Disability Affirmation Predicts Employment Among Adults With Visual Impairment and Blindness

Abstract

Empirical research on psychosocial correlates of employment among adults with visual impairment or blindness (VI) is limited, and previous studies involving psychological constructs have reported generally nonsignificant or mixed findings. Considering persistent disparities in employment outcomes faced by people with VI, further investigation is needed to understand the role of psychosocial factors in employment in this population. In the current study, stepwise logistic regression was used to explore the association between personal disability identity (i.e., disability affirmation and disability acceptance) and employment status in a sample of 180 working-age U.S. residents. Results indicated that employment status was predicted by higher disability affirmation (p = .001, odds ratio [OR] = 2.21) in a model that included demographic and impairment-related variables that have previously been examined in employment among adults with VI. The exploratory model predicting employment also included higher educational attainment (p < .001, OR = 6.03), self-reported visibility of disability (p = .006, OR = 4.22), age (p < .001, OR = 0.94), and use of mobility tool (p < .001, OR = 0.05). The regression model yielded a pseudo R² value of .32, indicating good fit. Important among these findings, the current study introduces disability affirmation, which involves disability pride and positive self-concept as a person with a disability, as a predictor of employment among working-age U.S. residents. Rehabilitation counselors may benefit from a consideration of the role of disability affirmation in consumer employment outcomes.

Keywords

disability identity disability affirmation blindness visual impairment employment invisible disabilities

In 2016, only 44% of working-age Americans with visual disabilities were employed, compared with 77% of working-age Americans without disabilities (Kraus et al., 2018). Compared with the general U.S. population, individuals with visual impairment or blindness (VI) also have lower annual earnings, lower annual household income, and higher poverty rates (Erickson et al., 2018). Considering disparities in employment and income, researchers, practitioners, and policymakers have sought to understand factors contributing to employment, underemployment, and unemployment among individuals with VI. Most studies on predictors of employment in adults with VI have examined environmental, impairment, and demographic factors. In their systematic review, Lund and Cmar (2019) noted a critical absence of psychosocial variables in the literature on employment in adults with VI. Their findings further indicated a near absence of studies exploring psychosocial factors that are specific to disability status and the experience of disability. Research is needed on population-general psychosocial variables, such as self-esteem, perceived support, and anxiety, as well as disability-specific psychosocial variables, such as disability identity.

Correlates of Employment in Adults With VI

In their recent systematic review, Lund and Cmar (2019) examined research from 1990 to 2018 on correlates of employment among adults with VI who were living in the United States. Overall, Lund and Cmar found that, generally speaking, demographic and impairment-related variables were not significant predictors of employment outcomes; they noted only a few inconsistent exceptions. Lund and Cmar found that studies generally supported the importance of education in attaining favorable employment outcomes. The level of education was positively associated with employment in seven of the nine studies, and education was found to be positively associated with earnings in one additional study. The review by Lund and Cmar further indicated that being male generally predicted higher employment.

As reported by Lund and Cmar (2019), the association between age and employment outcomes was examined in six studies. Age was a significant predictor in two of these studies (Bell & Silverman, 2018; Cmar et al., 2018), but the direction of the relationship differed in each study. Lund and Cmar attributed inconsistent findings to reliance on small, nonrepresentative, and nonnational samples. Severity of visual impairment was a significant predictor of employment in two of the eight studies: people who were partially sighted were employed in lower level positions than those who are blind (Leonard et al., 1999) and persons with mild vision loss were more likely to work full-time than persons who are totally blind; this relationship was not observed among participants who reported moderate, severe, or profound vision loss (odds ratio [OR] = 2.82; Cmar et al., 2018).

Age at onset of VI was a significant predictor of employment in two of the five studies. Congenital blindness was associated with full-time or self-employment (Bell & Silverman, 2018; OR = 1.41), and age at onset was negatively associated with full-time employment (Cmar et al., 2018). Presence of multiple disabilities was examined in four studies and was found to have a small, significant, negative relationship with full-time or self-employment in Bell and Silverman (2018) but a nonsignificant, negative association with employment in Capella-McDonnall (2005). Variables related to mobility devices were examined in two studies. Use of cane was associated with both competitive employment and earnings (Bell & Mino, 2013), and type of cane had a small, significant association with competitive employment and earnings (Bell & Mino, 2013).

