Examining Naturally Occurring Language- and Literacy-Learning Opportunities for Children with Disabilities Across ECE/ECSE Contexts

Abstract

In this study, we examined the language- and literacy-learning opportunities (LLLO) that children with disabilities experienced across early childhood general (ECE) and special education (ECSE) classroom contexts and compare those opportunities to those experienced by their typically developing peers. We examined differences in the amount of LLLO that children with (n = 258) and without (n = 1,274) identified disabilities experienced in 315 ECE and 116 ECSE classrooms, and whether the association between child status and LLLO varied as a function of differences in these classroom contexts. Results indicated that average LLLO ranged from less than 1 min to 7 min. There were some differences in the LLLO based on context, and the reading opportunities that children experienced varied based on both child status and classroom context. Overall, children with identified disabilities were not provided with increased LLLO, as theory and practical recommendations suggest. Results and implications are discussed.

Keywords

children with disabilities language and literacy early childhood special education early childhood education

Young children’s language and literacy skills are widely recognized as a critical component of early learning. Both are foundational for later academic achievement; early language skills are associated with later literacy and social and emotional development, whereas early literacy skills are associated with later literacy and math scores (e.g., National Early Literacy Panel [NELP], 2008). These language and literacy skills are equally important for children with and at risk for disabilities (e.g., Goldstein, 2011), for whom stronger early skills may serve as protective factors against later academic and behavioral risks. Research and policy recommendations emphasize the importance of providing all children with language- and literacy-learning opportunities (LLLO) to support the development of these critical skills (e.g., National Governors Association, 2013; NELP, 2008), and considerable evidence links the provision of LLLO (e.g., shared reading, code-focused instruction, language-focused interventions) with gains in young children’s language and literacy scores (NELP, 2008). Similar patterns hold true for children with and at risk for disabilities, particularly when instructional factors such as pacing and intensity are appropriately adjusted (e.g., Harn et al., 2011; Justice et al., 2015). Despite this, there are well documented gaps in the skills of children with disabilities from an early age (e.g., Ferrer et al., 2015), underscoring the importance of understanding the LLLO provided in early childhood contexts.

There is a large body of work that considers the provision of LLLO in highly specific research and learning contexts; however less is known about the amount of opportunities that occur during children’s typical experiences in early childhood classrooms. For instance, much of the literature focuses on understanding LLLO is in the context of specific interventions or curricular programs (e.g., Simmons et al., 2007; Weiland, 2016). In other studies, LLLO assessments are limited by a given measurement approach, such as linguistic discourse analyses (e.g., Girolametto et al.,2000), observations of academic engagement and opportunities to respond, (e.g., Greenwood et al., 2002), and global ratings of classroom quality (e.g., Justice et al., 2008). These studies provide important insights into these specific contexts but do not always capture the full range of LLLO that occur and impact children’s classroom learning. A growing body of work focused on measuring classroom processes indicates that there is significant variability in the instruction provided within and across early childhood contexts, and that the instructional opportunities that do occur often happen at less than optimal levels (e.g., Connor et al., 2020; Weiland et al., 2023). There is less research focused on the authentic, typically occurring LLLO of young children with and at risk for disabilities who may be served in either early childhood general education (ECE) or early childhood special education (ECSE) contexts. As such, in this article, we present the results of a study in which we examined the differences in LLLO experienced by children with disabilities across ECE and ECSE contexts, using a more comprehensive measurement approach, given that both contexts are expected to support children’s language and literacy learning and that systematic differences hold important implications for families, policymakers, and researchers alike.

LLLO for Children with Disabilities

To date, a handful of studies have examined the general quality of instruction offered in ECSE classrooms, finding that this is typically in the low range (e.g., Guo et al., 2013, 2014, Pelatti et al., 2016). Such limited instructional support may be problematic for children with for disabilities, who in general require additional instruction to master skills and access curriculum. There are many elements of instruction in addition to global quality that must be considered when developing effective instructional supports for children with disabilities (e.g., an explicit focus on segmenting complex skills, sequencing, practice opportunities; Johnson et al., 2019), however one of the key elements is instructional time (Harn et al., 2008). For instance, there is a large body of work documenting that increased instructional time can lead to positive language and literacy outcomes, particularly for children with lower entry-level skills (e.g., Simmons et al., 2007). Although instructional time in and of itself is not the only indication of instructional intensity, measures of the amount of instruction, particularly when focused on specific content areas, may be one important aspect of instruction (Connor, Piasta, et al., 2009).

Much of the work aimed at increasing language and literacy skills in early childhood contexts has been conducted within intervention frameworks for both children with and at risk for disabilities, in which specific elements of language and literacy instruction are intensified via individual or small-group intervention delivered by researchers or specially trained interventionists, outside of the context of children’s typical educational experiences (e.g., Piasta et al., 2023). Although this type of work is critical for identifying evidence-based practices, children with disabilities also require increased programming for generalization, so that the skills they acquire in intervention-based contexts generalize to other contexts and skill areas (Albin & Horner, 1988). Of late, there has been a push in ECSE for more embedded, naturally occurring instruction to support this process (Division for Early Childhood, 2014). Although this has led to several studies of embedded interventions focused on increasing the LLLO experienced by all children, particularly those with disabilities (e.g., embedding print focused read-alouds or explicit phonological awareness intervention; Justice et al., 2015), we know very little about the typical classroom-based LLLO of children with disabilities that occur in children’s daily educational experiences outside of specific interventions. Collectively, the literature highlights a critical need for research that examines the classroom-based LLLO for children with disabilities across the various early childhood contexts in which they are served.

Early Childhood Service Contexts

Children with disabilities can receive instruction in either ECE or ECSE contexts, which may differ in significant ways. For instance, the number of children with disabilities may vary greatly; ECE classrooms may have one to two children with disabilities, whereas an ECSE classroom can include a majority of children with disabilities or equal numbers of children with disabilities and typical peers. The number of educators can also vary across ECE and ECSE contexts. This is often aligned with the needs of children (e.g., children may receive related services or have an assigned one-on-one educator) but may also be an artifact of the structure of the program (Odom et al., 2011). In addition, the educators who work with children across ECE and ECSE contexts may also differ. Educators in ECE contexts may only be required to hold an associate degree, whereas educators in ECSE classrooms are typically required to hold at least a bachelor’s degree and ECSE credentials (Geiger et al., 2003). Although differences in education alone have not been shown to have a strong direct association with educators’ instruction (e.g., Mashburn et al., 2008), educational experiences may lead to differences in beliefs and knowledge, which may be associated with differences in instruction (e.g., Spear et al., 2018).

Given this variability, and the critical importance of supporting all children’s language and literacy skills, more information is needed on language- and literacy-specific instruction that children with disabilities experience across ECE and ECSE contexts. This information is critical as families and stakeholders consider placement options and weigh the extent to which children with disabilities experience appropriate LLLO within the least restrictive environment. Moreover, understanding the LLLO currently provided across contexts may help researchers and policymakers better support practitioners in providing LLLO that align with recommendations.

The Present Study

Collectively, the literature indicates that children with disabilities should be receiving at least as much, if not more, LLLO than their typical peers, yet we know very little about the instructional opportunities that are provided to children with disabilities within ECE and ECSE classroom contexts. The purpose of the present study was to provide a snapshot of the LLLO that children with disabilities experience across ECE and ECSE contexts. In addition, given research recommendations that children with disabilities may require more instruction than their typically developing peers, we also examined how the LLLO experienced by children with disabilities differed from those of experienced by typical peers. We addressed several research aims. First, we descriptively examined the LLLO that children with disabilities experienced across ECE and ECSE contexts. Next, we examined whether there were differences in the LLLO (a) for children with and without identified disabilities and (b) by classroom context (i.e., ECE/ECSE); we also considered (c) whether the association between child disability status and LLLO varied as a function of classroom context (i.e., interaction effect). Recognizing the fact that this study is not a randomized controlled trial and educators in ECE and ECSE contexts often differ (e.g., Spear et al., 2018), we also examined these differences in LLLO after controlling for multiple measures of educators’ beliefs, knowledge, and background characteristics, all of which were selected based on prior research in the field (e.g., Schachter et al., 2016). We included these educator covariates to determine whether any differences that we might observe were a function of the context or differences in the educators who staff those contexts, and to obtain unbiased estimates regarding the relationships between classroom context, child disability status, their interaction, and LLLO outcomes.

