Joint Book Reading in the Second Year and Vocabulary Outcomes

Transcription

1 Journal of Research in Childhood Education 2007, Vol. 21, No. 3 Copyright 2007 by the Association for Childhood Education International /07 Joint Book Reading in the Second Year and Vocabulary Outcomes W. Allen Richman Macon State College John Colombo University of Kansas Abstract. The positive effects of joint book reading (JBR) between parents and children have been widely presumed. However, the impact of JBR has not been widely investigated in toddlers, nor has the extant literature considered the context of JBR reading sessions (e.g., reader, prompt, occasion, who chose the book, etc.). The present study examined the occurrence of JBR in middle-class homes, using two methodologies (i.e., questionnaire and daily journal), and then examined variables derived from these methods with children s vocabulary, as assessed by the MacArthur-Bates Communicative Development Inventory. Indicants of the frequency of JBR and the variety of titles read were strongly correlated with vocabulary outcomes, especially expressive vocabulary. In addition, JBR was determined to occur in two independent contextual situations: one that occurred as a matter of regular routine, and another that occurred more irregularly and spontaneously; the former was marginally associated with receptive vocabulary, while both factors contributed significantly and independently to the prediction of expressive vocabulary. These findings are in accord with the hypothesis that JBR promotes language skills, and extends them to toddlers. Much attention has been recently devoted to parent-child interactions that involve the reading of printed material; such interactions have been labeled joint book reading (JBR). It has been widely assumed that JBR promotes language, literacy, and overall cognitive and intellectual development. Reports in both parenting magazines (Bush, 1990; Parent Teacher Association [PTA], 1987) and scholarly publications (e.g., Burns, Griffin, & Snow, 1999; Elley, 1989; Eller, Pappas, Note: This research was supported by a Graduate School Summer Fellowship awarded to the first author and by grant HD35903 awarded to John Colombo. Address correspondence and reprint requests to W. Allen Richman, Division of Education, Macon State College, Macon, GA ; arichman@mail.maconstate.edu & Brown, 1988; Neumann, Copple, & Bredekamp, 2000; Pemberton & Watkins, 1987; Senechal & Cornell, 1993; Snow & Goldfield, 1983; Strickland, 1993) have touted the importance of JBR as a vital component of a child s development. Indeed, several investigations have reported 1) gains in language or vocabulary as a function of exposure to specific words during reading (Elley, 1989; Senechal & Cornell, 1993), 2) enhancement of parent-child interactions during reading sessions (Arnold, Lonigan, Whitehurst, & Epstein, 1994; Dale, Crain-Thoreson, Notari-Syverson, & Cole, 1996; Valdez- Menchaca & Whitehurst, 1992; Whitehurst et al., 1988), and 3) significant correlations between early exposure to books and later reading achievement (Scarborough, Dobrich, 242

2 Joint Book Reading & Hager, 1991). Furthermore, the Commission on Reading of the National Academy of Education (Anderson, Hiebert, Scott, & Wilkinson, 1985) identified JBR as the most important activity in which parents can engage their children that will influence later academic success. Recent findings suggest that parents are heeding these messages and read regularly to their children who are between 1 and 6 years of age (High et al., 1999; Yarosz & Barnett, 2001). At the same time, there is still some debate about the relative impact that JBR has on a child s development. Two meta-analyses (Bus, van Ijzendoorn, & Pellegrini, 1995; Scarborough & Dobrich, 1994) concluded that the frequency of JBR accounts for about 8 percent of the variance in outcomes such as vocabulary, reading, and school achievement; the figure of 8 percent is interpreted as highly positive by some, and discouraging by others. These meta-analyses, however, provide insight into at least two gaps in our current understanding of JBR. The first of these gaps is consideration that other factors beyond merely the frequency of JBR could impact the observable outcomes from the JBR session. The studies analyzed by Bus et al. (1995) and Scarborough and Dobrich (1994) were primarily focused on the frequency of JBR sessions, and as such were dissociated from any contextual factors that might impact the reading environment or setting. Indeed, the literature provides evidence to suggest a number of contextual factors that may influence the observable gains from a JBR session. In a series of studies, Whitehurst and colleagues have shown that the reader s style can significantly