Reflection, paper

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=worg20

Journal of Organizational Behavior Management

ISSN: 0160-8061 (Print) 1540-8604 (Online) Journal homepage: https://www.tandfonline.com/loi/worg20

Changing Less-Preferred Duties to More-Preferred: A Potential Strategy for Improving Supervisor Work Enjoyment

Carolyn W. Green , Dennis H. Reid , Susan Passante & Vicki Canipe

To cite this article: Carolyn W. Green , Dennis H. Reid , Susan Passante & Vicki Canipe (2008) Changing Less-Preferred Duties to More-Preferred: A Potential Strategy for Improving Supervisor Work Enjoyment, Journal of Organizational Behavior Management, 28:2, 90-109, DOI: 10.1080/01608060802100899

To link to this article: https://doi.org/10.1080/01608060802100899

Published online: 08 Sep 2008.

Submit your article to this journal

Article views: 790

Citing articles: 8 View citing articles

Journal of Organizational Behavior Management, Vol. 28(2) 2008 Available online at http://jobm.haworthpress.com

© 2008 by The Haworth Press. All rights reserved. 90 doi:10.1080/01608060802100899

WORG0160-80611540-8604Journal of Organizational Behavior Management, Vol. 28, No. 2, May 2008: pp. 1–27Journal of Organizational Behavior Management

RESEARCH ARTICLE

Changing Less-Preferred Duties to More-Preferred: A Potential Strategy

for Improving Supervisor Work Enjoyment Research ArticleJOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT Carolyn W. Green

Dennis H. Reid Susan Passante Vicki Canipe

ABSTRACT. We evaluated a strategy for making highly nonpreferred work duties more preferred as a potential means of enhancing work enjoy- ment among supervisors in a human service setting. Repeated preference ratings and rankings were completed by 4 supervisors during baseline to identify their most disliked work tasks. These tasks were then altered by

Carolyn W. Green is affiliated with the J. Iverson Riddle Center, Morganton, North Carolina.

Dennis H. Reid is affiliated with the Carolina Behavior Analysis and Support Center.

Susan Passante and Vicki Canipe are affiliated with J. Iverson Riddle Center, Morganton, North Carolina.

Address correspondence to: Dennis H. Reid, Ph.D. P.O. Box 425, Morganton, NC 28680 (E-mail: [email protected]).

Research Article 91

soliciting each supervisor’s opinion regarding why the respective task was so disliked and what could be done to make the task more preferable, removing disliked stimuli associated with the tasks, and adding liked stim- uli during completion of the tasks. Following the task alterations, ratings of task likeability increased for all 4 supervisors. Preference rankings of target tasks also increased, including from rankings as among the least preferred tasks to among the most preferred tasks. All supervisors rated the task- alteration process as making their quality of work life better, and all chose to continue the intervention. These results occurred without any observed decrease in work quality. Results are discussed regarding the potential contribution of behavioral interventions with subjective concepts, such as enjoyment, that are viewed as important by the general populace. However, qualifications with relying on verbal reports also are noted.

KEYWORDS. Preference assessments, supervisors, work enjoyment

INTRODUCTION

A relatively small but persistent focus of behavioral research is the work performance of human service staff. Since the first reports of behavioral applications with staff behavior in human service settings in the 1970s, a considerable number of investigations have demonstrated means of pro- moting desired performance (for reviews, see Phillips, 1998; Reid, O’Kane, & Macurik, in press). Such research has centered on two impor- tant and related areas. One area has been training human services staff with relevant work skills (Jahr, 1998). The second area has been managing or improving day-to-day work performance, generally involving situations in which staff have the necessary work skills, but routine performance does not reflect acceptable application of those skills (e.g., Harchik, Sherman, Sheldon, & Strouse, 1992; Parsons, Rollyson, & Reid, 2004).

Another important area involving human service staff that has not been emphasized in behavioral research is quality of work life, such as the enjoyment staff experience with their work. Areas potentially related to work enjoyment have been addressed somewhat, such as staff acceptance of behavioral supervisory interventions (Parsons, 1998). However, work enjoyment has not been targeted as a dependent variable per se for behav- ioral intervention.

Investigating means of enhancing staff work enjoyment would appear to represent a useful undertaking for behavioral research. The significance of

92 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

targeting subjective concepts that typically are perceived as very important by society at large (e.g., happiness, satisfaction) has long been recognized in applied behavior analysis, though relatively understudied (Wolf, 1978). Work enjoyment appears to be one such concept. Reference to the degree of enjoyment associated with work is commonly made when staff describe their work situations, reflecting perceived importance of work enjoyment among workers (Martin, 2005). Supervisors, likewise, frequently report that it is important to help staff enjoy their work, though the difficulties in enhancing work enjoyment are also noted (Parsons, Reid, & Crow, 2003). In light of the pervasive importance of work enjoyment among the work- ing populace, the lack of investigatory attention on this topic has resulted in criticism regarding a limited focus of behavioral research in work set- tings and is considered a significant impediment to the relevance and future of organizational behavioral research (Martin, 2005).

