Subjective wellbeing of autistic adolescents and young adults: A cross sectional study

Subjective wellbeing (SWB) represents an individual's perception of wellness that is supported by homeostatic mechanisms. These mechanisms are proposed to be maintained by low negative affect and high positive affect, although less is known about these mechanisms and SWB in autism. The current cross‐sectional study aimed to compare patterns of positive affect, negative affect (Positive Affect and Negative Affect Scale), and SWB (Personal Wellbeing Index‐School Children) between autistic (n = 53) and non‐autistic (n = 49) individuals aged 10–22 years (Mage = 13.97, SD = 3.13). Between‐group t‐tests revealed that compared with same‐age peers, autistic participants scored lower SWB overall (p < 0.001). In both groups average SWB scores fell into the higher range, however, autistic participants were three‐times more likely to fall below this range when compared to non‐autistic participants. Negative affect had a higher intercept in the autistic sample, but no difference in slopes were observed. A hierarchical multiple regression revealed that diagnosis, positive affect, and negative affect significantly predicted SWB in our sample. Between‐group t‐tests found no significant difference in positive affect or negative affect across age between the autistic and non‐autistic samples. In autistic participants, positive affect increased across age as SWB decreased, whilst negative affect remained stable, a pattern inconsistent with homeostatic SWB. The current study is overall consistent with the homeostatic explanation for SWB within autism; however, we identified potential differences between autistic and non‐autistic participants in the contribution of positive affect and negative affect to homeostatic protect mood across development.


