Contexts of reading instruction: Implications for literacy skills and kindergarteners’ behavioral engagement

Early Childhood Research Quarterly 26 (2011) 157–168

Contents lists available at ScienceDirect

Early Childhood Research Quarterly

Contexts of reading instruction: Implications for literacy skills and kindergarteners’ behavioral engagement夽 Claire Cameron Ponitz ∗ , Sara E. Rimm-Kaufman University of Virginia, United States

a r t i c l e

i n f o

Article history: Received 25 November 2009 Received in revised form 7 October 2010 Accepted 16 October 2010 Keywords: Behavioral engagement Child by literacy instruction interactions Classroom observations Emergent word-reading Kindergarten Phonological awareness

a b s t r a c t This observational study examined kindergarteners’ (n = 170) exposure to literacy instruction in their classrooms (n = 36), child-by-instruction interactions, and behavioral engagement in relation to literacy skills. Time spent in four instructional contexts was coded according to who managed children’s attention (teacher-managed, TM or child-managed, CM), and the content focus (basic skills such as teaching letters and their sounds, or meaning-focused such as discussing a book); children’s behavioral engagement and off-task behavior were also coded live five times over the year. Word-reading and phonological awareness skills were assessed in fall and spring. Hierarchical Linear Modeling results indicated that kindergarteners with lower initial skills gained more in word-reading, but not phonological awareness, when they were exposed to relatively more time in TM basic skills instruction. In contrast, more time in CM meaningfocused instruction did not interact with initial skills to predict either outcome. Engagement analyses indicated that students were more likely to be off-task in CM than in TM contexts. Children who spent more time off-task during TM contexts had lower spring scores on both outcomes. Discussion explores the implications of this work for both literacy learning and behavioral engagement in the transition year of kindergarten. © 2010 Elsevier Inc. All rights reserved.

Learning to read is the principal task of the elementary school years. Many children struggle with this task, and those who fall behind early face an uphill challenge as they proceed through school. Since No Child Left Behind (NCLB) was signed into law in 2002, school accountability policies have heavily targeted elementary reading achievement. Some states responded to NCLB by requiring research-based literacy curricula and teacher practices, mandating minimum amounts of classroom time devoted to reading instruction, and providing targeted instruction for struggling readers early in their school experience. Perhaps as a consequence of these efforts, the 2007 National Assessment of Educational Progress (NAEP) showed overall improvement in Grade 4 reading compared to 2005 results (Lee, Grigg, & Donahue, 2007). Yet, reading problems are far from resolved. NAEP gains from 2005 to 2007 occurred in only 18 states; in the other 32 states, no change

夽 The Institute of Education Sciences, U.S. Department of Education is gratefully acknowledged for its fellowship support to the first author through the University of Virginia (R305B060009). This work was also funded by a grant from the National Science Foundation-Developmental and Learning Sciences #0418469 to the second author. The opinions expressed are those of the authors and do not represent views of the U.S. Department of Education or the National Science Foundation. ∗ Corresponding author at: CASTL, University of Virginia, 2200 Old Ivy Road, Charlottesville, VA 22903, United States. Tel.: +1 434 982 6965; fax: +1 434 243 0533. E-mail address: [email protected] (C.C. Ponitz). 0885-2006/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ecresq.2010.10.002

occurred. Further, over 30% of fourth-graders still do not read at a basic level. Multiple explanations for the discrepancy between policy, instructional mandates, and literacy outcomes are possible. We apply ecological perspectives from research on early literacy acquisition (Connor, Son, Hindman, & Morrison, 2005; Morrison, Bachman, & Connor, 2005). Ecological theorists seek to understand how child and environmental factors separately, and through complex interactions, contribute to development (Cairns & Rodkin, 1998). To learn to read, children must master the oral and written language system that humans use to communicate (Bialystok, 1995). This system includes component skills that vary within children, including vocabulary, syntactic knowledge, phonological awareness, orthographic/alphabet knowledge, insight to the alphabetic principle, and automaticity with written code (Catts, Fey, Zhang, & Tomblin, 1999). A growing base of evidence suggests that most children require deliberate instruction to acquire the component skills involved in reading (Adams, 1994; Foorman, Francis, Fletcher, Schatschneider, & Mehta, 1998; Rayner, Foorman, Perfetti, Pesetsky, & Seidenberg, 2001). Furthermore, researchers have described, with increasing detail and specificity, the environmental (i.e., instructional) experiences associated with literacy competence (Connor, Piasta, et al., 2009; Morrison et al., 2005). The centrality of instruction in reading development turns our attention to the social system of homes, classrooms, and schools in

158

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

Table 1 Examples of kindergarten literacy activities using a dimensional approach to instruction. Teacher-managed (TM)

Child-managed (CM)

Basic skills

Teacher leads a rhyming game with children Teacher holds up a picture and children name the beginning letter of the noun represented in the picture (e.g., what letter does dog start with?) Teacher leads children in the alphabet song

Children complete worksheets where they circle words that start with a specific first letter (e.g., circle the words that start with “B”) Children complete a worksheet finding the objects on a page that include a certain sound Children complete a hand-writing worksheet

Meaning-focused

Teacher reads a book with children and asks them to predict what will come next Teacher explains the meaning of a new vocabulary word

Children look at books silently on their own

Teachers lead the children in acting out a story

which children learn to read (Coyne, Kame’enui, & Simmons, 2001). Most work has used parent- and teacher-reported practices or has targeted the amount and type of instruction in classroom contexts to predict children’s literacy outcomes (Connor et al., 2005; Sénèchal, 2006; Wasik, Bond, & Hindman, 2006). For example, an emerging base of both observational and experimental research shows that the optimal approach to literacy instruction depends upon children’s baseline language and/or literacy abilities (Connor, Morrison, & Katch, 2004; Connor, Morrison, & Slominski, 2006; Connor, Piasta, et al., 2009; Juel & Minden-Cupp, 2000). Yet children are not passive recipients of various instructional experiences but rather, are active participants in the learning process (Greenwood, Horton, & Utley, 2002). The present study examines children’s behavioral engagement in four contexts of literacy instruction, and contributions to emergent word-reading and phonological skill improvement during kindergarten, a developmentally challenging transition year (Pianta, Cox, & Snow, 2007). 1. Dimensions of literacy instruction Like others, we use a multi-dimensional approach to conceptualize and measure reading instruction (Connor, Morrison, et al., 2009). The first of two dimensions – basic skills versus meaningfocused – pertains to the content of what is being taught. 1.1. Basic skills versus meaning-focused instruction Basic skills instruction focuses explicitly on the skills that underlie fluent word-reading such as rhyming, segmenting, and matching letters to sounds (i.e., the alphabetic principle). Instruction in phonological skills seems particularly important for later reading proficiency, especially for children at risk for poor literacy outcomes (Juel, 1988; Rayner et al., 2001). Observational and experimental studies have shown that first graders with low initial word-reading skills improve more when teachers provide direct, explicit instruction in basic decoding, such as helping children to recognize letters and the sounds in letters and words (Connor, Morrison, & Katch, 2004; Connor, Piasta, et al., 2009; Juel & Minden-Cupp, 2000). In contrast, meaning-focused activities explicitly target skills such as listening and reading comprehension, problem-solving through text, and extracting meaning from what is read. In an advantaged preschool sample, Connor et al. (2006) found that preschoolers with strong initial word-reading (i.e., alphabet knowledge, word recognition) gained more in word-reading skill when they spent relatively more time in teacher-managed activities where comprehension and meaning-making was the goal (e.g., teacher-led book-readings). Note that this dimension varies according to its explicit focus (Connor, Morrison, et al., 2009). In other words, it is possible for students to learn new vocabulary during a rhyming exercise, and they may learn about sound-symbol correspondence as they read for understanding (Olson, Wise, Forsberg, Sameroff,

Children create illustrations that correspond to a sentence provided by the teacher Children write in “journals” about their experiences

