Investigating affordance of opportunity for young children’s language interactions in a nursery setting: How can small group talk act as a forum for language learning?

Abstract

The importance of research on the unique nature of the communication supporting environment in nurseries has been heightened by growing evidence of the significance of early language skills for later academic and social development. This study focussed on children’s language use during small group times. Opportunities to hear and practise language were examined to uncover variation in conversational experiences for children with differing language needs. In this mixed-methods study, different measures were used to examine the relationship between participation and language level. Participants were an Early Years practitioner and 19 3- to 4-year-olds in two cohorts. Children’s language levels were measured using the Clinical Evaluation of Language Fundamentals Pre-School (2) UK. Quantitative analysis of interaction rates was made from video recordings of small group conversations. This was followed by detailed qualitative examination of talk during episodes of more sustained conversation. Differences were revealed in affordance of opportunity for children according to language level. Children’s interaction rates were positively correlated with scores on the Clinical Evaluation of Language Fundamentals Pre-School (2) UK at the start. Analysis showed conversational features of both formal and informal talk. Combining features from each was found to be associated with episodes of sustained conversation. Patterns of turn-taking were associated differently with participation for children with higher and lower language levels. Findings support the role for small group times as a forum for language development, facilitating opportunities for children differently according to their language needs. This has important implications for practice in supporting children to make the transition from informal to formal talk in the educational setting.

Keywords

communication development early years practitioner interactions language learning nursery talk young children

Introduction

Evidence increasingly indicates that early language development has a strong association with later academic and social development. In particular, research has linked early language difficulties to later reading difficulties (Botting et al., 2006; Catts et al., 2002; Dickinson and Tabors, 2001; Griffin et al., 2004; Loban, 1964; Muter et al., 2004; Roth et al., 2002; Snowling et al., 2003). Youngsters with language difficulties also appear more likely to display behavioural, social and emotional difficulties (Howlin et al., 2000; Lindsay et al., 2007; Stringer and Clegg, 2006; Stringer and Lozano, 2007). The argument for promoting children’s language skill in the early years is further strengthened by research on social disadvantage indicating that language input and language outcomes in the early years differ significantly for youngsters from more and less advantaged backgrounds (Hansen and Joshi, 2007; Hart and Risley, 1995; Roulstone et al., 2011). One solution to this has been to offer children earlier access to nursery education. There has been little research, however, on the specific nature of language development in the nursery environment. Studies have addressed adult use of language in the classroom and with groups of children (Cabell et al., 2011; Dickinson and Tabors, 2001; Girolametto et al., 2003; Pence et al., 2008; Turnbull et al., 2009; Wasik et al., 2006). Few studies though have approached the topic from the point of view of the child’s own use of language in context.

Language in the nursery environment

There is some evidence that children’s use of language in the nursery and at home differs (Flewitt, 2005; Tizard and Hughes, 2002 [1984]; Wells, 1981). One key difference is likely to be that the child is, possibly for the first time, one member of a group when interacting in the nursery. This offers increased opportunities for experiencing the modelling of language, but also presents more challenges in, for example, gaining a turn to talk and having something relevant to say. The opportunities offered by the nursery may be more similar to those in communal cultures or the extended family, where children are exposed to frequent polyadic interactions, rather than the dyadic interactions often studied in Western language development (Dunn and Shatz, 1989; Goldfield and Reznick, 1990; Hoff, 2010; Ochs and Schieffelin, 1995; Shneidman and Goldin-Meadow, 2012; Strapp and Federico, 2000).

Being part of a small group conversation may be of benefit even where children do not necessarily contribute verbally themselves. Several experimental studies, where vocabulary can be introduced under controlled conditions, have demonstrated that young children can show understanding of words they have overheard between others, without the child being directly engaged in the talk (Akhtar, 2005; Akhtar et al., 2001; Floor and Akhtar, 2006; Foster and Hund, 2012; Gampe et al., 2012; Scofield and Behrend, 2011; Shneidman et al., 2009). Nevertheless, Hoff (2006) reminded us that both access to a language model that can be analysed as input and opportunity for communicative experience are necessary components for language learning. Children require the opportunity to use the language they have heard to communicate at some point. The impetus to communicate might need to be stronger in the group setting though, as the child has to compete with others for a turn.

Recent government-sponsored initiatives in the United Kingdom reflect an increased emphasis within the Early Years Curriculum on oral language skills (Department for Children Schools and Families (DCSF), 2008; Department for Education and Skills (DfES), 2006; I-CAN, 2012). While research has demonstrated some evidence that practitioners can change their interactions in response to training and children can make attendant progress, what precisely is helpful and to which particular children remains far from clear (Cabell et al., 2011; Girolametto et al., 2003; Lonigan et al., 2011; Wasik et al., 2006). Sylva et al. (2007, Effective Provision of Pre-School Education (EPPE) Project) found more adult–child interactions and social conversations in day care that was associated with better cognitive development and later achievements for children. Further, these settings were characterised by more episodes of sustained shared thinking, where adult and child worked together in an intellectual way to solve a problem, clarify a concept, evaluate activities or extend narratives (Siraj-Blatchford, 2009a; Siraj-Blatchford et al., 2003). Jordan (2009) developed a similar model of co-construction in her work, drawing from previous ideas about scaffolding and collaboration (Bruner, 1974; Rogoff, 1998, 2003; Vygotsky, 1986 [1934]). Working with teachers to identify and work with their own best dialogues with children, Jordan (2009) identified three different types of interaction each with its own place and purpose. In co-construction, adult and child act as equal partners in interactions. The child’s own knowledge and experience is validated by the adult and the adult shares their own thinking with the child. This can happen whether the child or the adult initiated the topic. Other interaction types were characterised either by the adult leading and controlling the discussion or by the child directing, controlling and leading the discussion. Jordan (2009) argued that where co-construction occurred it allowed for continuity of learning as links were made between the children’s different experiences. Home and school became more connected and adults and children could both contribute from their own fund of experience.

