Parent-reported Strengths and Positive Qualities of Adolescents and Adults with Autism Spectrum Disorder and/or Intellectual Disability

Abstract

Despite increasing attention to understanding strengths in those on the autism spectrum, few studies have explored this in older individuals. The present study provides a qualitative analysis of parent-reported strengths and positive qualities of 68 15-30-year-olds with autism and/or intellectual disability (ID). Most parents (97%) reported at least one positive quality. Themes were consistent with previous studies; Personality characteristics (82%) and Sociability (53%) were reported most frequently. A Work Ethic/Motivation category not previously reported emerged. Autistic individuals (with/without ID) were more likely than those with ID-only to have a Specific Skill. Findings highlight the importance of recognizing strengths of autistic adults. Research is needed to understand how to best leverage positive qualities to promote goal achievement and quality of life.

Historically, the majority of autism research has focused on studying deficits associated with autism spectrum disorder (ASD) and how they pose challenges for autistic individuals. Though an understanding of challenges is necessary to inform service development and individual support planning, it is equally important to characterize the strengths of autistic individuals. Overemphasis on difficulties can have detrimental impact on the self-confidence and mental health of individuals on the autism spectrum (Cage et al., 2018; Cooper et al., 2017) and may limit our ability to identify ways to support well-being and growth. Researchers and self-advocates have called for more strengths-based approaches to our understanding of the profiles of autism symptoms and how to improve quality of life of autistic individuals (Bal et al., 2016; Kapp, 2018; Burnham Riosa et al., 2017). While movement has been made in this direction in child and adolescent studies (e.g., Carter et al., 2020, Cost et al., 2021), few studies have explored positive qualities in adult samples.

Research tracking autism symptom trajectories highlights change and development throughout childhood and adolescence that continues into adulthood (e.g., Bal et al., 2019; Esbensen et al., 2009; Seltzer et al., 2003). Thus, difficulties in childhood captured on some symptom measures may actually be framed as strengths in adulthood. For example, although “stimming” is often classified as a stereotypic behavior, autistic adults highlight its positive role as a mechanism to regulate emotions (Kapp et al., 2019). However, absence of childhood difficulties may simply reflect limitations of instruments designed for use with children, rather than capturing true strengths of adults (Bal et al., 2019). Research specifically aimed at identifying positive qualities is needed to provide a more comprehensive profile of autistic adults. Strengths or positive features may help explain the variability in adult outcomes and have important implications for intervention development. Hence, it is essential to better understand strengths or positive qualities that may be associated with autism or vary from one individual to another. A comprehensive picture of autism is crucial to improve supports to achieve goals and maximize quality of life. Understanding of positive qualities may also have benefits to families and society more broadly, by fostering appreciation for and recognition of capabilities of people with autism (King et al., 2012). This might increase understanding and acceptance of autism in the general public and may also serve to improve the autistic adults’ mental health and well-being (Cage et al., 2018). The current study aims to describe caregiver-reported positive qualities in a sample of autistic adolescents and adults to inform future measure development to support more systematic investigation of the strengths of these individuals.

Savant skills have historically been the focus of strengths-based research with autistic individuals. Savant skills are generally defined as extraordinary talents that are both out of keeping with the person’s general abilities and stand out relative to the general population (Treffert, 2009). While savant skills are often thought of in the context of intellectual disability (ID; e.g., exceptional skill in a certain domain paired with intellectual disability; O'Conner & Hermelin, 1988; Pring 2005), studies suggest that these extraordinary talents are observed in autistic individuals with and without intellectual disability (Bal et al., 2021; Howlin et al., 2009; Miller, 1999) and those with and without savant skills have comparable intellectual profiles (Bölte & Poustka, 2004). Savant skills studied in autism primarily fall into five categories: music, art, calendar calculating, mathematics, and mechanical or spatial skills (Treffert, 2009). Memory is considered a separate talent in some studies (Bal et al., 2021; Bölte & Poustka, 2004; Meilleur et al., 2015) but others consider it an integral part of all other categories (Treffert, 2009). The reported proportion of autistic individuals with a savant skill varies from 10-63% (Meilleur et al., 2015; Howlin et al., 2009; Rimland, 1978) depending on how they are measured. A recent study using the Simons Simplex Collection, a non-epidemiological sample, found that parents of autistic children without intellectual disability reported extraordinary talents in 46% of the sample and an additional 22% of the sample had personal strengths (i.e., skills that stood out relative to their other abilities, but were not above the general population; Bal et al., 2021).

As greater emphasis is placed on strengths-based approaches in supporting autistic individuals, researchers have broadened the focus of strengths research to include positive qualities beyond extraordinary talents. For example, Colavita and colleagues (2014) did a qualitative analysis of parent responses to the open-ended question, “What are your child’s greatest strengths?” in a sample of 141 children with developmental disabilities ages 3 to 19 years. Over half of the parents reported personality characteristics (e.g., loving/affectionate, caring, active) and social personality characteristics (e.g., sense of humor, helpful, playful) as the strengths of their children. Other themes which emerged included Particular Skills, Cognitive Functioning, and Behavioral Characteristics/Coping Mechanisms. While over half of the sample (n=84 of 141) were autistic children, the authors did not report comparisons between younger children and adolescents or those with and without autism.