affect the observed gains (Arnold et al., 1994; Valdez-Menchaca & Whitehurst, 1992; Whitehurst et al., 1988). The type of book has also been shown to alter the mother-child interactions that occur during JBR (Stadler & McEvoy, 2003). These findings suggest that the frequency of JBR alone may not capture all of the variance that may impact child outcomes. Also apparent from these meta-analyses is the relative dearth of research with very young children. There is some evidence in the literature to suggest that earlier onsets of JBR are related to vocabulary (DeBaryshe, 1993) and emergent literacy (Allison & Watson, 1994). Theoretically, JBR should provide infants or toddlers with a wealth of cognitive, linguistic, and educational advantages (Moerk, 1985) that would greatly help them in perceiving, parsing, and eventually recognizing various aspects of the language stream (Aslin, Saffran, & Newport, 1998; Jusczyk, 1997; Jusczyk & Hohne, 1997) at a critical juncture in the developmental course of language acquisition. However, in over 30 years of research, only two very early studies in the area (Irwin, 1960; Wells, 1985) report findings with children below 18 months of age. Irwin (1960) provided books to an experimental group of 24 parents of 13-month-old children and then instructed these parents to read to their children for 15 to 20 minutes a day. From 17 to 30 months of age, the experimental group showed increases in phonemic gains relative to a control group, with the experimental group showing a significantly more diverse phoneme repertoire than the controls. The Bristol Language Development Project (Wells, 1985) followed 128 children from 15 to 60 months of age and reported a number of significant relationships between the home environment and language skills, with the frequency of JBR sessions being one of the strongest predictors of language skills. Thus, the literature provides some evidence to suggest that reading to very young children is beneficial. At the same time, the evident gaps in the literature are the absence of considering the impact of contextual factors on the efficacy of JBR and the limited amount of research performed with young children (i.e., below 18 months of age). One final limitation of the extant literature is that it is largely based on retrospective self-reports of JBR from caregivers; lacking in the literature are estimates of JBR with younger samples of children that are collected in a prospective or concurrent manner. 243

3 Richman and Colombo The present study sought to close some of these empirical and methodological gaps by assessing the occurrence of JBR with a sample of children younger than 18 months old by using two methodologies (a questionnaire and a daily journal) in a sequential format. Children s vocabulary also was collected as an outcome variable. The goal of the study was to obtain a better picture of JBR with younger children, and to determine the relationship between an expanded consideration of the early parameters (frequency, variety, and context) of JBR with early vocabulary development. Method Participants Participants were recruited using a database of all viable births in the suburban areas of the Kansas City metropolitan area. The authors sent a questionnaire to all parents within the county in the database who had children 10 to 17 months of age. A total of 1,058 questionnaires were distributed. From these, 168 responded appropriately to the questionnaire, and a subset of 48 participants continued with the JBR journal portion of the study. All but three of these participated in the vocabulary assessment. All caregivers recruited for the sample reported that they were already reading to their children at the point of contact. Table 1 shows the demographic data for both the questionnaire and JBR journal, as well as comparable data from the county census. Caregivers agreeing to continue with subsequent phases were representative of the original 168 participants who agreed to participate. With the exception of level of education, the information included on the questionnaire and journal sample was not appreciably different from that on the county census data. Measures The Questionnaire. The questionnaire was devised to provide data on family demographics, caregiver reading habits, and frequency and onset of joint book reading practices. Family demographics included Table 1 Family Demographics Data for Questionnaire, Journal, and County Census Data Sample: Questionnaire JBR Journal County Age (years) Ethnicity (percent) Education (percent) Primary Caregiver 32.4 (4.7) 32.7 (6.3) Secondary Caregiver (5.2) 33.4 (5.3) African American Asian Caucasian Hispanic Multi-racial HS degree College degree Graduate degree Median Household $50,000 $50,000 $59,870 Income Note. The data in column 4 are from State and County QuickFacts, by the U.S. Census Bureau, n.d. Retrieved December 1, 2001, from states/20/20091.html

