State of Current Training for Severe Problem Behavior: A Survey

Abstract

Ensuring that Board Certified Behavior Analysts (BCBAs) receive sufficient training in various forms of practice is an area of clinical importance that necessitates additional attention. Given that there is relatively limited information available on the extent to which BCBAs receive ongoing training and support, Reed and Henley (Behavior Analysis in Practice, 8, 16–26, 2015) conducted a study to evaluate the various types of training offered to BCBAs and aspiring certificants. As such, the purpose of the current investigation was to extend the findings of Reed and Henley by surveying BCBA respondents who practice (a) primarily in the home setting and (b) with clients who engage in severe problem behavior. The results are discussed with respect to 3 socially significant findings. First, 43% of respondents had been assigned their first severe case without initial or ongoing support. Second, only 35.2% of respondents received training on functional analyses more than 1 time. Finally, 5 respondents (4%) answered that the majority of their work was with clients 19 years old or older.

There has been a recent increase in publications related to improving the standards for (a) graduate training programs in applied behavior analysis (ABA; Behavior Analyst Certification Board, 2017; Carr, Nosik, & Luke, 2016) and (b) supervision of individuals pursuing certification through the Behavior Analyst Certification Board (BACB; Hartley, Courtney, Rosswurm, & LaMarca, 2016; Sellers, Valentino, & LeBlanc, 2016; Turner, Fischer, & Luiselli, 2016). Carr (2016) suggested a potential deficit of over 55,000 Board Certified Behavior Analysts (BCBAs) to meet the growing need for behavior-analytic services, making discussions related to the quality of graduate training and BCBA trainee supervision extremely important.

The BACB states that BACB credentials are entry level credentials, indicating that individuals have met minimum standards for training and experience in the discipline of behavior analysis. In an effort to ensure they remain current on developments in the discipline, certificants should acquire additional training beyond the requirements for certification. (Behavior Analyst Certification Board, 2019, p. 1). As such, in addition to the increased focus on graduate training and supervisory practices for individuals seeking certification, we also need to focus on ensuring continuous and ongoing skill development for individuals who have already completed graduate school and have obtained their BACB credential.

Reed and Henley (2015) conducted a study to document the various types of staff and supervisory training and performance management procedures offered to BACB certificants and aspirants working in applied settings. The survey included responses from 382 individuals covering five major content sections: (a) demographics, (b) preservice training, (c) ongoing training, (d) incentives, and (e) training on staff supervision. Results suggest several areas that require increased attention; in particular, nearly half of the respondents indicated that they did not receive initial preservice training before working independently, and nearly one third of respondents indicated a lack of ongoing training in their work setting. These outcomes are especially concerning given that the provision of effective staff training and performance management is an ethical obligation of BACB certificants (Behavior Analyst Certification Board, 2014). Further, inadequate staff training increases the likelihood of inadequate treatment implementation, which, by definition, compromises a client’s right to effective behavioral treatment (Van Houten et al., 1988).

Although the results of Reed and Henley (2015) contribute to the available research on current staff training and performance management practices, there are some areas worth extending. First, Reed and Henley (2015) focused on evaluating current training and supervision practices with respect to structure and format (e.g., the authors only included questions about the amount of training that was provided and the training modalities); the authors did not include survey questions about the respondents’ client profiles (e.g., degree of deficit, severity of problem behavior). Specifically, out of the 33 survey questions posed, only one question targeted the actual content of the training by asking respondents to identify the “direct relevance of training topics” (Reed & Henley, 2015, p. 21). It may be useful to gather information about whether the content of the staff training aligns with the presenting clinical assessment and programming needs of the trainees’ target client populations. Specifically, are staff receiving the necessary and required training to support a given client profile? For example, if a clinician primarily focuses on the assessment and treatment of severely challenging behaviors, is that clinician receiving specific training on how to conduct functional analyses (FAs) and designing related function-based interventions?