One means of potentially enhancing aspects of staff work enjoyment may be through assessment and alteration of staff preferences for certain work duties. If respective work tasks were identified as strongly nonpre- ferred by staff and then alterations could be made to make the tasks more preferred, then one component of work enjoyment may be enhanced (i.e., a less enjoyable task could be made more enjoyable). Such an approach has been used to improve features of quality of life among people with develop- mental disabilities, such as by improving the preferred nature of a nonpre- ferred task that occasions problem behavior in order to reduce the behavior associated with the task (Foster-Johnson, Ferro, & Dunlap, 1994). Although attention has been directed to identifying reinforcers for increasing work behaviors among staff (e.g., Green & Reid, 1991; Wilder, Therrien, & Wine, 2005), researchers have not assessed staff work preferences with the intent of increasing the preferred nature of tasks that are nonpreferred.

The purpose of this investigation was to evaluate a means of making highly nonpreferred work tasks more preferred among staff supervisors in a human service setting as a means of enhancing the supervisors’ work enjoyment. Supervisors were targeted because they fulfill a critical role in human service agencies, in that they are in large part responsible for the proficiency of staff work performance and can exert significant influence on how well staff work (Ivancic & Helsel, 1998; Sturmey, 1998). However, supervisors also frequently report a lack of enjoyment and dissatisfaction with their jobs (Hewitt et al., 2004; Sturmey, 1998). The focus of the investigation was on the supervisors’ verbal reports related to work enjoyment. What staff say about their work is recognized as an important variable affecting the significance of

Research Article 93

behavioral investigations in work settings (Olson, Laraway, & Austin, 2001). However, verbal reports also represent a risky measure for behav- ioral research (for a cogent discussion, see Wolf, 1978). Consequently, several different types of verbal reports were obtained (e.g., preference ratings and rankings, ratings of quality of work life) in an attempt to increase the likelihood of valid measures of work enjoyment using a congruent validity approach (cf. Cushing, Horner, & Barrier, 2003).

METHOD

Social Validity Survey

As just indicated, a rationale for intervening to change a highly nonpre- ferred work task to a more preferred task among supervisors was to attempt to improve an aspect of supervisory quality of work life in terms of work enjoyment. To initially assess the social validity of this means of attempting to improve work enjoyment (Schwartz & Baer, 1991), a sur- vey was administered to 29 working supervisors attending a training workshop on behavioral intervention procedures. The supervisors repre- sented 12 community agencies and 1 state institution providing services for people with developmental disabilities (none of the survey respon- dents worked in the agency in which the intervention in this study was conducted). Twenty-one of the supervisors supervised direct support staff in their agencies, and eight supervised supervisors of direct support staff. The survey question of relevance for purposes here asked supervisors to rank on a 7-point Likert scale how much their quality of work life would be better or worse if their supervisors found a way to make their least preferred work task more enjoyable to perform. Survey options ranged from 1 (extremely better) to 7 (extremely worse), with a neutral value of 4 (neither better nor worse). Results indicated all supervisors reported that such an action would make their quality of work life better (59% reported extremely better, 38% reported a lot better, and 4% reported a little bet- ter). These results appear to provide some support from individuals employed in supervisory positions that the process of making a least-pre- ferred work task more preferred would improve their quality of work life.

Settings and Participants

The setting was three residences for people with severe multiple disabilities operated by a large multiservice agency. Between 8 and 10

94 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

individuals lived within each residence. There were always at least three direct support staff working within each residence each day, with a day and evening shift supervisor of the direct support staff also present. The individuals with severe multiple disabilities received, among other supports, educational services from the agency’s adult education department. Licensed special education teachers developed skill-acquisition teaching programs that the direct support personnel implemented with consultative assistance from the teachers.

Participants were 4 supervisors, each of whom had a high school education. Three of the participants were front-line supervisors, meaning they directly supervised direct support staff in respective residences. Ms. Tome had 4 years of experience as a supervisor; Ms. Jones had 6 years of supervi- sory experience; and Mr. Davis had 3 years. The other supervisor, Ms. Noel, had a mid-level supervisory position, meaning she supervised other supervi- sors of direct support staff. Ms. Noel had 9 years of supervisory experience. These supervisors were selected for the investigation because they had super- visory responsibilities, either directly or indirectly through other supervisors, over staff working with individuals for whom the agency’s adult education department was currently responsible for their education services.

