Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Jan 8;19(2):e70161. doi: 10.1002/aur.70161

Measurement Invariance of the PROMIS Family Relationships Scale Among Autistic and General Population Adolescents

Rachel M Benecke 1,2,✉, Zachary J Williams 3, Laura Graham Holmes 4, Judith S Miller 1,5,✉, Elizabeth A Kaplan‐Kahn 1
PMCID: PMC12945458  NIHMSID: NIHMS2142511  PMID: 41508413

ABSTRACT

Social relationships are a key component of quality of life, a high‐priority outcome for autistic people, and family relationships are critical in adolescence. The PROMIS Family Relationships scale has been well validated for use with the general population, but psychometric validation in the autistic population is lacking. This study investigated measurement invariance of the PROMIS Family Relationships among autistic and general population adolescents. The scale demonstrated scalar invariance between the groups, providing evidence that it measures the same construct equivalently and scores can be meaningfully compared between groups. With a well‐validated self‐report measure, researchers can ask autistic teens directly about their experiences of their family relationships, rather than relying solely on parent proxy report.

Keywords: adolescence, autism, family relationships, measurement invariance, patient‐reported outcome measures, PROMIS

Summary

Autistic people and their families value research about quality of life, including family relationships. When researchers use tools that were originally developed for the general population, rather than those designed specifically for autistic people, they must first make sure that the tool measures the same thing in the same way for autistic people. This study tested whether a tool that works well for the general population, the PROMIS Family Relationships scale, also works well for autistic teens. We found that the scale did measure the same thing in the same way for autistic teens, meaning that researchers can confidently do research using this scale with autistic teenagers and compare answers from both groups. This is important because most research on autism and family relationships asks questions of parents, but this scale can be used directly with autistic teens themselves to ask about their experiences.

1. Introduction

Autistic people and their families value quality of life (QoL) as a research priority and clinical outcome (Benevides et al. 2020; Frazier et al. 2018; Gotham et al. 2015; Pellicano et al. 2014). Among adolescents in general, one factor associated with more positive QoL is strong family relationships (Grevenstein et al. 2019). Members of the autism community agree that family relationships are also a key element of QoL for autistic people (Graham Holmes et al. 2020). Among autistic teens, family relationships are known to influence life satisfaction (Lam et al. 2020; Øverland et al. 2024) and transition‐to‐adulthood outcomes (Test et al. 2014). The importance of family relationships may be even more pronounced for autistic adolescents than their non‐autistic peers (Krieger et al. 2018). However, there remain significant gaps in our field's understanding of the role that family relationships may play in supporting the QoL of autistic adolescents.

A striking gap in the literature is a lack of studies utilizing self‐report measures to evaluate autistic adolescents' own perspectives on their family relationships. Research about autism and family relationships tends to focus on how parents or siblings relate to an autistic family member (e.g., Factor et al. 2019; Karst and Van Hecke 2012; Orsmond et al. 2009), rather than investigating this relationship from the autistic person's perspective. When considering autistic teens' perspective of family relationships, researchers often rely on parent‐proxy reports (Schwartz et al. 2018), wherein the parent responds on behalf of their child. However, since adolescence is marked by individuation for autistic and non‐autistic individuals alike (Test et al. 2014; Wood et al. 2018), parental insight into teens' subjective experiences may be limited, and parent proxy report is inherently biased for the construct of family relationships. Therefore, to accurately understand teens' perspectives of their family relationships, researchers should rely on self‐report whenever possible.

Self‐report is often overlooked in autism research (Schiltz et al. 2024). Although some have interpreted discrepancies between proxy and self‐report as evidence that autistic people are incapable of accurate self‐report, most autistic people are able to reflect and reliably report on their internal experiences (Kim and Lecavalier 2024). However, the measures themselves play a role, and it is important that researchers carefully consider whether the instruments they use are sufficiently validated to support their findings, including specifically with autistic people if originally developed for the general population.

The Patient‐Reported Outcomes Measurement Information System (PROMIS) Family Relationships scale has demonstrated strong psychometric properties for general population adolescents (Bevans et al. 2017). Although the parent proxy version has been successfully used with parents of autistic children ages 5–12 (Schwartz et al. 2018), further psychometric validation of the self‐report PROMIS Family Relationships scale with autistic adolescents is necessary. Specifically, researchers cannot presume that a measure which has been validated for the general population will function identically with autistic teenagers. Measurement invariance tests whether a scale measures an underlying construct the same way between groups, or more specifically, whether any differences in observed scores are explained by true group differences (Chen 2008). To that end, we evaluated measurement invariance of the PROMIS Family Relationships scale among autistic and general population teens.

