School-Entry Skills and Early Skill Trajectories Predict Reading after 1 Year

By: Tracy A. Cameron
Department of Psychology, University of Otago
Elizabeth Schaughency;
Department of Psychology, University of Otago
Mele Taumoepeau
Department of Psychology, University of Otago
Craig McPherson
Department of Psychology, University of Otago
Jane L. D. Carroll
Department of Psychology, University of Otago

Acknowledgement: This study was conducted as part of the PhD research of Tracy A. Cameron and Honours research of Craig McPherson. Portions of these data were included in a symposium at the Educational Psychology Forum (2018) and an abstract accepted for presentation at the Twenty-Seventh Annual Society for the Scientific Study of Reading Conference (2020). The authors thank the children, parents, and participating schools whose participation made this research possible. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported, in part, by the University of Otago Doctoral Scholarship and a Graduate Women New Zealand Harriette Jenkins Award, both to Tracy A. Cameron.
Appendices containing computer code used in the modified latent change score modeling have been submitted to accompany this article.

Research demonstrates the importance of successful reading acquisition; earlier skills set the stage for later skill development. Prereading skills at the age of 5 years, before reading instruction, predict reading fluency at the end of first grade, which, in turn, contributes to second-grade reading comprehension (van Bergen et al., 2021). Furthermore, literacy skill growth in the early years predicts later reading achievement, with disparities in reading-related skills widening across development (C. E. Cameron et al., 2015). These findings highlight the significance of the early years for reading development and need for methods to study this developmental change. Latent change score (LCS) approaches provide educational researchers with a flexible framework for studying intraindividual development and individual differences in development (Grimm et al., 2016). In this study, we developed a novel LCS to index children’s trajectories of early literacy and beginning reading development derived from progress monitoring across the first 6 months of reading instruction. We then used these modified LCSs in multivariate analyses investigating contributions of school-entry skills and developing early literacy skills during beginning reading instruction to children’s literacy progress after 1 year of school and literacy instruction in New Zealand (NZ).

Early Literacy Skills in the First Year of School

Learning to read is a developmental process (Ehri, 2014). Early literacy skills predict, and are theoretical precursors of, later reading performance (Petscher et al., 2020; Shanahan & Lonigan, 2010). However, describing skills in development is like trying to hit a moving target (Speece, 2005), with implications for what skills are assessed, at what point, and over what time frame (Petscher et al., 2020). Moreover, learning to read is situated within cultural–linguistic contexts, which may also influence predictors of literacy development (e.g., Zugarramurdi et al., 2022). In this study, we focus on reading acquisition in English-medium instruction in NZ.

Overall, NZ’s young people achieve adequately in reading: Mean performance of NZ 15-year-olds in the Programme for International Study Assessment (PISA) study was above the Organisation for Economic Co-operation and Development average for reading, most (80%–85%) NZ participants achieving at or above PISA’s minimum target (Level 2; OECD, 2019). However, PISA findings also reflect variability in reading achievement; 15%–20% of NZ participants performed below Level 2, suggesting poor reading performance (OECD, 2019).

In NZ, children typically commence primary school and formal literacy instruction on or around their fifth birthday. Traditionally, systemic screening of literacy progress occurs 1 year after school entry (age 6), conducted to inform supplemental instruction usually provided between ages 6 and 7 (Schluter et al., 2020). At this 1-year assessment, teachers consider children’s overall reading progress and whether they have achieved instructional book level targets in leveled readers (Ministry of Education [MoE], 2010a).

Recently, the NZ MoE Learning Support Action Plan prioritized earlier identification of, and intervention for, literacy acquisition difficulties (MoE, 2019), with the aim of preventing the literacy challenges experienced by the significant minority of young people in NZ (McNaughton, 2020). Empirically identified skills that forecast later reading achievement may be useful for identifying risk for reading difficulties (Petscher et al., 2020). To inform assessment and instructional approaches for supporting reading achievement, innovative research approaches are needed that build on the relevant knowledge base in reading acquisition and combine rigorous educational and developmental research methodologies.

Skills at School Entry

At age 5 (or U.S. kindergarten), early literacy skills that predict later literacy achievement in English include phonological awareness and alphabet knowledge (National Early Literacy Panel, 2008) and oral language (e.g., Dickinson et al., 2019). Phonological awareness is a metacognitive or metalinguistic skillset involving sound units in spoken words, progressing developmentally from awareness of larger units (e.g., words comprising compound words) to smaller units (e.g., phonemes, the individual sounds within words; Anthony & Francis, 2005). Phoneme awareness, in particular, is associated with literacy acquisition (Ehri, 2014); however, at school entry in English-speaking countries, many children have only minimally developed phoneme awareness (Anthony & Francis, 2005; Clayton et al., 2020). Alphabet knowledge includes the names and sounds of letters (i.e., letter-sound correspondence; Shanahan & Lonigan, 2010). Preschool children in English-speaking contexts do not typically induce the relationship between graphemes (e.g., letters) and the sounds they represent unless this awareness is intentionally fostered (Castles et al., 2018). Subsequently, although awareness of letter names and sounds generally correlate in U.S. kindergarten, children often have greater letter-name knowledge at school entry (Nese et al., 2017). Combined, preschool phonological awareness and letter knowledge are conceptualized as code-related skills, predicting later word recognition (e.g., Hjetland et al., 2020).

Oral language is multifaceted and includes both receptive and expressive skills (Dickinson et al., 2019; Schaughency et al., 2017). Preschool oral language and early literacy skills are interrelated, both predicting children’s reading at the end of U.S. first grade (Dickinson et al., 2019). Linguistic comprehension, along with code-related skills involved in word reading, is posited to be important for reading comprehension (Gough & Tunmer, 1986, see Castles et al., 2018, for discussion). Given this relationship, listening comprehension tasks are recommended, but rarely included, to assess comprehension processes in the early years (Hjetland et al., 2020; Lonigan & Burgess, 2017).

Developing Skills in Beginning Reading Instruction

Once children enter school and begin formal reading instruction, they begin their journeys toward becoming nascent readers. Repeated measurement affords opportunities to model growth of skills in development (e.g., Clayton et al., 2020; Clemens et al., 2018).

Phonemic Awareness

Phonemic awareness typically grows during beginning reading instruction (Castles et al., 2018). Literacy researchers have employed tools to tap developing phonemic awareness during early schooling in English-speaking contexts, assessing earlier developing onset phoneme awareness (T. A. Cameron et al., 2020; Oslund et al., 2017) to more advanced phoneme segmentation skills (Ritchey & Speece, 2006; Tindal et al., 2015). Phoneme segmentation tasks may be prone to floor effects at school entry (Tindal et al., 2015), whereas onset phoneme tasks continue to correlate with letter-sound knowledge at midyear (Burns et al., 2018; T. A. Cameron et al., 2020).

Letter-Sound Knowledge

Children commencing U.S. kindergarten typically do not perform well on letter-sound correspondence tasks (Nese et al., 2017), although they typically demonstrate growth over the year (Nese et al., 2017; Sáez et al., 2016). Early kindergarten may be a potentially important developmental window for letter-sound correspondence growth (T. A. Cameron et al., 2020; Clemens et al., 2020).

Word Reading

Reading research documents the importance of accurate and efficient word reading (Petscher et al., 2020). Word reading tasks have been recommended to index reading development in older children (e.g., U.S. first grade; Fuchs et al., 2004; NZ Year 2: Schaughency et al., 2015), with recent research examining word reading development of younger children. Clayton et al. (2020) followed 4- to 5-year-old children starting school in London each term over a year: There were floor effects on word reading during the first assessment in the fall (Term 1), but strong growth on word reading from winter to spring (Terms 2–3). Moreover, growth in word reading across in the latter half of U.S. kindergarten predicts important reading outcomes (Clemens et al., 2018).