Dependent Measures

There were two primary dependent measures. The first measure was each supervisor’s preference ratings for specific job tasks. Each supervi- sor’s primary job duties were delineated into specific tasks. The delineation was based on the written job description for each supervisor as well as discussion with agency executives. For the 3 supervisors of direct support staff, 12 job tasks were specified (Table 1). For the midlevel supervisor, 8 tasks were delineated (also presented in Table 1). A 7-point Likert rating scale was developed with which each assigned job task could be rated, with 7 indicating that a task was liked as much as possible and 1 representing a task was liked the least possible. (4 on the scale represented the neutral value of “neither liked nor disliked.”) To initiate a rating by a supervisor, a representative of the agency’s education depart- ment (experimenter) met individually with the supervisor and explained that the purpose was to obtain information regarding preferences for existing job duties. The experimenter explained how the Likert scale was to be completed and then left the rating form with the supervisor with a request to complete the form and return it to the education department office. The experimenter also informed the supervisor that there would be

Research Article 95

continued requests to complete the same form in the future. For subse- quent ratings, the experimenter simply met with the supervisor, provided another rating form, and asked the supervisor to complete the form and again provide it to the education department office.

The second primary dependent measure was a partial preference ranking of the delineated job tasks. The focus of the ranking was on the most- and least-preferred job tasks of each supervisor. A supervisor was provided with a list of all delineated job tasks and asked to rank the most-preferred and second most–preferred job tasks as well as the least- preferred and second least–preferred job tasks. The ranking was presented to each supervisor in the same manner as the preference rating, and immediately following the preference rating. Presentation of the rankings and ratings to each supervisor was distributed across weeks (see Baseline), during arbitrarily selected work days within the respective week. Each presentation was provided at a time when the supervisor was not involved with a staff meeting or interaction with an individual staff member.

TABLE 1. Examples of supervisor job tasks

Direct support staff supervisor

Observe staff teaching programs Observe staff work routines other than teaching Record (new) staff completion of orientation training assignments Plan consumer trips Complete vocational time sheet recordings Attend designated meetings Transport consumers to off-campus appointments Complete monthly staff evaluation packets Complete monthly progress notes Complete daily checklists regarding staff work assignments Prepare consumer activity schedules and assign groups of consumers Complete other duties as assigned

Mid-level supervisor

Complete staff work schedules and assign staff Review and sign monthly time sheets of staff Assign staff for classroom duty Check program books, environmental checks, etc., for accuracy and timeliness Meet monthly with probationary staff for performance review Attend consumer team meetings Conduct shift-change meetings with staff Observe quality of staff teaching performances with consumers

96 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

Because the usefulness of increasing supervisory preference for performing a highly nonpreferred work task may be questionable if a simultaneous decrease in the quality of performing that task occurred, a secondary dependent measure involved assessments of the quality of the supervisors’ target task completion. Toward the end of the baseline condi- tion (see Experimental Conditions), the respective work task ranked most frequently as the least preferred for each supervisor was selected for the intervention. An evaluation of the supervisor’s routine completion of that task was then undertaken prior to initiating the intervention (and subse- quently after the intervention had been in effect). The same job task was ranked during baseline as the least preferred for 2 supervisors (Ms. Tome and Ms. Jones)—that of writing monthly progress notes regarding consumer performance on teaching and other behavior change programs conducted by direct support staff whom the supervisors supervised. The task ranked least preferred for Ms. Noel was reviewing monthly time sheets to assess staff accuracy in completing the time sheets. The least preferred task for Mr. Davis was conducting monthly evaluative observa- tions of staff carrying out teaching programs with consumers.

To evaluate the quality of monthly progress reviews for Ms. Tome and Ms. Jones, a 5-item checklist was developed that defined desired compo- nents of progress note reviews based on agency policy (e.g., were all notes reviewed by the 10th of each month, were increases or decreases in program data accompanied by an explanatory narrative). Subsequently, a sample of progress note reviews was assessed with the checklist during baseline (11 notes for each supervisor) and while the intervention was in effect (11 notes for each supervisor). The dependent measure was percent of desired components completed in accordance with the criteria on the checklist.

For Ms. Noel, the quality of her reviews of staff completion of monthly time sheets involved asking her supervisor to assess the quality of her reviews once during baseline and once after the intervention had been in effect. Ms. Noel’s supervisor completed a 5-point Likert scale consisting of 3 questions pertaining to the timeliness of Ms. Noel’s reviews, whether she ensured each time sheet was signed by the respective staff person, and whether Ms. Noel ensured each time sheet was completed accurately. The rating points on the Likert scale ranged from outstanding performance to unsatisfactory performance.

For Mr. Davis, a sample of his completed program observations was assessed to ensure all required information was recorded in accordance with agency policy. Five components of his program observations were

Research Article 97

assessed (e.g., were observations completed for all assigned staff activities? was all information provided on the observation forms as dictated by agency policy? were all forms dated appropriately?). Three program observation forms were assessed during the last month of the baseline condition (representing 38% of the total monthly observations conducted) and 4 during a month within the intervention period (representing 50% of the total monthly observations). The dependent measure was the percent- age of components expected to be completed that were completed according to agency criteria.

