Autistic Adult Knowledge of the Americans with Disabilities Act and Employment-Related Rights

Abstract

As the autism community continues to grow, it is increasingly important to evaluate strategies to teach workplace-related skills to the members of this community. One of these skills is self-advocacy, defined by Test et al. as consisting of four components: self-knowledge, knowledge of rights, communication, and leadership. This study aimed to address the second component of this definition, knowledge of rights. Three autistic adults learned information about key terms from the Americans with Disabilities Act. Participants practiced identifying legal rights related to discrimination and inability to provide certain accommodations in a series of short text scenarios, using instructor-modeled rehearsal of similar scenarios as the independent variable. Results show that each participant entered the study with an existing level of knowledge and ability to interpret the rights of various scenarios (30–60% correct during baseline) and that training resulted in small improvements (a 20–40% increase) in accurate scenario interpretation. Future research should include additional techniques, such as role play, to enhance performance accuracy and examine long-term retention and generalization of the skills acquired. Social validity feedback suggests that the training used in this study was acceptable to participants and may be amenable to future studies.

1. Introduction

Individuals with autism spectrum disorder (ASD) have often reported difficulties surrounding finding and maintaining employment [1,2,3]. Approximately 40% of autistic individuals are unemployed [4], a rate higher than other groups of individuals with psychological diagnoses, though this rate does seem to be improving [5]. This lower overall employment rate may be due to a combination of factors. For example, social aspects of employment such as applying for jobs, correspondence with prospective employers, and succeeding in interview settings may be difficult for members of the autism community [6,7,8,9]. Furthermore, autistic workers may find it difficult to retain a job once it is earned [10,11]. Factors outside of a worker’s control, such as employer inflexibility and stereotype threat, may affect workplace performance and satisfaction, and additional attributes (such as social differences, sensory sensitivities, and time management) may produce various positive and negative effects on workplace experience and performance [12,13,14,15,16,17,18,19,20,21]. Even seemingly secondary skills, such as self-transportation, can play a role in gaining and maintaining employment [22,23,24].

Previous research suggests that if these difficulties can be addressed, autistic workers can flourish, and these individuals have a great deal to offer employers. For example, based on the job and the individual, an autistic worker who takes an interest in workplace tasks may deeply engage and dedicate themselves to these tasks in a manner that other workers may struggle to do [25,26]. Turnover is also less common in this community; as indicated by Scott et al. [27], an autistic worker who finds gainful employment is more likely than non-autistic workers to stay with an employer for a prolonged period. Finally, autistic workers have self-described several advantages, including forthright communication, increased efficiency and focus, and a strong sense of logic and pattern recognition [25]. While the concept of an overarching “autism advantage” is empirically dubious [28,29,30], these and other studies clearly demonstrate that autistic workers may perform just as well as other employees, if not exceeding them in key areas.

Furthermore, various changes and skills have empirical support in improving workplace satisfaction and performance. Social obstacles have been overcome with the help of a job coach, behavioral skills training, and technological assistance [31,32]. It was also found that regular feedback could reduce performance anxiety, and auditory prompting systems may improve performance without the need for increased supervision [33]. Additional supports, such as visual aids, can be used in training and may be retained if needed in the work environment [34].

Using these supports and changes, however, may be more difficult than anticipated. The autism community is tremendously diverse in factors such as sex, co-occurring diagnoses, interests, abilities, and many more [35,36,37]. Without knowledge of an individual’s unique skills and needs, employers may incorrectly assume what an “autistic” worker can or cannot do, potentially harming their work experience and productivity. Thus, a key skill, that of self-advocacy, may be very important in helping autistic workers access support, break down assumptions, and improve their overall employment experience.

1.1. Self-Advocacy

Self-advocacy, when discussed in the context of students with disabilities, includes four key elements [38]. First, self-knowledge describes an individual’s understanding of their strengths and weaknesses, as well as their understanding of their diagnoses and the impact(s) it may have. Second, knowledge of rights describes an individual’s understanding of what they deserve and have guaranteed by law, as well as the ability to evaluate a situation to determine if these rights are being met. Third, communication is the individual’s ability to take the initiative to communicate their wants and needs for change. How the individual communicates varies based on individual ability and context. For example, a vocal request “I need help” may be sufficient for classroom self-advocacy, while for a professional, a textual response such as a formal email may be more appropriate. Fourth is leadership, described as an individual taking initiative for their own circumstances and actively advocating for their community, organizations, and groups, potentially on a larger scale.

This skillset can be tremendously useful to autistic adults. The self-advocating individual can successfully describe obstacles or rights violations and request alterations or assistance; even if the requested changes are not feasible, the request can highlight an issue for consideration by an employer. Autistic individuals have historically not been consulted in decisions about their education or training [39], yet their experiences, perceptions, and goals can have a tremendous impact in these endeavors. Self-advocacy may also differ from “autism advocacy” by non-autistic individuals, such as parents [40], and self-advocacy topics and techniques may vary by the setting one is self-advocating in (e.g., school or workplace).

To the authors’ knowledge, research on the topic of self-advocacy has not yet established a unifying definition of self-advocacy among workers with disabilities. For example, Schena et al. [41] conducted a literature review on teaching self-advocacy skills and found that most studies examined self-advocacy exclusively in an educational, rather than employment-based, setting. However, while the definition proposed by [38] is designed to apply to individuals in academic settings, it could apply to workplace self-advocacy as well. Whether it be with fellow co-workers or individuals in a greater industry, or even all autistic workers, self-knowledge, legal knowledge, communication, and collective self-advocacy in a workplace environment are both feasible and necessary. The focus of this study was on the second component of self-advocacy defined by Test et al. [38], “knowledge of rights”, with these rights derived from the Americans with Disabilities Act (ADA).