Method

Participants

Data were collected in 2011 to 2014 as part of an evaluation of a statewide professional development for early childhood educators in one Midwestern state (see Piasta et al., 2017, 2020). Participating educators (a) worked as lead, co-lead, or assistant educators in an early childhood classroom, (b) directly taught at least one child who was 4 years of age, and (c) were willing to participate in all study activities. Up to five children were randomly selected from each educator’s classroom, and participating children were those who (a) had parental consent, (b) were proficient in English, and (c) were not diagnosed with severe/profound disabilities such that they could not complete selected study assessments. We prioritized enrollment for those children expected to matriculate to kindergarten the following year. The analytic sample for the present study comprised 431 educators and 1,532 children; we included all participants for whom we had classroom context and child disability status data. The data used in the current study were collected at the fall timepoint, prior to the completion of professional development opportunities.

Educators

Participating educators were 42 years old on average (SD = 10.86). Educators identified as female (97%), 81% Caucasian, 17% African American, 1% Hispanic/Latina; less than 1% identified as multiracial or Asian. In terms of education, 35% of educators had less than a bachelor’s degree, 32% had a bachelor’s degree and 26% had a graduate degree; 73% of educators had an education-related major in their highest degree. Educators had an average of 11 years of experience (range = 0–36) working in early childhood education. Fifty-one percent of educators taught in public programs, with 36% in Head Start programs; 37% taught in center-based programs, and 5% taught in home-based programs. Fifty-two percent of educators reported using Creative Curriculum; 9% used High Scope, 14% reported another curriculum, and 16% reported not using any curriculum.

Educators were identified as teaching in an ECE (n = 315) or ECSE (n = 116) classroom context based on their responses to a background questionnaire. As there were limited guidelines or regulations differentiating ECE and ECSE contexts within the state (Operating Standards for [location masked for blind review] Educational Agencies Serving Children with Disabilities, 2014), classrooms were identified as ECSE contexts if educators (a) responded positively to at least two out of three indicators of ECSE context (i.e., self-identifying as a special educator, labeling classroom as ECSE or a as preschool special education classroom), (b) indicated that they worked in a classroom with eight or more children with an IEP, which under state law requires an early childhood special educator, or (c) indicated that they worked in a classroom with at least six children with an IEP and responded positively to at least one of the three indicators of ECSE context. All other classrooms were identified as ECE contexts.

Children

According to questionnaires completed by a family member, participating children were between 42 and 87 months of age (M = 55 months, SD = 4.89) when fall data were collected. Fifty-one percent of children were boys, and 97% of children spoke English as their primary home language. Children were identified as 73% Caucasian, 16% African American, 1% Asian, 1% American Indian/Alaskan, and 5% Hispanic/Latinx; 6% were identified as multiracial and 2% reported their race as other. Thirty-nine percent of children lived in households with an annual income of less than $25,000, 24% of children lived in households with an annual income between $25,000 and $50,000, and 34% of children lived in households with an annual income above $50,000. In terms of maternal education, the majority of mothers had a high school diploma as the highest degree (53%); 5% had less than a high school diploma, 13% had an associate’s degree, 15% had a bachelor’s degree, and 12% had a graduate degree.

Classification of children with (n = 258) and without identified disabilities (n = 1,274; typically developing peers) was based on families’ responses to a background questionnaire. Children were identified as having disabilities if (a) their families indicated that they had an IEP, or (b) their families indicated that they had a known diagnosis and they were taught in classrooms that we classified as ECSE. Children were identified as typically developing peers if (a) there were no reports of an IEP or known diagnosis, or (b) they were missing IEP or diagnosis data across either ECE or ECSE contexts. Overall, classrooms had four participating children on average, with a range of one to five children. In ECE classrooms, there were 64 children with disabilities and 1,041 typically developing peers. In ECSE classrooms, there were 194 children with disabilities and 233 typically developing peers.

Measures

Data for the current study were collected in the fall of the year during which children participated. LLLO data were coded by research staff from videotaped classroom observations. Information about educator beliefs, knowledge, and background characteristics were collected through educator surveys. Each measure is described below.

LLLO

To measure LLLO, the current study used an adaptation of the Individualizing Student Instruction (ISI) classroom observation system (Connor, Piasta, et al., 2009), which examines a range of instructional opportunities and classroom-based activities experienced by individual children. ISI measures the amount of time in minutes that each individual child spends in classroom activities, including those that target language and literacy content (see Pelatti et al., 2014 for more detail). It is important to distinguish the ISI measure from traditional measures of instructional quantity (e.g., time on task, opportunities to respond, instructional time allocated) that typically focus on either child- or teacher-directed quantities of instruction. ISI accounts for learning opportunities across multiple modalities and interactions, and includes codes for non-instruction (e.g., transitions, snack time) and other instructional opportunities (e.g., math, science); any literacy or language opportunities that occurred in these other contexts was coded as an LLLO. As such, the ISI captures LLLO that occur at any point during an observation, regardless of whether it was during a non-instructional or instructional activity. Given the focus of our analyses, we include all LLLO opportunities but not other non-instructional variables,

The ISI content dimension includes 12 distinct language and literacy content codes that were adapted to reflect recommended early childhood instruction. In the current study, we created four composite scores that aligned with specific domains of language and literacy instruction: (a) code, which included six codes targeting emergent skills that are associated with later decoding skills (e. g., phonological awareness, alphabet knowledge); (b) oral language and vocabulary (language), which included two codes targeting oral and written language and vocabulary; (c) reading, which included three codes targeting reading (e.g., print or text concepts, children or educators reading connected text) or comprehension instruction; and (d) writing, which included codes related to emergent writing. For instance, instruction that aimed to increase children’s ability to manipulate individual speech sounds (e.g., identifying rhymes for the word “hop”) would be categorized as “code” using a phonological awareness indicator in ISI; instruction that aimed to familiarize children with structures of print (e.g., pointing to capital letters a book title), would be categorized as “reading” with a print-text-concept indicator.

Observation Procedures

Trained field assessors videotaped each classroom’s entire instructional window based on days and times identified by educators as representative of their typical instruction. Observations ranged from 24 to 220 min (M = 91.27, SD = 25.36); this is consistent with literature indicating that instructional time varies greatly across contexts (Early et al., 2010), and can generally be attributed to differences in the duration of the instructional day identified by educators or set by the program (e.g., half- vs. full-day; see Preliminary Analyses below for treatment of duration differences). Each classroom was videotaped using two high-quality cameras to capture the full range of activities of all children throughout the classroom. These video pairs were then coded simultaneously with the ISI coding system by trained coders using Noldus Observer Pro software (Noldus Information Technology, 2010). Following standard ISI procedures, instruction was coded at the individual child level for the children participating in the study. For more details on this coding and training process, see Connor, Morrison, et al. (2009) and Pelatti et al. (2014). Twenty percent of videos were randomly selected for double coding. Interrater reliability, as measured by one-way random, single-measure intraclass correlation coefficients (ICCs), was .89 for code, .90 for language, .97 for reading, and .77 for writing, which indicate good to excellent reliability (Cicchetti, 1994).

Educator Covariates

We used multiple measures to examine educator background characteristics, beliefs, and knowledge; all were self-reported via questionnaires. This included educators’ race and ethnicity, gender, age, level of education, and years of experience working with preschool children. Level of education was dummy coded into three categories: (a) less than a bachelor’s degree, (b) a bachelor’s degree, or (c) a graduate degree, as these align with the variable education credentials often held by educators in ECE and ECSE contexts.