Developing Early Literacy Skills and Beginning Reading

Phoneme awareness and letter-sound knowledge, and growth in letter-sound knowledge, in the first half of U.S. kindergarten predict word reading performance and growth later in the year (Clemens et al., 2020). Theoretically, phoneme awareness and letter-sound correspondence contribute to developing word reading (Ehri, 2014); therefore, Clemens et al. (2018) posit following word reading growth concurrently with these early literacy skills that provide information about children’s emerging word reading skills in beginning reading instruction. To investigate this proposition, T. A. Cameron, Carroll, et al. (2022) concurrently assessed onset phoneme awareness, letter-sound correspondence, and word reading across the first 6 months of school and reading instruction in NZ. All measures depicted general growth. In addition, time series analyses indicated intraindividual increase in either onset phoneme awareness or letter-sound correspondence predicted intraindividual increase in word reading over this period (T. A. Cameron, Carroll, et al., 2022), consistent with theorized associations between code-related early literacy skills and word reading (Hjetland et al., 2020). Combined, these results suggest the first 6 months of school may be an important window for literacy skill development. However, T. A. Cameron, Carroll, et al. (2022) did not test whether children’s individual early learning trajectories in the first half of the year contributed to prediction of progress at the end of the year above and beyond oral language and early literacy skills assessed at school entry.

Methodological Considerations

Investigating contributors to children’s early learning and later progress in reading acquisition raises a number of methodological considerations.

Design and Measurement Considerations

Reciprocal influences between early literacy skills and reading acquisition create challenges for studying predictive relations between these constructs. Subsequently, longitudinal studies are needed that assess skills prior to reading acquisition (Zugarramurdi et al., 2022). NZ enjoys a high participation rate in early childhood education and care (ECEC); in 2021, 97% of children participated in ECEC prior to school entry (MoE, 2021). NZ ECEC takes a variety of forms, from center-based to home-based (a.k.a., family daycare) services. Despite the varied forms of ECEC, registered services follow a common child-centered, largely play-based, curriculum (MoE, 2017). Subsequently, assessment at school entry occurs prior to formal literacy instruction in NZ.

In most countries, children vary in age at the start of school due to cut-off dates for school entry (Mavilidi et al., 2022). Beginning schooling in NZ, in which children enter school, and begin reading instruction, typically on their fifth birthday, provides a unique opportunity for following children’s literacy development, controlling for potential relative age effects in early learning (where relatively older students may be advantaged over younger students; Mavilidi et al., 2022). To our knowledge, T. A. Cameron, Carroll, et al. (2022) are the first to follow children’s early literacy and word reading acquisition across the first 6 months of school. Growth mixture modeling (GMM) of these data revealed differing broad trajectories (i.e., latent classes) of skill development for each measure.

T. A. Cameron, Carroll, et al. (2022) found associations between these latent classes (categories) of growth trajectories and criterion measures included after progress monitoring. In this study, we use multivariate analyses to simultaneously model relations between children’s skills at school entry, early literacy learning, and a theoretically informed array of measures after 1 year of school. In particular, reading and spelling acquisitions share foundational phoneme awareness and letter-sound correspondence skills (Clemens et al., 2014; Ehri, 2014), and children’s spelling at the end of U.S. kindergarten predicts their later reading (Treiman, Hulslander, et al., 2019). Therefore, in this study, we also include a measure of pseudoword spelling as well as measures of word reading, broad early reading skills, and school-used indices of reading progress.

Modeling Early Literacy Skill Development Over Time

Repeated measurement of skills can depict learning over time, with different approaches to analyze longitudinal data contributing to understanding reading development. Path analysis can elucidate direct and indirect contributions of predictors to later performance (e.g., van Bergen et al., 2021). However, characterizing progress monitoring data of skills in development poses analytic challenges (Petscher et al., 2016). Children’s level and growth are important considerations when analyzing progress monitoring data. Two children may display similarly limited growth but very different initial scores, with, for example, one child performing near ceiling and the other very poorly and making limited progress. Analytic approaches such as GMM capture this variation in level and growth through the identification of broad classes of growth (see Grimm et al., 2017) but do not provide individualized indices of intraindividual growth for each child.

LCS models are a relatively new approach that can be used to investigate individual differences in change over time (Grimm et al., 2016; Petscher et al., 2016). LCS models view change as a dynamic process (Serang et al., 2019). Rather than modeling the outcome y as level of performance at a given time, LCS uses the change in y from time t − 1 to time t as the modeled outcome (Grimm et al., 2016). In particular, “dual change” LCS models provide a flexible approach that simultaneously estimates average growth (α) and proportional change (β), allowing consideration of individual differences in growth (Petscher et al., 2016). However, dual change LCS models typically capture performance (level/intercept) and growth over time as separate variables (e.g., Serang et al., 2019). Subsequently, research to date has modeled these components separately rather than capturing both level and growth in a single metric. To address this shortcoming in analytic methods, our interest was in modifying the LCS model to combine level and growth into a single score, which would allow us to track children’s growth, while accounting for their initial level.

This Study

This study modeled contributions of school entry and developing early literacy skills to literacy progress after 1 year of school in NZ. We analyzed data collected from children who completed early literacy measures during three data collection phases (see T. A. Cameron, Carroll, et al., 2022): (a) at school entry, assessing children’s oral language and early literacy skills; (b) five probes, administered every fourth school week, across the first 6 months of school, assessing phoneme awareness, letter-sound correspondence, and beginning word reading; and (c) after 1 year of school, researcher-administered measures of reading-related skills and school-used measures of literacy progress.

Data were analyzed to address two research questions (RQ): (RQ1) Do individual LCSs resulting from LCS models, modified to capture both the intercept and growth indices, correspond with children’s empirically derived trajectory class for progress monitoring measures? Serang et al. (2019) provide code for a dual change LCS model, yielding latent true (intercept) and change (growth) scores. To create a variable capturing individual children’s progress over time, we adjusted their code to obtain a single latent variable combining these components. We hypothesized that modified LCS for each progress monitoring measure would correspond with children’s trajectory class derived from growth mixture models (T. A. Cameron, Carroll, et al., 2022) and predict children’s performance after 1 year of school.

(RQ2) Do school entry and/or early literacy skill growth indexed by our mLCS predict literacy progress after 1 year of school? Based on previous research, we hypothesized that both school-entry skills and early literacy skill development would predict literacy skills and reading progress after 1 year of school. Moreover, we hypothesized that path analyses would elucidate indirect effects from skills at school entry to 1-year outcomes via early literacy skill development. Such findings would point to the first 6 months of school as a potentially important window for monitoring children’s early learning. However, based on research and theory on the pathways to reading (e.g., Castles et al., 2018; Hjetland et al., 2020), we also considered that specific results might vary across outcome measures (e.g., specific researcher-administered or school-used measures).

Method

This study was approved by our institution’s Human Ethics Committee. Participating schools distributed letters of invitation to participate in this longitudinal study to parents of children entering their school. Parents’ written consent and children’s verbal assent were obtained prior to participation.

Participants

Participants were 105 children (n = 55 boys) starting their first year of school (M = 5.05 years old [4.92, 5.67], SD = 0.11, NZ Year 0/1), in eight state and state-integrated (i.e., publicly funded, New Zealand Immigration, 2020) schools in a NZ South Island city. Schools served children from a variety of socioeconomic backgrounds as reflected in MoE assigned decile rankings (2–10; MoE, 2022), with most participants attending mid–high decile schools (Crooks et al., 2010). All but three children (97.14%) attended some form of teacher-led ECEC prior to primary school, consistent with national ECEC participation rates (MoE, 2021). Reported maternal education was postsecondary for the majority (n = 77; 73.33%), with the remainder attending or completing high school (n = 21; 20.00%) or leaving school after primary or intermediate (n = 7; 6.67%).