A second area worth extending from Reed and Henley (2015) is that only 18.85% of the respondents reported that they work in the home setting (the remainder worked in various other locations, such as centers, hospitals, and schools). Many aspects of service provision are directly influenced by the service setting, and home-based services seem to pose specific difficulties. Dixon et al. (2017) conducted a between-group analysis of home- and center-based services for children diagnosed with a developmental disability and found that participants mastered targets much faster when services were rendered in a center. Although further research is needed to identify the factors that contributed to the outcome differences, these data seem to indicate certain treatment barriers within the home setting. This may be a problem given that not all clients who need services can receive them from a center. Factors such as transportation, client age, adult client refusal, problem behavior associated with the home setting, and so on may require services in the home. Thus, service providers may be compelled to work in the individual’s home.

Additionally, caregivers may wish to mitigate severe problem behavior in the home before sending their son or daughter to another setting (e.g., hospital, clinic, residential home). This is in line with the BACB’s Professional and Ethical Compliance Code for Behavior Analysts, which instructs behavior analysts to recommend the least restrictive procedures likely to be effective (Behavior Analyst Certification Board, 2014). This is compounded by the fact that those who engage in severe problem behavior may have fewer options as the federal government continues to urge for more community-based intervention while closing many developmental centers across the United States (Lerner & Pollack, 2015). If behavior analysts are able to intervene on severe problem behavior in the home, it may reduce the need for more intrusive interventions. As such, it may be valuable to gather more survey outcomes from certified clinicians who primarily work in the home setting.

As the demand for behavior-analytic services increases, there is a growing need to ensure appropriate initial and ongoing staff training practices. Although there is a plethora of available research on effective staff training practices (Clayton & Headley, 2019; Hine, 2014; Jimenez-Gomez, McGarry, Crochet, & Chong, 2019; Parsons, Rollyson, & Reid, 2013), more information is needed on how training is provided and the extent to which certified clinicians are accessing training that is directly related to their target client populations and settings. As such, the purpose of the current investigation was to conduct a partial replication and extension of Reed and Henley (2015) by surveying BCBAs and BCBA-Ds (BCBA–Doctoral level) who have experience supporting individuals with severe problem behaviors in the home setting.

Method

Participants

Participants in the study included BCBAs and BCBA-Ds who responded to an anonymous survey, which was distributed through the BACB directory to certificants living in the United States. The number of individuals who received the survey is unknown, given that certificants can block solicitations; thus, no response rate could be calculated. Three hundred twenty-seven individuals opened the survey and consented to participate. The inclusion criteria entailed three aspects, such that participants (a) completed at least 30% of the survey, (b) primarily provided services in home settings, and (c) had experience working with individuals who demonstrate severe problem behavior. One hundred twenty-five individuals were included in the final data analysis (38%).

Instrumentation

To evaluate the quality of behavior-analytic services provided to individuals who engage in severe problem behavior in the home setting, questions were hosted online through SurveyMonkey (SurveyMonkey, Inc., 2018). To prevent individuals from completing the survey more than once, the survey concluded with a question asking participants to confirm that they had not previously completed the survey.

The survey questions were divided into three sections. The first section pertained to respondents’ personal demographics (see Table 1) and included questions regarding the respondents’ ages, graduate degrees, program areas, program settings, credentials, years as a behavior analyst, states of residence, primary work settings, and their number of years in the field. The second section pertained to company information (see Table 2) and included questions regarding the respondents’ work environments, including whether they worked for a two- versus three-tiered company (i.e., whether or not the BCBA provided the majority of direct supervision hours); the number of clients assigned to their caseloads; the age range of their clients; and whether clients’ behavior posed no harm (defined as “off-task or stereotypic behavior”), minimal harm (defined as “poses no harm to self or others but disruptive to the environment”), moderate harm (defined as “poses harm to self or others but can be redirected”), or severe harm (defined as “often results in tissue damage to self or others and puts the client at risk of incarceration/hospitalization”).

Table 1.