INTRODUCTION
Autism spectrum disorder (hereafter autism) is a neurodevelopmental condition that persists throughout the lifespan (Guthrie et al., 2013).Worldwide prevalence rates of autism diagnoses are reported to be between 1.0% (Zeidan et al., 2022) and 1.68% (Baio et al., 2018), with estimates in the USA as high as 2.78% (Maenner et al., 2020).Autism is characterized by differences in social communication together with repetitive and restrictive behaviors (American Psychiatric Association, 2022).Autism is typically identified and diagnosed in early childhood (Carbone et al., 2020), however, characteristics of autism can remain unrecognized until an individual's social abilities no longer appear to meet age-normed expectations (Lauritsen, 2013) or they demonstrate difficulties with affective control and emotion dysregulation (e.g., intense meltdowns; Samson, Hardan, Podell, et al., 2015).Autistic individuals are more likely to experience mental health difficulties, including depression (Hedley & Uljarevi c, 2018), chronic physical health conditions (Cashin et al., 2018), reduced quality of life (Graham Holmes et al., 2020), and increased mortality (Hwang et al., 2019) compared to non-autistic individuals.There is a need for research that examines the mechanisms leading to these adverse outcomes for autistic individuals.In this study, we examined potential predictors of subjective wellbeing (SWB), otherwise defined as a selfreported internal sense of wellness (Cummins, 2010), in autistic and nonautistic adolescents and young adults.
SWB is a homeostatic mechanism (i.e., an internal system that regulates wellbeing within set limits, like body temperature; Torday, 2015) protected by cognitive resources (i.e., personal interpretation of life stressors; Campbell et al., 2013) which help to reduce negative mood and support positive mood.Together, these mechanisms and resources are thought to protect against depression (Cummins, 2018), promote better physical health and longevity (Steptoe et al., 2015).Richardson et al. (2016) proposed that some populations, such as those vulnerable to poorer mental health outcomes, may be predisposed to lower SWB.The causal relationship between SWB and mental health continues to be a focus of research in many populations (Gargiulo & Stokes, 2009;Li et al., 2023), yet in terms of autism, there is limited research concerning SWB and the mechanisms underpinning it.Albeit limited, there is some evidence that autistic people experience significantly lower SWB compared to non-autistic people from childhood (Begeer et al., 2017) through to adulthood (Thorpe, 2018), with the exception of late adolescence as the literature has focused on early adolescence.Further, higher SWB is associated with lower depressive symptoms among autistic adults (Hedley et al., 2019), which suggests that homeostatic SWB may promote better mental health outcomes in autism.Should positive SWB be protective for autistic individuals, interventions that promote positive SWB could be implemented in an effort to improve longterm outcomes within this population.
According to Cummins' homeostatic theory of SWB, a balance of high positive affect and low negative affect (Blore et al., 2011;Busseri, 2018;Tomyn & Cummins, 2011) maintains SWB within a narrow range of scores that represents Homeostatic Protective mood, a genetically determined positive mood (Cummins, 2013).This theory proposes that most people have a biological predisposition for positive wellbeing, which when measured on a 100-point scale (International Wellbeing Group, 2013) is typically represented by scores above 50-points, otherwise known as the positive range (Cummins, 2013).Westernized populations most commonly score within a set-point range, 70-90 points (Cummins, 2018), and individuals may fluctuate within this range in response to stressful life events.Positive affect is described as pleasant arousal that is characterized by being alert and active and is associated with positive emotions.Negative affect accesses negative emotions and is considered as unpleasant arousal represented by anxiety and anger.(Cummins, 2018;Russell, 2009;Tugade & Fredrickson, 2007).Thus, positivity, alertness, happiness, and contentment are proposed to underpin Homeostatic Protective mood and maintain SWB, which promotes both emotional stability and stable wellbeing (Gonz alez-Carrasco et al., 2016a).Russell (2009) theorizes that positive affect and negative affect are influenced by the intense shifts in emotion and mood experienced by people in response to daily life events, which is likely to cause fluctuations in Homeostatic Protective mood.Some individuals respond to negative life events with strong negative emotions (Yang et al., 2021), which if maintained, can develop into negative mood (i.e., sustained negative state), causing Homeostatic Protective mood to weaken, and in turn SWB may drop (Cummins, 2018).Thus, at times, when negative events persist and resources that support positive affect are insufficient, sustained low SWB may develop (Cummins, 2010) leading to adverse outcomes (Cummins, 2018;Steptoe et al., 2015).
The stability of Homeostatic Protective mood and SWB is protected by the cumulative effect of resources referred to by Cummins (2018) as external and internal buffers, which provide homeostatic resilience by supporting low negative affect and high positive affect (see Figure 1).These proposed external buffers are (i) money, which provides individuals access to necessary resources, (ii) relationships, which offer support, connection, and belonging, and (iii) achieving in life, which promotes purpose and value (Cummins, 2018).The proposed internal buffers are (i) self-esteem, which protects from negative experiences (Thorpe, 2018), (ii) optimism, which drives expectations of a positive future (Carver et al., 2010), and (iii) perceived control over life events (Lai & Cummins, 2013).When resource access is limited longterm, Homeostatic Protective mood is weakened and less able to defend SWB following challenging life events (Cummins, 2018).
For autistic adolescents, aged 11-to 17-years-old, fluctuations in positive affect and higher negative affect are implicated in a greater number of depressive symptoms (Dallman et al., 2022).This may be because the quality of one's social interactions can disrupt the balance of positive affect and negative affect momentarily in the non-autistic population (Dallman et al., 2022).Where autistic social characteristics can lead to difficulties in socializing, it is likely that repeated negative social interactions, for some autistic people, from adolescence (O'Hagan & Hebron, 2017) and into adulthood regularly disrupt affect, and thus, resource acquisition (Bailey et al., 2020).It is therefore plausible that some autistic individuals may be more prone to reduced resources, frequent negative experiences combined with less stable mood may weaken Homeostatic Protective mood and low SWB when faced with challenging situations.Cummins (2010) argues lifelong stability of SWB; however, some evidence suggests a decrease in SWB from childhood, throughout adolescence (Buecker et al., 2023), and into early adulthood (e.g., Tomyn & Cummins, 2011) in both autistic and non-autistic people.In a sample of 940 non-autistic adolescents, Gonz alez-Carrasco, Casas, Malo, et al. (2016) found that SWB gradually decreased for non-autistic adolescents aged 11to 15-years-old.Similar patterns are observed among autistic children and young adolescents (Begeer et al., 2017) as well as young adults (Kojima, 2020), yet there is a lack of evidence for SWB patterns from mid to late adolescence.It is likely that SWB may be vulnerable to developmental changes that occur during adolescence and young adulthood; for example, increased complexity in social interactions, changes in parent and sibling relationships (Steinberg & Morris, 2001), executive function maturation, puberty (Steinmayr et al., 2018), and adaptive functioning (Picci & Scherf, 2015).This may in turn contribute to limited acquisition of the resources, which are known to protect SWB (Bailey et al., 2020).Thus, we contend that SWB is more likely to be overwhelmed during adolescence for autistic individuals, yet this hypothesis is so far unexamined within the literature.We propose that understanding the patterns of SWB, positive affect and negative affect in adolescence and young adulthood is necessary to inform psychological and other supports that may be most beneficial for young autistic people, thereby enhancing health and wellbeing outcomes in adulthood.
F I G U R E 1 Homeostatic Defense of SWB in Relation to the Strength of Challenge Experienced.Stronger life challenges necessitate stronger resources to protect the homeostatic mechanisms that underpin SWB.Stronger resources help to maintain SWB within an individual's positive set point range to protect against adverse outcomes.Adapted from Cummins (2014).