We used two measures to examine educators’ instructional orientation, both with response options that ranged from 0 to 4 (0 = strongly disagree, 4 = strongly agree). The Appropriate Practices in Preschool Survey (adapted from Stipek & Byler, 1997) has two subscales (Basic Skills Orientation [APB; 15 items]; Child-Centered Orientation [APC; 11 items]) that reflect educators’ agreement with teacher-directed and child-led instruction. In the current sample, internal consistencies (α) were .84 for APB and .79 for APC (note that we report ordinal alphas for these and other measures utilizing similar scales as these are more accurate for ordinal data; Gadermann et al., 2012). We also used an abbreviated 10-item version of the Modernity Scale (adapted by Justice et al., 2008 from Shaefer & Edgerton, 1985) to characterize educators’ beliefs as more adult or child centered (α = .75). We used two subscales of a self-efficacy measure (Justice et al., 2008, adapted from Bandura, 1997) to measure educators’ general instructional self-efficacy (SEI; eight items; α = .86), and specific language and literacy self-efficacy (SELL; five items; α = .95). Response options ranged from 0 (low feelings of efficacy) to 4 (strong feelings of efficacy). We used the Preschool Teacher Literacy Beliefs measure (PTLB; Hindman & Wasik, 2008; Seefeldt, 2004) to examine educators’ overall endorsement of beliefs consistent with research concerning emergent literacy development and pedagogy. The measures consists of 30 items for which response options range from 0 (strongly disagree) to 4 (strongly agree; α = .87). We also examined these beliefs across four subscales: (a) code (α = .45), (b) language (α = .77), (c) reading (α = .61), and (d) writing (α = .68). Although the internal consistency for code-related beliefs is less than ideal, this may be a function of our sample, as reliability across subscales range from .60 to .87 in the literature (Hindman & Wasik, 2008). An average score was used in analyses for all scales.

We examined educators’ knowledge for practice (KFP) with the Teacher Knowledge Assessment of Early Language and Literacy Development (Neuman & Cunningham, 2009). This measure consists of 70 multiple choice and true-false items, assessing knowledge of children’s development and relevant pedagogical practices (α = .85). We examined educators’ language- and literacy-specific disciplinary content knowledge (DCK) with the Teacher Knowledge Assessment (Cunningham et al., 2009). This measure consists of 19 multiple choice or short answer items assessing knowledge of oral and written language (α = .86).

Analysis Approach

As this was not a randomized control trial, we began our modeling approach with the a priori identification of theoretically important covariates, given that LLLO outcomes were observed in a natural setting where multiple confounding factors may influence LLLO variation. By incorporating teacher and classroom context variables as covariates in the full model, we aimed to account for their potential influence and obtain unbiased estimates regarding the relationships between classroom context, child status, their interaction, and LLLO outcomes. In the base model, we examined only the main effects of context, status, and their interaction. In the full model, we included these effects while also controlling for these a priori identified teacher and classroom context variables.

We conducted a series of preliminary analyses to examine the distribution, nesting, and missingness within our data. Skewness and kurtosis statistics as well as histograms indicated that LLLO outcome data were non-normally distributed with positive skew. The scores of each LLLO outcome exhibited right-skewed distributions, with zero values comprising 18.7% in Code, 14.8% in Reading, 16.7% in Language, and 68.4% in Writing. Shapiro–Wilk statistical tests, adjusted for large samples (Royston, 1982), confirmed the non-normal distribution of all the outcome measures: code (W = 0.70, df = 1,532, p < .001), reading (W = 0.87, df = 1,532, p < .001), language (W = 0.77, df = 1,532, p < .001), and writing (W = 0.43, df = 1,532, p < .001).

Given the hierarchical structure of the data (i.e., LLLO were measured for individual children nested in classrooms), and the fact that data represented counts of minutes spent in various LLLO, our analytic approach involved fitting multilevel negative binomial models using Mplus version 7 with maximum likelihood as the method of estimation (Muthén & Muthén, 1998–2015). Such models are more flexible than multilevel Poisson models for counts and include an error term to account for possible over-dispersion relative to the Poisson (Gardner et al., 1995). The Poisson assumes equi-dispersion (i.e., mean = variance), while all four LLLO outcomes exhibit variance greater than its mean, indicating overdispersion and supporting use of the negative binomial model. Negative binomial models are a generalized linear model based on the log link function; predictions are thus estimates for the natural log of the counts conditional on values of the predictors (O’Connell et al., 2022). As mentioned above, each classroom’s instructional window (as identified by educators) was videotaped and the duration of observation periods ranged from 24 to 220 min, indicating that each child may have different levels of exposure to LLLO (i.e., experience LLLO for different lengths of time) due to the differences in the duration of observation in their individual classrooms. There is an important difference between a child who experienced 3 min of LLLO in a classroom with 24 min observation duration and a child who experienced 3 min of LLLO in a classroom with 220 min observation duration. Assuming all other factors remain the same, a longer observation window would potentially allow for more opportunities for LLLO to occur. In order to account for variation in observation duration for children in different classrooms, duration of observations was log-transformed (to maintain same scale as the log transformed counts in the negative binomial model; Hox et al., 2018) and treated as an “offset” predictor variable in the models in Mplus via a regression coefficient that was fixed to 1 (McCullagh & Nelder, 1989). This approach is typically used when the count variable does not share the same exposure window for every case in the study. In doing so, the dependent variable of the model was converted from “counts” in minutes (raw number of minutes) to “rates” of minutes proportional to the duration of observations (log [counts of minutes spent for each child/duration of observations for that child’s classroom]), so that, by using rates, the counts in minutes were adjusted to be comparable across children from classrooms with different observation durations. This analysis approach ensured that LLLO was evaluated in terms of its relative occurrence rather than being confounded by total observation time given that the duration of observation periods varied across classrooms.

With respect to missingness, child outcomes were completely observed but some observations were missing for educator-level variables. Missingness ranged from <1% (years of experience) to 6% (DCK). Based on further examination of the missing data mechanisms in our data set, the missingness of some variables were associated with observed data. Thus, we treated the missing data mechanism as missing at random (Enders, 2010). We used multiple imputation to handle missing values. Twenty imputed datasets were generated using the mice package in R (van Buuren & Groothuis-Oudshoorn, 2011) for use in the inferential analyses described below.

Results

Descriptive Analyses

Table 1 presents descriptive information characterizing the LLLO that children experienced overall and across ECE and ECSE contexts. Descriptively, children with disabilities experienced more code (Identified disabilities = 4.41min; Typically developing = 3.54 min), reading (Identified disabilities = 7.86 min; Typically developing = 6.71 min), language (Identified disabilities = 3.59 min; Typically developing = 3.24 min), and writing (Identified disabilities = 0.97 min; Typically developing = 0.77 min) opportunities than their typically developing peers. Considering LLLO across classroom contexts, ECSE contexts appeared to provide more LLLO on average across the board. As illustrated in the Table 1 descriptives, children in ECSE contexts experienced more code (ECE = 3.51 min; ECSE = 4.15 min), reading (ECE = 6.36 min; ECSE = 8.32 min), language (ECE = 3.00 min; ECSE = 4.07 min), and writing (ECE = 0.65 min; ECSE = 1.22 min) opportunities on average than children in ECE contexts. Descriptive statistics on educator covariates are presented as Supplemental Material (see Supplemental Table 1).

Table 1.

Descriptive Statistics of LLLO Outcomes in Minutes by Child Status and Classroom Context.