& Haith, 1996). In the present study, however, the basic skills or meaning-focused distinction is based on the explicit (versus implicit) focus of observed instructional activity. 1.2. Teacher-managed versus child-managed instruction The second dimension of instruction designates who manages children’s attention—the teacher or the child. In teacher-managed (TM) instruction, the teacher focuses the child’s attention on the target activity, such as in whole-class discussion or individualized tutoring. In contrast, child-managed (CM) instruction occurs when students themselves focus their own attention and are responsible for carrying out an activity. Examples include looking at books in the library corner or completing worksheets independently. This distinction differs from the commonly used terms of teacher- and child-directed, which pertain to who selected the activity, regardless of how instruction is actually delivered (Stipek et al., 1998). Because of our interest in kindergarteners’ attention and participation, we use the teacher- versus child-managed distinction to highlight activities that are initiated and sustained by the teacher or the child. Taken together, the two dimensions – basic skills or meaningfocused, and teacher- or child-managed instruction – create four possible instructional contexts with regard to literacy: (1) TM basic skills (2) TM meaning-focused, (3) CM basic skills, and (4) CM meaning-focused. Examples of kindergarten literacy activities within these four contexts are shown in Table 1. 1.3. Links among dimensions of literacy instruction and word-reading in early childhood The dimensional approach to reading instruction has been used to explain how instruction relates to literacy outcomes at multiple grade levels, including preschool (where children would be considered pre-readers or emergent readers), first grade (with some readers and some non-readers), and third grade (where most children would be considered readers but with varying levels of comprehension) (Connor, Morrison, & Katch, 2004; Connor, Morrison, & Petrella, 2004; Connor et al., 2006). These observational and experimental studies suggest that literacy instructional dimensions interact with child characteristics to predict achievement. In general, spending more time in CM meaning-focused contexts predicts fall-spring improvement for children who have a sturdy base of word-reading skill (Connor, Morrison, & Katch, 2004). This makes sense because students who already identify letters and decode new words can rely on that knowledge even when basic skills are not an explicit focus of instruction, as in sustained silent reading (SSR) or shared story-book reading. In contrast, research has shown that greater exposure to TM basic skills instruction predicts improvement for students with weaker initial word-reading skills (Connor, Morrison, & Katch, 2004; Juel & Minden-Cupp, 2000).

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

In addition to focusing on the specific literacy skills that beginning readers need to acquire, TM basic skills instruction may be important for another reason: by directing students’ attention, the teacher alleviates the cognitive burden on children of having to plan and decide for themselves what to look at or do, as in CM instruction (Lillard, 2005). Instead, in TM activity, the teacher provides a highly structured setting in which behavioral and academic expectations are clear and often overlapping. For example, in a whole-group book-reading activity, children sit on the floor and interact with the teacher as she reads to them. Content-related and behavioral instructions are interwoven elements of this common activity (Gerde & Powell, 2009). In contrast, in CM literacy instruction, academic expectations may be routine (“go in the library corner and look at your book”) but behavioral expectations opaque, such as an unspoken teacher expectation that children will not engage in rough and tumble play. Or, because the teacher is absent in the activity’s actual execution, young children may forget the instructions and become distracted (Rimm-Kaufman, Pianta, & Cox, 2000). In one study of context across content areas, more kindergarteners were observed off-task during CM time as compared to TM time (Rimm-Kaufman, La Paro, Downer, & Pianta, 2005); however, that study did not focus on literacy.

2. Academic and attentional implications of kindergarten Kindergarten is an important year not only because of its implications for reading achievement, but also because classroom demands stress the young child’s emerging ability to self-regulate and effectively engage in learning activities (McClelland, Morrison, & Holmes, 2000; Pianta et al., 2007; Rimm-Kaufman et al., 2000). Students enter school with a wide range of prior experiences, which relate to variability in how well children manage the school transition (Morrison et al., 2005; Pianta et al., 2007). Many children must first develop behavioral and attention skills in order to achieve the literacy competence that is an explicit focus of early elementary school (McClelland et al., 2000; Piotrkowski, Botsko, & Matthews, 2000; Stipek, 2006). Preschool and kindergarten teacher reports of engaged behaviors, as well as performance on individual tests of self-control and self-regulation, predict a host of outcomes including literacy skills, even after initial skill levels are controlled (Bodovski & Farkas, 2007; Fantuzzo et al., 2007; Matthews, Ponitz, & Morrison, 2009; McClelland et al., 2000; Ponitz, Rimm-Kaufman, Brock, & Nathanson, 2009; Ready, LoGerfo, Burkam, & Lee, 2005). Further, children identified as cooperative and participating in kindergarten showed the steepest scholastic growth between ages five and 13 (Ladd & Dinella, 2009). Separate literatures point to the importance of appropriate instructional experiences, but also suggest that children must actually participate in provided activities if they are to acquire literacy (Pressley et al., 2001). Behavioral engagement in reading activity, specifically, is an important contributor to literacy achievement, but this has been primarily documented in the later elementary school years (Guthrie, Schafer, & Huang, 2001). Behavioral engagement means that when literacy instruction is offered, students attend to and execute the intended activities, rather than pursuing off-task activities such as gossiping or looking out the window. This definition of behavioral engagement is situated in the framework of research on the broad engagement construct. In a review of available studies, primarily with students in late elementary and middle school, Fredricks, Blumenfeld, and Paris (2004) describe time-on-task or behavioral engagement (our present focus), as well as emotional and cognitive engagement. Fredricks et al. note that of the three types, behavioral engagement is easiest to observe, whereas measuring emotional and cognitive engagement typically requires reliable self-reports from students, which would lack

159

validity with kindergarten-age children. This paper examines the extent to which kindergarteners’ observed behavioral engagement in afforded learning opportunities varies by instructional context and predicts reading achievement.

3. Rationale and hypotheses Growing evidence indicates that child characteristics interact with instruction to predict literacy achievement, though few studies have explored this issue in kindergarten (Connor & Morrison, 2005; Connor, Piasta, et al., 2009). One available inquiry used data from the Early Childhood Longitudinal Study of Kindergarten (ECLS-K; Xue & Meisels, 2004). Children with higher initial vocabulary, comprehension, and basic word-reading skills, whose teachers reported relatively greater use of meaning-focused activities (TM or CM context undetermined), improved more on a standardized measure of word-reading, compared to kindergarteners with lower initial language and literacy ability. Teacher-reported use of basic skills activities did not interact with initial ability to predict achievement. ECLS-K data relied on teacher reports of their instructional approach however, which were not individualized to child participants. The present study sought to bridge a gap between the literatures documenting child-by-literacy-instruction interactions and children’s behavioral engagement during a milestone year. Because of the importance of decoding skills early in the reading trajectory (Aram, 2005; Rayner et al., 2001; Spira, Bracken, & Fischel, 2005), outcomes were kindergarteners’ performance on assessments of emergent word-reading skill (e.g., letter knowledge, word recognition) and phonological awareness (e.g., rhyming, syllabic segmentation). This observational study was conducted in an under-examined, typically under-resourced population (Brody & Flor, 1998), participants living in a rural area, to replicate and extend prior studies of child-by-instruction interactions found in urban and suburban settings (Connor, Morrison, & Katch, 2004; Juel & Minden-Cupp, 2000). Many residents of rural communities lack access to social and infrastructural services that are prevalent in urban areas, including high-quality early education programs and safe areas to play (De Marco, 2008; Hessler, 2009). Furthermore, relatively little research has been done within rural populations (Tickamyer & Duncan, 1990). One recent investigation, also using the ECLS-K sample, reported that reading scores were lower in rural as compared with urban areas, for the reasons described above (Durham & Smith, 2006). However, results also indicated complex relations among SES, rurality, and reading, which varied by ethnic group. Low SES was not as detrimental for reading scores of white children living in rural as compared to urban areas, whereas the opposite pattern was true for African American children. Even within the present sample of primarily Caucasian rural families, child and family characteristics varied. Research suggests socio-demographic factors tend to be negatively associated with children’s achievement, and contribute to child outcomes through additive effects (McLoyd, 1998; Sameroff, Bartko, Baldwin, Baldwin, & Seifer, 1998). Thus, our analyses controlled for children’s age and gender, plus a composite of four indicators of sociodemographic risk (unmarried parent, educational attainment below high school, low income status, and not attending preschool). We posed three research questions: First, does instructional time spent in four contexts (TM basic skills, TM meaning-focused, CM basic skills, CM meaning-focused) predict kindergarten word-reading and phonological awareness? If so, do relations operate through child-by-instruction interactions similar to those found in preschool and first grade (Connor, Morrison, & Katch, 2004; Connor, Morrison, Fishman,

160

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

Schatschneider, & Underwood, 2007)? Based on prior work, we predicted that children with weak word-reading skills at kindergarten entry would show greater word-reading achievement at year’s end when they were observed in relatively more TM basic skills activity such as teacher-led rhyming. Following from Xue and Meisels (2004), kindergarteners who began school with strong word-reading were expected to show greater spring achievement when they participated in relatively more CM meaning-focused activity, such as SSR. Though no studies of which we are aware have examined child-by-instruction interactions in relation to phonological awareness, because these are also basic literacy skills, we tentatively hypothesized the same pattern of findings for phonological awareness as for word-reading. Second, does children’s behavioral engagement vary depending on the four literacy contexts? TM contexts offer explicit support for student engagement in learning, whereas CM contexts hold students responsible for their own participation (Rimm-Kaufman et al., 2005). In this study of kindergarteners, we expected to see greater behavioral engagement in conditions with relatively more support and structure. Thus, we hypothesized that a larger proportion of children’s TM time would be on-task time, compared to CM time. Third, does behaviorally engaged time, rather than overall time, in the four literacy contexts predict achievement through child-byinstruction interactions? We expected that accounting for the time children spent engaged (versus off-task) in the four literacy contexts would explain additional variance, beyond the contribution of total time spent in each context.

