Coaching Teachers for Emergent Literacy Instruction Using Performance-Based Feedback

Abstract

Coaching has garnered support as a professional development approach that helps teachers use research-based instruction to teach emergent literacy skills to young children. However, approaches to coaching vary widely, as do the backgrounds and training of the teachers included in different studies. This study investigated the influence of skill-focused coaching on certified teachers’ use of three clusters of instructional skills. Discussion of visually represented observation data provided the primary framework for coaching. After coaching, teachers in the intervention group used significantly more of the skills in two of three clusters of instructional skills than did those in the control group. Group differences also were found on the language and literacy sections of a general measure of the classroom literacy environment, indicating that targeted skill coaching may have more general effects on classroom quality. Results are discussed from the perspectives of professional development and research on coaching.

Keywords

emergent literacy intervention coaching

Kindergarten marks the beginning of a child’s formal journey toward becoming an educated, contributing member of society. What is gained from kindergarten and what is carried forward to future schooling are heavily influenced by the skills that children bring into their kindergarten year, and children with early difficulties are unlikely to catch up (National Institute for Literacy [NIL], 2008). Children from low-income families are especially at risk, often entering preschool and kindergarten already behind their more advantaged peers in areas that predict later literacy learning and achievement (Halle et al., 2009; NIL, 2008). Accordingly, increasing attention has been given to the critical role of prekindergarten programs in laying the foundations for emergent literacy learning, particularly for young children who are at risk because of low family income. However, whether children gain important early language and literacy skills during their preschool years depends on the quality of the preschools that they attend (Hamre & Pianta, 2007). Both descriptive (Howes et al., 2008; Justice, Mashburn, Hamre, & Pianta, 2008; Mashburn et al., 2008) and intervention (Justice, Meier, & Walpole, 2005; Wasik, Bond, & Hindman, 2006) studies have demonstrated relationships between children’s learning and teachers’ use of particular types of instructional strategies (e.g., language expansions, open-ended questions). Although many early childhood classrooms provide adequate support for emotional and social development, fewer demonstrate the kinds of instructional interactions associated with children’s language and emergent literacy learning (Howes et al., 2008; LoCasale-Crouch et al., 2007). Children from low-income families, who are most at risk, are more likely to attend lower quality preschools (LoCasale-Crouch et al., 2007; Pianta et al., 2005). Direct attention to early childhood teachers’ instructional skills as a focus of professional development is warranted.

Emergent Literacy Coaching

The National Staff Development Council (2001) recommended that professional development be sustained over time, linked to curriculum and student outcomes, collaborative, and interactive. Coaching that responds directly to what is happening in individual classrooms appears to be central to changing instructional interactions (Powell & Diamond, 2011). But what occurs in coaching varies widely (Deussen, Coskie, Robinson, & Autio, 2007), and there are few studies of the active ingredients of effective coaching as they relate to different types of outcomes (Sheridan, Edwards, Marvin, & Knoche, 2009; Wasik & Hindman, 2011).

Several studies indicate that coaching may add value to other forms of professional development such as in-service courses (Jackson et al., 2006; Neuman & Cunningham, 2009) or training for curriculum implementation (Assel, Landry, Swank, & Gunnewig, 2007; Lonigan, Farver, Phillips, & Clancy-Menchetti, 2011) in helping teachers acquire knowledge and skills for teaching emergent literacy. Furthermore, a small number of studies has assessed the effects of coaching as a primary or stand-alone approach to professional development in which instructional interactions related to language and emergent literacy were among the desired outcomes. Building on their earlier study (Neuman & Cunningham, 2009), Neuman and Wright (2010) directly compared the effects of coaching alone with those of an in-service course. Community-based and prekindergarten teachers were randomly assigned to a course, coaching alone, or a wait-list. Coaches took active roles in the classrooms, assisting by coplanning, coteaching, demonstrating, and side-by-side prompting. Although both intervention groups obtained higher scores than the wait-list group on a test of emergent literacy knowledge, teachers in the coaching group had significantly higher ratings on the Language & Literacy Classroom Observation Toolkit; (ELLCO; Smith & Dickinson, 2002) than either of the other groups. However, there were no differences in the ELLCO dimensions most related to teacher–child interactions (interactional environment, support for learning, and teaching strategies).

A more direct focus on skills for teaching emergent literacy was undertaken in a single-subject study in which coaching was implemented as a stand-alone intervention (Hsieh, Hemmeter, McCollum, & Ostrosky, 2009). Skills were grouped into three clusters for the purposes of coaching (book reading, phonological awareness and alphabetic principle, print concepts and writing) and introduced in random order to five child care and two public school prekindergarten teachers. Coaches’ written observations provided examples of what teachers said (verbatim) and did in relation to each skill within the cluster being coached. The percentage of skills observed in the cluster also was noted on the observation sheet and graphed over time. Sharing these data provided the framework for feedback and discussion subsequent to each observation. After approximately 6 weeks of twice-weekly coaching, all teachers used more of the strategies in each cluster.

