JWJustin Williams

Justin Williams

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Justin Williams has a total of 19 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Parenting stress in caregiver‐mediated interventions for toddlers with autism: An application of quantile regression mixed models

    by Andrew Schlink, Justin Williams, Maria Pizzano, Amanda Gulsrud, Connie Kasari

    Traditional longitudinal modeling approaches require normally distributed data and do not account for sample heterogeneity. Parenting stress, in particular, can be difficult to model across time without transforming the data as it is usually high for caregivers of children with ASD. This study used novel linear quantile mixed models (LQMMs) to model non‐normal parent stress scores across two caregiver‐mediated interventions involving toddlers with ASD. The sample included 86 caregiver‐child dyads who were randomized to either a parent‐only psychoeducational intervention or hands‐on parent training in a naturalistic developmental intervention. Child and parent‐related domains of the Parenting Stress Index (PSI) were the primary outcomes in this study. The PSI was collected at entry, 10‐week exit, 3‐month follow‐up, and 6‐month follow‐up periods. Separate LQMMs were used to model five specific quantiles ( of the two PSI domains across the complete intervention timeline. These five quantiles effectively modeled the entire conditional distribution of parenting stress scores. The LQMMs indicated that child‐related parenting stress decreased across all quantiles within both interventions, with no difference in the rate of parenting stress change between the intervention groups. For parent‐related parenting stress, the effect of intervention depended on the group's stress level; some parents increased their perceived stress within the hands‐on intervention at the 3‐month follow‐up. Overall, this study demonstrated that the use of LQMMs yielded additional information, beyond traditional longitudinal models, regarding the relationship between parenting stress within two caregiver‐mediated intervention protocols. This study also discussed the methodological contributions and potential future applications of LQMMs.Lay Summary: This study used a newer longitudinal modeling technique to examine how parenting stress changed across two caregiver‐mediated interventions for toddlers with ASD. Results showed that certain parents in the JASPER condition might require additional support as they exit the study and enter into their first follow‐up period. It was also determined that this new modeling technique could be a valuable tool to analyze highly variable data often present in ASD intervention studies.


  • Who gains and who loses? Sociodemographic disparities in access to special education services among autistic students

    by Alexandra Sturm, Justin Williams, Connie Kasari

    Little is known about differences in the allocation of special education services to students with autism compared with students with other primary learning differences (e.g., intellectual disability [ID], specific learning disability [SLD]) and the comparative impact of sociodemographic factors on special education service receipt. The present study aimed to compare allocation of services (i.e., quantity and types) between students eligible for special education services under autism, SLD and ID, and to identify differences in sociodemographic predictors (e.g., race, neighborhood income) of service allocation. Data were culled from special education administrative records from a large urban, primarily Hispanic, Latino, or Spanish‐serving school district from academic years 2011–2012 to 2016–2017. Participants included N = 76,428 students (Nautism = 18,151, NSLD = 54,001, NID = 4,276) ages 2–18. Results showed that autistic students received, on average, more services than students with SLD. Services received by students served under autism and SLD eligibility were more congruent with areas of need (e.g., language, occupational therapy) relative to students served under ID. Student‐level socioeconomic status (free and reduced lunch) was more positively predictive of the number of special education services received than the neighborhood income of the school the student attended. Finally, the most significant racial disparities in service allocation were observed among students served under autism eligibility. The present study demonstrates the critical role of economic resources in the quantity and types of services received, and the desirability of a public education for well‐resourced families who are possibly best situated to navigate special education services.Lay Summary: An eligibility of autism in special education confers a significant advantage in the number of services a student receives. Disparity in the quantity of special education services exists among autistic students across many sociodemographic factors. Most notably, greater allocation of services to non‐Hispanic, Latino, or Spanish, White American, European American, or Middle Eastern American students and students who do not receive free and reduced lunch.


