Understanding Disability Biases in Undergraduate Rehabilitation Students: An Exploratory Study

Abstract

Shaping and cultivating positive attitudes toward people with disabilities is an important task for all programs that prepare future rehabilitation practitioners. So too, is identifying potentially problematic attitudes or biases about ability status. With the growth of undergraduate rehabilitation programs, it is imperative for educators to understand the factors that may influence students’ biases about people with differing ability status. Many training programs address explicit biases (e.g., those measured via self report), but overlook the influence and existence of implicit biases among students. Furthermore, many trainings focus on stimulating awareness of potential bias, rather than diving into causes and impacts. In order to develop effective curricula for mitigating bias and training infused with social justice, it is critical to understand this phenomena. Using a quantiative design, the current study examines explicit and implicit biases of students in undergraduate rehabilitation programs at three institutions across the United States. Results support that explicit and implicit biases are two separate constructs and should be treated as such (i.e., educational activities about bias are not sufficient). In addition, the results were indicative of the importance of addressing biases as multidimensional, and the potential utility of contact experiences as a factor for mitigating bias.

Keywords

attitudes toward disability rehabilitation education social barriers professionalism/professional issues

Rehabilitation practitioners, as frontline service providers for people with disabilities, are equipped with skills and knowledge to help clients improve their quality of life and achieve independence. It is particularly important for these practitioners to be aware of their biases toward individuals with disabilities lest they risk perpetuating the status quo of systemic inequality and institutionalized oppression (Kelsey & Smart, 2012). Master’s level rehabilitation counselor education is “quality controlled” via individual and programmatic accountability measures represented by the Council for Accreditation of Counseling and Related Educational Programs (CACREP, 2016), and promotes the Code of Professional Ethics for Rehabilitation Counselors (Commission on Rehabilitation Counseling Certification, 2016) and American Counseling Association (2014) Code of Ethics. These accountability measures, accreditation procedures and mandated code of ethics, serve the function of gatekeeping: protecting the profession and future clients (Brear & Dorrian, 2010).

Unfortunately, such accountability measures described above do not exist in undergraduate rehabilitation programs. Although these undergraduate programs have existed for decades, in recent years, there has been a movement toward undergraduate rehabilitation education (Evenson & Holloway, 2007; Herbert et al., 2010). In reviewing posted curricula of programs on the CACREP (2019) registry of undergraduate programs, it is clear that these programs are similar in content, even though the program titles vary (e.g., Rehabilitation and Human Services, Rehabilitation and Disability Studies, and Rehabilitation Services). Although efforts have been made to identify the biases of undergraduate rehabilitation students (e.g., Hunt & Hunt, 2000), there is a dearth of research that explores the biases of this group.

Biases About Disability

The relationship between attitude and behavior has been extensively studied in the psychology literature (Ajzen & Fishbein, 1977; Boysen & Vogel, 2008; Livneh, 1982; Lord et al., 1991; Todd et al., 2011), and is particularly challenging to measure (Antonak & Livneh, 2000). Defining attitudes has historically been controversial; however, it is clear that conceptually, attitudes encompass the underlying psychological phenomena of biases given that biases are learned, charged with emotion, and impact behavior (Chubon, 1992; Fishbein & Ajzen, 1975). In seeking to better understand this phenomenon, Shih et al. (2013) discuss that stereotypes can be broken down into explicit biases and implicit biases. Although explicit bias may be amended by learning (Shih et al., 2013), they are susceptible to social desirability (Stier & Hinshaw, 2007) and may not reflect an individual’s true inner feelings. On the other hand, implicit bias is developed through experience (Gawronski & Bodenhausen, 2006) and operates outside of one’s conscious level (Devine, 1989). As such, although more difficult to access, implicit bias is a missing piece to demystifying discriminatory behaviors.

Implicit Biases

The pervasiveness of implicit, or unconscious, biases is well-documented (Antonak & Livneh, 2000; Devine, 1989; Nosek et al., 2007). Implicit biases lie below the conscious mind and are curated from early childhood to reflect social stereotypes that individuals rarely would claim to perpetuate (Baron & Banaji, 2006; Chapman et al., 2013; Rutland & Killen, 2015). A suggested cause for the intangible nature of implicit biases is that one’s lack of awareness of the bias leads it to remain intact and unchecked (Devine, 1989). Stated more directly, individuals are only concerned about their implicit biases after they have been made aware of their existence (Devine et al., 2012). This can be visualized as an individual who is opposed to ableism, but does not recognize that they are complicit in maintaining an ableist society and its norms.

Implicit bias and disability

A commonly held notion is that the passage of the Americans With Disabilities Act (ADA) in 1990 alleviated discrimination based on ability status (Thomas et al., 2014). The goal of such legislation is to encourage integration of people with disabilities into society, and to provide them with the same access to experiences that people without disabilities are privileged to. Where the ADA falls short is in its focus on distributive justice, and lack of concern with social or institutional reframing of the treatment of people disabilities in a holistic way. Attempts at distributive justice, such as the ADA, emphasize the allocation of material goods and positions (i.e., jobs, benefits), while ignoring the social contexts which contribute to inequity and institutional oppression (Young, 1990).