Ethnic composition of the student population at participating schools varied. The sample’s ethnic composition is described using the total response method (Statistics New Zealand, 2005). Parents identified 18.10% of participating children as belonging to more than one ethnic group, with 91.43% identified as NZ European, 13.33% Māori, 2.86% Pacific, and 11.43% other (e.g., Asian, European), falling within the range of the population from which it was drawn. All participants spoke English as their primary language.

Literacy instruction was in English, following the NZ curriculum for English-medium instruction (MoE, 2009). At the time of data collection, early literacy instruction in NZ primary schools generally included guided reading, using leveled readers (MoE, 2010a).

Measures

School entry, progress monitoring, and researcher-administered 1-year measures were undertaken by four psychology graduate students whose first language was NZ English, after training in administration and scoring. Training for each measure included reviewing manuals and training materials, followed by video-recorded practice assessments with at least three nonparticipating target-age children, with videos viewed by a clinical psychologist and speech–language therapist for administration integrity. Participants’ assessments were video-recorded with independent double scoring of a randomly selected subset of children to evaluate interscorer reliability (intraclass correlation coefficient = 0.96–1.00, 95% CI [0.92, 1.00], p < .001). Time frames for data collection are depicted in Figure 1 and described in the sections below.

School Entry

Lack of familiarity with structured assessment contexts presents potential barriers to young children’s performance (Pretti-Frontczak et al., 2014). Assessment in NZ ECEC typically involves observations or narrative descriptions (learning stories) of children’s learning experiences (MoE, 2017); subsequently, most NZ children have limited experience with structured assessments prior to school entry. Kaminski et al. (2018) developed the Preschool Early Literacy Indicators (PELI®) to assess oral language and early literacy skills in 3- to 5-year-old children, embedding assessment tasks in a 10- to 15-min shared reading activity. As in other English-speaking countries (Kaplan & Walpole, 2005), NZ preschool children typically have experienced shared reading with an adult (Morton et al., 2017); consequently, PELI® provides a child-friendly approach for assessing skills at school entry.

Typically within 2 weeks of starting school, children were assessed on the PELI® story (form) recommended for end of preschool, “A Day at the Beach” (Kaminski et al., 2018), adapted for NZ English (see T. A. Cameron, Carroll, & Schaughency, 2022, for details). Composite scores were calculated using PELI®‘s manual-provided formula. In this sample, PELI®NZ demonstrated concurrent validity with criterion measures of similar constructs (rs = .52–.82; T. A. Cameron, Carroll, & Schaughency, 2022) and predictive validity for later early reading skills (rs = .53–.64; T. A. Cameron, Carroll, et al., 2022).

First 6 Months of School: Progress Monitoring

Beginning approximately 1.5 months after starting school, children were assessed every fourth school week for the next 5 months on alternate-form 1-min fluency-based measures. Assessment order was: Dynamic Indicators of Basic Early Literacy Skills First Sound Fluency (FSF) (Good & Kaminski, 2011), AIMSweb Letter Sound Fluency (LSF) (Shinn & Shinn, 2002), and NZ Word Identification Fluency (NZWIF-Y1) (T. A. Cameron, Carroll, et al., 2022).

FSF and LSF

FSF assesses children’s onset phoneme awareness. Children were asked to identify the first sound of up to 30 words read aloud by the administrator for 1 min. LSF assesses children’s letter-sound correspondence. Children were asked to provide the sounds for as many letters as they could from a list of 100 randomly selected lowercase printed letters. Prior NZ research using FSF and LSF with children in their first year of school has shown good alternate forms reliability, growth over time, and predictive validity for later literacy measures (T. A. Cameron et al., 2020).

NZWIF-Y1

NZWIF-Y1 is a word reading fluency task designed specifically for children in their first year of school in NZ (T. A. Cameron, Carroll, et al., 2022). Children had 1 min to read aloud as many words as they could. Children’s scores were the number of correctly read words. If a child read all 50 words in under a minute, their score was prorated.

In this sample, FSF, LSF, and NZWIF-Y1 showed good alternate forms reliability (rs = .70–.95) and predictive validity (rs = .46–.79) for later literacy measures (see T. A. Cameron, Carroll, et al., 2022).

After 1 Year of School: Criterion Measures

Researcher-Administered Measures

Wechsler Individual Achievement Test–Australian and New Zealand Standardised, Third Edition (WIAT-IIIA&NZ): Early Reading Skills (Joshua et al., 2016)

This untimed 6- to 12-min subtest includes 34 items covering early literacy and reading skills (e.g., alphabet knowledge, rhyming, starting and ending sounds, and reading). We used age-based standardized scores. Joshua et al. (2016) reported split-half reliability (.94; 6-year-olds) and concurrent correlations with reading measures (.48–.55; ages 4–8 years) in the Australian and NZ standardization sample.

Test of Phonological Awareness: Early Elementary “Letter Sounds” (Torgesen & Bryant, 2004)

For this untimed 10- to 15-min pseudoword spelling task, the administrator read a script asking children to write the names of “funny animals.” The administrator read each name aloud and provided a rhyming real word for a total of 18 names. This small-group task was administered to three to five children, positioned so they could not see other children’s responses. Various metrics for scoring spelling correlate with reading and spelling development (Clemens et al., 2014; Treiman, Hulslander, et al., 2019); however, in the first year of school, nonbinary scoring may provide information about emerging spelling knowledge beyond dichotomous scores of words spelled correctly or incorrectly (Treiman, Kessler, & Caravolas, 2019). Therefore, we used a “correct sounds” scoring metric to measure children’s ability to “represent the phonological features of a word” (Clemens et al., 2014; Ritchey et al., 2010, p. 81). Children earned 1 point per correct letter sound (e.g., for “yit,” correct letter sounds were /y/, /ĭ/, and /t/, for a possible 3 points). Correct spellings included those provided by the test author and other plausible representations for the target sound in NZ English, with oversight provided by a speech–language therapist with expertise in early literacy.

New Zealand Word Identification Fluency (Schaughency et al., 2015)

NZWIF is comprised of high-frequency words introduced across the first 2 years of school in NZ. Administration of NZWIF is similar to NZWIF-Y1 used in progress monitoring. However, 50 words were randomly presented on one page with a smaller (18-point) font. Previous NZ research using NZWIF indicates strong concurrent correlations with instructional book level (.86) at the start of Year 2 and predictive relations to a variety of reading measures from beginning to end of Year 2 (.78–.88; Schaughency et al., 2015).

New Zealand School-Used Indicators of Reading Progress

School-used indicators of reading progress were obtained from children’s teachers via a brief questionnaire, timed to correspond with each child’s 1-year anniversary of starting school.

Book Level

The NZ Ministry of Education provides instructional book-level targets for children in the first 3 years of school, with Level 12 (Green) the target after 1 year of school (MoE, 2010a). Given the importance of book levels for educational decision-making in NZ, book levels have been used in NZ school-based research as a socially valid indicator of a child’s progress in reading connected text (Schaughency & Suggate, 2008). Research suggests that children’s book level after 1 year of school correlates robustly with pseudoword decoding at the end of Year 1 (T. A. Cameron et al., 2020) and predicts untimed word reading accuracy and oral reading fluency at the end of Year 3 (Schaughency et al., 2015).

Overall Teacher Judgment of Reading Progress (ReadingOTJ)

National Standards implemented during data collection required teachers to make overall teacher judgment (OTJs) about children’s reading progress in the NZ curriculum, to be informed by their observations and students’ assessed performance (MoE, 2009, 2020). Therefore, we asked teachers for overall judgments regarding children’s reading progress after 1 year of school. We used the National Standards achievement categories relative to expected progress (Ward & Thomas, 2016), scored as described by O’Hare et al. (2021), that is, “Well below” (−2), “Below” (−1), “At (on track)” (0), or “Above” (1). Evaluations of OTJs after 1 year of school indicate general, but not perfect, alignment between teachers’ judgments of literacy progress and children’s book levels (MoE, 2010b).

