Fathers' involvement is crucial - a report to the ESRC

R000223309: Father involvement and outcomes in adolescence and adulthood
End of Award Report  24 October 2001

ESRC research provides evidence that the involvement of fathers in children's educational development is linked to academic success. The National Literacy Trust is grateful to the researchers of the University of Oxford and to the ESRC for their permission to display the summary of the research findings on this website. 

1. Summary of Research Results 

In recent years there has been considerable research from the US on positive outcomes for children whose fathers become 'involved' in their care. The purpose of this study was to ascertain whether there was similar evidence in the UK using longitudinal data from the National Child Development Study (NCDS). NCDS is an ongoing longitudinal study of some 17,000 children who were born in England, Wales and Scotland in one week in 1958. These children have been followed up at frequent intervals during their childhood and into adult life. In NCDS we were able to explore fathers' involvement with their children when the children were aged 7, 11, and 16 years of age.  

An 'involved' father, as defined in this research, is a father who reads to his child, takes outings with his child, is interested in his child's education and takes a role equal to mother's in managing his child. He may or may not live with the child's mother, and he may or may not be the biological father to the child. Below is a summary of our main findings. 
 

• Early father involvement with a child is associated with continuing involvement with that child throughout childhood and adolescence. Generally once fathers are 'involved' they remain involved with that child throughout childhood. At different ages, fathers related to their children in different ways, but the underlying concept of father involvement is a continuous one. At age 7, fathers are more likely to be 'involved' where mothers are also involved and when the child does not have emotional and behavioural problems, but the practice is less strong amongst manual workers. With older children father involvement is inversely related to family size and poor school performance. Although financial difficulties in the family are not related to whether father is involved at ages 7 or 11, with younger children, fathers are less likely to be involved where there is domestic tension in the home. Generally, the higher the level of a father's education, the greater is the likelihood of his involvement with his children. Maternal employment is associated with less paternal interest in education but only in younger children.
• Father involvement is associated with good parent-child relationships in adolescence and also with later satisfactory partnerships in adult life. Good father-child relations are associated with an absence of emotional and behavioural difficulties in adolescence and greater academic motivation. Teenagers feel close to their fathers when their fathers are involved with them as they were growing up. However, feelings of closeness to mother in adolescence are unrelated to the mother's early level of involvement with the child.  When the children grow up those who have felt close to their fathers in adolescence are more likely to have satisfactory adult marital relationships. 
• Children with involved fathers are less likely to be in trouble with the police. Boys in particular are less likely to be in trouble with the police where they have 'involved' fathers. This relationship persisted even when we took into account several factors which are associated with adolescent delinquency.
• Father involvement protects children in separated families against later mental health problems. Involvement of the father or a father figure has a significantly protective role against psychological problems in adolescents in families where parents have separated. This finding is independent of whether mothers are also involved. The association between father involvement in adolescence and psychological distress in adult life is stronger for daughters than for sons. Therefore, early father involvement has an important protective role against both later psychological maladjustment in children where parents have separated, and against adult psychological distress in women.
• Father involvement is strongly related to children's later educational attainment. Father involvement at age 7 and mother involvement at age 7 significantly and independently predict higher educational attainment by age 20 and this applies to both girls and boys. 
• Early father involvement protects against an adult experience of homelessness in sons of manual workers.  As might be expected absence of partners in women, and psychological distress and large families in men are the key predictors of welfare dependency. For both men and women low educational attainment predicts whether they would be living in council housing and whether they are receiving state benefits.  However, both father and mother involvement in families of manual workers is highly protective against an adult experience of homelessness in their sons.
• Early father involvement does not protect against unemployment in adult children when other factors are taken into account. However, men who had involved mothers when they were growing up were less likely to be unemployed than men with uninvolved mothers.

Background 

 Although there has been increased interest in the study of father-child relationships, fathers are still dramatically underrepresented in the research literature on the effects of parental involvement and influences on adolescent and child development, particularly in the UK.  Phares and Compas (1992) analysed child and adolescent research in eight journals from 1984 to 1991 and found that nearly one half of the studies dealt only with mothers and 1 per cent of the studies exclusively involved fathers.  

