Young Adult South African Daughters ’ Perceptions of Paternal Involvement and Nurturance

This study aimed to assess current and retrospective levels of reported and desired paternal involvement experienced by young adult daughters, as well as current and retrospective levels of paternal nurturance. A sample of 89, female, third year South African Psychology students completed self-administered questionnaires, consisting of a biographical questionnaire, four Father Involvement Scales and two Nurturant Father Scales. Daughters reported their fathers as having been involved and nurturing while growing up. Although they indicated that they perceived fathers as somewhat less involved in young adulthood; they reported being satisfied with the level of father involvement. Daughters also reported high current paternal nurturance. The findings therefore indicate that a group of middle to upper middle-class South African daughters perceived their fathers as relatively involved in their lives and suggest that their fathers’ involvement extends beyond traditional father roles.

Overall, South African fatherhood research is very limited and Black i low-income fathers have received the bulk of research attention.Few studies have been conducted on White, Coloured and Indian fathers.No published work on middle to higher income Black fathers, or research comparing fatherhood across cultures or socio-economic status, could be located.Also, no published studies on adult daughters and fathers could be located.There is thus a need for fathering research that incorporates the diversity of fathers in South Africa (Morrell, Posel, & Devey, 2003), and specifically focuses on father-daughter relationships.This study aimed to address this gap by exploring a group of adult daughters' perceptions of past and current father involvement.