1.2. The ADA and Workplace Rights

The workplace experience of workers with all forms of disability changed with the onset of the ADA [42,43]. Though not the first act of legislation ensuring the rights of workers with disabilities, this legislation outlined several key terms applied to a wide range of employers and employment fields. Title I of the ADA, and in subsequent legislation, broadly defines the term disability as when a person has a record of experience that inhibits one or more major life activities [44,45,46]. Differentiating employment based on disability status legally constitutes discrimination [44,45,46]. It is interesting to note that, with the onset of the ADA, overall employment of individuals with disabilities may have dropped in the early 1990s [47,48,49,50]. Researchers have attempted to explain this phenomenon, with the most popular explanation being an overall increased expense in hiring and maintaining workers with disabilities [48,49,50], though research to the contrary exists [51,52]. Given these reports, knowledge of the ADA’s provisions is an essential skill for autistic workers.

The ADA plays a significant role in the workplace experience of persons with disabilities. Skills and behaviors related to self-advocacy and the receipt of accommodations have been examined in previous studies. For example, Lindsay et al. [53] described a search of seven databases that produced 26 articles surrounding diagnosis disclosure behavior among autistic workers. However, based on the results of an earlier review [41], no studies have taught self-advocacy skills, nor examined them, in the context of employment or the ADA among the autism community.

The closest related study focused on teaching eight adults with various disability diagnoses to recognize violations of their civic rights and take action to express those rights [54]. These participants resided in a rehabilitation facility and had a wide range of diagnoses, none of which included autism. Participants in this study (four male, four female) were noted to lack skills necessary for employment or functional independence. During the study, participants met in a group setting to review civic rights. The researchers presented rights via a series of pre-written text, approximately four short sentences in length, with each scenario illustrating a violation or non-violation. The researchers then categorized the rights into four overarching themes (i.e., personal rights, community rights, human service rights, and consumer rights), each of which contained four to ten rights illustrated in scenarios (such as the right to marry, the right to vote, the right to privacy, the right to choose what to buy, etc.).

Through regular meetings in a small group, Sievert et al. [54] displayed a string of text scenarios one at a time to individual members of the group. Participants explained whether the scenario identified a civic rights violation, and if so, how to best respond to the violation. All eight participants in this study achieved mastery criteria of 90% or greater accuracy and maintained this level of performance one and three months following the end of instruction. This is a landmark study for the current study, in that the methods it used resulted in participants clearly identifying and explaining key rights. However, Sievert et al. [54] examined these rights in a more general sense and included almost no employment-specific rights or scenarios. The present study deployed a methodology like that used by Sievert et al. [54] to teach employment-related rights to three autistic adults. These rights come from Title I of the ADA and were used to identify some of the terms and situations that may be covered by the ADA.

2. Materials and Methods

2.1. Participants

To recruit participants, the researchers advertised the study using pre-existing email listservs, physical flyers, and announcements at the host university autism-centered events. Emails were sent to a wide variety of individuals, including autistic adults, family members of autistic individuals, healthcare providers, educators, and community organizations. Physical flyers (posted around the host university psychology building) detailed the anticipated length of study sessions (one hour per session), incentive for participation, and the inclusion criteria. Announcements of the study were also made at autism-centered events (including social events, research consultation meetings, and online gaming meetings).

Inclusion criteria for the study were as follows. Participants (1) were legal adults who identified as a person with autism, (2) were fluent in written and spoken English, (3) could complete the study independently (without prompts from a parent or other individual; for example, an individual who needed regular prompting to stay on-task, or who would not take steps without consulting another individual, would not be considered capable of completing the study independently), and (4) could attend to and understand the purpose of the study. This final criterion was assessed during the initial meeting by asking questions about the study following a brief introduction. Specifically, participants were required to answer the following questions: (1) “What will you be doing as a part of the study?” (correct answers including mention of the ADA or employment) and (2) “What can you do if, once you have started the study, you decide that you don’t want to participate anymore?” (correct answers including mention of “I can leave”, “I can withdraw”, or similar). All participants received a USD 40 electronic Amazon gift card and were eligible to enter a drawing for an additional USD 30 gift card at the end of study (i.e., end of maintenance). These amounts were decided based on prior research projects conducted by the first author with this population using gift cards of these amounts.

Three autistic adults were recruited for the study (see

Table 1. Participant Demographics.

Participant ID	Age (Years)	Ethnic Identity	Hispanic Cultural Origin	Gender Identity	Autism Diagnosis Received?	Autism Score (Out of 50)	Employed within the Past Year?
Christa	43	Non-Hispanic White or European American	No	Nonbinary	Yes	31	No
Nate	25	Hispanic	Yes	Male	Yes	29	No
Justin	20	Non-Hispanic White or European American	Yes	Male	Yes	24	No

for demographic information). Participant 1 (“Christa”) was a 43-year-old autistic nonbinary individual who identified as non-Hispanic white. Participant 2 (“Nate”) was a 25-year-old autistic male who identified as White Hispanic. Participant 3 (“Justin”) was a 20-year-old autistic male who identified as White Hispanic. No participant reported paid employment within the past year. This sample size was not pre-determined by power analysis but rather was a convenience sample whereby as many individuals as could be recruited in the time dedicated for the study were screened and included.

