Revisiting men's role in father involvement: the importance of personal expectations.

Using fathers' and mothers' reports of expectations (measured prenatally) and father involvement (measured postnatally), we examined how both parents influence the likelihood that new fathers would be involved in instrumental (feeding, bathing, and changing the infant) and affective (playing and reading to the infant) caregiving activities. The study employed a longitudinal design with 68 couples participating in both the prenatal (e.g., approximately three months before the infant was born) and postnatal (e.g., between three to six months after the birth of the infant) phases of the study. Results indicate that both parents' expectations are substantial predictors of instrumental involvement (as reported by both fathers and mothers), and that fathers' expectations are stronger than mothers' expectations for predicting affective involvement. Limitations follow.

Keywords: father involvement, expectations, transition to parenthood

**********

Research has documented the importance of mothers' attitudes when predicting fathers' involvement with their children (Allen & Hawkins, 1999; Fagan & Barnett, 2003; McHale & Huston, 1984). Mothers who have traditional attitudes about the division of caregiving responsibilities, for example, are known as "gatekeepers" (Allen & Hawkins, 1999) because their stereotypical attitudes hinder father involvement. While mothers can and sometimes do limit father involvement, it is doubtful that mothers are exclusively responsible when father involvement is minimal (Walker & McGraw, 2000).

This study focuses on personal expectations as another important factor when discussing how both mothers and fathers influence the likelihood of father involvement with their infant children. Since research indicates that the early years of parenthood are representative of future father involvement (Cowan & Cowan, 1998), this study examines the role that personal expectations prior to the birth of a child influence paternal involvement after the child is born.

FATHER INVOLVEMENT: FROM THOUGHT TO BEHAVIOR

We have chosen to narrow our focus of father involvement to the thought processes that precede involvement. These thought processes can generally be categorized as gender-role ideologies, or the extent that parents believe they (and others) should adhere to traditional expressions of behavior (e.g., Beitel & Parke, 1998; Bonney, Kelley, & Levant, 1999; McBride & Darragh, 1995). Those who believe men should be involved in the day-to-day caregiving of children, for example, are considered egalitarian or nontraditional in their ideology.

Mixed conclusions have surfaced about the likelihood of father involvement when using gender-role ideology as a predictor variable. There is support for the entire spectrum of possibilities, including (1) "gatekeeping," or that the mother's traditional ideology is an important factor in predicting father involvement (Allen & Hawkins, 1999; Fagan & Barnett, 2003; Hoffman & Moon, 1999; McHale & Huston, 1984), (2) that the father's but not the mother's egalitarian ideology is associated with his involvement (Bulanda, 2004; Wille, 1995), and (3) that neither parent's ideology seems to have an impact on the father's likelihood of involvement (Marsiglio, 1991).

Research has also shown that behavior deviating from one's ideology can create tension within the individual and the marriage (Barclay & Lupton, 1999) and that men (as parents or nonparents) will not participate in domestic labor unless both spouses hold egalitarian ideologies (Greenstein, 1996). Although a recent societal shift in ideology has occurred whereby men and women are encouraged to take equal roles in caregiving responsibilities (Hawkins & Dollahite, 1997; McBride & Darragh, 1995; LaRossa, 1997), fathers still tend to be less involved than mothers with regard to childcare tasks (Pleck, 1985). Due to the overall confusion in identifying what level of influence ideology has on father involvement, personal expectations are examined in this study as an alternative perspective for how thoughts are translated into behavior.

PERSONAL EXPECTATIONS

Whereas ideology reflects a societal perception of what should be done, personal expectations portray personal beliefs for what will be done. McBride and Darragh's (1995) research found that personal standards for paternal involvement may be more influential than ideology for some couples. Research has found that parents typically have very high expectations for how involved they will be with their children (Hooker, Fise, Jenkins, Morfei, & Schagler, 1996) and falling short of personal expectations can be detrimental for the psyche of the person as well as for the cohesion within the marriage (Cowan & Cowan, 1992; Litton-Fox, Bruce, & Combs-Orme, 2000; Strauss & Goldberg, 1999). Although the literature focuses on the negative consequences of having high expectations (Cowan & Cowan, 1992; Litton-Fox et al., 2000; Strauss & Goldberg, 1999), these expectations may motivate a behavioral response; the question still remains as to whether high personal expectations lead to greater levels of father involvement.

