Intervention: Dialogic Reading

Intervention: Dialogic Reading
Revised February 8, 2007

Appendices

Appendix A1.1 Study characteristics: Lonigan, Anthony, Bloomfield, Dyer, & Samwel, 1999 (randomized controlled trial)

Characteristic	Description
Study citation	Lonigan, C. J., Anthony, J. L., Bloomfield, B. G., Dyer, S. M., & Samwel, C. S. (1999). Effects of two shared-reading interventions on emergent literacy skills of at-risk preschoolers. Journal of Early Intervention, 22 (4), 306-322.
Participants	The study began with 110 children; 15 children left the child care centers, leaving a sample of 95 children. Most of the children were from low-income families. The mean age of the child participants was 45.1 months (range 25 to 64 months). Forty-six percent were female and 77% were African-American. Results for the 66 children who had been randomly assigned within center to the Dialogic Reading and no-treatment comparison conditions are included in this report.
Setting	The study took place in five child care centers in an urban area in Florida. Four centers served primarily children of families eligible for subsidized child care. The fifth center was affiliated with a church and approximately 25% of families served by the church center received a state child care subsidy.
Intervention	The study included two intervention groups: Dialogic Reading and typical shared book reading. The Dialogic Reading intervention is included in this review; results involving typical Shared Book Reading are included in the WWC Shared Book Reading report. In the Dialogic Reading condition, trained undergraduate volunteers engaged in Dialogic Reading intervention sessions for 10 to 15 minutes each day across a six-week period. Children were read to in small groups of three to five children in a location outside the classroom.
Comparison	Children in the no-treatment comparison group engaged in their standard preschool curriculum.
Primary outcomes and measurement	The primary outcome domains were children's oral language and phonological processing. The study used the following standardized measures: the Peabody Picture Vocabulary Test-Revised (PPVT-R), the Expressive One-Word Picture Vocabulary Test-Revised (EOWPVT-R), the Verbal Expression subscale of the Illinois Test of Psycholinguistic Abilities (ITPA-VE), and the Listening Comprehension subtest of the Woodcock-Johnson Psychoeducational Battery (WJ-LC). The study also utilized four measures of phonological processing: rhyme oddity detection, alliteration oddity detection, sound blending, and sound elision (see Appendices A2.1 and 2.3 for more detailed descriptions of outcome measures).
Teacher training	Undergraduate volunteer readers were trained in Dialogic Reading style using a videotape training method, which covered the two phases of Dialogic Reading. During the training, the trainees were presented with Dialogic Reading guidelines and watched vignettes of adult-child Shared Book Reading that followed or did not follow the guidelines. Trainees analyzed the vignettes and had one-on-one role plays with the trainer. The phase one and phase two training sessions lasted for 30 and 20 minutes respectively.

Top

Appendix A1.2 Study characteristics: Lonigan & Whitehurst, 1998 (randomized controlled trial)

Characteristic	Description
Study citation	Lonigan, C. J., & Whitehurst, G. J. (1998). Relative efficacy of parent and teacher involvement in a shared-reading intervention for preschool children from low-income backgrounds. Early Childhood Research Quarterly, 13 (2), 263-290.
Participants	The study began with 114 children from low-income households; 23 of these children left the child care center they were attending prior to the posttest, leaving 91 children in the sample. These 91 children were between 33 and 60 months of age at the time of pretest. Fifty-four percent were female and 91% were African-American and all children were from English-speaking homes. The children were randomly assigned within classroom to the intervention and comparison conditions.¹ Results for the 75 children who had been randomly assigned to the Dialogic Reading at school, Dialogic Reading both at school and at home, and comparison groups are included in this report.
Setting	The study took place in four child care centers in Nashville, Tennessee that served primarily children of families eligible for subsidized child care.
Intervention	The study included three intervention groups: Dialogic Reading at school, Dialogic Reading at home, and Dialogic Reading both at school and at home. The Dialogic Reading at home group is not included in this review because it is not center-based. The Dialogic Reading at school and the Dialogic Reading both at school and at home groups were combined for this review to reflect analyses conducted by the study authors and findings from the combined groups are used to determine the overall rating of effectiveness. Dialogic Reading was implemented over a six-week period. Teachers or aides conducted Dialogic Reading sessions with children in small groups of less than six children. Sessions were planned to take place every day for about 10 minutes. The study authors divided centers into low and high compliance centers based on the frequency level (i.e., high and low) of Dialogic Reading sessions. The WWC uses the findings for the low and high compliance centers combined to determine the overall rating of effectiveness; however, the WWC reports findings for the low and high compliance centers separately in Appendix A5.
Comparison	Children in the no-treatment comparison group did not participate in Dialogic Reading at home or at school.
Primary outcomes and measurement	The primary outcome domain was children's oral language use. Standardized tests included the PPVT-R, the EOWPVT-R, and the ITPA-VE. Lonigan and Whitehurst also included measures of verbal production (MLU, speech production, diversity, and semantic diversity) which are not included in this review because of attrition (see Appendix A2.1 for more detailed descriptions of outcome measures).
Teacher training	Teachers were trained in Dialogic Reading using a videotape training method which covered the two phases of Dialogic Reading. During the training, the trainees were presented with Dialogic Reading guidelines and watched vignettes of adult-child shared book reading on tape that followed or did not follow the guidelines. Trainees analyzed the vignettes and had one-on-one role plays with the trainer. The phase one and phase two training sessions lasted for 30 and 20 minutes respectively.
¹ Although the authors did some reassignment of children where necessary to obtain equality in pretest scores, the WWC did not downgrade the study because reassignment of children was minimal and because reassignment of children resulted in a more conservative test of the intervention effects.

Top

Appendix A1.3 Study characteristics: Wasik & Bond, 2001 (randomized controlled trial)

Characteristic	Description
Study citation	Wasik, B. A., & Bond, M. A. (2001). Beyond the pages of a book: Interactive book reading and language development in preschool classrooms. Journal of Educational Psychology, 93 (2), 243-250.
Participants	The study began with 127 children from low-income households ranging in age from 3.9 years to 4.6 years (mean age = 4.3 years). After assignment to groups, six children transferred to another school, leaving 121 children in the sample. The center that the children attended served primarily three- to five-year-old children who were eligible for free or reduced-price lunch. Among the children attending the center, 94% were African Americans. Four teachers were randomly assigned to the intervention and comparison conditions.
Setting	The study took place in a Title I early learning center in Baltimore, Maryland.
Intervention	The study examined the effect of interactive book reading plus extension activities reinforcing the use of target vocabulary in the book on children's language development. The interactive reading sessions were conducted by trained teachers with the whole class of children and the extension activities supporting the use of target vocabulary were conducted in various contexts. The intervention took place four days a week and lasted for 15 weeks in the children's classrooms. Two books were read per week; one book was read twice and the other book was read once.
Comparison	The comparison group teachers read the same books the same number of times that they were read in the intervention group. However, the comparison group teachers were not trained to use the interactive book reading strategies.
Primary outcomes and measurement	The primary outcome domain was children's oral language use. This domain was assessed with one standardized measure and two researcher-developed measures. The standardized measure was the PPVT-III. The researcher-developed measures were: (1) a receptive language measure, which was developed using a subset of vocabulary words presented in the interactive book reading intervention and (2) an expressive language measure, which contained pictures representing the words presented during the interactive book reading intervention. The PPVT-III is not considered in this review because an effect size could not be calculated with the information provided (see Appendix A2.1 for more detailed descriptions of outcome measures).
Teacher training	Teachers were trained in interactive book reading techniques and book reading extension activities by an experienced teacher trainer. During the training, teachers were instructed in how to use interactive book reading strategies (e.g., defining vocabulary words, asking open-ended questions, and giving children an opportunity to talk and be heard). The teacher trainer modeled these strategies in the intervention classrooms and assisted extension activities in the first four weeks of the intervention. In addition, each intervention teacher was given books and materials focused on early childhood topics and themes that would be used in the book reading and extension activities.

