Abstract
The development of children’s emergent literacy skills in kindergarten must be supported, given the significance of these skills for their future educational success. However, without further training, it may be challenging for educators to provide this support effectively. The current study aims to evaluate the impact of a relatively brief professional development program with several components on the number of emergent literacy learning opportunities offered to children. A multiple baseline design was employed with three educators of inclusive kindergarten classes for 4-year-olds. Data collection involved between 6 and 12 observation time points during the baseline phase and 18 time points for the intervention phase. Visual and statistical analyses indicated that professional development led to an increase in learning opportunities presented to children. Consequently, even if the professional development program provided was brief, it represents a promising option to support emergent literacy skills in kindergarten.
Before being formally taught to read and write, young children acquire emergent literacy skills through daily life processes encountered in physical and social settings (Giasson, 2011; Rohde, 2015; Saracho, 2017). Rohde (2015) defined emergent literacy through four interrelated components: oral language development (vocabulary), phonological awareness (manipulating and distinguishing sounds), print awareness (alphabet knowledge and print concepts), and writing (drawing, writing letters). Support for the development of these skills during preschool years is important because they are linked to literacy and educational success (Carpentier et al., 2022; Duncan et al., 2007; Purpura et al., 2011). In this regard, a systematic review by Chambers et al. (2016), which focused on studies conducted in preschool environments, underlined the effectiveness of comprehensive educational programs that combined support for child-initiated activities as well as adult-planned play activities targeting specific literacy skills, such as phonological awareness. Furthermore, many authors (Dubois et al., 2019; Piasta, 2016; Terrell & Watson, 2018) have stressed the importance of combining direct and indirect interventions to foster those skills. “Indirect interventions” refers to modification of a child’s physical environment to include writing tools and a variety of reading and print materials (Dynia et al., 2018; Guo et al., 2013). Such literacy-rich environments raise children’s interest and enhance engagement in activities targeting emergent reading and writing skills (Baroody & Diamond, 2016).
Physical environments alone are not sufficient to support the development of emergent literacy skills (Dubois et al., 2019; Piasta, 2016). Indeed, for opportunities to become real learning situations, educators must consciously integrate direct interventions (e.g., scaffolding, modeling, feedback, rephrasing) into ongoing interactions with a child (Dubois et al., 2019; Johnson et al., 2015; Piasta, 2016). Furthermore, in an inclusive education context, these direct and indirect intervention practices must be adapted to provide support that meets the diverse needs of children (United Nations Educational, Scientific and Cultural Organization [UNESCO], 2021). Following the natural proportion principle proposed by van Rhijn et al. (2019), a classroom is deemed an inclusive setting if the number of children with special needs reflects the prevalence in the target population.
However, results of recent studies conducted in Canada (Canadian Children’s Literacy Foundation & Canadian Child Care Federation, 2022; Drainville & Charron, 2021) and in the United States (Guo et al., 2013; Zhang et al., 2015) revealed gaps in writing environments as well as in literacy-supportive interactions between early-childhood educators (ECEs) and children.
Some authors (Elek et al., 2022; Jaskolski & Moyle, 2023; Lemire et al., 2021) have underlined the possibility for ECEs to improve their emergent literacy-promoting practices through professional development (PD). Moreover, results of Brunsek et al.’s (2020) meta-analysis showed positive associations between ECEs’ PD and improvements in young children’s literacy skills regarding sound and print awareness, letter knowledge, expressive vocabulary, and writing. PD extends beyond a theoretical presentation of concepts, as it targets the development of professional competencies that learners can implement in practice (Child Care Aware of America & National Association for the Education of Young Children, 2023; Dunst, 2015). PD that includes coaching—that is, a relationship between a more experienced person and someone who is learning and is supported while developing skills—has been significantly linked to expected outcomes (Brunsek et al., 2020; Markussen-Brown et al., 2017). Longer, more intense PD has also been associated with better results (Brunsek et al., 2020; Markussen-Brown et al., 2017), although multicomponent PD could be more beneficial than long and intense PD (Markussen-Brown et al., 2017). Furthermore, in a meta-analysis by Markussen-Brown et al. (2017), PD was shown to have more effect on structural quality (e.g., increased literacy material), which is related to indirect intervention, than on process quality (ECE-child interactions) or direct intervention.