4. Method Child participants were recruited to participate between April and September prior to their entrance into kindergarten in one of seven elementary schools in four rural districts in a midAtlantic state. The research team recruited families in person during kindergarten registration and fall open house meetings. The school districts were within a 90-min drive of the university town and thus represented a sample of convenience; steps were also taken to ensure that the sample represented the population. Parents of 333 children signed consent forms, representing roughly 60% of the children enrolled in kindergarten that fall. Between four or five children per classroom were selected randomly from this larger pool, resulting in a sample of 170 children. Selected and non-selected participants did not differ on gender, parent marital status, income, or maternal educational attainment, according to Chi-squared analyses. Child participants were 170 kindergarteners (including 92 males; 143 Caucasians, 23 African-Americans, and 5 children of other ethnic groups), who were on average 5.42 years old (range 4.67–6.33 years) at kindergarten entry. About one-third of families (n = 57 of 170) selected income ranges below $29,999; another onethird (n = 56 of 170) reported salaries between $30K and $49,999; and the final group of families (n = 54 of 170) reported salaries above $50K (three families did not answer this question). Maternal education ranged from less than high school (11 mothers and 29 fathers) to a doctoral degree (one mother); the median parent education level was a high school degree. Children were members of 36 classrooms, taught by 36 kindergarten teachers (35 females, 1 male; 35 Caucasian, 1 Hispanic). Teachers had on average 18.1 years of experience (with a range of 1–37 years). Thirty-one teachers were fully certified and licensed by the state; five had provisional licensure. All had attained Bachelor’s degrees and 11 had earned Master’s degrees. Teachers provided separate consent to participate.

4.1. Procedures Three sources of data were collected: parent and teacher surveys, classroom observations, and direct assessments. First, parents completed questionnaires at kindergarten registration and teachers completed a short demographic survey at the end of the year. Second, pairs of reliable research assistants collected observational data five times during the year. Research assistants coded multiple aspects of classroom activity, including the amount of time that study children spent in the four instructional contexts and their behavioral engagement in these contexts. Third, as part of a 25-min battery, research assistants administered standardized tests of literacy skills to individual students in fall and spring. See Table 2 for descriptive statistics and Table 3 for correlation coefficients among all independent and dependent variables. 4.2. Measures 4.2.1. Socio-demographic characteristics. Parents answered questions about child gender and family characteristics. To measure socio-demographic risk, a composite was created from four items, scored one if the risk was present: marital status of parents (risk = unmarried including single/never married, divorced, or widowed), years of maternal education (risk = a high school degree or below), household income (risk = less than $30K/year), and preschool attendance (risk = did not attend). Items were then summed, resulting in a composite score indicating cumulative risk. The risk variable for analyses had a mean value of 1.66 and ranged 0–4 (see Table 2). 4.2.2. Classroom observation of children’s literacy instruction and behavioral engagement Based on a review of the relevant literatures, an observational coding system and accompanying manual to document children’s exposure to and behavioral engagement in learning contexts was developed for the present study (Rimm-Kaufman, Klein, & Romberg, 2007). Observations were conducted at the child level and coded on laptops using a software program written for the study (A.P. Rimm-Kaufman & Rimm-Kaufman, 2006). Researchers identified the intended instructional context, noting the explicit purpose of the activity (basic skills or meaning-focused), who managed children’s attention (TM or CM), and children’s behavioral engagement (engaged, off-task, or other) simultaneously. A description of relevant codes from the coding manual is included in Table 4. Inter-observer reliability was established for all codes using 23 ten-minute observations and was strong (ICC = 0.97). To obtain a representative sample of children’s exposure to and participation in literacy activities, each child participant was individually coded in 10-min blocks on five different days throughout the year (two days in fall, two days in winter, and one day in spring). Observations were conducted in the morning, which included literacy instructional time. Other activities were also observed, such as transition and snack, which were coded but not a focus of the present study. The coding software tracked how many seconds children spent in each activity context until a new activity was coded; it also tracked how many seconds they spent in each type of behavioral engagement until a new type of engagement was identified. The software compiled data for each of the four instructional literacy contexts separately. Obtained variables represented the total number of seconds (converted to minutes) that children spent in the four literacy instructional contexts, the number of engaged minutes (including active and passive engagement), the number of off-task minutes (including off-task and not involved), and the number of minutes in another type of engagement (e.g., engaged in a different, non-intended academic activity, or engagement not determined; see Table 4). Except for seven children observed

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

161

Table 2 Descriptive statistics of all independent and dependent variables (N = 170). M (percent) Background

Teacher-managed Reading, basic skills

Reading, meaning-focused

Child-managed Reading, basic skills

Reading, meaning-focused

Literacy achievement

Fall age in years Male (coded 1; female = 0) Maternal education (below HS = 1; other = 0) Family income (below $30K/year = 1; other = 0) Not married (single/divorced/widowed) Did not attend preschool Sum of four sociodemographic risks

5.42 54% 68% 35% 26% 39% 1.66

Total minutes (% of total observed time) Engaged minutes Off-task minutes Other type of engaged minutes Total minutes (% of total observed time) Engaged minutes Off-task minutes Other type of engaged minutes Total minutes (% of total observed time) Engaged minutes Off-task minutes Other type of engaged minutes Total minutes (% of total observed time) Engaged minutes Off-task minutes Other type of engaged minutes Fall letter-word identification (w-score) Spring letter-word identification (w-score) Fall sound awareness (w-score) Spring sound awareness (w-score)

for 40 min, the sum was based on 50 possible minutes the child could have been observed in any activity, including literacy and non-literacy time; for total observed minutes, M = 49.59, SD = 1.99. ANOVAs indicated that the seven children observed for 40 min did not differ by any family risk variable or by gender, all ps > .30.

4.2.3. Direct assessments of emergent word-reading and phonological awareness Emergent word-reading skills were measured using two common subtests from the Woodcock–Johnson III Tests of Achievement (WJ III; Woodcock, McGrew, & Mather, 2001). Letter-word identification assessed letter knowledge of lower-case and upper-case letters, and recognition of actual words. Children’s knowledge of the sounds comprising words was measured using a composite score of the four subtests of the WJ III sound awareness. Part A focused on rhyming skills, such as identifying two rhyming words of three options. Part B required syllabic and phonemic deletion, such as saying “football” without the word “foot.” Part C tested substitu-

SD

Range

0.30 – – – – – 1.11

4.70–6.24 0–1 0–1 0–1 0–1 0–1 0–4

5.06 (10%) 4.54 0.43 0.09 6.54 (13%) 5.90 0.56 0.07

4.50 (9%) 4.05 0.80 0.22 5.27 (11%) 4.81 1.36 0.14

0–24.83 (0–50%) 0–21 0–5.15 0–1.85 0–21.27 (0–43%) 0–18.8 0–9.5 0–0.87

2.95 (6%) 2.42 0.43 0.10 4.79 (10%) 3.60 1.05 0.14 351.46 407.98 455.55 473.42

4.28 (9%) 3.65 1.03 0.55 5.82 (12%) 4.63 1.95 0.37 23.08 23.14 13.02 12.37

0–21.93 (0–44%) 0–17.15 0–6.9 0–5 0–22.22 (0–44%) 0–19.77 0–11.75 0–2.57 293–460 336–477 427–496 440–511

tion, such as changing the/sh/sound in “shoe” to/t/. Part D required children to reverse syllables and phonemes, such as saying football backwards or reversing “pat” to “tap.” For 5-year-olds, reliability coefficients on both tests are reported above .85 (McGrew & Woodcock, 2001). In the fall and spring, Form A and Form B of the WJ III were given, respectively. Items increased in difficulty until the child responded to a certain number of items incorrectly, when testing ceased.