Affordance of opportunity and participative models of language

Adopting an ecological perspective, language can helpfully be viewed as part of the context, medium or habitat in which people live, rather than a body of knowledge that exists in their heads (Bronfenbrenner, 1979; Van Lier, 2000). Skills develop through the individual’s interaction with the environment, which in turn affects their ability to access the potential meanings that exist in the language. Opportunities for learning are shaped by the scope that they afford to the child depending on the previous cultural and linguistic experiences and understandings of the learner and it is this that determines what affordances new learning experiences offer (Van Lier, 2000). Language is always about something, within a specific socio-cultural context. The exact meaning of that context, and so the learning opportunities offered, will differ depending on the constructions that the individual brings to the situation, affecting how the situation is perceived, including the tools that mediate (e.g. a scripted test or a conversation about family experiences). Framing teaching and learning in this way enables a shift from viewing the success or otherwise of the child’s engagement with an activity as a deficit or disadvantage in the child. Rather, it enables a focus on the activity itself and how this can be constructed to enable the child’s learning.

Taking such an interactionist perspective, conversation analysis (CA) has been usefully applied to educational contexts (Freebody, 2003; Have, 2007; Sacks et al., 1974). Detailed studies have included, for example, one-to-one scaffolding of learning, the effects of teacher and teaching assistant classroom talk in opening up or closing down talk and analysis of small group communication tasks for young children with specific language impairment (Pike, 2010; Radford et al., 2006, 2011). This work has helped to build up an understanding of how participants demonstrate and develop their understandings about the nature and purpose of the conversation. Children’s language development can be viewed from the point of view of creating shared understandings and how these understandings develop and influence conversation. This then, in turn, offers possibilities for looking at classroom practice and how shared understanding can be promoted to afford opportunities for children’s learning.

This approach has yet to be applied to investigation of the detail of moment-to-moment interactions between very young children and nursery practitioners. By learning more about the strategies that different children use and the mechanisms that may be supporting or inhibiting their language development within the nursery curriculum, this study aimed to add to a more detailed understanding of what is happening in the Early Years settings. It is only through such research that further understanding can be gained about the ways in which children respond to and employ language and its consequent effects on the development of communication in the nursery.

Rationale and design

This study combined individual and conversational measures, each offering its own particular affordances and ways of doing language, to address three research questions. First children’s verbal initiation and response rates during small group conversations were investigated in order to establish who talks when, with what frequency and the occurrence of extended turn-taking. Second, these interaction rates were compared to children’s language levels, to examine the extent to which children’s language skills were linked to the likelihood of their taking up the opportunities that were offered for interaction. The third question, addressed through a detailed analysis of selected conversations, was that of how conversational experiences differed for children of differing language ability and the potential implications for nursery practice.

Participants

Participants were drawn from children attending a 30-place nursery class, at a local authority-maintained primary school, in an economically disadvantaged metropolitan borough in the West Midlands (Social Disadvantage Research Centre, 2009). A total of 19 3- to 4-year-old children participated in the study in two cohorts working with the same key worker in two successive academic years. The key worker was a female nursery nurse who took specific responsibility for the group at registration and other key teaching times. Nine children in Cohort 1 attended a daily afternoon nursery session and 10 children, who attended morning sessions, were in Cohort 2. There were no significant differences in mean age, initial language score, gender or number of children who had English as an additional language between the two cohorts, as shown in Table 1.

Table 1.

Characteristics of participants.

	Cohort 1 (n = 9)	Cohort 2 (n = 10)	Whole sample (n = 19)	Test for significant differences between cohorts
Average age at initial assessment in months	45.2 (SD = 3.93)	45.5 (SD = 3.40)	45.4 (SD = 3.6)	t = −.16, p = .87
Age range at initial Assessment in months	40–52	42–52	40–52	–
Gender	5 boys; 4 girls	5 boys; 5 girls	10 boys; 9 girls	Two-tailed Fisher exact, p = 1.00
English as an additional language	3 (Punjabi)	3 (2 Punjabi; 1 Bengali)	6	–
Special educational needs	None	1	1	Two-tailed Fisher exact, p = 1.00
CELF Pre-School (2) UK Mean Standard Score	M = 85.1 (SD = 16.0)	M = 83.8 (SD = 15.5)	M = 85.2 (SD = 15.4)	t = .18, p = .86

CELF: Clinical Evaluation of Language Fundamentals; SD: standard deviation.

Procedures

Materials

The Clinical Evaluation of Language Fundamentals (CELF) Pre-School (2) UK (Wiig et al., 2006), a widely used standardised language assessment, was chosen as a measure of individual language level. The test is designed for use with 3- to 6-year-olds, to give a measure of the child’s general language ability as well as providing information about their language and communication strengths and weaknesses. The test gives a Core Language Score, made up from sub-tests that assess receptive language, use of grammatical structures/morphology and expressive single-word vocabulary.

Data collection

Preliminary testing using CELF Pre-School (2) UK, to give a baseline measure of children’s language skills, was carried out during the first half of the Spring Term, having given the children a term to settle into nursery. To gain naturalistic, sample observations of children’s participation in conversations, video observations were then made of the children during small group-focussed sessions with their nursery key worker. Such situations were noted from observation to be the time at which the children would have the most readily available access to quality talk with an adult. This arrangement also offered the potential for the same few children being present together with the adult on each occasion, allowing for building a bank of recordings of the same participants together. Filming took place between March and June. Between 10 and 12 sessions, varying from 15 to 40 minutes in length, were recorded for each cohort.

Analysis

A mixed-methods approach was adopted for analysis. Combining quantitative and qualitative methods enabled the relative strengths and weaknesses of each approach to complement each other (Lewis and Miller, 2011; Mercer, 2010). Quantitative methods allowed for the representativeness of the entire data set to be accounted for, without necessitating lengthy transcription, and permitting numerical and statistical comparison for relationships and differences across and within the data. Qualitative analysis was then applied to capture the content on the nature and meaning of the talk.