This distributive approach to legislation (including the more recent Workforce Innovation and Opportunity Act [WIOA], 2014) never gets at the foundation for discriminatory behavior: problematic internalized biases which remain unchallenged in perpetuating hegemonic paradigms in society. A byproduct is that students who are being trained to work with individuals with disabilities have likely internalized a significant amount of ableist and/or negative biases about people with disabilities. For example, consider the student disability services on a university’s campus are separate from other student services such as registration, food courts, or libraries generally housed at the heart of campus. Recruitment material rarely includes representations of students with disabilities, and it is not uncommon for historic buildings to have non-ADA compliant spaces, excluding students, staff, and faculty entirely from certain areas (Imrie & Kumar, 1998; Pearson & Samura, 2017).

Empathy and bias

Empathy, or the ability to understand the feelings of another, has been explored as an effective tool for mitigating implicit biases (Dolby, 2014; Shih et al., 2013; Todd et al., 2011). Researchers have indicated that there has been a decrease in empathy in college-aged students over time (Konrath et al., 2011); however, there is little research on the relationship between declining empathy in college-aged students and levels of bias. Given the implication that implicit biases may be the underlying fuel propelling discriminatory behaviors and intergroup conflict (Devine, 1989; Greenwald et al., 2009; Hu et al., 2015; Stanley et al., 2011), the belief that declining empathy in college-aged students requires immediate attention.

As students advance through higher education programs with the goal of working with individuals with disabilities, it is imperative that researchers and educators understand what biases they have, and actively work to encourage student reflection to, at a minimum, identify bias. The all too often used comments of “I don’t see color” or “I was raised not to view people differently” need to be challenged and identified as problematic (DiAngelo, 2011). It is also important to identify what, if any, relationship may exist between levels of empathy and levels of bias. As a starting point in this conversation, this study seeks to reveal more information about biases (explicit and implicit) around disabilities in undergraduate rehabilitation students. To better understand the phenomenon, students from three undergraduate rehabilitation programs located in East, South, and West regions of the United States were recruited. Specific research questions of this study include the following:

Research Question 1: What are the disability-related explicit and implicit biases of undergraduate rehabilitation students?

Research Question 2: What is the relationship between explicit and implicit biases?

Research Question 3: How do the students’ empathic self-efficacy and contact experiences predict the disability-related explicit and implicit biases?

Research Question 4: Are there group differences among students enrolled in three different university programs?

Method

Participants and Procedures

Participants for this study consisted of undergraduate students in a rehabilitation major (i.e., Rehabilitation and Disability Studies, Rehabilitation and Human Services, and Rehabilitation Services) from three different institutions from the East, South, and West areas of the United States. A total of 188 students completed the paper-and-pencil survey. Among them, 13 students did not complete a demographic questionnaire (n = 175). The sample primarily identified as White, female, and at the academic rank of Senior (see Table 1). No students reported having a nonbinary gender identity; one student reported a transgender identity; and 4% identified as a member of the Lesbian, Gay, Bisexual, Transgender, and Queer+ (LGBTQ+) community (n = 7); 79% identified as not being a member of the LGBTQ+ community (n = 136); and 17% identified as not a member of the LGBTQ+ community but do identify as an ally (n = 29). In addition, 31% of students identified as being a first-generation college student (n = 54).

Table 1.

Participant Demographics.

Individual characteristics	n	%
Gender^a
Male	23	13.1
Female	152	86.9
Race/ethnicity^b
White	118	67.8
African American	7	4.0
Asian	6	3.4
Other	7	4.0
Hispanic or LatinX	36	20.7
Academic year^a
Freshman	7	4.0
Sophomore	26	14.9
Junior	66	37.7
Senior	76	43.4
Type of experience with persons with disabilities^c
Work	102	60.0
Personal friendship	99	56.6
Family	83	47.4
Recreation	84	48.0
Other	24	13.7

Missing n = 13 for gender and academic year. ^bMissing n = 14 for race/ethnicity. ^cParticipants are allowed to provide more than one answer.

The data were collected during Fall 2018. After receiving institutional review board (IRB) approval from all institutions, course instructors who are teaching major courses were contacted for permission to recruit study participants during their class time. Following the agreement of the instructors, the researchers visited the classes to invite students to participate in the study. For one institution that was not possible for the researchers to visit, the survey packets with detailed instructions of the administration were mailed to the course instructors. The same instructions were used by the researchers who directly visited the classes to collect the data. A paper-and-pencil survey packet was given to students who were willing to participate.

Instruments

For the regression analyses, predictor variables used in this study were students’ perceived empathic self-efficacy and their contact experiences with individuals with disabilities, and the outcome variables were explicit and implicit biases toward individuals with disabilities. To measure such variables, this study used the following questionnaires: (a) demographic information, (b) Attitudes toward Disabled Person Scale–Form O (ATDP-O; Yuker et al., 1970), (c) Multidimensional Attitude Scale (MAS; Findler et al., 2007), (d) Disability Attitude Implicit Association Test (DA-IAT; Pruett & Chan, 2006), (e) Perceived Empathic Self-Efficacy Scale (PESE; Di Giunta et al., 2010), and (f) Contact with Disabled Persons Scale (CDP; Yuker & Hurley, 1987).

