Method

The survey and assessment sites consisted of the experimental site, or Reach Out and Read treatment, and a control site, without a Reach Out and Read program. At the time of the study, both sites were medical facilities serving young children, including those from underinsured families. Children and their parents came to the medical sites for routine well-child visits and health care.

The sample population came from a small city and nearby, surrounding communities. The city population make-up included 17.72% African American, 27.8% Hispanic, and 64.65 % White, according to the 2000 census. A total of 14,873 families were in the community with 15.5% of the families falling below the poverty line (Wikipedia, 2007).

Before the project began in each of the sites, the medical staffs expressed concern that there would be more participant families than could be logistically accommodated. Once at the sites, the randomized selection of families was abandoned for simply seeking participation from every family arriving for a six month well-child visit with the pediatricians in a time frame agreed upon by all involved. It was found that the staffs underestimated the number of families coming in for a six-month well-child visit right at six months of age.

Data Collection: RORA Parent Survey
Data collection was administered by the principal investigator at the child’s 6 month well-baby visit with the help of an assistant. A follow-up was completed with the principal investigator at the 12 month well-child visit. At both visits and both sites, a parent survey and a child assessment were completed.

The RORA parent survey was an updated and lengthened version of the Before and After Books and Reading (BABAR)  parent survey designed by ROR co-founder, Robert Needlman, MD (ROR, 2001). A complete survey can be found in the appendix. Questions that were replications from the BABAR parent survey included the following questions:

• What are your three favorite things to do with [child’s name] these days?
• What do you do to help prepare [name] for sleep at night?
• Is there anything you do with [name] now that will help him/her be successful when he/she goes to kindergarten?
• Do you ever read children’s books to [name]?
• How many books altogether do you have at home that you read to [name]?
• How many days each week do you read children’s books to [name]?

Other questions were modified in some way from the BABAR questions and included the following:

• What languages are spoken in (by) your home/family?
• How many years of school have you completed? Where?
• Altogether how many books for reading to your child have you received from the doctor or nurse here?

Additional questions were designed to elicit additional information about family literacy and included the following:

• Which of the following have you used or read in your home during the last week?
  ___ books   ___ magazines   ___newspapers   ___ paper and pen
• How many other adults are in the home?
• How many other children are in the home?
• Do any of the other adults (or older children) read to, sing to, or tell stories to [name]?

Questions designed to further validate the ROR model at the second visit included the following questions:

• Has [child’s name] doctor talked with you or your family about reading to [child’s name]?
• Did the doctor give you good ideas about how to read to [child’s name]?
• Have you ever seen volunteers reading to children in the waiting room here?
• Did your child listen to the volunteers reading in the waiting room?

Data Collection: PLS-4—Child Assessment
The PLS-4: Preschool Language Scale, Fourth Edition (Zimmerman, 2002) was used for the child assessment at the 6 month and 12 month ages. Subscales from the standardized test include auditory comprehension and expressive communication and yield norm-referenced scores. A total language score was also provided. Both the Spanish and English language versions were used.

The Sample
All families asked to participate at the first visit agreed and expected to be contacted again at the 12 month well-child visit. Compensation was two board books from ROR, an I Am Your Child video tape, and reading information from the National Institute for Literacy (the information and videos were provided in either Spanish or English).

Of the 22 families participating at the ROR site, one participant family was immediately removed from the sample, because his age was not close to six months. On completion of the whole project, three more participant ROR families were removed from the sample as their ages at one of visits was more than a month off of the others. Three families did not show up for the second visit. Families that did not show up for appointments did make it to the next scheduled appointment 50% of the time. Rescheduled appointments after that were not kept, or were made when the child was much older than six months. One family could not be contacted by phone or by mail. The final sample from the ROR site, completing both visits, was 14 families.

At the control site, one participant family was removed from the 20 families because the parent was on the staff at the clinic and had heard our presentation about the project. By the second visit, one participant family had moved away; three could not be contacted by phone or by mail; two could not be scheduled while the child was close to 12 months; two were no shows repeatedly; and one refused to participate because of time. Two of the final 10 participant families needed to be rescheduled for another appointment. These numbers indicate some of the limitations of getting a good sample for experimental research.