5. Results Before proceeding to the research questions, we examined descriptive data. These indicated that on average, children spent several minutes of the 50 possible observed minutes in each of the four literacy contexts, with wide individual variability (see Table 2). Older children and those with fewer socio-demographic risks had higher literacy achievement scores at fall or spring (see Table 3). Higher fall letter-word ID scores were modestly negatively

Table 3 Correlations among independent and dependent variables.

1. Age in years 2. Male 3. Sociodemographic risk composite 4. TM, reading basic skills (TOTAL) 5. TM, reading basic skills (engaged) 6. TM, reading analysis (TOTAL) 7. TM, reading analysis (engaged) 8. CM, reading basic skills (TOTAL) 9. CM, reading basic skills (engaged) 10. CM, reading analysis (TOTAL) 11. CM, reading analysis (engaged) 12. Fall letter-word identification 13. Spring letter-word identification 14. Fall sound awareness 15. Spring sound awareness *

p < .05.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

–

−.12 –

.04 −.01 –

−.03 .04 .11 –

−.01 .02 .11 .99* –

−.03 −.07 .03 −.13 −.13 –

−.03 −.09 .02 −.12 −.12 .97* –

−.03 .08 .08 .00 .00 .22* −.23* –

−.02 .04 .07 .00 .01 −.18* −.20* .97* –

.07 .02 −.11 −.27* −.28* −.14 −.13 .08 .06 –

.08 .00 −.14 −.27* −.28* −.15 −.14 .11 .10 .95* –

.23* −.12 −.28* −.20* −.20* .04 .02 −.12 −.11 .14 .16* –

.25* −.16* −.23* −.12 −.09 .05 .07 −.04 −.01 .09 .14 .64* –

.24* −.08 −.22* −.14 −.12 .01 .05 −.02 .02 .09 .13 .52* .56* –

.22* −.12 −.23* −.19* −.15* .02 .07 .02 .04 .21* .21* .50* .71* .75* –

162

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

Table 4 Excerpts from coding manual of literacy instructional contexts and engagement descriptions. I. Type of instruction refers to the instructional activities of the teacher. Typically, these reflect the instructional demands of the classroom and indicate “dose.” Reading: Refers to any activity in which there is an effort to improve and acquire skills in writing, reading, and reading comprehension. Reading activities include: reading groups, reading aloud, book buddies, decoding activities, silent reading, etc. Language includes activities and instructions in communication skills, e.g., oral reports, or drawing pictures to illustrate stories or journal entries. Child managed instruction: Child managed instruction includes those instructional activities where the student is primarily controlling his/her focus of attention, for example, reading independently (e.g., sustained silent reading) and completing worksheets independently. These activities can be conducted in small groups or alone. This category includes children carrying out an activity that has been planned or directed by the teacher. Teacher managed instruction: Teacher managed instruction includes activities in which the teacher or the assistant teacher is the primary director of the children’s attention, and the teacher is encouraging or teaching an academic skill or knowledge. For example, reading a book or teacher and student discussions surrounding a particular book would be considered teacher managed instruction. Code-focused instruction/basic skills: This includes explicit instruction in basic skills. Examples include reciting the alphabet, letter-sound recognition, initial consonant striping, word segmentation, spelling in reading, handwriting practice, and most alphabet type activities. Meaning-focused/inference analysis: The focus here is on comprehension of problems. In reading, this refers to activities to improve and acquire skills in text analysis, understanding, and interpretation and written or oral expression. This includes teacher read aloud as well as read-aloud/discussion combined, discussion about a story, conversation about print conventions, vocabulary activities, scaffolded sustained silent reading, and acting out a story. II. Engagement refers to the interest and participation level of the study child (SC). Engaged: Active engagement/academic: The SC is actively involved in a teacher sanctioned instructional activity. This includes writing and appearing engaged in seat work activities, participating in a game, or participating in play that may have some academic purpose. This code offers a clear behavioral indication that the child is receiving an active dose of instruction. Passive engagement, probably involved: The SC is passively involved in a classroom task. This may include sitting and listening to a book, watching or observing the teacher, or showing focused attention during circle time/calendar. This would be coded most often when the teacher is actually instructing the children but can also be coded during child-managed instruction. Off-task: Not engaged: The SC is predominantly uninvolved in the teacher sanctioned activity. This includes wandering attention, daydreaming, unoccupied behavior, or engagement in an inappropriate activity. Wandering without aim around the room is clearly off task behavior. If the SC is passively involved in a classroom task, but it is not clear that he/she is truly involved and engaged in the task at hand, the behavior is coded as not engaged. Probably not involved: The SC is passively involved in a classroom task, but it is not clear that he/she is truly involved and engaged in the task at hand. This may include sitting and listening to a book but looking elsewhere around the room while it is being read. Other: The study child is actively engaged in the activity in the classroom, but this activity clearly has no academic focus or value. For example, when a child gets materials (as instructed) for an activity or makes a transition when the teacher is requesting that the students make a transition, “engaged/doing what is expected” would be coded. At times, it is not clear what is expected of that child, in which case “unable to determine task” would be coded.

correlated with minutes spent in TM basic skills, and marginally positively correlated with minutes spent in CM meaning-focused, which suggests weak relations between children’s initial wordreading skills and the literacy activities to which they were exposed. We also examined whether teacher variables related systematically to the amounts of time children spent in the four literacy contexts. To do this, aggregate variables representing the average minutes spent in each context for all the children in each classroom were calculated. An independent sample t-test indicated that classrooms taught by teachers with a masters’ degree spent significantly more time in TM meaning-focused (8 min) compared to classrooms taught by teachers with a bachelor’s degree (5.8 min), t = −2.22, p < .05. Differences by teacher experience were also found, where classrooms taught by teachers with more years of experience spent less time in CM meaning-focused, r = −.38, p < .05, and more time in TM meaning-focused at a level of marginal significance, r = .33, p = .08. 5.1. Child characteristic by literacy instructional context interactions The first research question asked whether minutes spent in the four literacy contexts predicted kindergartens’ spring letter-word ID and sound awareness, controlling for background characteristics and fall scores. This aimed to replicate, in a rural kindergarten sample, child-by-literacy instruction interactions found in other studies for word-reading, and to test the presence of child-by-instruction interactions for phonological awareness. To control for individual variation in how children spent their observed time, total minutes observed in the four contexts were included in the models. In addition, because children were nested within classrooms, we used Hierarchical Linear Modeling (HLM) software (Raudenbush, Bryk, Cheong, & Congdon, 2004). Examination of the Intraclass Correlation Coefficient (ICC) confirmed 11% of the variance at the

classroom level in spring word-reading, 2 = 55.46, p < .05, and 6% for spring phonological awareness, 2 = 48.49, p = .06. A child-level outcome was specified using two equations: Yij = ˇ0j + ˇ1j + ˇ2j · · ·ˇnj + rij

(1)

ˇ0j = 00 + uj

(2)

Eq. (1) defined the outcome Y for child i in classroom j as the intercept, or mean spring score for all children (ˇ0j ) plus the contribution of child-level variables ˇ1−n , plus error for the individual child (rij ). The intercept ˇ0j was further defined as the fitted grand mean score ( 00 ) plus error for classroom j (uj ). All variables of interest were child-level; models were built incrementally by adding relevant variables of interest in three stages. The first stage included child background variables (with fall age, gender, and the socio-demographic risk composite); we then calculated the percent of original (fully unconditional model) variance explained by including these variables. Next, we added fall pretest and calculated the additional variance explained. Finally a model was tested with the four variables that represented the number of minutes each child was observed in the four instructional contexts, and child-by-instruction interactions. Interaction variables were computed by multiplying children’s fall scores by the minutes spent in each of the four contexts. All interaction variables were based on centered variables so they could be tested in the same model, following the work of Connor and colleagues in this area (Connor, Morrison, & Katch, 2004; Connor, Piasta, et al., 2009). Final HLM models (one per outcome) are reported in Tables 5 and 6, retaining all four instructional context minutes variables and a single interaction variable. Background characteristics explained 12% of the variance in letter-word ID (11% in sound awareness); fall score explained an additional 30% of the variance in letter-word ID (46% in sound awareness). Consistent with prior work, overall time spent in each of the four contexts was not significantly predictive of children’s

