Abstract
Behavioral social work is the application of behavior analysis to the field of social work. There are behavioral social work interventions for individuals, groups, and communities. Nevertheless, behavioral social work is far from a widely adopted approach among social work practitioners. A reason for the underuse might be seen in the fact that most interventions in behavioral social work aim at individual clients and groups. Social work could further benefit from the application of methods taken from behavioral community interventions. Behavioral community interventions modify the behavior of a larger group of people (e.g., pedestrians using a crosswalk, cafeteria patrons, students using a university building) with antecedent- and consequence-focused interventions. The article describes a training program for social work students in behavioral community interventions. Results of two projects, undergraduate students designed and performed, are presented. Behavioral community interventions are recommended as a valuable part of the education of social workers.
Keywords
There’s a natural fit between the methods of behavior analysis and the field of social work (Gambrill, 2013; Thyer, 1987). The application of behavior analysis to the field of social work is called behavioral social work. Its characteristics are, among others (Gambrill, 1995): Problems are translated to behaviors: Virtually every problem, humans encounter, is caused by or involves behavior: Someone has done something or failed to do something. The behavior is observed (or measured in another, objective way, e.g., via its results) in the situation. The functional relations between behaviors, its antecedents, and its consequences are essential for the understanding and modification of the behavior. The methods of behavior modification are based on applied research and high-quality scientific evidence. The leading question is, whether the life quality of the client is influenced in a positive way.
Behavioral social work interventions usually are among the most research-supported methods in the field of social work (Thyer, 2015a). Unfortunately, the use of research-supported interventions is neither a required component of our profession, nor common (Ghanem, Lawson, Pankofer, Maragkos, & Kollar, 2017; James, 2016; Palinkas et al., 2016; van der Zwet, Beneken genaamd Kolmer, & Schalk, 2016). One of the reasons for this research-practice gap (Gambrill, 2015a) is the education of social workers who often lacks adequate training in scientific research methods.
The scope of behavioral social work stretches from the micro- to the macrolevel. There are well-researched behavioral social work interventions for individuals, groups, and communities. Although interventions with individuals and groups are often borrowed from or identical with methods used by psychologist or other professionals, interventions on the community level are more specific to social work as a discipline in its own rights (Ohmer & Korr, 2006; Rothman & Thyer, 1984).
Behavioral community interventions target not necessarily a community in the strict sense (e.g., a municipal community) but a group of people (clients) who only share a few commonalities. The boundaries of this group are fuzzy, individual group membership often changes over time, sometimes individuals are group members only for a few moments. An example of a group targeted in a behavioral community intervention are the motorists using a road with a pedestrian crosswalk. The intervention could consist of different signs, pedestrians give to signal their intent to cross the street (Crowley-Koch, Van Houten, & Lim, 2011) or different versions of a street sign (Bennett, Manal, & van Houten, 2014). The dependent variable would be the number of near accidents at this crosswalk. The members of this group are the motorists. The research design of the studies is usually a single case design (where the subject serves as its own control), but unlike single case designs with the same individuals across all observational sessions, the members of this group change over time. Some motorists use this street only once, others several times, but probably no motorist uses this street at every observational session. Nevertheless, the studies in behavioral community interventions usually demonstrate the experimental control of the intervention.
Behavioral community interventions have been researched in a variety of situations, with different populations and intervention types (Luyben, 2009; Thyer, Himle, & Santa, 1986). Most of the interventions fall into the categories of antecedents-only interventions and interventions that combine antecedents and consequences. Examples are as manifold as the use of prompts to encourage litter removal among theatergoers (Burgess, Clark, & Hendee, 1971), reminders and incentives to encourage parents to bring their children to dental screening (Reiss & Bailey, 1982), verbal prompts to ask parents to buckle up their children in shopping carts (Ferrari & Baldwin, 1989), passive (a poster) and mediated prompts (a sign held by a research assistant) to request motorists to use the turn signal when leaving a parking garage (Clayton & Myers, 2007), and the use of feedback to reduce energy consumption in private households or university buildings (Clayton & Nesnidol, 2017; Fischer, 2008).
Training in Behavioral Social Work
The scope of behavioral social work corresponds to the vocational fields for social workers. Social workers work with individual clients and groups of clients. But social workers also work for the community. Unlike other professions, it’s part of the job description of a social worker to try to make this world a better place. The self-image of social work implies the willingness to take responsibility for the community (cf. National Association of Social Workers, 1996).