This lack of research is especially unfortunate given that there are several reasons why higher levels of paternal involvement should positively influence children's psychological and social development. First, fathers who are involved with their children are likely to engage their children in physical play and parent-child play, especially the parent's ability to be responsive to the child's initiative allowing for a nurturant give-and-take in their play is particularly important for the child's emotional and social development (Biller & Kimpton, 1997). Second, in families where fathers are involved, mothers are also involved (Amato, 1994), and therefore children raised in such families benefit from having two highly involved parents with the consequent diversity of stimulation and increase in social capital (Coleman, 1988). Third, fathers are more likely to be involved when the co-parental relation is good (Coiro & Emery, 1998; NICHD Early Child Care Research Network, 2000), and therefore in families where the father is involved the overall family context in which children are raised is positive, which is in an important factor contributing to positive child outcomes (Kelly, 2000). Finally, involved fathers are likely to financially support their children. Given the adverse child outcomes associated with poverty, fathers' involvement can indirectly influence their children by influencing the economic structure of the household (Cabrera, Tamis-LeMonda, Bradley, Hofferth & Lamb, 2000).  
 

Outcomes studies: father involvement and closeness 
Some of the findings in studies exploring father involvement effects have been impressive. For instance, Barnett, Marshall and Pleck (1992), showed that sons who reported a positive relationship with their mother or father had relatively low levels of psychological distress. In fact, only the father-child relationship not the mother-child relationship was significantly related to son's distress. More recently, Amato (1994) showed that regardless of the quality of the mother-child relationship, the closer children were to their father, the happier, more satisfied and less distressed they reported being. And whereas Barnett et al. (1992) demonstrated this effect only for sons, Amato (1994) showed that it holds true for both sons and daughters. More recent research has shown that children with involved fathers tend to be more psychologically well adjusted (Flouri & Buchanan, in press; Gould, Shaffer, Fisher & Garfinkel 1997; Sanford et al., 1995), to do better at school, and to engage in less antisocial behavior (Flouri & Buchanan, forthcoming). Father involvement and nurturance are also positively associated with childrenís intellectual development, social competence, internal locus of control and the ability to empathize (Fagan & Iglesias, 1999; Yongman, Kindlon, & Earls, 1995). Studies of adults also show supporting evidence.  
'Closeness' to fathers during childhood has been found to be positively related to adult daughters' and sons' educational and occupational mobility (Amato, 1994).  

Father absence 
Other research has explored the associations with father absence. Studies of father absence consistently suggest that father absence is a factor contributing to the lower educational attainment of children in mother-only families. Children in mother-only families, for example, score lower than other children on measures of academic achievement and cognitive ability (e.g. Mulkey, Crain & Harrington, 1992), and are more likely to drop out of school (Sandefur & Wells, 1999).  

The peripheral role of fathers 
On the other hand another clusters of studies, suggest that fathers play peripheral roles in their children's lives. Crockett, Eggebeen and Hawkins (1993) found that father absence has few consequences for children once economic factors are controlled, and King (1994) also found very limited evidence to support a link between 'visitation' by non-resident fathers and children's well-being. Although other studies have shown that even when these factors are adjusted father absence continues to be associated with an increased risk of child problems (Amato, 1993), there is sufficient disagreement to cast some doubt on fathers' non-economic contributions to children. Finally, objections have been raised as to whether fathers make an independent contribution above and beyond that of mothers. It is possible that families in which fathers are highly involved are also those in which mothers are involved, which means that the extra attention of fathers may be largely redundant once the mother's involvement is taken into account.  

Problems in fathering research 

Apart from this inconsistency in findings there is another serious problems in father involvement research. Most of the research carried out so far on father involvement is based on cross-sectional (correlational) data and therefore causality cannot be determined.  