2. Methods

2.1. Participants

The present study is a secondary data analysis of two datasets in which autistic and general population adolescents, ages 14–17, and their parents participated via an online survey. See Table 1 for participant demographics. All recruitment and study procedures were approved by the Institutional Review Board at the Children's Hospital of Philadelphia (CHOP).

TABLE 1.

Demographic information.

Variable Autistic group (%) General population group (%) p a
Totals 138 403
Age (mean[SD]) 15.48 (1.10) 15.50 (1.09) 0.85
14 34 (24.63%) 96 (23.82%)
15 36 (26.09%) 104 (25.81%)
16 36 (26.09%) 109 (27.04%)
17 32 (23.19%) 94 (23.33%)
Gender
Female 37 (28.81%) 199 (49.38%) < 0.01
Male 98 (71.01%) 202 (50.12%)
Transgender 3 (2.17%) N/A b
Missing 0 (0.00%) 2 (0.50%)
Race
Asian 0 (0.00%) 15 (3.72%) 0.01 c
Black 4 (2.90%) 29 (7.20%)
Native American 0 (0.00%) 3 (0.74%)
White 129 (93.48%) 322 (79.90%)
Other or more than one race 5 (3.62%) 34 (8.44%)
Ethnicity
Hispanic or Latino/a 8 (5.80%) 60 (14.89%) 0.01
Not Hispanic or Latino/a 130 (94.20%) 343 (85.11%)
Open in a new tab
a

Two‐sample t‐test.

b

Not reported for general population.

c

Pearson's chi‐squared test.

2.1.1. Autistic Sample

The autistic sample consisted of self‐report data of 138 teens from Graham Holmes et al. (2020). Full recruitment details are reported in the original study; in short, parents and teens were recruited through CHOP and the Interactive Autism Network, and eligibility included parent‐reported medical autism diagnosis and self‐reported ability to complete the survey independently. The sample included approximately three times as many males (n = 98; 71.01%) as females (n = 37; 26.81%), with three teens identifying as transgender. The sample was predominantly white (n = 129; 93.48%) and non‐Hispanic/Latino (n = 130; 94.20%).

2.1.2. General Population Sample

Our comparison group consisted of 403 teens from the PROMIS pediatric normative sample (Irwin et al. 2010). Parents and teens were recruited through the GfK Knowledge Panel; full recruitment details are reported in the original study. The sample was evenly split between males (n = 202; 50.12%) and females (n = 199; 49.38%); the general population sample was asked to identify as male or female without specifying cis‐ or transgender. As the panel aimed to be nationally representative, the gender ratio was balanced and racial/ethnic makeup more closely matched the broader US population (n = 322; 79.90% white and n = 343; 85.11% non‐Hispanic/Latino).

2.2. Measure

We used the short form pediatric self‐report version of the PROMIS Family Relationships scale, which consists of eight items scored on a scale of 1 (Never) to 5 (Always), with higher scores indicating a more positive relationship with one's family (Bevans et al. 2017). The PROMIS scales are unidimensional QoL measures available through the National Institutes of Health (NIH; Reeve et al. 2007) and were validated with representative samples of the US population (Cella et al. 2010). The scales can be administered individually or as part of a wider battery, such as the PAB‐L (Graham Holmes et al. 2020).

2.3. Data Analysis

We first fit confirmatory factor analysis (CFA) models to our samples using the lavaan package in R (Rosseel 2012). While a variety of models were considered, we ultimately proceeded with a unidimensional model, which was the best choice for alignment with theory, parsimony, and fit. We identified models by setting factor variance to 1. Given the ordinal nature of the PROMIS scales and low item‐level missingness (15 items across 11 general population participants, none for autistic sample), we used the robust diagonally weighted least squares estimator with pairwise deletion for missing values. To avoid potentially inflated standard fit indices (Xia and Yang 2019), we used robust categorical maximum‐likelihood estimators (Savalei 2021). Because RMSEA is often elevated for models with few degrees of freedom with few items and factors (Cheung and Rensvold 2002)—like the PROMIS scales, even for their validation samples (Bevans et al. 2017)—we anticipated higher RMSEA values and interpreted fit statistics holistically. To address inadequate baseline model fit, we considered model modification indices and correlated error terms where theoretically justified (Jorgensen 2017). For most items, no autistic teens endorsed the lowest response; for those items, the two lowest categories were collapsed for both groups (Tsai et al. 2024). To maximize our ability to detect non‐invariance, we did not collapse categories for items where there were sufficient responses in each category.