Two additional studies, in which the primary purpose was to evaluate the effects of electronic delivery of coaching, also lend support to the potential of coaching for influencing emergent literacy teaching skills. Powell, Diamond, Burchinal, and Koehler (2010) compared electronically delivered, remote coaching based on teacher-submitted videotapes to on-site coaching, with Head Start teachers randomly assigned to these two conditions and a wait-list. Teachers were coached on instructional skills agreed on by coach and teacher in advance. After seven sessions, both coaching groups obtained higher ratings on the ELLCO than did teachers on the wait-list, with no difference between teachers in the two coaching conditions. Use of instructional skills coded from audiotapes yielded larger gains for on-site coaching in code-focused instruction but not in language instruction.

Pianta, Mashburn, Downer, Hamre, and Justice (2008) also addressed the use of remote coaching, targeting the quality of teaching represented in the three domains (emotional support, classroom organization, and instructional support) of the Classroom Observation and Scoring System (CLASS; Pianta, La Paro, & Hamre, 2008), with the latter domain including many of the instructional skills identified as important in previous research on emergent literacy. Teachers in state-funded prekindergarten classrooms were randomly assigned to access to a hypermedia library illustrating teaching strategies or to remote coaching on their own videotaped interactions plus access to the hypermedia library. The coaching group received remote, biweekly, yearlong consultation. Coaches edited the videotapes to emphasize specific dimensions from the CLASS and provided written feedback and questions for the teacher to respond to. Teachers who received consultation had significantly higher rates of change on three of seven CLASS dimensions, including language modeling. As a group, these four studies indicate that coaching may be useful as a stand-alone or primary approach to professional development when the goal is to acquire emergent literacy teaching practices. Three of the four studies (Hsieh et al., 2009; Pianta, Mashburn, et al., 2008; Powell et al., 2010) demonstrated change in specific instructional skills and the fourth (Neuman & Wright, 2010) resulted in change on broader classroom measures of the emergent literacy teaching environment. Furthermore, while approaching coaching in very different ways, in each of the three studies that resulted in change in instructional skills, coaching was focused directly on these skills; each also used some form of visual representation of observational data. A review by Casey and McWilliam (2011) indicated that graphical and other forms of visual feedback (e.g., frequencies, percentages, verbatim examples) may be useful in changing teachers’ use of specific behavioral support or instructional strategies. Also, in the three skill-focused studies reviewed, the primary or only role of the coaches was to provide feedback based on observations. Despite these commonalities, the studies differed in how coaching targets were selected, defined, and measured, as well as in how feedback was provided to teachers. Teacher backgrounds and work contexts also differed across studies in ways that could affect the influence of coaching as well as interpretation of results (Assel et al., 2007; Lonigan et al., 2011; Pianta et al., 2005). There continues to be a need for research to understand whether and how different approaches to coaching bring about change in particular types of instructional practices in early childhood teachers from different backgrounds (Powell & Diamond, 2011; Sheridan et al., 2009; Wasik & Hindman, 2011; M. J. Zaslow, 2009).

Purpose and Research Questions

The purpose of the current study was to replicate the intervention employed in a previous, single-subject study of coaching (Hsieh et al., 2009) by using it with a more homogeneous group of teachers as well as to evaluate its usefulness under somewhat different coaching conditions. Given previous studies showing variations in training outcomes for teachers from different types of programs and with differing amounts of specialized training (e.g., Assel et al., 2007; Lonigan et al., 2011), this study limited variation among teachers by including only certified teachers from state-funded prekindergarten classrooms located in public schools. In contrast to the previous study (Hsieh et al., 2009), in which coaching occurred semiweekly over approximately 6 weeks and the amount of time spent on each cluster was based on change in individual teachers in accord with the single-subject design, the current study provided biweekly coaching over a full year. This study also included as an outcome a broader measure of quality (the ELLCO), to examine whether targeted skill coaching would have an effect on more global qualities of the language and literacy teaching environment in addition to specific teaching interactions.

Two research questions were addressed:

Research Question 1: Does coaching result in changes in teachers’ use of targeted literacy teaching skills? and

Research Question 2: Does coaching on specific literacy teaching skills result in changes in emergent literacy teaching environments?

Method

The study employed a group design in which teachers in 13 classrooms, representing three different state-funded prekindergarten programs within one county, were randomly assigned within programs to receive on-site, yearlong coaching (n = 7) or to a wait-list control group (n = 6). One program had three teachers in the intervention group and four in the control group, the second had two teachers in each group, and the third had one teacher in each group. Teachers were assigned within programs as all programs wanted immediate access to the intervention; teachers in the intervention group were asked to refrain from sharing what they were learning with other teachers in their programs. No teachers in either group had participated in any previous studies of emergent literacy teaching.

Participants

Teachers were recruited from two early childhood programs in a medium-sized midwestern community and one cooperative program that served the surrounding county. All recruitment procedures and study methods were approved by the university’s institutional review board (IRB). To recruit participants, the authors visited each program during the spring prior to beginning the study the following fall. In the two in-town programs, teachers who volunteered represented 60% of all teachers in each of those programs; in the county program, the two volunteers represented 30% of all teachers in the program. One teacher in the control group (a county teacher) was later dropped because she agreed to pilot-test an emergent literacy curriculum that was not part of the study. This resulted in 12 teachers, 7 of whom received coaching and 5 of whom did not. All classrooms served children with disabilities and children considered to be at risk for later academic difficulties based primarily on low family income. Based on a brief questionnaire and observations completed prior to starting the intervention, none of the teachers used a published curriculum. All teachers had similar schedules, with large and small group activities and substantial amounts of center time during which children could choose where to play. All teachers regularly read books to children, usually in a large group.