Quantitative analysis

Video analysis

Initially the entire footage was coded, except for some short sections moving between locations. The data on each child were analysed individually, direct from the video recordings, using a coding scheme designed to capture levels of initiation and response (King and Saxton, 2010). Coding categories were established empirically by examining the observations, rather than through the use of any pre-defined coding. This gave four measures of interaction: Adult Initiations, Child Initiations, Responses Received by the child and Responses Made by the child. Adult Initiations included opportunities to talk offered to the group in general as well as those addressed specifically to the individual child. Individual interaction rates were derived for each child on each of the four categories. Frequencies were totalled across sessions and converted to overall rates of initiation or response per minute, to take account of the differing lengths of time recorded for each child (see Box 1). Additionally, the frequency of verbal interactions in which the child took two or more successive turns was analysed, as a measure of the development of more extended conversation.

Box 1.

Example to show calculation of different interaction rates per minute for each child.

Interaction measures obtained	Example calculation for Child 1 Adult Initiation rpm
Adult Initiation (AI)	$\frac{Child 1 (Total AISessions 1 - 10)}{Child 1 (Total time in mins Sessions 1 - 10)} = Child 1 AI rpm$
Child Initiation (CI)
Responses Received (RR)
Responses Made (RM)

Four distinct types of discussions were identified within sessions: Child-led, Routine, Adult-led and Individual. (For definitions, see Appendix 1.) Data coded for interaction rate were also analysed by discussion type, to explore potential differences in pattern and purpose of interaction.

Inter-rater reliability

To determine the reliability of the coding guidelines and method, one entire session from each cohort was coded blind by an independent rater, following training by the researcher. Two sessions were chosen, as close as possible to 10 per cent of the time for that cohort, comprising 9 per cent of the total video footage in all. Comparison was made between the numbers of observations made within each coding category for each child. A percentage agreement between the two raters was calculated for each interaction category, using the following formula: number of agreements/number of agreements – disagreements × 100. Overall agreement was acceptable, ranging across the four interaction categories between 80.4 per cent and 96.6 per cent.

The reliability of the Discussion Type categories was tested through independent co-rating of 35 per cent of the sessions chosen at random. At a broad category level for the four types, there was 100 per cent agreement in the number of transitions identified and the category to which each sub-section was assigned. Kappa for the agreement between the ratings of Rater 1 and Rater 2 was, therefore, 1.00, indicating a highly significant agreement (p = .001). Further analysis of percentage agreement for length of the sub-sections identified indicated agreement between 93 per cent and 100 per cent.

Comparison of language measures

Interaction measures were compared to each other for the sample as a whole and across discussion types, using statistical analysis to examine opportunities offered for verbal interaction. Interaction patterns were then compared to initial CELF Pre-School (2) UK scores to analyse the relationship between children’s language levels and their verbal interaction patterns. For the purposes of further comparison, the data were divided into two groups based on initial CELF Pre-School (2) UK score, using a cut-off point of Standard Score below 80, being 1.5 standard deviations below the mean. Nine children fell below the cut-off point, while 10 scored above this level.

Qualitative analysis

Further analysis was undertaken in order to identify any differences in adult talk to different children, to explore how children received and responded to conversational cues and to illustrate how this may result in different outcomes according to children’s language levels.

Video analysis

The entire footage was reviewed to select cases for detailed content analysis using criteria based on definitions of co-construction and sustained shared thinking (Jordan, 2009; Siraj-Blatchford, 2009a, 2009b). Extended turns on a topic from a child/children, plus one or more of the following conditions was required: The children asked further questions or gave information about personal experiences and knowledge; the adult shared their own thinking or asked for further information from the child; there was a quality of curiosity, excitement or genuine interest from adult and children; one or more other children joined in and contributed without initial prompting by the adult.

Sequences of talk that met the specific criteria outlined above were then reviewed using the detailed transcription method employed in CA, a recognised and well-used method of transcription which is underpinned for analysis by a model of language as part of social construction (Have, 2007; Sacks et al., 1974). As sequences of talk form the focus for analysis, only episodes lasting more than 1 minute were included. The transcript was then interrogated line by line, through repeated viewings. The following questions informed the analysis: How is the speaker of the turn selected (e.g. by self-selection or by the previous speaker)? Who responds in a turn, in terms of taking up the topic offered or otherwise and how is this effected? What work does that turn achieve, in terms of what happens next?

Results

Interaction patterns in small group conversations

The majority of interactions took place between the adult and a child. Overall, only 15.7 per cent of verbal interactions initiated by a child were addressed to another child, although there was a wide range among children in this respect (0.0%–43.7%). Similarly, only 5.5 per cent of the responses made by children were made to another child (range: 0.0%–31.5%). Analysis suggested that the majority of talk was not part of an extended conversation for children. In total, only 28.6 per cent of child talk was part of an extended turn exchange.

Interaction patterns differed with discussion types across sessions, when the interaction rates for all children were considered together. A one-way repeated measures analysis of variance showed a significant effect for discussion type on the rate of Adult Initiations, Child Initiations and Responses Received, but not for Responses Made, as summarised in Table 2. Pair-wise comparisons indicated that the rate of Adult Initiations was higher during Routine and Adult-led discussion types than in Child-led or Individual Time. Rates of both Child Initiation and Responses Received were higher during Individual Time compared to all other types of discussion. The rate of Responses Made did not differ significantly overall between discussion types. That is, even where the discussion type was associated with more child initiations, such as in Individual Time, the rate of follow-up Responses Made by children did not differ significantly. Discussion type accounted for about 60 per cent of the variance in the rate of Adult Initiations and about 40 per cent of the variance in the rate of Child Initiations and Responses Received.

Table 2.

Summary of analysis of variance (ANOVA) for four measures of verbal interaction during different discussion types.