Next, we fit a series of increasingly restrictive multigroup CFA models using lavaan (Rosseel 2012) and semTools (Kline 2015; van de Schoot et al. 2012). Configural invariance was assessed by allowing all factor loadings and item intercepts to vary freely, confirming that the overall factor structure of the measure fit well for both groups. Next, metric invariance was assessed by constraining factor loadings to be equivalent across groups while allowing item intercepts to vary freely. Metric invariance was established if there was no significant difference in fit between the metric and configural CFA models, indicating that there was no difference in the pattern of item loadings across groups. Finally, scalar invariance was assessed by constraining both factor loadings and item intercepts to be equivalent across groups. Scalar invariance was established if there was no significant difference in fit between the scalar and metric CFA models.

In lieu of rule‐of‐thumb cutoffs, which vary in the literature and may be affected by sample size, imbalanced group size, and level of invariance (Chen 2007; Cheung and Rensvold 2002; Hu and Bentler 1999; Meade et al. 2008), we used permutation testing (Jorgensen et al. 2018; Kite et al. 2018) to evaluate measurement invariance. By permuting group assignment and generating null distributions for each level of measurement invariance testing, we could empirically evaluate whether the Δχ2 and ΔCFI (D'Urso et al. 2021) between each model was significant. An a priori threshold of α = 0.01 was chosen for statistical significance to reduce false positives and due to the moderate power of our sample to detect sizable invariance at this significance level (Kite et al. 2018; Lakens et al. 2018). If p‐values were non‐significant (≥ 0.01), fit for the more restrictive model did not deteriorate, thus demonstrating measurement invariance.

3. Results

3.1. Internal Reliability

A summary of reliability and means by group is presented in Table 2. The PROMIS Family Relationships scale demonstrated excellent internal consistency across both the autistic (ω T = 0.96) and general population (ω T = 0.95) groups.

TABLE 2.

Domain summaries by group.

Group n ω T M (SD)
Autistic teens 138 0.96 50.66 (9.71)
General population teens 403 0.95 47.92 (8.96)
Open in a new tab

Abbreviation: ω T, omega total (model‐based reliability).

3.2. Confirmatory Factor Analysis

Because the unmodified unidimensional Family Relationships model demonstrated borderline fit (results reported in Supporting Information), we examined modification indices for opportunities to improve fit. Since the items “My family treated me fairly” and “My family listened to me” capture highly related concepts that shared common variance, we correlated error terms for those items. Fit for the modified model was excellent for the combined sample (χ 2 = 74.466, CFI = 0.996, TLI = 0.994, SRMR = 0.022). Fit was good for the autistic sample (χ 2 = 52.629, CFI = 0.949, TLI = 0.925, SRMR = 0.044) and excellent for the general population sample (χ 2 = 61.794, CFI = 0.980, TLI = 0.970, SRMR = 0.025). Detailed results are reported in Table 3.

TABLE 3.

Confirmatory factor analysis model results.

Group χ 2 df p CFI TLI SRMR RMSEA (90% CI)
Combined sample 74.466 19 < 0.001 0.996 0.994 0.022 0.098 (0.076–0.122)
Autistic teens 52.629 19 < 0.001 0.949 0.925 0.044 0.156 (0.097–0.215)
General population teens 61.794 19 < 0.001 0.980 0.970 0.025 0.093 (0.066–0.120)
Open in a new tab

3.3. Measurement Invariance

Results of the measurement invariance analyses for the Family Relationships scale across groups are presented in Table 4. As shown by non‐significant permutation testing (all ps > 0.01) at each level of increasing equality constraints, the Family Relationships scale demonstrated configural, metric, and scalar invariance.

TABLE 4.