Top

Appendix A1.4 Study characteristics: Whitehurst, Arnold, Epstein, Angell, Smith, & Fischel, 1994 (randomized controlled trial)

Characteristic	Description
Study citation	Whitehurst, G. J., Arnold, D. S., Epstein, J. N., Angell, A. L., Smith, M., & Fischel, J. E. (1994). A picture book reading intervention in day care and home for children from low-income families. Developmental Psychology, 30 (5), 679-689.
Participants	The study began with 73 three-year-old children from low-income families; at immediate posttest 67 children remained in the sample. At entry into the study, the mean age of the children was 3.5 years. Forty-five percent were female, 55% were black, and 23% were Hispanic. The children were randomly assigned within classroom to intervention and comparison conditions.¹
Setting	The study took place in five day care centers in Suffolk County, New York, which served mainly children of families qualified for public subsidy of day-care costs under Title XX of the Federal Social Security Act.
Intervention	The study included two intervention conditions: a Dialogic Reading at school condition and a Dialogic Reading both at school and at home condition. The Dialogic Reading at school and the Dialogic Reading both at school and at home groups were combined for this review to reflect analyses conducted by the study authors and findings from the combined groups are used to determine the overall rating of effectiveness. However, the WWC reports findings for the two intervention groups versus the comparison group separately in Appendix A5.In the Dialogic Reading at school condition, the teacher or aide conducted the sessions in the classroom in small groups of no more than five children daily for about 10 minutes over a six week period. In the Dialogic Reading both at school and at home condition, children received similar small-group Dialogic Reading sessions at school and a one-on-one daily Dialogic Reading session at home with their parents.
Comparison	The comparison condition children participated in play activities in small groups of no more than five children daily for about 10 minutes. The play activities centered on construction toys that were not available in the classrooms before the study.
Primary outcomes and measurement	The primary outcome domain was children's oral language use. The study used the following standardized measures: the PPVT-R, the EOWPVT-R, and the ITPA-VE. The authors also utilized a researcher-developed measure called "Our Word" (see Appendix A2.1 for more detailed descriptions of outcome measures).
Teacher training	Teachers were trained in Dialogic Reading using a videotape training method, which presented the two phases of Dialogic Reading. During the training, the trainees were presented with a set of guidelines and taped vignettes of adult-child book reading that exemplified or did not follow the guidelines. Trainees critiqued the vignettes and had one-on-one role plays with the trainer. The phase one and phase two training sessions were presented three weeks apart and lasted for 30 and 20 minutes respectively. Parents were trained to use Dialogic Reading at home using the same videotape and similar training procedures that were used for teachers at their child's day-care centers.
¹ The authors did some reassignment of children where necessary to obtain equality in pretest scores. This reassignment of children did not lead the WWC to downgrade the study because reassignment of children was minimal and resulted in a more conservative test of the intervention effects.

Top

Appendix A1.5 Study characteristics: Whitehurst, Epstein, Angell, Payne, Crone, & Fischel, 1994 and Zevenbergen, Whitehurst, & Zevenbergen, 2003 (randomized controlled trial)

Characteristic	Description
Study citation	Whitehurst, G. J., Epstein, J. N., Angell, A. L., Payne, A. C., Crone, D. A., & Fischel, J. E. (1994). Outcomes of an emergent literacy intervention in Head Start. Journal of Educational Psychology, 86 (4), 542-555. Additional source: Zevenbergen, A. A., Whitehurst, G. J., & Zevenbergen, J. A. (2003). Effects of a shared-reading intervention on the inclusion of evaluative devices in narratives of children from low-income families. Journal of Applied Developmental Psychology, 24, 1-15. This study and its additional source are not included in the overall effectiveness rating because the intervention included a combination of Dialogic Reading and Sound Foundations, which does not allow the effects of Dialogic Reading alone to be determined. The study is also included in the WWC Sound Foundations intervention report.
Participants	The study began with 207 four-year-old at-risk low socioeconomic status children. Forty children did not complete the study, leaving 167 children in the final sample. The final sample of children was 46% Caucasian, 45% African-American, 8% Hispanic, and 1% Asian, and 44% of the sample was female. Fifteen classrooms¹ were randomly assigned to the intervention and comparison conditions.
Setting	The study took place in 15 classrooms from four Head Start centers in Suffolk County, New York.
Intervention	Children in the intervention conditions participated in an emergent literacy program at school (Dialogic Reading plus an adaptation of Sound Foundations) and one-on-one Dialogic Reading at home. Dialogic Reading occurred over a 30-week period and consisted of reading to children in small groups three to five times per week in the classroom and one-on-one reading at home with the same book. A different book was used each week and the researchers added hints to each book (e.g., wh- and recall prompts). Sound Foundations occurred at least two times per week for no more than 45 minutes per week over a 16-week period in the classroom. Children were introduced to seven consonant sounds at the beginning and ending of words, to two vowel sounds at the beginning of words, and to manuscript letters that correspond to curriculum sounds.
Comparison	Children in the no-treatment comparison group participated in their regular Head Start program.
Primary outcomes and measurement	Whitehurst, Epstein, et al. (1994) examined outcomes in the oral language, phonological processing, print knowledge, and early reading/writing domains. Children's oral language was measured by three standardized measures: PPVT-R; EOWPVT-R; and ITPA-VE. Children's literacy was measured by 18 subscales from the Developing Skills Checklist. Because of the large number of outcome measures (21), the study authors conducted a principal components analysis to reduce the data to four factors for the analyses: Language (oral language domain), Print concepts (print knowledge domain), Linguistic awareness (phonological processing domain), and Writing (early reading/writing domain) (see Appendices A2.1–2.4 for more detailed descriptions of outcome measures). Zevenbergen et al. (2003) tested additional oral language outcomes from the same study. They assessed children's narrative understanding by asking children to retell a story about a bus immediately after hearing the story. Their narrative was transcribed and was coded for general content (information) and children's use of narrative devices (references to character states, dialogue, and causal states) (see Appendix A2.1 for more detailed descriptions of outcome measures).
Teacher training	Parents and teachers were trained by the authors on Dialogic Reading using a 20-minute video which was combined with role-playing and discussion after viewing the video. Training occurred once at the beginning of the school year. Teachers and aides in the intervention classrooms were asked to keep a daily log of the reading activities. To observe compliance and provide feedback, each classroom was visited at least once every two weeks by one of the study authors. Specific training for Sound Foundations is not reported.
¹ Zevenbergen et al. (2003) reported that 16 classrooms were assigned randomly.