More recently, Zapata (2020) examined mobility tool use as a proxy for visibility of disability and found that participants who reported using a cane or guide dog had lower odds of being employed. Because Zapata did not account for the severity of visual impairment, findings could not be interpreted in the context of the literature on invisible disabilities. Invisible disabilities include both psychological and physical conditions that either (a) do not have visible manifestation or (b) have visible manifestations that observers misattribute to other causes. Sensory disabilities (e.g., vision and hearing loss) are among the most common invisible disabilities (Santuzzi et al., 2014).

Psychosocial Factors Associated With Employment Among Adults With VI

Findings from Lund and Cmar (2019) indicated that psychosocial variables were examined less frequently than demographic and impairment factors. Findings further indicated that psychosocial variables either (a) were not associated with employment or (b) yielded mixed results, with only a few exceptions. Five of the 13 studies that met inclusion criteria examined psychosocial variables; results from these studies were generally mixed or nonsignificant. Employment was significantly and positively associated with life satisfaction and support of friends (Cimarolli & Wang, 2006) and negatively associated with low self-esteem, cynicism (Hagemoser, 1996), overprotection, and peer conflict within the social network (Cimarolli & Wang, 2006). Anxiety was significantly and negatively associated with employment in Cimarolli and Wang (2006) but not in Hagemoser (1996). General self-efficacy was not a significant predictor of employment (OR = 1.03; Leonard et al., 1999); however, higher self-reported transportation self-efficacy predicted employment (Cmar et al., 2018). Psychosocial variables that were not demonstrated to significantly predict employment included depression, obsessiveness, Type A behavior, anger, family problems (Hagemoser, 1996), encouragement from family and friends, satisfaction with social contact (OR = 1.70; Leonard et al., 1999), and family support (Cimarolli & Wang, 2006; OR = 1.59). Only one empirical study examined disability-specific psychosocial factors (Jo et al., 2010). Jo and colleagues (2010) found that participants who focused on their remaining assets (e.g., functions, abilities) after vision loss were more likely to be employed.

Lund and Cmar (2019) concluded that, although some psychosocial variables were significantly associated with employment outcomes, definitive conclusions could not be drawn about psychosocial variables because they were infrequently examined. Further research is needed on psychosocial factors related to employment outcomes among adults with VI, especially factors related to disability-specific psychosocial phenomena.

Disability Identity

Identity is a conceptual tool for thinking about sameness and difference at the individual level as well as the level of group affiliation and social categorization (Hammack, 2015). Researchers have described disability as a minority identity category because people with disabilities are members of a group that faces stigmatization, prejudice, and discrimination (Gill, 1997). As members of a minority group, people with disabilities must determine the extent to which they will assimilate into the nondisabled dominant culture versus affiliate with the disability culture and express pride in this group membership (Gill, 1997). Positive disability identity refers to having “favorable or beneficial self-beliefs” as a person with a disability, which can be described as personal disability identity (PDI), as well as having “positive ties” to other members of the disability community, which can be described as group disability identity (GDI; Dunn, 2015, p. 95).

Researchers within the field of rehabilitation psychology have begun to explore the role of PDI in life adjustment of individuals with disabilities. In their systematic review of the literature on disability identity, Forber-Pratt et al. (2017) found only nine quantitative studies on disability identity and only two measures of PDI (Darling & Heckert, 2010; Hahn & Belt, 2004). PDI has been operationalized as feelings of disability pride and affirmation as well as disability acceptance versus denial. The most widely used measure of PDI operationalizes disability affirmation and disability acceptance versus denial (Hahn & Belt, 2004). Hahn and Belt (2004) described positive PDI as attitudes of “dignity and pride” rather than “denial and shame” (p. 455). Research using Hahn and Belt’s PDI subscales has demonstrated a positive association between disability identity and satisfaction with life (Bogart, 2014) and general self-efficacy (Zapata, 2018) as well as a negative association with anxiety and depression (Bogart, 2015). This is the first study to examine the role of PDI, as operationalized in measurement, in employment outcomes among individuals with disabilities.

The goal of this exploratory study was to first examine the association between employment status and PDI (i.e., disability acceptance and disability affirmation), demographic variables, and impairment-related variables in a sample of working-age U.S. residents with VI. The author hypothesized that disability affirmation and disability acceptance would be among significant predictors of employment status, after accounting for demographic and impairment-related variables that have previously been examined in the context of employment among adults with VI.

Method

Participants and Procedures

Participants included 180 working-age adults residing in the United States with self-reported VI. Participants (67% female; n = 120) ranged in age from 21 to 66 years and represented 41 states and Puerto Rico (n = 1). The following states were not represented: Alaska, Hawaii, Kentucky, North Carolina, Rhode Island, South Dakota, Vermont, West Virginia, and Wyoming. Additional participant characteristics are displayed in Table 1.