During the initial session, all participants completed the Autism Spectrum Quotient (ASQ). This is a 50-item self-rated questionnaire devised by Baron-Cohen et al. [55] to offer a quick, self-administered measure of an individual’s autism experience. Higher scores on the ASQ indicate greater self-perceived autism traits, with a mean of 35.8 for autistic respondents and a standard deviation of 6.5 and with a mean of 16.4 and standard deviation of 6.3 for non-autistic respondents. During intake, Christa received a net ASQ score of 31, within one standard deviation of the average of the autism community (0.74 standard deviations). Nate received an ASQ score of 29, just over one standard deviation below the average score for the autism community (1.05 standard deviations), and Justin received an ASQ score of 24, over one standard deviation (1.82 standard deviations) of the average score of the autism community’s original score.

2.2. Setting and Materials

Research sessions were held 1–2 times per week. Sessions were advertised to last no longer than 60 min; if a participant’s session reached 60 min (which happened once due to a mixture of internet difficulty and long participant response latency), the researcher presented the participant with the option to stop or to continue on past this limit, with sessions lasting an average of 40 min (range 30–120 min). Participants met individually with the first author to protect participant confidentiality. All sessions were scheduled to take place via Zoom and were video-recorded with participant knowledge and permission for later reference and data collection. The first author hosted all Zoom calls from private locations on the hosting university campus (various private offices), while participants typically joined meetings from their primary place of residence, such as their family home or dorm room.

During the first scheduled session, the first author gathered information from the participant using two forms in addition to reviewing the consent form. This initial session was guided by a script. The forms participants completed were the ASQ and a demographics questionnaire to gather information related to autism diagnosis, age, ethnic identity, cultural origin, gender identity, and previous employment. The study made use of Microsoft PowerPoint 2021 (version 2306) to present scenarios to participants during all baseline and training sessions. Every session meeting except for intake sessions included the use of at least one PowerPoint for testing, instruction, or both (as described in the section below). The researcher used a random number tool (a random number generator from Calculator.net, a site freely available on the internet) to randomly order test scenarios within sessions.

Test Scenarios

To follow the method of Sievert et al. [54], the first author identified a series of principles related to the ADA in an employment setting relevant to the participants of this study. The first author read Title I of the ADA in detail and requested consultation of a representative from the hosting university’s Disability Services office to identify the key principles used for this study (though this consultant did not later review the devised scenarios). The principles included two overarching categories: (1) detecting employment discrimination and (2) identifying situations where an employer could not provide the requested accommodation (see

Table 2. Principles Addressed in Training.

Discrimination-Related Principles	Inability to Accommodate-Related Principles
Refusal of Accommodation	Prolonged Absence
Retaliation	Removal of Essential Job Functions
Direct Threat	Facility Operation
Reassignment	Infringement on the Rights and Pay of Others
Interviewing	Financial Ability of the Employer
Exacerbating a Condition	Persistent Refusal of Accommodation
Timing of Disclosure	Workplace Redesign

for an itemization of these principles).

A series of text scenarios were created to use as probe items for all sessions. Sample text scenarios from Sievert et al. [54] were adapted to increase the similarity between the items in that study and the current study. Scenarios were of similar length and detail (maximum of five sentences; see Appendix A and Appendix B). Each principle included four scenarios that were derivations of one another. All scenarios described a hypothetical individual with a disability (physical and/or mental) who encountered or responded to the identified principle. The scenarios also included hypothetical employers, superiors, customers, or co-workers based on the specific scenario used. For example, in a scenario illustrating “direct threat” as part of the set of scenarios focused on “inability to accommodate”, one sub-scenario described a worker exhibiting direct threat in a manner that can be easily solved via accommodation, while the other illustrated the same worker exhibiting a direct threat that cannot be solved via reasonable accommodation. Scenarios were randomly sorted into two sets, one used for training, the other for baseline and post-training purposes. All scenarios are included in Appendix A (discrimination-related principles) and Appendix B (inability to accommodate-related principles).

A set of “testing scenarios” were presented in PowerPoint for baseline and post-training. This presentation was divided into two sections, one contained all scenarios used for “discrimination training” and the other for “inability to accommodate”. One scenario was placed per slide, and each was presented in successive order to participants all during sessions. Two introductory slides were included prior to presenting a set of test scenarios. These two slides consisted of (1) a title slide, describing the slides as either “Baseline Session” or “Post-training Session”, and (2) a slide that included two questions (i.e., “Is this a case of employment discrimination?” and “Why or why not?”) for scenarios related to employment discrimination and (“Is this a case where an employer could not accommodate an employee?” and “Why or why not?”) for scenarios related to inability to accommodate. Participants were required to respond to each test scenario. A pair of “instructional” presentations were also created, one to teach about the ADA and the employment discrimination scenarios, the other to teach about inability to accommodate. The first of these presentations (containing 44 slides) began with an overview of the ADA, as well as definitions and explanations of four key terms: “disability”, “employer”, “essential function”, and “reasonable accommodation”. The presentation then described a principle and included the two scenarios used for training purposes on each principle. This ordering was repeated until the end of the slide deck. The presentation that showed scenarios of “inability to accommodate” (containing 24 slides) began with the principles and was therefore shorter than the first. The first author followed a script during both presentations to ensure consistency in how the information was presented across all participants.

2.3. Experimental Design

This study employed a multiple-baseline-across-skillsets design to examine the effects of the training on participant performance. Due to the limited time the researchers had to complete the study, participants began with two pre-training probes. An initial performance criterion of less than 85% correct in both skillsets served as the inclusion criterion. Participants then learned about key employment-discrimination-related principles tested in baseline, rehearsed using a “training” set like testing scenarios with a variation of behavioral skills training (BST) in a single session (i.e., a mastery criterion was not required to continue to post-training probes). Participants were then exposed to an immediate post-training probe, followed by a baseline of the second skillset (i.e., inability to accommodate). Each skillset received two post-training probes and a one-month maintenance probe using the same materials described for post-training. Sessions occurred once or twice each week until study completion. Participants attended sessions on the same days each week but sometimes met on different days or at various times due to scheduling conflicts.