METHODOLOGICAL CONCERNS

To provide a more objective and comprehensive assessment of father involvement, Fagan and Barnett (2003) recommended that research include both parents' perceptions. Another concern, according to Fagan and Barnett, is that causal relationships between maternal attitudes (about their own role and the role of fathers) and paternal behavior are inferred in the absence of longitudinal data. Bulanda (2004), using longitudinal data from the National Survey of Families and Households, found that the father's (but not the mother's) gender-role ideology predicts both parents' reports of father involvement. This study seeks to examine the role that both parents' beliefs, measured as personal expectations, predict father involvement. Whereas statistical controls were used for Bulanda's (2004) assessment, we have opted to "control" for father involvement by examining how personal expectations measured during the prenatal period are linked to father involvement shortly after the child's birth. Although "involvement" with one's posterity is complex and can begin prior to the birth of a child (Sandelowski, 1994; Shapiro, 1987), in this study the notion of father involvement is limited to more traditional caregiving activities (e.g., feeding, changing, and bathing the child). Finally, whereas Bulanda's study included parents of children who were between five and 13 years of age, our study looks at father involvement shortly after the child is born (e.g., between three and six months after the birth).

METHODS

SAMPLE AND PROCEDURES

Expectant parents in their second or third trimester of pregnancy who attended various childbirth classes given by two hospitals in the Intermountain region were invited to participate in a study about preparing to become parents for the first time. Ninety-six couples agreed to participate, took their surveys home with them, and returned the completed surveys at the next scheduled prenatal class meeting. Initial incentives included raffle drawings for diapers and baby monitors, but a lack of participation led to a $10 incentive for the vast majority of those who later completed the surveys.

The couples were contacted again, usually by telephone, approximately three months after the anticipated birth date of their child and were invited to fill out another survey. Nearly all the surveys completed during the postnatal period were returned within one month, but some were returned up until the child was six months old. With an attrition rate of 29%, 68 of the initial 96 couples completed the second (or postnatal) survey and received $10 as compensation. The child's age of three months was chosen primarily for methodological reasons; three months was sufficient for establishing some kind of paternal involvement pattern. As reported in the prenatal survey, the ages of the final sample ranged from 17 to 49 years (Father M = 27.58, Mother M = 25.01), and the range for length of marital duration was nearly six months to seven and one-half years (M = 2.43).

The majority (93%) of the participants were non-Hispanic white, while the annual income distribution of couples varied considerably: 12% earned less than $12,000, 24% earned between $12,000 and $24,000, 37% earned between $24,001 and $36,000, and the remaining 27% of the participating couples earned more than $36,000 per year. One couple did not report their income. During the prenatal period, nearly half of mothers and more than three-fourths of fathers were working 30 hours or more per week. These figures dipped slightly during the postnatal period, with 41% of mothers and 74% of fathers working 30 hours or more per week. The gender distribution of the infants was nearly even with 44% being female. Birth weight ranged from 5.06 to 13 pounds with an average (mean) weight of 7.68 pounds.

MEASURES

The prenatal measures, collected from both husbands and wives during the prenatal assessment, consisted of demographics (age, hours worked per week) and expectations for father involvement. Items of father involvement include: "Number of times per week you expect to change baby's clothes, play with your child, read to your child, bathe your child, and feed your infant." (It was initially unclear whether "feed your infant" should be included in the analyses if mothers were nursing their infants. The decision to retain "feed your infant" in the analysis was determined after finding that 81% of the fathers in this sample reported some level of involvement on that item.) Table 1 includes descriptive statistics for both husbands' and wives' expectations of father involvement. Hours worked per week was measured at the ordinal level using a Likert scale that ranged from 1 = none to 7 = more than 40 hours per week. The postnatal survey was similar to the prenatal survey, except that father involvement (rather than the expected father involvement) was reported by both mothers and fathers between the child's age of three and six months. Table 1 provides descriptive information for how involved the fathers were according to reports from both husbands and wives.