The current study aims to assess the impact of relatively short PD of average intensity composed of several components. More specifically, the goal of the study was to implement activity-based intervention (ABI; Johnson et al., 2015) in kindergarten to support the development of children’s emergent literacy skills. ABI is a well-detailed and structured naturalistic educational approach combining both direct and indirect interventions to support the development of children at risk of learning difficulties, with or without disabilities. ABI integrates and links evaluation, choice of goals and objectives, intervention, and follow-up of a child’s progress; it also includes environmental adaptations to arouse interest and encourage the emergence of certain skills.
Central to ABI is the integration of numerous and diverse learning opportunities (LOs) proposed in an intentional and systematic way. LOs are adjusted to the children’s developmental needs and embedded through three activity contexts: (a) free play, (b) routines, and (c) planned activities. These LOs require interaction between ECEs and children to generate new knowledge or skills. Various authors (Gulboy et al., 2023; Rakap & Parlak-Rakap, 2011; Snyder, Rakap, et al., 2015) have put forward possibilities for using naturalistic approaches such as ABI to support the development of preacademic skills, including literacy.
The PD used in the current project is based on the work of Dunst and collaborators (Dunst et al., 2019; Dunst & Trivette, 2009) as well as on Practice-Based Coaching (Snyder et al., 2022; Snyder, Hemmeter, & Fox, 2015). This specific PD, carried out individually, consists of three stages: (a) introduction to new knowledge, (b) implementation and feedback, and (c) understanding of learning. During the first stage, ECEs are introduced to the material and knowledge associated with the practices to implement. The goal of this stage is for these individuals to commit to learning before new knowledge is transmitted to them. Hence, newly transmitted information is based on ECEs’ prior knowledge. In the second stage, learners apply the newly acquired knowledge through role-playing, exercises, and real-life interventions. As learners experiment, they are given feedback based on in situ observations during the implementation of the new intervention practices. In the third stage, learners self-assess their own knowledge and skill acquisition.
This article presents results from a broader project designed to support the implementation of ABI in inclusive kindergarten classes for 4-year-olds through PD for ECEs. The research question addressed in this article is the following: Does this relatively short PD with several components lead to an increased number of literacy LOs that involve interactions between ECEs and children?
Method
Design
Data are from a larger study on ABI implementation based on a core mixed-method design (Creswell & Plano Clark, 2018). This article’s research question was addressed using a single-case multiple baseline design. Single-case designs are characterized by the systematic use of repeated observation measures (Riley-Tillman et al., 2020). A single-case, multiple baseline design is recommended when an intervention is irreversible, as is the case in this project. In this type of single-case design, intervention (i.e., PD) delivery is staggered across time for each participant. The study meets What Works Clearinghouse (WWC) standards without reservations, as it meets the number of phases, number of data points per phase and concurrence requirements for this design type (WWC, 2022). There are two conditions (baseline phase and intervention phase) for three participants, with at least six observations per condition.
Participants
ECEs had to volunteer to participate as well as be in charge of an inclusive kindergarten class for French-speaking 4-year-olds in Québec (Canada). Three female Caucasian ECEs aged 33, 35, and 36 years old participated; all had a B.Ed. in Kindergarten and Elementary Education. Two ECEs had 11 to 16 years of experience, and one had 6 to 10 years. All participants had received training on emergent literacy (phonological awareness, shared book reading) in the past, but none had previously used or received training on ABI. Each ECE was responsible for an inclusive kindergarten class for 4-year-olds, for a total of 32 children (n = 10 [six girls and four boys], n = 12 [four girls and eight boys], and n = 10 [five girls and five boys]). The average age of the children was 56.56 months (SD = 3.56), and all spoke French at home. In each class, at least two children were identified as having special needs: they were either at risk of developmental delays (waiting for diagnostic assessment or receiving early intervention services such as speech therapy) or had a diagnosed disability (e.g., language disorder, hearing impairment). In Quebec, the prevalence of special needs among preschoolers is approximately 12% (Groleau & Aranibar Zeballos, 2022). Therefore, these classrooms are considered inclusive, based on the natural proportion principle (van Rhijn et al., 2019).