2.4. Independent Variable

The independent variable was discrimination training of rights violations via written and verbal instructions, modeling, practice, and feedback. This training occurred in a pre-designed sequence where each scheduled training session had a certain topic to cover. Although instructions, modeling, practice, and feedback were used for portions of the training, participants were not required to meet a predetermined mastery criterion to complete participation in the study. Session times averaged 45 min, with the shortest session (Nate’s final session) lasting 30 min and the longest (Justin’s first baseline session) lasting for 120 min. While some sessions contained more material than others, this variation was principally due to duration of participant response whereby some participants responded quickly while others took longer and asked questions during the sessions. The first author answered all participants’ questions to the best of their ability (including logistical questions (e.g., “Will we be doing this topic again next time?”), material-related questions (e.g., “Can you explain that principle again?”), and questions related to the ADA and personal experiences thereof (e.g., “So when my boss told me that I needed to come in at a certain time, was that an example of an essential function?”) and articulated if they did not know the answer when this was true. There was no firm “cut-off” time between sessions (e.g., there was not a rule that stated, “all sessions may last no longer than one hour”). However, during the use of test scenarios, participants were required to begin their replies no later than 90 s following the initial presentation of each scenario.

2.5. Dependent Variable

The primary dependent variable was participant responses to questions related to the scenarios presented. Participants answered questions (“Is this …?” and “Why or why not?”) based on each scenario presented. The first author scored the answer to both questions. Responses to the first question (“Is this …?”) were scored as a simple correct–incorrect dichotomy. Responses to the second question (“Why or why not?”) were scored based on a listing of response criteria developed for the study. The researchers elected to avoid a topographically restricted response such as a phrase that a participant must speak (e.g., “this is too expensive”). Instead, if a participant response (1) included the key information in a correct reply as described by the response criteria and (2) did not contradict that information within the same reply, the researcher scored this reply as correct. Participants were not restricted from including additional information or personal experiences (e.g., “this accommodation is too expensive; I’ve never asked for or had one that’s been more than like $50 in one sitting, this would be way more than that”), provided that the addition of this information did not contradict the key information in the answer. Appendix A and Appendix B contain each used scenario as well as correct responses to each scenario.

As participants finished a session, the first author calculated the session’s score using transcriptions and recordings from each session as needed. A final score was reported as the number of items the participant responded to correctly divided by the total available items per session.

2.6. Interobserver Agreement

An individual with experience in recording observational data was recruited to score a total of 50% of all sessions (i.e., the first, second, and fifth session for each participant) for the purpose of interobserver agreement (IOA). The second scorer viewed Zoom recordings of all sessions and received BST training from the first author before recording data for the purpose of IOA. Agreement was calculated by dividing the number of agreements by the total number of scorable instances for each skillset (instances of agreement plus disagreements) and was reported as a percentage. This produced an overall agreement of 100% for all scored sessions.

2.7. Procedure

Figure 1 contains a visualization of study proceedings. Upon entry into the study, participants first completed a consent and intake session, followed by two baseline sessions. During sessions three and four participants learned about the ADA principles, and post-training probes were conducted. This included an additional baseline test (for inability to accommodate principles in session three). The fifth session consisted of post-training probes and a social validity assessment, while the sixth session consisted entirely of maintenance probes.

2.7.1. Baseline

Baseline consisted of two to three sessions to evaluate individual performance while replying to the testing scenarios. During baseline, the first author presented one scenario selected for testing at a time to the participant and asked, “Is this a case of employment discrimination?” (if examining discrimination) or “Is this a case where an employer could not accommodate an employee?” (if examining undue hardship). After the participant provided a reply, the first author then asked “Why?” or “Why not?” During this process, the first author did not provide prompts or corrective feedback but did provide generalized and intermittent praise contingent upon participant engagement with the topics at hand (e.g., “OK”) after ending each scenario and before moving on to the next scenario. A total of 14 scenarios were presented in this fashion per category, with 28 scenarios across the two categories.

After each baseline session, the first author calculated the participant’s score, and data were evaluated for stability (defined as “within 10% of the score of the previous baseline session”) before moving to the instructional phase of the study. If this stability was not achieved, the first author planned to deploy a third baseline session before beginning instruction; however, this was not necessary for any of the participants in the study.

2.7.2. Instruction: Discrimination Training

After baseline levels were stable, participants met with the first author to review the principles related to ADA-related employment discrimination (see Table 2). The first author presented these rights in the following order using the scripted “training” PowerPoint presentation that defined key terms and modeled examples for each term resembling those in baseline and during testing. The steps of the training were as follows:

Step 1. Introduction to Topic. First, the participant was introduced to discrimination through an introduction to key terms from the ADA. These terms included (1) disability, (2) employer, (3) essential responsibility, and (4) reasonable accommodation.

Step 2. Definition of Key Term(s) of First Principle. Second, beginning with the first principle (i.e., promotion), the first author defined key terms present in the sub-principle. In this case, the first author defined “refusal of accommodation” and then discussed information regarding the connection between this principle and employment discrimination.

Step 3. Example of Interpretation of Scenario of Sub-Principle. Third, the first author read aloud an example scenario of the principle on the screen and answered and explained two questions: (1) whether the scenario did or did not indicate an instance of employment discrimination based on promotion and (2) the reason(s) why it did or did not indicate employment discrimination.