RESULTS

Two separate analyses were conducted, each using postnatal reports of father involvement from either the mother (mother reports father involvement with infant) or the father (father reports his level of involvement with infant). In each analysis, frequency scores (number of times per week that each activity occurs) for each of five activities were entered into a principal component analysis using Varimax rotation. The five activities were feeds the child, changes baby's clothes, plays with child, reads to child, and bathes child. As shown in Table 2, when mothers' reports of father involvement were used in the analysis, two separate and distinct factors emerged. The first factor (accounting for 34.10% of the variance) consisted of three activities (bathes, changes clothes, and feeds the child) that appeared to represent more functional tasks for physical needs, while the second factor (25.85%) included two tasks (plays with child and reads to child) representing more interactive and emotive tasks.

The first factor of caregiving tasks (e.g., feeds, changes clothes, and bathes child) was labeled as "instrumental." The second factor consisted of more emotionally interactive activities (plays and reads to the child) and was labeled "affective." Combined, the two factors accounted for 59.95% of the variability in these five father-infant activities.

Two factors were also identified when father reports of his involvement were subjected to an identical factor analytical procedure. As shown in Table 3, the first factor (accounting for 36.59% of the variance) included three activities (reads, plays, feeds child), while the second factor (20.92%) included two tasks (changes clothes, bathes child). Altogether, the two factors accounted for 57.51% variability in these five fathering activities. To avoid confusion, we chose to use the factors that were identified from the analysis of mothers' responses for the remainder of the study. The mother's perception was also selected so that the results could be compared with those from previous studies, which have examined how the mother's beliefs influence the likelihood of father involvement.

PREDICTOR VARIABLES

Four separate expectation variables were created with the intent of mimicking the involvement factors that emerged from the factor analysis of mothers' reports of father involvement as reported above: fathers' instrumental expectations (bathes, changes clothes, feeds child); fathers' affective expectations (plays, reads to child); mothers' instrumental expectations for fathers (bathes, changes clothes, feeds child); and mothers' affective expectations for fathers (plays, reads to child). The activity scores within each of the two expectation variables were summed separately for fathers and mothers. All told, our eight predictor variables consisted of fathers' expectations for their Instrumental involvement, fathers' expectations for their affective involvement, mothers' expectations for their husbands' instrumental involvement, mothers' expectations for their husbands' affective involvement, mothers' age, fathers' age, mothers' work per week, and fathers' work per week.

CRITERION VARIABLES

Four involvement variables were created from the postnatal reports of actual father involvement. These consisted of mothers' reports of fathers' instrumental involvement (bathes, changes clothes, feeds child), mothers' reports of fathers' affective involvement (plays, reads to child), fathers' self-reported instrumental involvement (bathes, changes clothes, feeds child), and fathers' self-reported affective involvement (plays, reads to child). The involvement variables consisted of summed activity data within each of the four categories described above. Frequency distributions were generated for each of these four variables. For each variable, low involvement consisted of scores below the 50th percentile and high involvement included scores above the 50th percentile.

PRENATAL PREDICTORS OF POSTNATAL FATHER INVOLVEMENT

Discriminant analyses were used to determine the relation between high and low levels of father involvement and the prenatal variables described above. The decision to dichotomize the dependent variables was based upon our examination of frequency distributions. Both of the mothers' distributions were positively skewed, and both distributions included online outliers. In addition, the instrumental domain scores were bi-modal. Therefore, dichotomizing the father involvement variables was used to retain as many responses as possible, and at the same time eliminate the statistical nuances resulting from non-normal distributions and outliers.

Separate analyses were conducted for each of the four criterion variables (mothers' reports of fathers' instrumental and affective involvement; fathers' self-reported instrumental and affective involvement). For the analyses that employed mother reports of involvement, the mothers' age and employment variables were included as predictors. For the analyses that employed fathers' self-reported involvement, the fathers' age and employment variables were included as predictors. In all four of the analyses, both the fathers' and mothers' expectations for involvement were included as predictors.