Table 1.

Participant Characteristics.

Characteristic	Statistic
Age (SD)	46 (12)
Male	33% (60)
Race/Ethnicity
White	82% (146)
Black or African American	1% (8)
American Indian or Alaska Native	<1% (0)
Asian	3% (5)
Native Hawaiian or Pacific Islander	<1% (0)
Hispanic/Latino (may have endorsed others)	10% (18)
Mixed race/multiple ethnicities	3% (6)
Other race/ethnicity	8% (14)
Employment status
Part- or full-time employment	56% (101)
Unemployed	44% (79)
Income
Less than US$10,000	4% (6)
US$10,000 to US$19,999	4% (7)
US$20,000 to US$29,999	20% (34)
US$30,000 to US$39,999	18% (30)
US$40,000 to US$59,999	14% (23)
US$60,000 to US$79,999	9% (16)
US$80,000 to US$99,999	12% (21
US$100,000 to US$119,999	6% (11)
US$120,000 to US$139,999	13% (22)
US$140,000 or higher
Bachelor’s degree or higher	56% (101)
Married	59% (106)
Region of residence
Urban	30% (54)
Suburban	44% (78)
Rural	26% (47)
Cause(s) of VI
Stargardt disease	5% (9)
Retinitis pigmentosa	36% (65)
Macular degeneration	<1% (1)
Glaucoma	9% (16)
Cataracts	1% (2)
Diabetic retinopathy	2% (4)
Retinitis pigmentosa and cataracts	10% (18)
Glaucoma and cataracts	4% (8)
Multiple causes (nonspecified)	8% (14)
Other cause(s)	24% (43)
Age at diagnosis (SD)	22 (17)
Totally blind or light perception (vs. partial vision)	19% (34)
Congenital (vs. acquired) condition	18% (32)
More than one impairment	24% (44)
Family member with the same condition	42% (75)
Stable (vs. progressive) condition	16% (29)
Uses mobility tool (cane, guide dog, both)	56% (100)
Disability visible to others at least most of the time	35% (63)

Note. N = 180. Percentages might not sum to 100 as a result of rounding or multiple response selections. VI = visual impairment or blindness.

Participants were recruited through online forums and mailing lists for people with VI. A link to the study was posted to private Facebook groups of which the author—who has retinitis pigmentosa with associated VI—is a member. A link to the study was also sent through mailing lists of the National Federation of the Blind, a U.S.-based organization. Data were collected through Qualtrics. Participants completed a 15-min cross-sectional online questionnaire containing questions related to demographic, impairment-related, and attitudinal information. Participants were informed about study rationale, the voluntary nature of the survey, risks/discomforts, and benefits of participation. Participants did not receive financial compensation for their participation. The author disclosed status as visually impaired in the study invitation.

Participant responses were collected in the Spring 2019. Variables used in analyses were derived from a larger data set collected by the author that included additional attitudinal measures. This study was reviewed by the Office for the Protection of Human Subjects at the University of California, Berkeley. This study was funded by an award from Division 22 of the American Psychological Association’s Foundation for Rehabilitation Psychology.

Measures

Personal Disability Identity Scale: Affirmation and acceptance

The domains of disability affirmation and acceptance were measured using an 8-item instrument composed of two subscales (Hahn & Belt, 2004). A mean score was calculated for each subscale, including affirmation of disability (4-item) and acceptance of disability (4-item). Disability affirmation items included “Being a person with a disability is an important reflection of me” and “I feel proud to be a person with a disability.” Disability acceptance items included “I do not feel good about being a person with a disability” and “My disability sometimes makes me feel ashamed.” Items on acceptance were reverse scored; lower scores on items targeting denial of disability represented higher disability acceptance. Responses for both factors were reported on a 5-point Likert-type scale from 1 (strongly disagree) to 5 (strongly agree). Zapata (2018) reported modest to moderate internal consistency estimates for affirmation of disability (α = .65; ω = .71) and acceptance of disability (α = .74; ω = .79) scores. More recently, Zapata (2019) obtained the following internal reliability coefficients: disability affirmation (α = .82; ω = .83) and disability acceptance (α = .81; ω = .82). The 4 items of the Affirmation scale had factor loadings that ranged from .29 to .91. The 4 items of the Acceptance scale had factor loadings that ranged from .46 to .91 (Hahn & Belt, 2004; Zapata, 2018, 2019). Goodness of fit indices have generally supported the two-factor model of disability acceptance and affirmation (Hahn & Belt, 2004; Zapata, 2018, 2019).