Education in behavioral social work is far from being widespread (Dillenburger, Godina, & Burton, 1997; Gambrill, 2015b; Thyer, 2015b). Often behavioral social work education is restricted to the microlevel, reducing social work to therapy with individuals and small groups. This restricts the possibilities of social work. Social work is more than therapy, it includes working with the client in his environment and designing the environment to enable a better quality of life. This calls for an inclusion of behavioral community interventions in the education of social workers. Students should learn how behavioral community interventions work and how to conduct own research in the area of macrolevel social work (Thyer, 2008).
Project Seminar
Undergraduate students (N = 9) of a social work program in Germany took part in a project seminar on implementing behavioral community interventions in their immediate environment. The curriculum of this program adheres to the principles of behavioral social work. The participants of the seminar already knew about basic principles of behavior analysis and had practiced with individual clients, sometimes with groups of clients. The curriculum included lectures on applied behavior analysis, tutorials on behavior analysis and behavior modification, a one-semester self-modification project, and an internship with case work with individuals and groups.
Course of the Seminar
Introduction to Behavioral Community Interventions
After a rehearsal of behavioral methodology, students presented and discussed several sample studies on behavioral community interventions, among others: A study by Austin, Hatfield, Grindle, and Bailey (1993) about the use of signs to increase recycling in an office environment.
Anderson, Crowell, Hantula, and Siroky (1988) used feedback to increase cleaning in a student-managed university bar. A sign, stating “Thank you for stopping,” increased complete stopping at an intersection on the campus of a public university (Austin, Hackett, Gravina, & Lebbon, 2006). Feedback on energy consumption led to reduction in electric power consumption in a university residential hall (Bekker et al., 2010). A 10-year-old boy increased recycling in his neighborhood with prompts and feedback (Keller, 1991).
Especially the last study demonstrated that one can manage a behavioral community intervention study even with few or no training in research methods. This encouraged the students to try a project of their own.
From the studies discussed, the students compiled a list of possible techniques to use in a behavioral community intervention. In general, interventions are either antecedent- or consequences-based or both.
Antecedents are: task clarification, the “precise specification of the behavioral components of a job” (Crowell, Anderson, Abel, & Sergio, 1988, p. 65); training of knowledge and skills (Arthur, Bennett, Edens, & Bell, 2003; Perlow, 2001); equipment—the objects and their arrangement, necessary to perform the task; verbal (e.g., Ferrari & Baldwin, 1989), written (e.g., Okinaka & Shimazaki, 2011), and graphic (e.g., Reetz, Loukus, Taylor, & Dixon, 2014) prompts; goal-setting (Latham & Yukl, 1975; Locke & Latham, 1990) and voluntary commitments (e.g., Geller & Lehman, 1991).
Consequences are as follows: The results of the behavior: After cleaning up, the room is clean. Reactions of the social environment: Being friendly to others leads to friendly reactions from others. Verbal (e.g., Metz, Kohn, Schultz, & Bettencourt, 2017), written (e.g., Babcock, Sulzer-Azaroff, Sanderson, & Scibak, 1992), or graphic (e.g., Loewy & Bailey, 2007) feedback. Reinforcers, social (e.g., praise) and tangible (e.g., Mathews, Hodson, Crist, & LaRoche, 1992).
Antecedents-only interventions are generally easier to implement (e.g., comprise only the posting of signs), while interventions with antecedents and consequences combined typically are more effective (C. M. Johnson, Redmon, & Mawhinney, 2001; Komaki, Collins, & Penn, 1982).
Selection of Socially Relevant Problem
In a brainstorming session, the students had to decide on a socially relevant problem they wanted to target with their project. Students formed two project teams and assigned the different roles of project manager, contact person to external agents (e.g., university facility management, student services), record keeper, and so on.
After discussion, the students decided on these two projects, both from the field of environmental protection:
The amount of waste from disposable coffee mugs is considerable
1
: 7.6 Million disposable coffee mugs are used in Germany every day. Disposable coffee mugs can’t get recycled because they are made from a compound of paper and plastics. Every year, the equivalent of 43,000 trees is consumed by the production of disposable coffee mugs, additionally 1.5 billion liters of water and 22,000 tons of oil are consumed during the production of the mugs and the lids (made of polystyrene). Several communities in Germany have started awareness campaigns to reduce the waste resulting from disposable coffee mugs. The lecturing building of the university had only a small cafeteria with two coffee dispensers. The cafeteria staff had no facilities for cleaning of dishes, therefore only disposable coffee mugs were offered. It was possible to bring your own coffee mug to fill at the dispenser or to reuse a disposable coffee mug, but seemingly only few of the students, professors, and university staff in the building took advantage of this opportunity.