Verbal interaction measure	Discussion type								Analysis (one-way repeated measures ANOVA) and pair-wise comparisons (Bonferroni, p = .01)
	1. Child-led		2. Routine		3. Adult-led		4. Individual time
	M	SD	M	SD	M	SD	M	SD
Adult initiations	1.2	.65	2.42	.67	2.4	.29	.96	.60	Main effect: F(1.2, 22.1) = 29.3, p ≤ .001; Partial Eta² = .62 Pair-wise comparisons: 1:2 p ≤ .001, 1:3 p ≤ .001, 1:4 ns, 2:3 ns, 2:4 p ≤ .001, 3:4 p ≤ .001
Child initiations	.46	.53	.36	.33	.49	.46	.84	.59	Main effect: F(2.1, 38.1) = 13.2, p ≤ .001; Partial Eta² = .42 Pair-wise comparisons: 1:2 ns, 1:3 ns, 1:4 p ≤ .001, 2:3 ns, 2:4 p ≤ .001, 3:4 p = .001
Responses received	.64	.70	.48	.58	.56	.63	.97	.67	Main effect: F(2.1, 37.8) = 11.9, p ≤ .001; Partial Eta² = .40 Pair-wise comparisons: 1:2 ns, 1:3 ns, 1:4 p = .01, 2:3 ns, 2:4 p ≤ .001, 3:4 p = .001
Responses made	.78	.70	.91	.73	.71	.75	.84	.65	Main effect: F(1.9, 33.6) = 1.22, ns; Partial Eta² = .06 Pair-wise comparisons: 2:3 p = .01 otherwise ns

ANOVA: analysis of variance; SD: standard deviation.

Children’s interaction patterns and language levels

As shown in Table 3, a partial correlation analysis, controlling for age, revealed positive correlations between children’s rates of initiation, responses received and responses made. The higher a child’s rate of initiations, the higher the rate of responses received and made by them. There was no significant correlation between the rate of Adult Initiations and any of the other verbal interaction measures. Children’s verbal interaction rates varied as a function of their initial language levels. There was a positive correlation between children’s rates of initiation, responses received and made and their initial score on the CELF Pre-School (2) UK (see Table 3.). The higher a child’s initial CELF score, the higher their rates of interaction were likely to be. In contrast, the rate of Adult Initiations showed a negative correlation with children’s CELF scores. The lower a child’s score, the higher the rate of initiations received from the adult was likely to be.

Table 3.

Correlations between initial language measures and interaction rates.

	1	2	3	4	5
1 CELF Raw Score Time 1 (n = 19)	–
2 Adult Initiation Rpm (n = 19)	−.50*	–
3 Child Initiation Rpm (n = 19)	.51*	−.17	–
4 Responses Received Rpm (n = 19)	.50*	−.12	.92**	–
5 Responses Made Rpm (n = 19)	.52*	−.12	.91**	.99**	–

CELF: Clinical Evaluation of Language Fundamentals.

p ≤ .05; **p ≤ .01.

Conversational features – informal and educational talk

Using the questions framed by CA, different turn-taking patterns emerged, offering opportunities to analyse the features that opened up or closed down children’s participation. One prominent feature was the degree to which the adult managed the conversation in terms of accepting topics and managing turn-taking. Several of the episodes followed a pattern of natural or informal conversation, with adult and child contributing equally. There were also instances in which features of educational talk were more prominent, with a clear goal set and formal management of turn-taking by the adult. In addition, a third type of conversation was identifiable, in which formal and informal patterns combined. A transcript and commentary, selected as a proto-typical exemplar, is presented here to illustrate the three different types of pattern observed. Examples given reflect the questions used to interrogate the data, with a focus on points such as who is selected for a turn, the vocabulary used, overlaps and pauses. A list of transcription conventions used is given in Appendix 2.

Extract 1: Informal conversational pattern.

Context: Routine discussion; days of the week. The adult (A) was asking about yesterday and tomorrow. Child 7 then introduced information about his own personal experience (Line 21).

10 A: If tod[ay

11 C1: [Thursday]

12 C1: Thursday

13 A: Thursday (.) well done

14 C7: And then Friday

15 A: And then Friiiday (.) and then

16 C1: Saturday

17 C3: (xxx)

18 A: And

19 C1: Saturday

20 A: and Sss (0.3) Sunday

21 C7: And [then I having

22 C1: [xx

23 C7: then I (have) six weeks off school

24 A: Not just yet we’ve got a long while til our six weeks (.) but we do::n’t come to school on Saturday (.)

(( Nods head))

25 C7: Or Sunday

26 A: [And ]Sunday (.) cos that’s is our weekend and on the w[eekend we stay at home

27 C7: [But ] [ But

28 C7: But sometimes when I not at school I go (.) sometimes I’m not at school I go TO holidays

29 A: That’s right (.) you’ve just come back off your holidays haven’t you ((C7))?

30 C7: I’ve been for (six birthday) will be Mickey Mouse

31 A: Where did you go now to meet Mickey Mouse? (.) you went to? (.) Dis (.)

32 C7: [Disneyland]

33 C1: [I xx]

34 A: Di:::::sneyla:::nd, tha::t’s right (.) you’ve just come back off your holidays from Disneyland haven’t you

35 C7: I haven’t been to Disneyland yet ((shakes head))

36 A: You haven’t been (.) not yet.

37 C7: But I (xxx sixth birthday) will be Mickey Mouse

38 A: Aaahhh there we go then (.) Right then

The pattern of talk here showed many features of informal conversation. Child 7, a child of higher language ability, developed the conversation on his own topic, by linking it to the Routine topic and the adult allowed him to do so by accepting his personal topic (at Line 29) and extending his turns with further questions of her own. The next few turns then took place exclusively between the adult and Child 7, with Child 1’s self-selection for a turn not being accepted.

Extract 2: Educational conversational pattern.

Context: Adult-led topic. The adult introduced the topic by holding up the story book ‘The Very Hungry Caterpillar’ by Eric Carle (1974). The children were invited to contribute what they could remember of the story, which had been read to them on a previous occasion.

