Enhancing preschool educators’ ability to facilitate conversations during shared book reading

Abstract

The purpose of this study was to investigate whether professional development enhanced educators’ use of conversational strategies during shared book reading with small groups of preschoolers. Twenty preschool educators and small groups of children from each of their classrooms were randomly assigned to the experimental or control group. The 10 educators in the experimental group received instruction in shared book reading strategies as well as individual classroom coaching sessions. Each educator was video-recorded reading two storybooks to a small group of preschoolers at pretest and posttest. The video-recordings were transcribed and coded to yield the measures of the book-related talk. The findings revealed that the educators in the experimental group used a greater number of open questions, responsive statements and different words compared to the control group. The educators and children in the experimental group also maintained longer book-related conversations and had more conversations that were five or more turns in length compared to the control group. These findings suggest that professional development that includes group instruction and individual coaching can enhance educators’ ability to facilitate book-related conversations with preschool-age children.

Keywords

conversation educator preschool professional development shared book reading

Introduction

Many preschool children, particularly those from low-income homes, lack the language skills necessary to support literacy development (Biemiller, 2006; Hart and Risley, 1995, 2003; Ramey and Ramey, 1998). One way in which preschool educators can enhance these children’s language learning is by promoting adult–child conversations in their classroom (Dickinson and Tabors, 2001; Massey, 2004). The extent to which educators ask questions and are linguistically responsive in classroom conversations is strongly and positively correlated with children’s cognitive, linguistic and social-emotional outcomes (Howes et al., 1992; Mashburn et al., 2008; NICHD, 2002; Peisner-Feinberg et al., 2001). One important classroom context for conversation is shared book reading. During this activity, adults engage children in conversation about the story by asking questions or making comments that invite the children’s response. These short conversations create opportunities for children to practice using vocabulary and syntax from the story. Moreover, educators’ responses provide children with feedback on their communicative attempts and models of more complex language (Crain-Thoreson and Dale, 1999; Doyle and Bramwell, 2006; Hargrave and Sénéchal, 2000; van Kleeck, 2008; Wasik and Bond, 2001).

Recent research reveals that there is wide variability in the language environment in preschools, specifically in the amount of responsive language directed to children (Connor et al., 2006; Dickinson and Tabors, 2001; Turnbull et al., 2009) and the opportunities for children to talk (Massey, 2004; Wasik et al., 2006). Turnbull et al. (2009) reported that across a variety of classroom contexts, only one-third of educators’ utterances were language facilitating. For example, they reported that educators asked less than one open question per minute in classrooms serving at-risk children (0.7 per minute). The current study investigated whether preschool educators’ participation in a professional development programme would improve their ability to facilitate conversations during shared book reading with small groups of preschool children from poor socio-economic backgrounds. This study focused on book reading as the context for conversations because it is a regular curricular activity in preschool classrooms and provides a naturalistic context for asking children questions and facilitating conversation.

Conversation

Conversations between adults and children during the preschool years play a vital role in language and literacy development (Chall et al., 1990; Connor et al., 2006; Dickinson and Tabors, 2002; Hart and Risley, 1995, 2003; Justice et al., 2005; Wasik et al., 2006). Longitudinal studies have shown that variations in the language environment during the preschool years have far reaching effects and are predictive of children’s language and conventional reading outcomes in the higher grades (Dickinson and Tabors, 2002; Hart and Risley, 1995). For example, Hart and Risley (1995) found that by 3 years of age, preschool children’s vocabulary scores significantly correlated with both the amount of parent talk in the home and the length of the parents’ utterances. Children who heard greater amounts of language input, that contained more diverse vocabulary, had more advanced language ability. Importantly, children from low-income families heard significantly less talk, less varied vocabulary, and had significantly poorer language ability by 3 years of age, as compared to children from middle and upper income homes. A follow-up study 7 years later revealed a significant positive relationship between the children’s language ability in preschool and their receptive vocabulary, their language development, and reading comprehension at grade four (Hart and Risley, 2003). Thus, many preschool children from low-income homes may not have the language ability necessary to form a sound basis for literacy success (Hart and Risley, 1995). High quality preschool programmes that provide a responsive language environment may offer a compensatory boost to the language learning of these children before they begin their formal schooling. This boost may enhance children’s oral language skills, such as vocabulary, and better position them to profit from emergent literacy interactions (Dickinson and Tabors, 2002).

Socio-cultural theory

The importance of adult–child conversation is underscored by Vygotsky’s socio-cultural theory of child development. According to this theory, language is the key social tool via which learning takes place as participants co-construct meaning through conversation (Hetherington et al., 2005). When children are regularly engaged in adult–child conversations they can acquire knowledge gradually through such interaction (Vygotsky, 1978). Preschool educators do this by scaffolding both the children’s participation in the conversation and their understanding of the topic of the moment. Scaffolding refers to an instructional strategy whereby educators initially expect children to participate or respond at lower levels of knowledge, skill, or confidence, and gradually increase their expectations of the children’s participation (Pentimonti and Justice, 2010; Wood et al., 1976). Skilled educators can scaffold a child’s language learning by asking questions in a way that elicits an answer within the children’s zone of proximal development, or at a level just beyond the child’s current ability. At lower levels of language ability, educators may ask children closed questions that are literal in nature, such as a request to label a familiar item (e.g. Educator: ‘What is this?’ Child: ‘A bird.’) (Blewitt et al., 2009). This facilitates children’s engagement in book-related talk at a low level of demand that is likely to elicit a correct response, thus providing an opportunity for the educator to give positive feedback. One possible response might be for the educator to acknowledge the child’s contribution and expand on it by adding syntactic and semantic content (e.g. Educator: ‘Yes, that is a bird. It’s called a robin.’). The children’s participation in conversational turn-taking related to the book enables the educator to stimulate additional book-related talk at higher levels of linguistic and cognitive demand. For example, the educator can scaffold a higher level of response by following up with an open question (e.g. Educator: ‘What do you think robins like to eat?’). By expanding on the child’s utterance and scaffolding with an open question, the educator can facilitate the development of the child’s vocabulary and oral language (Hargrave and Sénéchal, 2000; Wasik and Bond, 2001). This active participation in conversations enables children to construct meaning, thus contributing significantly to their own cognitive and linguistic development (Rogoff, 1990). In accordance with a socio-cultural theoretical perspective, preschool educators who scaffold children’s language use and participation in conversation may promote or even accelerate children’s developmental progress.

The current study focuses on two strategies educators use to engage children in book-related conversations, namely, questions and contingent responses. When educators ask questions, they invite children to take a conversational turn, thus facilitating their engagement in dialogue (deRivera et al., 2005). Closed questions, such as invitations to label (e.g. ‘What is this?’), have been found to elicit constrained one- or two-word answers (deRivera et al., 2005). In comparison, open questions (e.g. ‘What do you think will happen next?’) are less constrained by the asker, typically generate multi-word responses (deRivera et al., 2005; Wasik and Bond, 2001; Wasik et al., 2006), and may provide opportunities for more linguistically and cognitively challenging talk (Massey et al., 2008). Educators’ contingent responses, including comments, expansions, imitations and acknowledgements, provide children with linguistic information related to a joint focus of attention (e.g. a picture in a book) or the topic of a story (Girolametto and Weitzman, 2002; Turnbull et al., 2009). Furthermore, educators have the opportunity to expand or elaborate on the children’s utterances by adding linguistic and cognitive content (Nelson et al., 1996). Additionally, when educators incorporate the children’s utterances into the conversation, they allow the children to influence the direction of the discussion, potentially enhancing their engagement (Culatta et al., 2010). This may increase the likelihood of children taking successive turns related to the ongoing topic of conversation, resulting in longer conversations (Crain-Thoreson and Dale, 1999; Girolametto and Weitzman, 2002; van Kleeck, 2008). Studies have shown that educators’ contingent responsiveness positively correlates with children’s increased expressive language (Dickinson and Tabors, 2001; Girolametto and Weitzman, 2002; Snow et al., 1998). Additionally, educator–child participation in longer conversations has been shown to correlate positively with enhanced language outcomes (Zimmerman et al., 2009). The National Early Literacy Panel’s meta-analysis found that shared book reading was an effective strategy for use with preschool children, yielding moderate effects for oral language skills (NELP, 2008).