Exposure to Instruction

Figure 1 shows slight variation in exposure to school-based instruction for children by assessment session. To control for this variability, weeks-in-school was included as a potential covariate in analyses (T. A. Cameron et al., 2020; Schaughency & Suggate, 2008).

Procedure

Assessment times were arranged with teachers. Assessments were administered individually—except for Test of Phonological Awareness: Early Elementary “Letter Sounds” (TOPA)—within a quiet space at children’s schools during the school day. If a child missed a session due to school absence, they were assessed soon after they returned to school. One child, whose family moved overseas, was not assessed at 1-year follow-up; school-used data were collected for this child, with parental permission.

Statistical Analyses

Data were analyzed with IBM SPSS Statistics for Windows Version 26 Version 3.4 (IBM Corp, 2019) with PROCESS add-on (Hayes, 2018) and MPlus Base Program and Combination Add-On, Version 8.1 (MPlus; Muthén & Muthén, 2019). The Muthén and Muthén (2002, p. 608) Monte Carlo simulation for 5-equidistant time points, time covariate, and zero “missingness” indicated that our sample size exceeded the minimum requirement for modeling growth of 54 children. Preliminary analyses examined distributions of scores and bivariate relations between study measures.

To address RQ1, we adapted the dual-change LCS MPlus code provided by Serang et al. (2019) to produce a single modified variable (mLCS) for each progress monitoring measure (controlling for weeks-in-school). To do so, we combined Serang et al. (2019) code for initial status, and constant and proportional change, to produce the final single LCS (see Appendix, for mLCS MPlus code and diagrams for each progress monitoring measure). To evaluate validity of resultant mLCS, we correlated mLCS with growth trajectory classifications derived from GMM (see T. A. Cameron, Carroll, et al., 2022) and criterion measures.

To address RQ2, we employed structural equation modeling (SEM) path and ordinal logistic regression analyses. SEM path analyses were undertaken in MPlus (following Barbeau et al., 2019; Keith, 1999), using bias-corrected bootstrap confidence intervals (10,000 draws) and maximum likelihood estimation. Figure 2 shows our a priori hypothesized path model, with (a) early learning growth (mLCS) and 1-year assessments predicted from school-entry PELI®NZ; (b) 1-year assessments predicted from early learning growth (mLCS); and (c) no covariance between any parameters.

Path analysis assumptions were checked; all parameters were continuous and normally distributed. Multicollinearity between NZWIF-Y1mLCS and LSFmLCS (r = .85) would result in the dominant variable subsuming effects of the other (Barbeau et al., 2019; Clemens et al., 2019). Therefore, separate path analysis models were undertaken modeling contributions of (a) onset phoneme awareness (FSFmLCS) and letter-sound correspondence (LSFmLCS) with school-entry skills and (b) onset phoneme awareness (FSFmLCS) and word reading (NZWIF-Y1mLCS) with school-entry skills. Model fit statistics were evaluated using chi-square (χ2; nonsignificance indicates good model fit), comparative fit index (CFI ≥ .95), Tucker–Lewis index (TLI ≥ .95), standardized root-mean-square residual (SRMR ≤ .05), and root-mean-square error of approximation (RMSEA ≤ .05).

Finally, ordinal logistic regressions tested the influences of school-entry and progress monitoring indicators on teachers’ ordinal 1-year judgments of children’s reading progress. Consistent with path analyses, two separate models were used and all required assumptions were met.

Results
Descriptive Statistics

Progress monitoring on FSF and LSF exhibited normal distributions (±2 skew and kurtosis). NZWIF-Y1 scores exhibited floor effects during earlier assessment sessions, resolving to acceptable limits (skew < 3 and kurtosis < 10; Kline, 2011) by the last session (skew: from 3.83 to 1.95; kurtosis: from 19.12 to 4.32). Table 1 shows descriptive statistics. Boys’ and girls’ scores did not differ on measures.

(RQ1) Do Individual LCSs Resulting From LCS Models, Modified to Capture Both the Intercept and Growth Indices, Correspond With Children’s Empirically Derived Trajectory Class for Progress Monitoring Measures?

As shown in Table 1, newly created FSFmLCS and LSFmLCS variables exhibited normal distributions. NZWIF-Y1mLCS was positively skewed (skew: 2.46; kurtosis: 8.56); therefore, it was square-root transformed.

Higher mLCS reflects better performance, whereas for GMM classes, lower values represent better progress (see T. A. Cameron, Carroll, et al., 2022). Thus, moderate to strong negative correlations between mLCS and respective GMM latent growth trajectory classifications support concurrent validity (ps < .01, with 95% 10,000 bootstrapping CI: FSFmLCS: rs = −.73 [−.62, −.82]; LSFmLCS: rs = −.87 [−81, −.91]; NZWIF-Y1mLCS: rs = −.88 [−.82, −.92]). Further mLCS validity evidence is shown in Table 2, with moderate to strong positive associations between mLCS variables and school-entry and 1-year criterion measures (rs = .56–.88, ps < .001).

(RQ2) Do School-Entry and/or Early Literacy Skill Growth Predict Literacy Progress After 1 Year of School?

Predicting 1-Year Measures From School-Entry Skills and Developing Onset Phoneme Awareness and Letter-Sound Correspondence

Table 3 and Figure 3 show results of the path analysis model predicting reading outcomes after 1 year from PELI®NZ, FSFmLCS, and LSFmLCS. Model fit statistics indicate good fit, χ2: p = .475, CFI = 1.0, TLI = 1.0, SRMR = .013, and RMSEA = 0. PELI®NZ at school entry had both direct and mediated effects (via FSFmLCS and LSFmLCS) on literacy measures after 1 year of schooling. Specifically, overall PELI®NZ performance at school entry contributed to developing onset phoneme awareness and letter-sound correspondence. Developing letter-sound correspondence in the following 6 months of school, in turn, predicted children’s performance on all 1-year literacy outcomes. Developing onset phoneme awareness contributed to children’s pseudoword spelling and word reading fluency. The final model explained 47.9%–64.7% of the variance in 1-year literacy measures.

Predicting 1-Year Measures From School-Entry Skills and Developing Onset Phoneme Awareness and Word Reading

Table 4 and Figure 4 show results of the path analysis model predicting reading outcomes after 1 year from PELI®NZ, FSFmLCS, and NZWIF-Y1mLCS. Model fit statistics indicate good fit, χ2: p = .292, CFI = .999, TLI = .993, SRMR = .010, and RMSEA = .047. Although direct paths between PELI®NZ performance at school entry to BookLevel (p = .107) and TOPA (p = .066) were nonsignificant, their removal worsened model fit; therefore, they were left in the final model. Again, school-entry PELI®NZ had both direct and mediated effects (via FSFmLCS and NZWIF-Y1mLCS) on literacy measures after 1 year of school. Specifically, overall school-entry PELI®NZ contributed to developing onset phoneme awareness and word reading. Developing word reading in the following 6 months of school, in turn, predicted children’s performance on all 1-year literacy outcomes. Developing onset phoneme awareness contributed to children’s pseudoword spelling and instructional book level. The final model explained 52.6%–78.5% of the variance in 1-year literacy measures.

Predicting 1-Year Teacher Judgments From School-Entry and Progress Monitoring Measures

Ordinal logistic regression models explored contributions of school-entry and progress monitoring measures to teachers’ 1-year OTJ of children’s reading progress (ReadingOTJ; Table 5). For Model 1, higher PELI®NZ and increased LSFmLCS scores uniquely contributed to increased odds of teachers making more positive judgments (approximate R2: Nagelkerke = 61.3%; PELI®NZ: odds = 1.01, Wald χ2(1) = 7.31, p = .007; LSFmLCS: odds = 1.23; Wald χ2(1) = 20.70, p < .001). For Model 2, increased NZWIF-Y1mLCS uniquely contributed to increased odds of teachers making more positive judgments (approximate R2: Nagelkerke = 69.2%; NZWIF-Y1mLCS: odds = 13.67; Wald χ2(1) = 29.77, p < .001).