Demographic information

Participants’ demographic information, such as age, gender, race/ethnicity, academic year, religious preference, being a first-generation college student, and the nature of experiences with individuals with disabilities, were collected to describe study participants.

ATDP-O

The ATDP-O is a unidimensional measure examining attitudes toward people with disabilities (Yuker et al., 1970), which was used to measure undergraduate students’ explicit biases. It consists of 20 items rated on a 6-point Likert-type scale (3 = I agree very much, −3 = I disagree very much), with higher scores indicating more positive attitudes toward individuals with disabilities. Yuker et al. reported test–retest reliability ranging from .66 to .89 for this form. Cronbach’s alpha coefficient in this study was .73.

MAS

Another measure used to examine explicit biases toward people with disabilities is the MAS, which assesses three dimensions of affect, cognition, and behavior (Findler et al., 2007). Participants are presented with a social scenario vignette describing an interaction between “Joseph/Michelle” and a person who uses a wheelchair, and then asked to respond to the 34 items on a 5-point Likert-type scale ranging from 1 (not at all) to 5 (very much). Higher scores indicate more negative attitudes toward individuals with disabilities. Findler et al. reported reliabilities of three components ranging from .83 to .90. Cronbach’s alpha in this study was following: .83 for affect, .89 for cognition, and .81 for behavior subscales.

DA-IAT

Adapting from implicit association test assessing racial attitudes, Pruett and Chan (2006) developed the DA-IAT which examines implicit attitudes toward people with disabilities. This paper-based DA-IAT consists of two sets of practice and actual tests (only the actual test set is used to calculate a score), based on the formula identified by previous studies (Lemm et al., 2008; Pruett & Chan, 2006). Each set includes one instruction page and two practice or test pages. The time given to respondents was 20 s per page, consistent with the administration time used by Pruett and Chan (2006). The practice test is given to help respondents be familiar with testing by asking them to categorize insect–flower or good–bad words properly. For the actual test, respondents are asked to categorize disability–nondisabled symbols or good–bad words as much as possible. Symbols and words are presented within a block where disability symbols are paired with good words and nondisabled symbols are paired with bad words (i.e., incongruent pairs), or vice versa (i.e., congruent pairs). It was assumed that an individual would find difficulty associating disability symbols with positive words if he or she has negative attitudes toward people with disabilities (Pruett & Chan, 2006). The test–retest reliability identified by Pruett and Chan study was .78.

PESE

The PESE is a 6-item questionnaire developed to examine individuals’ self-efficacy beliefs regarding their empathic responding to other people’s needs (Di Giunta et al., 2010). Participants are asked to respond on a 5-point Likert-type scale (1 = not well at all, 5 = very well). Di Guinta et al. identified internal consistency estimates of the PESE ranging from .76 to .80 for the U.S. samples. The Cronbach alpha coefficient in this study was .86.

CDP

The CDP is a unidimensional instrument developed to measure the amount of contact people have had with individuals with disabilities (Yuker & Hurley, 1987). It consists of 20 items on a 5-point Likert-type scale ranging from 1 = never to 5 = very often. Higher scores indicate more frequent contact experiences with individuals with disabilities. Collecting from five different samples, Yuker and Hurley (1987) reported median split-half coefficients of .87 and median coefficient alpha estimates of .92. In this study, Cronbach’s alpha of .93 was found on this scale.

Data Analyses

Once completed surveys were collected, the data were entered into an Excel file for data cleaning and were then imported to the IBM Statistics Package for the Social Sciences (SPSS) version 25.0 for analyses. Descriptive statistics were used to summarize demographic information and biases of the participants. Two-tailed Pearson correlation analyses were used to examine the relationships between explicit and implicit biases. Also, multiple regression analyses were conducted to examine the associations of students’ empathic self-efficacy and contact experiences on the explicit and implicit biases. Finally, an analysis of variance (ANOVA) and a multivariate analysis of variance (MANOVA) were used to identify differences on biases among gender, academic year, and undergraduate rehabilitation programs.

Results

Of the 188 participants in this study, 185 completed the ATDP with a M = 25.95 (SD = 12.12), 182 completed the MAS Affect with a M = 47.58 (SD = 10.11), 180 completed MAS Cognition with a M = 20.41 (SD = 6.27) and MAS Behavior with a M = 22.45 (SD = 5.77), 183 completed the PESE with a M = 23.98 (SD = 4.46), 183 completed CDP with a M = 51.73 (SD = 15.36), and 182 completed the DA-IAT with a M = 0.03 (SD = 8.02). To explore the correlations between explicit and implicit biases, multiple two-tailed Pearson correlation analyses were conducted. Table 2 presents zero-order correlations among data collected by multiple instruments. Specifically, the measure of implicit bias (i.e., DA-IAT) is not significantly correlated with explicit biases (i.e., ATDP and MAS Cognition) but is significantly correlated with other two explicit biases (i.e., MAS Affect and MAS Behavior). In addition, DA-IAT is not significantly correlated with empathic self-efficacy or contact with people with disabilities.