PRENATAL PREDICTORS OF FATHER'S SELF-REPORTED POSTNATAL INSTRUMENTAL INVOLVEMENT

As shown in Table 4, the analysis that differentiated high and low levels of Instrumental involvement (as reported by father) produced a statistically significant function, Chi Square (N = 59, 6) = 15.66; p < .05. The six expectation and demographic variables combined to explain 25% of the variability between fathers who reported high and low levels of instrumental involvement. The strongest predictor of fathers' reports of instrumental involvement was the mothers' prenatal expectations for affective involvement. However, the relationship was negative; as mothers' affective expectations increased, fathers reported less instrumental involvement. Not surprisingly, the fathers' prenatal expectations for both affective and instrumental involvement as well as the mothers' prenatal expectations for instrumental involvement were also predictor variables in the function. In all three instances fathers and mothers who reported higher expectations (36%, 9%, and 9%, respectively) also reported higher involvement. Fathers' age and employment did not contribute to the function's ability to differentiate between high and low levels of father involvement (as reported by the fathers).

PRENATAL PREDICTORS OF FATHER'S SELF-REPORTED POSTNATAL AFFECTIVE INVOLVEMENT

Table 5 summarizes the analysis that employed fathers' self-reported affective involvement with their infants as the criterion variable. In this analysis the six predictor variables combined to form a statistically significant function, Chi Square (N = 57, 6) = 21.60; p < .05, that explained nearly 34% of the variability between fathers in the low and high affective involvement groups. Not surprisingly, fathers' prenatal affective expectations emerged as the strongest predictor of low and high affective involvement with highly involved fathers expressing prenatal expectations for affective involvement that were 46% higher than those of fathers who reported low affective involvement. The next strongest predictor was fathers' instrumental expectations with highly involved fathers expressing higher (34%) instrumental expectations than fathers in the low involvement group. Fathers' work hours were negatively related; as work hours increased, fathers' reports of their affective involvement decreased. The fourth predictor was mothers' affective expectations. Mothers' instrumental expectations and fathers' age did not contribute to the function.

Prenatal Predictors of Father's Postnatal Instrumental Involvement as Reported by Mother. When mothers' reports of fathers' instrumental involvement were entered as the criterion variable, the mothers' prenatal expectations for their husbands' affective involvement emerged as the strongest predictor. Contrary to expectation, however, this relationship was negative: mothers who anticipated higher levels of affective involvement from their husbands reported lower levels of instrumental involvement by the fathers, whereas fathers who were more involved in the "instrumental" role had wives who held lower expectations for their husbands' affective involvement. Conversely, fathers who held higher expectations for their affective involvement (+35%) exhibited higher levels of instrumental involvement as reported by their wives. Both the mothers' and the fathers' prenatal expectations for high levels of instrumental involvement by the fathers were associated with higher levels of instrumental involvement by fathers. Additionally, mothers who reported lower levels of paternal instrumental involvement were younger and tended to work fewer hours per week than did mothers who reported higher levels of instrumental involvement by fathers. In this analysis, the six predictor variables combined to produce a statistically significant discriminant function, Chi square (N = 61, 6) = 16.71; p < .05, that explained more than 25% of the variability between low and high levels of paternal instrumental involvement as reported by new mothers (see Table 6).

Prenatal Predictors of Father's Postnatal Affective Involvement as Reported by Mothers. Mothers' observations of fathers' affective involvement were not significantly related to the six predictor variables used in the discriminant analysis, Chi square (N = 60, 6) = 5.14; p > .05, and in fact explained only 9% of the variability between fathers who exhibited high and low levels of affective involvement. As shown in Table 7, the fathers' own prenatal expectations for instrumental involvement emerged as the strongest predictor, where fathers who exhibited high levels of affective involvement had expectations for instrumental involvement that were 26% higher than similar expectations held by fathers who exhibited low levels of affective involvement. In every instance (namely, mothers' expectations for affective and instrumental paternal involvement; fathers' expectations for affective and instrumental paternal involvement), prenatal expectations were greatest for fathers who exhibited high levels of affective involvement. Additionally, mothers who reported higher levels of affective paternal involvement were slightly younger and tended to work fewer hours per week than mothers who reported low levels of affective involvement by fathers.