Personal Demographic Information

	N	%
Gender
Male	23	18.4
Female	102	81.6
Age
M (39)
Range (24–75)
Graduate degree
MS	48	38.4
MA	34	27.2
MEd	26	20.8
PhD	17	13.6
Program area
Applied behavior analysis	65	52
Special education	21	16.8
Psychology	17	13.6
Education	9	7.2
Clinical psychology	6	4.8
Counseling	5	4
Experimental analysis of behavior	1	0.8
Social work	1	0.8
Academic program type
On the ground	74	59.2
Online	31	24.8
Blended (on the ground and online)	20	16
BACB credential
BCBA	110	88
BCBA-D	15	12
Years certified
0–2	36	28.8
3–5	42	33.6
6–8	21	16.8
9–11	14	11.2
12–14	6	4.8
14+	6	4.8
State of residence
California	40	32
Florida	8	6.4
New York	8	6.4
Virginia	8	6.4
Connecticut	6	4.8
Arizona	5	4
Maryland	5	4
Illinois	4	3.2
Texas	4	3.2
Michigan	3	2.4
Nevada	3	2.4
North Carolina	3	2.4
Pennsylvania	3	2.4
Washington	3	2.4
Alabama	2	1.6
Colorado	2	1.6
Hawaii	2	1.6
Massachusetts	2	1.6
New Jersey	2	1.6
Alaska	1	0.8
Georgia	1	0.8
Indiana	1	0.8
Kentucky	1	0.8
Maine	1	0.8
Minnesota	1	0.8
Missouri	1	0.8
Nebraska	1	0.8
South Carolina	1	0.8
Tennessee	1	0.8
Utah	1	0.8
Wisconsin	1	0.8
Primary setting of services
Home	125	100
Years of experience with autism
<1	0	0
1–3	7	5.6
4–6	16	12.8
7–9	27	21.6
10–12	23	18.4
13–15	10	8
15+	42	33.6

Table 2.

Company Demographic Information

	N	%
Type of company
Three tiered	29	23.2
Two tiered	82	65.6
Other	14	11.2
Current client caseload
<5	30	24
6–9	35	28
10–15	37	29.6
16–19	7	5.6
20–29	8	6.4
30–49	3	2.4
50+	5	4.0
Age range of current clients
0–3	4	3.2
4–11	98	78.4
12–18	18	14.4
19–29	4	3.2
30+	1	0.8
Average percentage of problem behavior
No problem behavior		7.0
Problem behavior with no harm		30
Minimal problem behavior		30
Moderate problem behavior		29
Severe problem behavior		14

Before moving on to the third section of the survey, respondents were given a definition of severe problem behavior and asked if they had previously worked with such behavior. If they confirmed prior experience, they were permitted to answer the questions in the final section of the survey. The operational definition for severe problem behavior was as follows: any behavior that reasonably may result in harm, damage, or threat to the safety of oneself, another person, or property—wherein the result of the behavior may be or has been incarceration or hospitalization (including 5,150 or more restrictive placements).

This final section pertained to the respondents’ experience assessing and/or treating severe problem behavior (see Table 3) and included questions regarding the number of clients for whom the respondents had provided clinical services, the topographies of problem behavior addressed, the type of training received, the type of FA training received for severe problem behavior, the number of FA training opportunities on severe problem behavior, initial or ongoing support, the type of initial support, their supervisors’ years of experience, hours of ongoing support, the type of ongoing support, the recommendation for a more restrictive placement, whether the recommendation was denied or delayed, and the rationale for the delay.

Table 3.