In another study, Carter and colleagues (2015) used the Assessment Scale for Positive Character Traits- Developmental Disabilities (ASPeCT-DD; Woodard, 2009) and interviews to examine parent reported positive traits of their adolescent children (13-21 years). Among the sample of 427 parents, 183 parents had children with autism-only, 55 had children with autism and ID and the remaining 238 had ID-only. Parents of children with autism (with and without ID) reported fewer total strengths, positive relations, positive outlook, and active coping, relative to the children with ID only. When a subsample of 28 parents were asked, “What comes to mind when you think about [child name]’s strengths?”, they frequently described Interpersonal skills and Personality traits. Other themes included Hobbies, Cognitive abilities, and Particular skills. Identified themes were not associated with demographics (e.g., age, sex, race) or having a dual diagnosis of autism and ID. Speech as the primary communication mode was associated with a higher number of parent-endorsed strengths, whereas challenging behavior was associated with fewer strengths. More parent-reported strengths were also associated with higher levels of community involvement, highlighting that strengths likely play an important role in daily lives and warrant greater attention. One limitation to this study was that correlates of strengths were based on a small collection of parent-rated items about challenging behaviors or functional abilities. While these scales have been used in some of the authors’ other studies (i.e., Carter et al., 2011, 2012), the use of non-standardized scores limits comparison to other studies.

In a study of 300 youth (11-22 years) with ID (including 29% with autism+ID), parents were asked to rate six global statements though to comprise “thriving” (Weiss & Burnham-Rosa, 2015). Youth with autism+ID were rated as having lower overall thriving than youth with ID only, reflecting lower levels of parent-rated competence, confidence, caring and contribution, but no difference in connecteness or character. Social-communicative ability, in addition to involvment in home and school were mediators of thriving. Authors highlighted the need for interventions to support the individual and the environment to foster socially-inclusive environments.

Recently Cost and colleagues (2021) examined changes in parent responses to the Child Behavior Checklist (Achenbach & Rescorla, 2001) item, “Please describe the best thing about your child” in a sample of 153 autistic children assessed at approximately 3, 8, and 11 years old. Parent responses were coded using the VIA framework comprised of 6 categories (Wisdom & knowledge, Courage, Humanity, Justice, Temperance, and Transcendence; Samson & Antonelli, 2013; Kirchner, Ruch, & Dziobek, 2016) and 5 new categories from the study authors (Happiness, Interests, Appearance, Specific skills and Miscellaneous) added to capture themes observed in the responses. Of the 34 traits coded, Love, Happiness, and Kindness were the most prevalent, ranging from 31-69% of the sample across the three time points. Humor was also among the most prevalent at 3 and 11 years (27-36%), while Intelligence was among the most prevalent at 8 years (29%). Wisdom/knowledge, Justice, Temperance, and Transcendence, all categories of the VIA framework, were reported by less than 1% of the parents at all time points. Wisdom/knowledge and Courage categories were endorsed more frequently as the children aged, whereas Interests and Specific skills were reported less frequently, suggesting a developmental nature to these themes. The use of the VIA framework allows for comparison to research outside of autism, additional research is needed to validate its use in autistic individuals, as self-reported data from youth with disabilities suggests the theoretical stucture may not be invariant across youth with and without disabilities (Shogren et al., 2017; Shogren et al., 2018).

Moreover, the developmental nature of strengths highlighted by Cost and colleagues undrescored the importance of extend examinations of strengths and positive qualities into later adolescence and adulthood. Self-report studies of strengths using the VIA Inventory of Strengths (Samson & Antonelli, 2013) indicated that Open-mindedness, Love of learning, Fairness, Persistence, and Curiosity were ranked as their top 10 personal strengths by both autistic adults and controls (Kirchner et al., 2016). Autistic adults also ranked Authenticity, Creativity, Prudence, Bravery, and Modesty, while controls ranked Humor, Kindness, Love, Teamwork, and Gratitude. Teti and colleagues (2016) explored the strengths of autistic adults through Photovoice, a technique through which autistic individuals use pictures and discussions to convey their emotions and identities. Three themes emerged from analysis of the photos and group discussions: “special interests that cultivated positive emotions and coping strategies,” “skills and activities that evoked pride,” and “reframing ASD as special versus a disadvantage.” In a recent study, Carter and colleagues (2020) asked siblings of individuals with autism and/or ID to complete the ASPeCT-DD (80% aged 16 or older; 52% had autism). Siblings were rated highly across most items, with 99% of the sample having at least one strength endorsed; 55% were endorsed to have 21 or more strengths. Kindness to others, caring for others and sense of humor being endorsed as “very” or “extremely” characteristic of 70-75% of the sample. Higher ratings were associated with being male, less frequent challenging behaviors, verbal status, siblings not living together, and the sibling not having autism. Across adult studies, there has been limited use of standardized instruments to characterize samples (e.g., IQ, adaptive skills, emotional functioning), making it difficult to contextualize findings into the broader adult autism literature (Bal & Taylor, 2019).