The majority of teachers in both groups had master’s degrees (71% and 80%, coaching and control groups, respectively). All were state-certified in early childhood education with an endorsement in early childhood special education, as required in inclusive classrooms. Teachers in the coaching group had an average of 9 years of experience in early childhood (range = 1-20 years), whereas control teachers had an average of 14 years (range = 3-25 years). The groups did not differ in the proportion of teachers with bachelor’s and master’s degrees (χ² = 11, p = .74) or in years of experience (z = −.74, p = .53). Most (86%) of the teachers in the coaching group and all in the control group had received previous district-level in-service training on emergent literacy. All teachers were women, and all but two were Caucasian. At the beginning of the year, the majority of children in these teachers’ classrooms were 4 years of age. Children in the intervention and control classrooms did not differ significantly in age (t = 1.5, p = .12) and ethnicity (χ² = 5.54, p = .13) nor did classrooms differ in the proportion of children with Individualized Education Programs (IEPs; χ² = .24, p = .87).

Procedures

Skill clusters were the same as those in the Hsieh et al. (2009) study. For that study, instructional skills in the clusters were based on a comprehensive review of the literature, including research studies describing intervention approaches as part of their methodologies (e.g., Justice et al., 2005; Wasik & Bond, 2001) and research-to-practice books (e.g., Justice & Pence, 2005; McGee & Richgels, 2004) available at the time the clusters were developed. For coaching, skills were grouped into three clusters reflecting areas of emergent literacy found to be highly predictive of later reading and writing (Lonigan, 2006; Strickland & Riley-Ayers, 2006), similar to the three activity groupings used by Lonigan et al. (2011). As shown in Table 1, Cluster A focused on book-reading strategies to support language and comprehension (e.g., Wasik & Bond, 2001). Cluster B addressed sound awareness and the alphabetic principle, with particular focus on strategies to promote rhyme and beginning sound awareness (e.g., Ukrainetz, Cooney, Dyer, Kysar, & Harris, 2000); alphabet teaching also was included in Cluster B to link it closely with phonological awareness (Lonigan et al., 2011). Cluster C focused on skills to support print concepts and written language (e.g., Justice & Ezell, 2004). Within each cluster, strategies were included that would address the varied supports needed by children at different levels of learning (Hindman, Connor, Jewkes, & Morrison, 2008; Justice & Pence, 2005). For example, Cluster B included providing a model as well as asking for production on the assumption that, to support all children, the teacher should include use of both types of prompting within the same activity. When a particular skill could fit logically within more than one cluster, an effort was made to balance the number of strategies across the three clusters for the purposes of coaching; thus, underlining text to draw attention to print or talking about print features could just as easily fit within a book-reading session or a writing activity, but for purposes of coaching was included within Cluster C. Overall, Cluster A was primarily concerned with understanding of meaning in a situation where meaning talk occurs more naturally than in other contexts (Hindman et al., 2008), whereas Clusters B and C were primarily concerned with code-related skills. Skills listed within each cluster checklist were stated in lay language to ensure that they would be readily understood by teachers (e.g., using “thought questions” rather than “open-ended” and/or “decontextualized” questions in Cluster A).

Table 1.

Strategy Clusters.

Cluster	Teaching strategies
Cluster A—Vocabulary, information, comprehension, narrative structure	• Introduce content of book
	• Talk about physical features of book
	• Sustain interest
	• Ask thought questions
	• Expand children’s information
	• Involve children in reviewing/closing
	• Relate story concept and vocabulary to past or future experiences
Cluster B—Phonological awareness, alphabetic principle	• Use age appropriate activities to emphasize letters and sounds
	• Demonstrate and emphasize characteristics of sounds and words
	• Ask children to identify characteristics of sounds and words, given model or choice
	• Ask children to demonstrate characteristics of sounds and words
	• Point out letters associated with sound or word
Cluster C—Print concepts, written language	• Use print to represent ideas in a way that is obvious to children
	• Point out relationship between picture/object and associated print
	• Encourage use of writing in play
	• Point out print conventions
	• Model and explain how to use writing materials
	• Use words related to writing and reading

Source. Hsieh, Hemmeter, McCollum, and Ostrosky (2009).

Abbreviations: Cluster A, story comprehension and language; Cluster B, phonological awareness; Cluster C, print concepts and writing.

Prior to beginning the school year, teachers in the coaching group met on two consecutive days for orientation (10 hr total). They were introduced to important areas of emergent literacy and provided with a rationale for directly teaching emergent literacy skills to children who are at risk for later reading and academic disability (NIL, 2008). An overview of the three clusters of teaching skills was given. Handouts included written examples of what the skills might look like in practice (i.e., what teachers might say or do). Each teacher then selected one activity setting in which to practice and receive coaching on each cluster. The goal of the orientation was for teachers to understand enough about the skills to identify where in their daily schedules they were already using or could use similar skills, and to become familiar with the format in which they would be observed. A goal of achieving 80% or more of the skills in each cluster was established, drawing on the criterion used in a previous study (Hsieh et al., 2009). Three brief additional group meetings (1.5 hr each) were held during the year to share experiences; teachers also received information on useful websites and ways to facilitate parent involvement in early literacy learning. No additional training on clusters was provided.