Shared book reading

Shared book reading is an interactive method of reading books aloud to children during which the adult encourages the children’s engagement in book-related conversation (Blewitt et al., 2008; Crain-Thoreson and Dale, 1999; Doyle and Bramwell, 2006; Hargrave and Sénéchal, 2000; van Kleeck, 2008; Wasik and Bond, 2001). Children’s active involvement in shared book reading has been shown to be positively associated with a number of areas of child language development, including vocabulary (Gerde and Powell, 2009; Justice et al., 2005; Hargrave and Sénéchal, 2000; Lonigan and Whitehurst, 1998; Roberts et al., 2005; Wasik and Bond, 2001), expressive language (Crain-Thoreson and Dale, 1999), oral narrative skills (Lever and Sénéchal, 2011; Zevenbergen et al., 2003) and story comprehension (Dickinson and Smith, 1994). Further, studies report enhanced language learning outcomes for preschool children from poor socio-economic backgrounds (NELP, 2008) and for children learning English as a second language (Collins, 2005). However, while many educators use shared book reading strategies effectively, studies report considerable variation in educators’ ability to engage children in extended book-related conversations (Dickinson and Tabors, 2001; Girolametto et al., 2006; Massey et al., 2008; Zucker et al., 2010). Gerde and Powell (2009) reported that among 60 Head Start teachers, some asked no questions within a single book reading while others asked as many as 62 questions (M = 14, SD = 8.2, range = 0–62). Furthermore, several studies have reported that educators used a style of reading books that was instructional in nature (Hindman et al., 2008), using talk that was intended to elicit specific responses (Dickinson et al., 2002), manage children’s behaviour (Wasik et al., 2006) or control the topic by dominating the turn-taking (Girolametto et al., 2000). Other studies have reported that the majority of educators’ extra-textual talk during book reading is of low cognitive demand, requiring minimal responses from the children (Girolametto et al., 2007; Dickinson et al., 2002). Finally, many educators do not use shared reading to promote vocabulary learning, particularly when reading to children considered at risk of poor language outcomes (Biemiller and Slonim, 2001). In order to enhance the language learning opportunities for children in preschool classrooms, it is important to understand whether professional development can enhance educators’ ability to facilitate conversation and word learning (Dickinson and McCabe, 2001).

A number of studies have examined the effectiveness of shared book reading interventions in preschool classrooms and reported improved outcomes for standardized measures of child language outcomes for the children in the experimental group (Blewitt et al., 2009; Justice et al., 2005; Lever and Sénéchal, 2011; van Kleeck et al., 2006). However, few experimental studies have reported on the effects of professional development instruction on preschool educators’ use of strategies to engage children in shared reading (e.g. Girolametto et al., 2007; Wasik et al., 2006). Girolametto and his colleagues (2007) reported on the effects of a 2-day professional development programme that instructed educators in strategies to promote preschool children’s language development and the use of interactive book reading. At posttest, educators in the experimental group used more questions and comments that invited children to talk about the story and compared to a control group that received instruction in facilitating peer interaction. These changes in the educators’ talk elicited more talk from the children about emotions and the children’s experiences related to the story compared to the controls. Wasik et al. (2006) trained 10 Head Start educators to promote book-related conversation (e.g. ask open questions, promote vocabulary learning, and connect the story to children’s experiences). This professional development programme was delivered over 9 months and included a monthly instructional workshop and individualized coaching. Relative to a control group, the educators in the experimental group asked significantly more open questions during book reading and this result positively correlated with the children’s receptive language scores. Taken together, the results of both Girolametto et al. (2007) and Wasik et al. (2006) indicate that the educators made significant behavioral changes in the way they talked to the children during shared book reading following professional development. However, additional work is needed to identify the effects of professional development on educators’ use of questions, responsive language and length of conversations, as measured by the number of conversational turns related to a single topic. Additionally, further research is needed to explore the effects of professional development on educators’ use of explicit strategies to promote word learning during shared book reading.

The first objective of this study was to determine whether professional development would result in an increase in the educators’ rate of asking questions and responsive statements during shared book reading. It was predicted that educators would use more open questions and responsive statements based on the results of previous research (Wasik et al., 2006). The second objective of this study was to determine if educators in the experimental group differed in their use of strategies to promote word learning. It was predicted that educators would use a greater number of different words in their extra-textual conversation and that they would use more explicit strategies to promote word learning. This hypothesis was based on previous work demonstrating that educators used more diverse vocabulary (Wasik and Bond, 2001) and more explicit strategies to promote word learning (Wasik et al., 2006) following intervention. Finally, the third objective of this study was to identify whether educators and children would have more and longer book-related conversations during shared storybook reading following professional development. It was predicted that educators in the experimental group would engage children in more frequent and longer conversations relative to controls based on previous research, thus indicating increased child engagement in conversation when adults use more questions and contingent responses (Crain-Thoreson and Dale, 1999; Girolametto and Weitzman, 2002).

Method

Design

This experimental study used a pretest–posttest design with random assignment of preschool educators to the experimental and control groups. All educators were videotaped reading two narrative storybooks to a small group of three to four preschool children at both test times, separated by 8 weeks. The educators in the experimental group participated in an 18-hour emergent literacy professional development programme during the autumn of the preschool year. The programme consisted of four workshops (three full days and one half day) and three individual coaching visits to the educators’ classrooms between each workshop. The educators in the control group participated in all pretest and posttest sessions and remained on an active wait list to receive the professional development programme following the posttest data collection period.

Participants

The educators were randomly assigned to the experimental and control groups using a random numbers table. All educators completed a demographic questionnaire (e.g. age, years of study, experience). Table 1 displays descriptive data on the pretest characteristics of the preschool educators. At pretest, there were no significant differences between the experimental and control groups in terms of the educators’ age, years of education, years of experience or number of children in their classrooms, t(17 − 18) = −1.47 to 0.06, p = 0.16–0.95. Four educators, two in each group, had learned English as a second language and spoke a different language at home. However, all educators had completed their academic qualifications in early childhood education in English and spoke English fluently. None of the educators reported participating in previous professional development training on how to facilitate emergent literacy skills.

Table 1.

Pretest characteristics of the early childhood educators

Variable	Experimental group (n = 10)	Control group (n = 10)
Age (in years):
M (SD)	39.7 (10.3)	39.4 (9.2)
Min–Max	23–58	24–54
Years of education:
M (SD)	3.0 (1.3)	2.9 (1.1)
Min–Max	2–5	2–5
Years of experience:
M (SD)	14.1 (6.0)	18.5 (7.4)
Min–Max	2–25	6–26
Class size (# children)^a:
M (SD)	15.0 (6.0)	18.0 (6.3)
Min–Max	8–24	5–24

All classrooms maintained an adult–child ratio of one educator for every eight children.

The educators also completed the Educator Early Literacy Questionnaire (adapted from Boudreau, 2008) with regard to their classroom book reading practices. This questionnaire asked the educators to rate how often they read to children in small groups and how many books they typically read in one sitting. Across all 20 educators, the median rating of how often the educators read to small groups of children was 4.0 (i.e. daily) and the mean number of books read in one sitting was 1.9 books (ranging from 1.0 to 3.5 books). These responses indicated that the educators had regular experience of reading to small groups of preschool children and typically read two books in one sitting.

Children. A total of 76 preschool children participated in this study. The children were recruited from each participating educator’s classroom. Children were eligible to participate if they were from low-income homes, as determined by child care subsidy information. Supervisors who had access to subsidy data nominated all of the children in the educator’s classroom who were receiving subsidy. Consent and information forms were given to all eligible families. The participating children were randomly selected from those who returned signed consent forms, balancing for an equivalent number of boys and girls wherever possible. In some of the participating centres, information on subsidy was not available. In these cases, consent forms were sent home to all families and four children were randomly selected from those children whose families returned signed consents. The group size was set at four children, based on previous research reporting that young children are more interactive in small rather than large group settings (McCabe et al., 1996; Wasik, 2008). In some cases, where only three families had signed consent forms, the group size included the three available children (i.e. four groups had three children each; one in the experimental and three in the control condition).

Once consent to participate in the study was obtained from the families, parents were asked to complete a child and family demographic questionnaire. Additionally, all educators were asked to complete a short questionnaire consisting of four items from the Speech and Language Assessment Scale (SLAS; Hadley and Rice, 1993). This provided baseline information on each child’s language development. Table 2 provides summary data of the characteristics of the children. The average age of the children was approximately 55 months (ranging from 44 to 69 months). Approximately 75% of the children in both groups were receiving subsidy, indicating that they came from low-income homes. Consistent with large urban settings, 44% of the children in the experimental group and 51% of the children in the control group spoke a language other than English at home (i.e. Amharic, Cantonese, Dari, Farsi, French, Japanese, Mandarin, Polish, Portuguese, Romanian, Spanish, Tagalog, Tibetan, Tigrina, Twi or Yoruba). There were no significant differences between the two groups of children in terms of their age in months, t(74) = −1.10, p = 0.274, the length of time they had attended child care, t(71) = 0.76, p = 0.448, the proportion of children who spoke another language at home, χ(1) = 0.46, p = 0.646, or the proportion of children who were on subsidy, χ(1) = 2.943, p = 0.086. Also, there were no significant differences between the two groups in terms of their language ability as determined by the educators’ ratings of language skills on the Speech and Language Assessment Scale (SLA; Hadley and Rice, 1993), t(74) = −0.27, p = 0.788. The data in Table 2 indicate that the children’s language skills were, on average, within normal limits (i.e. a rating of 3 = normal for age). Further, an independent samples t-test indicated that there were no significant differences in the educators’ ratings of the boys’ and girls’ scores on the SLAS scale (i.e. experimental group, t(37) = −0.752, p = 0.457, and control group, t(35) = −1.622, p = 0.114).