While these few studies have begun to explore strengths in autistic adolescents and adults, there remains a gap in our understanding of parent-perceptions of their positive qualities. The broader literature on strengths in autistic individuals has tended to focus on children, or samples including a wider age range or individuals with other developmental disabilities, making it difficult to identify potentially unique strengths of older autistic individuals. Moreover, the limited use of standardized measures has limited exploration of factors associated with different strengths, leaving an incomplete picture of how positive qualities may fit within the larger individual profile. This study targets this gap by examining the association between parent-reported positive qualities and participant characteristics, such as diagnosis (ASD, ID), demographics (i.e., sex, age) and other clinical characteristics (i.e., IQ, language level, autism symptoms, emotional and adaptive functioning).

Methods

Participants

Participants were drawn from a sample of 76 adolescents and adults recruited for a research study aimed at adapting autism diagnostic tools for use with adults. They were included in the present sample if they had a diagnosis of ASD and/or Intellectual Disability (ID) and their parent completed the Adult Behavior Checklist (Achenbach & Rescorla, 2003) or the Child Behavior Checklist (Achenbach & Rescorla, 2001). Best estimate diagnoses were based on information obtained from comprehensive diagnostic assessments completed by advanced graduate students supervised by licensed clinical psychologists. Six people were excluded due to missing surveys, and two people were excluded for not having a diagnosis of either ASD or ID, leaving 68 individuals (M age in years=19.61, SD=3.87 years; 75% male). Demographic and clinical characteristics are provided in Table 1.

TABLE 1.

Participant Demographics

	Autism n=36	Autism + ID n=14	ID n=18
	n (%)	n (%)	n (%)
Male	29 (81)	11 (79)	11 (61)
White	25 (72)	10 (71)	15 (83)
Mat Ed BA+*	27 (75)	12 (86)	16 (89)
Verbal**	34 (97)	10 (67)	6 (46)
Externalizing ***	8 (23)	4 (31)	2 (11)
Internalizing ***	21 (60)	4 (31)	2 (11)
	M (SD)	M (SD)	M (SD)
Age	19.32 (3.34)	19.14 (4.88)	20.56 (4.05)
Verbal IQa,b	92.06 (33.21)	30.50 (23.66)	35.17 (20.03)
Nonverbal IQa,b	95.80 (23.96)	45.07 (17.63)	34.61 (19.16)
ADI-R Soc+Coma,b,c	12.91 (5.84)	17.85 (6.94)	7.80 (5.32)
ADI-R RRB	4.09 (2.50)	4.62 (1.61)	3.53 (1.77)
Vineland-II ABCa,b	65.20 (12.98)	47.43 (10.57)	48.31 (15.89)
# of reported qualities	3.81 (2.35)	3.36 (2.17)	3.67 (2.25)

Note. Bold = p<.05

^*indicates participants with mothers who have received a bachelor’s degree or above

^**Overall Level of Language on the ADI-R, phrase speech or more

^***indicates participants with borderline or clinical level problems on the ABCL/CBCL (T≥63); ID: intellectual disability; ABC adaptive behavior composite

^ap<.05 Autism vs. Autism+ID

^bAutism vs. ID

^cAutism+ID vs. ID.

Groups of individuals with ID are included to better elucidate the effects of autism on reported personal strengths and challenges. There exist many stereotypes about what an autistic person may or may not be good at, especially in the context of their apparent ability levels. By separating the sample into three groups based on autism diagnosis and intellectual ability, comparisons of emerging themes can be examined. Therefore, sample was divided into three diagnostic categories: Autism-only (n=36), ID-only (n=18), and Autism+ID (n=14).

Measures

Positive Qualities.

The Child Behavior Checklist (CBCL; Achenbach & Rescorla, 2001) and the Adult Behavior Checklist (ABCL; Achenbach & Rescorla, 2003) were used to capture parent-reported positive qualities. The CBCL was completed for participants under 18 years old and the ABCL for those 18 years and older. Both forms include open-ended questions about strengths and challenges: the CBCL prompts, “Please describe the best things about your child,” and the ABCL asks, “Please describe the best things about him/her.” Parent responses to the open-ended questions were coded as described below.

Emotional or Behavioral Concerns.

The CBCL/ABCL Internalizing and Externalizing Problems scales were also used to identify individuals with borderline or clinical levels of emotional or behavioral concerns (T≥63). Both measures are well-validated (Achenbach & Rescorla, 2001; Achenbach & Rescorla, 2003) and widely used in autism research, with reliability and validity reported in some studies supporting their use in autistic samples (e.g., Pandolfi et al., 2012; Taylor et al., 2016; McCauley et al., 2020).

Autism Symptoms.