Discussion

This longitudinal study modeled contributions of school-entry skills and developing early literacy skills to children’s literacy progress after 1 year of school and literacy instruction in NZ. To describe skill development across the first 6 months, we created a novel mLCS metric, adapted from Serang et al. (2019). The advantage of our mLCS metric is integration of children’s initial performance and growth over time into one continuous score, allowing for contributions of individual differences in children’s progress to be empirically modeled. This novel mLCS approach showed good correspondence with GMM derived developmental trajectories (T. A. Cameron, Carroll, et al., 2022) and provided for investigation of children’s early literacy progress in multivariate analyses.

Overall, multivariate analyses suggested skills assessed at school entry and skill development over the first 6 months of school contributed to children’s 1-year performance and teachers’ ratings of literacy progress. Path and ordinal regression analyses revealed consistencies, and nuanced differences, in contributions as a function of predictors and outcomes included in models. Children’s overall performance at school entry contributed directly and indirectly to 1-year progress in path models, suggesting that oral language and emergent literacy skills at school entry contribute to literacy progress in the first year of school. In the first path model, PELI®NZ contributed to developing onset phoneme awareness (FSFmLCS) and letter-sound correspondence (LSFmLCS). FSFmLCS, in turn, contributed to pseudoword spelling and word reading at year-end, whereas PELI®NZ and LSFmLCS contributed unique variance to all 1-year outcomes. For the second path model, PELI®NZ again contributed to skill acquisition for both progress monitoring measures—onset phoneme awareness (FSFmLCS) and word reading (NZWIF-Y1mLCS). FSFmLCS, in turn, again contributed to pseudoword spelling and reading reflected in instructional book level, whereas NZWIF-Y1mLCS contributed to all 1-year outcomes. Moreover, PELI®NZ contributed unique variance to some 1-year outcomes (WIAT-IIIA&NZ and NZWIF).

Ordinal regressions predicting teachers’ overall judgments (OTJ) of literacy progress add to path analytic results. In the model including letter-sound awareness, PELI®NZ and LSFmLCS contributed unique variance to OTJ ratings. However, for the model including word reading, only NZWIF-Y1mLCS contributed unique variance to OTJ.

Not surprisingly, given the role of instructional book level as a marker of literacy progress in NZ (MoE, 2010a), teachers’ ratings were in line with children’s book levels: Children judged to be “At (on track)” (book level M = 13.9) or “Above” (M = 18.44) expected progress surpassed the NZ book level 12 target after 1 year of school, whereas children judged to be “Below” (M = 7.67) or “Well below” (M = 2.11) were achieving around or below the book level 6 target for 6 months of schooling (MoE, 2010a). Such findings are consistent with previous evaluations of the alignment of OTJs and children’s book level at age 6 (MoE, 2010b). In the first year of school, teachers commonly use observations of children’s oral reading (e.g., running records) to determine instructional book level (T. A. Cameron et al., 2019). These opportunities to observe children’s reading of high-frequency words in instructional texts likely contribute to the specific role of NZWIF-Y1mLCS in predicting teachers’ judgments.

In sum, results support contributions of school-entry oral language and early literacy skills to reading acquisition (Dickinson et al., 2019; Petscher et al., 2020). Moreover, findings for progress monitoring variables add to the evidence base on the importance of skill development early in the first year of school (e.g., Clemens et al., 2020). Progress-monitoring findings are consistent with the proposition that word reading develops concomitantly with developing phoneme and letter-sound correspondence (e.g., Ehri, 2014, see T. A. Cameron, Carroll, et al., 2022). Findings for LSFmLCS and NZWIF-Y1mLCS in predicting year-end outcomes, in particular, are consistent with, and extend, research during U.S. kindergarten (Clemens et al., 2020), pointing to potential roles for these variables as markers of literacy development early in the first year of school (Clemens et al., 2018). FSFmLCS contributed unique variance to pseudoword spelling and reading measures in path analyses, consistent with theorized roles of phoneme awareness in orthographic mapping in literacy learning (Ehri, 2014). These FSFmLCS results are potentially important given predictive relations between spelling at end of U.S. kindergarten and later word reading (Treiman, Hulslander, et al., 2019).

Implications

This research has implications for research and practice in beginning reading, indicating the value added of following developing skills in the context of early instruction (Speece, 2005). Our mLCSs provide a potential metric for educational and developmental researchers to employ when considering intraindividual development over a particular educational window.

For practice, findings indicate potential roles for school-entry screening, combined with progress monitoring of developing skills in the first 6 months of school, for the early identification of need for additional supports in beginning literacy instruction (T. A. Cameron et al., 2020). Results align with prioritizing screening and early identification of learning needs in reading in NZ (MoE, 2019). Our findings add to research on PELI®NZ at school entry (Cameron, Carroll, & Schaughency, 2022) and progress monitoring with FSF, LSF, and NZWIF-Y1 over the first 6 months (T. A. Cameron, Carroll, et al., 2022), supporting predictive validity to literacy progress after 1 year of school. The mean word reading fluency gap between children with mid/upper quartile versus bottom quartile PELI®NZ performance doubled from midyear (M = 10.5 words) to year-end (M = 21.6). Such results illustrate the prediction from school-entry screening but also signpost the dynamic nature of development. Differentiated early literacy instruction may be indicated to offset this gap in beginning reading.

To understand skill development over time and respond to children’s educational needs, educators need tools that reflect growth (Clemens et al., 2018; Fuchs, 2004). Literacy progress in the first year of school in NZ is currently monitored via instructional book level (T. A. Cameron et al., 2019). Tools such as NZWIF-Y1 may provide teachers with a sensitive and efficient indicator of whether children are on track toward progressing in their book level.

Given potential floor effects with word reading measures, sublexical indicators, such as LSF, may be important to include for students obtaining low scores on word reading measures (Clemens et al., 2018). Children displaying limited progress in FSF and LSF may be particularly at risk for difficulties with literacy acquisition (T. A. Cameron et al., 2020), signaling the need for further assessment and appropriate early instructional support.

Strengths, Limitations, and Future Work

Our study specifically focused on contributions of early skill trajectories to literacy progress after 1 year of school. We built on research from the science of reading and employing methods to tap into the dynamic nature of skills in development, while considering the instructional context of our work (see Fien et al., 2021). Future work needs to test whether predictions from school-entry skills and early skill growth hold true for reading achievement, and reading comprehension, in later years. Future work could evaluate the efficacy of our proposed screening approach for earlier identification of, and provision of instructional intervention for, literacy difficulties. Given the importance of oral language as well as word reading for children’s reading comprehension (Hjetland et al., 2020), future work should also focus on developing effective yet efficient means of monitoring children’s developing oral language competencies (Silberglitt et al., 2016; Silverman et al., 2020). Ultimately, work is needed on how to support teachers in linking screening and progress monitoring results to instructional supports for oral language (Silverman et al., 2020) and decoding-related skills (Petscher et al., 2020) to individualize student instruction (Hindman et al., 2020).

Our study was limited to English-medium instruction in one instructional context. Future work is needed with a larger number of students, classrooms, and schools to directly examine school and instructional influences on findings, and whether predictors vary for students from diverse socioeconomic and cultural backgrounds. Moreover, future work should also consider other linguistic contexts as specific findings may vary across languages (Zugarramurdi et al., 2022).

Conclusion

Our results indicate contributions of school-entry skills and early skill trajectories to beginning reading. Findings support further research examining feasibility and utility of tools used here for assessing and measuring children’s progress as additions to researchers’ and teachers’ toolkits during beginning reading instruction. School psychologists can contribute to the development of systems and resources to support practitioners’ use of data generated from early progress monitoring in educational decision-making to meet students’ learning needs in beginning instruction.