Demographic characteristics

The variable for employment status was dichotomized (without employment = 0, part- or full-time employment = 1). Participants who reported being full-time students were excluded due to their potential confounding influence on findings related to employment. Five demographic independent variables were included in the study. Participants self-reported gender (female = 0, male = 1), education level (less than bachelor’s degree = 0, bachelor’s degree or higher = 1), and marital status (not married = 0, married = 1). Participants older than 66 years were excluded from analyses based on the standard retirement age for social security benefits in the United States (National Academy of Social Insurance, 2019). Finally, participants reported annual household income. Response options included nine ordinal categories increasing from (1) less than US$10,000 to (9) US$140,000 or more.

Impairment-related characteristics

Six impairment-related independent variables were included in the study. Participants self-reported presence of multiple disabilities (condition = 0, at least two conditions = 1), visibility of disability (“never, rarely, or sometimes visible to others” = 0, “often or almost always visible to others” = 1), mobility tool use (do not use mobility tool = 0, use cane, dog, or both = 1), and stability of disability (progressive condition = 0, stable condition = 1). Participants self-reported age at diagnosis. First, nonnumeric responses such as “at birth,” “congenital,” “0,” and values less than 1 year were recoded as “1.” Then, responses were dummy coded to create a variable for congenital onset (age at diagnosis >1 = 0, age at diagnosis <2 = 1). Participants reported whether a family member has the same condition (no family member = 0, at least one family member = 1). The variable for the presence of a family member with disability was intended as a proxy for family support, in the absence of more comprehensive support indicators.

Results

Descriptive Statistics

All analyses were conducted using Stata (Version 16). Participant scores on disability affirmation and disability acceptance were generally normally distributed. Table 2 displays statistics on skewness and kurtosis as well as means and standard deviations for PDI subscales. Cronbach’s alphas were above .80, indicating good internal consistency of PDI subscale scores (see Table 2).

Table 2.

Descriptive Statistics for Key Variables.

Variables	M	SD	α	Skewness	Kurtosis	Disability affirmation
Disability affirmation	3.09	1.07	.82	.16	2.53	—
Disability acceptance	2.94	0.94	.83	.18	2.10	.69

Table 3.

Regression Models Predicting Employment.

Predictor variable	B	SE B	z	OR	[95% CI] OR	SE OR
Initial exploratory model
Bachelor’s degree	1.80*	.42	4.33	6.06	[2.68, 13.68]	2.52
Disability affirmation	0.73*	.23	3.19	2.08	[1.33, 3.27]	0.48
Age	−0.06*	.02	−3.64	0.94	[0.91, 0.97]	0.02
Visible condition	1.42*	.47	3.05	4.17	[1.67, 10.43]	1.95
Uses mobility tool	−2.93*	.53	−5.55	0.05	[0.02, 0.15]	0.03
Constant	1.12	.57	1.22	3.06	[0.51, 18.46]	2.81
Observations		180
LR chi-square (df)		78 (13)
Pseudo R²		.32
Final confirmatory model
Bachelor’s degree	1.80*	.42	4.26	6.03	[2.64, 13.79]	2.54
Disability affirmation	0.79*	.24	3.27	2.21	[1.37, 3.54]	0.53
Age	−0.06*	.02	−3.75	0.94	[0.91, 0.97]	0.02
Gender	−0.35	.42	−0.82	0.41	[0.31, 1.61]	0.30
Visible condition	1.44	.52	2.74	4.22	[1.51, 11.79]	2.21
Uses mobility tool	−3.01*	.54	−5.60	0.05	[0.02, 0.14]	0.03
Stable condition	−0.08	.64	−0.12	0.93	[0.26, 3.27]	0.60
Congenital condition	−0.41	.55	−0.75	0.66	[0.23, 1.94]	0.36
Blind versus partial vision	0.08	.66	0.13	1.09	[0.30, 3.99]	0.72
Constant	1.12	.57	1.22	3.06	[0.51, 18.46]	2.81
Observations		180
LR chi-square (df)		80 (9)
Pseudo R²		.32

Note. Initial model was derived from stepwise logistic regression (backwards method); final model includes severity indicators and gender. B = unstandardized coefficient; OR = odds ratio; CI = confidence interval; LR chi-square = likelihood ratio chi-square test.

p < .004.