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

163

Table 5 HLM results for spring letter-word Identification, with and without time off-task. Fixed effects

Coefficient (df)

Intercept (spring letter-word ID) Fall age Gender (1 = male; 0 = female) Risk composite Fall letter-word ID w-score TM reading, basic minutes TM reading, meaning-focused minutes CM reading, basic minutes CM reading, analysis minutes TM reading, basic minutes × fall letter-word ID TM reading, basic off-task minutes TM reading, meaning-focused off-task minutes CM reading basic, off-task minutes CM reading, meaning-focused off-task minutes

t-ratio *

409.43 (35) 7.56 (160) −2.80 (160) −9.42 (160) 0.60 (160) −0.02 (160) 0.09 (160) 0.16 (160) −0.07 (160) −0.04 (160) – – – –

182.64 1.48 −1.26 −2.58* 11.95* −0.10 0.40 0.60 −0.34 −4.19* – – – –

Coefficient (df)

t-ratio

408.83 (35) 6.23 (156) −1.80 (156) −7.98 (156) 0.62 (156) 0.24 (156) 0.30 (156) 0.41 (156) −0.02 (156) −0.04 (156) −2.88 (156) −1.75 (156) −1.39 (156) 0.04 (156)

181.40* 1.21 −0.78 −2.11* 12.57* 0.73 1.20 1.14 −0.06 −4.12* −1.65† −2.04* −1.20 0.03

Random effects

Variance component (df)

2

Variance component (df)

2

Level-2, U0 Level-1, r

56.95 (35) 222.41

78.88*

51.89 (35) 215.28

76.18*

* †

p < .05. p < .10.

spring scores on either subtest (see Table 5 for letter-word ID and Table 6 for sound awareness results). After entering the four interaction variables for letter-word ID (representing fall score multiplied by the minutes in each literacy context), one hypothesized interaction was significant and explained an additional 4% of the variance in spring letter-word ID. This interaction, shown in Fig. 1, indicated that children with low fall word-reading skills (1 SD below the mean) who spent relatively more time in TM basic skills activities had higher spring performance, compared to children with low fall skills who spent less time in TM basic skills. The opposite pattern was true for children with high fall scores (1 SD above the mean); their scores were highest when they spent less time in TM basic skills activities. The second hypothesized interaction for letter-word ID, between CM meaning-focused and children’s initial skills, was not significant, t = 0.63, p > .05. No other child-by-instruction interactions were significant in the models for either letter-word ID or sound awareness. 5.2. Differences in engaged time in TM or CM literacy contexts To unpack findings from the first research question, the second research question examined the amount of time spent in each

instructional context according to children’s engagement (engaged, off-task, or other). To assess whether engagement was more likely in TM than CM contexts, we conducted paired-samples t-tests for the percent of time children were engaged in TM and CM, for both basic skills and meaning-focused activities. However, ICCs for these data also showed evidence for nesting, so classroom-level variables were created that represented the classroom average percent of engaged time in comparison paired-samples tests. Results were consistent with the child-level findings, which are reported for parsimony. These analyses used the subsamples of children who were observed in both TM and CM: for the basic skills comparison, n = 75; for the meaning-focused comparison, n = 101. This was necessary because in this study of natural variation in literacy instruction provided across classrooms, not every child was observed in all four literacy contexts within the 50-min sampling period. Results were consistent with the hypothesis that children spent more time engaged in TM contexts than in CM contexts (see Fig. 2). In basic skills activities, children were engaged 89% of the time during TM, compared with 83% of the time engaged in CM, t(74) = 2.10, p < .05. In meaning-focused activities, children were engaged 88% of the time during TM versus only 75% of the time during CM, t(100) = 9.38, p < .01. While the absolute number of minutes that

Table 6 HLM results for spring sound awareness, with and without time off-task. Fixed effects

Coefficient (df)

t-ratio

Coefficient (df)

t-ratio

Intercept (spring sound awareness score) Fall age Gender (1 = male; 0 = female) Risk composite Fall sound awareness w-score TM reading, basic minutes TM reading, meaning-focused minutes CM reading, basic minutes CM reading, analysis minutes TM reading, basic minutes × fall sound awareness TM reading, basic off-task minutes TM reading, meaning-focused off-task minutes CM reading basic, off-task minutes CM reading, meaning-focused off-task minutes

474.32 (35) 1.37 (160) −1.64 (160) −.62 (160) .67 (160) −.16 (160) .05 (160) .13 (160) .25 (160) .01 (160) – – – –

524.04* .56 −1.35 −1.23 14.74* −1.26 .42 .89 2.50* 1.39 – – – –

474.14 (35) 1.03 (156) −1.30 (156) −.74 (156) .65 (156) .05 (156) .17 (156) .20 (156) .12 (156) .01 (156) −2.33 (156) −.92 (156) −.24 (156) .69 (156)

559.09* .42 −1.23 −1.47 14.94* .28 1.41 .95 .46 1.27 −2.41* −3.06* −.31 1.19

Random effects

Variance component (df)

2

Variance component (df)

2

Level-2, U0 Level-1, r

.91 60.13

36.87

2.20 54.35

42.11

*

p < .05.

164

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

420

Spring LW score

415

410 Fall LW score +1 SD 405

Average fall LW score Fall LW score -1 SD

400

395

390 0 minutes

5 minutes

10 minutes

Minutes in TM basic skills Fig. 1. LW = letter-word. Children with low fall letter-word scores (1 SD below the mean) had higher spring scores when exposed to relatively more TM, basic skills literacy instruction. The opposite is true for children with high letter-word scores (1 SD above the mean).

children spent in CM and TM contexts was fairly similar (shown in Table 2), Fig. 2 shows the percent of that total time that children spent off-task was greater in CM than TM contexts. 5.3. Does engaged time explain additional variance in spring reading achievement? The third research question used the same analytic approach as research question 1. However, instead of the variables that represented total minutes of teacher-intended time in the four instructional contexts, the number of minutes that children were behaviorally engaged in the four instructional contexts was included, plus the four child-by-minutes engaged interactions. Time engaged in each context was calculated by subtracting time spent off-task or in another type of engagement from the total time provided in each context. Results were virtually identical to those shown in Tables 5 and 6. Notably, the correlations among engaged time and total time in each context were quite high, ranging r = .95 to .99 (see Table 3). That is, our hypothesis that accounting for children’s engaged time would explain more variance relative to total time was not supported. The high correlations among engaged time and total time indicated that for all contexts, children were engaged the majority of time. Furthermore, most children (on average, 73%) were observed as engaged at some point (i.e., they had a value other than zero for their engaged minutes). In contrast, averaging across the contexts, only 43% of children were ever coded off-task. Further, correlations between total time in each context and minutes spent off-task in that context were moderate, ranging r = 0.44–0.65. In other words, whereas engaged behavior and total time were highly correlated (essentially, they represented the same variables), off-task behavior was distinct from total time because not all children were observed off-task. Based on this, in two final analyses (one per outcome), we entered the number of minutes spent off-task in each of the four contexts. With all the original variables from research question 2 retained, time off-task explained additional variance in both outcomes. Children who spent more time off-task during TM meaning-focused (t = −2.31, p < .05) and TM basic skills (t = −1.76, p = .08) had lower spring letter-word ID scores; this explained an additional 2% of the variance. The coefficient of −1.75 for TM

meaning-focused means that, holding all other variables constant at their mean or reference value, for every 1.75 min (of the 50 total minutes) that a child was observed off-task, their spring letterword ID score decreased 1 point. Similarly, more time off-task during TM meaning-focused, t = −3.06, p < .01, and TM basic skills, t = −2.41, p < .05, predicted lower spring sound awareness scores, explaining an additional 3% of the variance. 6. Discussion This study examined four contexts of literacy instruction, children’s behavioral engagement, and literacy achievement during kindergarten, with three results of import. First, kindergarteners who began school with low word-reading skills had higher spring word-reading scores if they were observed in relatively higher amounts of teacher-managed basic skill instruction; however, TM basic skills minutes did not interact with initial score to predict phonological awareness. Second, participants spent a greater percentage of their time engaged when they were observed in teacher-managed contexts compared to child-managed contexts. Third, the more minutes that children were observed off-task during teacher-managed instruction, the lower their scores on both literacy assessments. 6.1. Contexts of literacy instruction, initial word-reading skills, and achievement Children in our study spent more time in teacher-managed than child-managed contexts, an unsurprising finding in light of recent recommendations that teachers provide significant amounts of direct reading instruction during the kindergarten year (Kame’enui, Carnine, Simmons, Dixon, & Coyne, 2002). In a study of 223 kindergarten classrooms in three states, children spent over twice as much time in TM (defined as structured, teacher-managed) as in CM activities like seatwork (Pianta, La Paro, Payne, Cox, & Bradley, 2002). Results from the present rural sample in the Southeastern United States are consistent with this other work. Also, children spent more time in meaning-focused activities such as looking at books and listening to book-readings, compared to basic skills activities like rhyming, letter-sounds, and worksheets. Teachers with more years of experience were also slightly more likely to provide teacher-managed meaning-focused activ-