4 Joint Book Reading identification of the primary caregiver of the child, household income, and education. Finally, a number of questions focused on the JBR environment. The primary questions for consideration were about the onset, frequency, average length, and normal time at which JBR practices occurred. In general, these questions were designed to elicit specific answers on a continuous scale (e.g., On average, how many times a week is some type of written material read to the child?). The JBR Journal. A journal was developed to record the JBR practices of our sample on a daily basis for a 30-day period. Participants were asked to record every title that was read to the child, along with the start and end time for each title read. Beyond the title and author of the book, caregivers were also asked to record the ISBN when possible. The reason for using ISBN was to obtain a measure of the caregivers data collection accuracy. This information allowed for the tabulation of the frequency and duration of JBR sessions, and allowed us to calculate the number of unique titles read that month. Beyond duration and frequency, however, the journal captured descriptive data on the contextual circumstances surrounding each JBR session. The contextual variables of interest included the following: 1) The identity of the reader in the JBR session 2) The occasion of the JBR session (e.g., bedtime, naptime, mealtime, playtime, or other) 3) The prompt for the JBR session, which referred to whether the JBR was a regularly scheduled (i.e., routine) occasion, a JBR session requested by the child, or a spontaneous JBR session 4) The activity that characterized the nature of what was done with the book during the JBR session (summarize/paraphrase book, read all of the book, read part of the book, looked at pictures) 5) The selector of the book used during the JBR session (child or reader). MBCDI. The MacArthur-Bates Communicative Development Inventory (MBCDI) is a well-recognized and widely used instrument for the assessment of emergent vocabulary; its validity and reliability have been established in a number of studies (Fenson et al., 1993). Given the age of the population being assessed in this study, the MBCDI presents itself as an efficient means of assessing the vocabulary of this sample of children. Following the completion of the journal, the MBCDI Words and Gestures form was sent to participants. The Words and Gestures (normed for 8 to 16 months of age) was sent, because it provides an estimate of both receptive vocabulary (the words a child can understand, even if he or she does not use them) and expressive vocabulary (the words a child can understand and uses). Procedures Timeline Questionnaires were sent to all eligible caregivers when children between 10 and 17 months of age. The final two questions of the questionnaire requested continued participation for the JBR journal phase. Once the questionnaires were returned, participants who indicated their willingness to continue participation with the JBR journal phase were sent the JBR journal, in which caregivers logged JBR activities for the next 30 days. After completion of the JBR journal, caregivers returned the journal materials and then were mailed the MBCDI form to complete. The MBCDI was completed an average of days (SD 15.81; range days) after the questionnaire; and 45.3 days (SD 8.5; range days) after the completion of the JBR journal. Processing the Journal Data The journal data was entered into a custom database program designed specifically for recording this data set. Along with the direct coding of data from the journal, the database allowed for the calculation of the two frequency measures, unique titles and book readings. Figure 1 diagrams how these two measures would be derived from three 245

5 Richman and Colombo hypothetical days in which JBR occurred. In addition, the duration for each book read was calculated by simple subtraction of start times from stop times. On occasion, when readers inserted multiple titles on a single line, the total duration was divided by the number of books read. All titles and ISBNs (or author s name) were recorded in the database for future reference and for an indirect verification of the reading session. Although the presence of an ISBN itself did not serve as definitive verification that the reading session occurred, the presence of a valid ISBN increased our confidence in the veracity of the provided data. Verification of the book ISBN was obtained for 93% of the titles recorded. Results The JBR Questionnaire Frequency, Onset, and Duration of JBR. Questionnaire data