Footnotes

1  National Standards were discontinued at the end of 2017 (Hipkins, 2017).

2  Parameter bias (0.1%, 9.6%) and standard error bias (1.8%, 4.4%) below 10% for all model parameters; standard error bias below 5% for parameter for power assessed; coverage between .91 and .98 (.92, .94); power over .80 (.97).

References

Anthony, J. L., & Francis, D. J. (2005). Development of phonological awareness. Current Directions in Psychological Science, 14(5), 255–259. 10.1111/j.0963-7214.2005.00376.x

Barbeau, K., Boileau, K., Sarr, F., & Smith, K. (2019). Path analysis in Mplus: A tutorial using a conceptual model of psychological and behavioral antecedents of bulimic symptoms in young adults. The Quantitative Methods for Psychology, 15(1), 38–53. 10.20982/tqmp.15.1.p038

Burns, M. K., Maki, K. E., Helman, L., McComas, J. J., & Young, H. (2018). Contributions of the components of phonemic awareness to letter-sound knowledge with kindergarten students in high-poverty urban elementary schools. Reading & Writing Quarterly, 34(5), 409–418. 10.1080/10573569.2018.1468835

Cameron, C. E., Grimm, K. J., Steele, J. S., Castro-Schilo, L., & Grissmer, D. W. (2015). Nonlinear Gompertz Curve Models of achievement gaps in mathematics and reading. Journal of Educational Psychology, 107(3), 789–804. 10.1037/edu0000009

Cameron, T. A., Carroll, J. L. D., & Schaughency, E. (2022). Concurrent validity of the Preschool Early Literacy Indicators (PELI) with a New Zealand sample of 5-year-olds entering primary school. International Journal of School & Educational Psychology, 10(2), 208–219. 10.1080/21683603.2020.1805382

Cameron, T. A., Carroll, J. L. D., Taumoepeau, M., & Schaughency, E. (2019). How do New Zealand teachers assess children’s oral language and literacy skills at school entry?New Zealand Journal of Educational Studies, 54(1), 69–97. 10.1007/s40841-019-00133-4

Cameron, T. A., Carroll, J. L. D., Taumoepeau, M., & Schaughency, E. (2022). Patterns of early literacy and word reading skill development across the first six months of school and reading instruction [Manuscript under review].

Cameron, T. A., Taumoepeau, M., Clarke, K., McDowall, P., & Schaughency, E. (2020). Describing patterns of early literacy skill development in the first year of school and reading instruction in a New Zealand sample. The School Psychologist, 35(4), 243–254. 10.1037/spq0000370

Castles, A., Rastle, K., & Nation, K. (2018). Ending the reading wars: Reading acquisition from novice to expert. Psychological Science in the Public Interest, 19(1), 5–51. 10.1177/1529100618772271

Clayton, F. J., West, G., Sears, C., Hulme, C., & Lervåg, A. (2020). A longitudinal study of early reading development: Letter-sound knowledge, phoneme awareness and ran, but not letter-sound integration, predict variations in reading development. Scientific Studies of Reading, 24(2), 91–107. 10.1080/10888438.2019.1622546

Clemens, N. H., Hsiao, Y. Y., Simmons, L. E., Kwok, O. M., Greene, E. A., Soohoo, M. M., Henri, M. A., Luo, W., Prickett, C., Rivas, B., & Otaiba, S. A. (2019). Predictive validity of kindergarten progress monitoring measures across the school year: Application of Dominance Analysis. Assessment for Effective Intervention, 44(4), 241–255. 10.1177/1534508418775805

Clemens, N. H., Lee, K., Henri, M., Simmons, L. E., Kwok, O. M., & Al Otaiba, S. (2020). Growth on sublexical fluency progress monitoring measures in early kindergarten and relations to word reading acquisition. Journal of School Psychology, 79, 43–62. 10.1016/j.jsp.2020.01.003

Clemens, N. H., Oslund, E. L., Simmons, L. E., & Simmons, D. (2014). Assessing spelling in kindergarten: Further comparison of scoring metrics and their relation to reading skills. Journal of School Psychology, 52(1), 49–61. 10.1016/j.jsp.2013.12.005

Clemens, N. H., Soohoo, M. M., Wiley, C. P., Hsiao, Y. Y., Estrella, I., Allee-Smith, P. J., & Yoon, M. (2018). Advancing Stage 2 research on measures for monitoring kindergarten reading progress. Journal of Learning Disabilities, 51(1), 85–104. 10.1177/0022219416688171

Crooks, T. J., Smith, J. K., & Flockton, L. C. (2010). Mathematics assessment results 2009 (National Education Monitoring Report 52). Educational Assessment Research Unit, University of Otago. https://nemp.otago.ac.nz/PDFs/maths_09/NEMP_Mathematics_2009.pdf

Dickinson, D. K., Nesbitt, K. T., & Hofer, K. G. (2019). Effects of language on initial reading: Direct and indirect associations between code and language from preschool to first grade. Early Childhood Research Quarterly, 49, 122–137. 10.1016/j.ecresq.2019.04.005

Ehri, L. C. (2014). Orthographic mapping in the acquisition of sight word reading, spelling memory, and vocabulary learning. Scientific Studies of Reading, 18(1), 5–21. 10.1080/10888438.2013.819356

Fien, H., Chard, D. J., & Baker, S. K. (2021). Can the evidence revolution and multi-tiered systems of support improve education equity and reading achievement?Reading Research Quarterly, 56(S1), S105–S118. 10.1002/rrq.391

Fuchs, L. S. (2004). The past, present, and future of curriculum-based measurement research. School Psychology Review, 33(2), 188–193. 10.1080/02796015.2004.12086241

Fuchs, L. S., Fuchs, D., & Compton, D. L. (2004). Monitoring early reading development in first grade: Word identification fluency versus nonsense word fluency. Exceptional Children, 71(1), 7–21. 10.1177/001440290407100101

Good, R. H., & Kaminski, R. A. (2011). Dynamic indicators of basic early literacy skills: DIBELS next assessment manual. Dynamic Measurement Group. https://www.mccracken.kyschools.us/Downloads/DIBELS%20Next%20Assessment%20Manual.pdf

Gough, P. & Tunmer, W. (1986). Decoding, reading, and reading disability. Remedial and Special Education, 7(1), 6–10. 10.1177/074193258600700104

Grimm, K. J., Davoudzadeh, P., & Ram, N. (2017). Developmental methodology: IV. Developments in the analysis of longitudinal data. Monographs of the Society for Research in Child Development, 82(2), 46–66. 10.1111/mono.12298

Grimm, K. J., Mazza, G. L., & Mazzocco, M. M. M. (2016). Advances in methods for assessing longitudinal change. Educational Psychologist, 51(3–4), 342–353. 10.1080/00461520.2016.1208569

Hayes, A. F. (2018). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (2nd ed.). Guilford Press.

Hindman, A. H., Morrison, F. J., Connor, C. M., & Connor, J. A. (2020). Bridging the science of reading to preservice elementary teachers: Tools that bridge research and Practice. Reading Research Quarterly, 55(S1), S197–S206. 10.1002/rrq.345

Hipkins, C. (2017). National standards ended. https://www.beehive.govt.nz/release/national-standards-ended

Hjetland, H. N., Brinchmann, E. I., Scherer, R., Hulme, C., & Melby-Lervag, M. (2020). Preschool pathways to reading comprehension: A systematic meta-analytic review. Educational Research Review, 30, Article 100323. 10.1016/j.edurev.2020.100323

IBM Corp. (2019). SPSS statistics for Windows (Version 22.0.0.1) [Computer software]. https://www.ibm.com/analytics/spss-statistics-software

Joshua, N., Wilson, C., & Bendrups, N. (2016). Wechsler Individual Achievement Test, third edition: Australian and New Zealand standardised (WIAT-III A&NZ). Pearson.