Table 2.

Zero-Order Correlations Between Explicit and Implicit Biases, Self-Efficacy, Contact With People With Disabilities, and Social Desirability.

Measure	ATDP	MAS Affect	MAS Cognition	MAS Behavior	PESE	CDP	DA-IAT
ATDP	1
MAS Affect	−.15*	1
MAS Cognition	−.16*	.45**	1
MAS Behavior	−.14	.50**	.35**	1
PESE	.27**	−.01	−.20**	−.08	1
CDP	.29**	.00	−.13	−.07	.26*	1
DA-IAT	.08	−.15*	−.02	−.16*	−.10	−.11	1

Note. ATDP = Attitudes Toward Disabled Person Scale; MAS = Multidimensional Attitudes Scale Toward Persons with Disabilities; PESE = Perceived Empathic Self-Efficacy Scale; CDP = Contact with Disabled Persons Scale; DA-IAT = Disability Attitude Implicit Association Test.

p < .05. **p < .01.

Multiple regression analyses were conducted to examine the associations between PESE, CDP, and explicit/implicit biases. A total of five models were analyzed fixing PESE and CDP as predictor variables and rotating explicit and implicit biases as outcome variables. The first model (PESE-CDP on ATDP) yielded a statistically significant result with an F(2, 180) = 12.89 (p = .00), and an R² = .126; both PESE (b = .57, t = 2.92, p = .00) and CDP (b = .19, t = 3.31, p = .00) were found to be statistically significant predictors for the model. The second model (PESE-CDP on MAS Affect) was not found significant with an F(2, 178) = .00, p > .05. The third model (PESE-CDP on MAS Cognition) yielded a statistically significant result with an F(2, 176) = 4.40 (p = .01), and an R² = .04; PESE (b = −.25, t = −2.37, p = .02) was found statistically significant whereas CDP (b = −.035, t = −1.14, p > .05) was not. The fourth model (PESE-CDP on MAS Behavior) was not found significant with an F(2, 177) = .88, p > .05. The fifth model (PESE-CDP on DA-IAT) was not found significant with an F(2, 175) = 1.58, p > .05.

Finally, three analyses of variance (i.e., two analysis of variance [ANOVA] and one multivariate analysis of variance [MANOVA]) were performed to explore between groups differences on the explicit and implicit biases construct. Specifically, a three-way between-groups design (three schools × two genders × four academic levels) was used as fixed factors for the analyses. The unidimensional explicit bias (i.e., ATDP) and implicit bias (i.e., DA-IAT) were used as the dependent variables (DV) for each of the two ANOVAs while three MAS subscales were used as the DV for the MANOVA. For the purpose of anonymity, instead of listing the actual school names, three schools are labeled as school #1, #2, and #3 in the results. All three models yielded nonsignificant results on the homogeneity test, indicating the appropriateness for the subsequent analyses (Field, 2005). As a result, a statistically significant between groups difference was found in the ANOVA-ATDP model with an F = 5.44, p = .00, R² = .16. This finding has a small effect size, η² = .136 (Cohen, 1988). However, the only statistically significant between-group factor was the schools. A post hoc test was performed to explore the difference in detail and can be found in Table 3.

Table 3.

Post Hoc Comparison Under Bonferroni Correction on ATDP.

Schools	Mean difference	SE	p
School #1–School #2	11.60	2.50	.000
School #1–School #3	9.66	2.05	.000
School #2–School #3	−1.94	2.18	1.000

Note. ATDP = Attitudes Toward Disabled Person Scale.

The ANOVA–DA-IAT was not found as significant with an F = 1.508, p = .18, R² = .05. No further post hoc was performed for this model. Finally, the only statistically significant between-group difference in the MANOVA-MAS model was the school factor with a Wilks’ Lambda = .90, F(6, 324) = 3.01, p = .01. Both gender, Wilks’ Lambda = .97, F(3, 162) = 1.91, p = .13, and academic level, Wilks’ Lambda = .94, F(9, 394) = 1.22, p = .28, were not significant. Subsequent multiple ANOVAs were performed to explore detailed differences in the school factor. Among three MAS constructs, two (MAS Cognition and MAS Behavior) yielded statistically significant results. Specifically, MAS Affect yielded an F = .85, p = .43; MAS Cognition yielded an F = 3.88, p = .02; and MAS Behavior yielded an F = 3.56, p = .03. For more detailed differences on the MAS Cognition and MAS Behavior among three schools refer to Table 4.

Table 4.

Post Hoc Comparison on MAS Cognition and MAS Behavior.

Scale	Schools	Mean difference	SE	p
MAS Cognition	School #1–School #2	−3.0575	1.36395	.079
	School #1–School #3	−2.5056	1.11534	.078
	School #2–School #3	0.5519	1.18694	1.000
MAS Behavior	School #1–School #2	−2.9270	1.30136	.077
	School #1–School #3	−1.9934	1.06416	.188
	School #2–School #3	0.9336	1.13248	1.000

Note. Bonferroni correction is used. MAS = Multidimensional Attitudes Scale Toward Persons with Disabilities.