Literature Review
Traditional parental roles typecast the father as the breadwinner and the mother as the carer or child raiser (LaRossa, as cited in Cabrera et al., 2000).As a result, maternal involvement was uncritically accepted in the past as the more important influence on child development (Cabrera et al., 2000).The preceding four decades have seen a change in the fathering role (Lamb, 2000) and has brought about an interest in the influence of paternal involvement on child development (Amato, 1994;Finley et al., 2008;Harris et al., 1998).An accumulating body of research now supports the idea that paternal involvement is important and makes a unique contribution to the development of their children (Goncy & Van Dulmen, 2010;van Schaick & Stolberg, 2001).
Recent literature has brought to light that father involvement continues to be important even after children reach adolescence (Mitchell, Booth, & King, 2009).Paternal involvement in adolescence is associated with higher levels of well-being in adolescents (Cookston & Finlay, 2006;Mitchell et al., 2009).Flouri and Buchanan (2003) found the relationship between paternal involvement and adolescent well-being statistically significant, and the findings of other studies found that reported father involvement is associated with less behavioural problems during adolescence (Amato & Rivera, 1999;Harris et al., 1998;Mitchell et al., 2009).Interestingly, Flouri and Buchanan (2003) found that paternal involvement had a stronger effect than maternal involvement on adolescent happiness and was equally predictive for sons' and daughters' reported levels of happiness.Furthermore, closeness to fathers during childhood and adolescence is also related to adjustment and well-being in adulthood (Amato, 1994).Schwartz and Finley (2006) found that this holds true even if the paternal figure is not the biological father.
Converging evidence concerning the importance of father involvement in childhood and adolescence can be found in literature regarding the negative effects of father absence.Father absence has been linked to lower engagement and performance in academic achievement, increased delinquent behaviour and riskier sexual practices in adolescents (Cabrera et al., 2000;Ellis et al., 2003).Studies that have been done in the South African context support the idea that paternal absence is negatively associated with educational outcomes (Mboya & Nesengani, 1999;Timaeus & Boler, 2007).Father involvement is linked to specific positive health outcomes in adolescent daughters regarding self-esteem, perceived body image, alcohol and drug use, and depression (Bayens, 2007;Cookston & Finlay, 2006;Goncy & Van Dulmen, 2010;Videon, 2002).A South African study found that living with a biological father could act as a protective factor against teenage pregnancy (Vundule, Maforah, Jewkes, & Jordaan, 2001).This is of particular relevance, as teenage pregnancy is associated with serious health problems in South Africa (Vundule et al., 2001).Phares et al. (2009) found that fathers are less involved with adolescent daughters than are mothers.Residential fathers are more involved with sons than with daughters (Harris et al., 1998).There is, however, some disagreement if this is also true for non-residential fathers (Mitchell et al., 2009).Non-residential fathers are on the increase (Cabrera et al., 2000) and non-residential fathers are less involved than residential fathers (Schwartz & Finley, 2005).Father involvement usually decreases with divorce (Agar et al., 2010;Schwartz & Finley, 2005;van Schaick & Stolberg, 2001) and marital conflict (Harris et al., 1998).
Paternal involvement can differ according to socio-economic group.Family values differ between cultures and is often influenced by socio-economic status (SES) (Spjeldnaes, Moland, Harris, & Sam, 2011).Difficulties in fathering may also be uniquely influenced by SES. Spjeldnaes et al. (2011), for example, highlight how absent fathers are prevalent in low SES families due to migratory labour, imprisonment and death.
South African fatherhood research indicates that fathers are important in the well-being of children (Richter, Chikovore, & Makusha, 2010;Spjeldnaes et al., 2011) and that the construct of fatherhood is evolving from the father as patriarchal breadwinner to the new involved caring co-parent (Hinckley, Ferreira, & Maree, 2007;Mkhize, 2006;Morrell et al., 2003;Roy, 2008) Furthermore, there is agreement among fatherhood researchers that fathers are an underutilised resource in terms of the contribution they could potentially make in the lives of their children (Desmond & Desmond, 2006;Montgomery, Hosegood, Busza, & Timaeus 2006;Morrell et al., 2003).
Many South African young men and fathers express that they want to be good fathers.They want to be involved in their children's lives, and they see fatherhood as extending beyond traditional roles to include themes of love, social support and guidance (Madhavan & Roy, 2012;Rabe, 2006;Spjeldnaes et al., 2011).However, black lowincome fathers are often depicted as uninvolved and irresponsible (Madhavan & Roy 2012;Montgomery et al., 2006), and many black fathers have to pursue work opportunities that are geographically separated from family life (Desmond & Desmond, 2006;Madhavan et al., 2008;Townsend, Madhavan, & Garey, 2005).Black low-income fatherhood is therefore often characterised by absence (Madhavan & Roy, 2012;Morrell et al., 2003;Richter et al., 2010;Roy, 2008;Spjeldnaes et al., 2011).Not only migratory labour, but unemployment, poverty, HIV/AIDS and strained personal relationships disempower men and can act as barriers to men fulfilling the new fatherhood role (Desmond & Desmond, 2006;Mkhize, 2006;Montgomery et al., 2006;Rabe, 2007).Despite these barriers, some researchers have found that fathers are involved in various ways in the lives of their children (Madhavan et al., 2008) and in various aspects of childcare (Montgomery et al., 2006).It needs to be kept in mind that coresidence is not a good indicator of father involvement and fatherhood has to be conceptualised differently (Rabe, 2006;Townsend et al., 2005).There are various ways for fathers to be connected with and support their children, as well as being involved in their lives, without permanently and physically living with them (Madhavan et al., 2008;Morrell, 2006;Rabe, 2007;Richter et al., 2010).
Only two studies focussing on White South African fathers could be found.Khunou (2006) conducted a qualitative study with fathers who are struggling to gain access to their children.These men expressed a great desire to be part of their childrens' lives and discussed the difficulties fathers face in the context of custody policies' apparent favouring of mothers as caregivers (Khunou, 2006).The second study focussed on how Afrikaans speaking, Paternal Involvement and Nurturance Christian fathers were represented in a men's magazine (Viljoen, 2011).Viljoen found that the magazine included themes of new fatherhood, with a focus on how to be a good father (2011).

Theoretical Framework
Although conceptualisations differ, the majority of the authors support the idea of conceptualising father involvement as a multidimensional construct (Hawkins & Palkovitz, 1999;Schoppe-Sullivan, McBride, & Ho, 2004).In the absence of a universally accepted conceptualisation, we will approach father involvement as a multidimensional construct for the purpose of this study.This includes the father's presence or "being there", either physically or emotionally, in a variety of domains in the child's life.This study will only consider daughters' accounts of involvement and the focus is on perceived paternal involvement.Finley and Schwartz (2004) argue that the positive developmental and behavioural outcomes are a product of the perceived involvement, regardless of actual involvement.
There is thus evidence to support the utility of assessing paternal involvement in terms of young adults' perceived paternal involvement (Finley & Schwartz, 2004;Rohner & Britner, 2002).

Definitions of Concepts -
For the purpose of this study, the term father refers to the male figure the participant most identifies with as being her father.The inclusion of non-biological fathers is motivated by research that found high involvement of a paternal figure could have the same positive outcomes as those associated with high involvement of biological fathers (Agar et al., 2010;Amato & Rivera, 1999;Schwartz & Finley, 2006).Perceived involvement refers to the extent that young adult daughters experienced their fathers as being present, in a physical and emotional sense, in various domains in their lives.Young adulthood refers to the transitional period between adolescence and adulthood.