The current study
The aim of the current study was to examine the agerelated patterns of SWB, positive affect and negative affect that occur in autistic and nonautistic adolescents and young adults.As such, we hypothesized that (H1) autistic participants' overall SWB scores will be lower, the increase with older age will be smaller (the slope), and the mean SWB scores at the youngest age (the intercept) will also be lower than in nonautistic participants', (H2) the positive affect increase over age (the slope) will be smaller, and the mean at the youngest age will be lower (the intercept) in the autistic sample compared to non-autistic group, and (H3) autistic participants' negative affect change over time (the slope) will be larger, and the intercept will be higher than nonautistic participants'.

METHOD Design
A cross-sectional design was utilized to analyze the agerelated differences in SWB, positive affect and negative affect between the autistic and non-autistic adolescents and young adults.Analyses involved t-tests to compare group differences in SWB, positive affect and negative affect and a multiple regression analysis to determine group differences in slope and intercepts over SWB, positive affect, and negative affect.Data for this project were collected between 2013 and 2019 from autistic and nonautistic adolescents and young adults using purposive sampling (i.e., by recruiting autistic and nonautistic participants matched by age and sample size).This study received ethical approval from the Deakin University Human Research Ethics Committee.

Participants
The final sample included 102 adolescents and young adults aged between 10 and 22 years old, 53 of whom were autistic (Mage = 13.89,SD = 3.11; 36 male), and 49 non-autistic (Mage = 14.06,SD = 3.19; 28 male).All autistic participants reported a clinical autism diagnosis without intellectual disability (i.e., >70 intelligence quotient).Data for gender were not captured, only sex reported at birth.

Materials
The Personal Wellbeing Index-School Children (PWI-SC) fourth edition (Cummins & Lau, 2023) is a modified version of the Personal Wellbeing Index-Adult fifth edition (International Wellbeing Group, 2013), which is a self-report measure of SWB.Participants rate seven items on an 11-point Likert scale where each item represents one of seven life domains (i.e., standard of living, personal health, achievement in life, personal relationships, personal safety, feeling part of the community, and future security), which comprise the seven subscales of the PWI-SC.For all scale items, responses range from 0 (very sad) to 10 (very happy).A participant's overall SWB score is derived from the sum of the seven domains, divided by seven, and multiplied by 100; higher scores indicate higher levels of SWB.Individual SWB scores fall on a continuous scale of between 0 and 100, and both individual and group scores are referenced against normative data for interpretation (International Wellbeing Group, 2013).This measure has been validated for use in both non-autistic and autistic samples (Tomyn et al., 2013).
The Positive and Negative Affective Scale (PANAS) is a self-report measure of mood as characterized by two orthogonal scales of positive affect and negative affect (Watson et al., 1988).Twenty terms that represent feelings and emotions (e.g., distressed, interested, active and proud) and are rated by how strongly they are being experienced by the participant in that moment on a 5-point Likert scale between 1 (very slightly or not at all) and 5 (extremely).Ten terms relate to positive affect and negative affect respectively, with scores ranging from 10 to 50 for each affective type, and higher scores reflect higher degrees of each affect type.This scale has been validated for use in both the autistic population and nonautistic population (Samson, Hardan, Lee, et al., 2015).