The best way to conserve resources is to avoid waste. The second-best way is waste separation to enable recycling of resources. Although waste separation in Germany is relatively well organized for private households, in public space, there is still room for improvement. While there were different waste baskets for paper and residual waste in the offices of the university the project took place, in the lecturing building, there were no separate paper bins. There were residual waste bins in every seminar room and in the halls. According to the university’s facility management, a prior attempt to introduce paper recycling failed because there was too much residual waste mixed with the paper waste, which made recycling virtually impossible. Ideally, paper waste should contain no more than 3% of residual waste to make recycling profitable (DIN EN 643:2014-11, 2014).
Definition of Problem Behavior
Project 1
The problem behavior (behavioral excess) can be defined as following: “The cafeteria patron takes a disposable coffee mug when buying a coffee (or another hot beverage).”
There are two incompatible alternative behaviors (behavioral deficits): “The cafeteria patron uses a reusable coffee mug when buying a coffee (or another hot beverage)” and “the cafeteria patron reuses a disposable coffee mug when buying a coffee (or another hot beverage).”
Project 2
The problem behavior (excess) is “the student or lecturer (user of the building) throws a recyclable piece of paper waste in the residual waste bin.”
The incompatible alternative behavior (deficit) wasn’t possible at all, prior to the intervention: “The student or lecturer (user of the building) throws a recyclable piece of paper waste in the paper waste bin.”
If there were paper waste bins available, there could occur a second problem behavior (excess): “The student or lecturer (user of the building) throws a piece of residual waste in the paper waste bin.”
Functional Assessment of Problem Behavior
In preparation of the intervention design, the students undertook an assessment of the problem behavior including its functional aspects. The assessment was guided by the Performance Diagnostic Checklist (PDC) by Austin (2000). This assessment tool involves asking questions and collecting data in each of four areas: (1) antecedents, (2) equipment and processes, (3) knowledge and skills, and (4) consequences. The PDC has been used in several interventions in the field of organizational behavior management but is adaptive to behavioral community interventions as well (Johnson, Casella, McGee, & Lee, 2014).
The students summarized the results of the assessment in contingency charts, using the Antecedents–Behavior–Consequences (A–B–C format).
Project 1
For the problem behavior of using a disposable coffee mug and the alternative behavior of using a reusable mug, the students identified the contingencies, shown in Table 1.
Contingency Analysis of Project 1.
aA consequence is considered strong if it is “big” enough, if it follows immediately on the behavior, and if it follows with high probability.
In short, the problem behavior of using a disposable coffee mug is strengthened by strong positive consequences (most of them related with the low-response effort of using a disposable mug). Additionally, by interviewing the cafeteria patrons, the students found out that many patrons (students as well as professors) thought that disposable coffee mugs are recyclable paper waste (which isn’t the case). The alternative behavior of using a reusable mug is weakened by strong negative consequences (most of them related with high-response effort). Additionally, one must plan for using a reusable coffee mug well in advance (you must take your coffee mug with you, when leaving home, even if you drink your coffee only a few hours later).
Project 2
Project Team 2 chose a different format to summarize the results of their assessment because the alternative behavior wasn’t possible prior to the intervention. The results are shown in Table 2.
Contingency Analysis of Project 2.
The sine qua non of paper recycling is the availability of a paper waste bin (basically, one could also carry paper waste to the next paper bin—e.g., at home—but the effort was considered too high to take this possibility into account). If it was available, one must be able to identify which bin is for paper and which is for residual waste. One must know which items are paper and which items are residual waste (see above: many users of the building thought that disposable coffee mugs are paper waste). Depending on the location of the paper waste bins (cf. Fritz et al., 2017; Ludwig, Gray, & Rowell, 1998), the additional effort of separately discarding paper and other waste was considered relatively low.