Autism symptoms were measured by the current behavior algorithm of the Autism Diagnostic Interview-Revised (ADI-R; Rutter, Le Couteur, & Lord, 2003). Consistent with previous studies (Hus & Lord, 2013), the sum of 21 items from the Social and Nonverbal Communication domains were used as an estimate of social-communicative impairments consistent with DSM-5 criteria for ASD (APA, 2013) and the RRB total (8 items) was used to reflect restricted, repetitive patterns of behaviors and interests. Social and Communication domain scores, and Restrictive and Repetitive Behavior domain scores were missing for 5 participants due to parents not completing an ADI-R.

Cognitive and Language Ability.

Verbal and nonverbal IQ scores were collected from a hierarchy of standardized cognitive assessments: Wechsler Abbreviated Scale of Intelligence, Second Edition (WASI-II; Wechsler, 2011), Differential Ability Scales-II (DAS-II; Elliott, 2007) and Mullen Scales of Early Learning (Mullen, 1995). If an individual could not attain either a verbal or nonverbal IQ score on the test normed from their age, the next test in the hierarchy was administered to derive age equivalents that were then used to compute ratio IQs (Bishop, Farmer, and Thurm, 2015). One person in the Autism-only group was missing cognitive data.

Verbal language ability.

Parent-reported language level was measured by item 30 of the ADI-R. Participants were classified as “verbal” if they scored a 0, reflecting daily use of phrase speech, and “minimally verbal” if they scored a 1 or 2, reflecting use of single words or less (Bal et al., 2016).

Adaptive Behavior.

Adaptive behavior was measured using the Vineland Adaptive Behavior Scales - 2^nd Edition Survey Interview (Vineland-II; Sparrow et al., 2005). The Vineland-II assesses parent-reported adaptive behavior in three domains: Communication, Daily Living Skills, and Socialization. The Adaptive Behavior Composite (ABC) is an age-standardized score (M=100, SD=15) which provides an estimate of adaptive functioning across the three domains. The Vineland-II is used extensively to measure adaptive behavior in autistic samples (Chatham et al., 2018; Perry et al., 2008; Yang et al., 2016) including adults (Matthews et al., 2015) and has been validated in autism samples (Sparrow et al., 2005). Adaptive behavior scores are missing for 3 participants due to parents not completing a Vineland-II.

Analyses

Coding Procedure.

Qualitative content analysis was used to derive themes from parent responses to the open-ended CBCL/ABCL questions (Elo & Kyngäs, 2008; Hsieh & Shannon, 2005). Responses generally contained multiple sentences. Sentences were parsed into phrases, each of which would be assigned a strength category. Authors L.T.V.V. and Z.L. reviewed transcripts and then used inductive coding to generate initial themes independently. Both reviewers then integrated their themes to develop coding categories (e.g., Personality characteristics) and subcategories (e.g., Kind, Humorous/playful). Revisions to the categories were made in consultation with S.W. and V.B. A codebook with operational definitions and examples for each category and subcategory was created before coding commenced (Table S1).

To establish reliability, both research assistants coded the first ten transcripts together. The next ten transcripts were coded independently and then compared; exact agreement was 92.6%. The remaining transcripts were divided between coders; every fifth transcript was independently coded and compared to protect against coder drift. Overall, 24% of transcripts were double coded. Exact agreement was 98% for categories and 93% for subcategories. Inconsistencies were reviewed and discussed until a consensus was reached between coders.

Statistical Analysis.

Analysis of Variance or T-tests were performed to assess whether the number of parent-reported positive qualities differed by diagnostic group (Autism-only, Autism+ID, ID-only), gender or levels of internalizing or externalizing problems. Pearson correlations were also used to examine associations between number of reported qualities and age, IQ and adaptive behavior. Chi-square and t-tests were used to examine associations between endorsement in each category of positive qualities and diagnostic group, as well as demographics (gender, age), and clinical characteristics (level of internalizing and externalizing problems; IQ; verbal level, autism symptoms and adaptive function) for categories endorsed for at least 15 participants. Post hoc Chi-square analyses were also performed to explore diagnostic group differences for subcategories with at least 15 participants.

Examined assumptions (normality of distribution and equal variances) were met for statistical analyses comparing groups. Due to small sample and varying group size, effect sizes were computed for each comparison to facilitate interpretation. Hedge’s g was used to estimate effect size for means comparisons of differently sized groups (g=0.2, 0.5, 0.8 were considered small, medium and large, respectively; Hedges, 1981), Cramer’s V(ϕ_c) for comparisons across categorical groups of any size (i.e., 2x3, where ϕ_c = 0.1, 0.3, 0.5 indicate small, medium and large; Cramér, 1946).

Results

On average, parental responses were 20 words (SD=21.33). The majority of parents reported at least one positive quality (n=66; 97%), with most parents reporting multiple (89%; M=3.68; SD=2.26; range=0-14). There was a tendency for parents completing the ABCL to report more strengths (M=4.05, SD=2.54) than parents completing the CBCL (M=3.04, SD=1.54; t(66)=−1.80, p=0.08, g=0.48).