Model comparison statistics (w/permutation).

Model Model comparison statistics
Comparison Δχ 2 p ΔCFI p
1. Configural Null vs. 1 53.777 0.064 > 0.999 0.063
2. Metric 1 vs. 2 23.257 0.186 < −0.001 0.184
3. Scalar 2 vs. 3 4.704 0.232 < 0.001 0.449
Open in a new tab

4. Discussion

Evaluating measurement invariance is an important step in the psychometric validation necessary to use instruments developed for the general population in clinical and research contexts with autistic people. Autistic people may interpret items or response options differently from their non‐autistic peers, and in some cases, the construct of interest is intertwined with the autistic experience such that an instrument does not measure the same construct as in the general population (Nicolaidis et al. 2020). Particularly with a subjective outcome like quality of family relationships (Schiltz et al. 2024), before comparing findings across groups, it is vital to show that researchers are truly measuring the same construct in both groups and that the chosen measure functions similarly in a new population as the one in which it was initially tested. While the parent‐proxy PROMIS Family Relationships scale has been validated for use with parents of autistic children (Schwartz et al. 2018), our study is the first to present evidence that the self‐report version short form functions similarly for autistic teens and general population teens.

Self‐report is vital to our understanding of autistic people's lived experience and underutilized in autism research, often due to concerns that autistic people are less able to reflect and report on their own experiences than their non‐autistic peers—an assumption rooted in ableism (Schiltz et al. 2024). Psychometrically sound PROMs facilitate research and clinical interventions which center the voices of autistic people themselves. The necessity of self‐report holds especially true for teenagers, who are working to increase independence and define their own identity and goals. Particularly in the transition to adulthood, autistic adolescents note the influence of family on subjective and objective QoL (Lam et al. 2020; Øverland et al. 2024), and supportive family relationships constitute an important factor in later outcomes (Test et al. 2014). In this life stage, parent proxy report may not capture the depth of the individual's opinion on their own life, so researchers require a well‐validated PROM shown to work well with autistic teens.

Our results show configural, metric, and scalar invariance for the PROMIS Family Relationships scale. Because the scale demonstrated scalar invariance, we can conclude that the construct is being measured in a similar way for autistic and general population teens. Thus, researchers can confidently answer questions within the group of autistic teens or those that require comparing scores across autistic and non‐autistic groups.

Future research with the PROMIS Family Relationships scale should encompass two paths: further psychometric validation and applied research. Since autistic people across the lifespan value family relationships as a QoL outcome (Graham Holmes et al. 2020), follow‐up work should aim to establish measurement invariance for this scale with other age groups, equating scores in adolescents with those in younger children and adults. Furthermore, while self‐report is ideal when assessing subjective experiences, there are many autistic people for whom parent proxy report may remain more realistic, such as those who are very young or unable to read or respond verbally to questions about their QoL. Thus, future research should aim to establish “equivalent measures” of various constructs that span multiple reporters in addition to investigating the validity of each alternative measure with subgroups of the autistic population to ensure psychometric adequacy.

Now that the PROMIS Family Relationships scale has been validated for autistic adolescent self‐report, researchers have a useful tool to directly ask autistic teens about their lived experience to understand the impact of family relationships on autistic QoL. The existing body of literature on autism and family relationships has been historically one‐sided, largely focused on parents and sometimes siblings. Reporting from only one half of a dyadic dynamic perpetuates an ableist construction of knowledge that privileges the neurotypical experience. While the pediatric parent proxy PROMIS Family Relationships scale has been used in autism research (Schwartz et al. 2018), to our knowledge, our study is the first to investigate self‐report with autistic teens. Autistic teens deserve for their voices to be prioritized in research that claims to be about them. Applied research should seek input from autistic adolescents themselves to identify protective factors for family relationships or adapt family‐related therapeutic interventions.

4.1. Limitations

We note two measure‐related limitations to our study. First, it should be acknowledged that although our results show the PROMIS Family Relationships scale measures the same underlying construct of family relationship quality in the same way among autistic teens as their general population peers, measurement invariance cannot reveal whether it captures the most important aspects of that construct for autistic people, and researchers should strive to incorporate autistic perspectives in future scale development. Additionally, collapsing the two lowest response categories to account for no “Never” responses in the autistic group limits our ability to evaluate the differential response threshold of that specific item category for those items between groups.

