Search tips
Search criteria 


Logo of pediatricsLink to Publisher's site
Pediatrics. 2012 September; 130(3): 492–499.
PMCID: PMC3428755

The Impact of a Healthy Media Use Intervention on Sleep in Preschool Children



Although observational studies have consistently reported an association between media use and child sleep problems, it is unclear whether the relationship is causal or if an intervention targeting healthy media use can improve sleep in preschool-aged children.


We conducted a randomized controlled trial of a healthy media use intervention in families of children aged 3 to 5 years. The intervention encouraged families to replace violent or age-inappropriate media content with quality educational and prosocial content, through an initial home visit and follow-up telephone calls over 6 months. Sleep measures were derived from the Child Sleep Habits Questionnaire and were collected at 6, 12, and 18 months after baseline; repeated-measures regression analyses were used.


Among the 565 children analyzed, the most common sleep problem was delayed sleep-onset latency (38%). Children in the intervention group had significantly lower odds of “any sleep problem” at follow-up in the repeated-measures analysis (odds ratio = 0.36; 95% confidence interval: 0.16 to 0.83), with a trend toward a decrease in intervention effect over time (P = .07). Although there was no significant effect modification detected by baseline sleep or behavior problems, gender, or low-income status, there was a trend (P = .096) toward an increased effect among those with high levels of violence exposure at baseline.


The significant effects of a healthy media use intervention on child sleep problems in the context of a randomized controlled trial suggest that the previously reported relationship between media use and child sleep problems is indeed causal in nature.

KEY WORDS: sleep, media, preschool-aged children, randomized controlled trial, violence

What’s Known on This Subject:

Although observational studies have consistently reported an association between media use and child sleep problems, it is unclear whether the relationship is causal or if an intervention targeting healthy media use can improve sleep in preschool-aged children.

What This Study Adds:

This study demonstrates that a healthy media use intervention can improve child sleep outcomes and adds evidence that the relationship between media and sleep in preschool-aged children is indeed causal in nature.

Observational studies have consistently shown an association between media use and child sleep problems. Although this finding has been consistently replicated with high levels of use, bedtime use, and violent or frightening media content,17 effects have been observed across cultures and in all media formats (television,6,810 video games,1012 and computers11,13,14) and across the age spectrum, including preschoolers,2,79 school-aged children,1,36 and adolescents and adults.1315 Much of the existing research linking media use to child sleep problems has been cross-sectional, however, raising the possibility that the causality is reversed; that is, that sleep problems are leading to increased media use, evening media use, and exposure to violent media content, rather than the other way around. Although longitudinal studies have demonstrated that early childhood media use is associated with behavioral and developmental effects years later,8,1619 there is still the concern that these findings may be at least partly due to residual confounding by third factors, such as child behavior problems or household instability, that may lead to both problem media use and child sleep problems.8,1618,20

In this article, we leverage the experimental design of a randomized controlled trial of healthy media use in preschool-aged children to ascertain whether changes in media use in the intervention group led to differences in child sleep between the intervention and control groups. The randomization of a sample of this size eliminates most confounding, because the random assignment leads to equal balance between study groups in child and family characteristics. The longitudinal nature of the 18-month follow-up period also allows us to examine to what degree any effects observed are transient versus lasting.


We conducted a randomized controlled trial of a healthy media use intervention for families of preschool-aged children. This work was supported by grant 5R01HD056506 from the National Institute of Child Health & Human Development to Dr Christakis, and the study was approved by the institutional review board of the Seattle Children's Research Institute. Although the present analysis focuses on sleep outcomes, the primary outcomes of the trial were decreased aggressive behavior and increased prosocial behaviors, as is stated in the clinical trial registration; the analysis of these primary outcomes will be presented in a separate article.


Letters describing the study were sent to families with age-eligible children (3–5 years) enrolled in community pediatric practices in the Seattle metropolitan area, without regard to whether the child had been seen in the clinic recently. Participating clinics were selected to capture a representative sample of the Seattle population and included clinics that have substantial Medicaid populations. To be eligible, children needed to consume at least some media each week and to have English-speaking parents. Families were given the opportunity to opt out of further recruitment efforts and also had the option to opt in by returning a postage-paid mailer. In a separate analysis, the differences between these 2 groups were found to be minimal.21 Those who neither opted out nor in were contacted by way of telephone and asked to participate. Attempts were made to oversample low-income families at the initial contact stage, as identified by Medicaid status or zip code of residence. Randomization was stratified by low-income status and study case manager. After enrollment, study staff collected the survey and media diary during a home visit at the start of the intervention. For all follow-up surveys, families had the option of returning them by mail or submitting them online.

Because the primary objective of the randomized controlled trial was to decrease aggressive behaviors and increase prosocial behaviors, the sample size was based on power calculations for these aims. Of the 3334 families contacted and assessed for eligibility (Fig 1), 40% did not meet inclusion criteria, 35% declined to participate, and the remaining 25% (N = 820) were randomly allocated to the intervention or control arm within the DatStat Discovery Research Management System (DatStat, Inc, Seattle, WA) by using a computer-generated list of random numbers; of these, 617 (75%) completed the baseline visit between March 2009 and April 2010, and 565 of these (92%) completed at least 1 follow-up survey over the following 18 months and had sufficient data to be included in this analysis.

Consolidated Standards of Reporting Trials diagram.


The fundamental approach of this study was to try to get parents to substitute prosocial and educational content in place of violent content, because considerable research has shown that violent media exposure can lead to aggression and other behavioral and emotional problems in young children.1719,2225 The intervention framework was based on social cognitive theory2629 and sought to increase parental outcome expectations and self-efficacy around making healthy media choices for the child, with a specific emphasis on replacing violent or age-inappropriate content with age-appropriate educational or prosocial content.

Although the intervention addressed all screen time (television, DVDs/videos, computers, video games, and handheld devices), the primary focus was on television and videos because these formats account for the most screen time engaged in by preschool-aged children. No attempt was made to reduce the total number of screen time hours, nor was evening use or any other aspect of sleep hygiene specifically targeted.

Case managers received previous training in motivational interviewing techniques, the effects of violent television on children, and the use of parental controls in televisions and other media devices to monitor and modify viewing. Intervention sessions began with the initial home visit, in which the family’s assigned case manager collected assessment materials, discussed the child’s current media use with the parent, shared intervention handouts that were specific to the family’s needs, and engaged the parent in goal setting and anticipatory problem-solving around barriers. The home visit was then followed by mailings and follow-up telephone calls with the case manager for 12 months. During both the home visit and the telephone calls, case managers used their training to try to motivate and empower parents to replace violent and age-inappropriate media content with content that was age-appropriate and educational or prosocial in nature. Parents also were encouraged to engage in co-viewing and to discuss media content with their children, because co-viewing can increase parent awareness of the media content consumed, and some evidence suggests that co-viewing or discussion may enhance the positive effects of educational and prosocial media,30,31 although it may not mitigate the negative effects of violent or scary media.32

The monthly mailings included a program guide tailored to the family’s available channels with recommended educational and prosocial television shows and schedules and a newsletter with tips and reinforcement. The criteria for recommended programming were based on previous research regarding the positive benefits of educational and prosocial media for young children3342 and were as follows: a TV Guidelines rating of TV-Y, being rated high on educational and/or prosocial value by, and being currently available on network or cable television. In creating our final list, we sought to include programming that featured a diversity of gender, race, ethnicity, and culture and balance in terms of topical focus. Representative recommended shows included Curious George, Sesame Street, and Dora the Explorer.

The first 6 mailings also included DVDs with 5- to 10-minute clips of suggested educational and prosocial shows. During the monthly telephone calls, the case manager reviewed progress made on the parent’s goals since the last encounter, coached the parent through problem-solving around barriers as needed, and worked with the parent to set new goals as appropriate. The control group received a nutrition intervention, with analogous materials, including monthly mailings that encouraged families to decrease consumption of sugary drinks and increase fruit and vegetable intake.


Sleep measures were a subset of the Child Sleep Habits Questionnaire (CSHQ), assessing the frequency of sleep-onset latency, repeated night wakings, nightmares, difficulty waking in the morning, and daytime tiredness.43,44 Frequency choices were “2: Usually (5–7 days per week),” “1: Sometimes (2–4 days per week),” and “0: Rarely (0–1 day per week).” The items are summed after inverting the reversed item (latency), with total score ranging from 0 to 10 and higher scores indicating more problematic sleep, which is the same scoring method used in the full Child Sleep Habits Questionnaire. The primary outcome dichotomized the sleep scale at 0 to 1 (59% at baseline) versus 2 to 10 (41%), so that children were defined as having “any sleep problem” if the parent answered that at least 1 sleep problem occurred 5 to 7 days per week, or at least 2 sleep problems occurred 2 to 4 days per week. The sleep measures were collected at baseline and each follow-up time point (6, 12, and 18 months after beginning the intervention).


Demographic variables were collected in the baseline parent survey, including child gender and age in months at baseline. Children were defined as living in a low-income household if the parent-reported income category placed the family at or below twice the federal poverty level for their household size. For continuity, families were asked to choose 1 parent to receive the intervention and complete the baseline and all follow-up surveys; whether this respondent was the child’s mother as opposed to another primary caregiver (such as father or grandmother) is included in the analyses as a covariate. We also collected data on child internalizing (such as anxiety and depression) and externalizing (such as hyperactivity and aggression) symptoms using the Social Competence and Behavior Evaluation (SCBE), parent version. The SCBE is a validated measure with subscales for internalizing (anxious, depressive, and withdrawn) and externalizing (angry, aggressive, and oppositional) behaviors.45 Neither SCBE scale includes any questions about child sleep. One-week media diaries were collected at baseline, in which parents prospectively recorded all media use for the child. For each instance of media use, parents recorded format (television, DVD/video, computer, video game), content title, start time and duration, and co-use. Media diaries were coded for content as previously described.2


Missing data for the questions contributing to the sleep, internalizing, and externalizing scales were imputed by using the Stata (Stata Corp, College Station, TX) impute command for subjects with no >20% of items missing in the scale. Imputation enabled us to still include in the analyses children whose parents had left blank 1 or 2 items within a scale, which is especially important given that low-income families were more likely to have missed at least 1 item within a scale (9% vs 4% for the externalizing behavior scale, P = .01), and 1 of the goals in imputation in these situations is to minimize the degree to which nonrandom missingness biases the final results.4648 Each item used in the analysis had <2% of values missing. Descriptive and bivariate statistics were calculated, with t tests used to compare continuous variables between the intervention and control groups, and χ2 tests for categorical variables. The main intervention effect was tested in an intent-to-treat analysis using a repeated-measures logistic regression, with whether the child met criteria for having “any sleep problem” as the primary outcome. A repeated-measures regression analysis utilizes data from each available follow-up assessment per subject but takes into account the fact that different measurements across times within the same child will be more highly correlated with each other than measurements across children. The results of a repeated-measures analysis can be interpreted as the overall effect of the intervention on the child.

The intervention effect was then examined separately for each follow-up time point by using linear regression, with the child’s sleep problem score at that follow-up as the outcome. Secondary analyses examined the individual components of the sleep problem score, also by using repeated-measures logistic regression models, with whether the given sleep problem occurred ≥2 days per week as the outcome. To test for potential effect modification, the model from the primary analysis also was conducted by using interaction terms by each of gender, baseline sleep problem status, low-income household, and whether the child was in the top quartile for internalizing or externalizing behavior problem scores. The Wald test was used to compare the interaction terms and test the hypotheses regarding effect modification.

All regression models controlled the child's baseline sleep problem score, the child's gender and age in months, low-income household status, and whether the mother was the respondent. In the repeated-measures analyses, we also included covariates for the time point to account for the natural change in prevalence of sleep problems as children age, and an interaction term for a time × intervention to assess the decay of the intervention effect over time. We also tested whether gender, baseline media use, or SCBE internalizing and externalizing behavior problem scores were a significant contribution to the model. All analyses were conducted by using Stata/SE, version 10.


A total of 565 families completed at least 1 follow-up and had sufficient data to be included in this analysis, which is 92% of those completing the baseline visit (Fig 1). The population was representative of the Seattle metropolitan area and well balanced across study arms (Table 1), although somewhat more children in the control group (43%) had an older sibling as compared with the intervention group (34%). There were no significant differences between study arms at baseline in child sleep or media use. The most common sleep problem at baseline was difficulty with sleep-onset latency, with 26% of children taking >20 minutes to fall asleep for 2 to 4 nights a week and 12% taking >20 minutes to fall asleep 5 to 7 nights a week (Fig 2). In the primary analysis, gender, baseline media use, and SCBE internalizing and externalizing behavior problem scores did not contribute significantly to the model; because the P values for these covariates were >.20 and their inclusion did not significantly alter the intervention effort or improve model fit, they were not included in any of the final models.

Study Sample
Baseline sleep problems.

The intervention had a positive impact on sleep (Table 2), with children in the intervention group having significantly lower odds of “any sleep problem” in the repeated-measures analysis (odds ratio [OR] = 0.36; 95% confidence interval [CI]: 0.16 to 0.83). As expected, the effect size for the time variable (OR = 0.59; 95% CI: 0.46 to 0.77) shows a decrease in sleep problems over the 18 months of follow-up. The time × intervention interaction term, however, demonstrates a trend toward a decrease in intervention effect over time (OR = 1.40; 95% CI: 0.97 to 2.02). When linear regression models were run separately for each follow-up time point, similar effect sizes are seen at 6 months (difference in score between study arms of −0.16, 95% CI: −0.35 to 0.02) and 12 months (−0.14; 95% CI: −0.31 to 0.02) after the baseline visit, with a decay in effect at the 18-month follow-up (−0.08; 95% CI: −0.27 to 0.10). When examining the repeated-measures logistic regression models for each separate component of the sleep problem score, the greatest difference between study arms was for the question asking whether the child had a difficult time becoming awake and alert in the morning (OR = 0.40; 95% CI: 0.16 to 1.01).

Regression Results

In examining potential effect modification, no significant differences were seen in effect by whether the child had sleep problems at baseline (P = .13). Likewise, there was no significant effect modification detected by gender (P = .12), low-income household (P = .24), whether the child had a television in the bedroom at baseline (P = .25), high levels of evening media use at baseline (P = .27), or by internalizing (P = .97) or externalizing (P = .76) behavior problems. There was a trend (P = .096), however, toward an increased effect among those with high levels of violence exposure at baseline (daily mean ≥30 minutes) compared with those with less violent content, with OR = 0.23 (95% CI: 0.09 to 0.62) versus OR = 0.43 (95% CI: 0.19 to 1.00), respectively.


A healthy media use intervention in preschool-aged children, aimed at replacing violent content with age-appropriate, educational, and prosocial content, led to improved sleep in the intervention group as compared with the control group. The lack of significant effect modification by baseline sleep problems suggests that health media choices may help treat existing behavioral sleep problems and be a useful preventive measure. The observed trend toward effect modification by baseline levels of violent media exposure, coupled with the fact that the intervention targeted content rather than quantity and had no effect on mean daily usage, suggests that the intervention effects at reducing violent media use may have been the initial mechanism by which the intervention improved child sleep outcomes. Although effect sizes were similar at both 6 and 12 months after the initial intervention encounter, there was a trend toward a decrease in effect by the 18-month follow-up. Given that the intervention itself ended at 12 months, the potential decay at 18 months suggests that families may need supportive maintenance after the active intervention or that the intervention protocol may need to be revised to ensure that families are mastering the skills needed to continue making healthy media choices as their child continues to grow older and media options evolve.

Although an intervention to change the child’s media diet positively impacted sleep, it remains unknown how this relationship is mediated; the fact that the intervention targeted content choices rather than quantity of use, and in fact had no impact on total minutes of daily use, suggests that content indeed plays a key role. In our previous article describing the relationship between media use and child sleep at baseline in this study, we found not only that evening media use was associated with increased sleep problems but so was violent media use earlier in the day.2 If the intervention affected child sleep by reducing violent media consumption across the day, the effects may have been mediated by decreases in factors such as fears and state anxiety,20 arousal state at bedtime, or hyperactivity.

This study has a number of limitations that warrant mention. First, the sleep measure used was only a brief excerpt from a validated measure, because the study was designed primarily to examine outcomes of the healthy media use intervention around aggression and prosocial behaviors. The use of such a rough measure increases the chances of misclassification of child sleep problems, with both false-positive and false-negative results; however, we would expect this misclassification to be nonselective in nature, and to therefore bias our findings toward the null. We hope that future studies will be able to replicate these findings by using more robust measures of child sleep, such as sleep diaries or actigraphy. Second, it is not possible to blind the subjects of behavioral intervention studies; however, the study was presented to the parents as targeting media and aggression, not children’s sleep, so there is no reason to expect that social desirability or response bias would have led to differences in the sleep measure between study arms beyond that due to the intervention alone. Third, we drew our sample from clinics in a single urban area; however, because the children were eligible for inclusion regardless of whether they had been seen in the clinic over the past year, the study more closely approximates a population-based sample than a clinical sample. Still, the extent to which the results can be generalized to children from other communities is unknown. Finally, we did not have enough case managers to examine the degree to which interventionist characteristics may have impacted the effects observed in their panels and may not have had sufficient sample sizes of some subpopulations (such as low-income children or those with a television in the bedroom at baseline) to adequately test for effect modification by these factors. Perhaps future studies will be able to selectively target and oversample these populations and explore the potential for effect modification.

Despite these limitations, the findings have 2 important implications. First, these results in the context of a randomized controlled trial suggest that the previously reported relationship between media use and child sleep problems is indeed causal in nature. Second, clinicians and parents should be mindful that healthy media use choices could be a valuable strategy in treating and preventing child sleep problems. Given that early childhood sleep problems have been associated with a range of deleterious outcomes, both acute and long-term, including increased injuries, behavioral and emotional problems, difficulties in school, and obesity, the availability of useful, feasible strategies is critical.


confidence interval
odds ratio
Social Competence and Behavior Evaluation


This trial has been registered at (identifier NCT01459835).

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

FUNDING: This work was supported by grant 5R01HD056506 from the National Institute of Child Health & Human Development to Dr Christakis. Funded by the National Institutes of Health (NIH).


1. Owens JA, Maxim R, McGuinn M, Nobile C, Msall M, Alario A.. Television-viewing habits and sleep disturbance in school children. Pediatrics. 1999;104(3). Available at: [PubMed]
2. Garrison MM, Liekweg K, Christakis DA.. Media use and child sleep: the impact of content, timing, and environment. Pediatrics. 2011;128(1):29–35 [PMC free article] [PubMed]
3. Li S, Jin X, Wu S, Jiang F, Yan C, Shen X.. The impact of media use on sleep patterns and sleep disorders among school-aged children in China. Sleep. 2007;30(3):361–367 [PubMed]
4. Nixon GM, Thompson JM, Han DY, et al. . Short sleep duration in middle childhood: risk factors and consequences. Sleep. 2008;31(1):71–78 [PubMed]
5. Oka Y, Suzuki S, Inoue Y.. Bedtime activities, sleep environment, and sleep/wake patterns of Japanese elementary school children. Behav Sleep Med. 2008;6(4):220–233 [PubMed]
6. Paavonen EJ, Pennonen M, Roine M, Valkonen S, Lahikainen AR.. TV exposure associated with sleep disturbances in 5- to 6-year-old children. J Sleep Res. 2006;15(2):154–161 [PubMed]
7. Thompson DA, Christakis DA.. The association between television viewing and irregular sleep schedules among children less than 3 years of age. Pediatrics. 2005;116(4):851–856 [PubMed]
8. Mistry KB, Minkovitz CS, Strobino DM, Borzekowski DL.. Children’s television exposure and behavioral and social outcomes at 5.5 years: does timing of exposure matter? Pediatrics. 2007;120(4):762–769 [PubMed]
9. Mindell JA, Meltzer LJ, Carskadon MA, Chervin RD. Developmental aspects of sleep hygiene: findings from the 2004 National Sleep Foundation Sleep in America Poll. Sleep Med. 2009;10(7):771–779 [PubMed]
10. Alexandru G, Michikazu S, Shimako H, et al. . Epidemiological aspects of self-reported sleep onset latency in Japanese junior high school children. J Sleep Res. 2006;15(3):266–275 [PubMed]
11. Van den Bulck J.. Television viewing, computer game playing, and Internet use and self-reported time to bed and time out of bed in secondary-school children. Sleep. 2004;27(1):101–104 [PubMed]
12. Dworak M, Schierl T, Bruns T, Strüder HK.. Impact of singular excessive computer game and television exposure on sleep patterns and memory performance of school-aged children. Pediatrics. 2007;120(5):978–985 [PubMed]
13. Eggermont S, Van den Bulck J.. Nodding off or switching off? The use of popular media as a sleep aid in secondary-school children. J Paediatr Child Health. 2006;42(7-8):428–433 [PubMed]
14. Mesquita G, Reimão R.. Nightly use of computer by adolescents: its effect on quality of sleep. Arq Neuropsiquiatr. 2007;65(2B):428–432 [PubMed]
15. Johnson JG, Cohen P, Kasen S, First MB, Brook JS.. Association between television viewing and sleep problems during adolescence and early adulthood. Arch Pediatr Adolesc Med. 2004;158(6):562–568 [PubMed]
16. Christakis DA, Zimmerman FJ, DiGiuseppe DL, McCarty CA.. Early television exposure and subsequent attentional problems in children. Pediatrics. 2004;113(4):708–713 [PubMed]
17. Zimmerman FJ, Christakis DA.. Associations between content types of early media exposure and subsequent attentional problems. Pediatrics. 2007;120(5):986–992 [PubMed]
18. Zimmerman FJ, Glew GM, Christakis DA, Katon W.. Early cognitive stimulation, emotional support, and television watching as predictors of subsequent bullying among grade-school children. Arch Pediatr Adolesc Med. 2005;159(4):384–388 [PubMed]
19. Christakis DA, Zimmerman FJ.. Violent television viewing during preschool is associated with antisocial behavior during school age. Pediatrics. 2007;120(5):993–999 [PubMed]
20. Cantor J, Byrne S, Moyer-Gusé E, Riddle K. Descriptions of media-induced fright reactions in a sample of US elementary school children. Journal of Children and Media. 2010;4(1):1–17
21. Myaing MT, Garrison MM, Rivara FP, Christakis DA. Differences between opt-in and actively recruited participants in a research study. Journal of Clinical Medicine and Research. 2011;3(5):68–72
22. Christakis DA, Zimmerman FJ. The Elephant in the Living Room: Make Television Work for Your Kids. Emmaus, PA: Rodale; 2006
23. Bushman BJ, Huesmann LR.. Short-term and long-term effects of violent media on aggression in children and adults. Arch Pediatr Adolesc Med. 2006;160(4):348–352 [PubMed]
24. Anderson DR, Huston AC, Schmitt KL, Linebarger DL, Wright JC.. Early childhood television viewing and adolescent behavior: the recontact study. Monogr Soc Res Child Dev. 2001;66(1):I–VIII, 1–147 [PubMed]
25. Robinson TN, Wilde ML, Navracruz LC, Haydel KF, Varady A.. Effects of reducing children’s television and video game use on aggressive behavior: a randomized controlled trial. Arch Pediatr Adolesc Med. 2001;155(1):17–23 [PubMed]
26. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall; 1986
27. Baranowski T. How individuals, environments, and health behavior interact In: Glanz K, Rimer BK, Lewis FM, eds. Health Behavior and Health Education: Theory, Research, and Practice. 3rd ed. San Francisco, CA: Jossey-Bass;2002:165–184
28. Sharma M, Wagner DI, Wilkerson J.. Predicting childhood obesity prevention behaviors using social cognitive theory. Int Q Community Health Educ. 2005-2006;24(3):191–203 [PubMed]
29. Story M, Neumark-Sztainer D, French S.. Individual and environmental influences on adolescent eating behaviors. J Am Diet Assoc. 2002;102(suppl 3):S40–S51 [PubMed]
30. Mendelsohn AL, Brockmeyer CA, Dreyer BP, Fierman AH, Berkule-Silberman SB, Tomopoulos S.. Do verbal interactions with infants during electronic media exposure mitigate adverse impacts on their language development as toddlers? Infant Child Dev. 2010;19(6):577–593 [PMC free article] [PubMed]
31. Collins WA, Sobol BL, Westby S.. Effects of adult commentary on children’s comprehension and inferences about a televised aggressive portrayal. Child Dev. 1981;52(1):158–163 [PubMed]
32. Paavonen EJ, Roine M, Pennonen M, Lahikainen AR.. Do parental co-viewing and discussions mitigate TV-induced fears in young children? Child Care Health Dev. 2009;35(6):773–780 [PubMed]
33. Friedrich LK, Stein AH.. Aggressive and prosocial television programs and the natural behavior of preschool children. Monogr Soc Res Child Dev. 1973;38(4):1–64 [PubMed]
34. Friedrich LK, Stein AH. Prosocial television and young children: the effects of verbal labeling and role playing on learning and behavior. Child Dev. 1975;46:27–38
35. Friedrich-Cofer LK, Huston-Stein A, Kipnis DM, Susman EJ, Clewett AS. Environmental enhancement of prosocial television content: effects on interpersonal behavior, imaginative play, and self-regulation in a natural setting. Dev Psychol. 1979;15(6):637–646
36. Mares ML, Woodard E. Positive effects of television on children's social interactions: a meta-analysis. Media Psychol. 2005;7(3):301–322
37. Forge KLS, Phemister S. The effect of prosocial cartoons on preschool children. Child Study J. 1987;17(2):83–88
38. Silverman LT, Sprafkin JN. The effects of Sesame Street's prosocial spots on cooperative play between young children. Journal of Broadcasting. 1980;24:135–147
39. Collins WA, Getz SK.. Children’s social responses following modeled reactions to provocation: prosocial effects of a television drama. J Pers. 1976;44(3):488–500 [PubMed]
40. Rice ML, Huston AC, Truglio R, Wright JA. Words from “Sesame Street”: learning vocabulary while viewing. Dev Psychol. 1990;26(3):421–438
41. Wright JC, Huston AC, Murphy KC, et al. . The relations of early television viewing to school readiness and vocabulary of children from low-income families: the early window project. Child Dev. 2001;72(5):1347–1366 [PubMed]
42. Linebarger DL, Walker D. Infants' and toddlers' television viewing and language outcomes. Am Behav Sci. 2005;48(5):624–645
43. Goodlin-Jones BL, Sitnick SL, Tang K, Liu J, Anders TF.. The Children’s Sleep Habits Questionnaire in toddlers and preschool children. J Dev Behav Pediatr. 2008;29(2):82–88 [PubMed]
44. Owens JA, Spirito A, McGuinn M.. The Children’s Sleep Habits Questionnaire (CSHQ): psychometric properties of a survey instrument for school-aged children. Sleep. 2000;23(8):1043–1051 [PubMed]
45. Kotler JC, McMahon RJ.. Differentiating anxious, aggressive, and socially competent preschool children: validation of the Social Competence and Behavior Evaluation-30 (parent version). Behav Res Ther. 2002;40(8):947–959 [PubMed]
46. Perneger TV, Burnand B.. A simple imputation algorithm reduced missing data in SF-12 health surveys. J Clin Epidemiol. 2005;58(2):142–149 [PubMed]
47. Peyre H, Leplège A, Coste J.. Missing data methods for dealing with missing items in quality of life questionnaires: a comparison by simulation of personal mean score, full information maximum likelihood, multiple imputation, and hot deck techniques applied to the SF-36 in the French 2003 decennial health survey. Qual Life Res. 2011;20(2):287–300 [PubMed]
48. Shrive FM, Stuart H, Quan H, Ghali WA.. Dealing with missing data in a multi-question depression scale: a comparison of imputation methods. BMC Med Res Methodol. 2006;6(Dec 13):57. [PMC free article] [PubMed]

Articles from Pediatrics are provided here courtesy of American Academy of Pediatrics