Researchers
The principal investigator, who has a PhD in psychoeducation (research orientation), provided training and coaching to the ECEs. She had 15 years of experience as a professional in early intervention and childhood education and 7 years of experience as a clinical supervisor. She and a PhD student in psychology (concentration in family studies), who is also an ECE, assessed the children’s literacy skills using the Literacy for 0- to 6-year-olds observation grid of the Programme ÉIS-3 Évaluation, intervention, suivi T.01 (ÉIS-3; Bricker & Johnson, 2023), the French adaptation of the Assessment, Evaluation, and Programming System for Infants and Children (AEPS-3; Bricker et al., 2021). These assessment results are used during PD, as described below. Then, two master’s students (one in psychoeducation and one in education) scored the recorded observation moments, without the involvement of the principal investigator. These students received training on video coding and practiced with training videos until they attained a combined inter-rater reliability of at least 80%. All students involved in the project were conducting study projects and/or had experience in the field of early childhood education and intervention.
Process, Data Collection and Tools
The project began in the fall of 2022 and ended in the summer of 2023. For both phases (baseline and intervention), 10-min observation moments were recorded at least twice for each activity context: (a) free play, (b) routine (snack), and (c) play planned by the adult. The recorded observation sessions took place in the mornings to minimize disruption to the children’s afternoon naps.
Observation Grid for LOs
Video recordings were used to complete the Observation grid for LOs. Each 10-min recording was divided into equal intervals of 15 s. A first 15-s period was used to collect data on LOs, including the number of LOs and their description. The next 15 s was not used, and so on, for the entire duration of the 10-min video. A LO refers to an intervention carried out by the ECE to encourage one or more children to acquire or practice literacy skills included in the literacy observation grid. For each LO, antecedents (verbatim or description of the intervention action, adaptation of the environment), response of the child(ren), reinforcement/consequence and targeted skill were documented in the literacy observation grid.
Literacy for 0- to 6-Year-Olds
The literacy observation grid was used to assess the children’s skills prior to the experimental phase (Bricker & Johnson, 2023). The grid included 36 skills (11 goals and 25 objectives) across five levels: (a) print awareness (e.g., participate in group reading), (b) phonological awareness (e.g., identify words that rhyme), (c) knowledge of the alphabet (e.g., can say the letters of his first name), (d) vocabulary, and understanding a story, and (e) writing. Data from this assessment were used by the ECEs in the training session to select literacy intervention/teaching objectives for the children in their class group. The preliminary version of this grid had shown good psychometric qualities in previous studies (Lemire et al., 2014, 2015). Inter-rater reliability results indicate excellent agreement (κ = 0.87) in the assessment of literacy skills (Lemire et al., 2015). Furthermore, according to the content validity study, the observation grid's items are judged to be mostly relevant, functional, and representative of the concept (Lemire et al., 2014).
Baseline Phase and Data Collection
For Phase 1, data collection (baseline) started at the same moment for each participant. The number of observation moments varied between 6 and 12, depending on the ECE, because PD (training) did not start at the same time for each ECE. The duration of the baseline phase had to be planned because while each ECE was receiving training, a replacement teacher had to be called in. The order in which the PD started was determined randomly.
Intervention Phase and Data Collection
The intervention consisted of individual PD, as described above. It included 1 day (6 hr) of training. Three to four weeks after this individual training, a first 1-hr coaching session was provided, followed by a second one 4 to 5 weeks later. Between these two coaching sessions, an on-site recorded observation period was conducted. Following the training and each coaching session, six 10-min observation moments were recorded in 3 to 4 days across 1 to 2 weeks, for a total of 18 recorded observation moments per participant. In all, video recording time (baseline and intervention phases) was 240 min for the first ECE receiving training, 280 min for the second one and 300 min for the third.
PD content dealt with the main components of ABI, structured around three stages (Dunst & Trivette, 2009): introduction, implementation and feedback, and understanding of learning. In the introduction stage, ECEs became acquainted with the children’s literacy assessments and completed a questionnaire consisting of five open-ended questions, enabling them to reflect on their literacy support practices. The first question targeted how each ECE situated the children’s literacy skills level and adjusted the provided LOs to that level. The other three questions documented the LOs provided in each targeted context, namely free play, routines, and planned activities (e.g., “Are literacy learning opportunities offered during routines? If so, how?”).