Using longitudinal data from the National Child Development Study (NCDS), a longitudinal study of 17,000 children born in England, Scotland and Wales between 3-9 March 1958, we carried out this study to a) address this inconsistency and, b) attempt to determine causality, providing, at the same time, a British perspective in an area dominated by American research.  

Objectives 
According to our submitted proposal to the ESRC, our study aimed to address the following three questions:  

1) What are the relationships, while controlling for other factors, between father involvement and outcomes for children?  
2) What are the processes whereby, and stages when, in the childís development 'involved' fathering exerts its beneficial influence?  
3) What are the factors that appear to keep fathers 'involved' and factors that place this relationship at risk?  

To answer the first question, we proposed to look at the relationship between father involvement and the following child outcomes at age 16 and age 33:  
Adolescence: psychological adjustment (Rutter); academic performance (exam successes); personal relationships (self-reported relationships with parents etc); welfare and police involvement (child/family involvement with social services and/or police). 
Adulthood: psychological adjustment (Malaise Inventory); adult work history (periods of unemployment/employment); personal relationships (Locke-Wallace Marital Adjustment Scale if partnered); welfare and police involvement. We believe we have successfully addressed these objectives in the 8 papers we have written for peer-reviewed journals.  

In particular (see Tables in pp. 14-15), we have explored the relationship between father involvement and mental health outcomes at age 16 and 33 (paper #3), academic performance (paper #1), personal relationships at age 16 and 33 (paper #5), police involvement at age 16 (paper #6), labour force participation at age 33 (paper #2), as well as the relationship between father involvement and the following 3 indicators of social and economic disadvantage at age 33: subsidized housing, experience of homelessness and state benefits receipt (paper #4). In all these articles we also explored the processes whereby father involvement is associated with child outcomes, thus addressing our second aim.  

Finally, our third aim, the investigation of the determinants of father involvement, was fully addressed in paper #7 which explored factors associated with father's involvement at different stages of the child's life. 

Methods 
We addressed our research objectives by analysing data from different sweeps of the NCDS using SPSS. We mainly carried out multiple regression and logistic regression analyses. We also carried out non-response bias analyses, and we checked for moderator and mediator effects.  

Results 
With regard to the role of father involvement in later child outcomes, our research showed that: 
1. Father involvement in childhood is associated with both good father-child relations in adolescence and later marital satisfaction in adult life, even after controlling for mother involvement, mother-child relations and known confounds.  
2. Father involvement in childhood was negatively associated with adolescent delinquency in boys, even after controlling for mother involvement and known confounds. 
3. Father involvement in childhood protected both against psychological maladjustment in adolescents in non-intact families, and against psychological distress in women in adult life, even after controlling for earlier psychological problems, mother involvement and known confounds. 
4. Father involvement in childhood was strongly related with later educational attainment even after controlling for mother involvement and known confounds.  
5. Father involvement in childhood was not independently related to use of state benefits receipt and subsidized housing in adult life when controlling for known confounds and mother involvement in childhood. However, father involvement in childhood was negatively related to an adult experience of homelessness in sons of manual workers.  
6. Father involvement in childhood was not independently related to labour force participation in adult life when controlling for known confounds and mother involvement in childhood. Interestingly, however, men who had involved mothers when they were growing up were less likely to be unemployed than men with uninvolved mothers. 

Finally, with regard to the factors associated with father involvement our research showed that father involvement was predicted from different factors at different developmental stages but generally it was continuous and multidimensional, and strongly associated with mother involvement. Low parental socio-economic status and child emotional and behaviour problems were negatively related to father involvement at age 7. With older children father involvement was inversely related to family size and poor school performance. Financial difficulties in the family were not related to father involvement at either age. Domestic tension was negatively related to certain aspects of fathers' involvement with younger children. Father's education was generally related to father's involvement but maternal employment was only related to low father's interest in child's education at age 7 and 11.  