Given no evidence in Hsieh et al. (2009) of order effects for introduction of clusters, clusters were coached in the same order and began and ended at the same time for all teachers. Cluster A, book reading, was introduced first because it was a familiar part of each classroom day for all teachers. Because Cluster C had been least familiar to teachers in the previous study and required the most adjustment in use of space and time, it was introduced last. Coaching occurred biweekly, with five visits on each of the three skill clusters (total of 15 visits per classroom). Before initiating each cluster, the coach and teacher reaffirmed the specific activity context that would be used to practice and receive feedback on the cluster. The week before introducing the cluster, the coach collected data to use as a basis for discussion during the initial meeting on that cluster. Observed examples of each skill in the cluster were written onto the observation sheet as they occurred. As coaching was initiated, the coach briefly (10-15 min) reviewed the specific skills in the cluster using these written, verbatim examples from the teacher’s classroom, illustrating the skills using what the teacher was already doing as well as with other examples of how the skills could have been used. The goal was to build understanding of the cluster and also support a sense of competence. The criterion of achieving 80% of the skills in the cluster by the end of five visits was reviewed.

At each visit, coaching followed a clinical supervision cycle (Krajewski, 1993). Each session included (a) a brief preobservation discussion to reestablish the focus and context of the observation and to review percentage data from the previous session, (b) teaching-observing, (c) meeting to view and discuss the data and compare it with previous observations and with the 80% criterion, and (d) looking forward to what the teacher might do differently on the next visit. Joint inspection and discussion of data obtained during the observation served as the primary strategy for coaching. Data shown to the teacher included the actual data sheet with the coach’s written observations and the percentage of total skills demonstrated. As percentage data emerged across sessions, they were graphed and shown to the teacher as well. Feedback sessions occurred shortly after the observation and lasted approximately 15 min. Coaches did not participate in any other way in the classrooms unless directly requested by the teacher.

Four researchers provided coaching to the teachers. Three had doctorates in early childhood special education and the other was a doctoral student. Two had served as emergent literacy coaches in a previous project; all had experience as university practicum supervisors and in early childhood professional development. To ensure fidelity to the coaching protocol, each coach completed a checklist of coaching steps at each visit, noting each step as it was accomplished. These checklists were reviewed in coach meetings, which initially occurred approximately once a week, and then once a month as the coaching procedures became more fluent.

Instrumentation

Teacher observation checklists

Each checklist included the list of skills associated with that cluster as well as multiple exemplars demonstrating different ways that teachers could demonstrate the skill (Hsieh et al., 2009). Data were collected at the skill level, with each skill scored as yes or no based on occurrence of a specified number of exemplars. For some skills, the observer made a decision of observed/not observed based on the entire session (e.g., linking the story to a future or familiar event). Other skills were scored only if the teacher used multiple examples of the skill; the expectation was not that the teacher would use every exemplar but rather the minimum number of exemplars specified for the skill (e.g., at least three different ways of sustaining interest). The coach also took notes on what the teacher said and did in relation to each skill, to use as a guide to scoring as well as for display and discussion with the teacher.

All coaches were trained to use the data sheets using videotapes from the previous study. Each coach then achieved a minimum reliability of 85% on two additional tapes and subsequently in a classroom not involved in the study. Prior to coach training, two coaches and a reliability checker from the previous study reestablished reliability with one another and then trained the new coaches. During intervention, the reliability checker co-observed all clusters a minimum of twice per coach. Using the criterion for each skill cluster, each skill was scored independently by the coach and reliability checker as occurring or not occurring during the observation. These data were converted into percentages by dividing the number of skills scored as “yes” by the total number of skills in the cluster and multiplying by 100 (Kazdin, 1982). Interobserver reliability on teachers’ use of the skills across the year averaged 97% (range = 83%-100%).

Language and literacy classroom observation

Pre- and postcoaching observations were conducted using the ELLCO. This research version of the ELLCO includes four sections: a Learning Environment Checklist (LEC); two sets of one to five qualitative rating items grouped into General Classroom Environment (GCE) and Language, Literacy, and Curriculum (LLC); and a Literacy Activities Rating Scale (LARS) to record the frequency and duration of nine literacy behaviors. The ELLCO has internal consistency alphas ranging from .73 to .90 across subscales (Pianta, La Paro, et al., 2008). Prior to data collection, two experienced criterion coders observed together to achieve a minimum of 85% reliability on two observations, then trained three other observers to the same level in classrooms not part of the study. These three ELLCO observers, graduate students not otherwise involved in the study, were randomly assigned to intervention and control classrooms at each observation period. Observations occurred in mid-late September and early-mid May, prior to and after the coaching intervention.

Results

All analyses were performed using SPSS-X, Version 11.0. Because of the small number of teachers in each group, nonparametric statistics were used for all comparisons. Data from the initial and final observations for each cluster were summarized as the percentage of items occurring on each cluster, as described earlier. Total ELLCO scores were summarized as shown in the manual for each of the four separate sections of the instrument. The first research question addressed the effects of coaching on teachers’ use of the skills in each of the three instructional skill clusters. Pre- and postintervention scores are shown in Table 2. Despite random assignment, scores in all three clusters were higher in the coaching group prior to intervention, with a significant preintervention difference on Cluster A (Mann–Whitney U = 5.00, p = .048). Furthermore, teachers in both groups used a larger percentage of the skills in each cluster post intervention than pre intervention. Nevertheless, Mann–Whitney U tests indicated significant postintervention differences between groups for both Cluster B (U = .00, p = .003) and Cluster C (U = 3.5, p = .018). Although group differences were no longer significant for Cluster A (U = 12.00, p = .432), teachers receiving coaching continued to use a higher percentage of the skills in this cluster as well.