Interobserver agreement checks were conducted by a second observer independently reviewing a sample of the monthly progress notes assessed for Ms. Tome and Ms. Jones, and a sample of the observation forms for Mr. Davis. Interobserver agreement checks were conducted for 43% of the assessed progress notes each for Ms. Tome and Ms. Jones, and 43% of the assessed observation forms for Mr. Davis, including during both experimental conditions for each supervisor. Interobserver agreement was assessed on an item-by-item basis, with an agreement scored only if both observers recorded that the item was completed correctly or incorrectly, respectively. Interobserver agreement was 100% for all interobserver agreement checks.

Experimental Conditions

Baseline. During baseline, supervisors carried out work duties in accordance with their existing responsibilities (see Table 1), with the addition of being asked to complete the preference ratings and rankings. At least 3 ratings and rankings were requested of the supervisors during baseline (each supervisor completed each rating and ranking that was requested). In order to obtain a representative sample of preferences over time with existing work duties, the ratings and rankings were distributed across several weeks. Specifically, for Ms. Tome, Ms. Jones, Ms. Noel, and Mr. Davis, the ratings and rankings were requested and completed across 5, 6, 3, and 4 weeks, respectively.

Task Enjoyment Motivation Protocol. The intervention designed specifi- cally for attempting to increase preferences for the nonpreferred work tasks consisted of application of the Task Enjoyment Motivation Protocol (TEMP). The protocol involved two components. The first component was a participative management process (Reid & Parsons, 2006, chap. 6) that was implemented for two purposes. The first purpose was to obtain information from a participating supervisor to help determine what aspects of a job task

98 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

the supervisor did not prefer and what might be done to make the task more preferred. The second purpose was to take advantage of the acceptability factor often associated with participative management strategies. That is, management strategies that incorporate staff opinion in decisions that affect how work duties should be performed tend to be more acceptable to staff than strategies that do not include their input (Phillips, 1998).

Hence, a participative element was included in the overall intervention in an attempt to enhance the nature with which the highly nonpreferred job tasks of respective supervisors were altered. A meeting was con- ducted with the supervisor by the director of the agency’s adult education department and a teacher who consulted with the supervisor and the supervisor’s staff. During the meeting, results of the preference ratings and rankings were reviewed with the supervisor with a focus on the most nonpreferred task. The supervisor’s opinion was then solicited in terms of why the task was so nonpreferred and what might be done to make per- forming the task more preferable.

The second component of the TEMP intervention involved attempting to alter the performance of the target job task in two ways. The first way involved removing stimuli associated with the task that were reported by each supervisor to be disliked. Regarding completion of monthly progress notes for Ms. Tome and Ms. Jones and reviewing time sheets for Ms. Noel, all reported that the primary aspect that made the task highly nonpreferred was frequent interruptions when they tried to spend time reviewing the notes or time sheets, such that completing the task each month required much more time than it should. Interruptions were removed by scheduling a specific time to complete the monthly reviews in an office away from the supervisor’s usual work station. Interruptions were avoided in this manner because the supervisor’s staff (who were responsible for the interruptions) did not have access to the office arranged for the supervisors (the supervisors’ supervisors agreed that it would not be problematic for the former individuals to be away from their usual work station for several hours each month to complete progress notes or time sheets). The office selected for the supervisors was adjacent to an experimenter’s office, which allowed inconspicuous observations regarding whether the supervisors actually used the office. On each opportunity, each supervisor used the newly arranged office.

Removing nonpreferred stimuli from the target work task for the fourth supervisor, Mr. Davis, involved different actions than with the other 3 supervisors due to the different nature of his highly nonpreferred task of completing monthly observations of staff teaching duties.

Research Article 99

Mr. Davis reported that conducting the observations was nonpreferred because it appeared that the staff disliked having their work perfor- mance observed. Hence, an attempt was made to make the observation process more desirable for the staff by conducting a performance lottery based on results of the observation. Performance lotteries involve allow- ing staff whose performance meets a designated criterion to become eligible for a lottery with the chance to win desired items or privileges (Iwata, Bailey, Brown, Foshee, & Alpern, 1976; Shoemaker & Reid, 1980).

Each month, those staff whose performance met a previously existing criterion based on Mr. Davis’s observations had their names placed in the lottery. Names of staff were then randomly drawn. As each name was drawn, the winning staff person was provided a choice of winnings that included gift certificates from local stores, re-arrangement of certain work duties, and alterations of aspects of work schedules. In this manner, each staff person meeting the criterion won something. However, those staff whose names were drawn early in the process had a wider selection of winnings (and presumably, more desired winnings) relative to the staff whose names were drawn later in the process. An experimenter made periodic probe observations during the times the lottery was scheduled to occur with staff, which indicated that the supervisor conducted the lottery as intended.

