzer-Azaroff, B., Sanderson, M., & Scibak, J. (1992). Increasing nurses’ use of feedback to promote infection-control practices in a head-injury treatment center. Journal of Applied Behavior Analysis, 25, 621-627.

Burgio, L. D., Engel, B. T., Hawkins, A., McCormick, K., Scheve, A., & Jones, L. T. (1990). A staff management system for maintaining improvements in continence with elderly nursing home residents. Journal of Applied Behavior Analysis, 23, 111-118.

Carter, N. Holmstrom, A., Simpanen, M., & Melin, L. (1988). Theft reduction in a grocery store through product identification and graphing of losses for employees. Journal of Applied Behavior Analysis, 21, 385-389.

Crowell, C. R., Anderson, D. C., Abel, D. M., & Sergio, J. P. (1988). Task clarification, performance feedback, and social praise: Procedures for improving the customer service of bank tellers. Journal of Applied Behavior Analysis, 21, 65-71.

DeVries, J. E., Burnette, M. M., & Redmon, W. K. (1991). AIDS prevention: Improving nurses’ compliance with glove wearing through performance feedback. Journal of Applied Behavior Analysis, 24, 705-711.

George, J. T., & Hopkins, B. L. (1989). Multiple effects of performance-contingent pay for wait persons. Journal of Applied Behavior Analysis, 22 , 131-141.

Harchik, A. E., Sherman, J. A., Sheldon, J. B., & Strouse, M. C. (1992). Ongoing consultations as a method of improving performance of staff members in a group home. Journal of Applied Behavior Analysis, 25, 599-610.

Herndon, E. J. & Mikulas (1996). Using reinforcement-based methods to enhance membership recruitment in a volunteer organization. Journal of Applied Behavior Analysis, 29, 577-580.

Ingham, P. & Greer, D. (1992). Changes in student and teacher responses in observed and generalized settings as a function of supervisor observations. Journal of Applied Behavior Analysis, 25, 153-164.

Johnson, S. P., Welsh, T. M., Miller, L. K., & Altus, D. E. (1991). Participatory management: Maintaining staff performance in a university housing cooperative. Journal of Applied Behavior Analysis, 24, 119-127.

Malott, R. W. (1993). A theory of rule-governed behavior and organizational behavior management. Journal of Organizational Behavior Management, 12, 45-65.

Malott, R. W. (1993, October). The Three-Contingency Model of Performance management and Support in Higher Education. Educational Technology, 21-28.

Malott, R. W., Malott, M. E., & Shimamune, S. (1993). Comments on rule-governed behavior. Journal of Organizational Behavior Management, 12, 91-101.

Malott, R. W., Shimamune, S., & Malott, M. E. (1993). Rule-governed behavior and organizational behavior management: An analysis of interventions. Journal of Organizational Behavior Management, 12, 103-116.

Malott, R. W. (1998). Performance management and welfare reform: The three-contingency model of performance management applied to welfare reform. Behavior and Social Issues, 8, 109-140.

Rogers, R. W., Rogers, J. S., Bailey, J. S., Runkle, W., & Moore, B. (1988). Promoting safety belt use among state employees: The effects of prompting and a stimulus-control intervention. Journal of Applied Behavior Analysis, 21, 263-270.