Six categories of qualities emerged from the qualitative analysis: Personality characteristics, Sociability, Work ethic/motivation, Specific skills, Interests/activities, and Other. Strengths that fell into the Other category include responses such as, “physically fit” and “always on time.” Across diagnostic groups, parent-reported qualities most commonly fell into the Personality characteristics (82%) and Sociability (53%) categories (Table 2). Parents often reported multiple Personality characteristics, accounting for 59% of the total number of positive qualities reported. The number of positive qualities reported was not associated with diagnostic group (F(2, 65)=.19, p=.83) or any demographic or clinical characteristics (Table S2 and Table S3).

TABLE 2.

Categories and subcategories of strengths and challenges.

Category	%	Subcategory	Examples
Personality characteristics (82%)	22	Affectionate	“loving”
	32	Kind	“good heart”
	27	Humorous/playful	“loves to have fun”
	27	Positive affect	“happy”, “smiles and laughs a lot”
	3	Curious	“curious”
	4	Creative/unique	“he is his own person”
	12	Moral compass	“clear concept of right and wrong”
	16	Intelligent	“intelligent”, “smart”
	24	Amenable/flexible	“easy going”, “agreeable”
	16	Amiable	“people love him”
Sociability (53%)	22	Helping behaviors	“helps with what he can”
	6	Social skills	“great conversationalist”
	9	Emotion awareness	“in-tune to other people’s feelings”
	27	Seeking interaction	“seeks people out to have conversations”
Work ethic/motivation (25%)	7	Wants to succeed	“enjoys accomplishing a goal”
	9	Work ethic	“sets his mind and makes it happen”
	9	Perseverance	“he picks himself up and keeps going”
	7	Independence	“tries to be independent”
Specific skills (27%)	3	Technological skills	“very technologically savvy”
	4	Artistic skills	“a great artist”
	2	Musical skills	“know all kinds of music”,
	4	Visual-spatial skills	“he can see a piece and know its spot”
	12	Memory	“an impeccable memory”
	2	Problem solving	“good problem-solving skill”
	7	Academic	“is very smart especially in world affairs”
	2	Coping skills	“copes with bullying”
Interests/activities (13%)	13	Hobbies	animals, cooking, travel, movies, sports
Other (4%)			“physically fit”, “always on time”

Note. Category (%) indicates proportion of entire sample endorsed in that category; percentages in % column indicate proportion endorsing the subcategory within those endorsing that category.

Associations between positive qualities and diagnostic groups

Personality characteristics

As shown in Table 3, diagnostic groups did not differ in the proportion of individuals with reported Personality characteristics (X²(2)=.27, p=.87). Those with a strength endorsed in this area had higher levels of repetitive behaviors on the ADI-R (M=4.29, SD=2.18) than those without positive qualities in this category (M=2.60, SD=1.84; t(59)=−2.30, p=0.03, g=0.79). No other participant characteristics were associated with this quality (Table S4).

TABLE 3.

Categories and subcategories by total sample and diagnostic group

		Total Sample n=68	Autism n=36	Autism +ID n=14	ID n=18
Category	Subcategory	%	%	%	%
Personality characteristics		82	81	86	83
	Affectionate	22	17	36	22
	Kind	31	39	29	17
	Humorous/playful	25	25	36	17
	Positive affect	27	17	29	44
	Curious	3	3	0	6
	Creative/unique	4	8	0	0
	Moral compass	12	19	7	0
	Intelligent	16	22	7	11
	Amenable/flexible	22	19	21	28
	Amiable	16	11	21	22
Sociability		53	56	29	67
	Helping behaviors	22	28	14	17
	Social skills	6	8	0	6
	Emotion awareness	7	14	0	6
	Seeking interaction	25	17	14	44
Work ethic/motivation		25	33	21	11
	Wants to succeed	7	11	0	6
	Work ethic	9	8	14	6
	Perseverance	9	14	0	6
	Independence	7	6	14	6
Specific skills		27	33	36	6
	Technological skills	3	6	0	0
	Artistic skills	4	6	7	0
	Musical skills	2	3	0	0
	Visual-spatial skills	4	3	7	6
	Memory	12	14	14	6
	Problem – solving	2	0	7	0
	Academic	7	11	7	0
	Coping skills	2	3	0	0
Interests/activities	Hobbies	13	6	14	28
Other		4	6	0	6

Bold = p<.05; Italics = p<.10; %s indicate proportion of that category/subcategory endorsed within the whole sample or each diagnostic group. Some participants endorsed traits within more than one subcategory; therefore the sum of subcategory endorsements may exceed % of overall category.

As shown in Table 2, the most commonly reported subcategories were Kind (32%), Positive affect (27%) and Humorous/playful (27%). Endorsements did not differ by diagnostic group for any positive quality in the Personality characteristics category (Table 3).