Aims and Objectives
The objectives of the study were to assess (i) the level of paternal involvement experienced by young adult daughters during childhood and adolescence, (ii) the level of paternal involvement currently experienced, (iii) the level of paternal involvement desired by young adults during childhood and adolescence, (iv) the current level of desired involvement, (v) the level of paternal nurturance experienced by young adult daughters during childhood and adolescence (vi) as well as the current level of paternal nurturance experienced by young adult daughters, attending a South African university.As a secondary objective, this study also examined the relationships between demographic variables and retrospective and current levels of nurturance, reported involvement and desired involvement.

Method Participants
Data was collected from a convenience sample of female third year Psychology students, between the ages of 19 and 22, who had a father or a father figure in their lives.See Table 1 for a detailed description of the biographical variables.A total of 89 participants completed the questionnaires.The average age of participants in the sample was 21 years with a standard deviation of .686.The majority of the participants were white (n = 77, 86.5%), Afrikaans speaking (n = 57, 64%), their parents were still married (n = 70, 78.7%), well-educated (n = 72, 80.9% of fathers and n = 67, 75.3% of mothers obtained tertiary education) and fell in the middle to higher income group (n = 54, 60.7% of participants reported their monthly household income to exceed R25 000 or $2 900 approximately).

Research Design
A cross sectional research design was used.Data was collected via self-report questionnaires that were completed by a convenience sample of young adult daughters who agreed to participate voluntarily in the study.