Following the above finding, a one-way ANOVA was conducted, using CDP scores as the DV, and school as the independent variable (IV). The results of this ANOVA indicated that school had a statistically significant effect on CDP scores, F(2, 180) = 13.306, p = .00. This finding has a small effect, η² = .129 (Cohen, 1988). Post hoc results indicate that CDP scores at School 1 (M = 61.50, SD = 13.89) were significantly different from CDP scores at School 2 (M = 48.62, SD = 14.70) and School 3 (M = 48.64, SD = 14.54).

Discussion

This study sheds light onto many facets of biases in undergraduate students and is a strong starting point for future research on this topic. Previous literature, and this paper, supports the notion that explicit and implicit biases are not correlated (Lu et al., 2018; Pruett & Chan, 2006). Results of this study support this; however, it should be noted that two subscales of the MAS, Affect and Behavior, were significantly correlated with the DA-IAT. Few other studies assessing implicit biases specifically about disability have been found to use a multidimensional approach to explicit bias; most commonly, the ATDP is the preferred instrument (Antonak & Livneh, 2000). Results of this study indicate that, at multiple points, certain aspects of the MAS were found to be significant while others were not.

Implicit bias was also found as not significantly correlated with levels of perceived empathic self-efficacy or level of contact with people with disabilities, contradictory to previous research and hypotheses related empathy to implicit bias (Shih et al., 2013). However, empathic self-efficacy was significantly, positively correlated with scores on the CDP, indicating that more contact with people with disabilities related to higher levels of empathic self-efficacy. Similarly, Barr and Bracchitta (2015) indicated that more contact experiences with individuals with disabilities were significantly related to more positive attitudes toward people with disabilities, due to a more accurate understanding and viewpoint about the experiences of people with disabilities. Interestingly, perceived empathic self-efficacy and gender were significantly correlated with MAS cognition, indicating that participants who identified as male had more bias in the cognitive dimension, and less self-reported empathic self-efficacy. These findings are consistent with previous study results. For example, Findler et al. (2007) indicated that female participants reported more positive behavioral intention than male participants, and men with low self-esteem displayed more negative cognition toward individuals with disabilities compared with those with high self-esteem. Other studies also confirmed that women had more positive attitudes toward individuals with disabilities than their male counterparts (Goreczny et al., 2011; Vilchinsky et al., 2010). Contrary to the results of this study, however, Di Giunta et al. (2010) found no significant gender difference on the level of empathic self-efficacy.

It was hypothesized that students’ perceived empathic self-efficacy as well as contact experiences would account for a significant amount of variance in the explicit and implicit biases. Results demonstrate that empathic self-efficacy and previous contact with people with disabilities did not account for a significant amount of the variance in levels of implicit bias, or explicit bias as measured by MAS affect and behavior. On the other hand, empathic self-efficacy does appear to account for a significant amount of the variation in MAS cognition, and both CDP and PESE were significant predictors of scores on the ATDP. These findings again indicate that implicit biases are multifaceted, and it is imperative to approach bias mitigation at multiple levels and over time, as demonstrated by Devine et al. (2012).

Results of the study indicate that between-school differences should be given further attention. School 1 was found to have significant differences on the ATDP (i.e., demonstrating less bias). A follow-up analysis was done to explore possible causes for the differences on ATDP. The follow-up revealed that students at School 1 reported more contact experiences with people with disabilities as compared with the other two universities (e.g., had higher scores on the CDP). This finding supports previous research and hypotheses that contact with people with disabilities may mitigate explicit bias (Barr & Bracchitta, 2015; Hein et al., 2011). However, MAS cognition and behavior scores were statistically significantly different in the ANOVA, but head-to-head comparisons in post hoc testing did not show any significant differences between schools. These contrasting results demonstrate the difficulty of measuring biases, but underscore the importance of a multidimensional approach.

Limitations

There are several limitations in the study that should be taken into consideration when interpreting the study results. First, although participants are recruited in various regions in the United States, the sample was only collected from three universities and therefore the results cannot be generalized to the entirety of undergraduate students in rehabilitation programs. Second, given that the majority of the study participants were female and at a higher academic level, gender and academic level groups were not evenly distributed. Therefore, the results examining group differences should be examined with caution. Third, although the paper-based IATs have been considered a useful measure of examining implicit belief and attitude, the paper-based DA-IAT used in this study might have less accuracy compared with the computer-based DA-IAT (Pruett & Chan, 2006). Fourth, this study utilized the ATDP-O and MAS as measures of explicit biases, which only examine attitudes toward individuals with disabilities in general or those who are using a wheelchair. Given the fact that attitudes and biases will vary across different disability conditions, the study findings may not properly explain explicit biases toward people who have different characteristics and conditions depending on their disability type (e.g., individuals with intersectional identities). Finally, this study used self-report survey conducted in a participants’ classroom. The survey format and the environment where the study was conducted might have influenced students’ responses in a socially desirable way.