Procedure
Participants were recruited using advertisements in psychology clinics and on social media.After written consent was obtained by both participants and, where relevant, the caregiver of participants under 18-years of age, participants completed a suite of self-report pencil and paper measures over a one-hour period.Data were collected by two trained researchers in the participant's home or at the University campus using a standardized protocol.Only measures pertaining to the research question were analyzed.
Variable means, standard deviations, and Cronbach's alpha reliability coefficient for measures are presented in Table 1.The PWI-SC reliability was only at acceptable levels for non-autistic participants.Reliability for negative affect and positive affect for the entire sample were acceptable, and the intercorrelation of positive affect and negative affect was moderate (r = À0.55).
All correlations between variables were lower in the autistic group compared to the non-autistic group, as shown in Table 2. Correlations between positive affect and negative affect in both groups were moderate and positive.In the non-autistic group PWI-SC scores were negatively correlated with negative affect and positively correlated with positive affect, and both correlations were small.In the autistic group, a small positive correlation was found between PWI-SC scores and both negative affect and positive affect.
To test the age-related differences in slope and intercept between the autistic and non-autistic groups in SWB, positive affect and negative affect, tests of slope and of intercept were used (Zar, 1999).Table 3 reports a significant group difference in the intercept of PWI-SC for age, revealing a moderate effect size.At the intercept of age, negative affect is significantly higher in the autistic group than the non-autistic group.There was no significant group difference in either slope or intercept in positive affect for age.
Figure 2 shows that while the two groups significantly differ at the intercept, they trend towards each other in PWI-SC scores across age, as scores in the autistic group (b = 0.02) increase and scores in the non-autistic group (b = À0.19)decrease.Figure 2b shows the autistic group slope is apparently greater across age than the non-autistic group.Figure 2c shows that negative affect in the non-autistic group (b = 0.02) slope increase is steeper than the autistic group (b = 0.01).
Figure 3 shows that the effect of negative affect on PWI-SC was higher at the intercept for the non-autistic group compared to the autistic group.Negative affect scores decreased as PWI-SC scores increased in the nonautistic group and negative affect scores increased as PWI-SC scores increased in the autistic group.In both groups positive affect scores increased as PWI-SC scores increased.
To compare the proportion of autistic adolescents and young adults to non-autistic adolescents and young adults scoring <70 in SWB a proportions test was conducted (Zar, 1999).Autistic participants were 2.99 (+/À 95% CI 1.24-7.18)times more likely to score < 70 points on the PWI-SC than non-autistic participants (Z = 2.37, p < 0.01).
Multiple regression analysis exploring diagnosis on SWB after controlling for positive affect, negative affect, age, and sex.Age and sex were entered at step one, with diagnosis added at step two, then positive affect at step three, and negative affect at step four.Results revealed that diagnosis explained 9% of variance (F(2, 100) = 10.37,p = 0.002).Positive affect contributed a further 4% of variance, (F(2, 100) = 7.77, p = <0.001),while negative affect added 3% of variance (F(2, 100) = 6.63, p = <0.001).Overall, the model explained 16% of variance.
To explore PWI-SC subscale differences by diagnosis supplementary t-tests analysis were conducted.Table 4 reveals that the autistic adolescent group scored significantly lower on five subscales of the PWI-SC compared to the non-autistic adolescent group.These included personal health, personal relationships, personal safety, community connectedness, and future security.