Table 2.

Pretest characteristics of the children in the experimental and control groups

Variable	Experimental group (n = 39)	Control group (n = 37)
Gender
Male	14	20
Female	25	17
Age (months)
M (SD)	54.2 (6.8)	55.7 (5.0)
Min–Max	44–68	47–69
SLAS score^a
M (SD)	3.8 (0.7)	3.6 (0.6)
Min–Max	2.4–4.9	2.5–4.9
Time in child care^b
M (SD)	27.3 (14.5)	24.3 (18.5)
Min–Max	1–57	1–60
Exposure to another language
Yes	17	19
No	22	18
Subsidy
Yes	30	28
No	9	9

Speech and Language Assessment Scale (a score of 3 = normal for age; 5 = beyond normal).

Length of time in months that child has attended the child care centre.

Procedure

The same testing procedures were used before and after the professional development programme. A research assistant travelled to the child care centre and videotaped educator–child interaction during small group book reading. The children were seated on the floor in the designated book centre, where book reading typically took place in the classroom. The instructions given to the educators were: ‘The purpose of this videotape is to observe how children communicate during a typical storybook reading activity. Please read these two books to the children the way you usually read to them.’ The research assistant provided the educator with two unfamiliar storybooks, Little Yellow Dog Gets a Shock (Simon, 2003) and Don’t Forget to Come Back (Harris, 1978). Both books followed a narrative sequence (e.g. beginning, middle, end, with a problem and a resolution). A previous feasibility study indicated that educators were able to read the two books comfortably within this time period (Girolametto et al., 2007). The groups were videotaped for approximately 15 minutes of the storybook reading using a portable video-camera with a directional microphone.

Immediately following the videotaping, the early childhood educators rated the representativeness of their interactions using a 5-point scale (1 = very typical; 3 = typical; 5 = not typical). These ratings were made in order to obtain the educators’ perceptions of how typical the interactions were and to determine whether the videotapes were typical or not from the educators’ point of view. This procedure has been used in previous studies of educator–child interaction (Girolametto et al., 2007). At pretest, educators rated their amount of talk and rate of speech as typical (mean amount of talk = 3.0 and mean rate of speech = 3.2). In addition, the educators rated their comfort level during the book reading as typical of unobserved interaction (mean rating = 3.2). Similar ratings were obtained at posttest (mean amount of talk, 2.7; mean rate of speech, 3.1; and mean comfort level, 3.4). These ratings provide some assurance that the educators believed their interactions to be similar to other unobserved small group book reading sessions in the child care centre.

Professional Development Programme. The professional development programme utilized in this study was (ABC and Beyond™: The Hanen Program® for Building Emergent Literacy in Early Childhood Settings (Weitzman and Greenberg, 2010)). This training included three full day and one half day workshops and three individual classroom visits. All four workshops were taught by a speech-language pathologist who had extensive experience of consulting with educators in early childhood settings. The teaching methods used in all four workshops were similar and included: (a) a review of the previous week’s workshop content, (b) interactive lectures with examples and videotapes selected to illustrate key strategies, (c) small group discussions to analyze videotaped examples or discuss ideas for strategy implementation in the classroom, (d) role-plays of strategy implementation, and (e) completion of action plans, in which educators wrote out the strategies they intended to use in the activity that would be videotaped during the following classroom coaching visit.

The first 5-hour workshop included strategies for facilitating conversation during storybook reading. These strategies included: promoting conversation related to the book content, elaborating on word meaning, following the children’s lead, and engaging with all the children in the group. The professional development programme included instruction in strategies to instigate longer book-related conversations, one referred to as ‘Strive for five’ (Dickinson, 2003; Weitzman and Greenberg, 2010). This strategy encouraged educators to maintain and deepen a topic of conversation for five or more conversational turns by different speakers within the group. One conversational turn was defined as all consecutive utterances of the same speaker. For example, if the educator expressed six consecutive utterances this would constitute one turn. To promote word learning through conversation, the educators were taught to isolate words that are important for the children’s comprehension of the story, to elaborate on the meaning of those words, and to relate those words to the events in the story or to the children’s personal experience. The specific strategies were: (1) select and stress words that are rare or important for comprehension of the story, (2) explain those words using a definition or synonym, (3) relate the words to the events in the story or the children’s real world experience, and (4) repeat those words in conversation during the book reading. The second 5-hour workshop focused on the use of language or talk that went beyond the text, e.g. to include children’s experiences or make predictions. The third 5-hour workshop taught educators how to use print and sound references to build print concepts, alphabet knowledge and sound awareness. Educators were taught to use both storybook reading and post-story writing activities as a context for modelling this information. The final 3-hour workshop was a half day review. Between each workshop, educators received individual classroom visits from a speech language pathologist (a total of three visits) that provided them with an opportunity to individualize the workshop content to their classrooms and receive feedback on their use of programme strategies.

Classroom coaching sessions were conducted by one of two speech-language pathologists who adhered to a consistent protocol during these one hour classroom visits. Each visit included: (a) videotaping the educator interacting with a small group of children in the classroom, (b) coaching the educator, as needed, during the videotaping session, and (c) reviewing the videotape and discussing the use of strategies taught in the previous workshop. The videotapes were only used to assist learning during the professional development programme and were not used for outcome measures. Prior to videotaping, the speech-language pathologist reviewed the educator’s planned activity for the videotaped interaction and commented, if appropriate. For example, following the workshop on promoting conversation, the educator and speech-language pathologist might discuss how to use a selected children’s book to ask open questions to engage children in more cognitively challenging conversation (e.g. ‘What do you think will happen next?’, ‘Can you think of a time when you felt like that?’ or ‘Why do you think he is saying that?’). During videotaping, the speech-language pathologist paused the camera from time to time to make suggestions for engaging the children in the conversation, and then resumed videotaping. For example, if an educator was observed to ask questions that focused primarily on the text, the speech-language pathologist might coach the educator to ask one or two specific prediction questions. Additional information related to professional development can be found in (Girolametto et al., 2012).

Treatment Fidelity. In order to ensure treatment fidelity was maintained during the delivery of this professional development programme, three strategies were used (Kaderavek and Justice, 2010). First, each educator received a programme manual ABC and Beyond™: The Hanen Program® for Building Emergent Literacy in Early Childhood Settings (Weitzman and Greenberg, 2010) that provided information specific to the content of each workshop of the programme, as well as several examples of related strategies. Second, data were collected on the educators’ attendance at the four workshops and three classroom visits. Eight of the 10 educators attended all four workshops; however, two educators in the experimental group had an unavoidable scheduling conflict for one workshop session and the programme leader provided these two educators with an individual session that covered the missed content. All 10 educators in the experimental group participated in the three classroom visits and all visits adhered to the established protocol. Finally, the videotaped interactions that were filmed during the three individual classroom visits provided additional information about the extent to which the educators’ used programme strategies while being observed. The coaches completed checklists for each educator for each of the three classroom visits. A frequency response rating scale was used to determine if an educator used a programme strategy consistently (score of 2), sometimes (score of 1) or not at all (score of 0). These ratings were applied to the interaction prior to coaching (i.e. if an educator required coaching to use a strategy consistently, it received a rating of 0 or 1). The total number of points possible for the three sessions was 16, 14 or 6. The mean fidelity ratings for the three classroom visits were 14.7/16, 13.2/14 and 5.5/6. These average ratings indicated a high frequency of use of the programme strategies during classroom activities that were observed and coached by a speech-language pathologist.

Measures

The pretest and posttest videotapes of educator–child shared book reading were transcribed and coded by the primary author of this study, who was blind to the group assignments of the participants and times of the recordings (i.e. pretest and posttest). All shared book reading interactions were transcribed using Transcript Builder (Moore, 2000) and Systematic Analysis of Language Transcripts (SALT; Miller and Chapman, 2002). A trained research assistant, who achieved 90% accuracy in transcription on three training videotapes, conducted agreement reliability on 100% of the transcripts. Following a consensus procedure used by Johnston (2001), the research assistant viewed the videotapes while reading the completed transcripts, entering queries and corrections directly onto transcripts. The primary investigator and the research assistant then met to resolve all discrepancies. Where a discrepancy could not be resolved, an X was entered on the transcript. Agreement reliability was calculated using the following formula: number of agreements / (the number agreements + disagreements) × 100 (Sackett, 1978) and yielded 99.9% for utterance boundaries, 99.6% for words and 99.5% for allocation of speakers. Overall agreement was 99.7%. Agreement reliability indicated the extent to which the verifier agreed with the original transcription prior to making any changes to the transcripts.