LLLO	Child status	ECE			ECSE			Across contexts
LLLO	Child status	M	SD	Range	M	SD	Range	M	SD	Range
Code	Identified disabilities	3.60	3.99	0.00–14.60	4.68	7.11	0.00–63.56	4.41	6.49	0.00–63.56
	Typically developing	3.51	4.53	0.00–49.86	3.70	4.78	0.00–31.25	3.54	4.58	0.00–49.86
	Across status	3.51	4.50	0.00–49.86	4.15	5.97	0.00–63.56	3.69	4.96	0.00–63.56
Reading	Identified disabilities	7.56	6.88	0.00–33.31	7.96	6.77	0.00–40.22	7.86	6.78	0.00–40.22
	Typically developing	6.28	6.08	0.00–45.83	8.62	6.87	0.00–33.93	6.71	6.30	0.00–45.83
	Across status	6.36	6.14	0.00–45.83	8.32	6.82	0.00–40.22	6.90	6.40	0.00–45.83
Language	Identified disabilities	2.43	2.86	0.00–11.85	3.97	4.59	0.00–22.24	3.59	4.27	0.00–22.24
	Typically developing	3.04	3.77	0.00–31.67	4.15	4.67	0.00–22.47	3.24	3.98	0.00–31.67
	Across status	3.00	3.73	0.00–31.67	4.07	4.63	0.00–22.47	3.30	4.03	0.00–31.67
Writing	Identified disabilities	0.65	1.71	0.00–9.91	1.07	2.47	0.00–19.85	0.97	2.31	0.00–19.85
	Typically developing	0.65	1.77	0.00–19.52	1.34	3.08	0.00–21.64	0.77	2.09	0.00–21.64
	Across status	0.65	1.77	0.00–19.52	1.22	2.82	0.00–21.64	0.81	2.13	0.00–21.64

Note. LLLO are measured at the individual child level. The scores of each LLLO outcome exhibit a right-skewed distribution, with zero values comprising 18.7% in code, 14.8% in reading, 16.7% in language, and 68.4% in writing. LLLO = language and literacy learning opportunities; ECE = early childhood educators; ECSE = early childhood special educators.

Status by Context Analyses

To determine whether the LLLO experienced by children depended on their disability status or classroom context, we fit multilevel negative binomial models. For each LLLO outcome, we fit a base model and a full model. The base model included our three predictors of interest: child disability status (SPED; reference group = typically developing peers), classroom context (ECSE; reference group = ECE), and the interaction between these two variables to examine whether the association between child disability status and learning opportunities varied as a function of classroom context. In addition to these variables, the full model also controlled for educator beliefs (i.e., modernity, APB, APC, overall literacy, code, reading, language, writing), efficacy (i.e., SEL, SELL), knowledge (i.e., KFP, DCK), and background characteristics (i.e., years’ experience, education). One program-related variable (i.e., if classroom was public-school based, center-based, or family/home-based) was also included. The full list of predictors in the base and full models can be seen in Table 2. To assist in interpreting the significant interaction effect between SPED and ECSE, we calculated the rate ratio in the LLLO outcome for children with identified disabilities versus typically developing peers within ECSE versus ECE classrooms by exponentiating the negative binomial fixed effect [e^γ].

Table 2.

Multilevel Negative Binomial Results for the Four LLLO Outcomes.

	Base model				Full model
Parameters	Code	Reading	Language	Writing	Code	Reading	Language	Writing
Fixed effects
Intercept (γ₀₀)	−4.02**	−3.12**	−3.99**	−7.40**	−4.02**	−3.14**	−3.94**	−7.61**
ECSE (γ₀₁)	0.15	0.20	0.20	0.83*	−0.03	0.19	0.12	0.26
SPED (γ₁₀)	−0.02	0.05	−0.08	0.04	−0.01	0.05	−0.09	0.05
Interaction of ECSE × SPED (γ₁₁)	0.02	−0.14*	0.01	−0.26	0.01	−0.14*	0.01	−0.27
Modernity (γ₀₂)					−0.18	0.17	0.14	−0.73
Basic skills orientation (APB) subscale (γ₀₃)					0.33	−0.24	0.29	0.27
Child-centered orientation (APC) subscale (γ₀₄)					−0.35	0.07	0.09	−0.21
Self-efficacy: instruction subscale (γ₀₅)					0.13	−0.34*	−0.05	0.29
Self-efficacy: language and literacy subscale (γ₀₆)					0.14	0.33*	0.16	−0.34
Preschool teacher literacy beliefs (PTLB) (γ₀₇)					−0.62	−2.68	0.06	0.80
PTLB code subscale (γ₀₈)					−0.35	0.33	0.02	−0.98*
PTLB reading subscale (γ₀₉)					0.17	0.31	0.26	0.83
PTLB language subscale (γ₀₁₀)					0.39	0.80	−0.56	−1.38
PTLB writing subscale (γ₀₁₁)					0.61	0.78	0.10	1.01
Knowledge for practice (KFP) (γ₀₁₂)					0.01	−0.01	−0.00	0.05
Disciplinary content knowledge (DCK) (γ₀₁₃)					0.04	0.02	0.07*	0.03
Years of experience (γ₀₁₄)					−0.03**	−0.01	−0.01	−0.05*
Educators’ education level BA (γ₀₁₅)					−0.24	−0.02	−0.22	0.05
Educators’ education level MA+ (γ₀₁₆)					−0.00	0.21	0.19	0.51
Educators’ education level other (γ₀₁₇)					−0.29	−0.12	−0.10	−0.81
Center-based program (γ₀₁₈)					0.34*	−0.11	−0.04	0.51
Home-based program (γ₀₁₉)					0.17	0.24	0.38	0.53
Random effects
Intercept (τ₀₀)	1.92**	1.45**	1.56**	6.91**	1.64**	1.39**	1.44**	6.02**
Dispersion ( $α$ )	0.06*	0.00	0.03	0.14	0.06*	0.00	0.03	0.14

Note. In Mplus, the variance for the negative binomial regression model is defined as $λ$ (1 + $λ$ α) with a dispersion parameter “α” that allows the variance to exceed the mean $λ .$ If the dispersion parameter is not statistically different from zero, the model converges to a Poisson regression model. ECSE = early childhood special education educator (referent group = ECE); SPED = identified special education student (referent group = typically developing peer); BA = bachelor’s degree, MA+ = master’s degree or higher; other = other degree (referent group = no BA); for program type the referent group = public-school based program; all continuous covariates are grand-mean centered.

p < .05. **p < .01.

In the base model, intercept terms represented the log expected rates of minutes spent per observation duration of LLLO for typically developing peers in ECE classrooms (i.e., when ECSE = 0 and SPED = 0). When exponentiated, these terms were 0.02, 0.04, 0.02, and 0.001 which represent the expected event rates of code, reading, language, and writing opportunities respectively for typically developing peers in ECE classrooms. We multiplied the exponentiated values by 90 min, which was an average duration for a classroom’s entire instructional window, to support interpretation. Thus, in a classroom with a 90 min observation duration, the expected length of time (in minutes) for code, reading, language, and writing opportunities for typically developing peers in ECE classrooms were 1.62, 3.98, 1.66, and 0.06, respectively.

In the base model, the main effect of classroom context (i.e., ECSE in Table 2) significantly predicted the amount of writing opportunities (γ₀₁ = 0.83, z = 2.37, p < .01), indicating that children experienced more writing opportunities in ECSE classrooms than in ECE classrooms. However, when we controlled for covariates in the full model, classroom context no longer significantly predicted writing opportunities (γ₀₁ = 0.26, z = 0.56, p = .57). In the base model, the interaction of child disability status and classroom context (i.e., Interaction of ECSE and SPED in Table 2) significantly predicted the amount of reading opportunities (γ₁₁ = −0.14, z = −2.07, p = .04) and this interaction term for reading opportunities remained significant after we controlled for covariates in the full model (γ₁₁ = −0.14, z = −2.04, p = .04). Moreover, the regression coefficient of the interaction term for reading opportunities remained unchanged (γ₁₁ = −0.14) from the base model to full model. The main effects of ECSE and SPED, as well as their interaction effects, were not significant for code or language opportunities.

This significant interaction of child disability status and classroom context on reading opportunities indicated that the amount of reading opportunities that typically developing peers and children with identified disabilities experienced differed based on classroom context (see Figure 1). Within ECSE contexts, children with identified disabilities experienced 0.91 times [exp(γ₁₀ + γ₁₁) = exp(0.05 + (−0.14)) = 0.91)] or 9% fewer [(0.91 − 1) × 100% = −9%] reading opportunities than typically developing peers. For every 90 min of observation duration (and with all covariates at 0, i.e., at their grand mean for continuous variables), children with identified disabilities experienced 4.31 min and typically developing peers experienced 4.71 min of reading opportunities. However, within ECE contexts, children with identified disabilities experienced 1.05 times [exp(0.05) = 1.05] or 5% more [(1.05 − 1) × 100% = 5%] reading opportunities than typically developing peers. For every 90 min of observation duration (with all covariates at 0, or their grand mean), children with identified disabilities experienced 4.10 min and typically developing peers experienced 3.90 min of reading opportunities.