Top

Appendix A1.6 Study characteristics: Crain-Thoreson & Dale, 1999 (randomized controlled trial with attrition problems)

Characteristic	Description
Study citation	Crain-Thoreson, C., & Dale, P. S. (1999). Enhancing linguistic performance: Parents and teachers as book reading partners for children with language delays. Topics in Early Childhood Special Education, 19 (1), 28-39.
Participants	The study began with 37 children with mild to moderate language delays. All children were eligible for early childhood special education services. Five children did not complete the study, leaving 32 children remaining in the sample.¹ The mean age of the remaining children was 51.6 months (range 39 to 66 months) and 31.3% of the remaining sample was female. Results for the 22 children who were randomly assigned to the staff/practice and comparison conditions are included in this report.
Setting	The study took place in five classrooms in three school districts in the Pacific Northwest.
Intervention	The study included two intervention groups: a staff/practice group and a parent group. The staff/practice group is included in this review; the parent group was not included in the review because it was not center-based. Dialogic Reading occurred over an eight-week period during which staff engaged in book reading with individual children at least four times per week.
Comparison	Staff in the comparison group were trained on Dialogic Reading, but children did not participate individually in Dialogic Reading.
Primary outcomes and measurement	The primary outcome domain was children's oral language use which was measured by four non-standardized measures: mean length of utterances; number of utterances; number of different words used; and ratio of child participation. Children's vocabulary knowledge was measured by two standardized tests: PPVT-R and the EOWPVT-R (see Appendix A2.1 for more detailed descriptions of outcome measures).
Teacher training	Staff were trained on the Dialogic Reading program in two 1.5 hour instructional sessions held four weeks apart. Videotape training, live demonstration, and role-play were used to train teachers to use Dialogic Reading. In addition to learning Dialogic Reading principles, staff were instructed to pause and give children time to respond, and they kept logs of their shared reading activities.
¹ The study was downgraded by the WWC due to differential attrition between the intervention and comparison groups. The attrition was 0% for the staff/practice intervention group and 18% for the comparison group. The difference in attrition between groups was 18%.

Top

Appendix A2.1 Outcome measures in the oral language domain

Outcome measure	Description
Peabody Picture Vocabulary Test—Revised (PPVT-R)	A standardized measure of children's receptive vocabulary that requires them to identify pictures that correspond to spoken words (as cited in Crain-Thoreson & Dale, 1999; Lonigan & Whitehurst, 1998; Lonigan et al., 1999; Whitehurst, Arnold, et al., 1994).
Expressive One-Word Picture Vocabulary Test— Revised (EOWPVT-R)	A standardized measure of children's expressive vocabulary that requires them to name pictures of common objects, actions, and concepts (as cited in Crain-Thoreson & Dale, 1999; Lonigan & Whitehurst, 1998; Lonigan et al., 1999; Whitehurst, Arnold, et al., 1994).
Our Word	A researcher-developed measure designed to resemble the EOWPVT-R and measure children's knowledge of novel vocabulary introduced in books in the study (as cited in Whitehurst, Arnold, et al., 1994).
Illinois Test of Psycholinguistic Abilities—Verbal Expression Subscale (ITPA-VE)	A standardized measure of children's verbal fluency that requires them to describe four common objects (as cited in Lonigan & Whitehurst, 1998; Lonigan et al., 1999; White-hurst, Arnold, et al., 1994).
Number of child utterances	The number of child utterances during videotaped book sharing measured using the Computerized Language Analysis Programs (CLAN) (as cited in Crain-Thoreson & Dale, 1999).
Ratio of child participation	The number of child utterances divided by the sum of child and adult utterances during videotaped book sharing measured using the Computerized Language Analysis Pro-grams (CLAN) (as cited in Crain-Thoreson & Dale, 1999).
Mean length of utterances	The mean length of utterances during videotaped book sharing measured using the Computerized Language Analysis Programs (CLAN) (as cited in Crain-Thoreson & Dale, 1999).
Lexical diversity	The number of different words spoken by the child during videotaped book sharing measured using the Computerized Language Analysis Programs (CLAN) (as cited in Crain-Thoreson & Dale, 1999).
Woodcock-Johnson Psychoeducational Battery—Listening Comprehension Subtest (WJ-LC)	A standardized measure of children's listening comprehension that requires children to finish incomplete sentences by providing the missing word (as cited in Lonigan et al., 1999).
Language factor	A factor derived from a number of outcome measures (subscales from the Developing Skills Checklist, ITPA-VE, PPVT-R, and EOWPVT-R) using a principal components analysis. Factor loadings for the language factor were high for EOWPVT-R, PPVT-R, ITPA-VE, Tell a Story in Sequence, and Identify Function of Words-Numbers (as cited in Whitehurst, Epstein, et al., 1994).
Information/general content score	Children heard an adapted version of the Bus Story (Renfrew, 1969 as cited in Zevenbergen et al., 2003) and then retold the story. Researchers coded transcripts of the children's narrative to rate the general content of the children's story.
Reference to character states	Children heard an adapted version of the Bus Story (Renfrew, 1969 as cited in Zevenbergen et al., 2003) and then retold the story. Researchers coded transcripts of the children's narrative to determine the number of times children referred to the internal states of the characters in the story.
Dialogue	Children heard an adapted version of the Bus Story (Renfrew, 1969 as cited in Zevenbergen et al., 2003) and then retold the story. Researchers coded transcripts of the children's narrative to determine their usage of dialogue.
Causal statements	Children heard an adapted version of the Bus Story (Renfrew, 1969 as cited in Zevenbergen et al., 2003) and then retold the story. Researchers coded transcripts of the children's narrative to determine their usage of causal statements.
Receptive language measure	The measure was developed by the researchers using a subset of vocabulary words presented in the books used during the intervention and children were asked to identify the picture that corresponded to the target word (as cited in Wasik & Bond, 2001).
Expressive language measure	The measure was developed by the researchers and contained pictures representing the words in the books used during the intervention. Children were shown the pictures and asked to name the objects (as cited in Wasik & Bond, 2001).

Top

Appendix A2.2 Outcome measures in the print knowledge domain

Outcome measure	Description
Print concepts factor	A factor derived from a number of outcome measures (subscales from the Developing Skills Checklist, ITPA-VE, PPVT-R, and EOWPVT-R) using a principal components analysis. Factor loadings for the print concepts factor were high for Name Letters, Blend CVC Words, Rhyming, Identify People Reading, Distinguish Words-Pictures-Numbers, Identify Functions of Words-Numbers, and Identify Components of Writing (as cited in Whitehurst, Epstein, et al., 1994).