Measuring Instruments
The self-administered questionnaires included a biographical questionnaire as well as past and present Father Involvement Scales (FIS) and past and present Nurturant Fathering Scales (NFS).The biographical questionnaire included variables such as age, home language, ethnicity, parents' marital status, whether divorced parents were remarried, father's residence, parents' level of education, monthly household income and monthly allowance.
The FIS is a retrospective measure, which measures young adults' perceived and desired father involvement in 20 domains throughout their lives.It employs a Likert-type scale ranging from 1, not at all involved, to 5, very involved, for reported involvement, and 1, much less involved, to 5, much more involved, for desired father involvement.Participants can obtain a minimum score of 20 and a maximum score of 100 for reported and desired father involvement respectively (Finley & Schwartz, 2004).Finley and Schwartz (2004) do not offer guidelines for how these averages should be interpreted, except that a low score indicates low perceived paternal involvement and a high score indicates high perceived paternal involvement.The desired involvement scale has a curvilinear answering pattern.In this case 3 (it was just right) indicates highest satisfaction where a low score indicates less desired involvement (dissatisfaction) and a high score indicated more desired involvement (also dissatisfaction).
Exploratory and confirmatory factor analysis done by Finley and Schwartz (2004) indicated that a three-factor structure provided the best fit for the data.The three subscales were Expressive involvement, Instrumental involvement and Mentoring/advising.In the original study on the psychometric properties of the FIS, high internal consistency was reported for all three subscales (Finley & Schwartz, 2004).Exploratory and confirmatory factor analysis for desired father involvement yielded a two-factor structure, namely Expressive and Instrumental involvement.
These findings support the multidimensionality of father involvement (Finley & Schwartz, 2004).Finley and Schwartz (2004) found a high intercorrelation between the NFS and the three Reported Father Involvement subscales.However, they could not be collapsed into a single factor, lending some support for the discriminant validity of the NFS and the three subscales of the Reported Father Involvement (Finley & Schwartz, 2004).
The NFS assesses young adults' perceived quality of their relationship with their fathers.The NFS is a retrospective, 5-point Likert-type scale, consisting of nine items.Possible scores range from 9 to 45.As with the FIS, no normative data exists for the scale, However, considering the way the items are scored, a low score would indicate low nurturance and a high score high nurturance.Finley and Schwartz (2004) found that the NFS had a high internal consistency, and a unidimensional factor structure that explained 69.4% of the variance.The NFS and FIS have been found to have high internal consistency over a diverse sample of American young adults (Finley & Schwartz, 2004;Finley et al., 2008;Schwartz & Finley, 2005, 2006).
Although it is important to keep possible bias in mind when working with retrospective reports (Schwartz & Finley, 2009), the purpose of this study is to assess the degree to which young adult daughters perceived their fathers as being involved, irrespective of actual involvement.As Finley and Schwartz (2004) point out, "if an adolescent or adult child perceived that her father was highly involved in her life, then that father's impact on his child is a consequence of this perception of high involvement-independent of the veridicality of that perception" (p.145).
However, there is a need for research on the current levels of perceived and desired father involvement and nurturance, considering the absence of such research.For this purpose, we adapted the FIS to measure current levels of perceived and desired involvement and the NFS has been adapted to measure current levels of nurturant fathering.Adaptation involved changing the questions from past tense to present tense, for example "When you needed your father's support, was he there for you?" is changed to "When you need your father's support, is he there for you?" The FIS and the NFS have not been tested for use in the South African context.Consequently, a subset of the population (n = 32) were asked to participate in a pilot study.High Cronbach's alphas were found for all four measures of paternal involvement and both measures of paternal nurturance: Past Perceived Involvement = .97,Past Desired Involvement = .98,Present Perceived Involvement = .97,Present Desired Involvement = .97,Past Nurturance = .95,Present Nurturance = .96.
Internal consistency was also calculated for the main sample (N = 89).The reliability for both paternal nurturance scales were high (see Table 2).For both scales, exploratory factor analysis (extraction method: principal component analysis) showed that all nine items loaded highly on a single factor, indicating unidimensionality of the scales.
The factor loadings for past paternal nurturance ranged from 0.562 to 0.938 and the factor explained 68.045% of variance.The factor loadings for present perceived nurturance ranged from 0.566 to 0.923 and explained 76.170% of variance.Due to the small sample size and the number of items in each scale, it was not statistically meaningful to conduct a factor analysis on the father involvement measures.Each scale was subdivided into three subscales.The decision to subdivide the scales was based on the idea that father involvement is not a unidimensional construct, as demonstrated by a number of authors (Finley & Schwartz, 2004;Hawkins & Palkovitz, 1999;Lamb, Pleck, Charnov, & Levine, 1985;Schoppe-Sullivan et al., 2004).The subscales were based on previous conceptualisations, and previous factor analyses (see Finley & Schwartz, 2004).The first subscale, involvement in development, includes all items that ask participants to report the amount of paternal involvement experienced or desired in a particular domain of development.Items include intellectual, emotional, social, ethical/moral, spiritual, physical, and career development, as well as developing responsibility, independence and competence.The second subscale includes all items addressing expressive involvement.Expressive involvement can be seen as the emotional or caring dimension of involvement.This subscale includes items measuring leisure/fun/play, sharing activities/interests, mentoring/teaching, caregiving, school/homework, and companionship.The third subscale includes all items associated with instrumental involvement.Instrumental involvement can be regarded as the dimension of involvement that deals with parental responsibilities.This domain includes those items that are associated with the traditional role of the father, namely: providing income, being protective, advising, and discipline.All four measures of father involvement and their respective subscales had high reliability for this sample (Table 2).

Data Collection Procedure
We obtained institutional permission to conduct this research with university students, as well as ethical clearance from the relevant university ethics committee.The research objectives and procedure, as well as the ethical rights of participants were explained to the psychology third year students during one of their Psychology lectures.They were asked to voluntarily participate in the study.Students were informed of their ethical rights.All students who did not meet the criteria or chose not to participate, were free to leave the venue and volunteers were asked to stay behind.Supervised by the researchers, these latter students completed informed consent forms, along with the biographical questionnaires, the four involvement scales and two nurturance scales.

Data Analysis
SPSS was used to compute a descriptive level analysis for the data.This included calculating population statistics.
Correlation coefficients were calculated to assess the strength of the relationship between the various involvement and nurturance variables.Correlation coefficients were also calculated to assess the strength and the direction of the relationship between demographic variables and retrospective and current levels of nurturance, reported involvement and desired involvement.