Discussion

The purpose of the present study was to examine the LLLO experienced by children with disabilities across ECE and ECSE contexts. This represents an important contribution to the literature, as very little is known about the amounts of LLLO that children with disabilities experience, whether there are differences in these opportunities across ECE and ECSE contexts, or how these learning opportunities compare to those experienced by children who are typically developing. Given that children with disabilities may require more LLLO than their typical peers to master skills and access curriculum, these results hold important implications for policy, research, and practice concerning ways to best support the language and literacy learning of children with disabilities. Key findings and implications are discussed in turn.

Descriptive Examination of LLLO for Children with Disabilities

Our first research aim was to examine the LLLO that children with disabilities experienced across ECE and ECSE contexts. When we looked specifically at differences in LLLO for children with disabilities compared to their typically developing peers, we saw that on average, children with disabilities experienced more LLLO than their typically developing peers, though these LLLO were notably small in many cases. Children with disabilities experienced approximately 4 min of code and language opportunities, 8 min of reading opportunities, and less than 1 min of writing opportunities.

Collectively, these results indicate that children with disabilities may be receiving less than optimal LLLO. Children with disabilities experienced a range of approximately 1 to 8 min of specific LLLO on a typical day, regardless of context. We note that there is no agreed upon recommendation of the amounts of LLLO that young children should receive in general, or more specifically for children with disabilities. There are, however, some areas of instruction for which research offers a clearer picture. Recommendations for phonological awareness instruction alone are typically 15 to 20 min per day, for 3 to 5 days a week (e.g., Rice et al., 2022); the 4 min of code opportunities children with disabilities in our sample experienced, which included phonological awareness and other code-based skills, are certainly short of this mark.

A brief review of intervention research offers additional insights. Decades of research has focused on identifying effective interventions to promote various language and literacy skills, however there is still a need for more research to identify the ideal intensity (e.g., dosage, duration, frequency; Warren et al., 2007) of specific approaches to maximize language and literacy outcomes. This is especially important given well documented variability of intervention intensities (e.g., Goldfeld et al., 2022), and recent research around thresholds of instruction, or the idea that children may benefit from specified amounts of intervention but have diminishing returns beyond that threshold. Indeed, recent meta-analyses utilizing non-linear methodologies to explore optimal language and literacy intervention dosage for children with and at-risk for disabilities provide evidence of these types of threshold effects based on factors such as hours of intervention, group size, content, and learner characteristics. For example, Roberts et al. (2022) found that intervention effects declined after approximately 40-hours across a range of language and literacy outcomes for K-3rd grade students with reading disabilities, however when interventions were delivered 1:1, there were no threshold effects, and students continued to benefit. Erbeli et al. (2024), on the other hand, found a threshold of approximately 10-hours for small group and 1:1 interventions targeting phonemic awareness for both PK-1st graders with and without disabilities, however when the intervention combined phonemic awareness skills with letter knowledge, that threshold did not hold.

Although there is a need for more research across these areas, these studies highlight the complex interactions of content, intensity, and learner needs that may interact to ensure that literacy and learning opportunities have the maximum effect, and the importance of research that can document these types of interactions. It is important to distinguish these intervention-based indicators from the classroom LLLO we observed in the current study, but these do offer a useful metric for comparison, and our findings provide a starting point, or baseline view, of LLLO for the field. We recognize that these classroom-based LLLO also most likely differ from intervention work in terms of explicitness and other delivery features (e.g., segmenting, sequencing; Johnson et al., 2019); however, they do represent children’s typical classroom-based learning opportunities. Here again there are no clear guidelines for recommended amounts. On the one hand, recent evidence suggests that children with disabilities benefit as much, if not more, than their typically developing peers from a full year of preschool (Weiland, 2016). This may indicate that children with disabilities do not inherently require more LLLO, assuming the instruction is high quality (i.e., adequate dosage, evidence-based curriculum). On the other hand, studies that have examined effective interventions in a multitiered system of support framework with school-aged children often frame these interventions as supplemental learning opportunities that are delivered in addition to comparable or higher amounts of typical classroom-based instruction (e.g., 20 min interventions in addition to a 60 min reading block; Harn et al., 2011).

Collectively, our findings add to the literature and highlight the fact that more research is needed to examine necessary levels of LLLO for supporting young children with disabilities. It is possible that some children with disabilities will learn and progress with the same, uniformly delivered high quality LLLO as their typical peers, but given the variability of early childhood experiences (e.g., Connor, Morrison, et al., 2009), and the fact that children with disabilities often need additional instruction to achieve generalizable effects (e.g., Simmons et al., 2007), the amount of LLLO that children receive in typical classroom experiences is a critical instructional variable.

Differences in LLLO by Child Status

In addition to exploring the LLLO experienced by children with disabilities descriptively, across contexts, we were also interested in whether there were statistical differences in LLLO for children with and without identified disabilities. We did find a significant interaction between child status and context in predicting reading LLLO. Therefore, we include the discussion of the interaction term in the discussion below. Regarding LLLO opportunities for code, language, and writing, we did not find significant differences between the LLLO experienced by children with disabilities and their typical peers. We also removed the non-significant interaction term to observe whether the main effect changed, but the findings remained consistent.

This lack of main effects for most LLLO by child status is notable, as it indicates that in many critical aspects of early literacy, educators are providing comparable amounts of learning opportunities to all children, rather than differentiating and providing children with disabilities increased opportunities that may support them in meeting long-term learning outcomes. The need for differentiation and support may be particularly important for children with disabilities. Under special education law (Individuals with Disabilities Education Improvement Act, 2004), children with disabilities are entitled to a free and appropriate public education; given the research recommendations, our results may signal that the LLLO children with disabilities are receiving may not in fact be “appropriate” for them to make adequate progress. These findings hold particular importance given the 2017 Supreme Court ruling that schools need to support children in making more than “de minimis” progress when providing an appropriate education (Endrew F. vs. Douglas County School District, 2017).

Differences in LLLO by Classroom Context

In our multilevel negative binomial models (base model), we found that the main effects of classroom context on LLLO were statistically significant only for writing. When we controlled for educator covariates, these main effects were no longer significant, indicating that the variation in writing outcomes were explained by differences in the educators who staff ECSE classrooms (see Spear et al., 2018 for additional examinations of educator-level differences). Here again, the lack of main effects for code, language, and writing LLLO across contexts is notable. On the one hand, this finding indicates that the LLLO that children receive are often comparable, regardless of whether they are in an ECE or ECSE context. All children, regardless of context, should have access to high quality, effective early literacy LLLO, and the finding that code, language, and writing LLLO are comparable across contexts may indicate a level of parity around these aspects of early literacy instruction regardless of classroom or educator factors.

On the other hand, given the notably low amounts of code, language, and writing LLLO that occurred across contexts, and the general recommendation that children with disabilities typically will benefit from increased opportunities and practice, these findings again may indicate that children with disabilities are not receiving “appropriate” LLLO, regardless of context. Special education law mandates that children with disabilities are receiving access to education in the least restrictive environment in ways that offer access to full academic participation. This means ensuring that children with disabilities are receiving adequate LLLO to meet their needs. If persistently low and undifferentiated amounts of code, language, and LLLO are common across ECE and ECSE contexts as part of children’s typical educational experiences, this may represent a serious area of concern.