Logistic Regression Analyses

Results were analyzed using a significance level of .05. To control for Type I error, the Bonferroni adjustment was calculated, yielding a critical alpha of .004. ORs were interpreted in accordance with recommendations of Rosenthal (1996): ORs below 1.50 constitute negligible effects; ORs between 1.50 and 2.49 constitute small effect; ORs between 2.50 and 3.99 constitute medium effects; and ORs at or above 4.00 constitute large effects. See Table 3 for results of logistic regression.

Exploratory regression analysis

Prior research on correlates of employment among adults with VI has found that demographic, impairment-related, and psychosocial variables are generally not significant predictors of employment outcomes, with only a few inconsistent exceptions (see Lund & Cmar, 2019). Thus, in a backwards stepwise logistic regression, employment status (i.e., having part- or full-time employment) was regressed on disability identity (i.e., disability affirmation, disability acceptance), demographic variables, and impairment-related variables. Stepwise logistic regression (backwards method) derives a best-fitted model by conducting a series of multiple logistic regressions and removing the least correlated variables.

Results from the initial stepwise regression analysis are displayed in Table 2. In the derived model, employment status was predicted by higher disability affirmation (p = .001, OR = 2.08), visibility of disability (p = .002, OR = 4.17), mobility tool use (p < .001, OR = 0.05), higher educational attainment (p < .001, OR = 6.06), and age (p < .001, OR = 0.94). The remaining variables were not found to be statistically significant contributors to model fit and were removed from the regression model: disability acceptance (p = .31), gender (p = .44), congenital onset (p = .29), being blind versus partially sighted (p = .82), stability of condition (p = .90), being married (p = .63), presence of multiple disabilities (p = .96), and having a family member with the same condition (p = .88). McFadden’s R² (i.e., pseudo R²) was .32, indicating a well-fitted model (McFadden, 1977).

Confirmatory regression analysis

Considering previous research noting the absence of severity indicators as a limitation to the interpretation of findings regarding invisible versus visible disabilities (Zapata, 2018), the author conducted a second logistic regression on employment that included the initial derived model (described above) as well as severity-related indicators (i.e., stability of condition, congenital status, visual status). The severity indicators were intended to serve as control variables to contextualize findings related to visibility of disability. In addition, the author reintroduced gender into the model as a control variable, given previous findings that male gender predicts positive employment outcomes (Lund & Cmar, 2019). The final confirmatory model is displayed in Table 2.

For every one-unit increase in mean self-reported disability affirmation, participants had 121% higher odds of being employed (p = .001, OR = 2.21). Participants who reported that their disability was visible to others at least “most of the time” had 322% higher odds of being employed (p = .006, OR = 4.22). Employment status was predicted by mobility tool use but in the opposite direction: Participants who reported using a mobility tool had 95% lower odds of being employed (p < .001, OR = 0.05). Participants who had earned a bachelor’s degree or higher had 503% higher odds of being employed (p < .001, OR = 6.03). Finally, participants had 6% lower odds of being employed for every additional year of age (p < .001, OR = 0.94). Employment status was not predicted by gender (p = .41, OR = 0.71), visual status (p = .90, OR = 1.09), congenital onset (p = .45, OR = 0.66), or stability of condition (p = .91, OR = 0.93). McFadden’s R² (i.e., pseudo R²) was .32, indicating a well-fitted model.

Discussion

Findings of the current study indicated that disability affirmation is predictive of employment status among working-age U.S. residents with VI, after accounting for education, age, gender, mobility tool use, visibility of disability, and severity of vision loss. Disability affirmation has been conceptualized as a psychological construct of disability identity (Hahn & Belt, 2004). Hahn and Belt (2004), who developed the PDI subscales used in the current study, referred to disability affirmation as a shift in attitudes from “shame” to “pride.” Attitudinal statements reflecting disability affirmation include considering disability “an important reflection of me,” feeling “proud to be a person with a disability,” and “being glad” to be a person with a disability. Supporting the role of positive disability identity in life adjustment among adults with VI, results indicated that participants who reported higher feelings of disability pride were more likely to be employed. Rehabilitation counselors and other practitioners working on employment goals with consumers who have VI should consider how to support positive disability identity formation, with particular attention to how individuals with disabilities come to adopt disability-affirming attitudes. Relevant literature within the field of rehabilitation psychology includes such topics as disability-affirmative therapy and case formulation (Olkin, 2017), disability advocacy and culturally competent practice (Andrews, 2020), comprehensive reviews of the disability identity literature (Dunn & Burcaw, 2013; Forber-Pratt et al., 2017), and discussion of the experience of disability from within a social psychological perspective (Dunn, 2015, 2020).