Outcome measure

Description

Print concepts factor

A factor derived from a number of outcome measures (subscales from the Developing Skills Checklist, ITPA-VE, PPVT-R, and EOWPVT-R) using a principal components analysis. Factor loadings for the print concepts factor were high for Name Letters, Blend CVC Words, Rhyming, Identify People Reading, Distinguish Words-Pictures-Numbers, Identify Functions of Words-Numbers, and Identify Components of Writing (as cited in Whitehurst, Epstein, et al., 1994).

Top

Appendix A2.3 Outcome measures in the phonological processing domain

Outcome measure	Description
Rhyme oddity detection	A researcher-developed measure designed to measure children's understanding of words that rhyme (as cited in Lonigan et al., 1999).
Alliteration oddity detection	A researcher-developed measure designed to measure children's understanding of differences at the beginnings of words (as cited in Lonigan et al., 1999).
Sound blending	A researcher-developed measure designed to measure children's ability to combine word elements to form a new word (as cited in Lonigan et al., 1999).
Sound elision	A researcher-developed measure designed to measure children's ability to take away parts of words and say the word that is left over (as cited in Lonigan et al., 1999).
Linguistic awareness factor	A factor extracted from a number of outcome measures (subscales from the Developing Skills Checklist, ITPA-VE, PPVT-R, and EOWPVT-R) using a principal components analysis. Factor loadings for the linguistic awareness factor were high on Identify Sounds and Letters, Identify Same-Different Sounds, Segment Sentences, and Segment Words (as cited in Whitehurst, Epstein, et al., 1994).

Top

Appendix A2.4 Outcome measures in the early reading/writing domain

Outcome measure	Description
Writing factor	A factor derived from a number of outcome measures (subscales from the Developing Skills Checklist, ITPA-VE, PPVT-R, and EOWPVT-R) using a principal components analysis. Factor loadings for the writing factor were high for Print in Left-Right Progression, Print First Name, and Write Message Mechanics (as cited in Whitehurst, Epstein, et al., 1994).

Top

Appendix A3.1 Summary of study findings included in the rating for the oral language domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (classrooms/children or children)	Dialogic Reading group³	Comparison group³	Mean difference⁴ (Dialogic Reading - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
Lonigan et al., 1999 (randomized controlled trial)⁸
PPVT-R	2-5 year olds	66	84.40 (17.28)	85.19 (14.01)	-0.79	-0.05	ns	-2
EOWPVT-R	2-5 year olds	66	88.51 (10.57)	87.97 (15.11)	0.54	0.04	ns	+2
ITPA-VE	2-5 year olds	66	45.46 (8.27)	40.81 (10.95)	4.65	0.48	Statistically significant	+18
WJ-LC	2-5 year olds	29	8.51 (3.84)	7.29 (4.27)	1.22	0.29	ns	+11
Average⁹ for oral language (Lonigan et al., 1999)						0.19	ns	+8
Lonigan & Whitehurst, 1998 (randomized controlled trial)¹⁰
PPVT-R-Form M	3-4 year olds	75	80.95 (10.54)	81.80 (13.35)	-0.85	-0.07	ns	-3
EOWPVT-R	3-4 year olds	75	87.37 (9.69)	86.92 (11.19)	0.45	0.04	ns	+2
ITPA-VE	3-4 year olds	75	109.09 (16.01)	102.60 (12.25)	6.49	0.43	ns	+17
Average⁹ for oral language (Lonigan & Whitehurst, 1998)						0.14	ns	+5
Wasik & Bond, 2001 (randomized controlled trial) ¹¹
Receptive language	3-4 year olds	4/121¹²	nr	nr	nr	1.58	Statistically significant	+44
Expressive language	3-4 year olds	4/121¹²	nr	nr	nr	2.05	Statistically significant	+48
Average⁹ for oral language (Wasik & Bond, 2001)						1.81	Statistically significant	+47
Whitehurst, Arnold, et al., 1994 (randomized controlled trial)¹³
EOWPVT-R	3 year olds	67	89.89 (13.40)	85.18 (16.73)	4.71	0.32	Statistically significant	+13
PPVT-R	3 year olds	67	86.49 (13.81)	83.68 (15.83)	2.81	0.19	ns	+8
Our Word	3 year olds	67	10.18 (5.49)	8.91 (7.00)	1.27	0.21	ns	+8
ITPA-VE	3 year olds	67	100.06 (15.60)	100.11 (16.98)	-0.05	0.00	ns	0
Average⁹ for oral language (Whitehurst, Arnold, et al., 1994)						0.18	ns	+7
Crain-Thoreson & Dale, 1999 (randomized controlled trial with attrition problems)¹⁴
Number of child utterances	3-5 year olds	22	43.36 (22.10)	36.43 (21.30)	6.93	0.31	ns	+12
Ratio of child participation	3-5 year olds	22	0.30 (0.11)	0.26 (0.12)	0.05	0.40	ns	+15
Mean length of utterances	3-5 year olds	22	3.00 (0.93)	2.73 (0.97)	0.27	0.27	ns	+11
Lexical diversity	3-5 year olds	22	55.29 (23.20)	52.56 (38.70)	2.73	0.09	ns	+3
PPVT-R	3-5 year olds	22	63.70 (21.80)	59.54 (16.20)	4.16	0.20	ns	+8
EOWPVT-R	3-5 year olds	22	70.12 (11.30)	71.73 (10.60)	-1.61	-0.14	ns	-6
Average⁹ for oral language (Crain-Thoreson & Dale, 1999)						0.19	ns	+7
Domain average⁹ for oral language across all studies						0.50	na	+19
ns = not statistically significant na = not applicable nr = not reported ¹ This appendix reports findings considered for the effectiveness rating and the average improvement indices. Subgroup and follow-up findings from the same studies are not included in these ratings, but are reported in Appendices A5.1and 5.2. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are: a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. For Lonigan and Whitehurst (1998) and Whitehurst, Arnold, et al. (1994) the intervention group standard deviations were calculated by the WWC based on subgroup standard deviations. ³ For Lonigan et al. (1999), the intervention group mean equals the comparison group mean plus the mean difference. For Lonigan and Whitehurst (1998), the means reported in the table reflect those from high and low compliance centers combined. For Wasik and Bond (2001), the effect sizes were calculated based on the sample sizes and F (1,120) = 76.61 and F (1,120) = 128.43. For Crain-Thoreson and Dale (1999), the posttest means are covariate-adjusted means provided by the study authors. For Lonigan and Whitehurst (1998) and Whitehurst, Arnold, et al. (1994), the intervention group means were calculated by WWC by combining the school and school plus home conditions. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. For Lonigan et al. (1999), the mean differences were computed by the WWC and took into account the pretest difference between the study groups. The resulting effect sizes may overestimate the intervention's effects when the intervention group had lower pretest scores than the comparison group, and underestimate the intervention's effect when the intervention group had higher pretest scores than the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools and for multiple comparisons. For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In the case of Lonigan et al. (1999), a correction for multiple comparisons was needed, but the significance levels do not differ from those reported in the original study. ⁹ The WWC-computed average effect sizes for each study and for the domain across studies are simple averages rounded to two decimal places. The average improvement indices are calculated from the average effect size. ¹⁰ In the case of Lonigan and Whitehurst (1998), no corrections for clustering or multiple comparisons were needed. The findings in this table represent a comparison of children in the combined Dialogic Reading at school and at school plus home group versus children in the comparison group in both high and low compliance centers. ¹¹ In the case of Wasik and Bond (2001), corrections for clustering and multiple comparisons were needed, but the significance levels do not differ from those reported in the original study. ¹² For Wasik and Bond (2001), four teachers were randomly assigned to the intervention and comparison groups. ¹³ In the case of Whitehurst, Arnold, et al. (1994), a correction for multiple comparisons was needed, so the significance levels may differ from those reported in the original study. ¹⁴ In the case of Crain-Thoreson and Dale (1999), no corrections for clustering or multiple comparisons were needed.