Coding System. The 40 transcripts (20 pretest and 20 posttest) were coded by the first author who was blinded to group assignment and test time. Coding of transcripts was conducted in random order using a random numbers table. First, the transcripts were coded at the utterance level. An utterance was defined as the words spoken by a speaker that were not interrupted by an extended pause, terminal intonation contour or another speaker’s utterance. Appendix A displays the coding system for utterances, with examples. The codes included open questions, closed questions, comments, expansions, imitations and acknowledgments. Utterances that did not fit the definitions for these codes were classified as Other. These codes were applied to the educators’ utterances. Twenty per cent of the transcripts were randomly selected and coded independently by the second author of this study to provide reliability estimates for each of the codes. Inter-rater reliability was calculated using the formula: number of agreements / (the number agreements + disagreements) × 100 (Sackett, 1978). The reliability for the codes was: 85% for open questions, 95% for closed questions, 96% for comments, 88% for expansions, 85% for imitations, 87% for commands and 91% for acknowledgements. Overall inter-rater reliability for the coding system was 91%.

A second pass through the transcripts identified specific vocabulary teaching strategies used by the educators for 10 words that were preselected, five from each of the two storybooks. The 10 words were: shock, share, leap (or leapt), snuggle, friend, babysitter, South Pole, moose, thunderstorm and scare. SALT’s Explore feature was used to retrieve a list of all of the utterances in the transcripts that included each of those 10 words, including all possible morphological variations of them. In order to determine the context of use of a word, three utterances before and five utterances after the utterances that included each word were considered. One point was awarded for the use of a vocabulary teaching strategy (i.e. stress, explain, relate or repeat) for each of the 10 words. Appendix B includes a definition of each strategy, with examples. A total score for the use of word learning strategies was generated for each educator for the 10 selected words at pretest and posttest.

A third pass of the transcripts identified the turn length of all conversations. A conversation was defined as a sequence of utterances on a book-related topic that consisted of two or more conversational turns by two or more speakers. A conversational turn included all consecutive utterances by one speaker. A conversation ended when the educator began reading the text or when there were interruptions to the book-related talk, such as behavior management or off-topic talk. Twenty per cent of the transcripts were randomly selected and coded independently by a second coder to provide inter-rater reliability for number and length of conversations, which was 90% and 91%, respectively.

Outcome Measures. The frequency of each code was calculated from the transcripts using SALT. A variable labelled responsive comments was created by summing comments, imitations, expansions and acknowledgements. The frequency of each code was divided by the elapsed time of each reading session to yield the rate per minute of each code. This was done because the length of book reading time varied across educators. In addition, SALT analysis was used to calculate the number of different words used by each educator. Further, by using the Explore function of SALT, each occurrence of the 10 target vocabulary words was located in the transcripts along with three utterances before the utterance and five utterances following it. This made it possible to identify the use of each word in context. The use of these target words was scored according to the four vocabulary instructional strategies taught in the workshops (i.e. select/stress, explain, relate and repeat). Finally, the transcripts were coded to identify the number of conversations that were five turns or longer to yield the frequency of ‘strive for five’ conversations.

Results

The results are presented in three sections. The first section examines the differences between the two groups of educators on shared book reading codes. The second section reports on group differences in educators’ use of different words and vocabulary teaching strategies. Finally, the third section examines group differences in the number and length of book-related conversations that occurred between the educators and their small groups of children.

In order to compare groups, a series of one-way repeated-measures analyses of variance was used with group (experimental, control) as the between-subjects factor and time (i.e. pretest, posttest) as the within-subjects factors. The two-tailed alpha was set at 0.05. Partial eta squared (ηp²) was used to estimate the effect sizes and values were interpreted using Cohen’s (1992) guidelines for judging the importance of eta squared (0.02 = small; 0.13 = medium; 0.26 = large). These guidelines are widely used in many educational and psychological studies of behaviour (e.g. Bakeman, 2005).

The first question of this study examined whether the educators in the experimental and control groups differed in terms of their rate of open questions, closed questions and responsive comments. Table 3 presents the summary data for the educators’ rate of open questions, closed questions and responsive comments. A repeated measures ANOVA revealed that there was a significant Group × Time interaction for the rate of open-ended questions, F(1,18) = 4.419, p = 0.05, partial eta squared (η_p²) = 0.197. This effect size was medium. An examination of the data in Table 3 indicates that educators in the experimental group increased in their rate of open questions from 1.35 per minute to 1.64, whereas the control group decreased their use from 1.50 to 1.04 per minute. There was no main effect of time for this variable, F(1,18) = 0.208, p = 0.654, η_p²= 0.011. The second repeated measures ANOVA yielded a significant Group × Time interaction for the rate of closed questions, F(1,18) = 8.934, p = 0.008, η_p²= 0.332. This effect size was large. An examination of the data in Table 3 indicates that, contrary to expectations, educators in the experimental group increased their use of closed questions from 2.80 per minute to 3.80, whereas the control group decreased their use of closed questions from 3.03 per minute to 2.38. There was no main effect for time for closed questions, F(1,18) = 0.407, p = 0.531, η_p²= 0.022. The third repeated measures ANOVA examined the rate of responsive comments. There was a significant Group × Time interaction for this variable, F(1,18) = 12.061, p = 0.003, η_p²= 0.401, with a large effect size. The data in Table 3 indicate that educators in the experimental group increased their use of responsive comments from 4.76 per minute at pretest to 7.27 at posttest, whereas the control group decreased their use of responsive comments from 4.87 to 4.60 per minute. The main effect for time for responsive comments was also significant, indicating that both groups increased their use of responsive comments over time, F(1,18) = 8.366, p = 0.010, η_p²= 0.317. In summary, the educators in the experimental group differed significantly from the control group on two qualitative indices of language input (i.e. questions and responsive statements) following participation in the professional development programme.

Table 3.

Means and standard deviations for measures of book-related talk

Variable	Experimental group¹ Mean (SD) (n = 10)	Control group^a Mean (SD) (n = 10)	p-values for Group × Time interactions
Educators’ rate^b of:
Closed questions
Pre	2.80 (1.5)	3.03 (1.8)	0.008
Post	3.80 (1.0)	2.38 (1.0)	0.008
Open questions
Pre	1.35 (1.0)	1.50 (0.5)	0.05
Post	1.64 (0.8)	1.04 (0.3)	0.05
Responsive Comments
Pre	4.76 (1.6)	4.87 (2.0)	0.003
Post	7.27 (2.2)	4.6 (1.2)	0.003
Overall book-related talk
Pre	11.05 (4.7)	11.63 (4.5)	0.002
Post	15.41 (3.5)	10.36 (3.3)	0.002

Note: Pre = pretest; Post = posttest.

The experimental group included 10 educators and 39 children; the control group included 10 educators and 37 children.

Rate was used because 11 transcripts were less than 15 minutes in length (five at pretest and six at posttest).

The second question of this study looked at the lexical diversity of the educators’ utterances and their use of explicit strategies to promote word learning. Table 4 presents the summary data for the educators’ rate per minute of different words spoken during extra-textual book-related conversations. A repeated measures ANOVA on the educators’ rate of different words revealed that there was a significant Group × Time interaction for this variable, F(1,18) = 18.019, p = 0.001, η_p²= 0.500. The data in Table 4 indicate that the educators in the experimental group increased their rate of different words from 13.09 at pretest to 16.41 at posttest. In comparison, the educators in the control group used 14.57 different words per minute at pretest and 13.66 different words per minute at posttest. Moreover, there was a main effect for time for this variable F(1,18) = 5.834, p = 0.027, η_p²= 0.245, indicating that educators in both groups increased their use of different words between pretest and posttest. A second repeated measures ANOVA examined the educators' use of strategies to teach vocabulary (i.e. stress, explain, relate, repeat). The results revealed that there was no significant Group × Time interaction for this variable, F(1,18) = 0.516, p = 0.482, η_p²= 0.028, nor was there a main effect for time, F(1,18) = 2.148, p = 0.160, η_p²= 0.107. An examination of the data in Table 4 indicates that educators in the experimental group decreased their use of explicit strategies to promote word learning from 14.20 to 10.40, while the control group educators used, on average, 15.5 word teaching strategies at pretest and 14.20 at posttest.

Table 4.