Training for Severe Problem Behavior Cases

	N	%
Number of severe problem behavior cases
1–2	26	20.8
3–5	32	25.6
6–9	13	10.4
10–15	22	17.6
16–19	4	3.2
20+	28	22.4
Topography of severe problem behavior
Hitting	123	98.4
Biting	116	92.8
Elopement	116	92.8
Object throwing	115	92.0
Pulling others’ hair	111	88.8
Kicking	110	88.0
Scratching	110	88.0
Self-hitting	110	88.0
Pinching	108	86.4
Grabbing	106	84.8
Excessive screaming/yelling	106	84.8
Head butting	104	83.2
Self-biting	104	83.2
Breaking objects	103	82.4
Pushing	102	81.6
Spitting at others	99	79.2
Skin picking	96	76.8
Slapping	93	74.4
Ingesting objects/pica	93	74.4
Disrobing	92	73.6
Severe threats	81	64.8
Fecal smearing	75	60.8
Self-hair pulling	70	56.0
Licking and/or mouthing others	69	55.2
Masturbating in public	62	49.6
Body slamming	58	46.4
Grabbing others’ genitals	58	46.4
Excessive eating	58	46.4
Bowel or urine retention	58	46.4
Rumination	57	45.6
Rectal digging	54	43.2
Operant vomiting	54	43.2
Eye gouging/poking	51	40.8
Choking	49	39.2
Genital rubbing against others	49	39.2
Violent ideation	48	38.4
Harm to nonhuman animals	42	33.6
Fecal throwing	35	28.0
Coprophagia	31	24.8
Self-restraint	29	23.2
Self-cutting	23	18.4
Fire setting	19	15.2
Formal training for severe problem behavior
In-field clinical supervision	86	68.8
Other	10	8.0
Practicum project in a graduate program	9	7.2
Never received formal training for severe behavior	8	6.4
Attended a workshop	8	6.4
Read a research article	4	3.2
Formal training for functional analysis (FA)
In field on multiple occasions	44	35.2
Class in graduate school	27	21.6
In field for one occasion	24	19.2
Never received formal training	19	15.2
Center-based clinic training	11	8.8
FA training opportunities
Never received formal training	23	18.4
1	19	15.2
2–3	44	35.2
4–6	21	16.8
7–9	4	3.2
10+	14	11.2
Assigned a case without training
Yes	54	43.2
No	71	56.8
Type of initial support for first severe problem behavior case (n = 120)
Several overlaps	42	35
Some overlap	26	21.7
Clinic meeting (one to one)	22	18.3
No initial support	15	12.5
Group meeting	9	7.5
Discussion of key research	5	4.2
Provision of research without discussion	1	0.8
Experience of supervisor (in years) providing support (n = 120)
0	4	3.3
1–2	9	7.5
3–5	24	20.0
6–9	19	15.8
10+	38	31.7
Unknown	26	21.7
Hours of ongoing support for first severe case (n = 120)
0	22	18.3
1–2	36	30.0
3–5	33	27.5
6–9	11	9.2
10+	18	15
Type of ongoing support (n = 120)
Regular overlaps in field	49	40.8
In-office one-on-one meetings	36	30.0
Does not apply	19	15.8
In-office group meetings	15	12.5
Provided research	1	0.8
Duration of services for first severe case (n = 118)
<1	8	6.8
2–6 months	24	20.3
7–12 months	32	27.1
1–2 years	40	33.9
3–5 years	10	8.5
5+	4	3.3
Recommendation for a more restrictive placement (n = 118)
No	53	44.9
Group home (short term)	31	26.3
Group home (long term)	26	22.0
Hospitalization	34	28.8
Other	19	16.1
Denied or delay recommendation (n = 65)
Yes	15	23.1
No	50	75.9
Reason for denial or delay (n = 15)
Special request by funding source	2	13.3
Caregiver issues	0	0
Organizational reasons	3	20.0
Litigation	1	6.7
Unknown	5	33.3
Other	4	26.7

Response Measurement and Data Analysis

The percentage of response selection was calculated per question (27 total questions). Some questions permitted participants to select multiple answers, thereby calculating totals that exceeded 100%. Some questions also presented an option for the participant to write in an answer. Similar answers were grouped together, whereas unique answers were counted separately (and noted for future research). For example, some respondents wrote that they received training on severe problem behavior through “mostly in field” or “training on the job.” Given the question similarity, responses were grouped within the category of “in-field clinical training.” SurveyMonkey also provides filters that were used in various ways to analyze specific relationships among parameters (e.g., years of experience, age range of clientele).

Procedures

The survey link was e-mailed to potential recipients through the BACB directory to those who elected to receive solicitation. Upon opening the link, the online consent form was presented.

Results

Tables 1, 2, and 3 summarize the participants’ responses to the survey questions.