Top

Appendix A3.2 Summary of study findings included in the rating for the phonological processing domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (children)	Dialogic Reading group³	Comparison group³	Mean difference⁴ (Dialogic Reading - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
Lonigan et al., 1999 (randomized controlled trial)⁸
Rhyme oddity detection	2-5 year olds	61	3.74 (1.40)	3.90 (1.42)	-0.16	-0.11	ns	-4
Alliteration oddity detection	2-5 year olds	61	3.93 (1.30)	2.28 (1.28)	1.65	1.26	ns⁹	+40
Sound blending	2-5 year olds	61	2.37 (6.03)	2.83 (5.27)	-0.46	-0.08	ns	-3
Sound elision	2-5 year olds	61	2.85 (3.48)	3.55 (4.61)	-0.70	-0.17	ns	-7
Domain average¹⁰ for phonological processing						0.22	ns	+9
ns = not statistically significant ¹ This appendix reports findings considered for the effectiveness rating and the average improvement indices. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are: a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. ³ The intervention group mean equals the comparison group mean plus the mean difference. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. The mean differences were computed by the WWC and took into account the pretest difference between the study groups. The resulting effect sizes may overestimate the intervention's effects when the intervention group had lower pretest scores than the comparison group, and underestimate the intervention's effect when the intervention group had higher pretest scores than the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools and for multiple comparisons. For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In the case of Lonigan et al. (1999), no corrections for clustering or multiple comparisons were needed. ⁹ This statistical significance level was reported by the study authors. It differs from, but is more accurate than, the significance level based on the effect size presented in this table, which is an approximate effect estimate computed by the WWC. ¹⁰ This row provides the study average, which in this instance is also the domain average. The WWC-computed domain average effect size is a simple average rounded to two decimal places. The domain improvement index is calculated from the average effect sizes.

Top

Appendix A4.1 Summary of findings for Dialogic Reading plus Sound Foundations for the oral language domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (classrooms/children)	Dialogic Reading + Sound Foundations group³	Comparison group³	Mean difference⁴ (Dialogic Reading + Sound Foundations - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
Whitehurst, Epstein, et al., 1994 and Zevenbergen et al., 2003 (randomized controlled trial)⁸
Language factor	4 year olds	15/167	-0.02 (1.00)	-0.10 (1.00)	0.08	0.08	ns	+3
Character states	4 year olds	16/123	1.42 (1.82)	0.67 (0.86)	0.75	0.50	ns	+19
Dialogue	4 year olds	16/123	1.56 (1.44)	0.96 (0.92)	0.60	0.48	ns	+18
Causal state	4 year olds	16/123	0.18 (0.41)	0.33 (0.58)	-0.15	-0.30	ns	-12
Information/general content	4 year olds	16/123	87.54 (14.32)	87.40 (11.50)	0.14	0.01	ns	0
Domain average⁹ for oral language						0.15	ns	+6
ns = not statistically significant ¹ This appendix presents a summary of study findings for measures that fall in the oral language domain for a study that is not included in the overall effectiveness ratings. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are: a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. For Whitehurst, Epstein, et al. (1994), the standard deviations are not reported by the study author, but are reported as 1.00 by the WWC because standardized factor scores have a mean of zero and a standard deviation of one. ³ For Whitehurst, Epstein, et al. (1994), the intervention and comparison group means were estimated by the WWC from the y-axis of figure 2 in the Whitehurst, Epstein, et al. (1994) article. For Zevenbergen et al. (2003), the posttest means are covariate-adjusted means provided by the study authors. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools and for multiple comparisons. For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In the case of Whitehurst, Epstein, et al. (1994) and Zevenbergen et al. (2003), corrections for clustering were needed, but the significance levels do not differ from those reported in the original studies. ⁹ This row provides the study average, which in this instance is also the domain average. The WWC-computed domain average effect size is a simple average rounded to two decimal places. The domain improvement index is calculated from the average effect sizes.

Top

Appendix A4.2 Summary of findings for Dialogic Reading plus Sound Foundations for the print knowledge domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (classrooms/children)	Dialogic Reading + Sound Foundations group³	Comparison group³	Mean difference⁴ (Dialogic Reading + Sound Foundations - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
Whitehurst, Epstein, et al., 1994 (randomized controlled trial)⁸
Print concepts factor	4 year olds	15/167	0.26 (1.00)	-0.38 (1.00)	0.64	0.64	Statistically significant	+24
Domain average⁹ for print knowledge						0.64	Statistically significant	+24
¹ This appendix presents a summary of study findings for measures that fall in the print knowledge domain for a study that is not included in the overall effectiveness ratings. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are: a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. The standard deviations are not reported by the study author, but are reported as 1.00 by the WWC because standardized factor scores have a mean of zero and a standard deviation of one. ³ The intervention and comparison group means were estimated by the WWC from the y-axis of figure 2 in the Whitehurst, Epstein, et al. (1994) article. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools and for multiple comparisons. For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In the case of Whitehurst, Epstein, et al. (1994), a correction for clustering was needed, but the significance level does not differ from the one reported in the original study. ⁹ This row provides the study average, which in this instance is also the domain average. The WWC-computed domain average effect size is a simple average rounded to two decimal places. The domain improvement index is calculated from the average effect size.