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

a

ENGAGED

OFF-TASK

165

OTHER

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Teacher-managed

Child-managed

Basic skills (n=75)

b 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Teacher-managed

Child-managed

Meaning-focused (n=101) Fig. 2. Comparing the average percent of time children spent engaged, off-task, or in another type of engagement for TM versus CM. Sample sizes are based on (a) n = 75 children who were observed in both TM and CM basic skills instruction and (b) n = 101 who were observed in both TM and CM meaning-focused instruction.

ities as opposed to having children look at books on their own in child-managed activity; teachers with less experience provided the opposite pattern (more CM, with slightly less TM). Though speculative, this may reflect differences in training between newer and more seasoned teachers. For example, teachers with more experience are more likely to have taught during the Whole Language era of literacy instruction, and may continue to use practices that emphasize meaning-making and exposing children to books through whole-group read-aloud activities (Rayner et al., 2001). Results corroborate existing reports from urban and suburban samples as well as meta-analytic work from the National Early Literacy Panel (2008). Namely, emergent readers improve more in word-reading when they experience relatively more teachermanaged instruction that targets basic skills. In other words, previously documented interactions between TM basic skills and initial word-reading, found for first grade readers with low initial skills, replicate in this sample of rural kindergarteners (Connor, Morrison, & Katch, 2004; Juel & Minden-Cupp, 2000). Research has suggested that direct, intensive, and explicit instruction is

required in the code-based skills that forecast later reading achievement (Adams, 1994; Juel & Minden-Cupp, 2000; National Reading Panel, 2000; Rayner et al., 2001). Documenting the same pattern of results in diverse populations is one way to establish generalizability. Our observational data do not support another prior result in the literature on child-by-instruction interactions, where strong entering readers improved over the first grade year in wordreading when their classrooms spent more time in child-managed meaning-focused activities such as SSR (Connor, Morrison, & Katch, 2004). There are probably not enough kindergarteners in our sample who began school already reading at a level where they would acquire additional proficiency from looking at books on their own. Indeed, 88% of children had fall WJ III letter-word identification raw scores of 16 or below, which includes upper- and lower-case letter identification (13 items) and identifying familiar sight words such as “cat” or “on” (3 items); only three children of the sample of 170 proceeded to the part of the test where they were required to read words comprised of more than three letters. In other words,

166

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

most of the children in the sample were not prepared for extensive child-managed instruction. Comparing these scores to national norms provides insight into generalizability. The sample mean fall w-scores of 351 on letter-word identification and 455 on sound awareness are highly comparable to nationally normed scores of 353 and 455, respectively (McGrew & Woodcock, 2001). However, the average fall age in this sample is 5.42 years, compared to the five-year-old norming sample, which suggests the present, mostly Caucasian, sample is slightly behind national norms. This corresponds with a sociological study indicating that Caucasians in rural areas, though disadvantaged, perform relatively better than other ethnic groups on literacy assessments (Durham & Smith, 2006). Future work with a larger number of rural children from ethnic minority backgrounds would be important for better understanding our results. Overall, the present findings are most similar to a pattern reported in preschool for word-reading, where the strongest and most consistent child-by-instruction interactions emerge for children starting behind their peers (Connor et al., 2006). We can directly compare our findings with Connor et al.’s, because that correlational study used WJ III letter-word identification as well. In both studies, more minutes spent in teacher-managed basic skills was associated with higher word-reading scores in the spring for children with w-scores below about 369 (raw scores below 16). In addition, in Connor et al.’s study, more TM meaning-focused time predicted word-reading improvement for preschoolers with high initial skills. Similarly, kindergarteners in the present study who spent less time off-task during TM activity scored higher on both literacy assessments, compared to children who were more often off-task during TM instruction. 6.2. Literacy instruction, behavioral engagement, and off-task behavior Child-managed activity such as looking at books and completing worksheets means children are responsible for their own attention, which tends to be quite challenging for kindergarten-age students (McClelland et al., 2000; Rimm-Kaufman et al., 2000, 2005). On average, we find that children are behaviorally engaged for a larger percentage of time in TM than CM contexts. TM literacy contexts in kindergarten are those where the teacher talks with a large or small group and reads a book to children, guides the class in a rhyming exercise, or provides explicit instruction about vowel sounds. Throughout this type of instruction, teachers talk regularly to their students, direct their focus to the meaningful aspects of the activity, and by doing so, they help regulate children’s attention (Thompson, 1994). In addition, teachers may rely on TM activity because young children are likely to be on-task and it provides fewer classroom management challenges than small-group or center time activities (Rimm-Kaufman et al., 2005). Finally, some TM literacy activities, such as rhyming games or teacher-led songs, might be more interesting than child-managed activities like completing worksheets. Important to note, despite this difference in engagement between TM and CM contexts, children in the present sample are predominantly on-task across contexts, which may not be the case in a sample with more variability in child risk factors related to school difficulty (Downer, Rimm-Kaufman, & Pianta, 2007). Counter to our expectations, knowing whether a child was behaviorally engaged or not does not predict literacy scores through an interaction with initial skill, above and beyond the type of instruction provided. This is due to the high degree of overlap between the variables that represent total time in each context and those that represent engaged time in each context. This suggests on-task behavior as the normative state during literacy instruction, whereas off-task behavior may occur only for a sub-

group. More varied levels of engagement may have occurred during classroom activity not included in this study, such as transitions (Arlin, 1979). Interestingly, children who spend a higher percent of time off-task during multiple contexts have significantly lower spring word-reading as well as phonological awareness scores. Thus, behavioral engagement and off-task behavior operate in distinct ways; moreover, off-task time is worth considering in studies of early literacy, over and above child characteristic-by-instruction interactions. Also of note, as with the interaction between fall wordreading and teacher-managed basic skills, the off-task variables that most strongly predict spring scores are during TM (not CM), further reiterating the importance of TM activity for kindergarten literacy learning. Nonetheless, the child-managed context may be relevant for other unmeasured skills, such as children’s interest in text, comprehension, or social connections with peers and teachers (Baker, 1996). One testable hypothesis is that spending time with books may develop print concepts–where children learn how to hold books and turn pages (McGinty & Justice, 2009). 6.3. Implications: an argument for considering instruction and behavioral engagement These findings in a kindergarten sample expand the literature on children’s emergent literacy outcomes by highlighting the attentional and behavioral demands of different learning contexts (Hadeed & Sylva, 1999; Kontos & Wilcox-Herzog, 1997). The present results indicate that children who are not engaged in the learning process have poorer end-of-year performance, controlling for background variables and overall time exposed to instruction. This work adheres to a conceptual framework that acknowledges the importance of children’s engagement in the classroom (Ladd, Birch, & Buhs, 1999). Further, an implication for teachers emphasizes that they continually assess children’s participation in literacy instruction as well as provide strong classroom organization in the early years of school (Cameron, Connor, & Morrison, 2005; Cameron, Connor, Morrison, & Jewkes, 2008; Pressley, Rankin, & Yokoi, 1996). Successful kindergarten teachers are able to consider that children vary not only in their aptitude and skills, but also in the extent to which they need support of their attention to learning tasks. To the extent that teachers are responsible for the instructional activities that children experience, the pattern of relatively more teacher-managed than child-managed time may indicate teachers’ awareness that young children benefit from interacting with adults (Curby, Rimm-Kaufman, & Ponitz, 2009; Mashburn et al., 2008) or may reflect more time spent in activity settings that offer easier management of children’s behaviors (Rimm-Kaufman et al., 2005). However, the provision of relatively more meaning-focused activities than basic skills activities suggests a possible lack of attention to the specific literacy experiences that research suggests prereaders require in order to learn how to read. This explanation may also underlie the absence of statistical interaction between initial phonological awareness and instructional exposure in relation to children’s spring phonological awareness: if teachers’ main focus is on helping children learn to read words, they may spend as much time on non-phonological activity, such as recognizing letters and sight words, as on deliberate decoding instruction. This hypothesis would be interesting to test in the context of an intervention where teachers focus on sound-symbol correspondence. Another possible explanation for phonological awareness results might have to do with the relatively narrow range of fall-spring improvement on the WJ sound awareness subtest (1.4 SD, as compared to an average gain of 2.5 SDs on the letter-word ID subtest). If children do not show much overall improvement from fall to spring, it will be more difficult to predict this improvement with specific instructional experiences.