Kaminski, R. A., Abbott, M., Aguayo, K., & Good, R. H. (2018). PELI assessment manual. Dynamic Measurement Group. https://acadiencelearning.org/acadience-reading/prek-peli/

Kaplan, D., & Walpole, S. (2005). A stage-sequential model of reading transitions: Evidence from the early childhood longitudinal study. Journal of Educational Psychology, 97(4), 551–563. 10.1037/0022-0663.97.4.551

Keith, T. Z. (1999). Structural equation modeling in school psychology. In C. R.Reynolds & T. B.Gutkin (Eds.), The handbook of school psychology (3rd ed., pp. 78–107). Wiley.

Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). Guilford Press.

Lonigan, C. J., & Burgess, S. R. (2017). Dimensionality of reading skills with elementary-school-age children. Scientific Studies of Reading, 21(3), 239–253. 10.1080/10888438.2017.1285918

Mavilidi, M. F., Marsh, H. W., Xu, K. M., Parker, P. D., Jansen, P. W., & Paas, F. (2022). Relative age effects on academic achievement in the first ten years of formal schooling: A nationally representative longitudinal prospective study. Journal of Educational Psychology, 114(2), 308–325. 10.1037/edu0000681

McNaughton, S. (2020). The literacy landscape in Aotearoa New Zealand: What we know, what needs fixing, what we should prioritise: Report. Office of the Prime Minister’s Chief Science Advisor. https://dpmc.govt.nz/sites/default/files/2022-04/PMCSA-20-15_The-Literacy-Landscape-in-Aotearoa-New-Zealand-Full-report-final.pdf

Ministry of Education. (2009). The New Zealand curriculum reading and writing standards for years 1–8. Learning Media. https://nzcurriculum.tki.org.nz/Archives/Assessment/Reading-and-writing-standards

Ministry of Education. (2010a). The literacy learning progressions: Meeting the reading and writing demands of the curriculum. Learning Media. https://literacyprogressions.tki.org.nz/Media/Files/Literacy-Progressions-PDF

Ministry of Education. (2010b). The observation survey and the national standards. https://assessment.tki.org.nz/Media/Files/TUV-files/The-Observation-Survey-and-the-National-Standards

Ministry of Education. (2017). Te whāriki: He whāriki mātauranga mō ngā mokopuna o Aotearoa: Early childhood curriculum. https://www.education.govt.nz/assets/Documents/Early-Childhood/ELS-Te-Whariki-Early-Childhood-Curriculum-ENG-Web.pdf

Ministry of Education. (2019). Learning support action plan: 2019–2025. https://conversation.education.govt.nz/assets/DLSAP/Learning-Support-Action-Plan-2019-to-2025-English-V2.pdf

Ministry of Education. (2020). Overall teacher judgement. http://assessment.tki.org.nz/Overall-teacher-judgment

Ministry of Education. (2021, October). Early learning participation. Education Indicator: Education and Learning Outcomes. https://www.educationcounts.govt.nz/statistics/participation

Ministry of Education. (2022, May). School deciles. https://www.education.govt.nz/school/funding-and-financials/resourcing/operational-funding/school-decile-ratings/

Morton, S. M. B., Grant, C. C., Berry, S. D., Walker, C. G., Corkin, M., Ly, K., de Castro, T. G., Atatoa Carr, P. E., Bandara, D. K., Mohal, J., Bird, A. L., Underwood, L., & Fa’alili-Fidow, J. (2017). Growing up in New Zealand: A longitudinal study of New Zealand children and their families. Now we are four: Describing the preschool years. Growing Up in New Zealand. https://cdn.auckland.ac.nz/assets/growingup/research-findings-impact/GUiNZ_Now%20we%20are%20four%20report.pdf

Muthén, L. K., & Muthén, B. O. (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 9(4), 599–620. 10.1207/S15328007SEM0904_8

Muthén, L. K., & Muthén, B. O. (2019). MPlus base program and combination add-on (Version 7.31) [Computer software]. https://www.statmodel.com/programs.shtml

National Early Literacy Panel. (2008). Developing early literacy: Report of the National Early Literacy Panel. https://lincs.ed.gov/publications/pdf/NELPReport09.pdf

Nese, J. F., Kamata, A., & Tindal, G. (2017). A two-step sampling weight approach to growth mixture modeling for emergent and developing skills with distributional changes over time. Journal of School Psychology, 61, 55–74. 10.1016/j.jsp.2016.12.001

New Zealand Immigration. (2020). The school system. https://www.newzealandnow.govt.nz/living-in-nz/education/school-system

O’Hare, K., White, N., Harding, R., Galland, B., Sellbom, M., Shine, B., & Schaughency, E. (2021). Sluggish cognitive tempo and daytime sleepiness mediate relationships between sleep and academic performance. Journal of Developmental and Behavioral Pediatrics, 42(8), 637–647. 10.1097/DBP.0000000000000948

Organisation for Economic Co-operation and Development. (2019). PISA 2018 results (Volume 1): What students know and can do. 10.1787/5f07c754-en

Oslund, E. L., Hagan-Burke, S., Simmons, D. C., Clemens, N. H., Simmons, L. E., Taylor, A. B., Kwok, O. M., & Coyne, M. D. (2017). Predictive validity of curriculum-embedded measures on outcomes of kindergarteners identified as at risk for reading difficulty. Journal of Learning Disabilities, 50(6), 712–723. 10.1177/0022219416664866

Petscher, Y., Cabell, S. Q., Catts, H. W., Compton, D. L., Foorman, B. R., Hart, S. A., Lonigan, C. J., Phillips, B. M., Schatschneider, C., Steacy, L. M., Terry, N. P., & Wagner, R. K. (2020). How the science of reading informs 21st-century education. Reading Research Quarterly, 55(S1), S267–S282. 10.1002/rrq.352

Petscher, Y., Koon, S., & Herrera, S. (2016). Using latent change score analysis to model co-development in fluency skills. In K. D.Cummings & Y.Petscher (Eds.), The fluency construct: Curriculum-based measurement concepts and applications (pp. 333–364). Springer. 10.1007/978-1-4939-2803-3_12

Pretti-Frontczak, K. L., Grisham-Brown, J. L., & Hallam, R. A. (2014). Introduction to the topical issue on measurement and mismeasurement. Topics in Early Childhood Special Education, 34(2), 69–70. 10.1177/0271121414532139

Ritchey, K. D., Coker, D. L., Jr., & McCraw, S. B. (2010). A comparison of metrics for scoring beginning spelling. Assessment for Effective Intervention, 35(2), 78–88. 10.1177/1534508409336087

Ritchey, K. D., & Speece, D. L. (2006). From letter names to word reading: The nascent role of sublexical fluency. Contemporary Educational Psychology, 31(3), 301–327. 10.1016/j.cedpsych.2005.10.001

Sáez, L., Nese, J. F., Alonzo, J., & Tindal, G. (2016). Individual differences in kindergarten through Grade 2 fluency relations. Learning and Individual Differences, 49, 100–109. 10.1016/j.lindif.2016.05.020

Schaughency, E., McLennan, K. M., & McDowall, P. S. (2015). Development and preliminary evaluation of an adaption of word identification fluency for beginning readers in New Zealand. Assessment for Effective Intervention, 40(2), 67–80. 10.1177/1534508414544541

Schaughency, E., & Suggate, S. (2008). Measuring basic early literacy skills amongst year 1 students in New Zealand. New Zealand Journal of Educational Studies, 43(1), 85–106. https://www.researchgate.net/publication/289689655_Measuring_basic_early_literacy_skills_amongst_year_1_students_in_New_Zealand

Schaughency, E., Suggate, S., & Reese, E. (2017). Links between early oral narrative and decoding skills and later reading in a New Zealand sample. Australian Journal of Learning Difficulties, 22(2), 109–132. 10.1080/19404158.2017.1399914

Schluter, P. J., Audas, R., Kokaua, J., McNeill, B., Taylor, B., Milne, B., & Gillon, G. (2020). The efficacy of preschool developmental indicators as a screen for early primary school-based literacy interventions. Child Development, 91(1), e59–e76. 10.1111/cdev.13145

Serang, S., Grimm, K. J., & Zhang, Z. (2019). On the correspondence between the latent growth curve and latent change score models. Structural Equation Modeling: a Multidisciplinary Journal, 26(4), 623–635. 10.1080/10705511.2018.1533835

Shanahan, T., & Lonigan, C. J. (2010). The national early literacy panel: A summary of the process and the report. Educational Researcher, 39(4), 279–285. 10.3102/0013189X10369172

Shinn, M. M., & Shinn, M. R. (2002). Administration and scoring of early literacy measures for use with AIMSweb. Edformation.