141 A: Caterpillar (.) what happened next in our story?

142 C1: He’s has to (.) he’s have some

143 A: You tell me

144 C1: Have [so:::me pear and plums

145 C7: [An chocolate cake and ice-cream

146 A: Chocolate cake and ice cream

147 C1: And (lollipops)

148 A: Lollipops

149 C5: ~(xxxx)~

150 A: And apple Well done C5 (.)

151 C7 : And sausage

152 A: Sausage

153 C1: I SAID A APPLE

154 A: Can anybody else remember

155 C1: Apple

156 A: Apple

157 C7: Orange

158 A: Orange

159 C9: Cheese

160 A: Cheese GO::d girl we hadn’t said cheese

161 C1: And [Chocolate

162 C7: [WatermELON

163 A: Wa::::termelon

164 C1: ((leans right in))

165 A: and chocolate cake

166 C5: (And milk)

167 A: Shall we have a look see if you’[re right ((turns page))

168 C5: [(milk)]

169 C5: (Milk)

170 C5: (Milk)

171 A: Some milk? Did he have some milk too?

172 C7: (xxx more) pears

173 A: We didn’t say plums

174 C7: Or that?

175 A: What’s that?

176 C1: I said

177 C (xx) ((several children talk at once))

178 A: Strawberry

179 C1: I SAID PLUMS

180 A: We’d forgotten strawberry hadn’t we

181 C4: I know (.) I know what’s that one ((points at picture))

182 A: What’s that one?

183 C4: A dark berry

184 A: Uhh A bLUEberry.

185 A: Let’s have a look over the page (.) let’s see if you rem::embered (.) shall we do all these together ((C1)) yes so [we can ((turns page))

Here, the adult took alternate turns, managing the conversation by repeating children’s responses that gave new information and offering children the opportunity to take a turn should they self-select. Throughout this extract, children self-selected turns, sometimes gaining the floor through overlapping speech such as Child 7 at line 145, or louder speech as shown by Child 1 at line 153/155 and 179. Five of the children contributed without prompting, including children with lower language levels: Child 1 and Child 5. Throughout Extract 2, children’s contributions were accepted but not expanded upon with further questioning, in contrast to the more informal conversation in Extract 1.

Extract 3: Mixed conversational pattern.

Context: Adult-led topic. The group had been introduced to cornflour and water play and eight children were playing together with the adult. The adult led the topic, by demonstrating what happened when the gloop was picked up and by adding colouring.

151 A: Mmmmm if we pick it up we can make it hard and then when we open our ha::nds what happens to it ((Holds it out towards C19))

152 C19: It me::lts

153 A: It melts

154 C19: Its like when you eat like when you put your ice cream in the sun it me?lts

155 A: It does that’s exa:ctly what it looks like (3)

156 C13: Look (1) Melting on my hands

157 C11: ((Laughs))

158 C11: Melting on ((C13))’s hands

159 C19: ((C18)) doesn’t want to touch it I’ll put it over here

160 A: Hold your hand out ((C11)) ((Puts some into C11’s hands))

161 C19: Ahhaha I’m trying to put it on ((C18))’s hands (4)

The adult introduced a topic but the pattern differed from that in Extract 2 in that the adult did not take every other turn, with turns being taken by several children before the adult would take another turn. In this example, Child 19 (higher language ability) extended the adult’s turn by adding new vocabulary at Line 152 and Line 154. This new vocabulary was then picked up by Child 11 (lower language ability) and Child 13 who used it themselves (Lines 156 and 158).

Reviewing the three types of conversational patterns showed examples of children picking up on previous turns in the conversation, but in different ways. In these examples, children with higher language levels and higher interaction rates developed ideas from the previous turns and generated their own ideas to build on them. For example, Child 7 extended from days of the week, to talk about his holidays in Extract 1. Child 19 added new vocabulary in Extract 3. These children elaborated spontaneously on their answers, without waiting for further questioning.

The direct repetition of vocabulary tended to be picked up from previous turns by children with lower language levels, for example, by Child 11 in Extract 3. This vocabulary came in some cases from other children and in other cases from the adult. It was more characteristic of talk from these children, that they did not spontaneously offer further information, tending only to provide a yes or no response. The adult management of turns encouraged these children to take part in the conversation. The adult would accept their turn, often expanding it, as, for example, for Child 5 in Extract 2. In certain cases, this would necessitate everyone being offered a turn by the adult or children being selected individually for a turn.

Discussion

Focussing particularly on children’s own use of language as well as that of the adult, this study addressed three research questions with regard to small group verbal interactions in the nursery. First, investigating who talks when and with what frequency, analysis of verbal initiation and response rates by adult and children revealed that the majority of interactions took place between adult and child and there were few extended interactions on the part of children. The second question investigated to what extent the likelihood of children taking up the opportunities offered to talk was linked to their language level on initial assessment. A positive correlation was found between children’s own rates of initiation, responses made and received as well as their initial language score. The higher a child’s language level, there was a tendency for them to speak more and to receive and make more responses in turn. Third, the question of how conversational experiences differed, then, for children of differing language levels was addressed. Qualitative analysis revealed different types of conversational pattern, varying in the degree to which the adult managed the topic and turn-taking. Children were shown to respond in different ways to informal or more formal features of conversation, indicating some differences in affordance of opportunity according to language level.

Turn-taking patterns and sustained conversation

Previous studies employing CA have found that a salient feature of conversations is their degree of formality (Drew and Heritage, 1992; Freebody, 2003). Conversations in the current study were also exemplified with regard to features of formality and informality. Jones and Thornborrow (2004) distinguished interactions as being more or less formal/educational by degree, rather than necessarily being categorised exclusively as one or the other. Similarly in the present study, as well as showing obvious features of either everyday or institutional patterns, some episodes illustrated a mixed pattern with features of each. Distinctive here was that the adult managed the turn-taking and control of topic at some points, but would allow contributions out of turn, and picked up on children’s contributions to permit a change in direction of topic. The adult’s role was one of intermediary: combining responses from different children, allowing children to direct responses on topics from other children through her and weaving them together. Although rare, at times in such conversations several turns took place between the children themselves, without the adult taking a turn. It was these episodes, showing features of both natural and educational conversation, which most closely resembled the co-constructing of learning with children or sustained shared thinking (Jordan, 2009; Siraj-Blatchford, 2009a).

The conversations selected as examples of sustained conversation included many of the features associated in the literature as encouraging children’s contribution to conversation and learning (e.g. Girolametto et al., 2003; Justice and Sofka, 2010). Strategies such as following the children’s lead, using comments to encourage turns, scanning the group to involve everyone, labelling, expanding and extending, pausing to wait for a response and use of open questioning were all shown on occasion to open up conversation and encourage children’s contributions. However, there were also occasions where these strategies were used, but children did not respond. The patterns seen here were a mix of natural conversation, presumably similar to conversation experienced at home, and more formal educational talk similar to patterns that are likely to be experienced as the child moves up through school. Where the two patterns combine, this may offer the best chance for extended conversation in the nursery. These patterns of talk could also be regarded as transitional for the child moving into the educational system and perhaps help to explain differences in language use between home and school (Flewitt, 2005; Tizard and Hughes, 2002 [1984]).