In the initial exploratory model, results indicated that participants who reported having a disability that is visible to others at least “most of the time” had four times higher odds of being employed, after accounting for mobility tool use, stability of condition, congenital versus acquired onset, and visual status (blind vs. partially sighted). Importantly, the relationship between visibility of disability and employment status approached but did not reach statistical significance in the final model (p = .006), when severity indicators were reintroduced, but the effect size was still strong (OR = 4.22). A postulation that requires further empirical investigation, this finding potentially indicates that working-age U.S. residents living with an invisible visual disability face additional or unique barriers to employment attainment compared with adults who self-report having a disability that is visible to others. According to Santuzzi and colleagues (2014), the challenges faced by people with invisible disabilities can be manifested in (a) the extent to which people can effectively conceal disability status, (b) the reactions of others to visible manifestations of disability, (c) the domain of work or life influenced by disability, and (d) how disability identity formation is influenced by manifestations of disability that are visible to others. To inform best practices for serving the employment-related needs and goals of the invisible disability subpopulation, researchers should continue to examine the association between invisible disability and employment outcomes.

Results regarding educational attainment are consistent with the literature. As Lund and Cmar (2019) recommended based on findings from their systematic review, practitioners should encourage individuals with VI who do not possess a college degree to consider options for postsecondary education. The current study found that, in general, older participants had lower odds of being employed. Six previous studies investigated the association between age and employment outcomes in this population, with two studies finding significance but in opposite directions (see Lund & Cmar, 2019). Lund and Cmar attributed mixed findings regarding age to the use of small, nonnational samples. Consistent with Zapata (2020), the current study found that participants who reported using a cane or guide dog had lower odds of being employed, after controlling for visibility of disability and other vision loss severity indicators. In contrast to findings regarding mobility tool use, which Zapata (2020) used as a proxy for visibility of disability, participants who reported having a disability that is visible to others had higher odds of being employed. As discussed in the following section, more research on these contrary findings is warranted. Results regarding age and mobility tool use were significant (p < .004), but effect sizes were negligible and should be interpreted with due consideration of their limited practical significance.

Limitations and Future Directions

Causal research methods, including longitudinal and qualitative methods, could be used to deepen and expand correlational findings in the current study by exploring explanatory hypotheses, such as (a) the direction of the relationship between pride and employment and (b) the potential role of stigmatization in employment disparities based on visibility of disability.

Santuzzi and Waltz (2016) proposed that various work-related factors create variability in disability identity at multiple levels. Santuzzi and Waltz conceptualized disability identity in alignment with Bogart (2014), who used the same PDI subscales as those in the current study (Hahn & Belt, 2004). The theorized levels include the intraindividual (e.g., internalized stigma, connection of impairment to disability label), interpersonal (e.g., anticipated stigma, perceived legitimacy of disability label, situational relevance of disability, strength of other identities), organizational (e.g., culture of inclusion, workplace stress, job demands and changes), and legal, medical, and cultural (e.g., medical definitions, sociocultural attitudes and stereotypes). Future studies could examine the association between employment and disability visibility in the context of Santuzzi and Waltz’s multilevel variables. Future studies could also use these multilevel variables to examine what may be a reciprocal influence between disability pride and employment outcomes, such that feelings of disability pride influence employment experiences and employment experiences influence feelings of pride.

Results of the current study indicated that mobility tool use predicted lower odds of employment, whereas self-reported visibility of disability predicted higher odds of employment among working-age adults with VI. Conjectures for future empirical investigation, these seemingly paradoxical findings may be explained by such participant characteristics as (a) visual acuity, which was not among available impairment factors, (b) perceived stigma, (c) frequency and situational specificity of mobility tool use (i.e., when, where, and how often participants use a mobility tool), or (d) the context imagined by participants in response to the item regarding visibility of disability to others (i.e., who participants imagined as the “others”).

The research on correlates of employment among adults with VI would benefit from further examination of disability affirmation in larger and more diverse samples to substantiate the findings of the current study. First, the effect size of this finding was small (but not negligible); this may have been associated with the small sample size or threats to internal validity. Second, this study did not account for race/ethnicity of its participants, the majority of whom reported being “White” (82%). People with disabilities represent a heterogeneous population that may be considered the world’s largest minority group (Olkin, 2002). Considering the diversity of the experience of disability, future intersectional exploration of race/ethnicity and other cultural factors will be important for the external validity of the current findings. Third, stepwise regression, which was chosen for initial analysis based on the mixed findings in the literature and the exploratory objective of the study, is sensitive to the issue of model overfitting based on sample characteristics, which can limit external validity (Henderson & Denison, 1989). Thus, future research on the role of disability affirmation in employment is warranted.