Sociability

There was a marginal effect of diagnostic group (X²(2)=5.67, p=.05, ϕ_c=.29), reflecting a lower proportion of the Autism+ID group having Sociability as an endorsed strength relative to other groups (Table 3). No other clinical characteristics were associated with endorsements in the Sociability category (Table S5).

Among the four Sociability subcategories, Helping behaviors (22%) and Seeking interaction (27%) were the two most commonly endorsed. As shown in Table 3, a greater proportion in the ID-only group reported Seeking social interaction than either Autism group (X² (2)=7.77, p=0.02, ϕ_c=0.34). The other subcategories did not differ by diagnostic group (p-values=.15-.52).

Work ethic motivation

Diagnostic groups did not differ in the proportion of individuals with reported Work ethic/motivation (X²(2)=3.21, p=.20, ϕ_c=.22; Table 3). Those with qualities endorsed in this category were older (M=21.29, SD=4.20; t(64)=−2.09, p=0.04, g=.59) than those without Work ethic/motivation endorsed (M=19.07, SD=3.62) and had higher NVIQ (M=84.31, SD=36.47; t(63)=−2.09, p=.04, g=.60) than those without the endorsement (M=63.29, SD-34.46). Having a reported strength in Work ethic/motivation was also associated with being male (X²(1)=4.73, p=.03, ϕ_c=.27), verbal (100% vs. 71%; X²(1)=4.23, p=0.04, ϕ_c=31) and having borderline/clinical level of internalizing problems (X²(1)=4.10, p=.04, ϕ_c=.25). Endorsements in Work Ethic/motivation were not associated with other demographics or clinical characteristics (Table S6).

Specific Skills

As shown in Table 3, Specific skills were somewhat more likely to be reported in the Autism and Autism +ID groups than the ID-only (X²(2)=5.29, p=.07; ϕ_c=.28); when the two Autism groups were collapsed, the effect of diagnosis (Autism vs. ID-only) was significant (X²(1)=5.28, p=0.02, ϕ_c=0.28). Those with Specific Skills were more likely to have a borderline/clinical level of internalizing problems (72% vs. 37%; X² (1)=6.46, p=.011), and higher nonverbal IQ (M=84.29, SD=26.06) than those without ( M=62.40, SD=37.44; t(64)=−2.27, p=.027, g=.63). Endorsement of Specific skills was not associated with demographics or other clinical characteristics (Table S7).

Interests or activities

Diagnostic groups did not differ in the proportion of individuals with reported Interests or activities (X²(2)=2.84, p=.07, ϕ_c=.29; Table 3). Given that only 9 participants endorsed this category, further analyses were not conducted.

Discussion

Consistent with previous studies (Cost et al., 2021; Carter et al., 2020), positive qualities were endorsed for the majority of participants, with 97% of all parents reporting at least one and an average of 3 strengths reported across groups. The number of positive qualities did not differ by participant demographics or clinical characteristics, including verbal level and IQ. Previous studies have found that having speech and fewer challenging behaviors predicted more strengths reported by caregivers (Carter et al., 2015). Similarly, siblings reported higher ratings of overall strengths on the ASPeCT-DD when their sibling was verbal, male, had less frequent challenging behaviors and did not have autism (Carter et al., 2020) and lower levels of parent-reported thriving were associated with social-communicative abilities (Weiss & Burnham-Riosa, 2015). Differences may be attributable to differences in measurement (i.e., use of standardized scales of internalizing and externalizing symptoms or the clinician-rated ADI-R to quantify verbal ability and social-communication), compared to ratings of a single item or set of survey items to characterize speech, challenging behavior or social-communicative abilities in previous studies). The present study may also be underpowered to detect group effects. Nonetheless, the prevalence and variability of positive qualities across all groups highlights the importance of assessing strengths for all individuals across the ability range.

Six categories emerged from qualitative analysis of their reponses, highlighting a range of strengths and positive qualities. Themes were similar to previous studies (Carter et al., 2015; Colavita et al., 2014), though arrangement of positive qualities into categories varied somewhat from previous studies. Notably, the Sociability category emerging from our qualitative analysis is similar to Carter and colleagues’ (2015) qualitative analyses, in that the focus includes behavioral features of sociability (e.g., helping behaviors). In contrast to Carter’s analyses of the ASPeCT-DD, which includes traits such as kindness in social categories (i.e., Positive Relations), we placed dispositions (e.g., loving, good heart) in a broader category of Personality Characteristics. The latter have been termed social personality characteristics (Colavita et al., 2014) or fall under the Humanity category in the VIA framework (Cost et al., 2021) in other studies. Because this was a first analysis specifically of autistic adolescents and adults, we felt that it was useful to remain open to new categories and broad in our conceptualizations, rather than restrict analyses to predefined frameworks. Larger studies with sufficient sample size to conduct empirical analyses will be needed to inform categorization of traits.