DISCUSSION
The current study compared the age-related patterns of autistic and non-autistic adolescents' and young adults' SWB, positive affect, and negative affect, and examined how strongly positive affect and negative affect explained the SWB of our sample.We hypothesized that (H1) compared to non-autistic participants, autistic participants' overall SWB scores would be lower, that the increase in SWB with older age would be smaller, that the mean SWB scores at the youngest age would be lower.We further hypothesized (H2) that the positive affect increase over age would be smaller, and that the intercept would be lower in the autistic group compared to the non-autistic group.Finally, we hypothesized (H3) that autistic participants' negative affect change over time would be larger, and that the intercept will be higher than non-autistic participants.
Hypothesis one was partially supported; significantly lower SWB was found in the autistic group compared to the non-autistic group suggesting they may be more vulnerable to lower SWB.This may be associated with difficulties in developing important internal and external buffers.Additionally, autistic participants had a significantly lower intercept for SWB than non-autistic participants, consistent with the decrease in SWB observed in autistic early adolescents, aged 8-to 14-years old, compared to non-autistic peers (Begeer et al., 2017).Together, these findings suggest that the development of resources necessary for Homeostatic Protective mood and SWB may be delayed from childhood among autistic individuals.Inconsistent with H1, the autistic group's SWB increased across age compared to a decrease in the nonautistic group, however these findings hide meaningful comparisons between our groups.Consistent with previous findings (e.g., Tomyn et al., 2015) the mean non-autistic group' SWB scores fell within the center of the Australian normative SWB range of 70-to 90-points.In contrast, the mean autistic participant SWB scores fell at the bottom of this range and weaken Homeostatic Protective Mood.Furthermore, the proportion of autistic participants who scored below 70 points (i.e., lower end of the dissatisfied-satisfied range) was significant, with autistic participants' being nearly three times more likely to score below the expected positive SWB range.
Hypothesis two was unsupported by our data whereby there were no significant between-group differences in positive affect at intercept or slope.The unexpected similarities found between groups may reflect diagnostic differences in life experiences.Steinmayr et al. (2018) suggest that adolescence may be characterized by a gradual reduction in positive emotional experiences, which may be due to the increased challenges and stressors associated with this developmental period.Compared to non-autistic adolescents, autistic adolescents tend to spend more time engaged in preferred interests that elicit positive affective responses (Kovac et al., 2016;Rivard et al., 2018).Thus, although most autistic adolescents have fewer positive social experiences than non-autistic adolescents (Maïano et al., 2016), they often spend more time alone, engaged in preferred activities (Folta et al., 2022) which may lead to similar patterns of positive affect to non-autistic adolescents.
Hypothesis three was mostly unsupported by the current results.Findings from our study supported significantly higher negative affect in autistic participants at the intercept, however, a non-significant difference was found across age between diagnostic groups.The slopes for both groups appeared to be largely similar, and the age-related change was not as anticipated based on findings in 12-to 20-year-old adolescents and young adults (Tomyn et al., 2015;Tomyn & Cummins, 2011).Although significant, positive affect and negative affect were minimal predictors of SWB with diagnosis a stronger significant predictor, however the negative affect mean was lower in our autistic sample.Overall, when compared to non-autistic participants, autistic participants demonstrated an inconsistent pattern of negative affect and positive affect, a variation which requires further research given the group differences in SWB also found within our study.
Autistic participants in our sample were more likely to have significantly lower scores in four out of the seven subscales of the PWI-SC compared to non-autistic participants, including personal health, personal relationships, personal safety, community connectedness, and future security.These results are consistent with Roestorf et al. (2022) who found autistic adults scored lower on these same four subscales (as measured on the Personal Wellbeing Index-Adult) than expected based on general population scores, and somewhat consistent with Thorpe (2018) who found significantly lower scores across all seven subscales in autistic adults when compared with non-autistic adults.Although the PWI-SC subscales are not a direct measure of the SWB buffers, it is notable that our autistic participants differed significantly from the non-autistic participants on subscales that measured happiness with relationships; a known external buffer for positive SWB.As proposed by Richardson et al. (2016), SWB may be less protected in the presence of reduced internal and external buffers.Thus, we theorize that being autistic and having poorly established external and internal buffers possibly resets SWB to a lower set point, in turn leaving them more vulnerable to poorer long-term physical health and wellbeing, as well as to depression.
The importance of adolescence and early adulthood in developing these resources and the lack of opportunity afforded to autistic adolescents in doing so effectively highlights preadolescence as a crucial time to (i) screen for resource deficiency and (ii) apply appropriate interventions in response.
The unexpected patterns of positive affect and negative affect on SWB results found within the current sample of autistic and non-autistic participants are consistent with previous non-significant differences found between autistic and non-autistic adolescents in positive affect (Kovac et al., 2016) and autistic and non-autistic adolescents and young adults in negative affect (Ilen et al., 2023).This suggests that the mechanisms underlying Homeostatic Protective mood during adolescence may not be a balance of high positive affect and low negative affect, as proposed by others (e.g., Blore et al., 2011).Rather, perhaps there is an alternative pattern to the mechanisms that may underpin Homeostatic Protective mood and maintain SWB at a lower set point in autistic people.Adolescence (Meinhardt-Injac et al., 2020) and young adulthood (Wood et al., 2018) are critical periods for social development which may lead to lower SWB, and this vulnerability may be further complicated by autistic social differences.Some autistic individuals have lower self-esteem (van der Cruijsen & Boyer, 2021), optimism (Kuzmanovic et al., 2019), and autonomy (Hodgetts et al., 2018) compared to nonautistic people.As such, autistic individuals may be more reliant on strong external buffers to either compensate for, or boost, lower internal buffers to strengthen Homeostatic Protective mood and to maintain SWB within a positive set-point range.
The unexpected results in positive affect, negative affect, and SWB found in the autistic sample suggest that positive affect, negative affect, and SWB may be affected by the challenges associated with both adolescent development and autism.For all people, adolescence is a turbulent developmental period when affective control and emotion regulation are often challenged (Schweizer et al., 2020), which can lead to difficulty in responding to negative life events (Flouri & Mavroveli, 2013).It may be that adolescence is even further challenged for autistic individuals as they are often at greater risk of negative life events such as bullying and victimization (Maïano et al., 2016) compared to their non-autistic peers.Taken together, this suggests that autistic adolescents may be more vulnerable to negative life events, which may impose sustained lower positive affect and high negative affect and lead to reduced SWB.These results may be a consequence of differences that some autistic people show in emotion regulation, including (i) cognitive appraisal (i.e., personal interpretation of life events) and (ii) management of negative emotions (Cai et al., 2018), although this relationship requires further investigation.We suggest that the impact of persistent low positive affect and high negative affect on SWB may be cumulative for autistic people, especially when internal and external buffers are poorly established.Over time, these effects may force Homeostatic Protective mood into defeat, and as a consequence plunge SWB into a sustained low set-point range, with life-long implications (Cummins, 2010) in autistic people.To better understand the cumulative effects on SWB of positive affect and negative affect over time requires longitudinal research.