Implications

Research

This study has significant applications for research in rehabilitation counselor education and undergraduate education. Biases are a phenomena that require much further study, including increasing our understanding of the link between biases and behavior; especially in all professionals who will work with people with disabilities in the future. The relationship between perceived empathic self-efficacy and biases is an area that needs to be at the forefront of future research. This study is the first to explore any relationship between perceived empathic self-efficacy and bias. The significant correlation found between scores on the PESE and MAS Cognition scale warrants further exploration. Studies which replicate this one will be beneficial for identifying the connection between the MAS Cognition and PESE constructs. Research can also provide additional information on the relationships between implicit/explicit biases and other psychosocial or experience variables (e.g., working alliance, clinical judgments, and successful counseling outcomes). The consistent finding that there is no relationship between explicit and implicit biases is of particular importance for the development of effective interventions for mitigating biases. Furthermore, improvements in the measurement of attitudes are greatly needed. The DA-IAT presently only exists in a paper–pencil format. The nature of the paper–pencil format of an IAT make traditional reliability measures (e.g., Cronbach’s alpha) near impossible as the measure does not have a predefined number of items; participant scores are partially based on completing items quickly (Bardin et al., 2016). A validated electronic version of this measure will allow for reliability testing, testing with more participants, and increased accessibility. Furthermore, it would be beneficial for further exploration of biases to update and revise the MAS and the ATDP with more current language (e.g., person first), and to include a variety of disability characteristics. Multiple iterations of the instruments may aid in clarifying biases about specific groups of people with disabilities.

It is imperative to emphasize that working to mitigate implicit biases for one group needs to include any intersectional identities of that group. Intersectionality of multiple marginalized identities can lead to the compounding of discriminatory behavior aimed at a particular individual. Rehabilitation research must be more intentional in the exploration of experiences of people with disabilities who have additional marginalized identities. This extends to this study and its future iterations as well. In addition, educators and faculty must be assessed in regard to their own levels of bias about disability and individuals with intersectional identities. All people immersed in society are prone to have implicit biases; this must be a holistic examination, not simply a top-down directive for interventions directed at students. Although pretesting and posttesting using curricular interventions will shed light into what will be most effective for most students, if educators are not actively engaged with this process, it will not be adopted in the broad ways that are needed.

Education

The recent passage of the WIOA (2014) has decreased the educational requirement of vocational rehabilitation (VR) service providers in the state–federal VR programs to a bachelor’s degree, emphasizing the significance of providing quality education and training to undergraduate students. Up until the reauthorization of WIOA, vocational rehabilitation counselors were expected to have specialized training for working with individuals with disabilities via a master’s degree in rehabilitation counseling and eligibility for the CRC credential. With the advent of undergraduate rehabilitation programs, it is necessary to better understand the needs of this new group of trainees. Training students to work with individuals with disabilities should require facilitating the knowledge, skills, and attitudes needed to work with individuals with disabilities, but also to advocate and challenge discriminatory behavior toward this population. Advocacy and social justice are inherent in the profession of rehabilitation counseling, but educators must remain vigilant in problematizing dominant narratives and student biases. Batson and Ahmad (2009) described that engaging in a cognitive process of self and other perspectives taking can be significantly related to increasing empathic concern, which may eventually lead to reduced stereotyping and displaying more positive attitudes about out-group members. Therefore, self-reflection journaling activities, presenting intentionally challenging case scenarios, and providing feedback and resources are steps that can be taken toward addressing some levels of bias in one’s classroom.

In addition, undergraduate programs should liaise with community organizations to facilitate student experiences in working with people with disabilities at various times throughout their time in the training program via volunteer opportunities, internships, or entry-level jobs. Given the findings that males had less favorable attitudes toward people with disabilities, increased contact with individuals who have various types of disabilities would a have significant influence on changing biases among male students (Goreczny et al., 2011). Finally, paying specific attention to the types of voices being shared through course materials and videos is an area requiring much attention of educators. Showing videos of people with disabilities only insofar as they are used to describe that person’s disability and its symptoms do not aid students in understanding a perspective of being a person with a disability.

Conclusion

Educators and leaders in the field must recognize that decreasing bias is more than an interesting research topic. There is an ever-growing divisiveness within the United States at this time, and our students are growing up with a front-row seat to hateful, discriminatory, and frightening news headlines. Developing implicit biases is a natural aspect of the human experience, and it is our responsibility to ensure that our students are challenged to insightfully reflect on the biases that they might have. This study is a step toward uncovering more information about implicit biases in general, and specifically those of undergraduate students in rehabilitation education. Of note are the implications for further evaluation of multidimensional components of implicit bias, and their potential influence on behavior.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Ajzen

Fishbein

(1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84(5), 888–918. https://doi.org/10.1037/0033-2909.84.5.888

American Counseling Association. (2014). ACA code of ethics. https://www.counseling.org/resources/aca-code-of-ethics.pdf

Antonak

R. F.

Livneh

(2000). Measurement of attitudes towards persons with disabilities. Disability and Rehabilitation, 22(5), 211–224.