Children’s language level and affordance of opportunity

This study offers an indication of how conversational experiences may differently affect children’s language development. A simple count revealed higher rates of children’s own interactions associated with higher language levels. Lower language levels were, in turn, associated with a higher rate of initiations from the adult. Initial coding revealed that, even in Individual time where children made the highest rate of initiations and received the highest rate of responses, the overall rate of responses made by children in return was not higher. Further qualitative analysis using CA provided insight on this finding, however, in as much as children with higher language levels were observed to give follow-up information, both in response to the adult and often spontaneously. It was those children with lower language levels who more frequently gave one word responses, rarely offering more information even where they had initiated the interaction in the first place. It appeared that similar strategies were used by the adult initially, but as responses from the children were different according to the children’s language levels, the results were either more or less effective.

Support is offered for the presence of two important drivers in developing children’s language skills: opportunities to hear a rich vocabulary and to practise extended conversation (Dickinson et al., 2008; Hoff, 2006). Both factors feature strongly here, but with relative importance for different groups of children. Findings suggest that, in the absence of the impetus to talk, hearing language used and modelled in a carefully structured situation can be very supportive to children’s language development. It seems likely at least that these situations will stimulate language development more than persisting with attempts at individual conversation. Of relevance here also is the finding by Girolametto et al. (2006) that although educators showed increased frequency of talk to children with lower language levels (in this case with a language disorder), the result of this was more directions to behaviour and fewer language modelling utterances at an appropriate level of understanding. In these small group conversations, rather than stifling the child, the adult’s initiations can be seen as an attempt to encourage interaction, which the child may opt not to respond to.

In this study, a few of the children had very low rates of interaction and were observed to be reluctant to speak or shy. This is likely to have influenced both their test performance and their participation in conversation, resulting in lower levels on both measures (Crozier and Hostettler, 2003; Spere and Evans, 2009). For these children, the small group conversations did not afford frequent opportunities for verbal interaction. Martin (2009) described watching language being used as a form of apprenticeship towards becoming a language user. It would appear that these small group conversations may act as an illustration of polyadic conversations, supportive of children’s language without them necessarily participating verbally (cf. Floor and Akhtar, 2006). The group situation here acted to benefit children in different ways. Those children with lower language levels or less impetus to communicate were able to hear language modelled and follow models from others to make a contribution. Children with higher language levels spoke in more sustained shared conversation, learning to follow a given topic and listen to other’s contributions.

Limitations and further developments

Conclusions must be limited by the lack of ability to control for the effects of other situations in which children will be hearing, using and developing language. It is not possible to know what other opportunities for language learning the children may have been benefitting from at the same time. It is also important to remember that language will be influenced by context in both nursery observations and language assessments and can only act to demonstrate the children’s language skills within that particular context. There were also limitations in the degree to which the different affordances offered by each measure were taken into account. In using a standardised test, for instance, factors around children’s individual understandings of the test situation were not investigated.

Attempting to address the issue of language development through applying a mixed-methods approach also presented limitations. Compromises had to be made over sample size, to keep the observational data manageable while giving sufficient power to statistical comparisons required for quantitative analysis. This restricted the scale of the study, while possibly also restricting the in-depth exploration possible of individual needs and responses. Certain limitations need to be acknowledged in employing CA, which was not originally designed to look at learning as such and does not offer a theory about learning (Larsen-Freeman, 2004; Markee and Kasper, 2004). It focusses on the conversational process and the inter-subjective space as a site for potential learning, and makes observable particular types of behaviour which may be relevant to learning. Its value is in providing new insights and identifying affordances in a particular context. This in itself, however, does not necessarily indicate that anything carries forward from one context to the next, in the form of learning. It has been suggested that longitudinal studies would be a way of strengthening the potential of CA as a tool for exploring the processes involved in the learning of language (Larsen-Freeman, 2004; Larsen-Freeman and Cameron, 2008).

The value of this study lies in the evidence it provides about what can happen within particular circumstances, rather than in claims of generalisability. The adult involved may not have been typical of Early Years Practitioners in that she volunteered to take part, having an interest in developing good communication skills. The study does however demonstrate possibilities for patterns of interaction given similar circumstances. It is also likely that practitioners would need to take and adapt definitions, for instance of the discussion types. It is quite possible that working with other practitioners in other settings, a different system for categorisation may arise from the sessions observed.

Working with only one adult provided a constant factor in all the interactions, so that variation was due to factors in the situation other than differences between adult conversational partners. The fact that some of the practices used by the key worker in this study were the same as those identified as potentially facilitative in previous studies, lends weight to their potential representativeness. Involving only one key worker, however, also placed limits on the study in exploring the degree to which children’s own responses may vary in response to different conversational partners (see, for example, Ridley et al., 2002). With its focus on children’s own use of language, this study was restricted in its ability to compare responses to variation between adults in conversational initiation and response. Such comparison may act as a further source to illuminate the type of responses that can encourage children to develop conversational competence.

Conclusions and implications for practice

Considering affordance of opportunity, the present study helped to shed new light on how children may respond differently to the strategies used by the adult, perhaps because of their own language skills, preferred response style and understanding of the situation. This important finding shifts the focus from the adult to the whole context as a facilitator of communication. The value of the small group in enabling conversation to continue, particularly for those children with lower language levels, needs to be fully recognised as a vital aspect of the communication supportive Early Years environment. Practitioners need to use a skilful mix of informal and educational talk, supportive of turn-taking and models to copy, develop topics that are familiar and related to children’s own experiences and interests and act as intermediary to allow children to make connections. Providing a combination of natural conversation and more formal educational talk, such interactions also play a potentially important role in transition from home to nursery, as children are learning new ways of doing language.