Conclusions and Implications

Few studies have examined the association between psychosocial variables and employment outcomes among adults with VI (Lund & Cmar, 2019). Findings of the current study indicated that, among working-age adults with VI, participants who reported stronger feelings of disability pride were more likely to report being employed. Both practitioners and researchers within the field of rehabilitation counseling may benefit from further consideration of disability pride versus shame among working-age adults with VI. Consistent with the literature on the positive association between education and employment in this population (see Lund & Cmar, 2019), participants who reported having at least a bachelor’s degree had six times higher odds of being employed (p = .001). Participants who reported having a visible disability had four times higher odds of being employed (p = .006), after controlling for mobility tool use and severity of VI. This preliminary and nonsignificant finding suggests that people with invisible VI may face additional or unique barriers to obtaining employment, a possible explanation that should be addressed in future empirical investigation. Age and mobility tool use were significantly and negatively associated with employment status. However, effect sizes based on ORs were negligible; thus, future research on these associations is warranted.

Results of the current study also have implications for working-age adults who represent various disability groups. A 2005 review of disability-related employment experiences concluded with a call for more research on psychological processes underlying such processes as disability discrimination (Colella & Stone, 2005). In a more recent review of the literature on disability-related employment experiences, Santuzzi and Waltz (2016) found that the majority of research on employment discrimination considered the perspective of observers rather than the internal experiences of workers with disabilities. The current study contributes to the literature on psychological processes underlying employment outcomes, as told from the perspectives of people with disabilities. Positive disability identity has been theorized as a protective factor against disability-related life stressors, which includes negative employment-related experiences. Emerging research is beginning to support the association between PDI and positive life outcomes (see Forber-Pratt et al., 2017), and this is the first study to examine the association between PDI and employment among people with disabilities. Providing support for the association between PDI and positive life outcomes among people with disabilities, findings indicated that disability affirmation (feelings of pride vs. shame related to disability) predicts having employment (part- or full-time) among working-age adults with VI who live in the United States.

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 disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by an award from Division 22 of the American Psychological Association’s Foundation for Rehabilitation Psychology.

ORCID iD

Mercedes A. Zapata

References

Andrews

E. E.

(2020). Disability as diversity: Developing cultural competence. Oxford University Press.

Bell

E. C.

Mino

N. M.

(2013). Blind and visually impaired adult rehabilitation and employment survey: Final results. Journal of Blindness Innovation and Research, 3(1). https://dx-doi-org.web.bisu.edu.cn/10.5241/2F1-35

Bell

E. C.

Silverman

A. M.

(2018). Rehabilitation and employment outcomes for adults who are blind or visually impaired: An updated report. Journal of Blindness Innovation and Research, 8(1). https://dx-doi-org.web.bisu.edu.cn/10.5241/8-148

Bogart

K. R.

(2014). The role of disability self-concept in adaptation to congenital or acquired disability. Rehabilitation Psychology, 59, 107–115. https://doi.org/10.1037/a0035800

Bogart

K. R.

(2015). Disability identity predicts lower anxiety and depression in multiple sclerosis. Rehabilitation Psychology, 60, 105–109. https://doi.org/10.1037/rep0000029

Capella-McDonnall

M. E.

(2005). Predictors of competitive employment for blind and visually impaired consumers of vocational rehabilitation services. Journal of Visual Impairment & Blindness, 99, 303–315. https://doi.org/10.1177/0145482X0509900505

Cimarolli

V. R.

Wang

S. W.

(2006). Differences in social support among employed and unemployed adults who are visually impaired. Journal of Visual Impairment & Blindness, 100, 545–556. https://doi.org/10.1177/0145482X0610000906

Cmar

J. L.

McDonnall

M. C.

Crudden

(2018). Transportation self-efficacy and employment among individuals with visual impairments. Journal of Vocational Rehabilitation, 48, 257–268. https://doi.org/10.3233/JVR-180925

Colella

Stone

D. L.

(2005). Workplace discrimination toward persons with disabilities: A call for some new research directions. In Dipboye

Colella

(Eds.), Discrimination at work: The psychological and organizational bases (pp. 227–253). Mahwah, NJ: Erlbaum.

10.

Darling

R. B.

Heckert

D. A.

(2010). Orientations toward disability: Differences over the lifecourse. International Journal of Disability, Development and Education, 57(2), 131–143. https://doi.org/10.1080/10349121003750489

11.

Dunn

D. S.

(2015). The social psychology of disability. Oxford University Press.

12.