The finding that the main effects of classroom context on writing LLLO were no longer significant when we controlled for educator covariates is also noteworthy. In many ways, the fact that the increased writing LLLO offered in ECSE contexts were explained by differences in educators is expected, given the notable variability of the early childhood workforce in general (e.g., Phillips et al., 2016), and the educational, training, and pedagogical differences between educators in ECE and ECSE contexts specifically. It is important to highlight that these findings do not negate the fact that children are actually experiencing different amounts of LLLO across contexts; they do, however, point to a possible mechanism. Given the increased education and training of educators in ECSE contexts, our results indicate that ECEs in particular, who have more variability in education and experience due to differences in programming and licensure requirements, may require additional professional development opportunities in order provide more LLLO that support all children’s language and literacy outcomes. However, the fact that there were no significant differences between code and language LLLO across contexts should also be considered. This may be particularly concerning around language LLLO – given the role that language plays in the development of all other early literacy skills – and the fact that a large majority of young children with disabilities have challenges related to speech, language, and communication (e.g., speech and language impairments, developmental delays, autism; U.S. Department of Education, 2020). Effective professional development opportunities for both ECEs and ECSEs may be particularly important for ensuring that all children receive optimal LLLO.

Exploration of Interactive Effect of Status and Context in LLLO

Finally, given the complex interactions often at play between children’s disability status and the contexts in which they are served, we examined whether the associations between child disability status and LLLO varied as a function of classroom context. Our results highlight important differences in the reading LLLO children with disabilities receive across classroom contexts. Children with disabilities in ECSE contexts experienced fewer reading opportunities than their typical peers, whereas children with disabilities in ECE contexts experienced more reading opportunities than their typical peers. These findings hold interesting implications.

The difference in reading LLLO for children with disabilities across contexts is notable. One possible explanation may relate to the ways in which children’s disability-related needs and challenges impact instruction. Children with more significant disabilities may be more likely to be served in ECSE contexts, and these children’s individualized goals may not have aligned with language and literacy instruction (e.g., goals focused on functional communication, social skills), and they therefore experienced less LLLO but more individualized instruction targeting these other areas. Similarly, the fact that children with disabilities in ECE contexts experienced more reading LLLO than their typically developing peers may be related, as children served in ECE contexts may have less significant disabilities, and therefore have more language- and literacy-based goals. More research is needed to examine these patterns, as it is plausible that educator decisions around differentiation, instructional delivery, and child-level differences may all influence the amount of LLLO children with disabilities receive across early childhood contexts.

At the same time, the finding that children with disabilities served in ECSE contexts experienced less reading LLLO than their typically developing peers, even while educators in these contexts are offering comparably higher amounts of reading opportunities to other students in the classroom, raises some concerns, particularly given that children with disabilities may often benefit from more LLLO than their typically developing peers. Whereas it is possible that these are intentionally designed instructional decisions, as noted above, it is also possible that educators are providing reading LLLO differentially based more on their perceptions of responsiveness – providing more opportunities to typically developing children who may engage or react differently when presented with a reading LLLO, compared to a child with a more significant disability whose lower emerging literacy skills may lead to challenges with accessing or participating in the LLLO that are provided. It is possible that ECSE educators are recognizing the importance of reading LLLO, and providing them at the classroom level, but perhaps not including children with disabilities in meaningful ways. There is certainly evidence that children with disabilities benefit from interactions with their typically developing peers, however, as noted above, inclusive education in the least restrictive environment is predicated on meaningful inclusion (Odom et al., 2011). Our results, and measurement approach, provide insights that even in ECSE contexts where educators are providing more LLLO overall, there may still be barriers to meaningfully including children with disabilities in accessing LLLO. Our findings offer initial insights to support this line of study, but more research is needed to fully understand differences in LLLO for children with disabilities across contexts.

Limitations, Future Directions, and Conclusion

There are several limitations to the current work that warrant discussion. First and foremost, our data were collected prior to the COVID-19 pandemic, and subsequent increased attention to the “science of reading” that emerged with increased awareness of classroom practices and inequities in education. It is possible that notable shifts in LLLO have occurred since these data were collected, and future research should explore these patterns further. In addition, we did not include children with more profound disabilities, and we did not control for differences in child skill level for the children included in our sample. Future research is necessary to broaden the sample of children with disabilities, and to examine how child-level differences influence the LLLO children with disabilities receive across contexts.

Other limitations pertain to our measure. For instance, we focused specifically on the provision of LLLO, but other aspects of practice (e.g., literacy environment, quality of instruction, quality of interactions, explicitness, practice opportunities for children with disabilities) may also be critically important. There are strong theoretical reasons to focus on the amount of instruction children with disabilities experience, and time is an important component of this broader construct (e.g., dosage, intensity), however there are limitations to the way our measure captured time. In the current study, the duration of observation periods varied across classrooms. To address this, observation length was incorporated into the model as an offset to ensure that observed differences in LLLO were not simply artifacts of unequal observation times. When using this analysis approach, the average LLLO (in minutes) is interpreted as being proportional to the observation length. This approach does not assume that teachers will continue using LLLOs at a consistent rate across time or different classroom context. In natural settings, classrooms differ in schedules and program types, which may influence the overall session length and, consequently, the amount of LLLOs provided. To account for these differences and reduce potential biases, we included a classroom-related variable (i.e., program types) as a covariate in the analysis in addition to other teacher-related covariates. However, some unmeasured contextual factors may still influence the amount of LLLO provided, and future research is suggested to explore these aspects further. In addition, although our measure focused on the amount of LLLO time each student experienced, it did not account for the frequency of high-quality teaching episodes provided by educators within these periods. Variations in instructional efficiency, educator style, and child engagement may result in meaningful differences in the intensity of language and literacy instruction delivered. Future research should incorporate quality, frequency-based (count), and duration-based (time) measures to capture the intensity of high-quality instruction and the overall exposure to LLLO. This approach would provide a more comprehensive understanding of LLLO and to better capture the complexity of language and literacy instruction.

Finally, it is important to note that this work was descriptive and correlational. It represents a unique “snapshot” of naturally occurring LLLO experienced by children with and without disabilities across ECE and ECSE contexts; however further research is needed to explore these patterns in other contexts and determine whether the types of supports hypothesized as potential levers lead to increases in LLLO for children with disabilities.

Despite these limitations, our measure allowed us to detect differences between ECE and ECSE contexts that may have been missed when using more global measures of instruction, and also allowed us to examine differences based on individual students’ disability status. Indeed, this type of measurement approach represents an important contribution to the “next generation” of early childhood observation work (e.g., Gordon & Farran, 2022; Weiland et al., 2023) by providing insights into important variations in classroom experiences that shape child outcomes. Although it was beyond the scope of this project to explore child outcomes, to our knowledge this is one of the first studies to explore these naturally occurring variations specifically between children with and without disabilities in early childhood contexts, and this represents an important area for future research. Given the limitations noted here, we caution over-interpretation of these findings and encourage future researchers to explore whether these patterns are replicated in other contexts.

Collectively, our results offer important information about the classroom-based LLLO that children with disabilities are receiving across ECE and ECSE contexts, and hold important implications for families, policymakers, and researchers alike. Families who are considering different service options for their children with disabilities may want to consider the patterns of difference we found in reading LLLO across ECE and ECSE contexts, and weigh the unique needs and abilities of their children. Given the increased education and training of educators in ECSE contexts, researchers and policymakers may want to consider providing these types of training opportunities to all educators as a possible mechanism for increasing the provision of LLLO. However, given the low levels of instruction observed overall, these findings also highlight the critical need for research, training, and policy work that ensures that language and literacy are key components of all children’s instruction across all early childhood contexts.

Finally, our findings offer an important baseline view of LLLO and highlight the fact that more research is needed to examine necessary levels of LLLO for supporting young children with disabilities. Given these findings and increasing calls for the use of naturally embedded interventions for young children with disabilities across both ECE and ECSE contexts, we encourage additional research to develop clearer guidelines for recommended amounts of LLLO. This type of work is critical to inform efforts ensuring that all early childhood educators, across ECE and ECSE contexts, can provide sufficient classroom-based LLLO to support the language and literacy development of young children with and without identified disabilities.