Table 2.

Average Percentage of Skills Used in Emergent Literacy Teaching Clusters Pre- and Post-Coaching.

	Intervention (n = 7)				Control (n = 5)
	Pre		Post		Pre		Post
Cluster	M	SD	M	SD	M	SD	M	SD
Cluster A	.59^a	.24	.80	.30	.23^a	.26	.60	.32
Cluster B	.23	.27	.86^b	.10	.20	.20	.36^b	.22
Cluster C	.50	.27	.81^b	.18	.27	.25	.33^b	.31

Abbreviations: Cluster A, story comprehension and language; Cluster B, phonological awareness; Cluster C, print concepts and writing.

Significant group difference pre intervention (Mann–Whitney U). ^bSignificant group difference post intervention (Mann–Whitney U).

The number of teachers ending the year having achieved the minimum goal of 80% also demonstrated the effects of coaching. Prior to coaching, 3 of the 12 teachers used none of the skills in Cluster A, 5 used none in Cluster B, and 2 used none in Cluster C. Only 1 teacher in each of Clusters A and C met the criterion set for the intervention (80% of the skills) prior to coaching. After coaching, 6 of the 7 intervention teachers reached criterion on Cluster A, 5 of the 7 achieved criterion on Cluster C, and all achieved criterion on Cluster B. In the control group, no teachers were at criterion in either Clusters B or C at the final observation, and only 1 reached criterion on Cluster A.

The second research question asked whether coaching on specific skills would influence the quality of the broader language and literacy teaching environment, as measured by the ELLCO. Pre- and postintervention scores are shown in Table 3. No significant differences were found prior to intervention for any of the four sections of the ELLCO. Postintervention comparisons yielded significant group differences on all three sections most directly related to language and literacy (LEC, U = .50, p = .003; LLC, U = 1.00, p = .003; LARS, U = 1.0, p = .005). No differences were found for scores on the GCE (U = 13, p = .53).

Table 3.

Total Scores for Sections of ELLCO Pre- and Post-Coaching.

	Intervention (n = 7)				Control (n = 5)
	Pre		Post		Pre		Post
ELLCO section	M	SD	M	SD	M	SD	M	SD
LEC	21.00	4.16	28.57*	2.07	17.80	4.71	16.00*	5.92
GCE	20.86	2.27	22.86	2.61	19.40	2.51	21.60	2.10
LLC	25.71	3.86	31.14*	1.57	23.60	2.97	25.60*	2.70
LARS	6.86	3.85	10.71*	1.49	4.80	2.58	7.00*	1.58

Abbreviations: ELLCO, Language & Literacy Classroom Observation Toolkit; LEC, Literacy Environment Checklist; GCE, General Classroom Environment; LLC, Language, Literacy, and Curriculum; LARS, Literacy Activities Rating Scale.

Significant group difference post intervention (Mann–Whitney U), p < .05.

Discussion

This study contributes to the small body of research on the unique contributions of coaching to teacher practice by evaluating further the potential of a specific coaching approach that combined a focus on specific clusters of teaching skills with feedback and discussion based primarily on visually represented data related to the skills in the cluster being coached. To narrow the range of variation, only teachers with specialized training in early childhood, working within similar programmatic contexts, were included.

Effects of Coaching on Instruction

Previous studies have demonstrated a link between the quality of the instructional process and children’s learning (Mashburn et al., 2008), as well as the difficulty of changing instructional interactions between teachers and children (Justice, Mashburn, Pence, & Wiggins, 2008). One of the variations among coaching studies that has impeded interpretation of results is the variation in level of specificity in outcome measures (Mashburn et al., 2008). Burchinal, Vandergrift, Pianta, and Mashburn (2010) noted the need to more clearly align measures of the instructional process with the definition of the process being used. In the current study, the primary outcome measure was in exact alignment with the focus of coaching, and examining performance on the measure was a central part of the coaching. Visually represented data were the primary coaching strategy used to assist teachers to increase their use of three different clusters of teaching skills within the context of specific, self-selected activity settings. Results lend support to incorporating this type of data into the feedback process when coaching specific skills. Feedback using examples from their own instruction may help teachers isolate and understand particular skills, whereas summary data such as percentages may provide them with information to support comparison with a desired standard or with their own previous behavior. As a coaching mechanism, visually represented data aligned closely with the targets of coaching may provide a basis for nonjudgmental, transparent problem solving between teacher and coach, giving teachers a way to assess their own needs and growth (Sheridan et al., 2009).

Despite random assignment, there was a significant preintervention difference between groups on one cluster (Cluster A, story reading), with no significant difference post intervention. The lack of a postcoaching group difference for Cluster A may reflect a ceiling effect for teachers who received coaching. Despite the lack of a significant difference for Cluster A, the number of coached teachers who achieved the criterion of using 80% of the skills in the cluster increased from two to six of the seven teachers from pre- to post coaching. In the control group, although the average percentage of skills used from Cluster A increased from pre- to post coaching, only one teacher reached the 80% criterion level on the final observation.