Our study also has some sample limitations related to the nature of secondary data analysis. First, we intentionally refer to the PROMIS validation sample as “general population” rather than “typically developing” or “non‐autistic” because autism diagnoses were not reported for that sample. Given the representative nature of the sample drawn from the broader US population, we expect that a few autistic people may have been included, but true diagnostic status will remain unknown. Relatedly, we were unable to characterize participants in depth on cognitive functioning, verbal ability, or adaptive behavior skills. Future work which assesses these factors could further validate the generalizability of our findings accounting for the diversity in abilities among autistic people.

Our samples also share limitations commonly recognized as areas for growth across autism research as a whole, including racial and ethnic makeup more white and non‐Hispanic/Latino than the current US population (Giwa Onaiwu 2020; Jones et al. 2020; West et al. 2016); an imbalanced sex and gender ratio, which reflects diagnostic patterns but does not fully encompass the increased gender diversity noted on the autism spectrum (Mittertreiner et al. 2024); and underrepresentation of minimally verbal autistic people and/or those with co‐occurring intellectual disability (Maenner et al. 2021). While we did not exclude participants based on verbal ability or IQ, given the design of the questionnaire, we were able to establish measurement invariance only for teens in both groups capable of independent self‐report. Moreover, the autistic group was not large enough to evaluate differential item functioning (DIF) along any of these axes of sample diversity. Although the autistic group differed significantly from the general population group on demographic factors, it is worth noting that previous evidence for this measure has demonstrated no DIF by ethnicity, race, or gender (Bevans et al. 2017). Future studies may benefit from oversampling underrepresented groups to more fully represent autistic perspectives, as well as performing formal DIF analyses across subgroups. In addition, researchers should investigate accommodations to make PROMIS questionnaires more accessible to individuals who might struggle with standard self‐report.

5. Conclusion

This study is the first to investigate measurement invariance of the pediatric self‐report PROMIS Family Relationships scale with autistic and general population adolescents. By establishing scalar invariance, this study showed that this scale is valid for use with autistic teens and scores can be meaningfully compared across groups. With an accurate tool to investigate autistic teens' family relationships, researchers should follow a call to action to prioritize and include autistic perspectives rather than solely relying on proxy reports.

Funding

This work was supported by the McMorris Family Foundation, the Children's Hospital of Philadelphia Women's Committee, the National Institute of General Medical Sciences (T32‐GM007347), the National Institute on Deafness and Other Communication Disorders (F30‐DC019510), the National Institute of Mental Health (R01MH133838), and the Hondros Family.

Conflicts of Interest

R.M.B., L.G.H., J.S.M., and E.A.K.‐K. have no conflicts of interest to report. Z.J.W. has received consulting fees from Roche, Autism Speaks, and the Simons Foundation Autism Research Initiative. He is also Vice‐chair of Autistic and Neurodivergent Scholars Working for Equity in Research (ANSWER), a division of the Autism Intervention Research Network on Physical Health (AIR‐P).

Supporting information

Data S1: Supporting Information.

AUR-19-0-s001.docx (17.9KB, docx)

Acknowledgments

This work was supported by the McMorris Family Foundation and the Children's Hospital of Philadelphia Women's Committee. Z.J.W.'s time was supported in part by grants from the National Institute of General Medical Sciences (T32‐GM007347) and the National Institute on Deafness and Other Communication Disorders Grant F30‐DC019510. EAKK's time was supported in part by a grant from the National Institute of Mental Health (R01MH133838) and by funding from the Hondros Family.

Contributor Information

Rachel M. Benecke, Email: benecker@chop.edu.

Judith S. Miller, Email: millerj3@chop.edu.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