Supplemental Material

sj-docx-1-tec-10.1177_02711214251372539 – Supplemental material for Examining Naturally Occurring Language- and Literacy-Learning Opportunities for Children with Disabilities Across ECE/ECSE Contexts

Supplemental material, sj-docx-1-tec-10.1177_02711214251372539 for Examining Naturally Occurring Language- and Literacy-Learning Opportunities for Children with Disabilities Across ECE/ECSE Contexts by Caitlin F. Spear, Shayne B. Piasta, Meng-Ting Lo, Laura M. Justice and Ann A. O’Connell in Topics in Early Childhood Special Education

Footnotes

Acknowledgements

We also express deep appreciation for the early childhood agencies, administrators, educators, and PD facilitators involved, as well as the research staff members responsible for coordinating and executing data collection and other research activities.

Author Notes

The opinions expressed are those of the authors and do not necessarily represent the views of the Institute of Education Sciences, U.S. Department of Education, Early Childhood Quality Network, or Ohio Department of Education. Inquiries regarding study data or materials should be directed to the second author.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The research reported here was supported by the Institute of Education Sciences, U. S. Department of Education, through Grant R305E100030 to The Ohio State University. This research would not have been possible without collaboration from the Early Childhood Quality Network, including Dennis Sykes, Sharon Sullivan, Kathryn Rider, and Melissa Ross, or the cooperation of the Ohio Department of Education’s Office of Early Learning and School Readiness.

ORCID iDs

Caitlin F. Spear

Shayne B. Piasta

Meng-Ting Lo

Supplemental Material

Supplementary material for this article is available on the Topics in Early Childhood Special Education website at .

References

Albin

R. W.

Horner

R. H.

(1988). Generalization with precision. In Horner

R. H.

Dunlap

Koegel

R. L.

(Eds.), Generalization and maintenance: Life-style changes in applied settings (pp. 99–120). Paul H. Brookes Publishing.

Bandura

(1997). Self-efficacy: The exercise of control. W.H. Freeman.

Cicchetti

D. V.

(1994). Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychological Assessment, 6, 284.

Connor

C. M.

Adams

Zargar

Wood

T. S.

Hernandez

B. E.

Vandell

D. L.

(2020). Observing individual children in early childhood classrooms using optimizing learning opportunities for students (OLOS): A feasibility study. Early Childhood Research Quarterly, 52, 74–89. https://doi.org/10.1016/j.ecresq.2019.10.001

Connor

C. M.

Morrison

F. J.

Fishman

B. J.

Ponitz

C. C.

Glasney

Underwood

P. S.

Piasta

S. B.

Crowe

E. C.

Schatschneider

(2009). The ISI classroom observation system: Examining the literacy instruction provided to individual students. Educational Researcher, 38, 85–99.

Connor

C. M.

Piasta

S. B.

Fishman

Glasney

Schatschneider

Crowe

Underwood

Morrison

F. J.

(2009). Individualizing student instruction precisely: Effects of child× instruction interactions on first graders’ literacy development. Child Development, 80, 77–100.

Cunningham

Zibulsky

Callahan

(2009). Starting small: Building preschool teacher knowledge that supports early literacy development. Reading and Writing, 22, 487–510. https://doi.org/10.1007/s11145-009-9164-z

Division for Early Childhood. (2014). DEC recommended practices in early intervention/early childhood special education. http://www.dec-sped.org/recommendedpractices

Early

D. M.

Iruka

I. U.

Ritchie

Barbarin

O. A.

Winn

D.-M. C.

Crawford

G. M.

Frome

P. M.

Clifford

R. M.

Burchinal

Howes

Bryant

D. M.

Pianta

R. C.

(2010). How do pre-kindergarteners spend their time? Gender, ethnicity, and income as predictors of experiences in pre-kindergarten classrooms. Early Childhood Research Quarterly, 25, 177–193. https://doi.org/10.1016/j.ecresq.2009.10.003

10.

Enders

C. K.

(2010). Applied missing data analysis. Guilford Press.

11.

Endrew F. vs. Douglas County School District. (2017). 580 U.S.

12.

Erbeli

Rice

Bishop

M. E.

Goodrich

J. M.

(2024). A meta-analysis on the optimal cumulative dosage of early phonemic awareness instruction. Scientific Studies of Reading. Advance online publication. https://doi.org/10.1080/10888438.2024.2309386

13.

Ferrer

Shaywitz

B. A.

Holahan

J. M.

Marchione

K. E.

Michaels

Shaywitz

S. E.

(2015). Achievement gap in reading is present as early as first grade and persists through adolescence. The Journal of Pediatrics, 167(5), 1121–1125.

14.

Gadermann

A. M.

Guhn

Zumbo

B. D.

(2012). Estimating ordinal reliability for Likert-type and ordinal item response data: A conceptual, empirical, and practical guide. Practical Assessment, Research & Evaluation, 17(3), 1–13.

15.

Gardner

Mulvey

Shaw

(1995). Regression analyses of counts and rates: Poisson, overdispersed poisson, and negative binomial models. Psychological Bulletin, 118, 392–404.

16.

Geiger

W. L.

Crutchfield

M. D.

Mainzer

(2003). The status of licensure of special education teachers in the 21st century (COPSSE Document No. RS-7E). University of Florida, Center on Personnel Studies in Special Education.

17.

Girolametto

Hoaken

Weitzman

Lieshout

R. V.

(2000). Patterns of adult-child linguistic interaction in integrated day care groups. Language Speech and Hearing Services in Schools, 31, 155–168. https://doi.org/10.1044/0161-1461.3102.155.

18.

Goldfeld

Beatson

Watts

Snow

Gold

H. N.

Edwards

Connell

Stark

Shingles

Barnett

Eadie

(2022). Tier 2 oral language and early reading interventions for preschool to grade 2 children: A restricted systematic review. Australian Journal of Learning Difficulties, 27(1), 65–113.

19.

Goldstein

(2011). Knowing what to teach provides a roadmap for early literacy intervention. Journal of Early Intervention, 33, 268–280. https://doi.org/10.1177/1053815111429464

20.

Gordon

R. A.

Farran

(2022). Introduction to ECRQ special issue measuring quality in early care and education: Past, present, and future. Early Childhood Research Quarterly, 60, 374–378.

21.

Greenwood

C. R.

Horton

B. T.

Utley

C. A.

(2002). Academic engagement: Current perspectives on research and practice. School Psychology Review, 31(3), 328–349.

22.

Guo

Dynia

J. M.

Pelatti

C. Y.

Justice

L. M.

(2014). Self-efficacy of early childhood special education teachers: Links to classroom quality and children’s learning for children with language impairment. Teaching and Teacher Education, 39, 12–21. https://doi.org/10.1016/j.tate.2013.11.005

23.

Guo

Sawyer

B. E.

Justice

L. M.

Kaderavek

J. N.

(2013). Quality of the literacy environment in inclusive early childhood special education classrooms. Journal of Early Intervention, 35, 40–60. https://doi.org/10.1177/1053815113500343

24.

Harn

B. A.

Chard

D. J.

Biancarosa

Kame’enui

E. J.

(2011). Coordinating instructional supports to accelerate at risk first-grade readers’ performance. The Elementary School Journal, 112, 332–355.

25.

Harn

B. A.

Linan-Thompson

Roberts

(2008). Intensifying instruction: Does additional instructional time make a difference for the most at risk first graders? Journal of Learning Disabilities, 41, 115–125. https://doi.org/10.1177/0022219407313586

26.

Hindman

A. H.

Wasik

B. A.

(2008). Head start teachers’ beliefs about language and literacy instruction. Early Childhood Research Quarterly, 23, 479–492. http://dx.doi.org/10.1016/j.ecresq.2008.06.002

27.

Hox

Moerbeek

Van de Schoot

(2018). Multilevel analysis: Techniques and applications. Routledge.

28.

Individuals with Disabilities Education Improvement Act. (2004). 20 U.S.C. § 1400.

29.

Johnson

E. S.

Zheng

Crawford

A. R.

Moylan

L. A.

(2019). Developing an explicit instruction special education teacher observation rubric. The Journal of Special Education, 53(1), 28–40. https://doi.org/10.1177/0022466918796224

30.