An unexpected finding was that averages on skill clusters increased from beginning to end of the school year in both groups, particularly on Cluster A. It may be that as children gained vocabulary and familiarity with book-reading routines, teachers’ approaches to reading were adapted to children’s emerging ability to participate in different ways. The smaller pre to post changes in Clusters B and C may also represent adaptations to emerging abilities in children. Nevertheless, it is clear that teachers in the intervention group benefited from coaching in ways that went beyond adaptations to children’s developing skills.

Most teachers in the coaching group achieved the 80% criterion for most clusters within five coaching visits, indicating that this may be a realistic time frame to plan for coaching within the context of a school year. Nevertheless, one or two teachers in the current study did not achieve criterion within the 5 weeks designated for a given cluster. Although both this and the previous study provided a sustained focus on the same skill cluster over time (Powell & Diamond, 2011), for some skills, coaching may need to be extended for some teachers or supplemented with other coaching components such as modeling. In future studies, it will be important to evaluate how individual differences in teachers and classrooms may influence the effectiveness of different formats in which coaching is provided. For some teachers, individual skills within clusters (e.g., vocabulary teaching, use of open-ended questions) may need to be coached in more depth.

Effects of Coaching on Quality of Language/Literacy Environment

Research is lacking on whether targeted, skill-focused coaching might serve as a point of entry for influencing broader changes in classrooms, in contrast to its more typical use as an addition to courses or in-service training (Sheridan et al., 2009). Significant postintervention group differences were found on the three components of the ELLCO most related to language and literacy teaching, whereas no similar differences were found for ratings of the quality of the GCE. It appears that coaching on specific skills may have influenced more global ratings of the process environment beyond the teacher–child interaction. As noted above, implementation of Clusters B and C sometimes necessitated adding a new or modified activity segment to the classroom day. These additions may have been enough to yield higher ratings on the ELLCO. Although generalization of the skill clusters beyond the activity settings used for coaching was not addressed in this study, it is also possible that as teachers became more thoughtful about the skills included in the clusters, they began to offer additional opportunities for emergent literacy learning throughout the day. As implemented in this study, coaching was limited in scope and not intended to address broader goals. Nevertheless, being coached on instructional behavior may provide a foundation for deeper understanding (M. Zaslow, Tout, Halle, & Starr, 2011). The effects of coaching for targeted behavior change on broader aspects of teaching is worthy of additional research, particularly because coaches may lean toward working on broader classroom environments and conditions before addressing more specific instructional behaviors (Powell et al., 2010).

Limitations and Need for Future Research

The small number of classrooms in this study limited the options available for analysis of personal and contextual factors that may have contributed to differing outcomes for different teachers. Potentially important influences may include supportiveness of administrators (Baker, Kupersmidt, Voegler-Lee, Arnold, & Willoughby, 2010) and openness to coaching (Domitrovich, Gest, Gill, Jones, & DeRousie, 2009), neither of which were measured in this study. It should be noted that all teachers in this study were volunteers; it is not known how these teachers may have differed in motivation or other qualities from nonvolunteers. An additional threat to generalizability is that all coaches were faculty members or doctoral students. The usefulness of this coaching approach remains to be studied when used by in-school coaches or in-service providers. A further limitation is that no external observers collected fidelity of implementation data on coaching; rather, each coach used a checklist and noted when each step of the coaching process had been completed. External fidelity observers would further strengthen the confidence in the extent to which the independent variable was implemented as planned.

As coaching took place over most of the school year, maintenance observations could not be conducted within the time frame of the study. Coaching for maintenance deserves study on its own right to determine whether different coaching processes from those used to help teachers acquire instructional skills might also be useful for this different purpose (Sheridan et al., 2009). Finally, child outcomes that may have accrued from coaching were not included. Of the four studies of coaching reviewed above, two included child outcome measures. In one of these, children’s learning was affected in some but not all learning areas (Powell et al., 2010), whereas in the other, the design did not allow a conclusion as to whether the gains in children could be attributed to coaching (Hsieh et al., 2009). Many questions remain with respect to which child outcomes are influenced by which teaching skills and at what levels of teacher implementation. For some child outcomes, differences in even relatively low levels of specific instructional skills may be enough to influence children’s learning, whereas for others, a minimal threshold may be needed (Burchinal et al., 2010). Continuing research is necessary to study the effects of specific approaches to coaching in relation to both realistic and necessary levels of skills for bringing about change in different child outcome areas.

Practice Implications

Both this and other studies (Hsieh et al., 2009; Neuman & Wright, 2010; Pianta, Mashburn, et al., 2008; Powell & Diamond, 2011) indicate that coaching may be useful as a stand-alone option for professional development, rather than solely as a concurrent or subsequent addition to other professional development (Sheridan et al., 2009; M. J. Zaslow, 2009) and may in fact have effects on noncoached aspects of the classroom. Because the coaching process used was relatively straightforward, linking the process of coaching directly to observed examples of targeted outcomes, it could provide a useful coaching tool either applied on its own or integrated within more extensive coaching models.