Footnotes

Appendix 1

Appendix 2 Acknowledgements

Acknowledgement and thanks are extended to the staff, children and families at the nursery for taking part and allowing the privilege of listening in on their conversations.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Akhtar

(2005) The robustness of learning through overhearing. Developmental Science 8(2): 199–209.

Akhtar

Jipson

Callanan

(2001) Learning words through overhearing. Child Development 72: 416–430.

Botting

Simkin

Conti-Ramsden

(2006) Associated reading skills in children with a history of Specific Language Impairment (SLI). Reading and Writing 19: 77–98.

Bronfenbrenner

(1979) The Ecology of Human Development. Cambridge, MA: Harvard University Press.

Bruner

(1974) From communication to language: a psychological perspective. Cognition 3: 255–287.

Cabell

Justice

Piasta

. (2011) The impact of teacher responsivity education on preschoolers’ language and literacy skills. American Journal of Speech-Language Pathology 20: 315–330.

Carle

(1974) The very hungry caterpillar. Harmondsworth, Middlesex: Puffin Books.

Catts

Fey

Tomblin

. (2002) A longitudinal investigation of reading outcomes in children with language impairments. Journal of Speech, Language, and Hearing Research 45: 1142–1157.

Crozier

Hostettler

(2003) The influence of shyness on children’s test performance. British Journal of Educational Psychology 73(3): 317–328.

10.

Department for Children Schools and Families (DCSF) (2008) Every child a talker. Reference: 00854-2008DOM-EN. Nottingham: The National Strategies Early Years Department for Children Schools and Families.

11.

Department for Education and Skills (DfES) (2006) Communicating matters. Reference: 2026-2006PCL-EN. London: DfES.

12.

Dickinson

Tabors

(2001) Fostering language and literacy in classrooms and homes. Young Children 57(2): 10–18.

13.

Dickinson

Watson

Farran

(2008) It’s in the details: approaches to describing and improving preschool classrooms. In: Justice

Vukelich

(eds) Achieving Excellence in Preschool Literacy Instruction. New York: The Guilford Press, pp. 136–162.

14.

Drew

Heritage

(1992) Talk at Work: Interaction in Institutional Settings. Cambridge: Cambridge University Press.

15.

Dunn

Shatz

(1989) Becoming a conversationalist despite (or because of) having an older sibling. Child Development 60(2): 399–410.

16.

Flewitt

(2005) Is every child’s voice heard? Early Years 25(3): 207–222.

17.

Floor

Akhtar

(2006) Can 18-month-old infants learn words by listening in on conversations? Infancy 9(3): 327–339.

18.

Foster

Hund

(2012) The impact of scaffolding and overhearing on young children’s use of the spatial terms between and middle. Journal of Child Language 39(2): 338–364.

19.

Freebody

(2003) Qualitative Research in Education, Interaction and Practice. London; Thousand Oaks, CA; New Delhi, India: SAGE Publications.

20.

Gampe

Liebal

Tomasello

(2012) Eighteen-month-olds learn novel words through overhearing. First Language 32(3): 385–397.

21.

Girolametto

Weitzman

Greenberg

(2003) Training day care staff to facilitate children’s language. American Journal of Speech-Language Pathology 12: 299–311.

22.

Girolametto

Weitzman

Greenberg

(2006) Facilitating language skills: inservice education for early childhood educators and preschool teachers. Infants and Young Children 19(1): 36–49.

23.

Goldfield

Reznick

(1990) Early lexical acquisition: rate, content and the vocabulary spurt. Journal of Child Language 17: 171–184.

24.

Griffin

Hemphill

Camp

. (2004) Oral discourse in the preschool years and later literacy skills. First Language 24: 123–147.

25.

Hansen

Joshi

(2007) Millennium Cohort Study Second Survey: A User’s Guide to Initial Findings. London: Centre for Longitudinal Studies.

26.

Hart

Risley

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

27.

Have

(2007) Doing Conversation Analysis. 2nd ed. Los Angeles, CA; London; New Delhi, India; Singapore: SAGE Publications.

28.

Hoff

(2006) How social contexts support and shape language development. Developmental Review 26(1): 55–88.

29.

Hoff

(2010) Context effects on young children’s language use: the influence of conversational setting and partner. First Language 30(3–4): 461–472.

30.

Howlin

Mawhood

Rutter

(2000) Autism and developmental receptive language disorder – a follow-up comparison in early adult life, II: social, behavioural and psychiatric outcomes. Journal of Child Psychology and Psychiatry 41: 561–578.

31.

I-CAN (2012) Early language development programme. Available at: http://www.talkingpoint.org.uk/early-language-development-programme-eldp-0 (accessed 24th September 2014)

32.

Jones

Thornborrow

(2004) Floors, talk and the organization of classroom activities. Language in Society 33(3): 399–423.

33.

Jordan

(2009) Scaffolding learning and co-constructing understandings. In: Anning

Cullen

Fleer

(eds) Early Childhood Education: Society and Culture. London: SAGE Publications, pp. 39–52.

34.

Justice

Sofka

(2010) Engaging Children with Print: Building Early Literacy Skills Through Quality Read-Alouds. New York: The Guilford Press.

35.

King

Saxton

(2010) Opportunities for language development: small group conversations in the nursery class. Educational and Child Psychology 27(4): 31–44.

36.

Larsen-Freeman

(2004) CA for SLA? It all depends. The Modern Language Journal 88(4): 603–607.

37.

Larsen-Freeman

Cameron

(2008) Research methodology on language development from a complex systems perspective. The Modern Language Journal 92(2): 200–2013.

38.

Lewis

Miller

(2011) ‘Institutional talk’ in the discourse between an educational psychologist and a parent: a single case study employing mixed research methods. Educational Psychology in Practice 27(3): 195–212.

39.

Lindsay

Dockrell

Strand

(2007) Longitudinal patterns of behaviour problems in children with specific speech and language difficulties: child and contextual factors. British Journal of Educational Psychology 77: 811–828.

40.

Loban

(1964) Language Ability: Grades Seven, Eight and Nine. Berkeley, CA: University of California.

41.