The second way the TEMP intervention attempted to alter a supervi- sor’s completion of a highly nonpreferred task was by adding reportedly liked stimuli to the act of completing the task. Based on questions posed during the previous meetings with the first 3 supervisors, it was deter- mined that their tasks of completing monthly progress reviews or monthly time sheets might be made more preferred by providing a small snack or a bottle of drinking water for their consumption as they worked on the task in the designated office. An experimenter provided the snacks and water to each supervisor prior to the time her review of progress notes or time sheets was scheduled to occur. The fourth super- visor, Mr. Davis, had reported that his task of conducting observations of staff performance would be more preferred if the staff viewed the activity in a more preferable light. The performance lottery just described was considered to be the means for making the observations more pleasant for the staff.

In summary, the TEMP intervention involved first meeting with staff to obtain their opinion regarding what made a work task so highly nonpre- ferred and then what could be done to make it more preferred. Next,

100 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

arrangements were made to attempt to remove reportedly disliked stimuli associated with completing the respective tasks and adding stimuli that reportedly were liked, with the condition that such arrangements would not be likely to interfere with completion of the tasks. Preference ratings and rankings continued as in baseline, disbursed over time periods of 8, 7, 8, and 9 weeks across the 4 supervisors, respectively.

Acceptability Measures

To assess participant acceptance of the TEMP intervention—and as a measure of the social validity of the results obtained (Schwartz & Baer, 1991)—a survey was administered individually to each of the 4 supervisor participants after the TEMP program had been in effect for at least 3 weeks. There were 2 parts to the survey. The first part asked the partici- pant to rate on a 7-point Likert scale how much the specific intervention (e.g., for Mr. Davis, the program observations followed by a staff lottery) made the participant’s quality of work life better or worse. The scale ranged from “extremely better” (7) to “extremely worse” (1), with 4 rep- resenting the neutral value (“neither better nor worse”). Because some data suggest that rating forms may represent less valid measures of accept- ability of management interventions with staff relative to staff actually choosing to participate or not participate in a respective intervention (Reid & Parsons, 1995; van den Pol, Reid, & Fuqua, 1983), the second part of the acceptability survey asked the participant if he or she wanted the TEMP intervention to continue or not. Each participant’s response was subsequently honored in terms of whether the intervention continued with that participant.

Due to the nature of the intervention with 1 participant, Mr. Davis, an additional acceptability measure was conducted regarding his work. Because the intent was to make his observations more acceptable to staff through the performance lottery and thereby potentially make conducting the observations more enjoyable for Mr. Davis by reducing staff discon- tent with the observations, staff members were asked to give their opinions about the lottery. To assess whether the lottery made the obser- vations more acceptable to staff, 5 of Mr. Davis’s staff were arbitrarily selected to complete a 7-point Likert survey. The survey asked the staff person to rate how much the performance lottery impacted her quality of work life, with the same response options as just noted with the survey of the supervisors. The survey also asked the staff person if he or she wanted the performance lottery to continue.

Research Article 101

Experimental Design

The experimental design was a multiple probe across supervisors (and target tasks). Because not all required features of a concurrent, multiple probe design were met, however (i.e., baselines across each supervisor did not overlap time-wise), an additional control feature was incorporated. Specifically, for each supervisor, preference ratings of the target tasks were compared to the mean rating for all other work tasks during baseline and the TEMP intervention. In this manner, ratings for the other work tasks (for which no intervention occurred) served as controls for ratings of the respective target tasks.

RESULTS

Preference Ratings

Each supervisor’s preference ratings for the targeted, highly nonpre- ferred work task increased relative to baseline when the TEMP program was in effect, though the amount of increase varied across supervisors (solid line and data points on Figure 1). The least amount of increase was with Ms. Tome, whose ratings for the work task of reviewing monthly progress notes increased slightly, from a baseline average rating of 2.7 (range, 2 to 3) to a consistent rating of 4 during the TEMP program. Larger increases in preference ratings occurred for Ms. Jones, Ms. Noel, and Mr. Davis. Ms. Jones’ ratings for completing monthly progress notes increased from a baseline level of 1 to an average of 4.7 (range, 2 to 6) during the TEMP program. Ms. Noel’s ratings for reviewing monthly time sheets increased from a baseline average of 2.0 (range, 1 to 4) to a consistent level of 6 during the TEMP program. For Mr. Davis, his ratings for the task of conducting program observations were consistently the lowest possible (1) during baseline and the highest possible during the TEMP program (7). For the 2 supervisors for whom follow-up ratings were available, the increased ratings during the TEMP program main- tained (Ms. Tome) or increased further (Ms. Jones) across an 8-week period.