Intervention Design Based on the Results of the Functional Assessment
Project 1
According to the assessment, deficits of antecedents and consequences were responsible for the problem behavior. Many patrons didn’t know about the problems related with the usage of disposable coffee mugs and the consequences weren’t sufficient for a behavior change. The cafeteria patrons should get more information on the problem. They should experience that their decision is essential for the solution of the problem and valued by the community of students. The team decided for presenting three interventions in succession.
Other intervention options with antecedents (e.g., selling of reusable coffee mugs at the cafeteria) or consequences (e.g., granting a discount for not using a disposable mug) were discussed but had to be dismissed because of organizational obstacles.
Project 2
After baseline observations, the students decided for a two-stage intervention.
Revised poster from Project 2/Intervention 1. Poster indicates what items are recyclable and that disposable coffee mugs aren’t recyclable.
Poster from Project 2/Intervention 2. Poster shows the percentage of residual waste in paper waste bins, asking to reduce the percentage under 10%.
Research Design based on the Requirements for Experimental Control and Feasibility
Project 1
The students determined the number of used disposable mugs per day by counting each morning the remaining stock of mugs in the cafeteria. During the first 2 weeks of baseline observations, two students counted together. There was no disagreement between the two observers (the count is relatively simple); therefore, there were no further checks for observer reliability.
Unfortunately, it wasn’t possible to count the number of sold hot beverages. This would have allowed to calculate the percentage of hot beverages sold with and without a disposable mug, a much better measure of the dependent variable than the absolute number of mugs. The change in absolute numbers is also influenced by seasonal variables (e.g., with rising temperatures, the selling of hot beverages decreases), consisting a major threat to internal validity.
The research design is an A–B–BC–CD design: Baseline (A) was the number of mugs expensed every weekday prior to intervention, lasting 4 weeks. Fridays, Saturdays, Sundays, and holidays were excluded. The first intervention (B, 4 weeks) was the informational poster, the second intervention (BC, 1 week) included the poster and the boxes with waste mugs, the third intervention (CD, 1 week) consisted of the boxes and the feedback information.
The research design is relatively weak in terms of internal validity (e.g., causal inference), but it is appropriate for the simpler purpose of evaluating changes in the outcome measures. Due to time restraints and problems in the coordination with the operating organization of the cafeteria, a planned return to baseline wasn’t included.
Project 2
Throughout the project, the students counted every evening (at 5 p.m.) the items disposed in the waste bins of the first and the second floor of the building. They emptied the bins on a tarp and counted the number of paper and residual waste items. Students wore cut-proof gloves. Then, they calculated the percentage of residual waste within the whole waste (= residual waste + paper waste). The first floor served as a control condition, only the percentage of residual waste in the residual waste bins was recorded. The interventions took place on the second floor, where the paper waste bins were put in place. The students recorded the percentage of residual waste in the residual waste bins and (after intervention started) in the paper waste bins.
Students counted the waste items always in pairs of two. No formal interobserver reliability was calculated.
The students used a A–B/B′–B′C design and added a between-subject component (the first floor) with no intervention as a control condition. The baseline (A) was taken for 3 weeks. The intervention started with the setting up of the paper waste bins. The first intervention (B) included the posters stuck to the bins, explaining which items are paper waste. After a week, the posters were slightly changed (B′) to make the information more noticeable. This phase lasted 2 weeks. In the final intervention phase (B′C), the students posted the feedback information about the reduction of residual waste in the paper waste bins. This phase lasted 2 weeks.
This design allowed for a demonstration of experimental control because of the comparison with the control condition. The measure for the dependent variable was a percentage instead of absolute frequency, making the measure more reliable and immune against threats to internal validity (e.g., variability due to the number of persons in the building).
Results and Discussion
Project 1
During baseline, the students observed a declining trend in the use of disposable coffee mugs (cf. Figure 3). According to the cafeteria staff, they sell fewer hot beverages during summer than during winter term. The mean over the whole baseline observation period is 260 cups per day. From May 11 (Session 7) on, the baseline approached stability, with only a slightly decreasing trend. The project team decided to take 240 mugs as the baseline reference for the intervention phases.
Results from Project 1. Use of disposable coffee mugs.
The first intervention (posters) showed no effect at all, the second intervention (the boxes with used mugs) went along with an increase in mugs. The data points of the third intervention are completely within the range of the first intervention, but, in average, the use of mugs is lower (148.5) than in any other phase of the project (cf. Figure 3).