  • Open Access

    Modeling observations with a detection limit using a truncated normal distribution with censoring

    by Justin R. Williams, Hyung-Woo Kim, Catherine M. Crespi

    Background: When data are collected subject to a detection limit, observations below the detection limit may be considered censored. In addition, the domain of such observations may be restricted; for example, values may be required to be non-negative. Methods: We propose a method for estimating population mean and variance from censored observations that accounts for known domain restriction. The method finds maximum likelihood estimates assuming an underlying truncated normal distribution. Results: We show that our method, tcensReg, has lower bias, Type I error rates, and mean squared error than other methods commonly used for data with detection limits such as Tobit regression and single imputation under a range of simulation settings from mild to heavy censoring and truncation. We further demonstrate the consistency of the maximum likelihood estimators. We apply our method to analyze vision quality data collected from ophthalmology clinical trials comparing different types of intraocular lenses implanted during cataract surgery. All of the methods yield similar conclusions regarding non-inferiority, but estimates from the tcensReg method suggest that there may be greater mean differences and overall variability. Conclusions: In the presence of detection limits, our new method tcensReg provides a way to incorporate known domain restrictions in dependent variables that substantially improves inferences.


  • Open Access

    Modeling Observations with a Detection Limit using a Truncated Normal Distribution with Censoring

    by Justin R. Williams, Hyung-Woo Kim, Catherine M. Crespi

    Background: When data are collected subject to a detection limit, observations below the detection limit may be considered censored. In addition, the domain of such observations may be restricted; for example, values may be required to be non-negative. Methods: We propose a method for estimating population mean and variance from censored observations that accounts for known domain restriction. The method finds maximum likelihood estimates assuming an underlying truncated normal distribution. Results: We show that our method, tcensReg, has lower bias, Type I error rates, and mean squared error than other methods commonly used for data with detection limits such as Tobit regression and single imputation under a range of simulation settings from mild to heavy censoring and truncation. We further demonstrate the consistency of the maximum likelihood estimators. We apply our method to analyze vision quality data collected from ophthalmology clinical trials comparing different types of intraocular lenses implanted during cataract surgery. All of the methods yield similar conclusions regarding non-inferiority, but estimates from the tcensReg method suggest that there may be greater mean differences and overall variability. Conclusions: In the presence of detection limits, our new method tcensReg provides a way to incorporate known domain restrictions in dependent variables that substantially improves inferences.

  • Adolescents With Autism Spectrum Disorder and Social Skills Groups at School: A Randomized Trial Comparing Intervention Environment and Peer Composition

    by Michelle Dean, Justin Williams, Felice Orlich, Connie Kasari

    This study used a randomized controlled trial to compare two distinct models of group social skills interventions with adolescents with autism spectrum disorder (ASD). Participants had a confirmed diagnosis of ASD, had an IQ greater than or equal to 70, and were educated in the general education setting. Data from 62 adolescent participants who were randomized to one of two treatment conditions (skills versus engage) were analyzed. Skills participants had a diagnosis of ASD or social difficulties. Engage groups included adolescents with ASD and typically developing (TD) peer mentors. Skills and engage participants both improved joint engagement and reduced solitary engagement; however, skills participants reported higher social stress and lower quality interpersonal relationships at exit and increased emotional symptoms and problem behaviors at follow-up compared to the engage group. The findings suggest that within inclusive secondary school settings, it may be beneficial to include TD peers in social intervention groups.


  • Open Access

    microbiomeDASim: Simulating longitudinal differential abundance for microbiome data

    by Justin Williams, Hector Corrada Bravo, Jennifer Tom, Joseph Nathaniel Paulson

    An increasing emphasis on understanding the dynamics of microbial communities in various settings has led to the proliferation of longitudinal metagenomic sampling studies. Data from whole metagenomic shotgun sequencing and marker-gene survey studies have characteristics that drive novel statistical methodological development for estimating time intervals of differential abundance. In designing a study and the frequency of collection prior to a study, one may wish to model the ability to detect an effect, e.g., there may be issues with respect to cost, ease of access, etc. Additionally, while every study is unique, it is possible that in certain scenarios one statistical framework may be more appropriate than another. Here, we present a simulation paradigm implemented in the R Bioconductor software package microbiomeDASim available at http://bioconductor.org/packages/microbiomeDASim microbiomeDASim. microbiomeDASim allows investigators to simulate longitudinal differential abundant microbiome features with a variety of known functional forms with flexible parameters to control desired signal-to-noise ratio. We present metrics of success results on one particular method called metaSplines.