In contrast to the increases in preference ratings for the initially non- preferred work tasks targeted with the TEMP program, there were no consistent increases in ratings of the other (control) work tasks that were not targeted with the program (dashed line and open data points

102 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

on Figure 1). Average preference ratings for Ms. Tome, Ms. Jones, Ms. Noel, and Mr. Davis for all other work tasks during baseline were 3.9 (all average ratings of 3.9), 5.2 (range, 4.7 to 5.4), 5.0 (range, 4.9 to 5.0), and 4.6 (range, 4.1 to 5.2), respectively. During the TEMP program, ratings for all other work tasks were 3.9 (range, 3.9 to 3.9), 5.2 (range, 4.8 to 5.6), 5.1 (range, 5.0 to 5.1), and 4.5 (range, 4.4 to 4.6), respectively.

FIGURE 1. Preference ranking score for target tasks for each ranking for each supervisor and experimental condition.

Research Article 103

Preference Rankings

Results of the job-task rankings (i.e., rankings of the 2 most- and 2 least-preferred job tasks out of all job tasks) tended to coincide with the changes in preference ratings (Figure 2). A ranking of 3 as presented on Figure 2 indicates that the respective task was not ranked among the 2 most- or 2 least-preferred tasks (i.e., no strong preference). Rankings of

FIGURE 2. Preference rating score for target tasks and average rating score for control tasks for each rating for each supervisor and experimental condition.

104 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

5 and 4 represent a task being ranked as the mostpreferred and second most-preferred, respectively, and rankings of 1 and 2 represent a task being ranked as the least and second least preferred work task, respec- tively. For Ms. Tome, her target task was always ranked as the least- or second least-preferred during baseline, but never ranked as the least- or second least-preferred task during the TEMP program (though her target task was never ranked as the most or second most preferred task during the program). For both Ms. Jones and Ms. Noel, their target tasks were always ranked as the least or second least preferred task during baseline whereas during the TEMP program these tasks were always ranked as either the most or second most preferred tasks. For Mr. Davis, during baseline his target task was ranked as the least preferred on three rankings and as neither strongly nonpreferred nor preferred on one ranking, com- pared to a consistent ranking as the most preferred task during the TEMP program.

Work Quality Measures

There were no observed decreases in work quality during the TEMP program for any target task or supervisor. Work quality measures remained very high throughout baseline and the TEMP program for each work task and supervisor. Ms. Tome correctly completed 98% of the items on progress notes during baseline and 100% during the TEMP program and Ms. Jones correctly completed 100% of the progress note items during both experimental conditions. Ms. Noel’s supervisor rated her completion of monthly time sheet reviews as outstanding for each of the three performance areas related to the reviews during baseline and the TEMP program. For Mr. Davis, the assessments of his program observations indicated he correctly completed 80% of the assessed observation forms during baseline and 100% during the TEMP program.

Acceptability Measures

All 4 supervisor participants rated the respective TEMP intervention as impacting their quality of work life by making it extremely better (highest possible rating on the survey). All 4 participants also chose to have the TEMP program continue when given the choice of having it continued or not. Regarding the assessment of whether the lottery that constituted part of the TEMP program for Mr. Davis made observations of staff teaching performance more acceptable to the staff, all 5 staff to whom the survey was administered reported that the lottery made their quality of work life

Research Article 105

extremely better. All 5 staff also chose to have the lottery continued when provided with a choice of continuing it or not.

DISCUSSION

Results suggested that the TEMP intervention successfully changed strongly nonpreferred work tasks of supervisors to more preferred tasks, and these changes improved the supervisors’ reports of their quality of work life. Ratings of the likeability of the tasks increased for all 4 super- visors during the TEMP program, and improved from the disliked range to the liked range for 3 of the supervisors (for the fourth supervisor, the rating changed to the neutral value on the rating scale). Preference rankings likewise increased for all 4 supervisors. Rankings of the target tasks during baseline averaged among the least or second least preferred task for all 4 supervisors whereas during the TEMP program, the rankings were among the most- or second most-preferred task for 3 supervisors and neither strongly nonpreferred nor preferred for 1 supervisor. All 4 super- visors also rated their quality of life as being extremely improved by the intervention and all 4 chose to have the intervention continued. These changes occurred without any apparently detrimental effects on the qual- ity with which the supervisors completed their target work duty.

When considered in total, results just summarized suggest that the intervention improved an aspect of work enjoyment for each participat- ing supervisor (i.e., a disliked task became more liked, quality of work life improved). However, all of the measures were based on verbal reports (with somewhat of an exception with the choice measure) and as noted earlier, reliance on verbal behavior as a dependent measure can be a risky venture in behavioral research (Wolf, 1978). Hence, significant qualifications with the results are in order pending outcomes of future research.