Personal Demographics

The majority of respondents were female (82%), with a mean age of 39 years (range 24 to 75 years old). Much of the sample held a master’s degree (86%; combined master’s of arts [MA], master’s of science [MS], and master’s of education [MEd]), and the majority of respondents reported that they received their education from an ABA program (52%). Most respondents received their education through an on-campus program (59%), as opposed to an online (25%) or blended on-the-ground and online (16%) program. The most common type of certification was BCBA (88%), with most respondents having been certified within the last 5 years (63%). Given the inclusion criteria, all respondents provided services in the home setting (100%). The distribution of respondents across the United States showed a large representation within California (n = 40; 32%). Table 1 summarizes participant demographic information in more detail.

Company Demographics

A majority of the respondents reported to be working for a company that was set up in a two-tiered model (i.e., the BCBA provides the majority of the supervision hours directly to the behavior change agent; 65.6% of respondents). Of the respondents who resided in California, however, more claimed to be employed by a company that followed a three-tiered model (someone working under a BCBA provided most of the supervision hours; n = 40; 55%). More than a quarter of respondents (29.6%) reported having client caseloads that ranged from 10 to 15 clients. The primary age group of the reported clientele was 4–11 years (78.4%). A majority of the respondents reported working primarily with clients whose target problem behaviors posed no harm (30%) or minimal harm (30%). Moderate harm was selected by 29% of the respondents, and severe harm was selected by 14%. Table 2 summarizes the company demographic information in more detail.

Severe Problem Behavior

A quarter of respondents (25.6%) reported having worked with three to five severe problem behavior cases during their careers. A slightly smaller share of respondents (22.4%) reported having worked with 20 or more clients who engaged in severe problem behavior across their careers. This section also presented a list of potential topographies of severe problem behavior (see Table 3). Almost all the respondents selected the topography of “hitting others” (98.4%), whereas very few had experience with “fire setting” (15.2%). In-field supervision for the assessment and treatment of severe problem behavior was reported to be received by a large majority of respondents (68.8%). During initial training, direct supervision of FA completion was provided on more than one occasion to a minority (35%) of respondents. When asked about the exact number of opportunities for FA training, a small portion of the respondents (35.2%) stated that they received two to three opportunities. Close to half of the respondents indicated that they never received initial or ongoing support for their first client who engaged in severe problem behavior (43.2%). When support was provided, initial support was, for the most part, provided in the form of multiple overlaps per month (35%). It was reported that the supervisors who provided the most support had at least 10 years of experience in severe problem behavior (31.8%). Respondents reported ongoing, month-to-month support for 1–2 hr (30%) and 3–5 hr (27.5%) hours most frequently. Many respondents indicated that they received either in-field overlaps (40.8%) or one-to-one meetings in the office (30%) as part of their ongoing training. When providing services to their first client who engaged in severe problem behavior, a majority of respondents (54.2%) ended services at or before the first year (no additional information was provided on the justification for service termination). For those participants who did recommend a more restrictive placement for their clients, hospitalization (28.8%) and short-term residential placement (26.2%) were selected most frequently. For those participants who indicated that the recommendations were denied or delayed (n = 15; 12%), many were unsure of the reason (33.3%).

Subgroup Analysis

Two subgroups were pulled from the sample and analyzed with respect to specific training and supervision factors. The first subgroup (n = 54) entailed those respondents who (a) received initial or ongoing supports for severe problem behavior and (b) had more than one opportunity for FA training. First, of the respondents in this subgroup, more of them received education from an on-the-ground university when compared to the larger sample. Although the increase was small, the number of respondents went from 59.2% to 63%. A second finding of this subgroup is that more of them worked in a two-tiered as opposed to a three-tiered company. Respondents who worked in a two-tiered company increased from 65.6% to 74.1%.

The second subgroup of respondents (n = 33) included only those who received an on-the-ground education from an ABA master’s or doctorate program. The majority of respondents in this subgroup attained their BCBA certificate within 1 to 8 years of the survey, yet only one respondent did not receive in-field training in FA, and a larger majority had multiple opportunities to practice FA. The percentage of those who received more than one opportunity for FA training increased from 67% in the sample to 79% in the subgroup. The subgroup only had one respondent who failed to receive in-field FA training, whereas the sample had 23.