In the final sample of participant families for the RORA project, 60.9% were from the treatment site and 39.1% were from the control site. Of the 24 adult participants, 22 were Hispanic, one was African American, and one was White. All of the adult participants who primarily answered the parent survey questions were the mothers of the children. In the final sample, 78.3% of the participant families chose to have the parent survey and child assessment conducted in Spanish language, and 21.7% chose English language. Responses about the home language found at the ROR site found 14.3% were bilingual and 85.7% Spanish. Informal notes recorded three of the ROR site Spanish speakers had a small amount of English language. At the control site, the breakdown was 11.1% English; 33.3% bilingual; and 55.6% Spanish.

The ROR group reported an average of 7.6 years of school completed. The control group reported 8.4 years of school completed. At the ROR site, 85.7% of participants attended school in Mexico and 14.3% attended school in Texas. The control site found 55.6% attended school in Mexico and 44.4% attended school in Texas.

At the time of the study, the sample from the treatment (ROR) site represented a population that was underinsured. The control site had health services available for all, but in being one of the top providers for well-child check-ups, the population sample did reflect the underinsured. Informal conversations with the control site participants in the final sample revealed that at least three of the parents were currently enrolled in special programs for students with children in local school districts. The students were enrolled in child development classes. While the population samples were similar on the surface, they were not identical.

The sample reflected parent participants born in Mexico and the United States. Final statistics reflected the ROR site as having 85.7% of participants from Mexico and 14.3% from the United States. At the control site, 55.6% were from Mexico and 44.4% from the United States. This difference in the number of recent immigrants, contributed to contrasts in the data results.

Sixteen of the children were male and eight were female. All of the birth weights fell in the normal range. The average age of the children at the first visit was 6 months-8 days; 6 months-7 days for the treatment group; and 6 months-10 days for the control group. At the second visit, the average age was 12 months-6.6 days; 12 months-4.8 days for the treatment group; and 12 months-9 days for the control group.

The initial look at the 24 families in the parent survey revealed an outlier from the control site. The outlier was a family displaced to the area by a hurricane. In reporting the number of books at home during the second visit, the number of books reported for one child was completely out of line with the rest of the sample. The outlier was removed for the data analysis. Analysis of the data was conducted on 14 treatment site participant families and 9 control site participant families.

Interview Process
Disruption of the normal procedures at a well-child visit was minimized as much as possible, with the survey and assessment completed before the pediatrician completed the exams. Roughly two months were spent visiting each site for the first visits and, again, for the second visits. Cooperation and understanding from the nursing staff are critical to the success of such a project.

The parent survey was conducted as a verbal questionnaire, in Spanish or English, in a clinic exam room with all participants seated. Printed copies of the survey were available at the interview with questions and instructions in Spanish and English. During the first visit, introductions to the principal investigator and a bilingual assistant were made, permissions were secured, and questions about the purpose of the study satisfied. As the study sample involved mostly Hispanic families from Mexico, the primary bilingual assistant, a Hispanic woman, through nearly all of the first visits also needed to reassure the families about the purpose of the study. Without this careful introduction to RORA, completing the second visits may have been less successful.

Parent survey questions were asked by the assistant or the principal investigator and quickly recorded on the survey instruments, attached to clipboards. All of the assistants used in the RORA were trained ahead of time in asking the questions in a consistent, non-leading manner, without any misleading comments about responses, and remaining nonjudgmental.

During the questionnaire time, informal play with toys not involved with the child assessment were introduced by the principal investigator on a small rolling table placed in front of the child, who always sat on the mother’s lap.

Figure 2: RORA rolling table at ROR site
Photo by J. Gramann

In addition, the principal investigator and the assistant noted occasional qualitative observations informally on their clipboards during the interview and assessment process. As the parent questions finished, the PLS-4 assessment began by introducing new toy selections to the table and removing the other toys to shelves under the rolling table. The child assessment was conducted in an informal, low-key, play atmosphere with simple verbal cues only when needed. The parent was elicited into the play process as needed and provided responses to the questions about what the child has done at other times. The principal investigator conducted all of the child assessments, striving for consistency across all subjects.

Table of Contents | Next | Previous