Top

Appendix A4.3 Summary of findings for Dialogic Reading plus Sound Foundations for the phonological processing domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (classrooms/children)	Dialogic Reading + Sound Foundations group³	Comparison group³	Mean difference⁴ (Dialogic Reading + Sound Foundations - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
	Whitehurst, Epstein, et al., 1994 (randomized controlled trial)⁸
Linguistic awareness factor	4 year olds	15/167	0.08 (1.00)	0.06 (1.00)	0.02	0.02	ns	+1
Domain average⁹ for phonological processing						0.02	ns	+1
ns = not statistically significant ¹ This appendix presents a summary of study findings for measures that fall in the phonological processing domain for a study that is not included in the overall effectiveness ratings. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are: a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. The standard deviations are not reported by the study author, but are reported as 1.00 by the WWC because standardized factor scores have a mean of zero and a standard deviation of one. ³ The intervention and comparison group means were estimated by the WWC from the y-axis of figure 2 in the Whitehurst, Epstein, et al. (1994) article. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools and for multiple comparisons. For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In the case of Whitehurst, Epstein, et al. (1994), a correction for clustering was needed, but the significance level does not differ from the one reported in the original study. ⁹ This row provides the study average, which in this instance is also the domain average. The WWC-computed domain average effect size is a simple average rounded to two decimal places. The domain improvement index is calculated from the average effect size.

Top

Appendix A4.4 Summary of findings for Dialogic Reading plus Sound Foundations for the early reading/writing domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (classrooms/children)	Dialogic Reading + Sound Foundations group³	Comparison group³	Mean difference⁴ (Dialogic Reading + Sound Foundations - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
Whitehurst, Epstein, et al., 1994 (randomized controlled trial)⁸
Writing factor	4 year olds	15/167	0.20 (1.00)	-0.34 (1.00)	0.54	0.54	Statistically significant	+20
Domain average⁹ for early reading/writing						0.54	Statistically significant	+20
¹ This appendix presents a summary of study findings for measures that fall in the early reading/writing domain for a study that is not included in the overall effectiveness ratings. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are: a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. The standard deviations are not reported by the study author, but are reported as 1.00 by the WWC because standardized factor scores have a mean of zero and a standard deviation of one. ³ The intervention and comparison group means were estimated by the WWC from the y-axis of figure 2 in the Whitehurst, Epstein, et al. (1994) article. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools and for multiple comparisons. For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In the case of Whitehurst, Epstein, et al. (1994), a correction for clustering was needed, but the significance level does not differ from the one reported in the original study. ⁹ This row provides the study average, which in this instance is also the domain average. The WWC-computed domain average effect size is a simple average rounded to two decimal places. The domain improvement index is calculated from the average effect size.

Top

Appendix A5.1 Summary of subgroup findings for the oral language domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (children)	Dialogic Reading group³	Comparison group³	Mean difference⁴ (Dialogic Reading - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
Lonigan & Whitehurst, 1998 (randomized controlled trial; high compliance centers—Dialogic Reading at school)⁸
PPVT-R-Form M	3-4 year olds	31	80.80 (8.88)	80.70 (17.78)	0.10	0.01	ns	0
EOWPVT-R	3-4 year olds	31	91.20 (8.25)	86.80 (14.02)	4.40	0.37	ns	+14
ITPA-VE	3-4 year olds	31	106.80 (12.74)	100.20 (12.21)	6.60	0.52	ns	+20
Lonigan & Whitehurst, 1998 (randomized controlled trial; high compliance centers—Dialogic Reading both at school and at home)⁹
PPVT-R-Form M	3-4 year olds	23	79.00 (8.63)	80.70 (17.78)	-1.70	-0.10	ns	-4
EOWPVT-R	3-4 year olds	23	91.30 (7.00)	86.80 (14.02)	4.50	0.35	Statistically significant	+14
ITPA-VE	3-4 year olds	23	121.80 (16.69)	100.20 (12.21)	21.60	1.53	Statistically significant	+44
Lonigan & Whitehurst, 1998 (randomized controlled trial; high compliance centers—Dialogic Reading at school and both at school and at home)¹⁰
PPVT-R-Form M	3-4 year olds	38	80.23 (8.64)	80.70 (17.78)	-0.47	-0.03	ns	-1
EOWPVT-R	3-4 year olds	38	91.23 (7.71)	86.80 (14.02)	4.43	0.40	ns	+16
ITPA-VE	3-4 year olds	38	111.57 (15.46)	100.20 (12.21)	11.37	0.78	Statistically significant	+28
Lonigan & Whitehurst, 1998 (randomized controlled trial; low compliance centers—Dialogic Reading at school)¹¹
PPVT-R-Form M	3-4 year olds	27	80.10 (12.62)	83.40 (7.47)	-3.30	-0.29	ns	-12
EOWPVT-R	3-4 year olds	27	81.30 (11.59)	87.10 (8.97)	-5.80	-0.53	Statistically significant	-20
ITPA-VE	3-4 year olds	27	102.30 (15.91)	106.10 (14.28)	-3.80	-0.24	ns	-10
Lonigan & Whitehurst, 1998 (randomized controlled trial; low compliance centers—Dialogic Reading both at school and at home)¹²
PPVT-R-Form M	3-4 year olds	21	83.90 (11.33)	83.40 (7.47)	0.50	0.05	ns	+2
EOWPVT-R	3-4 year olds	21	88.60 (4.88)	87.10 (8.97)	1.50	0.20	ns	+8
ITPA-VE	3-4 year olds	21	114.50 (15.13)	106.10 (14.28)	8.40	0.55	ns	+21
Lonigan & Whitehurst, 1998 (randomized controlled trial; low compliance centers—Dialogic Reading at school and both at school and at home)¹³
PPVT-R-Form M	3-4 year olds	37	81.56 (12.06)	83.40 (7.47)	-1.84	-0.16	ns	-7
EOWPVT-R	3-4 year olds	37	84.11 (10.11)	87.10 (8.97)	-2.99	-0.30	ns	-12
ITPA-VE	3-4 year olds	37	106.99 (16.46)	106.10 (14.28)	0.89	0.06	ns	+2
Whitehurst, Arnold, et al., 1994 (randomized controlled trial; Dialogic Reading at school)¹⁴
EOWPVT-R	3 year olds	48	88.12 (10.43)	85.18 (16.73)	2.94	0.21	ns	+8
PPVT-R	3 year olds	48	85.73 (11.54)	83.68 (15.83)	2.05	0.15	ns	+6
Our Word	3 year olds	48	9.35 (5.00)	8.91 (7.00)	0.44	0.07	ns	+3
ITPA-VE	3 year olds	48	99.62 (14.50)	100.11 (16.98)	-0.49	-0.03	ns	-1
Whitehurst, Arnold, et al., 1994 (randomized controlled trial; Dialogic Reading both at school and at home)¹⁵
EOWPVT-R	3 year olds	41	92.32 (16.65)	85.18 (16.73)	7.14	0.42	ns	+16
PPVT-R	3 year olds	41	87.53 (16.72)	83.68 (15.83)	3.85	0.23	ns	+9
Our Word	3 year olds	41	11.32 (6.05)	8.91 (7.00)	2.41	0.36	ns	+14
ITPA-VE	3 year olds	41	100.66 (17.38)	100.11 (16.98)	0.55	0.03	ns	+1
ns = not statistically significant ¹ This appendix presents subgroup findings for measures that fall in oral language. Total group scores were used for rating purposes and are presented in Appendix A3.1. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are: a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. For Lonigan and Whitehurst (1998), the standard deviations for the Dialogic Reading at school and both at school and at home group for the low and high compliance centers were calculated by the WWC based on the standard deviations of the Dialogic Reading at school group and the Dialogic Reading both at school and at home group. ³ For Lonigan and Whitehurst (1998), the means for the Dialogic Reading at school and both at school and at home group for the low and high compliance centers were calculated by the WWC based on the means of the Dialogic Reading at school group and the Dialogic Reading both at school and at home group. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools (corrections for multiple comparisons were not done for findings not included in the overall intervention rating). For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In case of Lonigan and Whitehurst (1998) (high compliance centers—Dialogic Reading at school), no correction for clustering was needed. ⁹ In the case of Lonigan and Whitehurst (1998) (high compliance centers—Dialogic Reading both at school and at home), no correction for clustering was needed. ¹⁰ In the case of Lonigan and Whitehurst (1998) (high compliance centers—Dialogic Reading at school and both at school and at home), no correction for clustering was needed. ¹¹ In the case of Lonigan and Whitehurst (1998) (low compliance centers—Dialogic Reading at school), no correction for clustering was needed. ¹² In the case of Lonigan and Whitehurst (1998) (low compliance centers—Dialogic Reading both at school and at home), no correction for clustering was needed. ¹³ In the case of Lonigan and Whitehurst (1998) (low compliance centers—Dialogic Reading at school and both at school and at home), no correction for clustering was needed. ¹⁴ In the case of Whitehurst, Arnold, et al. (1994) (Dialogic Reading at school), no correction for clustering was needed. ¹⁵ In the case of Whitehurst, Arnold, et al. (1994) (Dialogic Reading both at school and at home), no correction for clustering was needed.