showed that our sample (n = 168) reported an early start to JBR and a relatively high current frequency of JBR. The mean reported age for first reading any written material to the child was 2.8 months of age (SD = 2.74), with 80 percent beginning by 4 months of age and 90 percent beginning by 6 months of age. Caregivers reported that their toddlers were engaged in an average of 13.0 JBR sessions per week (SD = 14.51; range 1-120). The duration of these sessions ranged from 1 to 30 minutes, with an average of 9.8 minutes (SD = 6.35) per session. Given the reported values of frequency and duration of JBR, the total time spent in JBR per week could be calculated for this sample; this came to an average of minutes per week (SD = ). The observed onset and frequencies of JBR obtained from the questionnaire were earlier and higher than what has been reported in the literature (e.g., Wells, 1985). As noted previously, this sample is likely skewed favorably toward JBR; the early onset and high frequency of JBR is likely attributable to this factor. At the same time, the fact that parents are making such reports far more concurrently than in any previous studies on this topic may also contribute to these estimates. Contextual Variables. Analyses of the contextual variation of JBR within this group also were conducted. Primary caregivers (in 95 percent of the cases, these were mothers) were the readers for the majority of the JBR sessions. They reported a mean weekly frequency of 9.7 (SD = 11.9) reading sessions Figure 1 246

6 Joint Book Reading per week; secondary caregivers (typically, fathers) reported JBR approximately 4.1 (SD = 3.8) times per week on average. Siblings and other readers accounted for an average of 2.3 (SD = 2.1) JBR sessions per week. Approximately half of the JBR sessions were reported to occur during play (M = 7.0 per week, SD = 10.6). Other JBR sessions occurred apparently as a component of a bedtime routine, occurring adjacent to naptime (M = 3.3 sessions per week; SD = 3.9), or prior to bedtime (M = 4.8 sessions per week; SD = 2.7). JBR occurred less frequently around mealtime (M = 0.6, SD = 1.7) or at other times (M = 2.2, SD = 4.1). Although the questionnaire data are from a sample that is skewed toward the conduct of JBR, the estimates obtained here for the frequency of JBR is remarkably close to that reported by Irwin (1960) in an earlier study. Caregivers in the current study reported approximately 15 JBR sessions per week; Irwin (1960) reported a value of 13 weekly sessions. The JBR Journal The journals provided data on 4,393 readings of 1,182 different titles, which were logged across the entire sample (N = 48) and across the entire period of recording (30 days). Frequency and Duration of JBR. The mean number of unique titles read in one month was 24.6 (SD = 14.5; range 5 to 67). It is widely reported that in JBR, children like to have the same book read to them multiple times; the data collected here strongly supported that claim, as caregivers reported an average of approximately 91.5 (SD = 52.4) book readings over the 30 days during which the journal was kept. Thus, on average, these children were being exposed to 3 book readings a day in JBR, only one of which was a unique title. Response Reliability Across Instruments. Given that journal respondents had provided questionnaire data during the prior phase of the study, we were able to assess how reliably caregivers were reporting on some aspects of JBR. The mean reported duration of JBR was available from both the questionnaire and journal instruments. The correlation for this variable was moderately strong and statistically significant (r =.65, p<.001). In addition, the frequency of JBR occasions was reported on the questionnaire, and the number of JBR book readings was derived from the journal. The correlation between these two reports was moderate but also significant (r=.40, p<.01), and the mean frequency of daily reading sessions from the questionnaire (M= 2.1; SD = 1.3) was in relatively good agreement for the mean frequency of book readings from the journal (M = 3.1; SD = 1.7). This level of agreement across these two measures can be regarded as encouraging, in light of the fact that caregivers did not have access to their JBR questionnaire reports when they were filling out the journal, and that the frequency from the journal was derived from an accumulation of reports over the 30 days. Contextual Variables From the JBR Journal. Among the primary goals of using a diary methodology was to attempt to obtain more information about the reading session, including the context in which the session occurred. With respect to the JBR reader, the journal showed that the mother was the individual most responsible for generating the literacy