References

  1. Benevides, T. W. , Shore S. M., Palmer K., et al. 2020. “Listening to the Autistic Voice: Mental Health Priorities to Guide Research and Practice in Autism From a Stakeholder‐Driven Project.” Autism 24, no. 4: 822–833. 10.1177/1362361320908410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bevans, K. B. , Riley A. W., Landgraf J. M., et al. 2017. “Children's Family Experiences: Development of the PROMIS Pediatric Family Relationships Measures.” Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care and Rehabilitation 26, no. 11: 3011–3023. 10.1007/s11136-017-1629-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cella, D. , Riley W., Stone A., et al. 2010. “The Patient‐Reported Outcomes Measurement Information System (PROMIS) Developed and Tested Its First Wave of Adult Self‐Reported Health Outcome Item Banks: 2005–2008.” Journal of Clinical Epidemiology 63, no. 11: 1179–1194. 10.1016/j.jclinepi.2010.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chen, F. F. 2007. “Sensitivity of Goodness of Fit Indexes to Lack of Measurement Invariance.” Structural Equation Modeling: A Multidisciplinary Journal 14, no. 3: 464–504. 10.1080/10705510701301834. [DOI] [Google Scholar]
  5. Chen, F. F. 2008. “What Happens if We Compare Chopsticks With Forks? The Impact of Making Inappropriate Comparisons in Cross‐Cultural Research.” Journal of Personality and Social Psychology 95, no. 5: 1005–1018. 10.1037/a0013193. [DOI] [PubMed] [Google Scholar]
  6. Cheung, G. W. , and Rensvold R. B.. 2002. “Evaluating Goodness‐Of‐Fit Indexes for Testing Measurement Invariance.” Structural Equation Modeling: A Multidisciplinary Journal 9, no. 2: 233–255. 10.1207/S15328007SEM0902_5. [DOI] [Google Scholar]
  7. D'Urso, E. D. , De Roover K., Vermunt J. K., and Tijmstra J.. 2021. “Scale Length Does Matter: Recommendations for Measurement Invariance Testing With Categorical Factor Analysis and Item Response Theory Approaches.” Behavior Research Methods 54, no. 5: 2114–2145. 10.3758/s13428-021-01690-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Factor, R. S. , Ollendick T. H., Cooper L. D., Dunsmore J. C., Rea H. M., and Scarpa A.. 2019. “All in the Family: A Systematic Review of the Effect of Caregiver‐Administered Autism Spectrum Disorder Interventions on Family Functioning and Relationships.” Clinical Child and Family Psychology Review 22, no. 4: 433–457. 10.1007/s10567-019-00297-x. [DOI] [PubMed] [Google Scholar]
  9. Frazier, T. W. , Dawson G., Murray D., Shih A., Sachs J. S., and Geiger A.. 2018. “Brief Report: A Survey of Autism Research Priorities Across a Diverse Community of Stakeholders.” Journal of Autism and Developmental Disorders 48, no. 11: 3965–3971. 10.1007/s10803-018-3642-6. [DOI] [PubMed] [Google Scholar]
  10. Giwa Onaiwu, M. 2020. ““They Don't Know, Don't Show, or Don't Care”: Autism's White Privilege Problem.” Autism in Adulthood 2, no. 4: 270–272. 10.1089/aut.2020.0077. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gotham, K. , Marvin A. R., Taylor J. L., et al. 2015. “Characterizing the Daily Life, Needs, and Priorities of Adults With Autism Spectrum Disorder From Interactive Autism Network Data.” Autism 19, no. 7: 794–804. 10.1177/1362361315583818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Graham Holmes, L. , Zampella C. J., Clements C., et al. 2020. “A Lifespan Approach to Patient‐Reported Outcomes and Quality of Life for People on the Autism Spectrum.” Autism Research 13, no. 6: 970–987. 10.1002/aur.2275. [DOI] [PubMed] [Google Scholar]
  13. Grevenstein, D. , Bluemke M., Schweitzer J., and Aguilar‐Raab C.. 2019. “Better Family Relationships—Higher Well‐Being: The Connection Between Relationship Quality and Health Related Resources.” Mental Health & Prevention 14: 200160. 10.1016/j.mph.2019.200160. [DOI] [Google Scholar]
  14. Hu, L. , and Bentler P. M.. 1999. “Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional Criteria Versus New Alternatives.” Structural Equation Modeling: A Multidisciplinary Journal 6, no. 1: 1–55. 10.1080/10705519909540118. [DOI] [Google Scholar]