Means and standard deviations for rate of different words spoken and vocabulary teaching strategies

Variable	Experimental group Mean (SD) (n = 10)	Control group Mean (SD) (n = 10)	p-value for Group × Time interaction
Educator’s number of different words^a
Pre	13.09 (3.8)	14.57 (3.0)	0.001
Post	16.41 (2.4)	13.66 (2.7)	0.001
Educator’s vocabulary Teaching strategies^b
Pre	14.20 (9.2)	15.50 (6.6)	0.482
Post	10.40 (5.9)	14.20 (10.5)	0.482

Note: Pre = pretest; Post = posttest.

Based on rate per minute.

Maximum of 40 points.

The third question of this study investigated the impact of the professional development programme on the educators’ and children’s joint participation in conversations. This measure of conversation examined the contributions of both parties to extending the conversations. Due to unequal variance in the data for number and length of conversations, independent samples Mann–Whitney U tests were conducted. Table 5 presents the summary data for these variables. There was no significant difference in the number of book-related conversations that occurred in the experimental and control group, either before, U = 40.50, z = −0.720, ns, r = −0.16, or after the educators’ participation in the professional development programme, U = 32.50, z = −1.33, ns, r = −0.30. However, the groups differed significantly from each other in the mean length of conversations and the number of conversations that were longer than five turns. At pretest, the two groups did not differ in the mean length of book-related conversations, U = 35.00, z = −1.13, ns, r = −0.25, or the number of conversations that were five turns or longer, U = 42.00, z = −0.606, ns, r = −0.14. However, at posttest, the experimental group had significantly longer conversations, U = 18.00, z = −2.42, p = .015, r = −0.54, with a large effect size. In addition, the number of conversations that were five turns or longer significantly differentiated the two groups, U = 15.50, z = −2.62, p = 0.009, r = −0.59. This effect size is large.

Table 5.

Mann–Whitney U-test summary data for the educators’ and children’s conversations during shared book reading

Variable	Experimental group¹ Median	Control group^a Median
Number of conversations
Pre	26.00	22.50	U = 40.50, z = −0.720, ns, r = −0.16
Post	26.00	22.50	U = 32.50, z = −1.33, ns, r = −0.30
Mean length of conversations
Pre	7.31	6.65	U = 35.00, z = −1.13, ns, r = −0.25.
Post	9.21	6.82	U = 18.00, z = −2.42, p = 0.015, r = −0.54.
Number of conversations greater than or equal to 5 turns
Pre	15.50	12.00	U = 42.00, z = −0.606, ns, r = −0.14.
Post	16.00	12.00	U = 15.50, z = −2.62, p = 0.009, r = −0.59.

Note: Pre = pretest; Post = posttest.

The experimental group included 10 educators and 39 children; the control group included 10 educators and 37 children.

Post hoc analysis

A post hoc analysis was conducted to explore the nature of the experimental group’s longer book-related conversations and determine if there was a particular focus or theme of the conversation that was associated with longer conversations during shared book reading with preschool children. The content of each book-related conversation was coded as either literal or inferential in nature, and also whether it was related to the text or if it included talk that was related to the children’s or educator’s personal experience. This resulted in four codes that were applied to all book-related conversations in each transcript: (a) literal text, (b) literal personal, (c) inferential text, and (d) inferential personal. A summary of the coding system can be found in Appendix C. Inter-rater reliability was conducted on 20% of the transcripts, resulting in 87% of inter-rater agreement. It was hypothesized that there would be a significant positive correlation between inferential personal conversations and the length of conversations, because educators received instruction in strategies that promote inferential talk (Weitzman and Greenberg, 2010). Further, Girolametto et al. (2007) reported that educators used significantly more inferential utterances related to the children’s personal experiences following professional development. Pearson product-moment correlations were used to assess the relationship between the four themes of conversations (i.e. literal-text, literal-personal, inferential-text and inferential-personal) and the number of conversations that were five turns or longer. The results indicated that there was a positive correlation between the number of inferential-text conversations and the number of long conversations, r(9) = 0.643, p = 0.045. Thus, longer conversations were associated with inferential-text conversations that focused children’s attention on the book, while asking them to predict or infer information that was not in the text.

Discussion

The current study yielded three important findings. The first major finding was that educators in the experimental group adopted a conversational interactive style characterized by questions and responsive statements during shared book reading. The second major finding of this study related to word use. Following professional development, the educators used a greater rate of different words in their book-related talk. Third, these behavioral changes in the educators’ extra-textual talk following professional development resulted in a higher mean length of conversation and more conversations that were at least five turns long. A post hoc examination of these data revealed that the increase in length of conversations was associated with inferential-text conversations.

Following participation in this professional development programme, the educators in the experimental group used significantly higher rates of open questions and responsive comments at posttest compared to the control group. Theoretically, this is an important finding, given the facilitative nature of both these utterance types for eliciting children’s book-related talk and maintaining children’s engagement. Open questions have been found to elicit longer child responses and more complex language use (Dickinson, 2003; de Rivera et al., 2005). Educators’ use of open questions has also been associated with improved child outcomes in oral language skills (Mol et al., 2008) and educator–child talk that is more cognitively challenging in nature (Zucker et al., 2010). Preschool educators’ linguistic responsiveness (i.e. contingent comments, imitations, expansions, acknowledgements) facilitates children’s talk and paying joint attention to a single topic (Girolametto and Weitzman, 2002; Turnbull et al., 2009). Further, educators’ responsive comments may expand or extend the topic, providing the children with linguistic models via which they can learn more complex language (Culatta et al., 2010). Importantly, studies have shown that educators’ responsiveness positively correlates with enhanced child language outcomes (Dickinson and Tabors, 2001; Girolametto and Weitzman, 2002; Snow et al., 1998). Although descriptive studies have reported on educators’ use of responsiveness (Girolametto and Weitzman, 2002), there are no known experimental studies that have demonstrated group differences in the educators’ use of responsive statements in shared book reading following professional development. Thus, this positive finding reveals that educators may increase these key features of language input that are facilitative of children’s language development.

This increase in the educators’ use of open questions and responsive comments during shared book reading may reflect an important shift from an instructional practice to a group discovery experience (Blank et al., 1978). When educators invite children to participate in book-related talk by asking open or unconstrained questions and then respond contingently to their utterances, they allow children to influence the direction of the discussion (Culatta et al., 2010). Book reading experience has the potential to become a dynamic interaction between adult and children, to the extent that they each contribute to the ensuing talk (Pardo, 2004; Price et al., 2009). Theoretically, these findings reflect basic tenets of the socio-cultural theory of child development (Vygotsky, 1978). For example, the findings suggest that these educators may be creating a culture of learning by valuing and reinforcing the children’s active participation as co-constructors in their own learning (Rogoff, 1990).

Contrary to our expectation, the educators in the experimental group asked significantly more closed questions at posttest compared to the control group. Given the constrained nature of these questions and their tendency to elicit minimal child responses (de Rivera et al., 2007) this was a surprising finding. However, a post hoc examination of the data revealed that the observed increase in closed questions was proportional to the educators’ overall increase in extra-textual talk. That is, both their rate of closed questions and their overall rate of talk increased by 33% relative to pretest. As a result, the educators in the experimental group maintained a consistent proportion of low demand or constrained questions. Blank et al. (1978) recommended that preschool educators aim for 70% of their questions in conversation to be literal (i.e. low demand) in order to provide children with sufficient opportunities to understand the conversation and actively participate in it. An examination of the literature indicates that approximately 66% of adults’ questions during book reading are closed questions (de Rivera et al., 2005; van Kleeck et al., 1997). It is possible that educators use closed questions to accommodate children with a variety of language learning needs. Pentimonti and Justice (2010) reported on educators’ use of a range of scaffolds that engage children in book-related conversation, e.g. first ask low-demand questions that provide opportunities for quieter children or those with lower levels of language to talk and later ask a greater number of high demand or open questions. In the current study, the educators’ use of closed questions may have played a role in scaffolding children’s ability to contribute to the conversation, and this resulted in longer conversations.

The second major finding of the study relates to the impact of the educators’ participation in professional development; this resulted in their use of more diverse vocabulary during shared book reading conversations. This finding suggests that these educators provided a richer linguistic environment for the children. Theoretically, this is an important outcome, as shared book reading has been shown to increase children’s incidental exposure to novel words (Sénéchal and Cornell, 1993). Vygotsky (1978) suggested that educators promote children’s implicit learning of words and word meanings when they use a highly diverse vocabulary. In doing so, preschool educators have the potential to model words that are common in the children’s language environment as well as words that are more complex.