Due to many threats to internal validity, the results of this project are open to interpretation. The main weaknesses of the project are the lack of experimental control and the measure for the dependent variable (absolute number of mugs), which is prone to many sources of variability. On the other hand, the project led to many informal discussions in the university and an article in the local newspaper. As a consequence, the operator of the cafeteria, the student services, began offering reusable coffee cups and sells reusable mugs at the cafeteria. Disposable mugs are no longer available.
Project 2
During baseline, about 40% of the items in the residual waste bins were recyclable paper waste (Figure 4 shows the percentage of residual waste in the residual waste bins and—only for the second floor, only after start of intervention—of residual waste in the paper waste bins). There were no differences between the first (control) and the second floor. During baseline, an increase in the percentage of paper waste in the residual waste bins occurred, which doesn’t lessen the interpretability of the results because the aim of the project was a decrease in the percentage of paper waste in the residual waste bins.
Results from Project 2. Percentage of residual waste in residual waste or paper waste bins. Arrow indicating the change from Poster 1 (B) to Poster 2 (B′).
The first intervention (positioning of paper waste bins on second floor + posters on bins) led to a decrease of paper waste in the residual waste bins on the second floor only. The percentage of paper waste in residual waste bins on the first floor—where no intervention took place—remained stable throughout the whole project period. The percentage of residual waste in the paper waste bins on the second floor dropped after 2 days from 30% to 15%. This was still too high for proper recycling of the paper waste. A second intervention (posting of feedback graphs, showing the percentage of residual waste in paper waste bins) unfortunately brought no further improvement.
Overall, the intervention was successful, although the informational measures didn’t succeed in bringing the percentage of residual waste in the paper waste bins under the threshold, required for the optimal recyclability of the paper waste. The research design allowed for evidence of experimental control. As with the other project, the topic of paper recycling was informally discussed in the university. After completion of the project, the facility management of the university decided to set up permanent paper waste bins in the halls of both floors.
Discussion and Applications to Practice
The seminar was evaluated, using an online survey and a group discussion format. Students were asked how they rated the course overall, how clear they found the structure and outline of the seminar, if the course content was related to their prior knowledge and if they thought that the seminar was a good preparation for the occupational field of social work.
Students anonymously rated the seminar as “good” (4.0 on a scale from 1 = bad to 5 = very good). The structure and requirements of the seminar were rated as good (4.3). They found their prior knowledge very helpful (4.9). They saw some relation of the seminar content to the occupational field of social work (3.6).
During the discussion, the students said that they liked the seminar rationale and the project, although it was a lot of work for them (especially compared to the workload in other seminars). They liked the practical aspects of the seminar, it showed them the universal applicability of behavioral principles beyond the behavior modification interventions for individual clients and groups.
From the instructor’s point of view, the seminar enhanced the students’ understanding of the research methods. They were critical (sometimes even too pessimistic) about the interpretability of the results of their projects and realized the methodological weaknesses.
The projects shouldn’t be assessed like proper research projects. The educational aspect is more important than the scientific value of the projects. Another problem was the time constraint. The students had to plan a project from the scratch, and conduct and analyze it in only 3.5 months, besides their other obligations. Time constraints are also the reason, why the students refrained from “bigger” interventions and a return to baseline. They preferred mostly antecedent interventions to consequential interventions because they imagined them as more time-consuming. For example, the instructor suggested to Project Team 1 to use a personal active prompt (Clayton & Helms, 2009). The students could have waited outside the cafeteria with a sign prompting the use of reusable coffee mugs. If a patron uses a reusable mug, the students could have flipped the sign, showing a message thanking the customer for using a reusable mug.
The results of the seminar shed light on deficits in the training of the students in behavior analytic research methods. Prior to the seminar, the students had little training in single case research methods. A lot of time was consumed for lecturing and discussing the principles and logic of single case research.
Overall, behavioral community interventions are a valuable but underused instrument in social work education. The seminar demonstrated that it is possible to bring this third practice field of social work (in addition to working with individual clients and groups of clients) to undergraduate social work students.
Footnotes
Acknowledgments
The author is grateful to Simone Barthel, Christina Busack, Tina Ehrenfeuchter, Veronika Frank, Marvin Lobozar, Pia Schmidhuber, Theresa Wendler, Jeannine Wendt, and Lukas Werner for their contribution to the research and data collection during the seminar. In addition, special thanks go to the anonymous reviewers for comments on an earlier draft of the article.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.