Dunn

D. S.

(2020). Understanding the experience of disability: Perspectives from social and rehabilitation psychology. Oxford University Press.

13.

Dunn

D. S.

Burcaw

(2013). Disability identity: Exploring narrative accounts of disability. Rehabilitation Psychology, 58, 148–157. https://doi.org/10.1037/a0031691

14.

Erickson

Lee

von Schrader

(2018). 2016 disability status report: United States. Yang-Tan Institute on Employment and Disability, Cornell University.

15.

Forber-Pratt

A. J.

Lyew

D. A.

Mueller

Samples

L. B.

(2017). Disability identity development: A systematic review of the literature. Rehabilitation Psychology, 62, 198–207. https://doi.org/10.1037/rep0000134

16.

Gill

C. J.

(1997). Four types of integration in disability identity development. Journal of Vocational Rehabilitation, 9, 39–46. https://doi.org/10.1016/S1052-2263(97)00020-2

17.

Hagemoser

S. D.

(1996). The relationship of personality traits to the employment status of persons who are blind. Journal of Visual Impairment & Blindness, 90, 134–144.

18.

Hahn

H. D.

Belt

T. L.

(2004). Disability identity and attitudes toward cure in a sample of disabled activists. Journal of Health and Social Behavior, 45, 453–464. https://doi.org/10.1177/002214650404500407

19.

Hammack

P. L.

(2015). Theoretical foundations of identity. In McLean

K. C.

Syed

(Eds.), The Oxford handbook of identity development. Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199936564.013.027

20.

Henderson

D. A.

Denison

D. R.

(1989). Stepwise regression in social and psychological research. Psychological Reports, 64, 251–257. https://doi.org/10.2466/pr0.1989.64.1.251

21.

S. J.

Chen

R. K.

Kosciulek

J. F.

(2010). Employment outcomes among individuals with visual impairments: The role of client satisfaction and acceptance of vision loss. Journal of Applied Rehabilitation Counseling, 41, 3–8.

22.

Kraus

Lauer

Coleman

Houtenville

(2018). 2017 disability statistics annual report. University of New Hampshire.

23.

Leonard

D’Allura

Horowitz

(1999). Factors associated with employment among persons who have a vision impairment: A follow-up of vocational placement referrals. Journal of Vocational Rehabilitation, 12, 33–43.

24.

Lund

E. M.

Cmar

J. L.

(2019). A systematic review of factors related to employment outcomes for adults with visual impairments. Journal of Visual Impairment & Blindness, 113, 493–517. https://doi.org/10.1177/0145482X19885211

25.

McFadden

(1977). Quantitative methods for analyzing travel behaviour of individuals: Some recent developments (Cowles Foundation Discussion Papers No. 474). Cowles Foundation for Research in Economics, Yale University. https://EconPapers.repec.org/RePEc:cwl:cwldpp:474

26.

National Academy of Social Insurance. (2019, August). Social security benefits, finances, and policy options: A primer. https://www.nasi.org/socialsecurityprimer

27.

Olkin

(2002). Could you hold the door for me? Including disability in diversity. Cultural Diversity and Ethnic Minority Psychology, 8, 130–137. https://doi.org/10.1037/1099-9809.8.2.130

28.

Olkin

(2017). Disability-affirmative therapy: A case formulation template for clients with disabilities. Oxford University Press.

29.

Rosenthal

J. A.

(1996). Qualitative descriptors of strength of association and effect size. Journal of Social Service Research, 21, 37–59.

30.

Santuzzi

A. M.

Waltz

P. R.

(2016). Disability in the workplace: A unique and variable identity. Journal of Management, 42(5), 1111–1135. https://doi.org/10.1177/0149206315626269

31.

Santuzzi

Waltz

Finkelstein

Rupp

(2014). Invisible disabilities: Unique challenges for employees and organizations. Industrial and Organizational Psychology, 7, 204–219. https://dx-doi-org.web.bisu.edu.cn/10.1111/iops.12134

32.

Zapata

M. A.

(2018). Personal disability identity in retinitis pigmentosa. Rehabilitation Psychology, 63, 512–520. https://doi.org/10.1037/rep0000238

33.

Zapata

M. A.

(2019). Personal disability identity measurement: Self-worth and personal meaning [Unpublished doctoral dissertation]. https://escholarship.org/uc/item/1s79b3gt

34.

Zapata

M. A.

(2020). An exploratory study of general self-efficacy and employment in adults with retinitis pigmentosa. Journal of Visual Impairment & Blindness, 114, 18–30. https://doi.org/10.1177/0145482X19900715