Then, each ECE received the same training: a theoretical presentation of the ABI’s main components (embedding LOs in the three target contexts), environment adaptations, direct and indirect intervention strategies, emergent literacy skills, enriched literacy environments, and planning of play activities. Videos were presented to illustrate the content of the presentation, and ECEs performed practical exercises to develop material corresponding to their group’s needs. First, ECEs chose intervention/teaching objectives based on the literacy developmental needs of their classes. To do this, they had access to all the children’s assessments previously carried out with the literacy observation grid. Next, based on their chosen objectives, each ECE was supported by the PD provider in creating a schedule of activities that embeds LOs in all three activity contexts (see Table 1 for an example). The ECEs implemented that schedule for 18 weeks. Finally, keeping in mind their groups’ literacy development needs, they designed an adult-planned play activity.
Schedule of Activities.
After training ended, during the “implementation and feedback” stage, the ECEs tested the activity schedule and planned play activities for 18 weeks. Using the literacy observation grid, the ECEs tracked the children’s progress through daily observations. Depending on the children’s progress, adjustments could be made to intervention/teaching objectives and new activities developed. Coaching began during this phase, along with the third PD stage, “understanding of learning.” Coaching involved two one-on-one in-person meetings lasting 1 hr each, at 4 and 12-weeks post-training, respectively. During the first meeting, ECEs worked out action plans and personal objectives, based on their own needs or on the implementation challenges encountered. The main challenges targeted by the ECEs were a lack of preparation time and difficulty embedding LOs (especially during free play). Between the two coaching sessions, a 15- to 20-min period of in-class observation was video-recorded, in accordance with the ECE’s action plan and personal objectives. The coach watched the video first and then presented it to the ECE during the second coaching session to offer supportive and constructive feedback, as defined by Snyder et al. (2022). At the end of the experimental phase, ECEs completed the questionnaire again, enabling them to assess their practices in line with the third PD stage, “understanding of learning.”
Interobserver Agreement
Interobserver agreement (IOA) data were collected for every 10-min video (30% of the coded 15-s intervals) of each ECE at each phase. Two previously trained research assistants independently filled out scoring grids for this 30%. The point at which the 30% observation began was determined at random. The average rate of agreement (number of agreements divided by the number of agreements and disagreements multiplied by 100) regarding the presence or absence of LOs was calculated. The average rate of agreement for all ECEs was 90% (84% for ECE1, 95% for ECE2, and 91% ECE3). These rates are within the interval of minimum acceptable values of 80% to 90% (Hartmann et al., 2004).
Data Analyses
The first and second authors independently conducted visual analyses based on the work of Barton et al. (2018) and Lobo et al. (2017), and then met to reach a consensus. The following elements were considered for each case: level, trend, variability and immediacy of the effect, overlap, and consistency of data organization across similar phases among the cases.
Non-overlap of all pairs (NAP; Parker & Vannest, 2009) was calculated for each participant. The following markers proposed by Parker and Vannest (2009) were used to measure the effect size of NAP values: weak (0.50–0.65), moderate (0.66–0.92), and strong (0.93–1). Then, between-case standardized mean difference (BC-SMD; Hedges et al., 2013) was computed to quantify the difference in the number of literacy LOs between the baseline and intervention phases. As Kratochwill et al. (2010) indicated, results regarding the range of effect to assess using different statistical analyses—in this case, NAP and BC-SMD—can be compared and allow for the determination if results are consistent. On one hand, the non-overlap index provided quantification for each participant. On the other hand, the BC-SMD provided an overall effect-size estimation comparable to Cohen’s d and interpreted in the same manner (Pustejovsky et al., 2014).