Step 4. Participant Interpretation of Scenario of Sub-Principle. Fourth, the first author provided the participant with a novel scenario describing the sub-principle. The participant was asked, “Is this scenario an example of discrimination?” The first author then either praised the participant’s answer as correct (if correct) or corrected it and asked for repetition of the correct answer (if incorrect). The first author then asked the participant “Why is this (not) an example of discrimination?” The participant’s answer was praised if correct. If the participant’s answer was incorrect, the first author first identified any correct parts of the answer, then identified the incorrect component(s). The first author then stated why this was incorrect, provided a correct response, and asked the participant to repeat the correct response. Participants continued to cycle through this model–response–feedback error correction procedure until the participant produced a correct response for each scenario presented during training.

Step 5. Repetition of Steps 2–4 for Each Remaining Sub-Principle. Fifth, once the participant described the previous scenario correctly, if there were no questions regarding the sub-principle, the first author moved on to the second principle, refusal of accommodation. This pattern continued until all seven principles were addressed and taught in this fashion.

Step 6. Immediate Post-Training Probe of Principles (Employment Discrimination). Sixth, the first author offered participants a short break (5 min) before proceeding to view the 14 scenarios for testing employment discrimination. The first author then asked in sequence for each principle whether the scenario presented an instance of discrimination and why it did or did not. The first author provided generic feedback based on engagement with the material (like baseline sessions) at this time.

Step 7. Baseline Test of Principles (Inability to Accommodate). Finally, the participants viewed the 14 scenarios illustrating the principles for inability to accommodate. Procedures for this stage were identical to those used during baseline.

2.7.3. Instruction for Inability to Accommodate Training

After baseline levels were stable, participants met with the first author to review the principles related to ADA-related inability to accommodate. These principles are shown in Table 2. The first author used a scripted PowerPoint presentation to introduce and practice these principles, using a procedure identical to that of the Employment Discrimination training. There were, however, three alterations. First, Step 1 (Introduction to topic) was significantly shorter, as there were no key terms to establish or practice with in this session. Second, Step 6 (Immediate post-training probe of principles) was a post-training probe of inability to accommodate principles rather than those of employment discrimination. Third and finally, Step 7 was an additional post-training probe of principles, this time examining the scenarios related to employment discrimination.

2.7.4. Second Post-Training Probe of Inability to Accommodate

The next session included post-training probes for inability to accommodate. The participant began by viewing 14 scenarios outlining the seven principles of inability to accommodate. In each case, the first author asked the participant whether the scenario indicated inability to accommodate or not and why this was the case. The participant viewed all 14 scenarios this way. As with previous post-training probes, the first author did not provide performance-specific feedback during this testing, only general feedback based on continued engagement during testing.

2.7.5. Social Validity

Following the second post-training probe, the first author presented the participant with an 8-item multiple-choice social validity questionnaire via a link in Zoom chat. The link led participants to a Qualtrics document (found in Appendix C). Participants were asked to provide feedback on topics covered, instructional style, timing of sessions, length of sessions, number of sessions, and self-rating of ADA as it pertained to employment. A final question was open-ended and invited participants to give additional advice or feedback to the researchers.

2.7.6. Maintenance

Participants were invited back one month following the final post-training session. Procedures for these sessions were identical to baseline and post-training sessions.

3. Results

Three participants entered and completed this study. Results show that all participants initially produced scores between 30 and 60% correct responding for both employment discrimination and inability to accommodate principles. These scores increased to approximately 80% following training. Participants scored higher on certain principles, such as timing of disclosure, both in baseline and in post-training.

3.1. Probe Performance

Results for Christa’s performance are presented in Figure 2. Christa initially scored 36% and 43% in baseline of employment discrimination scenarios. However, she scored higher in inability to accommodate baseline sessions, with scores of 46%, 54%, and 50%. After training, she scored 93% correct in her first post-training test on employment discrimination. Her second post-training resulted in a score of 86% for discrimination and 68% for inability to accommodate, while the final post-training resulted in a score of 71% on inability to accommodate. Christa was typically very engaged during instruction and frequently asked questions, made analogies, or tied a principle to real-world experience. She returned for a one-month maintenance probe and scored 89% correct in employment discrimination and 64% correct in inability to accommodate.

Results for Nate’s performance are presented in Figure 3. Nate initially scored 54% on both discrimination baseline sessions; he scored 43%, 46%, and 57% on inability to accommodate baseline sessions. Following training, his first post-training discrimination score was 68%. Afterwards, his next post-training score for discrimination scenarios fell to 54%, while his post-training inability to accommodate score rose to 84%. The final post-training probe of inability to accommodate produced another score of 84%. During the one-month maintenance probe, Nate scored 58% correct for employment discrimination and 75% correct for inability to accommodate performance. Finally, Nate returned for a one-month maintenance assessment, during which he scored 57% correct on employment discrimination scenarios and 75% on inability to accommodate scenarios. During some sessions, Nate experienced internet disruptions, which took the form of disconnection from sessions at varying points. While these interruptions were not timed, they could be as short as about 30 s (especially in later instances) or 2–3 min (the first time a disruption occurred). Nate described these disruptions when they occurred and was re-connected with a session’s Zoom link within a maximum of three minutes for each interruption that occurred.