Bardin

Perrissol

Fos

Souchon

(2016). Testing of a paper-and-pencil Personalized Single Category Implicit Association Test (SC-IAT-P). International Review of Social Psychology, 29(1), 31–44. http://doi.org/10.5334/irsp.35

Baron

A. S.

Banaji

M. R.

(2006). The development of implicit attitudes: Evidence of race evaluations from ages 6 and 10 and adulthood. Psychological Science, 17(1), 53–58.

Barr

J. J.

Bracchitta

(2015). Attitudes toward individuals with disabilities: The effects of contact with different disability types. Current Psychology, 34(2), 223–238. https://doi.org/10.1007/s12144-014-9253-2

Batson

C. D.

Ahmad

N. Y.

(2009). Using empathy to improve intergroup attitudes and relations. Social Issues and Policy Review, 3(1), 141–177. https://doi.org/10.1111/j.1751-2409.2009.01013.x

Boysen

G. A.

Vogel

D. L.

(2008). The relationship between level of training, implicit bias, and multicultural competency among counselor trainees. Training and Education in Professional Psychology, 2(2), 103–110. https://doi.org/10.1037/1931-3918.2.2.103

Brear

Dorrian

(2010). Gatekeeping or gate slippage? A national survey of counseling educators in Australian undergraduate and postgraduate academic training programs. Training and Education in Professional Psychology, 4(4), 264–273. https://doi.org/10.1037/a0020714

10.

Chapman

E. N.

Kaatz

Carnes

(2013). Physicians and implicit bias: How doctors may perpetuate health care disparities. Journal of General Internal Medicine, 28(11), 1504–1510.

11.

Chubon

R.A.

(1992). Attitudes toward disability: Addressing fundamentais of attitude theory and research in rehabilitation education. Rehabilitation Education, 6, 301–312.

12.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum.

13.

Commission on Rehabilitation Counselor Certification. (2016). Code of Professional Ethics for Rehabilitation Counselors. Schaumburg, IL: Author.

14.

Council for Accreditation of Counseling and Related Educational Programs. (2016). 2016 CACREP standards. http://www.cacrep.org/wp-content/uploads/2017/08/2016-Standards-with-citations.pdf

15.

Council for Accreditation of Counseling and Related Educational Programs. (2019). Registry of undergraduate rehabilitation programs, 2017-2018 academic year. https://www.cacrep.org/for-students/

16.

Devine

P. G.

(1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56(1), 5–18. https://doi.org/10.1037/0022-3514.56.1.5

17.

Devine

P. G.

Forscher

P. S.

Austin

A. J.

Cox

W. T. L.

(2012). Long-term reduction in implicit race bias: A prejudice habit-breaking intervention. Journal of Experimental Social Psychology, 48(6), 1267–1278.

18.

DiAngelo

(2011). White fragility. International Journal of Critical Pedagogy, 3(3), 54–70.

19.

Di Giunta

Eisenberg

Kupfer

Steca

Tramontano

Caprara

G. V

. (2010). Assessing perceived empathic and social self-efficacy across countries. European Journal of Psychological Assessment, 26(2), 77–86.

20.

Dolby

(2014). The future of empathy: Teaching the millennial generation. Journal of College and Character, 15(1), 39–44. https://doi.org/10.1515/jcc-2014-0006

21.

Evenson

Holloway

(2007). Undergraduate education: An essential rung on the rehabilitation career ladder. Rehabilitation Education, 21(2), 73–85. https://doi.org/10.1891/088970107805059760

22.

Field

(2005). Discovering statistics using SPSS (2nd ed.). SAGE.

23.

Findler

Vilchinsky

Werner

(2007). The multidimensional attitudes scale toward persons with disabilities (MAS) construction and validation. Rehabilitation Counseling Bulletin, 50(3), 166–176.

24.

Fishbein

Ajzen

(1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Addison-Wesley.

25.

Gawronski

Bodenhausen

G. V.

(2006). Associative and propositional processes in evaluation: An integrative review of implicit and explicit attitude change. Psychological Bulletin, 132(5), 692–731. https://doi.org/10.1037/0033-2909.132.5.692

26.

Goreczny

A. J.

Bender

E. E.

Caruso

Feinstein

C. S.

(2011). Attitudes toward individuals with disabilities: Results of a recent survey and implications of those results. Research in Developmental Disabilities, 32(5), 1596–1609. https://doi.org/10.1016/j.ridd.2011.02.005

27.

Greenwald

A. G.

Poehlman

T. A.

Uhlmann

E. L.

Banaji

M. R.

(2009). Understanding and using the Implicit Association Test: III. Meta-analysis of predictive validity. Journal of Personality and Social Psychology, 97(1), 17–41. https://doi.org/10.1037/a0015575

28.

Hein

Grumm

Fingerle

(2011). Is contact with people with disabilities a guarantee for positive implicit and explicit attitudes? European Journal of Special Needs Education, 26(4), 509–522. https://doi.org/10.1080/08856257.2011.597192

29.

Herbert

J. T.

Barrett

Evenson

Jacob

C. J.

(2010). Work roles and functions of undergraduate rehabilitation services alumni: A pilot study. Rehabilitation Research, Policy, and Education, 24(3/4), 149–166.

30.

Antony

Creery

Vargas

Bodenhausen

Paller

(2015). Unlearning implicit social biases during sleep. Science, 348(6238), 1013–1015.