  • Open Access

    Modeling Variables with a Detection Limit using a Truncated Normal Distribution with Censoring

    by Justin R. Williams, Hyung-Woo Kim, Catherine M. Crespi

    Background: When data are collected subject to a detection limit, observations below the detection limit may be considered censored. In addition, the domain of such observations may be restricted; for example, values may be required to be non-negative. Methods We propose a regression method for censored observations that also accounts for domain restriction. The method finds maximum likelihood estimates assuming an underlying truncated normal distribution. Results: We show that our method, tcensReg, outperforms other methods commonly used for data with detection limits such as Tobit regression and single imputation of the detection limit or half detection limit with respect to bias and mean squared error under a range of simulation settings. We apply our method to analyze vision quality data collected from ophthalmology clinical trials comparing different types of intraocular lenses implanted during cataract surgery. All methods tested returned similar conclusions for non-inferiority testing, but estimates from the tcensReg method suggest that there may be greater mean differences and overall variability. Conclusions: In the presence of detection limits, our new method tcensReg provides a way to incorporate known domain restrictions when modeling limited dependent variables that substantially improves inferences.

  • Open Access

    microbiomeDASim: Simulating longitudinal differential abundance for microbiome data

    by Justin Williams, Hector Corrada Bravo, Jennifer Tom, Joseph Nathaniel Paulson

    An increasing emphasis on understanding the dynamics of microbial communities in various settings has led to the proliferation of longitudinal metagenomic sampling studies. Data from whole metagenomic shotgun sequencing and marker-gene survey studies have characteristics that drive novel statistical methodological development for estimating time intervals of differential abundance. In designing a study and the frequency of collection prior to a study, one may wish to model the ability to detect an effect, e.g., there may be issues with respect to cost, ease of access, etc. Additionally, while every study is unique, it is possible that in certain scenarios one statistical framework may be more appropriate than another. Here, we present a simulation paradigm implemented in the R Bioconductor software package microbiomeDASim available at http://bioconductor.org/packages/microbiomeDASim microbiomeDASim. microbiomeDASim allows investigators to simulate longitudinal differential abundant microbiome features with a variety of known functional forms with flexible parameters to control desired signal-to-noise ratio. We present metrics of success results on one particular method called metaSplines.


  • Developmental screening and early intervention in a childcare setting for young children at risk for autism and other developmental delays: A feasibility trial

    by Amanda Gulsrud, Themba Carr, Justin Williams, Jonathan Panganiban, Felica Jones, Jackie Kimbrough, Wendy Shih, Connie Kasari

    Efforts to decrease disparity in diagnosis and treatment for under‐resourced children with developmental delays, such as autism spectrum disorder, have led to increased interest in developing programs in community settings. One potential setting that has already demonstrated feasibility in conducting universal screening is the childcare setting. The current study conducted developmental screening in a total of 116 children ages 16–80 months of age in an urban low‐income community childcare center. Parents of 20 children who screened positive were enrolled in the intervention phase of the study, where children received a staff‐delivered targeted early intervention or a waitlist control condition. Given the small and imbalanced sample sizes, confidence intervals from mixed effect models were used to measure changes across time for each group. Of the children who received treatment, there was an average increase in child initiated joint engagement, symbolic play, and language use. This study provides initial feasibility data for the implementation of a screening and early intervention program to service a predominantly low‐resource and ethnically diverse population within the childcare system in a large metropolitan city. Autism Res 2019, © 2019 International Society for Autism Research, Wiley Periodicals, Inc. Lay Summary Identifying and delivering treatment services for young children with developmental delays, such as autism spectrum disorder, may be most successful in community settings, especially for those children from under‐resourced areas. This study found preliminary evidence that the childcare setting is a good place to conduct screening and deliver early interventions for children at risk for autism and other developmental delays.