Finally, results for Justin’s performance are presented in Figure 4. Justin scored 46% and 43% during baseline sessions on employment discrimination. He scored 32%, 36%, and 57% on the inability to accommodate scenarios. After training, Justin’s scores on employment-discrimination-related scenarios improved to 54% and 75%. His post-training probes on the inability principles resulted in scores of 82% in both post-training sessions. Like Nate, Justin also periodically experienced internet disruptions, with these disruptions taking the form of visual blurring and discoloration of the screen. These issues were typically resolved within approximately 20 s.

3.2. Item Analysis

In addition to scoring individual participant responses to questions, the first author analyzed the principles and questions asked during the study. Individual participant scores and overall averages are shown in

Table 3. Item Analysis by Participant.

Item	Christa	Nate	Justin	Average Score
Baseline
Discrimination
Refusal of accommodation	0.25	0.50	0.50	0.42
Retaliation	0.00	0.50	0.50	0.33
Direct threat	0.50	0.50	0.50	0.50
Reassignment	0.00	0.50	0.50	0.33
Interviewing	0.00	0.05	1.00	0.35
Exacerbating a condition	1.00	0.75	1.00	0.92
Timing of disclosure	0.75	1.00	1.00	0.92
Average	0.36	0.54	0.71	0.54
Explanation
Refusal of accommodation	0.50	0.25	0.00	0.25
Retaliation	0.25	0.25	0.00	0.17
Direct threat	0.50	0.00	0.50	0.33
Reassignment	0.00	0.00	0.00	0.00
Interviewing	0.00	0.00	0.00	0.00
Exacerbating a condition	1.00	0.75	1.00	0.92
Timing of disclosure	0.75	0.75	1.00	0.83
Average	0.43	0.29	0.36	0.36
Inability to accommodate
Prolonged absence	0.75	0.50	0.50	0.58
Removing essential job functions	0.75	0.50	1.00	0.75
Facility operation	0.50	0.17	0.50	0.39
Infringement	0.75	1.00	1.00	0.92
Financial ability	0.50	0.67	0.83	0.67
Persistent refusal	0.25	0.00	0.00	0.08
Workplace redesign	0.50	0.67	0.50	0.56
Average	0.57	0.50	0.62	0.56
Explanation
Prolonged absence	0.50	0.30	0.30	0.37
Removing essential job functions	0.75	0.00	0.17	0.31
Facility operation	0.50	0.00	0.00	0.17
Infringement	0.75	0.67	1.00	0.81
Financial ability	0.00	0.67	0.67	0.45
Persistent refusal	0.25	0.00	0.30	0.18
Workplace redesign	0.25	0.30	0.00	0.18
Average	0.43	0.28	0.35	0.35
Post-training
Discrimination
Refusal of accommodation	1.00	0.75	0.75	0.83
Retaliation	1.00	0.75	0.50	0.75
Direct threat	0.50	0.50	0.50	0.50
Reassignment	0.75	0.75	0.50	0.67
Interviewing	1.00	0.25	0.00	0.42
Exacerbating a condition	1.00	0.75	1.00	0.92
Timing of disclosure	1.00	1.00	1.00	1.00
Average	0.89	0.68	0.61	0.73
Explanation
Refusal of accommodation	1.00	0.75	0.75	0.83
Retaliation	1.00	0.75	0.50	0.75
Direct threat	0.75	0.25	0.25	0.42
Reassignment	0.50	0.50	0.25	0.42
Interviewing	1.00	0.25	0.50	0.58
Exacerbating a condition	1.00	0.75	1.00	0.92
Timing of disclosure	1.00	1.00	1.00	1.00
Average	0.89	0.61	0.61	0.70
Inability to accommodate
Prolonged absence	1.00	1.00	0.75	0.92
Removing essential job functions	1.00	0.50	1.00	0.83
Facility operation	0.50	1.00	0.50	0.67
Infringement	1.00	1.00	1.00	1.00
Financial ability	1.00	1.00	1.00	1.00
Persistent refusal	0.00	0.00	0.50	0.17
Workplace redesign	0.50	1.00	0.75	0.75
Average	0.71	0.79	0.79	0.76
Explanation
Prolonged absence	0.75	1.00	0.75	0.83
Removing essential job functions	1.00	0.50	0.75	0.75
Facility operation	0.50	1.00	0.50	0.67
Infringement	1.00	1.00	1.00	1.00
Financial ability	1.00	1.00	1.00	1.00
Persistent refusal	0.00	0.00	1.00	0.33
Workplace redesign	0.50	1.00	0.75	0.75
Average	0.68	0.79	0.82	0.76

. During baseline, participants accurately identified scenarios as discrimination or nondiscrimination (or as able or unable to accommodate) at the same percentage as they accurately explained their responses (t(13) = 0.14, p = 0.44), with a correct average score of 55% in identifying answers and 56% in explaining them. This similarity persisted in post-training, with an average identifying score of 74% and explanation score of 73% (t(13) = 0.43, p = 0.33).

Furthermore, when comparing “discrimination” answers to “inability to accommodate” answers, inability to accommodate scores were slightly higher (46% in the pre-training and 76% in the post-training probe, compared to 45% and 71% in the discrimination condition). The researchers did not, however, find a significant difference in either the pre-training scores (t(13) = 0.07, p = 0.47) or post-training scores (t(13) = 0.045, p = 0.33).

When the post-training probe data were analyzed, the researchers found that certain items were much more likely to be scored or missed by participants. For example, during baseline, participants averaged a score of 92% correct in identifying discrimination in scenarios related to timing of disclosure, and successfully explained it 83% of the time. Conversely, items related to interviewing were only identified correctly 35% of the time, with no correct explanations. These values increased following post-training probes, but timing of disclosure remained at a higher accuracy rate (100% for both) compared to interviewing-related items (42% and 58%, respectively).