Discussion

The purpose of the current study was to partially replicate Reed and Henley (2015) and extend their findings by surveying BCBAs and BCBA-Ds who have experience supporting individuals who engage in severely challenging problem behaviors in the home setting. The results support the previous findings that training provided to BACB certificants does not always adhere to evidence-based standards (e.g., behavioral skills training, training to mastery, promoting generalization) based on the lack of initial and ongoing support, particularly with respect to treatment for severe problem behavior in the home setting.

The current study offers several potential contributions to the available related research. Specifically, the results can be considered with respect to (a) implications for initial and ongoing training for individuals who have already acquired their certificates as a BCBA or BCBA-D and (b) potential concerns related to providing high-quality ABA-based services to the adult population.

The current outcomes support the findings from Reed and Henley (2015) that new BCBAs are sometimes assigned clients without receiving initial or ongoing supports (i.e., 43.2% did not receive training). This is concerning given that the respondents specified that these were clients who engage in severe problem behavior (i.e., any behavior that reasonably may result in harm, damage, or threat to the safety of oneself, another person, or property—wherein the result of the behavior may be or has been incarceration or hospitalization, including 5,150 or more restrictive placements). The BACB Professional and Ethical Compliance Code for Behavior Analysts emphasizes that behavior analysts should maintain competence through professional development (Behavior Analyst Certification Board, 2014). To that end, recently certified BCBAs would likely benefit from ongoing evaluation from their previous and current supervisors through mentorship models such as traditional dyadic (one-to-one senior and protégé) and the relatively newer collaborative model (cohorts with a facilitator).

Pololi and Knight (2005) described these two mentorship models in the field of academic medicine, which might be able to carry over to the field of behavior analysis. First, the dyadic mentorship model is viewed as the traditional model where a senior professional works directly with a junior protégé. Dyadic models can be assigned formally (by an institution) or informally (spontaneously through factors such as mutual interest and a need for help). There are plenty of benefits of the dyadic method (e.g., mutually beneficial, one-on-one attention to problems), but Pololi and Knight also describe a relatively new model that might be easier to implement, more inclusive, and equally effective. The collaborative monitoring program groups a few peers with mutual interests together and assigns a senior facilitator. In this model, a cohort of juniors can rely on each other, whereas the facilitator (i.e., a senior in the field) helps provide a space for development. Future researchers should look to evaluate the effects of different mentorship models on the initial and ongoing support of BCBAs.

A second potential contribution of this study pertains to the information gathered on FA training and implementation. The survey showed that 84.8% of respondents had some form of FA training, but only a minority of them (35.2%) received training on more than one occasion. Roscoe, Phillips, Kelly, Farber, and Dube (2015) conducted a survey to evaluate the current state of FA training and implementation. The majority of the respondents identified FA as the most informative assessment procedure (i.e., 67.8% identified it as such); however, descriptive assessment procedures were reported to be used most frequently by the vast majority of their respondents (84.4%). If respondents are lacking multiple opportunities to receive training on what is considered to be such a useful assessment procedure, they might be missing a vital chance to better understand the analytic dimension of ABA (Baer, Wolf, & Risley, 1968).

FA methodology is highly effective and is regarded as the gold standard in the assessment and treatment of severe problem behavior (Beavers, Iwata, & Lerman, 2013; Hanley, Iwata, & McCord, 2003), but it is more than a methodology. Schlinger and Normand (2013) pointed out that this methodology is not only the standard in ABA and the experimental analysis of behavior but also the foundation for behavior analysis as a whole. FA involves the demonstration of any environment-behavior relationship. In other words, embedded in the methodology is the fundamental concept that behavior is a function of the environment. If practitioners are receiving limited opportunities to learn FA methodology, could this lack of exposure impact their understanding of a vital concept in behavior analysis?

A third potential contribution relates to the issue of adult services. Specifically, of the 125 respondents who participated in the survey, only five (4%) answered that the majority of their clients were over the age of 18, indicating a potential lack of experience with adult populations. Given the growing population of adults with severe problem behavior (Factor, Heller, & Janicki, 2012), many clinicians may not be sufficiently trained to address this problem.