The R package scdhlm (Valentine et al., 2016) and the website (https://jepusto.shinyapps.io/scdhlm/, Pustejovsky, 2016) were used to calculate the BC-SMD, which estimates the effect size “based on [Hedges’s] small-sample correction to the restricted maximum likelihood (REML) estimator” (Pustejovsky et al., 2014, p. 370). An analytical model allowing for variable intercepts, fixed treatment effects, and different baseline trends was chosen (Pustejovsky et al., 2014). REML was employed with random effects for the intercept and initial slope to account for individual differences in baseline levels and pre-intervention trajectories. Given that within-subject residual variance varied between the baseline and treatment phases, the model specified heteroscedastic errors at the session level (“variance differs across phases”). Because the intervention’s effect was assessed at the end of the treatment phase, the standardization relied on the residual variance measured during that phase. The aim was to avoid averaging the variability for each phase, which could distort the estimate due to the less stable baseline data. By using distinct residual variances for the baseline and intervention phases, the BC-SMD captured the actual size of the treatment effect more precisely, without being influenced by the instability present beforehand.
Results
The number of LOs observed for each ECE during the baseline phase ranged from low to moderate, with some variability. Such variability was expected because all the ECEs received some training on this subject before the project, and part of their mandate included fostering children’s emergent reading and writing skills. Figure 1 presents the number of LOs by observation moment and conditions for each ECE, and shows that after introducing PD, more literacy LOs were developed. The standardized mean difference among the participants indicated a significant increase in literacy LOs, with an effect that qualifies as strong (BC-SMD = 1.83, 95% confidence interval [CI] = [−0.84, 4.49]). The baseline instability observed for E1 increased the initial variability and supported the selection of the analytical model (random baseline slope) and the assumption of heteroscedasticity of variances.
Frequency of Learning Opportunities Observed by Time of Observation
For all three ECEs, the number of LOs rose rather quickly from the very beginning of PD, although some delay was observed for ECE1. The level of the intervention phase was moderate to strong for the three ECEs and characterized by a slight acceleration and data variability. For all three ECEs, the first coaching session was followed by lower numbers of observed LOs. Afterwards, for ECE1 and ECE3, the number of LOs started to rise again, while a certain stability was observed for ECE2. Both visual and statistical analyses indicate a functional relation between the PD and the increase in LOs provided by the ECEs.
ECE1
Baseline for ECE1 can be described as weak-to-moderate, stable with no acceleration. In Phase 2 (intervention), an increase in LOs with a slight acceleration was observed, and the level was moderate to strong. Variability in data is visible. There seemed to be some latency in the impact of the intervention. Indeed, Points 7 and 9 overlapped with some baseline points. From observation moments 8, then 10, 11 and 12, the number of LOs the ECE offered surpassed the highest baseline rates. NAP was 0.88, 95% CI = [0.61, 0.97], for an effect characterized as moderate. After the first coaching period, a slightly slower rate was noted for two observation moments, followed by acceleration and stabilization of LOs at higher values than those for the baseline phase.
ECE2
ECE2’s baseline was weak, despite a data point with a higher frequency, and stable with no acceleration. The intervention phase showed a moderate to strong level, with a slight acceleration. In this phase, variability was visible for the first half of observation moments, then some stability occurred. The impact of the intervention was quickly visible during the first four observation moments. Also, there was almost no overlap between the two phases except for point 7 (baseline), a recorded moment of shared reading that integrated a high number of literacy LOs. NAP was 0.87, 95% CI = [0.65, 0.95], a moderate effect. After the first hour of coaching, a slight deceleration was observed, but the level remained high, and then the number of LOs stabilized.
ECE3
ECE3’s baseline was overall weak and stable, despite two points with higher frequencies. The intervention phase was characterized by variability in data and a rather moderate level. If the trend analysis excluded point 13, which stood apart, there was a slight acceleration. The impact of the training was observed, with an immediate increase in the frequency of LOs. The first three points of the intervention phase were higher than the last three points of the baseline phase. For the first two observation moments that followed the first hour of coaching, the number of LOs overlapped three baseline points. An increase in LOs was then observed. After the second hour of coaching, two points (27; 30) overlapped four baseline points. Nonetheless, most data pairs did not overlap between Phase 1 and Phase 2. NAP was 0.86, 95% CI = [0.66, 0.95], a moderate effect.
In conclusion, there was data consistency among participants, baseline phases with a weak-to-moderate effect with some variability, intervention phases with a slight acceleration and moderate to high numbers of LOs despite variability in data, and low percentages of data overlap with comparable NAP for all ECEs. In short, visual and statistical analyses tended to show that PD had an impact on the number of literacy LOs offered to children, that is, there appeared to be more LOs following PD.