An important rider to this research, however, is that father involvement in situations of high family conflict may not be in children's best interests.  Related research by Ann Buchanan et al., suggests that children are vulnerable to mental health problems where parents, post separation or divorce involve the court because they cannot agree the arrangements for their children (Buchanan, Hunt, Bretherton and Bream 2001)  

Activities 
Dr Flouri gave a paper at the XXVI International Colloquium on Research in Economic psychology, a talk at the Centre for Research into parenting at Oxford, and submitted an abstract of a paper to be presented at the XXV International Congress of Applied Psychology in Singapore in 2002.  She has also attended a seminar on fathering organized by the Family Policy Studies Centre.  Dr Buchanan has played a secondary role in this research supporting Dr Flouri but has endeavoured to make opportunities for Dr Flouri to disseminate the research. 

Outputs 
We have submitted 8 papers in peer-reviewed journals, and we are currently negotiating a possible book on Fathering with John Wiley and Sons. In addition, we published an article in Community Care summarising our main research findings in non-technical language. A conference paper has been published in the Proceedings of an international conference and another one is going to be published in July 2002.  We were approached by the Sunday Times at the end of October 2001 who had seen the write-up in REGARD. They were interested in publishing an exclusive article on our findings, and we contacted the ESRC to check this out. In the event the article did not appear due to the heavy coverage on the war. A small article will appear in the Oxford University Newsletter BLUEPRINT in November. 

Impacts 
 Our experience with both Community Care and the Sunday Times suggests that we should anticipate considerable interest in this project when the End of Award Report is processed.  A press release will be prepared at this time. 
The book we are  proposing to John Wiley and Sons will be looking at children's social, educational and economic outcomes as well as mental health outcomes associated with involved fathering. As such, it will be of interest not only to anyone involved in developmental or clinical psychology or child custody but to a wide range of academics and practitioners as well. This book will consolidate most of the research we carried out under this project exploring both the determinants of and outcomes from involved fathering. Our research on the determinants of involved fathering will be of interest to all those who in one capacity or another are concerned with human development, parenting and socialisation. The section on outcomes from involved fathering will be of interest to an even wider audience because the outcomes we looked at cover such diverse areas. Below we outline to whom we think the outcomes from involved fathering will be of particular interest: 
1. Mental health in adolescence and in adult life (of interest to clinical psychologists, developmental psychologists, psychiatrists, social workers, psychology researchers) 
2. Educational outcomes (of interest to educational psychologists, teachers, sociologists, social scientists) 
3. Adolescent delinquency (of interest to probation officers, youth workers, psychologists, criminologists)  
4. Quality of relationships with parents in adolescence and partners in adult life (of interest to family therapists, family psychologists, social psychologists) 
5. Social and economic disadvantage (unemployment, homelessness, state benefits receipt, subsidised housing) in adult life (of interest to sociologists, economic psychologists and economists). 
 

Policy.  

As can be seen from the above the research has important policy implications for a number of government departments.  Firstly all government departments involved with the family need to consider that if fathers are to become involved with their children, they need the opportunity early in a child's life to spend quality life with their offspring. This has implications for the work-life balance and paternal leave and family policy. Other findings have relevance to particularly departments, for example Department of Health (mental health and adult relationship outcomes); Department of Education and Skills (educational outcomes); Home Office (findings on delinquency).  

Future Research Priorities 

A factor that is missing in this research, as mentioned earlier, is whether father involvement is to be advocated in situations of high conflict when parents post separation or divorce cannot agree the arrangements for their children. Apart from this, our research showed that father involvement is related with several positive child outcomes in the long term.  

Future funding would help us consolidate our journal submissions and write our book. At the same time, we acknowledge that NCDS is dated and therefore one should exercise caution in generalizing findings from these data. Future funding could help us compare the findings from this study with those of our next research project which is looking at Outcomes of, and determinants of, father involvement in contemporary Britain. Related to this, future funding could be useful in exploring determinants of and outcomes from father involvement in the more recent 1970 British birth cohort, and comparing these with the findings from this research.