Demographic and Disability Variables and Identity
This is a further analysis of the data described in the previous post. Here, I'm asking the questions of whether gender, race, sexual orientation or disability variables affect a person's view of their disability. (Are certain disabilities viewed more negatively than others? Does being the target of other forms of discrimination affect one's view of disability?)
Demographics
Ethnicity
Data on ethnicity was unavailable for 7 people. Five of them were because I hadn't added that question yet, and the other two declined the question (one left it blank and one answered 'human'). Of the others, 74% were white (I can't seem to get SPSS not to count the non-answers in the percentages, by the way). The 6 non-white people included three 'other' (mostly mixed-race), 1 middle-eastern, 1 native american and 1 black. Due to the small number of non-white people, I clumped them together as a single group.
Since there were two groups, I ran T-tests for Preencounter2, Internalization2 and Identity Subtype by ethnic category. Equal variances were found for all three variables. Preencounter2 and subtype showed no significant differences, but Internalization2 differed at p = .041. Looking at the means showed that whites (mean .6933) scored higher on this scale than non-whites (mean .4000).
Gender
For 6 people, data on gender was unavailable. Of the others, 47% were female, 37% were male and 2 chose 'other' (one with multiple personalities that differed in gender identity, and one biologically female but 'genderqueer'). Due to the rarity of 'other' responses, I dropped them from further analysis (it may be useful to do a further study to see if atypical gender identity affects disability identity).
No significant differences by gender were found on T-test of Preencounter2, Internalization2 or Identity Subtype.
Sexual Orientation
Data on sexual orientation was unavailable for 6 people. Of the others, 51% were straight, 14% were gay, 10% were bisexual, 8% were asexual and 2 identified as 'other' (one 'queer', one didn't clarify). I decided to combine gay and bisexual into one group and asexual and other into another group, reflecting which atypical orientations are widely recognized.
Since there were three categories, I ran a one-way ANOVA, using the Bonferroni and Scheffe post-hoc tests. The variables analyzed were the same as before. No significant differences were found.
General population estimates usually put the heterosexual population as a larger majority than they were in my study. Remafedi et al (1992) found 88% of their adolescent sample were heterosexual, and Smith (1991) found 80% of US adults were heterosexual. Ingudomnukul et al (2007) found that autistic women were significantly less likely than non-autistic women to be heterosexual (70% heterosexual as opposed to 97%), with bisexuality (13%) and asexuality (17%) being the next most common orientations.
So I decided to see if orientation differed depending on whether or not the person had an autism spectrum condition. I ran a T-test and found that there was no significant differences, though the test for equal variances approached significance (p = .075). This is likely a matter of sample size, since the number of autistic (26) and non-autistic (17) subjects both fell below the recommended sample size of 30 or more.
Disability Variables
As mentioned before, my sample included a lot of autistics. Here's the frequencies for each disability category (some had multiple disabilities, so percentages total more than 100%):
Number of disabilities ranged from 1-7, with 47% having only one disability. Note that if the person had multiple disabilities in the same category (for example, generalized anxiety and agoraphobia), these were counted as a single disability.
I also asked them, if they had multiple disabilities, to indicate which one they'd be focusing on while replying to the survey. One person decided to answer based on being 'multi' and 5 others with multiple disabilities left this question blank, so those 6 were classed as multi on this variable. In total, 55% were discussing autism, 12% were discussing multiple disabilities, 10% were discussing a physical disability, 6% were discussing a psychiatric disorder, 4% were discussing a chronic illness, and one each were discussing a developmental delay, deafness, epilepsy, learning disability/ADHD, an unknown disability and a trauma-related condition.
For the six most common disabilities, I ran T-tests specifically for those conditions. Significant differences were found for autism and LD/ADHD only.