3.3. Social Validity

Social validity responses are shown in

Table 4. Social Validity Feedback.

Participant	Christa	Nate	Justin
Style of teaching	Very well	Very well	Very well
Number of sessions	Liked number of sessions	Liked number of sessions	Liked number of sessions
Length of sessions	Liked length of sessions	Liked length of sessions	Liked length of sessions
Timing of sessions	Liked timing of sessions	Wish we had met a little more often	Liked timing of sessions
Knowledge of ADA	4/5	4/5	4/5
Knowledge of discrimination	4/5	4/5	4/5
Knowledge of inability to accommodate	4/5	4/5	5/5

Note: Specific wording and response options of each question included in Appendix C.

. When asked, Christa described satisfaction with the number, length, and timing of sessions and the teaching style. She did not leave any additional comments in the open-ended feedback section of the social validity form. Nate’s responses were like Christa’s, except that he expressed that he wished sessions were held more often. Nate left a short statement in his open-ended feedback, stating, “Very fun and informative sessions”. Finally, Justin also responded with high levels of satisfaction and left open-ended feedback: “I noticed throughout the testing on scenarios that all the scenarios would repeat, and I’d suggest that going forward to make the scenarios more interesting you could maybe randomize scenarios a bit more so that it doesn’t feel like it’s repeating”. All participants described their knowledge of each category as “a lot”, except for Justin, who rated his knowledge of inability to accommodate as “very much”.

4. Discussion

This study aimed to evaluate the use of an instructional method like that used by Sievert et al. [54] to teach members of the autism community about the ADA. Participants were exposed to discrimination training of rights violations via written and verbal instructions, modeling, practice, and feedback using a series of visually presented text scenarios. They responded to questions about the content of these scenarios and justified their responses. Results suggest that participant performance improved following training, and responses to a social validity questionnaire indicate that participants enjoyed the style of teaching, number, length, and timing of sessions. Participants also self-reported increased knowledge of the ADA including issues related to discrimination and inability to accommodate.

To our knowledge, no previous study has examined teaching ADA knowledge to individuals from the autism community. Instruction on and practice with self-advocacy behaviors may increase the rate at which an individual recognizes and responds to situations related to employment discrimination. Participants of this study may have not only become more knowledgeable of common situations related to employment discrimination but may also be in a better position to assess future work placements for these occurrences. However, the present study did not evaluate performance of self-advocacy in role-play scenarios. The results of this study partially replicate the findings of Sievert et al. [54] in several ways. First, participants in both studies began training with an existing skillset (i.e., baseline performance between 30 and 60% correct). However, Sievert et al. [54] reported participant performance increased to 90% or higher during post-training probes, while participants in the present study scored in the 65–85% range during post-training probes. This is likely due to the difference in training materials and procedure. For example, in their examination of “Personal Rights”, Sievert et al. [54] included seven principles, such as the right to marry, show physical affection, and vote. Participants of Sievert et al. [54] were required to achieve a mastery criterion (typically 2–5 consecutive correct responses to different scenarios) to advance to the next principle. While we elected to only use one practice point due to time constraints, this additional practice may help to explain the discrepancy between results of the two studies.

In further comparison to Sievert et al. [54], this study used several empirically validated methods of increasing response accuracy. These included instructor modeling, presentation of feedback, rehearsal based on researcher feedback, and use of topographically identical stimuli in different probes. These methods may have collectively produced the initial increase in post-training probe scores by participants. However, several factors also may explain the lack of a more notable increase, as well as the general drop in maintenance performance. First, the instructions were short. As compared to true BST, a mastery criterion was not enforced during instruction, though we did use a model–response–feedback error correction procedure. The researchers also allowed for significant response topography variation (a participant reply could be exceptionally long, short, personal, or objective, so long as it contained (and did not internally contradict) the correct response).

The discrepancy in accurate participant responses to certain types of scenarios is of interest. For example, participants scored an average of 80% correct to scenarios related to infringement in baseline but scored less than 30% on average to scenarios related to retaliation. This may be due to the way that certain scenarios were written. For example, for principles related to persistent refusal, participants tended to focus less on the essential or nonessential nature of the task being accommodated and more on the rejection of a suggested accommodation. Alternatively, it may be that participants had some history with certain categories of scenarios either directly or indirectly. For example, scenarios that dealt with exacerbating a condition resulted in 92% correct responses during baseline.

Regarding social validity, participants reported satisfaction with meeting once or twice a week for sessions of about an hour in length. Participants also self-rated their knowledge of target areas to be high, typically with a response of “a lot” (a score of four) when asked about their post-training knowledge. One participant (Christa) did not leave an open-ended response, another (Nate) left a short positive response, and the third (Justin) left a longer response that highlighted potential problems, most notably alluding to a rehearsal effect. These high ratings may be due to several factors including the responsiveness of scheduling according to participant availability and the presentation of materials during all sessions. Participants typically met at the same time and day each week, but these times and dates could change based on participant need, something the participants appreciated. Furthermore, the first author encouraged participants to ask questions, and to discuss points, to improve comprehension.

Finally, as noted above, this study did not teach participants how to respond to situations identified as violating rights once they were identified as discriminatory. To our knowledge, no peer-reviewed manuscript has examined this to date with members of the autism community, though based on the example of Sievert et al. [54], methods such as role play will be effective in future studies.