environment, logging 66 percent of the titles read, for a total of 285 hours of reading. Fathers read 20 percent of the titles, followed by the other category (10 percent). The identity of the individuals in this other category was captured on the log; examination of the data showed that 75 percent of these readings involved relatives (e.g., grandmothers, aunts, etc.; 300 readings), and 25 percent involved unrelated caregivers (e.g., baby-sitters, neighbors, etc.; 150 readings). Children chose the book in JBR sessions 62.3 percent of the time. Interestingly, the duration of JBR did not vary as a function of who chose the book, F (1, 47) =.11, ns. The most frequent occasion at which JBR occurred was play (58 percent of the sessions), and 38 percent of the reading sessions occurred at bedtime or naptime. This 247

7 Richman and Colombo frequency pattern replicated that which was obtained from the questionnaire (63 percent play; 37 percent nap/bedtime). Examination of the frequency of different prompts showed that a vast majority of JBR sessions fell into two categories: 42 percent of all sessions were initiated by the child, and 39 percent occurred as part of a scheduled (i.e., as part of a routine) JBR reading session. Finally, we analyzed what the dyad did during the JBR session (i.e., summarized book, read all of book, read part of book, looked at pictures, other). Unlike the other contextual variables, readers did not always consider the categories available for activity to be mutually exclusive; if multiple activities were checked (this occurred in less than 5 percent of the sessions logged), we counted them all. About 65 percent of the recorded sessions (3,000) involved reading the entire book. Read part (648, or 14.2 percent), looked at pictures (666, or 14.6 percent), or summarized (187, or 4.1 percent) all accounted for a much smaller proportion of the JBR activities reported. Extraction of Context-Based JBR Factors. Overall, data from the journal provided descriptive data on the contextual variables that characterize the nature of JBR in these homes. However, we sought to understand how these contextual JBR variables might covary to present more coherent clusters or factors of characteristics. Thus, a principal-components (PC) factor analysis was performed on the contextual variables. The sample size and construction of the items was appropriate to this analysis, and the PC approach is considered to be a simple and straightforward means of reducing data for the purposes envisioned (Stevens, 1996). The PC analysis yields a first solution that is unrotated; this solution provides commonalties, identification of specific factors, and factor loadings for each of the items involved. In this first solution, two factors were identified as having Eigenvalues over 1.00, which reflects the traditional Kaiser criterion for identifying significant factors. Table 2 Principal Components Analysis of JBR Contextual Variables After Varimax Rotation Contextual Factor Activity Occasion Prompt Contextual Variable: Book Selector 1 Routine- Based 2 Play-Based Eigenvalue: read all book read part of book looked at pictures bedtime naptime play child requested scheduled child parent

8 Joint Book Reading These two factors had eigenvalues of 4.1 and 2.9, and accounted for, respectively, 40.6 percent and 29.3 percent of the variance in the system. Once these initial factors were identified, the solution was rotated using the varimax option. According to Stevens (1996), this rotation is preferred for cleaning up factors from a PC analysis, and maximizes the interpretability of the results. The rotated solution yielded a number of items that loaded significantly on each of the two factors (see Table 2). Factor 1, hereafter referred to as Routine- Based JBR, was characterized by a scheduled reading occasion, most often occurring as part of a bedtime or naptime routine. The reader typically selects the book, and the book is read in its entirety. Factor 2, hereafter referred to as Play-Based JBR, was characterized by JBR occurring at playtime. The child typically initiated the session and selected the book, and this JBR factor was not associated strongly with any type of activity. JBR data and the MBCDI The observed range for receptive vocabulary (Words Understood from the MBCDI) was and for expressive vocabulary (Words Produced from the MBCDI) was These ranges were all below ceiling, as the total number of words on the MBCDI list is 396. MBCDI and Age. As might be expected, the number of words understood and produced from the MBCDI covaried significantly with age. The age of the child (calculated in days) was significantly correlated with both receptive vocabulary (r =.55, p<.01) and expressive vocabulary (r =.53, p<.01) measures. Thus, subsequent analyses were conducted statistically controlling for age. MBCDI and Questionnaire Data. Significant zero-order correlations between frequency and both receptive (r=.35, p<.05) and expressive vocabulary (r=.41, p<.01) attained statistical significance. When age was statistically controlled for in these analyses, the contribution of receptive vocabulary from JBR frequency approached significance (p=.10) and attained significance (p<.05) for expressive vocabulary (see Table 3). MBCDI and Journal Data. Significant zero-order correlations revealed significant relationships between the total number of books read for the month with receptive vocabulary (r=.39, p<.01) and expressive vocabulary (r=.52, p<.01) from the MBCDI. Significant correlations were also observed for the total number of unique title read for the month with both receptive (r=.43, p<.01) and expressive (r=.60, p<.01) vocabulary measures. Given that these two predictors are intercorrelated, the contributions of each of these were delineated in a hierarchical multiple regression. The order of entry for predictors was age (as both vocabulary measures covaried with age), total books read, and total unique titles for both outcome variables. Table 4 shows that when age is controlled for, the total number of books read per month accounts for unique variance in receptive vocabulary, but that Table 3 Regression Models Predicting Vocabulary From Frequency (Questionnaire) Dependent Step Predictors df R R 2 ΔR 2 F p value Receptive Vocabulary 1 Age 1, <.01 2 Frequency 1, Expressive Vocabulary 1 Age 1, <.01 2 Frequency 1,

9 Richman and Colombo unique titles made no further contribution to the predictive equation. This same set of analyses, which was repeated for the prediction of expressive vocabulary, shows that (again, statistically controlling for age) both the total number of books read in the month and the number of unique titles also accounted for a significant portion of the variance. The preceding analyses provide evidence in support of a strong association between reported JBR frequency measures and vocabulary in early childhood. However, a theme of this project has been the assessment of contextual information about the reading environment and about the reading session itself. A final set of analyses was conducted to address whether qualitative aspects of JBR were related to children s receptive and expressive vocabulary; this was done by employing the factors (i.e., Routine- Based JBR and Play-Based JBR) derived from the principal components analysis of contextual variables and determining their relationship with receptive and expressive vocabulary. After controlling for age, we inserted both Play-Based JBR and Routine- Based JBR factor scores into the model in alternating orders in order to determine the unique contributions of each of the two measures of vocabulary. Table 5 shows the results of these analyses. For receptive vocabulary, neither contextual factor was a strong predictor; Routine-Based JBR (irrespective of its position of entry into the equation) accounted for an additional 5% of the variance, which approached statistical significance. The contribution of Play- Based JBR to receptive vocabulary was not significant. For expressive vocabulary, however, both contextual factors accounted for significant and unique variance. Discussion The extant literature on joint book reading, both scholarly and popular, has strongly implicated the importance of JBR as a fundamental component of emergent literacy in the young child s environment. Limitations The current study is limited in two ways that bear repeating before a discussion of the findings presented here. First, the study is correlational; as such, the directional nature of the effects necessarily remain uncertain, although experimental studies show that supplementing JBR positively affects language skills (Elley, 1989; Senechal & Cornell, 1993). Second, while this study did obtain a representative sample of the population for the county, the sample was largely middle-class white families (i.e., 95 percent Caucasian, 56 percent of parents had a college degree, and all participants were reading to their children at induction into the study) (U.S. Census Bureau, n.d.). Clearly, replication of this line of study needs to be performed with a more nationally representative sample. This being said, however, one might expect that this situation would constrain variability and outcome; Table 4 Regression Models Predicting Vocabulary From Frequency (JBR journal) Dependent Step Predictors df R R 2 ΔR 2 F p value 1 Age 1, <.01 Receptive Vocabulary Expressive Vocabulary 2 Book Readings 1, Unique Titles 1, Age 1, <.01 2 Book Readings 1, <.01 3 Unique Titles 1,