Top

Appendix A5.2 Summary of follow-up findings for the oral language domain¹

			Authors' findings from the study
			Mean outcome (standard deviation²)		WWC calculations
Outcome measure	Study sample	Sample size (children)	Dialogic Reading group³	Comparison group³	Mean difference⁴ (Dialogic Reading - comparison)	Effect size⁵	Statistical significance⁶ (at α= 0.05)	Improvement index⁷
Whitehurst, Arnold, et al., 1994 (randomized controlled trial; Dialogic Reading at school and both at school and at home)⁸
EOWPVT-R	3 year olds	52	91.16 (11.79)	88.07 (17.49)	3.09	0.23	Statistically significant	+9
PPVT-R	3 year olds	53	81.07 (14.83)	83.21 (17.63)	-2.14	-0.13	ns	-5
ITPA-VE	3 year olds	52	105.13 (16.27)	104.23 (24.95)	0.90	0.05	ns	+2
Whitehurst, Arnold, et al., 1994 (randomized controlled trial; Dialogic Reading at school)⁹
EOWPVT-R	3 year olds	37	91.17 (10.36)	88.07 (17.49)	3.10	0.23	ns	+9
PPVT-R	3 year olds	37	79.52 (14.99)	83.21 (17.63)	-3.69	-0.23	ns	-9
ITPA-VE	3 year olds	36	102.28 (15.67)	104.23 (24.95)	-1.95	-0.10	ns	-4
Whitehurst, Arnold, et al., 1994 (randomized controlled trial; Dialogic Reading both at school and at home)¹⁰
EOWPVT-R	3 year olds	29	91.14 (13.74)	88.07 (17.49)	3.07	0.19	ns	+8
PPVT-R	3 year olds	30	83.31 (14.59)	83.21 (17.63)	0.10	0.01	ns	0
ITPA-VE	3 year olds	29	109.22 (17.12)	104.23 (24.95)	4.99	0.23	ns	+9
ns = not statistically significant ¹ This appendix presents six-month follow-up findings for both combined groups (Dialogic Reading at school and Dialogic Reading both at school and at home) and for subgroups (Dialogic Reading at school; Dialogic Reading both at school and at home) for measures that fall in the oral language domain. Immediate posttest scores for the combined group were used for rating purposes and are presented in Appendix A3.1. ² The standard deviation across all students in each group shows how dispersed the participants' outcomes are; a smaller standard deviation on a given measure would indicate that participants had more similar outcomes. For Whitehurst, Arnold, et al. (1994), the standard deviations for the Dialogic Reading at school and both at school and at home group were calculated by WWC based on standard deviations of the Dialogic Reading at school group and the Dialogic Reading both at school and at home group. ³ For Whitehurst, Arnold, et al. (1994), the means for the Dialogic Reading at school and both at school and at home group were calculated by WWC based on means of the Dialogic Reading at school group and the Dialogic Reading at school and at home group. ⁴ Positive differences and effect sizes favor the intervention group; negative differences and effect sizes favor the comparison group. ⁵ For an explanation of the effect size calculation, see Technical Details of WWC-Conducted Computations. ⁶ Statistical significance is the probability that the difference between groups is a result of chance rather than a real difference between the groups. ⁷ The improvement index represents the difference between the percentile rank of the average student in the intervention condition and that of the average student in the comparison condition. The improvement index can take on values between -50 and +50, with positive numbers denoting favorable results. ⁸ The level of statistical significance was reported by the study authors or, where necessary, calculated by the WWC to correct for clustering within classrooms or schools (corrections for multiple comparisons were not done for findings not included in the overall intervention rating). For an explanation about the clustering correction, see the WWC Tutorial on Mismatch. See Technical Details of WWC-Conducted Computations for the formulas the WWC used to calculate statistical significance. In the case of Whitehurst, Arnold, et al. (1994) (Dialogic Reading at school and both at school and at home), no correction for clustering was needed. ⁹ In the case of Whitehurst, Arnold, et al. (1994) (Dialogic Reading at school), no correction for clustering was needed. ¹⁰ In the case of Whitehurst, Arnold, et al. (1994) (Dialogic Reading both at school and at home), no correction for clustering was needed.

Top

Appendix A6.1 Dialogic Reading rating for the oral language domain

The WWC rates the effects of an intervention in a given outcome domain as positive, potentially positive, mixed, no discernible effects, potentially negative, or negative.¹

For the outcome domain of oral language, the WWC rated Dialogic Reading as having positive effects. The remaining ratings (potentially positive effects, mixed effects, no discernible effects, potentially negative effects, negative effects) were not considered, as Dialogic Reading was assigned the highest applicable rating.

Rating received
Positive effects: Strong evidence of a positive effect with no overriding contrary evidence. Criterion 1: Two or more studies showing statistically significant positive effects, at least one of which met WWC evidence standards for a strong design. Met. Three of the five studies that reported oral language outcomes found statistically significant positive effects and all three studies met WWC evidence standards for a strong design. Criterion 2: No studies showing statistically significant or substantively important negative effects. Met. No studies showed statistically significant or substantively important negative effects.
¹ For rating purposes, the WWC considers the statistical significance of individual outcomes and the domain-level effect. The WWC also considers the size of the domain-level effects for ratings of potentially positive or potentially negative effects. See the WWC Intervention Rating Scheme for a complete description.