  15. Irwin, D. E. , Stucky B. D., Thissen D., et al. 2010. “Sampling Plan and Patient Characteristics of the PROMIS Pediatrics Large‐Scale Survey.” Quality of Life Research 19, no. 4: 585–594. 10.1007/s11136-010-9618-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jones, D. R. , Nicolaidis C., Ellwood L. J., et al. 2020. “An Expert Discussion on Structural Racism in Autism Research and Practice.” Autism in Adulthood 2, no. 4: 273–281. 10.1089/aut.2020.29015.drj. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jorgensen, T. D. 2017. “Applying Permutation Tests and Multivariate Modification Indices to Configurally Invariant Models That Need Respecification.” Frontiers in Psychology 8: 1455. 10.3389/fpsyg.2017.01455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jorgensen, T. D. , Kite B. A., Chen P.‐Y., and Short S. D.. 2018. “Permutation Randomization Methods for Testing Measurement Equivalence and Detecting Differential Item Functioning in Multiple‐Group Confirmatory Factor Analysis.” Psychological Methods 23, no. 4: 708–728. 10.1037/met0000152. [DOI] [PubMed] [Google Scholar]
  19. Karst, J. S. , and Van Hecke A. V.. 2012. “Parent and Family Impact of Autism Spectrum Disorders: A Review and Proposed Model for Intervention Evaluation.” Clinical Child and Family Psychology Review 15, no. 3: 247–277. 10.1007/s10567-012-0119-6. [DOI] [PubMed] [Google Scholar]
  20. Kim, S. Y. , and Lecavalier L.. 2024. “Stability and Validity of Self‐Reported Depression and Anxiety in Autistic Youth.” Journal of Autism and Developmental Disorders 55: 3590–3604. 10.1007/s10803-024-06456-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kite, B. A. , Jorgensen T. D., and Chen P.‐Y.. 2018. “Random Permutation Testing Applied to Measurement Invariance Testing With Ordered‐Categorical Indicators.” Structural Equation Modeling: A Multidisciplinary Journal 25, no. 4: 573–587. 10.1080/10705511.2017.1421467. [DOI] [Google Scholar]
  22. Kline, R. B. 2015. Principles and Practice of Structural Equation Modeling. Guilford Press. [Google Scholar]
  23. Krieger, B. , Piskur B., Schulze C., Jakobs U., Beurskens A., and Moser A.. 2018. “Supporting and Hindering Environments for Participation of Adolescents Diagnosed With Autism Spectrum Disorder: A Scoping Review.” PLoS One 13, no. 8: e0202071. 10.1371/journal.pone.0202071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lakens, D. , Adolfi F. G., Albers C. J., et al. 2018. “Justify Your Alpha.” Nature Human Behaviour 2, no. 3: 168–171. 10.1038/s41562-018-0311-x. [DOI] [Google Scholar]
  25. Lam, G. Y. H. , Holden E., Fitzpatrick M., Raffaele Mendez L., and Berkman K.. 2020. ““Different but Connected”: Participatory Action Research Using Photovoice to Explore Well‐Being in Autistic Young Adults.” Autism 24, no. 5: 1246–1259. 10.1177/1362361319898961. [DOI] [PubMed] [Google Scholar]
  26. Maenner, M. J. , Shaw K. A., Bakian A. V., et al. 2021. “Prevalence and Characteristics of Autism Spectrum Disorder Among Children Aged 8 Years—Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2018.” MMWR Surveillance Summaries 70, no. 11: 1–16. 10.15585/mmwr.ss7011a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Meade, A. W. , Johnson E. C., and Braddy P. W.. 2008. “Power and Sensitivity of Alternative Fit Indices in Tests of Measurement Invariance.” Journal of Applied Psychology 93, no. 3: 568–592. 10.1037/0021-9010.93.3.568. [DOI] [PubMed] [Google Scholar]