The present study has several potential limitations. These are discussed with respect to Reed and Henley’s (2015) findings to identify consistencies across surveys. One potential limitation of this study is that the responses could not be verified through observation. This is inherent to the methodology of surveys because they require a verbal response about other behavior. Therefore, the responses from the survey are under different contingencies from responses in their natural context. For example, the difference is between a respondent’s answer about his or her accurate performance of FA methodology and their actual performance of FA methodology. To help ameliorate this problem, survey questions and answers are often compared across studies to ensure their reliability and validity (Ponto, 2015). The current study received similar answers when compared to the results of Reed and Henley (2015). Additionally, the BACB directory was used to compare the number of BCBAs and BCBA-Ds from the sample to the population. The current study was within 6% of the BACB directory. Although this does not eliminate this potential limitation, it provides evidence for the reliability of the survey questions.

A second potential limitation of the study was that SurveyMonkey was not able to block multiple responses to the survey from the same person. Although there is no evidence of an individual taking the survey more than once, it was possible. To help reduce this problem, the survey presented an initial question regarding prior completion. If respondents answered yes, they were omitted from the final data analysis. Reed and Henley (2015) used Qualtrics as a survey host, which blocked participants from completing the survey multiple times. Although the survey hosts were different, the replicated questions received similar answers, ensuring the reliability of the questions.

A third potential limitation was the sample size of 125 participants. This is a small sample size compared to the number of current BCBAs; therefore, the data may not be representative of the whole. Part of the purpose of the current study was to extend Reed and Henley (2015) by surveying individuals who primarily work in the home setting. Reed and Henley included 382 respondents, and 72 of them primarily worked in the home setting. The current study included nearly twice as many respondents who worked in the home setting and found similar results with respect to replicated questions.

A final potential limitation was that some questions had a low number of responses; specifically, the last two, which had 15 responses. This was a result of previous questions eliminating respondents from the survey. For example, the last question only included individuals who made recommendations for more restrictive placements. Future research should help clarify questions that received a low number of responses.

This area of study could benefit from future research that evaluates the variables pertaining to the subgroup analysis. Surveying larger groups of BCBAs with respect to on-the-ground education, ABA program area, company structure (two tiered, three tiered, or other), and FA training opportunities may help by showing which combination of training factors contributes to a quality skill set. For example, it is conceivable that those who worked at a two-tiered company while receiving their in-field experience hours had more direct contact with their BCBA supervisor, allowing for additional opportunities for training (which could include FA), or those who attended an on-the-ground university in an ABA program had more opportunities to openly discuss concepts such as FA. The subgroup analysis showed some differences compared to the sample population; however, the subgroups had a limited number of respondents. Future research may help determine the relationship (if any) among variables such as company structure, education modality and program area, support in the field, and FA training opportunities.

Overall, the results of the current study suggest several areas that could benefit from future research including increased sample sizes, inquiries into adult services, replication of survey question for reliability and validity, and relationships on training and educational factors (e.g., modality of education, program area, two- vs. three-tiered company). Finally, future research should use questions found in both Reed and Henley (2015) and the current study to maximize the ongoing reliability and validity.

Implications for Practice

• Many BCBAs are not receiving initial and/or ongoing training for severe problem behavior in the home setting. This might be related to the format of education and in-field training (e.g., on the ground vs. online, two tiered vs. three tiered).

• The lack of training includes procedures such as functional analysis (FA) methodology, in that a significant portion of the respondent sample only had one (or no) opportunity for FA training.

• Of the 125 respondents, only five (4%) had primarily worked with adults (19 years old or older) in the home setting.

• Given the growing need for severe problem behavior intervention in adult services, it is recommended that future surveys and experimental studies continue to examine the variables that contribute to the training of effective behavior analysts. Future surveys should attempt to address the limitations described in this study.

Funding Information

This study was financially supported in part by the Center for Applied Behavior Analysis for access to SurveyMonkey and the BACB registry.

Compliance and Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study.