Silberglitt, B., Parker, D., & Muyskens, P. (2016). Assessment: Periodic assessment to monitor progress. In S. R.Jimerson, M. K.Burns, & A. M.VanDerHeyden (Eds.), Handbook of response to intervention: The science and practice of multi-tiered systems of support (pp. 271–291). Springer. 10.1007/978-1-4899-7568-3_16

Silverman, R. D., Johnson, E., Keeane, K., & Khanna, S. (2020). Beyond decoding: A meta-analysis of the effects of language comprehension interventions on K-5 students’ language and literacy outcomes. Reading Research Quarterly, 55(S1), S207–S233. 10.1002/rrq.346

Speece, D. L. (2005). Hitting the moving target known as reading development: Some thoughts on screening children for secondary interventions. Journal of Learning Disabilities, 38(6), 487–493. 10.1177/00222194050380060301

Statistics New Zealand. (2005). Understanding and working with ethnicity data: A technical paper. 10.13140/RG.2.1.1347.0804

Tindal, G., Irvin, P. S., Nese, J. F. T., & Slater, S. (2015). Skills for children entering kindergarten. Educational Assessment, 20(4), 297–319. 10.1080/10627197.2015.1093929

Torgesen, J. K., & Bryant, B. R. (2004). Test of phonological awareness—Second edition: PLUS (TOPA-2+). Pro-Ed.

Treiman, R., Hulslander, J., Olson, R. K., Willcutt, E. G., Byrne, B., & Kessler, B. (2019). The unique role of early spelling in the prediction of later literacy performance. Scientific Studies of Reading, 23(5), 437–444. 10.1080/10888438.2019.1573242

Treiman, R., Kessler, B., & Caravolas, M. (2019). What methods of scoring young children’s spelling best predict later spelling performance?Journal of Research in Reading, 42(1), 80–96. 10.1111/1467-9817.12241

van Bergen, E., Vasalampi, K., & Torppa, M. (2021). How are practice and performance related? Development of reading from age 5 to 15. Reading Research Quarterly, 56(3), 415–434. 10.1002/rrq.309

Ward, J., & Thomas, G. (2016). National standards: School sample monitoring and evaluation project, 2010–2014. Report to the Ministry of Education. Ministry of Education. https://core.ac.uk/download/pdf/51343795.pdf

Zugarramurdi, C., Fernández, L., Lallier, M., Valle-Lisboa, J. C., & Carreiras, M. (2022). Mind the orthography: Revisiting the contribution of prereading phonological awareness to reading acquisition. Developmental Psychology, 58(6), 1003–1016. 10.1037/dev0001341

APPENDIX APPENDIX A: Modified “Dual Change Latent Change Score” MPlus CodeAdapted from Serang et al. (2019)

TITLE: Dual Change LCS Model;DATA: FILE IS “myData.csv”;VARIABLE: NAMES ARE  PM1-PM5  ControlVar_Time; USEVAR =  PM1-PM5  ControlVar_Time; MISSING = all (9999);ANALYSIS:COVERAGE = 0;MODEL = NOCOVARIANCES;MODEL:!Observed variables[PM1-PM5@0];PM1-PM5 (s2_e);!Latent True ScoreslPM1 BY PM1@1;lPM2 BY PM2@1;lPM3 BY PM3@1;lPM4 BY PM4@1;lPM5 BY PM5@1;!Altered in final single latent change score!See #COMBINED#![Initial Status (IPM1)]!c1 BY lPM1@1;![c1] (mu_C1);!c1 (s2_c1);[lPM1-lPM5@0];lPM1-lPM5@0;

!AutoregressionslPM2 ON lPM1@1;lPM3 ON lPM2@1;lPM4 ON lPM3@1;lPM5 ON lPM4@1;!Latent Change ScoresdPM2 BY lPM2@1;dPM3 BY lPM3@1;dPM4 BY lPM4@1;dPM5 BY lPM5@1;[dPM2-dPM5@0];dPM2-dPM5@0;!Constant Change Component!Altered in final single latent change score!See #COMBINED#![Change components (dPM2-dPM5)]!c2 BY dPM2-dPM5@1;![c2] (mu_C2);!c2 (s2_c2);!c1 WITH c2 (s_c1c2);!Proportional Change ComponentdPM2 ON lPM1 (C3);dPM3 ON lPM2 (C3);dPM4 ON lPM3 (C3);dPM5 ON lPM4 (C3);! #COMBINED#!Combines initial status (lPM1) and constant and proportional change components (dPM2-dPM5) in final single latent change score.LATPM BY lPM1@1 dPM2-dPM5@1;[LATPM] (mu_LATPM);LATPM (s_LATPM);LATPM ON ControlVar_Time;

Where PMx refers to progress monitoring score for time x. Italicized blocks of code depict code as originally developed by Serang et al. (2019) that were adapted and incorporated into the final block of code to yield the final single latent change scores. Adapted from “On the Correspondence Between the Latent Growth Curve and Latent Change Score Models,” by S. Serang, K. J. Grimm, and Z. Zhang, 2019, Structural Equation Modeling, 26(4), 623–635 (https://doi.org/10.1080/10705511.2018.1533835). Copyright 2019 by the Taylor & Francis Group. Reprinted with permission.

First Sound Fluency78.06 latfsf 1.00 lfsfl 1.0 fsfl 49.992.15 1.00 −.63 1.00pwis 1.00 dfsf2 1.00 lfsf2 1.00 fsf2 49.991.00 −.63 1.001.00 dfsf3 1.00 lfsf3 1.00 fsf3 49.99−.63 1.00dfsf4 1.00 lfsf4 1.00 fsf4 49.99−.63 1.00dfsf5 1.00 lfsf5 1.00 fsf5 49.99Letter Sound Fluency61.23 latlsf 1.00 llsfl 1.0 lsfl 34.401.84 1.00 −.39 1.00pwis 1.00 dlsf2 1.00 llsf2 1.00 lsf2 34.401.00 −.39 1.001.00 dlsf3 1.00 llsf3 1.00 lsf3 34.40−.39 1.00dlsf4 1.00 llsf4 1.00 lsf4 34.40−.39 1.00dlsf5 1.00 llsf5 1.00 lsf5 34.40New Zealand Word Identification Fluency: New Entrant Children (NZWIFY0/1)17.04 latwif 1.00 lwifl 1.0 wifl 10.021.31 1.00 −.29 1.00pwis 1.00 dwif2 1.00 lwif2 1.00 wif2 10.021.00 −.29 1.001.00 dwif3 1.00 lwif3 1.00 wif3 10.02−.29 1.00dwif4 1.00 lwif4 1.00 wif4 10.02−.29 1.00dwif5 1.00 lwif5 1.00 wif5 10.02

Note. Latfsf/latlsf/latwif = mLCS for each variable; pwis = progress monitoring weeks-in-school covariate; fsf1-5/lsf1-5/wif1-5 = progress monitoring scores for each measure; lfsf1-5/llsf1-5/lwif1-5 = latent true scores for each measure; dfsf1-5/dlsf1-5/dwif1-5 = proportional and constant change scores.

Submitted: February 1, 2022 Revised: January 16, 2023 Accepted: February 14, 2023