Limitations
The cross-sectional nature of this study offers a proxy indication of potential relationships between autistic and non-autistic individuals in both SWB and affect.However, we recognize this design does not account for individual, contextual, nor environmental differences that may affect SWB (Cummins, 2013).The current sample provided adequate data for a preliminary examination of SWB, positive affect, and negative affect in adolescents and young adults as well as in autism.Although, with more power and matched participant samples by agegroup, some exploratory outcomes may have increased differences between groups into significance.It is also important to recognize that the reliability coefficient for the PWI-SC and the negative affect were weaker than seen in other studies and results are to be interpreted with necessary caution.

Conclusion
The current study contributes evidence of the autistic populations' proximity to homeostatic defeat throughout adolescence and into early adulthood and that, there is likely increased pressure on SWB during these critical developmental periods.These findings support the applicability of the homeostatic theory of SWB among autistic adolescents and young adults.Further, that the homeostasis of SWB is more likely to be overwhelmed during adolescence for autistic individuals compared to nonautistic peers.The unpredictable fluctuations in affect found suggest that SWB may be vulnerable during adolescence and early adulthood, and for the autistic population.Furthermore, outcomes indicate preadolescence as an optimal developmental period to focus interventions aimed at building resources that protect SWB to reduce potential homeostatic defeat.Finally, this study raises questions surrounding both SWB and affect and suggests future research build on improving how to identify and support wellbeing during critical developmental periods.
It would be beneficial for clinicians to screen preadolescents for SWB and to monitor SWB throughout development, and preventatively target external buffers to support the expectant flow on effect in strengthening internal buffers to support positive wellbeing.It is recommended that research re-examine the role of positive affect and negative affect in SWB in both autistic individuals', and in adolescent and young adult populations.Including the confounding effects of positive affect and negative affect on Homeostatic Protective mood and SWB overtime, and the way that internal and external buffers influence this relationship.We also suggest that other mechanisms that may underpin Homeostatic Protective mood should be explored.To strengthen outcomes of the current study a longitudinal study would be beneficial to follow individual progress through adolescence and into adulthood.

AUTHOR CONTRIBUTIONS
All authors contributed to the written work presented in this paper and approved of the final manuscript.MAS and MH conceptualized this project and data collection.KR conceptualized the specific questions and hypotheses and drafted the manuscript.CM, DHD, MH, EW, DH, and MAS critically edited and contributed to the drafting of this manuscript.KR undertook the analysis under supervision of MAS and MH.MAS and MH were involved in the conceptual design of the study and the review of the study findings.Autistic family members, community providers, and agency leaders were involved in developing the research questions, study designs, implementation, and interpretation and dissemination of the findings.KR was involved in all aspects of this manuscript, and is a family member of autistic individuals, and is a community provider to the autistic community.MAS was involved in all aspects of this manuscript, and is a family member of autistic individuals, is a community

F
I G U R E 2 Comparison positive affect, negative affect, and Personal Wellbeing Index-School Childen on age by diagnosis.(a) Shows group differences in positive affect scores over age, (b) shows group differences in negative affect scores over age, and (c) shows group differences in Personal Wellbeing Index-School children scores over age.

F
I G U R E 3 The affect of positive affect and negative affect on Personal Wellbeing Index-School Children by diagnosis.(a) Shows group differences in the association between negative affect and Personal Wellbeing Index-School Children scores, (b) shows group differences in the association between positive affect and Personal Wellbeing Index-School Children scores.
Reliability, means and standard deviations of key measures.
Note: Alpha is Cronbach's alpha; Age = years.Abbreviations: NA, negative affect; PA, positive affect; PWI-SC, Personal Wellbeing Index-School Children.T A B L E 2 Correlations of key variables for non-autistic (upper panel) and the autistic sample (lower panel).
T A B L E 3 Regression analysis, slope, and intercept group comparisons.
T A B L E 4 Comparison of Personal Wellbeing Index-School Children subscales by diagnosis.