Rating received

Positive effects: Strong evidence of a positive effect with no overriding contrary evidence.

Criterion 1: Two or more studies showing statistically significant positive effects, at least one of which met WWC evidence standards for a strong design.

Met. Three of the five studies that reported oral language outcomes found statistically significant positive effects and all three studies met WWC evidence standards for a strong design.
Criterion 2: No studies showing statistically significant or substantively important negative effects.

Met. No studies showed statistically significant or substantively important negative effects.

¹ For rating purposes, the WWC considers the statistical significance of individual outcomes and the domain-level effect. The WWC also considers the size of the domain-level effects for ratings of potentially positive or potentially negative effects. See the WWC Intervention Rating Scheme for a complete description.

Top

Appendix A6.2 Dialogic Reading rating for the phonological processing domain

The WWC rates the effects of an intervention in a given outcome domain as positive, potentially positive, mixed, no discernible effects, potentially negative, or negative.¹

For the outcome domain of phonological processing, the WWC rated Dialogic Reading as having no discernible effects. It did not meet the criteria for positive effects, potentially positive effects, mixed effects, potentially negative effects, or negative effects, as no studies showed statistically significant or substantively important effects, either positive or negative.

Rating received
No discernible effects: No affirmative evidence of effects. Criterion 1: None of the studies shows a statistically significant or substantively important effect, either positive or negative. Met. The study did not show statistically significant or substantively important effects, either positive or negative.
Other ratings considered
Positive effects: Strong evidence of a positive effect with no overriding contrary evidence. Criterion 1: Two or more studies showing statistically significant positive effects, at least one of which met WWC evidence standards for a strong design. Not met. Only one study examined effects on phonological processing outcomes. Criterion 2: No studies showing statistically significant or substantively important negative effects. Met. The study did not have statistically significant or substantively important negative effects.
Potentially positive effects: Evidence of a positive effect with no overriding contrary evidence. Criterion 1: At least one study showing a statistically significant or substantively important positive effect. Not met. The study did not show statistically significant or substantively important positive effects. Criterion 2: No studies showing a statistically significant or substantively important negative effect and fewer or the same number of studies showing indeterminate effects than showing statistically significant or substantively important positive effects. Not met. The study did not have statistically significant or substantively important negative effects, but it did show indeterminate effects.
Mixed effects: Evidence of inconsistent effects as demonstrated through either of the following criteria. Criterion 1: At least one study showing a statistically significant or substantively important positive effect, and at least one study showing a statistically significant or substantively important negative effect, but no more such studies than the number showing a statistically significant or substantively important positive effect. Not met. The study did not show statistically significant or substantively important positive effects, or statistically significant or substantively important negative effects. Criterion 2: At least one study showing a statistically significant or substantively important effect, and more studies showing an indeterminate effect than showing a statistically significant or substantively important effect. Not met. The study did not have statistically significant or substantively important effects. The study did have indeterminate effects.
Potentially negative effects: Evidence of a negative effect with no overriding contrary evidence Criterion 1: At least one study showing a statistically significant or substantively important negative effect. Not met. The study did not show statistically significant or substantively important negative effects. Criterion 2: No studies showing a statistically significant or substantively important positive effect, or more studies showing statistically significant or substantively important negative effects than showing statistically significant or substantively important positive effects. Met. The study did not show statistically significant or substantively important positive effects.
Negative effects: Strong evidence of a negative effect with no overriding contrary evidence. Criterion 1: Two or more studies showing statistically significant negative effects, at least one of which met WWC evidence standards for a strong design. Not met. Only one study examined effects on phonological processing outcomes. Criterion 2: No studies showing statistically significant or substantively important positive effects. Met. The study did not show statistically significant or substantively important positive effects.
¹ For rating purposes, the WWC considers the statistical significance of individual outcomes and the domain-level effect. The WWC also considers the size of the domain-level effect for ratings of potentially positive effects. See the WWC Intervention Rating Scheme for a complete description.

Rating received

No discernible effects: No affirmative evidence of effects.

Criterion 1: None of the studies shows a statistically significant or substantively important effect, either positive or negative.

Met. The study did not show statistically significant or substantively important effects, either positive or negative.

Other ratings considered

Positive effects: Strong evidence of a positive effect with no overriding contrary evidence.

Criterion 1: Two or more studies showing statistically significant positive effects, at least one of which met WWC evidence standards for a strong design.

Not met. Only one study examined effects on phonological processing outcomes.
Criterion 2: No studies showing statistically significant or substantively important negative effects.

Met. The study did not have statistically significant or substantively important negative effects.

Potentially positive effects: Evidence of a positive effect with no overriding contrary evidence.

Criterion 1: At least one study showing a statistically significant or substantively important positive effect.

Not met. The study did not show statistically significant or substantively important positive effects.
Criterion 2: No studies showing a statistically significant or substantively important negative effect and fewer or the same number of studies showing indeterminate effects than showing statistically significant or substantively important positive effects.

Not met. The study did not have statistically significant or substantively important negative effects, but it did show indeterminate effects.

Mixed effects: Evidence of inconsistent effects as demonstrated through either of the following criteria.

Criterion 1: At least one study showing a statistically significant or substantively important positive effect, and at least one study showing a statistically significant or substantively important negative effect, but no more such studies than the number showing a statistically significant or substantively important positive effect.

Not met. The study did not show statistically significant or substantively important positive effects, or statistically significant or substantively important negative effects.
Criterion 2: At least one study showing a statistically significant or substantively important effect, and more studies showing an indeterminate effect than showing a statistically significant or substantively important effect.

Not met. The study did not have statistically significant or substantively important effects. The study did have indeterminate effects.

Potentially negative effects: Evidence of a negative effect with no overriding contrary evidence

Criterion 1: At least one study showing a statistically significant or substantively important negative effect.

Not met. The study did not show statistically significant or substantively important negative effects.
Criterion 2: No studies showing a statistically significant or substantively important positive effect, or more studies showing statistically significant or substantively important negative effects than showing statistically significant or substantively important positive effects.

Met. The study did not show statistically significant or substantively important positive effects.

Negative effects: Strong evidence of a negative effect with no overriding contrary evidence.

Criterion 1: Two or more studies showing statistically significant negative effects, at least one of which met WWC evidence standards for a strong design.

Not met. Only one study examined effects on phonological processing outcomes.
Criterion 2: No studies showing statistically significant or substantively important positive effects.

Met. The study did not show statistically significant or substantively important positive effects.

¹ For rating purposes, the WWC considers the statistical significance of individual outcomes and the domain-level effect. The WWC also considers the size of the domain-level effect for ratings of potentially positive effects. See the WWC Intervention Rating Scheme for a complete description.

Top

Institute of Education Sciences
What Works Clearinghouse

U.S. Department of Education