31.

Hunt

C. S.

(2000). Attitudes toward people with disabilities: A comparison of undergraduate rehabilitation and business majors. Rehabilitation Education, 14(3), 269–283.

32.

Imrie

Kumar

(1998). Focusing on disability and access in the built environment. Disability & Society, 13(3), 357–374. https://doi.org/10.1080/09687599826687

33.

Kelsey

Smart

J. F.

(2012). Social justice, disability, and rehabilitation education. Rehabilitation Research, Policy, and Education, 26(2/3), 229–240.

34.

Konrath

S. H.

O’Brien

E. H.

Hsing

(2011). Changes in dispositional empathy in American college students over time: A meta-analysis. Personality and Social Psychology Review, 15(2), 180–198. https://doi.org/10.1177/1088868310377395

35.

Lemm

K. M.

Lane

K. A.

Sattler

D. N.

Khan

S. R.

Nosek

B. A.

(2008). Assessing implicit cognitions with a paper-format Implicit Association Test. In Morrison

M. A.

Morrison

T. G.

(Eds.), The psychology of modern prejudice (pp. 123–146). Nova Science.

36.

Livneh

(1982). On the origins of negative attitudes towards people with disabilities. Rehabilitation Literature, 43, 338–347.

37.

Lord

C. G.

Desforges

D. M.

Ramsey

S. L.

Trezza

G. R.

Lepper

M. R.

(1991). Typicality effects in attitude-behavior consistency: Effects of category discrimination and category knowledge. Journal of Experimental Social Psychology, 27(6), 550–575. https://doi.org/10.1016/0022-1031(91)90025-2

38.

Webber

W. B.

Romero

Chirino

(2018). Changing attitudes toward people with disabilities using public media: An experimental study. Rehabilitation Counseling Bulletin, 61(3), 175–186.

39.

Nosek

B. A.

Smyth

F. L.

Hansen

J. J.

Devos

Lindner

N. M.

Ranganath

K. A.

. . .Banaji

M. R.

(2007). Pervasiveness and correlates of implicit attitudes and stereotypes. European Review of Social Psychology, 18(1), 36–88. https://doi.org/10.1080/10463280701489053

40.

Pearson

Samura

. (2017). Using a spatial lens to examine disability as diversity and diversity within disability on college campuses. In Aquino

K. C.

Kim

(Eds.), Disability as diversity in higher education: Policies and practices to enhance student success (pp. 89–103). Routledge.

41.

Pruett

S. R.

Chan

(2006). The development and psychometric validation of the Disability Attitude Implicit Association Test. Rehabilitation Psychology, 51(3), 202–213.

42.

Rutland

Killen

(2015). A developmental science approach to reducing prejudice and social exclusion: Intergroup processes, social-cognitive development, and moral reasoning. Social Issues and Policy Review, 9(1), 121–154.

43.

Shih

M. J.

Stotzer

Gutiérrez

A. S.

(2013). Perspective-taking and empathy: Generalizing the reduction of group bias towards Asian Americans to general outgroups. Asian American Journal of Psychology, 4(2), 79–83. https://doi.org/10.1037/a0029790

44.

Stanley

D. A.

Sokol-Hessner

Banaji

M. R.

Phelps

E. A.

(2011). Implicit race attitudes predict trustworthiness judgments and economic trust decisions. Proceedings of the National Academy of Sciences of the United States of America, 108(19). 7710–7715. https://doi.org/10.1073/pnas.1014345108

45.

Stier

Hinshaw

S. P.

(2007). Explicit and implicit stigma against individuals with mental illness. Australian Psychologist, 42(2), 106–117. https://doi.org/10.1111/ap.2007.42.2.106

46.

Thomas

Vaughn

E. D.

Doyle

Bubb

(2014). Implicit association tests of attitudes toward persons with disabilities. The Journal of Experimental Education, 82(2), 184–204. https://doi.org/10.1080/00220973.2013.813357

47.

Todd

A. R.

Bodenhausen

G. V.

Richeson

J. A.

Galinsky

A. D.

(2011). Perspective taking combats automatic expressions of racial bias. Journal of Personality and Social Psychology, 100(6), 1027–1042. https://doi.org/10.1037/a0022308

48.

Vilchinsky

Werner

Findler

(2010). Gender and attitudes toward people using wheelchairs: A multidimensional perspective. Rehabilitation Counseling Bulletin, 53(3), 163–174. https://doi.org/10.1177/0034355209361207

49.

Workforce Innovation and Opportunity Act (2014) Pub. L. No. 113-128 (29 U.S.C. Sec. 3101, et. seq.) 134(c)(3)(E).

50.

Young

(1990). Justice and the politics of difference. Princeton University Press.

51.

Yuker

H. E.

Block

J. R.

Young

J. H.

(1970). The measurement of Attitudes Towards Disabled Persons Scale. Ina Mend Institution.

52.

Yuker

H. E.

Hurley

M. K.

(1987). Contact with and attitudes toward persons with disabilities: The measurement of intergroup contact. Rehabilitation Psychology, 32(3), 145–154.