A primary qualification pertains to the degree to which the verbal reports (preference ratings and rankings, ratings of quality of work life) related to the concept of work enjoyment and represent valid changes. Verbal reports of typical workers are considered an important variable in research with staff as noted earlier and particularly with subjective concepts such as work enjoyment (Martin, 2005). However, enjoyment is in essence a private event and can only be measured indirectly such as through verbal reports. Although the different types of verbal reports in this investigation tended to coincide with each other across participants

106 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

and experimental conditions, thereby reflecting congruent validity, the reports still represent indirect measures of enjoyment. Consequently, the degree to which work enjoyment actually improved as a function of the intervention cannot be concluded definitively.

Another primary qualification that is related to the first qualification pertains to the possible existence of demand characteristics and their influence on the results (cf. Schwartz & Baer, 1991). That is, the partici- pants’ verbal reports may have been affected by previous reinforcement for saying what they perceived others (e.g., experimenters in this investi- gation) expected or desired them to say. In this regard, concern has been expressed previously that human service workers tend to respond verbally in ways they believe their supervisors want them to respond (Kemp, Miltenberger, & Lumley, 1996). In the current investigation one method- ological feature may have reduced the likelihood of this occurrence. Specifically, the educators who conducted the procedures with the supervi- sors had no supervisory or organizational authority over the participating supervisors (i.e., control of contingencies affecting the supervisors’ work assignments, schedules, pay, etc.), and the supervisors’ supervisors were not involved in the intervention with the exception of rating the quality of 1 participating supervisor’s work. Whether this feature reduced the possi- bility of demand characteristics is speculative, however; the results should be viewed with considerable caution pending outcomes of future research.

One means of addressing concerns over the validity of the results as just summarized would be to include other types of measures relating to work enjoyment in addition to verbal reports of preferences. For example, as verbal reports of preferences for work tasks change over time, simulta- neous measures could focus on counts of complaints regarding specific job duties, percentage of participant time directed to various work duties, and the relative order in which different tasks are completed over time. Addi- tion of these types of measures could possibly enhance the degree of credi- bility that preference-related verbal reports validly reflect work enjoyment.

A third qualification with the results pertains to a limitation of the experimental design employed. As noted previously, baseline measures across participants did not overlap time-wise. Simultaneous time-based measures is an important feature of a multiple probe design—a feature that was not adhered to in this investigation. Although the simultaneous measures of the preference ratings of the control tasks minimizes this experimental control problem somewhat, the lack of simultaneous mea- sures renders qualifications with the degree of functional control that can be attributed to the intervention.

Research Article 107

It should also be emphasized that our focus was on increased quality of work life as relating to work enjoyment. Research on quality of work life in human service organizations generally tends to focus on variables that are considered to reflect changes in work-life quality and impact an agency’s overall effectiveness, such as turnover rates (Howard & Gould, 2000). The latter type of research is of course important and should continue. Whether changes in the likeability of certain work tasks as addressed in this investigation will impact the latter areas is unknown. Our assumption is that the types of changes reported here will not likely impact these areas significantly, unless perhaps a broader array of work duties are made more likeable for supervisors or staff. Future research seems necessary to address this issue. Nonetheless, given the importance of work enjoyment among workers in human service agencies as referred to earlier, it seems a case can be made from a social significance perspective (Wolf, 1978) for improving enjoyment at work in its own right—provided there are no detrimental effects on an agency’s effectiveness.

Several other qualifications also exist that relate more to specific mea- sures and results associated with this investigation’s methodology. One qualification pertains to the relatively small number of preference ratings and rankings across experimental conditions such that a somewhat restricted sample of behavior was obtained. A concern existed that requesting staff to complete the ratings and rankings repeatedly for extended time periods (and particularly during baseline) may have reduced participant attention directed to completing the ratings and rank- ings and affect their validity. The finding that the preference ratings of the control tasks completed by the supervisors during intervention with the target tasks did not change relative to baseline ratings should reduce some concerns over the effectiveness of the intervention in this regard. Another qualification pertains to the finding that each supervisor completed the target tasks at an apparently high level of quality during baseline (and such quality continued during the intervention). It is not clear what effect the TEMP intervention would have had on the preference ratings and rankings as well as the work quality measures if a lower quality of work was observed during baseline.

In summary, results suggest that by conducting systematic preference assessments to identify nonpreferred tasks, soliciting staff opinion regarding what makes a work task nonpreferred, and altering the task by removing disliked stimuli and adding liked stimuli, highly nonpreferred work tasks can be made more preferred. Results also provide some sup- port for the suggestion that making the changes with nonpreferred tasks

108 JOURNAL OF ORGANIZATIONAL BEHAVIOR MANAGEMENT

represented a means of enhancing worker enjoyment through improved quality of work life. The latter suggestion is more tentative than the former though, given the qualifications noted earlier. For this reason, the TEMP strategy is viewed as a potential means of enhancing quality of work life pending results of additional research. Such research could expand on this investigation by addressing other work tasks and groups of human service personnel, and perhaps other measures of quality of work life. Investigations along these lines could further explore and pos- sibly substantiate contributions of behavior analysis for improving work enjoyment and other aspects of quality of work life considered important by human service staff.