Explicit instruction that defines and elaborates on word meaning can also be an important strategy for young children’s word learning (Maynard et al., 2010). However, the results of this study indicate that the educators in the experimental group did not use strategies for explicitly promoting word learning. The literature indicates mixed results for the effectiveness of interventions for enhancing educators’ use of vocabulary teaching strategies. Powell et al. (2010) found no difference in educators’ use of strategies to promote word learning or in children’s receptive language outcomes after either a Web-based or live course. In contrast, Wasik et al. (2006) reported significant improvements in educators’ use of explicit word learning strategies after a 9-month intervention programme. Three possible explanations for the lack of findings for the educators in the current study are tenable. First, the text in the narrative books used in the current study did not include many rare words and educators reading the books may have assumed that the words in the two books were familiar to the children. In future, this task might be enhanced by replacing common words with rare synonyms (e.g. replace stir with whisk), a technique used by Sénéchal and Cornell (1993). Second, the method of scoring the educators’ use of explicit strategies to promote word learning focused on 10 target words from the text that were deemed to be important for the children’s comprehension of the story. However, some educators used the strategies to promote word learning for different words found in the text, while others elaborated on words related to the pictures in the books. For example, one educator explained the phrase ‘vanity mirror’ after a child asked about the picture on the page. This word did not appear in the text. The technique in our current study focused on target words deemed important to the children’s comprehension of the story, but future work might consider all words that educators isolate and expand on. Third, this experimental design involved the educators reading the same two books at pretest and posttest. This may have potentially reduced the number of words the educators would expand on at posttest, given their prior experience of reading the books to the children. In future, it may be advisable to introduce a new book at posttest to provide a better indication of educators' use of strategies. Future work is needed to identify strategies that support educators in promoting explicit word learning with preschool children.

The third major finding of this study relates to the impact of the professional development programme on educator–child shared book reading conversations. By examining the length of joint conversations, this study was able to report on child engagement in shared reading conversations. Overall, the educators learned to engage the children in longer book-related conversations following professional development. These findings demonstrate positive effects of professional development on preschool educators’ ability to engage children in extended discourse and elicit greater children’s participation. Adult–child conversations are considered to make an important contribution to children’s language development (Dickinson, 2003, 2006; McCabe and Peterson, 1991; Peterson et al., 1999), and longer conversations have been shown positively to correlate with enhanced language outcomes (Ruston and Schwanenflugel, 2010; Zimmerman et al., 2009). One mechanism by which longer conversations promote language development resides in the supposition that longer conversations explore a topic in greater depth (Zimmerman et al., 2009). Within longer conversations, educators have more opportunities to repair or expand on children’s utterances by adding additional semantic and syntactic content that is within the children’s zone of proximal development. Dickinson recommended that educators aim to engage children in conversations that are five or more turns in length using a strategy referred to as ‘Strive for five’ (Dickinson, 2003; Greenberg and Weitzman, 2010). The purpose of this strategy is to remind educators to elaborate or deepen a topic beyond the typical educator–student exchange made up of ‘question–response–feedback’. This study indicates that, on average, educators and children in the experimental group were able to maintain conversations of at least five turns for more than half of their conversations.

Follow-up analyses were conducted to explore a possible mechanism or theme of conversation that accounted for the longer conversations. The working hypothesis was that longer conversations would correlate with inferential-personal talk. However, the results of the analysis indicated that longer conversations were related to inferential-text talk (r = 0.57) or conversations that focused on the events or characters in the story (e.g. how a character might feel about the events in the story), rather than the speaker’s personal experience. These findings provide support for instructing educators to use inferential talk in professional development. This type of conversation is considered to be more cognitively challenging, in that it covers topics in greater depth (Dickinson and Smith, 1994; Massey, 2004) and provides educators with opportunities to model decontextualized language (Crain-Thoreson and Dale, 1999; Doyle and Bramwell, 2006; Hargrave and Sénéchal, 2000; van Kleeck, 2008; Wasik and Bond, 2001). The results of this post hoc analysis are suggestive of the value of inferential talk as a mechanism that permits longer conversational topics to emerge and develop.

Enhancing educator quality

Preschool educators’ ability to engage children in longer conversations is a key feature of quality preschool programming and promotes language learning for children considered most at risk, particularly children from low-income families (Ramey and Ramey, 1998). Accordingly, preschool educators need to be equipped with the skills to provide a rich language environment for the children in their care. The National Institute of Child Health and Development – Early Child Care Research Network’s (2002) longitudinal study found that process measures of quality child care, such as the extent to which educators asked questions and were linguistically responsive in adult–child conversations, were most strongly and positively correlated with children’s cognitive, linguistic and social-emotional outcomes (Howes et al., 1992; Mashburn et al., 2008; NICHD, 2002). The current study indicates that educators’ participation in a professional development programme improved the quality of educator–child interactions during shared reading and supports the use of professional development to enhance preschool educators’ ability to facilitate conversation during book reading.

Historically, the effectiveness of professional development to bring about change in educators’ practice has been precarious (Guskey, 1986; Clarke and Hollingsworth, 1986). Guskey (2002) proposed that professional development is most effective when it addresses information that has immediate application to practice. The professional development programme used in this study (Weitzman and Greenberg, 2010) provided instruction in the use of shared book reading paired with individualized classroom coaching in order to ensure that the educators were able to use the strategies appropriately. Book reading is an activity that occurs regularly in preschool classrooms and these educators have opportunities to integrate shared book reading strategies into their daily routines for the benefit of all the children in their care. Importantly, the educators in the current study learned to use these strategies while working with a diverse group of learners, many of whom were from low-income families or homes where the language spoken is other than English. These children are at risk of lagging behind their classmates, but high quality educators who are proficient in the use of shared book reading strategies can provide these children with a rich language learning environment.

Working with diverse populations

Consistent with preschool classrooms in large urban centres, many of the children in this study were from low-income homes (76%) and many were English language learners (ELLs) (47%). Previous research indicates that children from low socio-economic backgrounds are at risk of poor academic outcomes (Biemiller, 2006; Chall et al., 1990; Hart and Risley, 1995, 2003; Ramey and Ramey, 1998). Preschool educators who use shared book reading as a regular instructional practice are, potentially, positioned to enhance these children’s language learning through this interactive book reading style. Lonigan, Anthony, Bloomfield and Samwell (1999) found that after only 3 to 5 hours of exposure to shared book reading (i.e. 19 sessions at 10–15 minutes per session), children from low-income families made significant gains in oral language skills, phonological awareness and listening comprehension as compared to a no-treatment control group of children. Further, the NELP reported that nine shared book reading intervention studies that included only children from low socio-economic (SES) backgrounds produced a large statistically significant effect size (i.e. 0.79). Shared book reading can be an important instructional strategy in preschool classrooms by providing these children with the necessary boost to enhance their school-readiness and potentially place them on a different trajectory for school success (Ramey and Ramey, 1998). The current study focused on the effects of professional development to improve educators’ use of shared book reading strategies that would potentially benefit all children in their classroom, including those children from diverse language backgrounds.

Preschool ELLs can be supported to learn English through regular opportunities to engage in conversation with the educator, to participate in extended classroom conversations that are intellectually stimulating, and to hear and learn more complex vocabulary words (Dickinson et al., 2009). Shared book reading provides a classroom context within which educators can provide these opportunities for ELLs (Restrepo and Towle-Harmon, 2008). Preschool educators can add rich explanations of word meanings, model language use and, importantly, provide feedback and reformulation of the ELL children’s utterances throughout book reading (Goldenberg, 2008). Further, academic language is essential for school success, similarly books, even those read to preschool children, that contain more complex language than children might typically experience in conversations (Sulzby, 1985). The results of the current study indicated that preschool educators in classrooms with children from diverse language backgrounds learned to engage the children in book-related conversations following participation in professional development compared to a control group of educators. The educators’ use of questions and contingent responsiveness facilitated the children’s engagement in book-related conversations resulting in longer conversations, or the children taking more turns related to each topic of conversation. However, the educators in the current study did not learn to use explicit strategies to promote vocabulary learning, a strategy that has been found to promote word learning in ELL preschoolers (Collins, 2005). Future research is needed to identify aspects of professional development that will enhance educators’ ability to incorporate explicit word learning strategies during shared book reading, particularly for ELLs. Furthermore, additional research is needed to identify the extent to which educators modify their language use when using shared book reading strategies with children from diverse language and economic backgrounds.

Limitations and implications

There are several limitations that should be considered when reflecting on the outcomes of this study and in looking ahead to future research. The current study was conducted with a small number of early childhood educators and children. All educators worked in classrooms with a one-to-eight ratio and were responsible for curricular planning for a classroom of 16–24 children. Thus the results of this study may not be generalizeable to preschool teachers with less experience or with larger group sizes than those used in the current study. Future studies are needed that investigate educators’ use of strategies with larger groups of children, a context that is reflective of typical classroom practice. Second, this study did not use standardized measures of children’s vocabulary and expressive language. Future studies may benefit from the use of standardized measures to provide an indicator of improved child language outcomes as a result of educators’ use of language-facilitating shared book reading strategies. Further, a longitudinal follow-up would permit the reporting of long-term maintenance of educators’ strategies and long-term impacts on children’s educational outcomes. Finally, approximately half of the children in the current study were English language learners who spoke another language at home. With respect to oral language development, the specific strategies recommended in this professional development programme, such as open questions and contingent responses, are important to facilitate these children’s engagement in conversation. Future research is needed that investigates the impact of professional programmes on the language and literacy development of bilingual children, particularly for preschool programmes that support language and literacy in the child’s two languages (Hammer et al., 2011).