Discussion
This article aims to assess whether the number of literacy LOs provided to 4-year-old children in inclusive kindergarten by ECEs increases after a short multicomponent PD on the ABI approach. Findings indicated that the number of LOs provided after PD increases for all ECEs. These findings are consistent with those of Lemire et al.’s (2021) study on ABI implementation to support emergent literacy-promoting practices of ECEs. As other studies have also shown (Cunningham et al., 2015; Elek et al., 2022; Jaskolski & Moyle, 2023), emergent literacy practices of ECEs working with young children can be improved through PD. However, Markussen-Brown et al. (2017) reported that PD is more effective in supporting indirect practices (e.g., adding materials) than direct practices. Results of our study tend to show that PD allows for an increase of LOs involving ECEs in their interactions with children (i.e., direct intervention). These findings are interesting because they indicate that to foster children’s emergent literacy skills, direct intervention should be offered in addition to indirect intervention (Dubois et al., 2019; Piasta, 2016).
Also, according to different authors (Brunsek et al., 2020; Markussen-Brown et al., 2017), to improve the practices of ECE, intensive PD programs are required that provide numerous training hours with frequent coaching sessions over a longer period. For example, in a recent study by Jaskolski and Moyle (2023), ECEs received a total of 43.5 hr of basic training, an additional 15 hr of specific training on phonological awareness, and monthly coaching. By comparison, PD in the current study is more concise: 6 hr of training (theoretical content, practical exercises) and two 1-hr coaching sessions, with an on-site recorded observation period in between. Thus, the results of the present study seem to be in line with some results of Markussen-Brown et al. (2017), suggesting that PD with multiple components might be more beneficial than long and intense PD.
Otherwise, despite an overall increase in LOs, a certain decline is observed after the initial coaching session, which could be explained by an expected latency effect. This delayed change does not question the impact of the intervention, since it is consistent among participants and predictable because it is a normal process of acquiring new skills or knowledge (Barton et al., 2018; Lieberman et al., 2010). In the current study, during the initial coaching meeting, each ECE had to set goals to achieve based on her identified needs. For example, one ECE wanted to increase the number of LOs during free play. To do so, the ECE targeted various actions to reorganize and improve symbolic play material. She also wanted to develop a planned play activity to introduce this new material to the children. The actions required to reach new objectives are time-consuming and could explain the decline in LOs. From the ECEs' perspectives, lack of time is a real challenge to implementing new practices (Lemire et al., 2022; Ohi, 2014). Also, our findings indicate data variability in the intervention phase for the three ECEs, which could be explained in part by unforeseen in-class events (e.g., conflicts between children, special activities, or illness). Such events are likely to reduce the readiness of ECEs to implement new practices.
Limitations
One study limitation is that because ECEs are aware they are being recorded, it may lead to a social desirability effect, increasing the number of LOs during the recording periods. However, the large number of recorded videos increases the likelihood that the ECEs became accustomed to it, and they then proceed with their actions as usual. Another limit is that, to our knowledge, there are no criteria or target number of LOs to reach. Also, the absence of a validated tool to assess if the implementation of the ABI proceeds as intended could limit the reliability assessment. Finally, the lack of a maintenance and generalization phase is among the limitations of the present study.
Future Research
Our study results are encouraging with respect to possibilities for increasing the number of literacy LOs through 1-day training and a bit of coaching. It should be highlighted that shorter PD could be more realistic in terms of time and money commitments, notably in regard to the teacher shortage crisis in the United States (García & Weiss, 2020), in Québec, Canada (Massous, 2023) and elsewhere in the world (United Nations, 2024). It is essential to continue to enhance knowledge to ensure the availability of PD programs adapted to the realities of preschool educational settings. In that sense, more studies could test the PD developed during this project, but in a group format with more participants. Studies could also document the impacts of ABI practices on young children’s literacy skills acquisition. Finally, it might be relevant for future research on the implementation of ABI to ask ECEs to assess children by including relevant content and practical exercises on the subject in the PD.
Footnotes
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study, including the research activities, authorship, and publication of this article, was conducted with financial support from the Fonds de recherche du Québec—Société et culture (FRQSC)/Quebec Research Fund—Society and Culture (FRQSC). This funding supported the implementation of the project and the dissemination of its findings.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.