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

6.4. Limitations Several limitations require mention. First, the sampling periods were limited to several 10-min episodes aggregated over the year, so some conclusions about how time is spent must be made tentatively. For example, not all participants were observed in all contexts, so information about off-task time in teacher-managed or child-managed activity is based on the subsamples of children who can be compared. This comparison appears to indicate that TM time is more common than CM time, and that children are more likely to be off-task during CM time; however, replication, or studies that actively manipulate time in TM or CM, would be required to draw strong conclusions. Second, the short observation blocks do not tell us about the temporal configuration of the school day. For instance, we cannot address whether teachers expect children to be engaged in an activity for 5 min, 10 min, or 60 min. Relatedly, because of the focus on the morning session where literacy instruction occurred, we also miss how instruction and engagement change over the course of a day or the year. For example, children with low ability levels might show higher levels of engagement in the morning than the afternoon, or students may show intense levels of initial engagement that decline over time (Lan et al., 2009). Third, though findings are consistent with randomized experiments (Connor, Piasta, et al., 2009), the present study was based on observational data gathered in a unique population. The present sample is rural but near a resourced university community and likely representative of other similar areas. Finally, this study did not examine the role of overall classroom quality, which likely contributes to children’s overall engagement and the nature of the instructional experiences offered, and needs to be examined in a future investigation (Mashburn et al., 2008). 6.5. Closing comments National goals demand that all children learn to read. Moreover, children’s behavioral engagement and acquisition of academic skills tend to be interdependent (Ladd et al., 1999). The typical kindergarten classroom, with its community of learners new to school, offers a lens from which to understand the relation between children’s participation in learning and their academic skills. Students who enter school with limited behavioral and academic skills may show difficulty when they are asked to attend to the teacher or persist on tasks above their level for long periods of time. Yet this study is consistent with others suggesting that literacy instruction managed by the teacher, which targets basic skills, seems especially important for emergent readers to make progress. Findings pave the way for future research that considers instructional literacy contexts with regard to their implications for reading achievement, as well as to the supports they offer in socializing young children and supporting their behavioral engagement in activities. Acknowledgements The authors would like to thank colleagues at the Center for Advanced Study of Teaching and Learning (CASTL) for their insight and assistance with the manuscript. Anita McGinty deserves special appreciation. References Adams, M. J. (1994). Beginning to read: Thinking and learning about print. Cambridge, MA: The MIT Press. Aram, D. (2005). Continuity in children’s literacy achievements: A longitudinal perspective from kindergarten to school. First Language, 25, 259–289. Arlin, M. (1979). Teacher transitions can disrupt time flow in classrooms. American Educational Research Journal, 16, 42–56.

167

Baker, L. (1996). Social influences on metacognitive development in reading. In C. Cornoldi, & J. Oakhill (Eds.), Reading comprehension difficulties: Processes and intervention (pp. 331–351). Mahwah, NJ: Erlbaum. Bialystok, E. (1995). Making concepts of print symbolic: Understanding how writing represents language. First Language, 15, 317–338. Bodovski, K., & Farkas, G. (2007). Mathematics growth in early elementary school: The roles of beginning knowledge, student engagement, and instruction. Elementary School Journal, 108, 115–130. Brody, G. H., & Flor, D. L. (1998). Maternal resources, parenting practices, and child competence in rural, single-parent African American families. Child Development, 69, 803–816. Cairns, R. B., & Rodkin, P. C. (1998). Phenomena regained: From configurations to pathways. In R. B. Cairns, L. R. Bergman, & J. Kagan (Eds.), Methods and models for studying the individual (pp. 245–265). Thousand Oaks, CA: Sage. Cameron, C. E., Connor, C. M., & Morrison, F. J. (2005). Effects of variation in teacher organization on classroom functioning. Journal of School Psychology, 43, 61–85. Cameron, C. E., Connor, C. M., Morrison, F. J., & Jewkes, A. M. (2008). Effects of classroom organization on letter-word reading in first grade. Journal of School Psychology, 46, 173–192. Catts, H. W., Fey, M. E., Zhang, X., & Tomblin, J. B. (1999). Language basis of reading and reading disabilities: Evidence from a longitudinal investigation. Scientific Studies of Reading, 3, 331–361. Connor, C. M., & Morrison, F. J. (2005). Examining causal implications of child × instruction interactions in early reading: U.S. Department of Education IES Grant # R205H04013. Connor, C. M., Morrison, F. J., & Katch, L. E. (2004). Beyond the reading wars: Exploring the effect of child–instruction interactions on growth in early reading. Scientific Studies of Reading, 8, 305–336. Connor, C. M., Morrison, F. J., & Petrella, J. N. (2004). Effective reading comprehension instruction: Examining child × instruction interactions. Journal of Educational Psychology, 96, 682–698. Connor, C. M., Son, S.-H., Hindman, A. H., & Morrison, F. J. (2005). Teacher qualifications, classroom practices, family characteristics, and preschool experience: Complex effects on first graders’ vocabulary and early reading outcomes. Journal of School Psychology, 43, 343–375. Connor, C. M., Morrison, F. J., & Slominski, L. (2006). Preschool instruction and children’s emergent literacy growth. Journal of Educational Psychology, 98, 665–689. Connor, C. M., Morrison, F. J., Fishman, B. J., Schatschneider, C., & Underwood, P. (2007). THE EARLY YEARS: Algorithm-guided individualized reading instruction. Science, 315, 464–465. Connor, C. M., Morrison, F. J., Fishman, B., Ponitz, C. C., Glasney, S., Underwood, P. S., et al. (2009). The ISI classroom observation system: Examining the literacy instruction provided to individual students. Educational Researcher, 38, 85–99. Connor, C. M., Piasta, S. B., Fishman, B., Glasney, S., Schatschneider, C., Crowe, E., et al. (2009). Individualizing student instruction precisely: Effects of child × instruction interactions on first graders’ literacy development. Child Development, 80, 77–100. Coyne, M. D, Kame’enui, E. J., & Simmons, D. C. (2001). Prevention and intervention in beginning reading: Two complex systems. Learning Disabilities Research & Practice, 16, 62–73. Curby, T. W., Rimm-Kaufman, S. E., & Ponitz, C. C. (2009). Teacher–child interactions and children’s achievement trajectories across kindergarten and first grade. Journal of Educational Psychology, 101, 912–925. De Marco, A. (2008). A qualitative look at child care selection among rural welfareto-work participants. Journal of Children & Poverty, 14, 119–138. Downer, J. T., Rimm-Kaufman, S. E., & Pianta, R. C. (2007). How do classroom conditions and children’s risk for school problems contribute to children’s behavioral engagement in learning? School Psychology Review, 36, 413–432. Durham, R. E., & Smith, P. J. (2006). Nonmetropolitan status and kindergarteners’ early literacy skills: Is there a rural disadvantage? Rural Sociology, 71, 625–661. Fantuzzo, J., Bulotsky-Shearer, R., McDermott, P. A., McWayne, C., Frye, D., & Perlman, S. (2007). Investigation of dimensions of social-emotional classroom behavior and school readiness for low-income urban preschool children. School Psychology Review, 36, 44–62. Foorman, B. R., Francis, D. J., Fletcher, J. M., Schatschneider, C., & Mehta, P. (1998). The role of instruction in learning to read: Preventing reading failure in at-risk children. Journal of Educational Psychology, 90, 37–55. Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004). School engagement: Potential of the concept, state of the evidence. Review of Educational Research, 74, 59–109. Gerde, H. K., & Powell, D. R. (2009). Teacher education, book-reading practices, and children’s language growth across one year of Head Start. Early Education & Development, 20, 211–237. Greenwood, C. R., Horton, B. T., & Utley, C. A. (2002). Academic engagement: Current perspectives on research and practice. School Psychology Review, 31, 328–349. Guthrie, J. T., Schafer, W. D., & Huang, C.-W. (2001). Benefits of opportunity to read and balanced instruction on the NAEP. Journal of Educational Research, 94, 145–162. Hadeed, J., & Sylva, K. (1999). Behavioral observations as predictors of children’s social and cognitive progress in day care. Early Child Development and Care, 154, 13–30. Hessler, K. L. (2009). Physical activity behaviors of rural preschoolers. Pediatric Nursing, 35, 253–346. Juel, C. (1988). Learning to read and write: A longitudinal study of 54 children from first through fourth grades. Journal of Educational Psychology, 80, 437–447. Juel, C., & Minden-Cupp, C. (2000). Learning to read words: Linguistic units and instructional strategies. Reading Research Quarterly, 35, 458–492.