Although encompassing somewhat different traits, consistent with previous studies (Carter et al., 2015; Colavita et al., 2014; Cost et al., 2021), Personality characteristics emerged as the most commonly reported strength, endorsed for more than 80% of each of the diagnostic groups. Importantly, this category was not associated with any demographic or clinical characteristics. Previous work identifying strengths of those with disabilities has argued the importance of identifying and building on these characteristics (Niemiec et al, 2017), so it is promising to see them readily identified by parents. Our exploratory approach also resulted in a new category not clearly covered in previous studies: Work ethic/motivation, which included subcategories capturing a desire to succeed, work ethic, perseverance and strides to be independent. The emergence of this category could be a reflection of our older, transition-age sample. Indeed, this was the only category for which we observed an effect of age, with those endorsing strengths in this area tending to be older. Higher IQ and verbal level were also associated with endorsement of this category, perhaps reflecting that some features (e.g., Wants to succeed) may be more readily observable in someone who can express them verbally.

While Cost and colleagues (2021) analyzed data from the same instrument/question, their application of the VIA framework and focus on younger school-age children makes comparisons somewhat difficult. It is interesting to note that while only 19% of their sample endorsed “Friendliness” (a trait which they added based on their qualitative analysis), 48% of our autistic sample (57% Autism-only; 29% Autism+ID) and 67% of our ID-only group were endorsed as demonstrating a strength in our Sociability category, which encompassed social-seeking and helping behaviors that likely comprise “friendly” behaviors. This is more consistent with Carter and colleagues (2015; 2020) analyses of the ASPeCT-DD, in which kindness and caring for others were amongst the endorsed for more than 80% of participants. The present study and both Carter studies included primarily adolescents and adults; thus, this may reflect a developmental difference in samples suggesting the emergence of prosocial behaviors as strengths as individuals with Autism and ID get older. Given how frequently Sociability and Work ethic/motivation were reported, these themes should be included in future assessment tools used to examine strengths and positive qualities of this population.

Considering that social difficulties are a core diagnostic feature of ASD, it may be somewhat surprising that Sociability was endorsed for more than half of the Autism-only group. A closer look at subcategories reveals that helping behaviors and seeking of interaction were the most commonly endorsed qualities. Social skills, on the other hand were endorsed by very few participants. These findings serve as an important reminder that while there is often emphasis on social “skill” (i.e., how skilled someone is in navigating interactions), the application of prosocial behaviors (e.g., trying to help or simply initiate interaction) may be just as important to promoting others’ positive perceptions of a person and facilitating interactions. While social “strengths” in autism are sometimes defined based on the absence of an expected difficulty or “symptom” (e.g., Bal et al., 2019), the themes that emerged in these analyses are a reminder that focusing on the presence of specific positive qualities will promote strengths-based approaches and provide insights into areas that can be further developed. It will also be important to recognize that autistic individuals may have ways of displaying prosocial behaviors and/or indicating social interest (e.g., seeking interaction) that may be missed or misinterpreted by parents, underscoring a need for autistic voices to be included in this research.

It has been widely documented that many autistic individuals have extraordinary talents or savant skills. The frequency of endorsed Special skills, especially in those with Autism (34%) or Autism+ID (36%) in this sample is within the range reported by other studies (Bal et al., 2021; Howlin et al., 2009; Meilleur et al., 2015; Rimland, 1978). Also consistent with previous studies (Bal et al., 2021; Treffert, 2009), memory was the most frequently reported, particularly for those with autism. While many subcategories were endorsed for only 1-2 participants and could have been collapsed (e.g., perhaps artistic skills and visual spatial skills), the variability of skills highlighted by parents is an important reminder that the autism spectrum is not only characterized by a range of difficulties, but also encompasses a heterogeneous group of strengths and talents.

An association between Borderline or Clinical levels of Internalizing symptoms and both Work ethic and Specific skills categories was observed. It is not immediately apparent what may explain the results of the present study. First these results must be interpreted with caution, considering studies suggesting that the CBCL may not demonstrate measurement invariance across children with and without ID (Dovgan et al., 2019) and that the ABCL may overidentify adults (Gotham et al., 2014). It is possible that those with high work motivation or specific skill levels are more prone to internalizing problems (e.g., depression or anxiety) if they do not have opportunities to work toward their goals and/or utilize their skills. Previous research with adults highlights many barriers to identifying appropriate employment and frustrations with available supports (Miller et al. 2018), which can lead to demoralization (Nicholas & Klag, 2020). Thus, the association in this study could reflect a mismatch between goals or skills and opportunities for pursuit. This is consistent with the emphasis on need for individual and environmental supports highlighted in Weiss & Burnham-Riosa’s (2015) study of thriving in youth with ID. Additional research is needed to explore how discrepancies between individual strengths and opportunities to meaningfully use those skills may influence mental health. Alternatively, the association may reflect something about how the individual’s presentation impacts parent reporting. Cost and colleagues (2021) reported that children with higher levels of internalizing and externalizing behaviors were less likely to have positive qualities endorsed in the Humanity category (i.e., traits such as kindness, love or friendliness). While the present relationship was opposite (i.e., more internalizing behaviors, more likely endorsement of positive qualities in Work ethic/motivation or Specific skills), more research is needed to understand whether these association reflects how internalizing or externalizing behaviors affect parents’ perceptions of their child or whether individuals with specific qualities may be at greater risk for emotional difficulties.