Lonigan

Allan

Lerner

(2011) Assessment of preschool early literacy skills: linking children’s educational needs with empirically supported instructional activities. Psychology in the Schools 48(5): 488–501.

42.

Markee

Kasper

(2004) Classroom talks: an introduction. The Modern Language Journal 88(4): 491–500.

43.

Martin

(2009) Language Disabilities in Cultural and Linguistic Diversity. Bristol; Buffalo, NY; Toronto, ON, Canada: Multilingual Matters.

44.

Mercer

(2010) The analysis of classroom talk: methods and methodologies. British Journal of Educational Psychology 80(1): 1–14.

45.

Muter

Hulme

Snowling

. (2004) Phonemes, rimes, vocabulary and grammatical skills as foundations of early reading development: evidence from a longitudinal study. Developmental Psychology 40: 665–681.

46.

Ochs

Schieffelin

(1995) The impact of language socialization on grammatical development. In: Fletcher

MacWhinney

(eds) The Handbook of Child Language. Oxford: Blackwell, pp. 73–94.

47.

Pence

Justice

Wiggins

(2008) Preschool teachers’ fidelity in implementing a comprehensive language-rich curriculum. Language, Speech, and Hearing Services in Schools 39: 329–341.

48.

Pike

(2010) Intersubjectivity and misunderstanding in adult-child learning conversations. In: Gardner

Forrester

(eds) Analysing Interactions in Childhood: Insights from Conversation Analysis. Chichester: Wiley-Blackwell, pp. 163–182.

49.

Radford

Blatchford

Webster

(2011) Opening up and closing down: how teachers and TAs manage turn-taking, topic and repair in mathematics lessons. Learning and Instruction 21: 625–635.

50.

Radford

Ireson

Mahon

(2006) Triadic dialogue in oral communication tasks: what are the implications for language learning? Language and Education 20(3): 191–210.

51.

Ridley

Radford

Mahon

(2002) How do teachers manage topic and repair? Child Language Teaching and Therapy 18(1): 43–58.

52.

Rogoff

(1998) Cognition as a collaborative process. In: Damon

Kuhn

Siegler

(eds) Cognition, Perception and Language: Handbook of Child Psychology. 5th ed., vol. 2. New York: Wiley, pp. 679–774.

53.

Rogoff

(2003) The Cultural Nature of Human Development. Oxford; New York: Oxford University Press.

54.

Roth

Speece

Cooper

(2002) A longitudinal analysis of the connection between oral language and early reading. Journal of Educational Research 95: 259–272.

55.

Roulstone

Law

Rush

. (2011) The role of language in children’s early educational outcomes. Research Brief for Department of Education, DfE-RB 134. Washington, DC: Department of Education.

56.

Sacks

Schegloff

Jefferson

(1974) A simplest systematics for the organisation of turn-taking for conversation. Language 50: 696–735.

57.

Scofield

Behrend

(2011) Clarifying the role of joint attention in early word learning. First Language 31(3): 326–341.

58.

Shneidman

Goldin-Meadow

(2012) Language input and acquisition in a Mayan village: how important is directed speech? Developmental Science 15(5): 659–673.

59.

Shneidman

Buresh

Shimpi

. (2009) Social experience, social attention and word learning in an overhearing paradigm. Language Learning and Development 5(4): 266–281.

60.

Siraj-Blatchford

(2009a) Conceptualising progression in the pedagogy of play and sustained shared thinking in early childhood education: a Vygotskian perspective. Educational and Child Psychology 26(2): 77–89.

61.

Siraj-Blatchford

(2009b) Quality teaching in the early years. In: Anning

Cullen

Fleer

(eds) Early Childhood Education: Society and Culture. London: SAGE, pp. 147–157.

62.

Siraj-Blatchford

Sylva

Taggart

. (2003) The Effective Provision of Pre-School Education (EPPE) Project Technical Paper 10: Intensive Case Studies of Practice Across the Foundation Stage. London: Institute of Education, University of London.

63.

Snowling

Gallagher

Frith

(2003) Family risk of dyslexia is continuous: individual differences in the precursors of reading skill. Child Development 74: 358–373.

64.

Social Disadvantage Research Centre (2009) Tracking Neighbourhoods: The Economic Deprivation Index 2008. Available at: https://www.gov.uk/government/statistics/tracking-economic-and-child-income-deprivation-at-neighbourhood-level-in-england-1999-to-2009 (accessed 24th September 2014)

65.

Spere

Evans

(2009) Shyness as a continuous dimension and emergent literacy in young children: is there a relation? Infant and Child Development 18: 216–237.

66.

Strapp

Federico

(2000) Imitations and repetitions: what do children say following recasts? First Language 20(60): 273–290.

67.

Stringer

Clegg

(2006) Language, behaviour and social disadvantage. In: Clegg

Ginsborg

(eds) Language and Social Disadvantage: Theory into Practice. Chichester: John Wiley & Sons, pp. 93–105.

68.

Stringer

Lozano

(2007) Under identification of speech and language impairment in children attending a special school for children with emotional and behavioural disorders. Educational and Child Psychology 24(4): 9–19.

69.

Sylva

Taggart

Siraj-Blatchford

. (2007) Curricular quality and day-to-day learning activities in pre-school. International Journal of Early Years Education 15(1): 49–65.

70.

Tizard

Hughes

(2002 [1984]) Young Children Learning. 2nd ed. Oxford: Blackwell.

71.

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 & Development 20(1): 53–79.

72.

Van Lier

(2000) From input to affordance: social-interactive learning from an ecological perspective. In: Lantolf

(ed.) Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press, pp. 245–259.

73.

Vygotsky

(1986 [1934]) Thought and Language (trans. Kozulin

). Cambridge, MA: MIT Press.

74.

Wasik

Bond

Hindman

(2006) The effects of a language and literacy intervention on head start children and teachers. Journal of Educational Psychology 98(1): 63–74.

75.

Wells

(1981) Learning Through Interaction: The Study of Language Development. Cambridge: Cambridge University Press.

76.

Wiig

Secord

Semel

(2006) Clinical Evaluation of Language Fundamentals Pre-School 2nd UK ed. Oxford, UK: Pearson.