168

C.C. Ponitz, S.E. Rimm-Kaufman / Early Childhood Research Quarterly 26 (2011) 157–168

Kame’enui, E. J., Carnine, D. W., Simmons, D. C., Dixon, R., & Coyne, M. D. (2002). Effective teaching strategies that accommodate diverse learners (2nd ed.). Upper Saddle River, NJ: Prentice Hall. Kontos, S., & Wilcox-Herzog, A. (1997). Influences on children’s competence in early childhood classrooms. Early Childhood Research Quarterly, 12, 247–262. Ladd, G. W., & Dinella, L. M. (2009). Continuity and change in early school engagement: Predictive of children’s achievement trajectories from first to eighth grade? Journal of Educational Psychology, 101, 190–206. Ladd, G. W., Birch, S. H., & Buhs, E. S. (1999). Children’s social and scholastic lives in kindergarten: Related spheres of influence? Child Development, 70, 1373–1400. Lan, X., Ponitz, C. C., Miller, K. F., Li, S., Cortina, K., & Perry, M. (2009). Keeping their attention: Classroom practices associated with behavioral engagement in first grade mathematics classes in China and the United States. Early Childhood Research Quarterly, 24, 198–211. Lee, J., Grigg, W., & Donahue, P. (2007). The Nation’s Report Card: Reading 2007 (NCES 2007-496). National Center for Education Statistics. Washington, DC: Institute of Education Sciences, U.S. Department of Education. Lillard, A. S. (2005). Montessori: The science behind the genius. New York, NY: Oxford University Press. Mashburn, A. J., Pianta, R. C., Hamre, B. K., Downer, J. T., Barbarin, O. A., Bryant, D., et al. (2008). Measures of classroom quality in prekindergarten and children’s development of academic, language, and social skills. Child Development, 79, 732–749. Matthews, J. S., Ponitz, C. C., & Morrison, F. J. (2009). Early gender differences in selfregulation and academic achievement. Journal of Educational Psychology, 101, 689–704. McClelland, M. M., Morrison, F. J., & Holmes, D. L. (2000). Children at risk for early academic problems: The role of learning-related social skills. Early Childhood Research Quarterly, 15, 307–329. McGinty, A. S., & Justice, L. M. (2009). Predictors of print knowledge in children with specific language impairment: Experiential and developmental factors. Journal of Speech, Language, and Hearing Research, 52, 81–97. McGrew, K. S., & Woodcock, R. W. (2001). Technical manual Woodcock–Johnson III. Itasca, IL: Riverside. McLoyd, V. C. (1998). Socioeconomic disadvantage and child development. American Psychologist, 53, 185–204. Morrison, F. J., Bachman, H. J., & Connor, C. M. (2005). Improving literacy in America: Guidelines from research. New Haven, CT: Yale University Press. National Early Literacy Panel. (2008). Developing early literacy: A scientific synthesis of early literacy development and implications for intervention. Jessup, MD: National Center for Early Literacy. National Reading Panel. (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. Washington, DC: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Institute of Child Health and Human Development (NIH Publication No. 00-4769). Olson, R., Wise, B., Forsberg, H., Sameroff, A. J., & Haith, M. M. (1996). The 5 to 7 shift in reading and phoneme awareness for children with dyslexia. The five to seven year shift: The age of reason and responsibility. Chicago, IL: University of Chicago Press., pp. 187–204. Pianta, R. C., La Paro, K. M., Payne, C., Cox, M. J., & Bradley, R. (2002). The relation of kindergarten classroom environment to teacher, family, and school characteristics and child outcomes. Elementary School Journal, 102, 225–238. Pianta, R. C., Cox, M. J., & Snow, K. L. (2007). School readiness and the transition to kindergarten in the era of accountability. Baltimore, MD: Brookes.

Piotrkowski, C. S., Botsko, M., & Matthews, E. (2000). Parents’ and teachers’ beliefs about children’s school readiness in a high-need community. Early Childhood Research Quarterly, 15, 537–558. Ponitz, C. C., Rimm-Kaufman, S. E., Brock, L. L., & Nathanson, L. (2009). Early adjustment, gender differences, and classroom organizational climate in first grade. Elementary School Journal, 110, 142–162. Pressley, M., Rankin, J., & Yokoi, L. (1996). A survey of instructional practices of primary teachers nominated as effective in promoting literacy. Elementary School Journal, 96, 363–384. Pressley, M., Wharton-McDonald, R., Allington, R., Block, C. C., Morrow, L., Tracey, D., et al. (2001). A study of effective first grade literacy instruction. Scientific Studies of Reading, 5, 35–58. Raudenbush, S. W., Bryk, A. S., Cheong, Y. F., & Congdon, R. (2004). HLM 6: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International, Inc. Rayner, K., Foorman, B. R., Perfetti, C. A., Pesetsky, D., & Seidenberg, M. S. (2001). How psychological science informs the teaching of reading. Psychological Science in the Public Interest, 2, 31–74. Ready, D. D., LoGerfo, L. F., Burkam, D. T., & Lee, V. E. (2005). Explaining girls’ advantage in kindergarten literacy learning: Do classroom behaviors make a difference? Elementary School Journal, 106, 21–38. Rimm-Kaufman, A. P., & Rimm-Kaufman, S. E. (2006). RK Coder. Unpublished software. Rimm-Kaufman, S. E., Pianta, R. C., & Cox, M. J. (2000). Teachers’ judgments of problems in the transition to kindergarten. Early Childhood Research Quarterly, 15, 147–166. Rimm-Kaufman, S. E., La Paro, K. M., Downer, J. T., & Pianta, R. C. (2005). The contribution of classroom setting and quality of instruction to children’s behavior in the kindergarten classroom. Elementary School Journal, 105, 377–394. Rimm-Kaufman, S. E., Klein, S. E., & Romberg, M. E. (2007). Social Development Lab-Kindergarten coding system (SDL-K). Unpublished work. Coding manual. University of Virginia. Sameroff, A. J., Bartko, W. T., Baldwin, A., Baldwin, C., & Seifer, R. (1998). Family and social influences on the development of child competence. In M. Lewis, & C. Feiring (Eds.), Families, risk, and competence (pp. 161–185). Mahwah, NJ: Erlbaum. Sénèchal, M. (2006). Testing the home literacy model: Parent involvement in kindergarten is differentially related to grade 4 reading comprehension, fluency, spelling, and reading for pleasure. Scientific Studies of Reading, 10, 59–87. Spira, E. G., Bracken, S. S., & Fischel, J. E. (2005). Predicting improvement after firstgrade reading difficulties: The effects of oral language, emergent literacy, and behavior skills. Developmental Psychology, 41, 225–234. Stipek, D. J. (2006). No Child Left Behind comes to preschool. Elementary School Journal, 106, 455–465. Stipek, D. J., Feiler, R., Byler, P., Ryan, R., Milburn, S., & Salmon, J. M. (1998). Good beginnings: What difference does the program make in preparing young children for school? Journal of Applied Developmental Psychology, 19, 41–66. Thompson, R. A. (1994). Emotion regulation: A theme in search of definition. Monographs of the Society for Research in Child Development, 59, 25–52. Tickamyer, A. R., & Duncan, C. M. (1990). Poverty and opportunity structure in rural America. Annual Review of Sociology, 16, 67–86. Wasik, B. A., Bond, M. A., & Hindman, A. (2006). The effects of a language and literacy intervention on Head Start children and teachers. Journal of Educational Psychology, 98, 63–74. Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock–Johnson III Tests of Achievement. Itasca, IL: Riverside. Xue, Y., & Meisels, S. J. (2004). Early literacy instruction and learning in kindergarten: Evidence from the Early Childhood Longitudinal Study— Kindergarten class of 1998–1999. American Educational Research Journal, 41, 191–229.