In conclusion, this study indicates that professional development can yield promising short-term outcomes and enhance educators’ ability to engage small groups of preschool children in book-related talk. This study has implications for the provision of high quality professional development to bring about positive changes in educator–child interaction during shared reading. These improvements in the educators’ ability to engage children in longer book-related conversations resulted in a significant change in the language environments within preschool programmes.

Footnotes

Acknowledgements

This study was financially supported by grants from the Canadian Language and Literacy Research Network and the H. J. Watson Ontario Graduate Scholarship. In addition, we wish to acknowledge and extend our appreciation to the preschool program supervisors, the early childhood educators, the children, and their families for their participation in this study.

References

Bakeman

(2005) Recommended effect size statistics for repeated measures designs. Behavior Research Methods 37(3): 379384.

Biemiller

(2006) Vocabulary development and instruction: A prerequisite for school learning. In: Dickinson

Neuman

(eds) Handbook of Early Literacy Research, 2nd ed. New York, NY: Guilford Press, pp. 4151.

Biemiller

Slonim

(2001) Estimating root word vocabulary growth in normative and advantaged populations: Evidence for a common sequence of vocabulary acquisition. Journal of Educational Psychology 93: 498520.

Blank

Rose

Berlin

(1978) Preschool Language Assessment Instrument, New York, NY: Grune and Stratton.

Blewett

Rump

Shealy

(2009) Shared book reading: When and how questions affect young children’s word learning. Journal of Educational Psychology 101(2): 294304.

Boudreau

(2008) Narrative abilities: Advances in research and implications for clinical practice. Topics in Language Disorder 28(2): 99114.

Burchinal

Roberts

Nabors

(1996) Quality of center child care and infant cognitive and language development. Child Development 67(2): 606620.

Burchinal

Roberts

Riggins

(2000) Relating quality of center-based child care to early cognitive and language development longitudinally. Child Development 71(2): 339357.

Chall

Jacobs

Baldwin

(1990) The Reading Crisis: Why Poor Children Fall Behind, Cambridge, MA: Harvard University Press.

10.

Clarke

Hollingsworth

(2002) Elaborating a model of teacher professional growth. Teaching and Teacher Education 18: 947967.

11.

Cohen

(1992) A power primer. Psychological Bulletin 112: 155159.

12.

Collins

(2010) ELL preschoolers’ English vocabulary acquisition from storybook reading. Early Childhood Research Quarterly 25(1): 8497.

13.

Connor

Morrison

Slominski

(2006) Preschool instruction and children’s emergent literacy growth. Journal of Educational Psychology 98(4): 665689.

14.

Crain-Thoreson

Dale

(1999) Enhancing linguistic performance: Parents and teachers as book reading partners for children with language delays. Topics in Early Childhood Special Education 19(1): 2839.

15.

Culatta

Blank

Black

(2010) Talking things through: Roles of instructional discourse in children’s processing of expository texts. Topics in Language Disorders 30(4): 308322.

16.

De Rivera C, Girolametto L, Greenberg J and Weitzman E (2005) Children's responses to educators' questions in day care play groups. American Journal of Speech-Language Pathology 14(1): 14–26.

17.

Dickinson

(2003) Why we must improve teacher-child conversations in preschools and the promise of professional development. In: Girolametto

Weitzman

(eds) Enhancing Caregiver Language Facilitation in Child Care Settings: Proceedings from the Symposium, Toronto, Canada: Hanen Early Language Program, pp. 4148.

18.

Dickinson

McCabe

(2001) Bringing it all together: The multiple origins, skills, and environmental supports of early literacy. Learning Disabilities Research and Practice 16(4): 186202.

19.

Dickinson

McCabe

Anastasopoulos

(2002) A framework for examining book reading in early childhood classrooms. In: Van Kleeck

Stahl

Bauer

(eds) On Reading to Children: Parents and Teachers, Mahwah, NJ: Erlbaum, pp. 95113.

20.

Dickinson

Smith

(1994) Long-term effects of preschool teachers' book readings on low-income children's vocabulary and story comprehension. Reading Research Quarterly 29(2): 104122.

21.

Dickinson

Tabors

(2001) Beginning Language with Literacy: Young Children Learning at Home and School, Baltimore, MD: Brookes Publishing.

22.

Dickinson

Tabors

(2002) Fostering language and literacy in classrooms and homes. Young Children 57(2): 1018.

23.

Doyle

Bramwell

(2006) Promoting emergent literacy and social–emotional learning through dialogic reading. The Reading Teacher 59(6): 554564.

24.

Gerde

Powell

(2009) Teacher education, book reading practices, and children's language growth across one year of Head Start. Early Education and Development 20(2): 211237.

25.

Girolametto L and Weitzman E (2002) Responsiveness of child care providers in interactions with toddlers and preschoolers. Language, Speech, and Hearing Services in Schools 33: 268–281.

26.

Girolametto L and Weitzman E (2007) Promoting peer interaction skills: Professional development for early childhood educators and preschool teachers. Topics in Language Disorders 27(2): 93–110.

27.

Girolametto L, Weitzman E and Greenberg J (2012) Facilitating emergent literacy: Efficacy of a model that partners speech-language pathologists and educators. American Journal of Speech-Language Pathology 21: 47–63.

28.

Girolametto L, Weitzman E and Greenberg J (2006) Facilitating language skills: In-service education for early childhood educators and preschool teachers. Infants and Young Children 19(1): 36–46.

29.

Girolametto L, Weitzman E, Lefebvre P and Greenberg J (2007) The effects of in-service education to promote emergent literacy in child care centers: A feasibility study. Language, Speech, and Hearing Services in Schools 38(1): 72–83.

30.

Girolametto L, Weitzman E, Van Lieshout R and Duff D (2000) Directiveness in teachers' language input to toddlers and preschoolers in day care. Journal of Speech, Language, and Hearing Research 43: 1101–1114.

31.

Greenberg J (2011) Making Hanen Happen Leaders Guide for Hanen Certified ABC and Beyond Trainers. Toronto: A Hanen Centre Publication.

32.

Goldenberg

(2008) Teaching English language learners: What the research does and does not say. American Educator 32(2): 844.

33.

Gusky

(1986) Staff development and the process of teacher change. Educational Researcher 15(5): 512.

34.

Gusky

(2002) Professional development and teacher change. Teachers and Teaching: Theory and Practice 8(3): 381391.

35.

Hadley

Rice

(1993) Parental judgements of preschoolers' speech and language development: A resource for assessment and IEP planning. Topics in Speech and Language 14(4): 278288.

36.

Hammer

Jia

Uchikoshi

(2011) Language and literacy development of dual language learners growing up in the United States: A call for research. Child Development Perspectives 5(1): 49.

37.

Hammett-Price

Van Kleeck

Huberty

(2009) Talk during book sharing between parents and preschool children: A comparison between storybook and expository book conditions. Reading Research Quarterly 44(2): 171194.

38.

Hargrave

Sénéchal

(2000) A book reading intervention with preschool children who have limited vocabularies: The benefits of regular reading and dialogic reading. Early Childhood Research Quarterly 15(1): 7590.

39.

Harris

(1978) Don't Forget to Come Back, Cambridge, MA: Candlewick Press.

40.

Hart

Risley

(1995) Meaningful Differences in the Everyday Experience of Young American Children, Baltimore, MD: Paul H. Brookes Publishing Co.

41.

Hart

Risley

(2003) The early catastrophe: The 30 million word gap by age 3. American Educator 27(1): 49.

42.

Hetherington

Parke

Schmuckler

(2005) Child Psychology: A Contemporary Viewpoint, 2nd edn. Toronto, Canada: McGraw-Hill Ryerson Ltd.

43.

Hindman

Connor

Jewkes

(2008) Untangling the effects of shared book reading: Multiple factors and their associations with preschool literacy outcomes. Early Childhood Research Quarterly 23: 330350.

44.

Howes

Burchinal

Pianta

(2008) Ready to learn? Children's pre-academic achievement in pre-kindergarten programs. Early Childhood Research Quarterly 23(1): 2750.

45.

Howes

Phillips

Whitebrook

(1992) Thresholds of quality: Implications for the social development of children in center-based child care. Child Development 63(2): 449460.