10 Joint Book Reading instead, considerable variability existed in JBR practice within this sample on both the questionnaire and journal, and this variability was predictive of language outcomes. As such, it seems reasonable to believe that the current report contributes significantly to the JBR literature by expanding the knowledge base of the JBR environment and JBR practice in toddlers during the second year of life. The current study protocol included a JBR questionnaire, followed by a journal in which daily JBR activities were logged and then by a vocabulary assessment using the MBCDI. Four findings serve to highlight the new information provided in this report. Reliability of Caregiver Report of JBR First, caregivers reliably reported on the frequency and duration of JBR sessions across the questionnaire and daily journal forms. This is an especially surprising finding, because the questionnaire variable was simply and directly reported and a similar variable from the journal was derived from 30 days of reporting, from which caregivers were unlikely to recall what was reported on the questionnaire, and highly unlikely to have fabricated their observations. Frequency and Variety of JBR and Vocabulary Outcomes Second, a number of variables that reflect the frequency and variety of JBR in the home significantly predicted to vocabulary measures from the MBCDI. The strongest and most consistent effects were seen for expressive vocabulary. From the questionnaire, caregivers report of frequency of JBR was significantly associated with expressive vocabulary, even after controlling for age. From the 30-day journal, the number of books read was consistently associated with vocabulary outcomes; in addition, the number of unique titles read accounted for significant unique variance in expressive vocabulary over and above that accounted for by the number of books read. This latter finding reflects an additive effect of both frequency and variety, as both variables were associated with different variance in the acquisition of labels; together, these two variables together accounted for 27 percent of the variance in expressive vocabulary. Table 5 Regression Models for Journal Contextual Variables Unique Contributions Dependent Variable Receptive Vocabulary Expressive Vocabulary Model Step Predictors df R R 2 ΔR 2 F p 1 Age 1, <.01 2 Routine 1, Play 1, Age 1, <.01 2 Play 1, Routine 1, Age 1, <.01 2 Routine 1, <.01 3 Play 1, Age 1, <.01 2 Play 1, <.04 3 Routine 1, <

11 Richman and Colombo Contextual Factors and Vocabulary Outcome The final two findings involved the quantitative documentation of different contextual scenarios in which JBR occurred. The third finding was that contextual variables from the 30-day journal allowed us to identify two separate factors describing JBR practice in these homes. One factor described JBR that occurred as a regular part of bed and nap routines, was driven primarily by the adult reader, and involved exposure to the full contents of a book. Another factor described JBR that occurred during play, primarily at the child s request and with a book chosen by the child, and during which exposure to the book s contents was variable. The fourth finding was that these two factors accounted for vocabulary in different ways. Only routine-based JBR tended to be associated with receptive vocabulary, but both JBR factors independently accounted for unique and significant amounts of variance in expressive vocabulary. Again, this an additive effect of exposure, suggesting that routinely scheduled JBR is important to vocabulary, but that even casual and spontaneous occurrences of JBR contribute to the language skills. Future research may be directed toward the question of whether different aspects of language may be fostered differentially by JBR occurring during these two scenarios. A Final Note A final consideration involves the use of the 30-day journal, in which caregivers logged each JBR session. We observed this to be a highly satisfactory means for collecting data on JBR in the home on a consistent basis. All caregivers who volunteered for this part of the study filled out the log dutifully; the requirement for logging the ISBN of the book read to the child allowed us to verify the title for each reading in over 90 percent of the cases. Although it would be desirable to conduct detailed and objective observations of JBR in the home, this research methodology would be impractical and costly, even with a moderate sample size. Our study suggests that future development and refinement of the journal method used here may allow for reliable, fine-grained study of JBR in natural and ecologically valid settings. References Allison, D. T., & Watson, J. A. (1994). The significance of adult storybook reading styles on the