REFERENCES

Cushing, L. S., Horner, R. H., & Barrier, H. (2003). Validation and congruent validity of a direct observation tool to assess student social climate. Journal of Positive Behavior Interventions, 5, 225–237.

Foster-Johnson, L., Ferro, J., & Dunlap, G. (1994). Preferred curricular activities and reduced problem behaviors in students with intellectual disabilities. Journal of Applied Behavior Analysis, 27, 493–504.

Green, C. W., & Reid, D. H. (1991). Reinforcing staff performance in residential facilities: A survey of common managerial practices. Mental Retardation, 29, 195–200.

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

Hewitt, A. S., Larson, S. A., Lakin, K. C., Sauer, J., O’Nell, S., & Sedlezky, L. (2004). Role and essential competencies of the frontline supervisors of direct support profes- sionals in community services. Mental Retardation, 42, 122–135.

Howard, B., & Gould, K. E. (2000). Strategic planning for employee happiness: A busi- ness goal for human service organizations. American Journal on Mental Retardation, 105, 377–386.

Ivancic, M. T., & Helsel, W. J. (1998). Organizational behavior management in large residential organizations: Moving from institutional to client-centered care. Journal of Organizational Behavior Management, 18(2/3), 61–82.

Iwata, B. A., Bailey, J. S., Brown, K. M., Foshee, T. J., & Alpern, M. (1976). A perfor- mance-based lottery to improve residential care and training by institutional staff. Journal of Applied Behavior Analysis, 9, 417–431.

Jahr, E. (1998). Current issues in staff training. Research in Developmental Disabilities, 19, 73–87.

Kemp, F. D., Miltenberger, R. G., & Lumley, V. A. (1996). Treatment acceptability and ‘faking good’: Are staff telling us what they think we want to hear. Behavioral Interventions, 11, 181–191.

Research Article 109

Martin, A. J. (2005). Perplexity and passion: Further consideration of the role of positive psychology in the workplace. Journal of Organizational Behavior Management, 24(1/2), 203–205.

Olson, R., Laraway, S., & Austin, J. (2001). Unconditioned and conditioned establishing operations in organizational behavior management. Journal of Organizational Behavior Management, 21(2), 7–41.

Parsons, M. B. (1998). A review of procedural acceptability in organizational behavior management. Journal of Organizational Behavior Management, 18(2/3), 173–190.

Parsons, M. B., Reid, D. H., & Crow, R. E. (2003). Best and worst ways to motivate staff in community agencies: A brief survey of supervisors. Mental Retardation, 41, 96–102.

Parsons, M. B., Rollyson, J. H., & Reid, D. H. (2004). Improving day-treatment services for adults with severe disabilities: A norm-referenced application of outcome manage- ment. Journal of Applied Behavior Analysis, 37, 365–377.

Phillips, J. F. (1998). Applications and contributions of organizational behavior manage- ment in schools and day treatment settings. Journal of Organizational Behavior Management, 18(2/3), 103–129.

Reid, D. H., O’Kane, N. P., & Macurik, K. M. (in press). Staff training and management. In W. Fisher, C. Piazza, & H. Roane (Eds.), Handbook of applied behavior analysis. New York: Guilford.

Reid, D. H., & Parsons, M. B. (1995). Comparing choice and questionnaire measures of the acceptability of a staff training procedure. Journal of Applied Behavior Analysis, 28, 95–96.

Reid, D. H., & Parsons, M. B. (2006). Motivating human service staff: Supervisory strate- gies for maximizing work effort and work enjoyment. Morganton, NC: Habilitative Management Consultants.

Schwartz, I. S., & Baer, D. M. (1991). Social validity assessments: Is current practice state of the art? Journal of Applied Behavior Analysis, 24, 189–204.

Shoemaker, J., & Reid, D. H. (1980). Decreasing chronic absenteeism among institutional staff: Effects of a low-cost attendance program. Journal of Organizational Behavior Management, 2(4), 317–328.

Sturmey, P. (1998). History and contribution of organizational behavior management to services for persons with developmental disabilities. Journal of Organizational Behavior Management, 18(2/3), 7–32.

van den Pol, R. A., Reid, D. H., & Fuqua, R. W. (1983). Peer training of safety-related skills to institutional staff: Benefits for trainers and trainees. Journal of Applied Behavior Analysis, 16, 139–156.

Wilder, D. A., Therrien, K., & Wine, B. (2005). A comparison between survey and verbal choice methods of identifying potential reinforcers among employees. Journal of Organizational Behavior Management, 25(4), 1–13.

Wolf, M. M. (1978). Social validity: The case for subjective measurement or how applied behavior analysis is finding its heart. Journal of Applied Behavior Analysis, 11, 203–214.