Results
For a complete summary of the mean score and range for each involvement and nurturance variable, see Table 2.The average Past Nurturance score was 33.75, indicating that the young adult daughters in the sample experienced relatively high levels of past paternal nurturance.Present paternal nurturance was also high and comparable to past paternal nurturance.Participants reported high past perceived paternal involvement.The average Past Perceived Paternal Involvement across subscales was 75.83.As with the Nurturant Father Scale, no normative data exists for the Perceived Father Involvement Scales.However, considering the way the items are scored, an average of 60 could logically be taken to indicate "somewhat involved" and 80 would indicate "often involved".An average score of 75.83 indicates that daughters did perceive their fathers as having been involved while growing up.The average score across subscales for Present Perceived Paternal Involvement is 67.56, indicating that young adult daughters reported their father to be slightly less involved compared to when they were growing up.
Scores for the Desired Father Involvement scales also range from 20 to 100.In this case, a score of 60 would indicate "it was/is just right", and 80 that they wanted him to be "a little more involved".The average score across subscales for Past and Present Desired Paternal Involvement was 70.16 and 67.28 respectively, signifying that young adult daughters were generally satisfied with the amount of past and present paternal involvement, although they would have wanted their fathers to be slightly more involved.
A bivariate correlation was run to assess the relationship between the fourteen paternal involvement and nurturance variables (Table 3).There was a significant positive correlation between the past and present perceived paternal involvement subscales, between past and present desired paternal involvement subscales, between the past and present nurturance scales, and between the past and present perceived paternal involvement subscales and the nurturance scales.Past and present desired involvement subscales had a significant negative correlation with past and present perceived involvement subscales and nurturance scales.
Correlation coefficients were also calculated between the biographical variables and the paternal involvement and nurturance variables.Six biographical variables correlated significantly with the paternal involvement and nurturance variables.These included age, marital status of parents, mother remarried, father remarried, father residence and level of education of the father.A partial correlation was run between each of the six biographical variables and the paternal involvement and nurturance variables, controlling for the other five biographical variables.There was a positive correlation between age and past perceived involvement in development (r(86) = .279,p = .013),past perceived expressive involvement (r(86) = .301,p = .007),past perceived instrumental involvement (r(86) = .357,p =. 001), and past nurturance (r(86) = .290,p = .009).There was a negative correlation between age and past desired involvement in development (r(86) = -.260,p = .021).Age also correlated with present perceived involvement in development (r(86) = .263,p = .019),present perceived expressive involvement (r(86) = .227,p = .045),present perceived instrumental involvement (r(86) = .233,p = .039).Marital status of parents was negatively correlated with present perceived involvement in development (r(84) = -.443,p < .001),present perceived expressive involvement (r(84) = -.322,p = .004),present perceived instrumental involvement (r(84) = -.432,p < .001)and present nurturance (r(84) = -.543,p < .001).This indicates that being a young adult daughter from a divorced family is associated with lower levels of present paternal involvement and nurturance.The remarried status of the father was positively correlated with present desired expressive involvement (r(86) = .230,p = .041).
Father residence was negatively correlated with present perceived involvement in development (r(84) = -.283,p = .012),present perceived expressive involvement (r(84) = -.3.16, p = .005),and positively correlated with present desired expressive involvement (r(84) = .256,p = .023).This indicates that the further the father lives from the daughter, the less involved she experiences him to be in the developmental and expressive domains of involvement and the more involved she wants her father to be in terms of expressive involvement.The level of education obtained by fathers was positively correlated with past perceived expressive involvement (r(86) = .226,p = .045)and negatively correlated with past desired involvement in development (r(86) = -.254,p = .024),past desired expressive involvement (r(86) = -.226,p = .045),past desired instrumental involvement (r(86) = -.350,p = .002),present desired involvement in development (r(86) = -.288,p = .010),present desired expressive involvement (r(86) = -.351,p = .001)and present desired instrumental involvement (r(86) = -.291,p = .009).The remarried status of the mother did not correlate significantly with any of the paternal involvement and nurturance variables, when controlling for age, marital status of parents, remarried status of the father, father residence and education of father.

Descriptive Results
The aim of this study was to assess young adult daughters' past and present perceived and desired paternal involvement as well as perceived past and present paternal nurturance.The total scores for past perceived involvement and nurturance indicate that young adult daughters experienced their fathers as having been involved and nurturing while they were growing up.Although there was no decline in the level of nurturance experienced, perceived involvement did decrease slightly in young adulthood.Average scores on all three subscales for past perceived involvement indicate that daughters reported fathers to have been often involved while they were growing up, while average scores indicate that they currently experienced their fathers as sometimes involved.The comparable means of the perceived past involvement and present involvement show that fathers were not reported to be more involved in the instrumental domain as originally proposed by Parsons and Bales (as cited in Finley & Schwartz, 2006).
Past and present desired involvement scores indicate that daughters were largely satisfied with the amount of paternal involvement experienced.There were differences in the level of paternal involvement desired across subscales for past desired involvement.Daughters reported that they were satisfied with the amount of instrumental involvement, but that they desired their fathers to be slightly more involved in the developmental and expressive domains.Although there is a slight decrease in reported paternal involvement during young adulthood, daughters still reported being satisfied with the level of paternal involvement currently experienced across all three subscales.