4.1. Limitations

There are limitations to this study. First, Justin’s comment (“I noticed throughout the testing on scenarios that all the scenarios would repeat, and I’d suggest that going forward to make the scenarios more interesting you could maybe randomize scenarios a bit more so that it doesn’t feel like it’s repeating”) highlights the potential problem of a rehearsal effect. This study re-used scenarios in baseline and post-training, and the scenarios used in training were topographically similar. Justin’s specific response points to a problem both of reduced engagement and a potential rehearsal effect. This may have formed an artifact in participant responses by including unnecessary (and potentially distracting or incorrect) details in multiple answers across sections. While the researchers re-used material to minimize the chance of item-specific effects that differed across probes, future research should aim to use varied items to address this problem.

Second, the scenarios used in the study could improve. These were modeled in length and content after scenario examples from Sievert et al. [54]. However, some scenarios may have lacked clarity or sufficient detail for participants to respond accurately. For example, one scenario testing reassignment (intended to highlight qualifications for such an event) included the phrase “denied based on his disability status” to describe the lack of an automatic, free transfer process for a worker with a disability. Some participant replies focused on this phrase and interpreted it as discriminatory. The item analysis also highlighted several scenarios that were particularly difficult or easy, which could be a good axis for future examination and scenario refinement.

Third, the first author met with participants on an individual basis. This was typically not a problem, and participants met weekly or semiweekly as needed. However, participants did sometimes miss sessions due to double-booking, illness, travel, or simple forgetting, which introduced increased latency between sessions as a potential confounding variable. The researchers have insufficient data to decisively identify any effects of this at the time, but more regular meetings with participants would remove this potentially confounding variable.

Fourth, this study was conducted via Zoom to increase accessibility and flexibility for participants, but it meant that each session had a reduced environmental control compared to what might be feasible with in-person sessions, especially given that participants often joined from locations they identified as their “house”, “apartment”, or “home”. For example, Christa periodically experienced interruptions from a pet, while Nate and Justin experienced session disruptions from their internet usage. The effects of this reduced environmental control are not clear.

Fifth, due to time constraints for the study, sessions were limited to baseline, training, and a maximum of two post-training probes for each response category. The small number of data points per condition limit the extent to which experimental control is demonstrated and therefore limit the extent to which beneficial changes can be attributed to the intervention. Although the effects were replicated across three participants and two response categories (i.e., discrimination-related principles and inability to accommodate-related principles), future research should extend post-training probes to demonstrate a higher level of experimental control.

Sixth, this study did not make use of Wolf’s [56] recommendations related to social importance of the effects of the intervention. The feedback acquired from participants via the social validity questionnaire developed for this study is promising, but additional questions explicitly related to their perceptions of the effectiveness of the training, as well as consultation with other stakeholders (e.g., varying autism community members, clinicians, and caregivers), will be beneficial in future studies.

Finally, this study was a preliminary step in addressing ADA knowledge in the workplace and as such was not a full replication of Sievert et al. [54]. Sievert et al. [54] made use of a second component in which participants role-played responses to legal rights violations, including assertive speech and communicating with individuals based on the success (or ineffectiveness) of communicating with others (e.g., if a doctor would not respect their legal right, participants might go to an office manager or review board). This study only aimed to imitate the first component of Sievert et al. [54], and as such instructional methods such as role play were not used. However, we encourage future researchers to use this method.

4.2. Future Directions

Despite the noted limitations, to the researchers’ knowledge, this is the first study to examine teaching ADA knowledge to autistic adults. There is accordingly ample room for future studies to investigate and iterate on the methods used herein. The researchers suggest potential avenues for future research.

First, future research should aim to recruit a wider demographic group, particularly regarding age, race, ethnicity, and neurodiversity. The participants of this study were all White, autistic, and English-speaking and could maintain focus for prolonged periods of time. Future research should measure the presence of other psychological diagnoses and compare autistic participants with one or more additional diagnoses, autistic participants without an additional diagnosis, and non-autistic participants.

Second, future research should re-examine these topics using different scenarios and use more than one scenario during practice sessions. It is possible that the elevated levels of performance by participants in Sievert et al. [54] was due to this additional practice with different scenarios, rather than minor topographical iterations like those used in the present study. This change may also increase generality and maintenance of the knowledge acquired during training.

Third, in Sievert et al. [54], participants not only verbally identified rights and principles but practiced advocating during role-play sessions. The researchers in Sievert et al. [54] provided a model for how to articulate a self-advocacy request and what to do if that request was met with resistance. Future research on this topic may benefit from the same proceedings. Participants in these future studies might learn how to phrase an accommodation request via email or how to disclose a diagnosis to justify an accommodation request. Future research could evaluate practice with this skill in different modes, including in writing, over digital communication, and face-to-face.

Fourth, it may be interesting to modify the questions that are asked of participants following each training scenario and, in turn, participant responses. Open-ended questions were included in this study to follow the example provided by Sievert et al. [54]; however, it is possible that close-ended questions can lead to better maintenance and generalization of the skills acquired. For example, rather than asking “Why or why not?”, an experimenter could ask a series of specific questions, such as “Is this a case of hiring discrimination?” and “What will the employee likely lose as a result of this scenario?” This may lengthen the duration of sessions but may be effective in training participants to systematically consider diverse types of employment discrimination. This may also have the added benefit of a lessened need for qualitative coding of open-ended questions.

Fifth, future research may wish to examine in-person instruction (as opposed to over Zoom), the effects of scenario refinement, scheduling training sessions more frequently, and scaling up the intervention by recruiting larger sample sizes and conducting the training in a small group format.

5. Conclusions

In conclusion, this study sets a foundation for teaching employment-related ADA knowledge to members of the autism community. Three autistic participants demonstrated slightly increased knowledge of detecting and explaining ADA violations following training. This study is preliminary, and future research should aim to address the limitations of the study to strengthen these findings.