Overall, it was somewhat remarkable how few associations were observed between clinical characteristics (e.g., IQ, adaptive function) and the positive quality categories. While small sample size may have limited statistical power to detect differences, lack of association with clinical characteristics further emphasizes a need to not only characterize strengths in autistic individuals, but to better understand how these positive qualities may relate to real-world outcomes (e.g., social participation). It is possible that positive qualities are key contributors to attainment of goals and desired outcomes and may even explain some of the variability in autistic adult outcomes. Thus, increasing focus on promoting strengths (as opposed to reducing impairments) likely will provide insights into new ways to support autistic adults to achieve their goals. Inclusion of strengths-based assessment tools is needed to gain a more holistic understanding of individuals and likely foster community participation (Carter et al., 2020). Moreover, helping autistic individuals to be aware of their own strengths has important implications for self-esteem and mental health.

While these results add to the growing literature on strengths and positive qualities in individuals on the autism spectrum and those with intellectual disability, findings must be considered in the context of several study limitations. First, this study was based on a small convenience sample, drawn from a larger study. The ABCL and CBCL were not specifically designed to measure positive qualities and therefore may not be sensitive to identifying strengths and positive qualities. The phrasing of “best” may have limited some parents to only reporting a few positive aspects about their son or daughter. Thus, lower levels or a lack of endorsement in some subcategories should not be interpreted as evidence that these individuals do not exhibit these qualities; rather, they may simply not be the most prominent features that come to parents’ minds. Additionally, the smaller answer space provided in the CBCL (compared to the ABCL) for parents to write their response may have affected reporting. Even though we found a marginal difference in the number of positive qualities (an average of 3 positive qualities for CBCL vs. 4 positive qualities for ABCL), the lack of age differences across analyses suggests this did not strongly affect results. Secondly, the small sample limited statistical power (particularly across three groups); given the limited research in this area and exploratory nature of this study, corrections for multiple comparisons were not made. Effect sizes are included to aid in the interpretation of the results. Further, the sample is primarily comprised of white males with college-educated mothers, and responses were limited to parent-report. Unfortunately, specific characteristics that may affect caregiver reporting, such as family composition and parental mental health were not available. Though number of children was not a significant predictor of parent-reported strengths (Carter et al., 2015), residential status has been shown to impact sibling reporting (Carter et al., 2020). As with much of autism research, greater effort is needed to recruit more diverse samples and future research should ask autistic adults and adults with other intellectual and developmental conditions to provide their perspectives on their own strengths. Where caregiver reports are gathered, further characterization of informants is needed. Finally, caution is warranted when interpreting associations with internalizing/externalizing scales, as previous research has suggested internalizing scales may overidentify autistic adults (Gotham et al., 2014) and possible measurement variance in children with and without ID (Dovgan et al., 2019).

Limitations notwithstanding, this study contributes important information on positive qualities of autistic individuals. Several categories, such as Specific skills and Interests/activities, have already been identified and explored in relation to autistic individuals, though historically through a lens of impairment (i.e., as circumscribed interests). Other categories, such as Work ethic and Personality characteristics, have not been as widely explored. These findings add to the growing body of literature calling for greater focus on strengths of autistic adults (e.g. Burnham Riosa et al., 2017; Kapp, 2018; Shogren et al., 2021) and underscore a need to systematically characterize positive qualities alongside symptoms or difficulties. Recent studies highlight a relationship between the number of positive attributes individuals associated with autism and their self-esteem (Cooper et al., 2020). While not all positive qualities highlighted here are directly related to autism itself, positive self-evaluations can contribute to feelings of self-worth and motivate individuals to persist in the face of challenge (Nguyen et al., 2020). Parents also highlight the positive contributions of their autistic children to their views of their family and society more broadly (King et al., 2012). Thus, helping others to recognize strengths in autistic individuals should be seen as an important element of fostering understanding of autism and promoting acceptance of autistic people. A more wholistic view may help to identify areas that can be leveraged to foster self-determination and personalized supports to promote goal-oriented outcomes (Shogren et al., 2021) and quality of life more broadly.

Though there is growing attention to the strengths of autistic individuals, this study is among the first, to our knowledge, to describe parent-reported strengths and positive qualities of autistic adults separately from adults with other neurodevelopmental conditions. Surprisingly few associations with demographics or clinical characteristics are a poignant reminder that everyone has strengths, regardless of their level of challenges. It is perhaps particularly important that research emphasizes the identification of positive qualities that capture the presence of specific behaviors or skills which can be built upon, rather than focusing on the absence of a specific difficulty commonly associated with autism. Better understanding and greater recognition of these positive qualities is likely to provide important insights into ways to promote positive self-images and mental health for autistic individuals and contribute to improved public understanding and greater acceptance of autistic individuals.