For autism, difference in variances was found for Preencounter2 and Identity Subtype, both showing much less variation among autistics than non-autistics. Those same two categories also differed on means, with non-autistics scoring higher on Preencounter2 and being more likely to be Unclassified/Preencounter Subtype. So, compared to people with mixed disabilities, autistics may be more positive about their disability on average. This fits what I've seen anecdotally, but has not previously been studied. (Interestingly, the opposite result has been found in parents of kids with disabilities, for example in Weiss [2002], where parents of autistic kids tend to have more negative views about their child's disability.)
For LD/ADHD, no significant differences in variance were found, but Preencounter2's variance approached significance (p = .077). Preencounter2 was also significantly higher in those with LD/ADHD, at p = .047. It's interesting to speculate about why this might be, especially since this is a sharp contrast with my findings about autism, despite high functioning autism* causing similar impairment to LD/ADHD. One big difference is that many high functioning autistics do well academically (Griswold et al [2002]), so maybe academic achievement is a more important predicter of disability identity than social functioning. (Maybe social functioning is more readily seen as dependent on other people as well as oneself.)
I ran a two-way ANOVA of Preencounter2 by autism and LD/ADHD to analyze this further. I found main effects for autism (p = .003) and LD/ADHD (p = .022) but no interaction (p = .103).
Then I analyzed the relationship between the three variables and total number of disabilities, using bivariate correlation. No significant relationships were found.
Disability discussed was next. For this, I lumped psychiatric disorder, chronic illness and the conditions dicussed by one individual each into an 'other' category, so I had autism, physical disability, multiple disabilities and other. I ran a one-way ANOVA on those four groups.
No significant differences were found, surprising since autism was observed to be associated with more positive attitudes. However, it's important to note that the percentage of my subjects who were on the autistic spectrum was higher than the percentage who chose to discuss autism, so this may indicate that multiply-disabled autistic individuals (who may have discussed autism, another disability, or 'multi') often viewed their disabilities positively.
* I have no way of knowing the functioning level of my autistic subjects, and some may have been lower functioning (eg assistive communication users), but on average it's probably reasonable to assume that most were high functioning.
Demographics
Ethnicity
Data on ethnicity was unavailable for 7 people. Five of them were because I hadn't added that question yet, and the other two declined the question (one left it blank and one answered 'human'). Of the others, 74% were white (I can't seem to get SPSS not to count the non-answers in the percentages, by the way). The 6 non-white people included three 'other' (mostly mixed-race), 1 middle-eastern, 1 native american and 1 black. Due to the small number of non-white people, I clumped them together as a single group.
Since there were two groups, I ran T-tests for Preencounter2, Internalization2 and Identity Subtype by ethnic category. Equal variances were found for all three variables. Preencounter2 and subtype showed no significant differences, but Internalization2 differed at p = .041. Looking at the means showed that whites (mean .6933) scored higher on this scale than non-whites (mean .4000).
Gender
For 6 people, data on gender was unavailable. Of the others, 47% were female, 37% were male and 2 chose 'other' (one with multiple personalities that differed in gender identity, and one biologically female but 'genderqueer'). Due to the rarity of 'other' responses, I dropped them from further analysis (it may be useful to do a further study to see if atypical gender identity affects disability identity).
No significant differences by gender were found on T-test of Preencounter2, Internalization2 or Identity Subtype.
Sexual Orientation
Data on sexual orientation was unavailable for 6 people. Of the others, 51% were straight, 14% were gay, 10% were bisexual, 8% were asexual and 2 identified as 'other' (one 'queer', one didn't clarify). I decided to combine gay and bisexual into one group and asexual and other into another group, reflecting which atypical orientations are widely recognized.
Since there were three categories, I ran a one-way ANOVA, using the Bonferroni and Scheffe post-hoc tests. The variables analyzed were the same as before. No significant differences were found.
General population estimates usually put the heterosexual population as a larger majority than they were in my study. Remafedi et al (1992) found 88% of their adolescent sample were heterosexual, and Smith (1991) found 80% of US adults were heterosexual. Ingudomnukul et al (2007) found that autistic women were significantly less likely than non-autistic women to be heterosexual (70% heterosexual as opposed to 97%), with bisexuality (13%) and asexuality (17%) being the next most common orientations.