The results of this study also make it clear that no assumptions should be made about the levels at which even highly qualified teachers, working in similar public school programs, implement emergent literacy instruction. The reasons for this low level of skill in these certified teachers should be examined. Emergent literacy is a relatively recent focus in prekindergarten, and may not yet have made its way into the preservice curriculum (Roskos, Rosemary, & Varner, 2006). Furthermore, state certification requirements that guide preservice curricula may not yet provide focused attention on emergent literacy. Given the number of children who could benefit from high-quality early literacy instruction, states and other entities responsible for preservice and in-service education of early childhood teachers would do well to include emergent literacy training among their priorities.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a field initiated research grant entitled “The effects of individual and group interventions on emerging literacy skills in young children” (Office of Special Education Programs, U.S. Department of Education, #H324C030114A).

References

Assel

M. A.

Landry

S. H.

Swank

P. R.

Gunnewig

(2007). An evaluation of curriculum, setting, and mentoring on the performance of children enrolled in pre-kindergarten. Reading and Writing, 20, 463–494.

Baker

C. N.

Kupersmidt

J. B.

Voegler-Lee

M. E.

Arnold

D. H.

Willoughby

M. T.

(2010). Predicting teacher participation in a classroom-based, integrated prevention intervention for preschoolers. Early Childhood Research Quarterly, 25, 270–283.

Burchinal

Vandergrift

Pianta

Mashburn

(2010). Threshold analysis of association between child care quality and child outcomes for low-income children in pre-kindergarten programs. Early Childhood Research Quarterly, 25, 166–176.

Casey

A. M.

McWilliam

R. A.

(2011). The characteristics and effectiveness of feedback interventions applied in early childhood settings. Topics in Early Childhood Special Education, 31, 68–77.

Deussen

Coskie

Robinson

Autio

(2007, June). “Coach” can mean many things: Five categories of literacy coaches in reading first (Issues & Answers Report, REL 2007-No. 005). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Education Laboratory Northwest. Retrieved from http://ies.ed.gov/ncee/edlabs

Domitrovich

C. E.

Gest

S. D.

Gill

Jones

DeRousie

R. S.

(2009). Individual factors associated with professional development training outcomes of the Head Start REDI program. Early Education and Development, 20, 402–430.

Halle

Forry

Hair

Perper

Wandner

Wessel

Vick

(2009). Disparities in early learning and development: Lessons from the Early Childhood Longitudinal Study-Birth Cohort (ECLS-B). Washington, DC: Child Trends.

Hamre

B. K.

Pianta

R. C.

(2007). Learning opportunities in preschool and early elementary classrooms. In Pianta

R. C.

Cox

M. J.

Snow

(Eds.), School readiness and the transition to school (pp. 49–84). Baltimore, MD: Brookes.

Hindman

A. H.

Connor

C. M.

Jewkes

A. M.

Morrison

F. J.

(2008). Untangling the effects of shared book reading: Multiple factors and their associations with preschool literacy outcomes. Early Childhood Research Quarterly, 23, 330–350.

10.

Howes

Burchinal

Pianta

Bryant

Early

Clifford

Barbarin

(2008). Ready to learn? Children’s pre-academic achievement in pre-kindergarten programs. Early Childhood Research Quarterly, 23, 27–50.

11.

Hsieh

W.-Y.

Hemmeter

M. L.

McCollum

J. A.

Ostrosky

M. M.

(2009). Using coaching to increase preschool teachers’ use of emergent literacy teaching strategies. Early Childhood Research Quarterly, 24, 229–247.

12.

Jackson

Larzelere

St. Clair

Corr

Fichter

Egertson

(2006). The impact of HeadsUp! Reading on early childhood educators’ literacy practices and preschool children’s literacy skills. Early Childhood Research Quarterly, 21, 213–226.

13.

Justice

L. M.

Ezell

H. K.

(2004). Print referencing: An emergent literacy enhancement strategy and its clinical applications. Language, Speech, and Hearing Services in Schools, 35, 185–193.

14.

Justice

L. M.

Mashburn

A. J.

Hamre

B. K.

Pianta

R. C.

(2008). Quality of language and literacy instruction in preschool classrooms serving at-risk pupils. Early Childhood Research Quarterly, 23, 51–68.

15.

Justice

L. M.

Mashburn

A. J.

Pence

K. L.

Wiggins

(2008). Experimental evaluation of a preschool language curriculum: Influence on children’s expressive language skills. Journal of Speech, Language, and Hearing Research, 51, 983–1001.

16.

Justice

L. M.

Meier

Walpole

(2005). Learning new words from storybooks: Findings from an intervention with at-risk kindergarteners. Language, Speech and Hearing Services in Schools, 36, 17–32.

17.

Justice

L. M.

Pence

K. L.

(2005). Scaffolding with storybooks: A guide for enhancing young children’s language and literacy achievement. Newark, DE: International Reading Association.

18.

Kazdin

A. E.

(1982). Single-case research designs: Methods for clinical and applied settings. New York, NY: Oxford University Press.

19.

Krajewski

R. J.

(1993). The observation cycle: A methodology for coaching and problem solving. In Anderson

R. H.

Snyder

K. H.

(Eds.), Clinical supervision, coaching for higher performance (pp. 99–112). Lancaster, PA: Technomic.