46.

Johnston

(2001) An alternative MLU calculation: Magnitude and variability of effects. Journal of Speech, Language, and Hearing Research 44: 156164.

47.

Justice

Kaderavek

(2002) Using shared storybook reading to promote emergent literacy. Teaching Exceptional Children 34(4): 813.

48.

Justice

Meier

Walpole

(2005) Learning new words from storybooks: An efficacy study with at-risk kindergartners. Language, Speech Hearing Services in Schools 36(1): 1732.

49.

Kaderavek

Justice

(2010) Fidelity: An essential component of evidence-based practice in speech language pathology. American Journal of Speech-Language Pathology 19: 369379.

50.

Lever

Sénéchal

(2011) Discussing stories: On how a dialogic reading intervention improves kindergartners’ oral narrative construction. Journal of Experimental Child Psychology 108: 124.

51.

Lonigan

Anthony

Bloomfield

(1999) Effects of two shared-reading interventions on emergent literacy skills of at-risk preschoolers. Journal of Early Intervention 22(4): 306322.

52.

Lonigan

Whitehurst

(1998) Relative efficacy of parent and teacher involvement in a shared-reading intervention for preschool children from low-income backgrounds. Early Childhood Research Quarterly 13(2): 263290.

53.

Mashburn

Pianta

Hamre

(2008) Measures of classroom quality in prekindergarten and children’s development of academic, language, and social skills. Child Development 79(3): 732749.

54.

Massey

(2004) Teacher-child conversation in the preschool classroom. Early Childhood Education Journal 31(4): 227231.

55.

Massey

Pence

Justice

(2008) Educators' use of cognitively challenging questions in economically disadvantaged preschool classroom contexts. Early Education and Development 19(2): 340360.

56.

Maynard

Pullen

Coyne

(2010) Teaching vocabulary to first-grade students through repeated shared storybook reading: A comparison of rich and basic instruction to incidental exposure. Literacy Research and Instruction 49(3): 209242.

57.

McCabe

Jenkins

Mills

(1996) Effects of play group variables on language use by preschool children with disabilities. Journal of Early Intervention 20(4): 329340.

58.

McCabe

Peterson

(1991) Getting the story: A longitudinal study of parental styles of eliciting narratives and developing narrative skills. In: McCabe

Peterson

(eds) Developing Narratives Structure, Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 217253.

59.

Miller

Chapman

(2002) Systematic Analysis of Language Transcripts, Madison, WI: University of Wisconsin-Madison.

60.

Mol

Bus

De Jong

(2009) Interactive book reading in early education: A tool to stimulate print knowledge as well as oral language. Review of Educational Research 79(2): 9791007.

61.

Moore

(2000) Transcript Builder, Eau Claire, WI: Thinking Publications Inc.

62.

National Early Literacy Panel (2008) Developing early literacy: Report of the National Early Literacy Panel. Available at: www.nifl.gov/publications/pdf/NELPReport09.pdf (accessed 8 July 2010).

63.

National Institute of Child Health and Human Development (2000) Report of the National Reading Panel. Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction. Available at: www.nichd.nih.gov/publications/nrp/smallbook.cfm (accessed 8 December 2008).

64.

NICHD Early Child Care Research Network (2002) Early child care and children's development prior to school entry: Results from the NICHD study of early child care. American Educational Research Journal 39(1): 133164.

65.

Nelson

Camarata

Welsh

(1996) Conversational recasting treatment on the acquisition of grammar in children with specific language impairment and younger language normal children. Journal of Speech, Language, and Hearing Research 39: 850859.

66.

Pardo

(2004) What every teacher needs to know about comprehension. Reading Teacher 58: 272280.

67.

Peisner-Feinberg

Burchinal

Clifford

(2001) The relation of preschool child-care quality to children’s cognitive and social development trajectories through second grade. Child Development 72(5): 15341553.

68.

Pentimonti

Justice

(2010) Teachers’ use of scaffolding strategies during read alouds in the preschool classroom. Early Childhood Education Journal 37: 241248.

69.

Peterson

Jesso

McCabe

(1999) Encouraging narratives in preschoolers: An intervention study. Journal of Child Language 26(1), 4967.

70.

Powell

Diamond

Burchinal

(2010) Effects of an early literacy professional development intervention on Head Start teachers and children. Journal of Educational Psychology 102(2): 299312.

71.

Ramey

(1998) Early intervention and early experience. American Psychologist 53(2): 109120.

72.

Restrepo MA and Towle-Harmon M (2008) Addressing emergent literacy skills in English-language learners. The ASHA Leader September. Available at: http://www.asha.org/Publications/leader/2008/080923/f080923a/ (accessed 20 July 2012).

73.

Roberts

Jurgens

Burchinal

(2005) The role of home literacy practices in preschool children's language and emergent literacy skills. Journal of Speech, Language, and Hearing Research 48: 345359.

74.

Rogoff

(1990) Apprenticeship in Thinking, New York, NY: Oxford University Press.

75.

Ruston

Schwanenflugel

(2010) Effects of a conversation intervention on the expressive vocabulary development of prekindergarten children. Language Speech Hearing Services in Schools 41: 303313.

76.

Sackett

(1978) Observing Behavior, Baltimore, MD: University Park Press.

77.

Sénéchal

Cornell

(1993) Vocabulary acquisition through shared reading experiences. Reading Research Quarterly 28(4): 360374.

78.

Sénéchal

Hargrave

(2000) A book reading intervention with preschool children who have limited vocabularies: The benefits of regular reading and dialogic reading. Early Childhood Research Quarterly 15(1): 7590.

79.

Simon

(2003) Little Yellow Dog Gets a Shock, London, UK: Orion Children's Books.

80.

Snow

Burns

Griffin

(1998) Preventing Reading Difficulties in Young Children, Washington, DC: National Academy Press.

81.

Statistics Canada (2006) Census tract (CT) profiles. (92/597-XWE) Available at: www12.statcan.ca/census-recensement/2006/dp-pd/prof/92-597/index.cfm (accessed 5 March 2009).

82.

Sulzby

(1985) Children’s emergent reading of favorite storybooks: A developmental study. Reading Research Quarterly 20: 458481.

83.

Turnbull

Anthony

Justice

(2009) Preschoolers exposure to language stimulation in classrooms serving at-risk children: The contribution of group size and activity context. Early Education and Development 20(1): 5379.

84.

Van Kleeck

(2008) Providing preschool foundations for later reading comprehension: The importance of and ideas for targeting inferencing in storybook-sharing interventions. Psychology in the Schools 45(7): 627643.

85.

Van Kleeck

Gillam

Hamilton

(1997) The relationship between middle-class parents' book-sharing discussion and their preschoolers' abstract language development. Journal of Speech, Language, and Hearing Research 40: 12611271.

86.

Van Kleeck

Vander Woude

Hammett

(2006) Fostering literal and inferential language skills in Head Start preschoolers with language impairment using scripted book-sharing discussions. American Journal of Speech-Language Pathology 15(1): 8595.

87.

Vygotsky

(1978) Mind in Society: The Development of Higher Psychological Processes, Cambridge, MA: Harvard University Press.

88.

Wasik

(2008) When fewer is more: Small groups in early childhood classrooms. Early Childhood Education Journal 35(6): 515521.

89.

Wasik

Bond

(2001) Beyond the pages of a book: Interactive book reading and language. Journal of Educational Psychology 93(2): 243250.

90.

Wasik

Bond

Hindman

(2006) The effects of a language and literacy intervention on Head Start children and teachers. Journal of Educational Psychology 98: 6374.

91.

Weitzman E and Greenberg J (2010). ABC and Beyond: Building Emergent Literacy in Early Childhood Setting. Toronto, ON.

92.

Whitehurst

Lonigan

(1998) Child development and emergent literacy. Child Development 69(3): 848872.

93.

Whitehurst

Lonigan

(2001) Emergent literacy: Development from prereaders to readers. In: Neuman

Dickinson

(eds) Handbook of Early Literacy Research, New York, NY: Guilford Press, pp. 1129.

94.

Wood

Bruner

Ross

(1976) The role of tutoring in problem solving. Journal of Child Psychology and Psychiatry 17: 89100.

95.

Zevenbergen

Whitehurst

Zevenbergen

(2003) Effects of a shared-reading intervention on the inclusion of evaluative devices in narratives of children from low-income families. Journal of Applied Developmental Psychology 24(1): 115.

96.

Zimmerman

Gilkerson

Richards

(2009) Teaching by listening: The importance of adult-child conversations to language development. Pediatrics 124: 342349.

97.

Zucker

Justice

Piasta

(2010) Preschool teachers’ literal and inferential questions and children’s responses during whole-class shared reading. Early Childhood Research Quarterly 25: 6583.