So I decided to see if orientation differed depending on whether or not the person had an autism spectrum condition. I ran a T-test and found that there was no significant differences, though the test for equal variances approached significance (p = .075). This is likely a matter of sample size, since the number of autistic (26) and non-autistic (17) subjects both fell below the recommended sample size of 30 or more.
Disability Variables
As mentioned before, my sample included a lot of autistics. Here's the frequencies for each disability category (some had multiple disabilities, so percentages total more than 100%):
- Autism Spectrum (65%)
- Psychiatric Disorder (27%)
- Chronic Illness (25%)
- LD/ADHD (22%)
- PTSD or other trauma-related condition (22%)
- Physical Disability (20%)
- Epilepsy or other episodic condition (16%)
- Deaf/Hard of hearing (8%)
- Blind/Visually Impaired (6%)
- Developmental Delay (4%)
Number of disabilities ranged from 1-7, with 47% having only one disability. Note that if the person had multiple disabilities in the same category (for example, generalized anxiety and agoraphobia), these were counted as a single disability.
I also asked them, if they had multiple disabilities, to indicate which one they'd be focusing on while replying to the survey. One person decided to answer based on being 'multi' and 5 others with multiple disabilities left this question blank, so those 6 were classed as multi on this variable. In total, 55% were discussing autism, 12% were discussing multiple disabilities, 10% were discussing a physical disability, 6% were discussing a psychiatric disorder, 4% were discussing a chronic illness, and one each were discussing a developmental delay, deafness, epilepsy, learning disability/ADHD, an unknown disability and a trauma-related condition.
For the six most common disabilities, I ran T-tests specifically for those conditions. Significant differences were found for autism and LD/ADHD only.
For autism, difference in variances was found for Preencounter2 and Identity Subtype, both showing much less variation among autistics than non-autistics. Those same two categories also differed on means, with non-autistics scoring higher on Preencounter2 and being more likely to be Unclassified/Preencounter Subtype. So, compared to people with mixed disabilities, autistics may be more positive about their disability on average. This fits what I've seen anecdotally, but has not previously been studied. (Interestingly, the opposite result has been found in parents of kids with disabilities, for example in Weiss [2002], where parents of autistic kids tend to have more negative views about their child's disability.)
For LD/ADHD, no significant differences in variance were found, but Preencounter2's variance approached significance (p = .077). Preencounter2 was also significantly higher in those with LD/ADHD, at p = .047. It's interesting to speculate about why this might be, especially since this is a sharp contrast with my findings about autism, despite high functioning autism* causing similar impairment to LD/ADHD. One big difference is that many high functioning autistics do well academically (Griswold et al [2002]), so maybe academic achievement is a more important predicter of disability identity than social functioning. (Maybe social functioning is more readily seen as dependent on other people as well as oneself.)
I ran a two-way ANOVA of Preencounter2 by autism and LD/ADHD to analyze this further. I found main effects for autism (p = .003) and LD/ADHD (p = .022) but no interaction (p = .103).
Then I analyzed the relationship between the three variables and total number of disabilities, using bivariate correlation. No significant relationships were found.
Disability discussed was next. For this, I lumped psychiatric disorder, chronic illness and the conditions dicussed by one individual each into an 'other' category, so I had autism, physical disability, multiple disabilities and other. I ran a one-way ANOVA on those four groups.
No significant differences were found, surprising since autism was observed to be associated with more positive attitudes. However, it's important to note that the percentage of my subjects who were on the autistic spectrum was higher than the percentage who chose to discuss autism, so this may indicate that multiply-disabled autistic individuals (who may have discussed autism, another disability, or 'multi') often viewed their disabilities positively.
* I have no way of knowing the functioning level of my autistic subjects, and some may have been lower functioning (eg assistive communication users), but on average it's probably reasonable to assume that most were high functioning.
posted by Ettina at 11:12 AM
1 Comments:
This comment has been removed by the author.
Post a Comment
<< Home