development of young children s emergent reading. Reading Research and Instruction, 34, Anderson, R. C., Hiebert, E. H., Scott, J. A., & Wilkinson, I. A. (1985). Becoming a nation of readers. Washington, DC: National Institute of Education. Arnold, D. H., Lonigan, C. J., Whitehurst, G. J., & Epstein, J. N. (1994). Accelerating language development through picture book reading: Replication and extension to videotape training format. Journal of Educational Psychology, 86, Aslin, R. N., Saffran, J. R., & Newport E. L. (1998). Computation of conditional probability statistics by 8-month-old infants. Psychological Science, 9, Burns, M. S., Griffin, P., & Snow, C. E. (Eds.). (1999). Starting out right: A guide to promoting children s reading success. Washington, DC: National Academy Press. Bus, A. G., van IJzendoorn, M. H., & Pellegrini, A. D. (1995). Joint book reading makes for success in learning to read: A meta-analysis on intergenerational transmission of literacy. Review of Educational Research, 65, Bush, B. (1990, November). Parenting s best kept secret: Reading to your children. Reader s Digest, 137, Dale, P. S., Crain-Thoreson, C., Notari-Syverson, A., & Cole, K. (1996). Parent-child book reading as an intervention technique for young children with language delays. Topics in Early Childhood Special Education, 16(2), DeBaryshe, B. D. (1993). Joint picture book reading correlates of oral language skill. Journal of Child Language, 20, Eller, R. G., Pappas, C. C., & Brown, E. (1988). The lexical development of kindergartners: Learning from written context. Journal of Reading Behavior, 20, Elley, W. B. (1989). Vocabulary acquisition from listening to stories. Reading Research Quarterly, 24, Fenson, L., Dale, P. S., Reznick, J. S., Thal, D., Bates, E., Hartung, J. P., et al. (1993). MacAr- 252

12 Joint Book Reading thur communicative development inventories. San Diego, CA: Singular Publishing Group. High, P. H., Hopmann, M., LaGasse, L., Sege, R., Moran, J., Gutierrez, C., et al. (1999). Child centered literacy orientation: A form of social capital. Pediatrics, 103, Irwin, O. C. (1960). Infant speech: Effect of systematic reading of stories. Journal of Speech and Hearing, 3, Jusczyk, P. W. (1997). Infant speech perception and the development of the mental lexicon. In J. C. Goodman & H. C. Nusbaum (Eds.), The development of speech perception: The transition from speech to sounds to spoken words. Cambridge, MA: MIT Press. Jusczyk, P. W., & Hohne, E. A. (1997). Infants memory for spoken words. Science, 277, Moerk, E. L. (1985). Picture-book reading by mothers and young children and its impact upon language development. Journal of Pragmatics, 9, Neuman, S. B., Copple, C., & Bredekamp, S. (Eds.). (2000). Learning to read and write: Developmentally appropriate practices for young children. Washington, DC: National Association for the Education of Young Children. Parent Teacher Association. (1987). How to encourage a love of reading. Redbook, 169, Pemberton, E. F., & Watkins, R. V. (1987). Language facilitation through stories: Recasting and modeling. First Language, 7, Scarborough, H. S., & Dobrich, W. (1994). On the efficacy of reading to preschoolers. Developmental Review, 14, Scarborough, H. S., Dobrich, W., & Hager, M. (1991). Preschool literacy experience and later reading achievement. Journal of Learning Disabilities, 24(8), Senechal, M., & Cornell, E. H. (1993). Vocabulary acquisition through shared reading experiences. Reading Research Quarterly, 28, Snow, C., & Goldfield, B. A. (1983). Turn the page please: Situation-specific language acquisition. Journal of Child Language, 10, Stadler, A. S., & McEvoy, M. A. (2003). The effect of text genre on parent use of joint book reading strategies to promote phonological awareness. Early Childhood Research Quarterly, 18, Stevens, J. (1996). Applied multivariate statistics for the social sciences. Mahwah, NJ: Erlbaum. Strickland, C. (1993). Look who s reading! Wilson Library Bulletin, 67, U.S. Census Bureau. (n.d.). State and county quickfacts. Retrieved December 1, 2001, from facts.census.gov/qfd / states/20/20091.html. Valdez-Menchaca, M. C., & Whitehurst, J. G. (1992). Accelerating language development through picture book reading: A systematic extension to Mexican day care. Developmental Psychology, 28, Wells, G. (1985). Language development in the preschool years. Cambridge, England: Cambridge University Press. Whitehurst, G. J., Falco, F. L., Lonigan, C. J., Fischel, J. E., DeBaryshe, B. D., Valdez- Menchaca, M. C., et al. (1988). Accelerating language development through picture book reading. Developmental Psychology, 24, Yarosz, D. R., & Barnett, W. S. (2001). Who reads to young children?: Identifying predictors of family reading activities. Reading Psychology, 22,