  28. Mittertreiner, E. J. , Ng‐Cordell E., McVey A. J., and Kerns C. M.. 2024. “Research Methods at the Intersection of Gender Diversity and Autism: A Scoping Review.” Autism 28: 13623613241245595. 10.1177/13623613241245595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Nicolaidis, C. , Raymaker D. M., McDonald K. E., et al. 2020. “Creating Accessible Survey Instruments for Use With Autistic Adults and People With Intellectual Disability: Lessons Learned and Recommendations.” Autism in Adulthood 2, no. 1: 61–76. 10.1089/aut.2019.0074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Orsmond, G. I. , Kuo H.‐Y., and Seltzer M. M.. 2009. “Siblings of Individuals With an Autism Spectrum Disorder: Sibling Relationships and Wellbeing in Adolescence and Adulthood.” Autism 13, no. 1: 59–80. 10.1177/1362361308097119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Øverland, E. , Hauge Å. L., Orm S., et al. 2024. ““I Have to Charge My Social Battery”: Perspectives From Autistic Young Adults on Quality of Life.” Autism 28, no. 6: 1369–1381. 10.1177/13623613241245578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pellicano, E. , Dinsmore A., and Charman T.. 2014. “What Should Autism Research Focus Upon? Community Views and Priorities From the United Kingdom.” Autism 18, no. 7: 756–770. 10.1177/1362361314529627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Reeve, B. B. , Hays R. D., Bjorner J. B., et al. 2007. “Psychometric Evaluation and Calibration of Health‐Related Quality of Life Item Banks: Plans for the Patient‐Reported Outcomes Measurement Information System (PROMIS).” Medical Care 45, no. 5: S22–S31. 10.1097/01.mlr.0000250483.85507.04. [DOI] [PubMed] [Google Scholar]
  34. Rosseel, Y. 2012. “ lavaan: An R Package for Structural Equation Modeling.” Journal of Statistical Software 48, no. 2: 1–36. 10.18637/jss.v048.i02. [DOI] [Google Scholar]
  35. Savalei, V. 2021. “Improving Fit Indices in Structural Equation Modeling With Categorical Data.” Multivariate Behavioral Research 56, no. 3: 390–407. 10.1080/00273171.2020.1717922. [DOI] [PubMed] [Google Scholar]
  36. Schiltz, H. K. , Williams Z. J., Zheng S., et al. 2024. “Measurement Matters: A Commentary on the State of the Science on Patient Reported Outcome Measures (PROMs) in Autism Research.” Autism Research 17, no. 4: 690–701. 10.1002/aur.3114. [DOI] [PubMed] [Google Scholar]
  37. Schwartz, J. , Huntington N., Toomey M., et al. 2018. “Measuring the Involvement in Family Life of Children With Autism Spectrum Disorder: A DBPNet Study.” Research in Developmental Disabilities 83: 18–27. 10.1016/j.ridd.2018.07.012. [DOI] [PubMed] [Google Scholar]
  38. Test, D. W. , Smith L. E., and Carter E. W.. 2014. “Equipping Youth With Autism Spectrum Disorders for Adulthood: Promoting Rigor, Relevance, and Relationships.” Remedial and Special Education 35, no. 2: 80–90. 10.1177/0741932513514857. [DOI] [Google Scholar]
  39. Tsai, C.‐L. , Wind S., and Estrada S.. 2024. “Exploring the Effects of Collapsing Rating Scale Categories in Polytomous Item Response Theory Analyses: An Illustration and Simulation Study.” Measurement: Interdisciplinary Research and Perspectives 23, no. 1: 66–89. 10.1080/15366367.2023.2288791. [DOI] [Google Scholar]
  40. van de Schoot, R. , Lugtig P., and Hox J.. 2012. “A Checklist for Testing Measurement Invariance.” European Journal of Developmental Psychology 9, no. 4: 486–492. 10.1080/17405629.2012.686740. [DOI] [Google Scholar]
  41. West, E. A. , Travers J. C., Kemper T. D., et al. 2016. “Racial and Ethnic Diversity of Participants in Research Supporting Evidence‐Based Practices for Learners With Autism Spectrum Disorder.” Journal of Special Education 50, no. 3: 151–163. 10.1177/0022466916632495. [DOI] [Google Scholar]
  42. Wood, D. , Crapnell T., Lau L., et al. 2018. “Emerging Adulthood as a Critical Stage in the Life Course.” In Handbook of Life Course Health Development, edited by Halfon N., Forrest C. B., Lerner R. M., and Faustman E. M.. Springer. [PubMed] [Google Scholar]
  43. Xia, Y. , and Yang Y.. 2019. “RMSEA, CFI, and TLI in Structural Equation Modeling With Ordered Categorical Data: The Story They Tell Depends on the Estimation Methods.” Behavior Research Methods 51, no. 1: 409–428. 10.3758/s13428-018-1055-2. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1: Supporting Information.

AUR-19-0-s001.docx (17.9KB, docx)

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Articles from Autism Research are provided here courtesy of Wiley

RESOURCES