Correlates of Parental Involvement and Nurturance
As higher levels of education are often associated with higher involvement of both parents in their children's lives and the sharing of parental roles (Ozgun & Honig, 2005), it is possible that our results reflect that our middle-class to upper middle class sample's fathers were adhering to the newer construction of fatherhood and were sufficiently involved with daughters.This explanation resonates with our finding that the greater the level of education obtained by the father, the more involved daughters reported their fathers to have been in the expressive domain.What is interesting is that the level of education obtained by the father was negatively correlated with all six past and present desired involvement subscales, indicating that daughters' whose fathers are highly educated, actually experienced their fathers as wanting to be too involved in their lives.
Our participants' satisfaction with paternal involvement could be due to them depending less on paternal involvement in their specific developmental stage, and thus they are more satisfied with the level of involvement.Another explanation could lie in daughters' socialisation regarding gendered expectations of fathers.These daughters could have reported satisfaction with the level of father involvement, because they interpret their fathers' involvement in terms of the level of involvement expected of fathers in society, rather than the level they personally desire.This explanation is especially relevant for our sample of mostly Afrikaans-speaking students.Paternal Involvement and Nurturance and vestiges thereof remains.It also needs to be taken into account here that the questionnaires used in this study did not measure the level of paternal involvement daughters believed they can reasonably expect from their fathers or how they evaluate the quality of paternal involvement.
Our results indicate that high past perceived paternal involvement and nurturance is associated with high reported current paternal involvement and nurturance; and high past desired paternal involvement is associated with high present desired paternal involvement.This suggests that paternal involvement stays relatively constant throughout the daughters' development.Furthermore, young adult daughters who reported low past and present levels of paternal involvement, reported high desired paternal involvement.This is in concurrence with Finley and Schwartz (2007), who also found that daughters desire father involvement.
Ethnicity, home language, educational level of mother, monthly household income and allowance did not influence perceived and desired paternal involvement and nurturance in this sample.However, the sample was not racially and socio-economically diverse as the majority of the participants were white and reported Afrikaans as their home language.
There was a slight decline in reported paternal involvement from retrospective to current accounts.However, there was a positive correlation between age and current reported involvement.It is possible that there are two processes occurring during young adulthood that could explain these seemingly contradictory results.Firstly, as daughters leave school and establish their autonomy, they experience their fathers to be less involved.Secondly, as they mature in young adulthood, they reconnect with their fathers.What is interesting is that age was also positively correlated with past perceived involvement.Theoretically, the current age of the young adult daughter cannot affect the level of paternal involvement experienced while growing up.However, it is possible that the increase in perceived paternal involvement associated with the increase in age in young adulthood, leads daughters to assess paternal involvement more positively retrospectively.This raises many questions about the utility of using retrospective accounts.
Not surprisingly, divorce was associated with lower levels of current perceived involvement and nurturance.In cases where fathers were remarried, young adult daughters expressed a greater desired for expressive involvement.
It is possible that this domain of involvement is affected more when the father remarries than the other two.
Whether the mother was remarried did not significantly influence perceived or desired paternal involvement and nurturance, when age, marital status of parents, father remarried, father residence and level of education obtained by father were controlled for.Daughters reported less expressive involvement and involvement in development the further away their fathers lived.This was coupled with an increase in desired expressive involvement.It is possible that daughters need their fathers to live near them, which likely means more visits and face-to-face contact, in order to experience their fathers as involved in the expressive and developmental domains.

Limitation
The present study has a number of limitations.Firstly, the sample was small and not representative of either the South African adult daughter, or the student population.Because South Africa is culturally and socio-economically diverse, these results that are based on mainly white, Afrikaans speaking, middle to higher income participants cannot be generalised to other populations.It is therefore important to research fatherhood across populations and socio-economic status.Secondly, it is possible that daughters who do not have satisfactory relationships with fathers may have been less likely to participate in the study.

Table 1
Summary of Biographical Variables

Table 2
Mean Scores, Range and Cronbach's Alphas for All Father Involvement and Nurturance Variables.

Table 3
Bivariate Correlation Between Paternal Involvement and Nurturance Variables