Justice

L. M.

Logan

J. A. R.

Kaderavek

J. N.

Dynia

J. M.

(2015). Print-focused read-alouds in early childhood special education programs. Exceptional Children, 81, 292–311. https://doi.org/10.1177/0014402914563693

31.

Justice

L. M.

Mashburn

A. J.

Hamre

B. K.

Pianta

R. C.

(2008). Quality of language and literacy instruction in preschool classrooms serving at risk pupils. Early Childhood Research Quarterly, 23, 1–22. https://doi.org/10.1016/j.ecresq.2007.09.004

32.

Mashburn

A. J.

Pianta

R. C.

Hamre

B. K.

Downer

J. T.

Barbarin

O. A.

Bryant

Burchinal

Early

D. M.

Howes

(2008). Measures of classroom quality in prekindergarten and children’s development of academic, language, and social skills. Child Development, 79, 732–749.

33.

McCullagh

Nelder

J. A.

(1989). Generalized linear models (Vol. 37). CRC Press.

34.

Muthén

L. K.

Muthén

B. O.

(1998–2015). Mplus user’s guide (7th ed.). Muthén & Muthén.

35.

National Early Literacy Panel (NELP). (2008). Developing early literacy: Report of the National Early Literacy Panel. National Institute for Literacy.

36.

National Governors Association. (2013). A governor’s guide to early literacy: Getting all students reading by third grade. National Governors Association. http://www.nga.org/files/live/sites/NGA/files/pdf/2013/1310NGAEarlyLiteracyReportWeb.pdf

37.

Neuman

S. B.

Cunningham

(2009). The impact of professional development and coaching on early language and literacy instructional practices. American Educational Research Journal, 46, 532–566. https://doi.org/10.3102/0002831208328088

38.

Noldus Information Technology, (2010). Version 10.5. Noldus.

39.

O’Connell

A. A.

Bhaktha

Zhang

(2022). Single and multilevel models for count outcomes. In O’Connell

A. A.

McCoach

D. B.

Bell

B. A.

(Eds.), Multilevel modeling methods with introductory and advanced applications (pp. 255–304). Information Age Publishing.

40.

Odom

S. L.

Buysse

Soukakou

(2011). Inclusion for young children with disabilities: A quarter century of research perspectives. Journal of Early Intervention, 33, 344–356.

41.

Operating Standards for Ohio Educational Agencies Serving Children with Disabilities. (2014, July 1). http://education.ohio.gov/Topics/Special-Education/Federal-and-State-Requirements/Operational-Standards-and-Guidance

42.

Pelatti

C. Y.

Dynia

J. M.

Logan

J. A. R.

Justice

L. M.

Kaderavek

(2016). Examining quality in two preschool settings: Publicly funded early childhood education and inclusive early childhood education classrooms. Child & Youth Care Forum, 45, 1–21. https://doi.org/10.1007/s10566-016-9359-9

43.

Pelatti

C. Y.

Piasta

S. B.

Justice

L. M.

O’Connell

(2014). Language- and literacy-learning opportunities in early childhood classrooms: Children’s typical experiences and within-classroom variability. Early Childhood Research Quarterly, 29, 445–456. https://doi.org/10.1016/j.ecresq.2014.05.004

44.

Phillips

Austin

L. J.

Whitebook

(2016). The early care and education workforce. The Future of Children, 26(2), 139–158.

45.

Piasta

S. B.

Farley

K. S.

Mauck

S. A.

Soto Ramirez

Schachter

R. E.

O’Connell

A. A.

Justice

L. M.

Spear

C. F.

Weber-Mayrer

(2020). At-scale, state-sponsored language and literacy professional development: Impacts on early childhood classroom practices and children’s outcomes. Journal of Educational Psychology, 112(2), 329–343. https://doi.org/10.1037/edu0000380

46.

Piasta

S. B.

Justice

L. M.

O’Connell

A. A.

Mauck

Weber-Mayrer

Schachter

R. E.

Farley

K. S.

Spear

C. F.

(2017). Effectiveness of large-scale, state-sponsored language and literacy professional development on early childhood educator outcomes. Journal of Research on Educational Effectiveness, 10, 354–378.

47.

Piasta

S. B.

Logan

J. A. R.

Zettler-Greeley

C. M.

Bailet

L. L.

Lewis

Thomas

L. J. G.

(2023). Small-group, emergent literacy intervention under two implementation models: Intent-to-treat and dosage effects for preschoolers at risk for reading difficulties. Journal of Learning Disabilities, 56(3), 225–240. https://doi.org/10.1177/00222194221079355

48.

Rice

Erbeli

Thompson

C. G.

Sallese

M. R.

Fogarty

(2022). Phonemic awareness: A meta-analysis for planning effective instruction. Reading Research Quarterly, 57(4), 1259–1289. https://doi.org/10.1002/rrq.473

49.

Roberts

G. J.

Dumas

D. G.

McNeish

Coté

(2022). Understanding the dynamics of dosage response: A nonlinear meta-analysis of recent reading interventions. Review of Educational Research, 92(2), 209–248. https://doi.org/10.3102/00346543211051423

50.

Royston

J. P.

(1982). An extension of Shapiro and Wilk’s W test for normality to large samples. Applied Statistics, 31(2), 115–124.

51.

Schachter

R. E.

Spear

C. F.

Piasta

S. B.

Justice

L. M.

Logan

J. A.

(2016). Early childhood educators’ knowledge, beliefs, education, experiences, and children’s language-and literacy-learning opportunities: What is the connection? Early Childhood Research Quarterly, 36, 281–294.

52.

Seefeldt

(2004). Preschool teacher literacy beliefs questionnaire. Johns Hopkins University Center for the Social Organization of Schools.

53.

Shaefer

E. S.

Edgerton

(1985). Parent and child correlates of parental modernity. In Sigel

I. E.

(Ed.), Parental belief systems (pp. 287–318). Lawrence Erlbaum.

54.

Simmons

D. C.

Kame’enui

E. J.

Harn

Coyne

M. D.

Stoolmiller

Santoro

L. E.

Smith

S. B.

Beck

C. T.

Kaufman

N. K.

(2007). Attributes of effective and efficient kindergarten reading intervention: An examination of instructional time and design specificity. Journal of Learning Disabilities, 40, 331–347. https://doi.org/10.1177/00222194070400040401

55.

Spear

C. F.

Piasta

S. B.

Ottley

J. R.

Yeomans-Maldonado

Justice

L. J.

O’Connell

(2018). Early childhood special and general educators: An examination of similarities and differences in beliefs, knowledge, and practice. Journal of Teacher Education, 69, 263–277. https://doi.org/10.1177/0022487117751401

56.

Stipek

D. J.

Byler

(1997). Early childhood education teachers: Do they practice what they preach? Early Childhood Research Quarterly, 12, 305–325. https://doi.org/10.1016/S0885-2006(97)90005-3

57.

U.S. Department of Education. (2020). EDFacts Data Warehouse (EDW): IDEA Part B child count and educational environments collection, 2018–19. http://go.usa.gov/xdp4T.

58.

van Buuren

Groothuis-Oudshoorn

. (2011). mice: Multivariate Imputation by Chained Equations in R. Journal of Statistical Software, 45(3), 1–67.

59.

Warren

S. F.

Fey

M. E.

Yoder

P. J.

(2007). Differential treatment intensity research: A missing link to creating optimally effective communication interventions. Mental Retardation and Developmental Disabilities Research Reviews, 13, 70–77.

60.

Weiland

(2016). Impacts of the Boston prekindergarten program on the school readiness of young children with special needs. Developmental Psychology, 52(11), 1763.

61.

Weiland

Moffett

Rosada

P. G.

Weissman

Zhang

Maier

Snow

McCormick

Hsueh

Sachs

(2023). Learning experiences vary across young children in the same classroom: Evidence from the individualizing student instruction measure in the Boston public schools. Early Childhood Research Quarterly, 63, 313–326. https://doi.org/10.1016/j.ecresq.2022.11.008

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB