AMetaA Meta-Analysis on Father Involvement andAnalysis on ... · Extensive peer interaction ......
Transcript of AMetaA Meta-Analysis on Father Involvement andAnalysis on ... · Extensive peer interaction ......
A MetaA Meta Analysis on Father Involvement andAnalysis on Father Involvement andA MetaA Meta--Analysis on Father Involvement and Analysis on Father Involvement and Early Childhood SocialEarly Childhood Social--Emotional DevelopmentEmotional Development
Robby HarrisDepartment of Applied PsychologyDepartment of Applied Psychology
New York University
A k l d tA k l d tAcknowledgementsAcknowledgements
Many thanks to Dr. Christy McWayne, Dr. Rodrigo Campos, and the entire McWayne Research Lab
Thanks to Dr. Jason Downer, University of Virginia
Development in Early Childhood andDevelopment in Early Childhood andDevelopment in Early Childhood and Development in Early Childhood and SocialSocial--Emotional AbilitiesEmotional Abilities
Transition to formal school environment Leave home for extended time, possibly for the first time Much less individual attention Extensive peer interaction
School readiness is a measure of a child’s likelihood to School readiness is a measure of a child s likelihood to succeed in this new setting Cognitive ability
Reading skills Language development Motor/physical development Reading skills
Math skills Motor/physical development Social-emotional abilities
Importance of developing social-emotional abilities to be successful in the transition (Raver & Zigler, 1997)
Importance of Parent InvolvementImportance of Parent InvolvementImportance of Parent InvolvementImportance of Parent Involvement
Parent involvement facilitates this transition Parent involvement facilitates this transition Academic achievement (Flouri & Buchanan, 2004) Social-emotional competence (Overbeek et al., 2007)p ( , )
Who is included in this research?M h i l d d f h Mothers are included, fathers are not.
What about the dads? Mothers are more likely to be the child’s primary caregiver. Mothers are more likely to be the child s primary caregiver.
Fathers are important.p
F th I l t i E l Childh dF th I l t i E l Childh dFather Involvement in Early ChildhoodFather Involvement in Early Childhood
Fathers help their preschool children develop: self-esteem
social competence social competence empathetic skills emotion regulation positive self-concept
Unique contributions, as distinct from mothers Unique contributions, as distinct from mothers
Discrepancies in the literature on how father involvement helps children develop these abilities
Influences on Father Involvement andInfluences on Father Involvement andInfluences on Father Involvement and Influences on Father Involvement and SocialSocial--Emotional DevelopmentEmotional Development
YES NO
Father involvement, regardless of SES
The influence of father involvement
YES NO
SES regardless of SES, helps children to develop the same social emotional
father involvement on developing social-emotional abilities in children
SES
Race/Ethnicity social-emotional abilities.
abilities in children differs according to the SES of the f the
Race/Ethnicity
Residential Status father.Residential Status
Addressing Conflicting FindingsAddressing Conflicting Findingsg g gg g g
Conducting a meta-analysis helps us to make sense of the literature. A meta-analysis examines every study that has been
conducted on a particular topic to assess the relationship between variables.
An index of the strength of the relationship, called an effect size, is calculated for each study.
Data are integrated to obtain a general estimate of theData are integrated to obtain a general estimate of the strength of the relationship between variables (Leary, 2008).
A meta-analysis is a study of studies.
Inclusion of moderating variables: SES Race/Ethnicity Father Residential Status
Research QuestionsResearch QuestionsResearch QuestionsResearch Questions
What is the relationship between father involvement and hild ’ i l ti l biliti i l hildh d?children’s social-emotional abilities in early childhood?
Does the relationship vary as a function of SES, / th i it f th id ti l t t ?race/ethnicity, or father residential status?
Social-Emotional Outcomes
Father Involvement
OutcomesSES
Race/Ethnicity/ y
Residential Status
MethodsMethods 3,775 titles collected.
Data collectionArticles of basic relevance to father involvement or school
di Keywords
e.g., father, early childhood, father involvement, preschool Sample size for this project
13 ti l
readiness, 2,750 articles remain.
Databases e.g., PsychINFO, ERIC, IBBS
Inclusion criteria
13 articles 73 effect sizes 817 total participants
Articles fit into specific inclusion criteria,
125 articles remain.
Relevance (FI or SR) Early childhood (ages 3-8) Published in years 1998-2008
p p
Relevance to early childhood school readiness outcomes,
Empirical (data collection) Quantitative Outcomes of FI, not determinants
30 articles remain.
Relevance to social-emotionalRelevance to social-emotional development,
13 articles remain.
S l Ch t i tiS l Ch t i tiSample CharacteristicsSample Characteristics
Dimensions of Father Involvement Dimensions of Father Involvement Active Father Involvement
Engagement/activity Communication Attachment Caregiving
Passive Father Involvement Presence Financial assistance
Dimensions of Social-Emotional Outcomes Positive: social skills emotion regulation Positive: social skills, emotion regulation Negative: behavior problems
Example of Included StudyExample of Included Study
Father Involvement Social-EmotionalFather Involvement Measures
Social-Emotional Measures
Activities with Your Child Temperament Assessment pBattery
Family Involvement Penn Interactive Peer Play ScaleFamily Involvement Questionnaire
Penn Interactive Peer Play Scale
D hi Q ti i E ti R l ti Ch kli tDemographic Questionnaire Emotion Regulation Checklist
9 total effect sizes from this study
(Downer & Mendez, 2005)
73 total effect sizes to be included in this meta-analysis
D t A l iD t A l iData AnalysisData Analysis
Calculated effect sizes within each study Calculated effect sizes within each study
Calculated mean effect sizes across each dimension of FI and S/E outcomes Random-effects model: we assume that included studies do not
embody “identical” true effect sizes (Borenstein, 2009).h d h l d d More weight was given to studies that included more variance
and larger sample sizes
Tests for heterogeneity and publication bias
Effects of moderating variables Effects of moderating variables
Summary of Effect Size ResultsSummary of Effect Size ResultsSummary of Effect Size ResultsSummary of Effect Size Results
Father Involvement
Active Involvement
Passive Involvement
Summary MeasuresOutcome
Summary Effect Size# of effect sizes
p-valuePositive
.22310 001
N/AN/AN/Ap value
Summary Effect Size# of effect sizes
lNegative
-.13320 069
-.14100 089
0.001 N/A
p-value 0.069 0.089
There was a significant effect between active fatherThere was a significant effect between active father involvement and positive social-emotional outcomes.
Active Father Involvement and Active Father Involvement and ct e at e o e e t a dct e at e o e e t a dPositive OutcomesPositive Outcomes
Correlation and 95% CICorrelation and 95% CISignificant summary
ffMean effectsize for each
study included
effect size of .22
study included
Summary effect size for
ll t di-1.00 -0.50 0.00 0.50 1.00-1.00 -0.50 0.00 0.50 1.00all studiesIndicative of no effect
Tests for Heterogeneity & Publication BiasTests for Heterogeneity & Publication BiasTests for Heterogeneity & Publication BiasTests for Heterogeneity & Publication Bias
Heterogeneity: how much of the variation is real? Heterogeneity: how much of the variation is real? Q statistic = 32.55, p-value = 0.002
Because the Q statistic was significant, that means there was h t it i th b d ff theterogeneity in the observed effects.
I2 = 60% (of the observed variation was real) There was enough variation to warrant moderation analyses. There was enough variation to warrant moderation analyses.
Publication Bias: are the studies included in this analysis representative of all of the studies executed y pon this topic or merely of those being published? There was no evidence of publication bias.
Moderating VariablesModerating Variablesgg
Is the moderator significant in
Moderators Outcome
significant in explaining any
variability?
Low Positive No
SES
LowMid/High
Positive No
LowMid/High
Negative
R / Minority Negative
No
NoRace/ Ethnicity
MinorityNon-minority
Negative
Residential Nonresidential Negative
No
NoResidential Status
NonresidentialResidential
Negative No
Findings & DiscussionFindings & Discussiongg
There was a significant positive effect between active father involvement and the development of youngfather involvement and the development of young children’s positive social-emotional abilities. There were significant trends between father involvement g
and children’s social-emotional problems.
Neither SES, race/ethnicity, or residential status served as significant moderators of this relationshipserved as significant moderators of this relationship. The relationship between father involvement and children’s
social-emotional outcomes may be the same across all levels of the moderators.
However, caution should be used in interpreting these results because of the great deal of data missing from theresults because of the great deal of data missing from the moderator variable analyses.
Future DirectionsFuture Directions
Directionality of resultsDoes father involvement affect child outcomes does child Does father involvement affect child outcomes, does child behavior affect father involvement, or is it transactional?
Longitudinal data on father involvement and child outcomes will help to address these questions.
Missing data from moderator variable analysesi d f ll i It is necessary to study context more purposefully in
relation to father involvement and child outcomes.
Measurements of father involvement Measurements of father involvement Need for a more explicit conception of father involvement More measures that directly sample fathers
School readiness meta-analysis
Dads Make A DifferenceDads Make A Difference
Thank You!Thank You!