20.

LoCasale-Crouch

Konold

Pianta

Howes

Burchinal

Bryant

. . . Barbarin

(2007). Observed classroom quality profiles in state-funded pre-kindergarten programs and associations with teacher, program, and classroom characteristics. Early Childhood Research Quarterly, 22, 3–17.

21.

Lonigan

C. J.

(2006). Development, assessment, and promotion of preliteracy skills. Early Education and Development, 17, 91–114.

22.

Lonigan

C. J.

Farver

J. M.

Phillips

B. M.

Clancy-Menchetti

(2011). Promoting the development of preschool children’s emergent literacy skills: A randomized evaluation of a literacy-focused curriculum and two professional development models. Reading and Writing, 24, 305–337.

23.

Mashburn

A. J.

Pianta

R. C.

Hamre

B. K.

Downer

J. T.

Barbarin

O. A.

Bryant

. . . Howes

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

24.

McGee

L. M.

Richgels

D. J.

(2004). Literacy’s beginnings: Supporting young readers and writers (5th ed.). Boston, MA: Pearson.

25.

National Institute for Literacy. (2008). Developing early literacy: Report of the National Early Literacy Panel. Jessup, MD: Author.

26.

National Staff Development Council. (2001). NSDC standards for staff development. Retrieved from http://www.nsdc.org/standards/

27.

Neuman

S. B.

Cunningham

(2009). The impact of professional development and coaching on early language and literacy instructional practices. American Educational Research Journal, 46, 532–566.

28.

Neuman

S. B.

Wright

T. S.

(2010). Promoting language and literacy development for early childhood educators: A mixed-method study of coursework and coaching. Elementary School Journal, 111, 63–86.

29.

Pianta

R. C.

Howes

Burchinal

Bryant

Clifford

Early

Barbarin

(2005). Features of pre-kindergarten programs, classrooms, and teachers: Do they predict observed classroom quality and child-teacher interactions? Applied Developmental Science, 9, 144–159.

30.

Pianta

R. C.

La Paro

K. M.

Hamre

B. K.

(2008). Classroom assessment scoring system (CLASS) manual: Pre-K. Baltimore, MD: Brookes.

31.

Pianta

R. C.

Mashburn

A. J.

Downer

J. T.

Hamre

B. K.

Justice

(2008). Effects of web-mediated professional development resources on teacher-child interactions in pre-kindergarten classrooms. Early Childhood Research Quarterly, 23, 431–451.

32.

Powell

D. R.

Diamond

K. E.

(2011). Improving the outcomes of coaching-based professional development interventions. In Neuman

S. B.

Dickinson

D. K.

(Eds.), Handbook of early literacy research (Vol. 3, pp. 295–307). New York, NY: Guilford.

33.

Powell

D. R.

Diamond

K. E.

Burchinal

M. R.

Koehler

M. J.

(2010). Effects of an early literacy professional development intervention on Head Start teachers and children. Journal of Educational Psychology, 102, 299–312.

34.

Roskos

K. A.

Rosemary

C. A.

Varner

M. H.

(2006). Alignment in educator preparation for early and beginning literacy instruction: A state-level case example. In Zaslow

Martinez-Beck

(Eds.), Critical issues in early childhood professional development (pp. 255–282). Baltimore, MD: Brookes.

35.

Sheridan

S. M.

Edwards

C. P.

Marvin

C. A.

Knoche

L. L.

(2009). Professional development in early childhood programs: Process issues and research needs. Early Education and Development, 20, 377–401.

36.

Smith

M. W.

Dickinson

D. K.

(2002). Early language and literacy classroom observation toolkit (Research edition). Baltimore, MD: Brookes.

37.

Strickland

D. S.

Riley-Ayers

(2006). Early literacy: Policy and practice in the preschool years (Preschool Policy Brief Issue No. 10). New Brunswick, NJ: National Institute for Early Education Research.

38.

Ukrainetz

T. A.

Cooney

M. H.

Dyer

S. K.

Kysar

A. J.

Harris

T. J.

(2000). An investigation into teaching phonemic awareness through shared reading and writing. Early Childhood Research Quarterly, 15, 331–355.

39.

Wasik

B. A.

Bond

M. A.

(2001). Beyond the pages of a book: Interactive book reading and language development in preschool classrooms. Journal of Educational Psychology, 93, 243–250.

40.

Wasik

B. A.

Bond

M. A.

Hindman

(2006). The effects of a language and literacy intervention on Head Start children and teachers. Journal of Educational Psychology, 98, 63–74.

41.

Wasik

B. A.

Hindman

A. H.

(2011). Identifying critical components of effective preschool language and literacy coaching intervention. In Neuman

S. B.

Dickinson

D. K.

(Eds.), Handbook of early literacy research (Vol. 3, pp. 322–336). New York, NY: Guilford.

42.

Zaslow

M. J.

(2009). Strengthening the conceptualization of early childhood professional development initiatives and evaluations. Early Education and Development, 20, 527–536.

43.

Zaslow

Tout

Halle

Starr

(2011). Professional development for early childhood educators: Reviewing and revising conceptualizations. In Neuman

S. B.

Dickinson

D. K.

(Eds.), Handbook of early literacy research (Vol. 3, pp. 425–434). New York, NY: Guilford.