DEMOGRAPHIC PREDICTORS

Age of mothers contributed to the function that predicted their reports of father involvement; the greatest reports of instrumental involvement and the least amount of affective involvement were reported by older mothers. Regarding employment, mothers reported more instrumental but less affective involvement when they worked more hours. For fathers' reports, older fathers were more likely to report greater instrumental involvement, but their age was not related to their reports of affective involvement. Fathers' employment was predictive of less involvement; as employment hours increased, fathers' reports of their instrumental and affective involvement decreased.

SUMMARY OF RESULTS

Results indicated that fathers' prenatal expectations were substantial predictors of involvement, as reported by both fathers and mothers. Mothers' expectations were also linked with their reports of father involvement, but in a manner that is different from the fathers' reports. Older mothers and fathers tend to report greater levels of Instrumental involvement, and yet older mothers are also likely to report less affective involvement. Age for fathers was not a predictor for more or less affective involvement. In general, employment was a stronger and more consistent predictor than age for father involvement; the more parents worked, the lower their reports of involvement tended to be. There is one exception to this: the more mothers are employed, the more they report their husbands are involved in instrumental activities.

The classification functions associated with the discriminant analyses reported above are summarized in Table 8. As shown, the analysis that employed father's reports of involvement produced greater prediction accuracy than either of the classification functions employing the mothers' results. The accuracy of predicting low involvement for all four classification results ranged from 69.7% to 86.5%. The accuracy of predicting high involvement, as reported by both parents, was slightly less accurate for each analysis, ranging from 55.6% to 80%.

DISCUSSION

Our findings emphasize the need to examine father involvement in the context of both parents (Bulanda, 2004; Fagan & Barnett, 2003). Given both parents' reports, it seems that new fathers influence their own involvement as much as (and sometimes more than) mothers. For fathers in this study, expectations for affective and instrumental involvement were each predictive of their instrumental involvement, and their expectations for affective involvement were predictive of their reports of subsequent affective involvement. These results suggest that paternal expectations motivate fathers' behavior. It may also be that expectations serve as filters for how involvement is viewed or that fathers tend to interpret and report their involvement in a way that is consistent with their biases (or expectations, in this case).

Mothers' expectations also play a role in father involvement, particularly when the mothers' expectations for affective caregiving (e.g., reading and playing with infant) are high. The direction of this relationship was negative, however. Mothers and fathers tend to report lower levels of instrumental (e.g., feeding, changing, and bathing infant) involvement when mothers' affective expectations were high during the prenatal phase. This supports previous studies showing that mothers' traditional attitudes, as indicated in this study by their high expectations for affective involvement, hinder paternal behavior (Allen & Hawkins, 1999; Fagan & Barnett, 2003; Hoffman & Moon, 1999; McHale & Huston, 1984).

Lower levels of instrumental involvement may also result from a third, unmeasured variable. Perhaps mothers and fathers negotiate father involvement during a time when the marriage is already likely to experience greater tension (Cowan & Cowan, 1992); parents learn to specialize in particular roles to avoid criticism and tension from the other parent. It is also possible that mothers with high expectations for instrumental involvement maintain certain standards for the method and quality of involvement, thereby discouraging paternal participation in those tasks.

LIMITATIONS

One of the more noteworthy concerns is that the sample was one of convenience, limited in size and ethnically homogenous. Future research is necessary to determine whether these results generalize to a more diverse population of new parents. Measurement issues are also a concern. It could be that the concept of "personal expectations" represents a more comprehensive composite of thought processes than assessed in this study. Separating expectations that are goal-oriented and active from expectations that are passive is one option and tends to follow in the spirit of the "generative father" framework (Hawkins & Dollahite, 1997) in that parents are active in how fathers get involved. Another approach is to study father involvement and expectations in terms of what each spouse hopes for (e.g., ideal), what each is willing to accept (e.g., acceptable), and what each spouse will not accept (e.g., unacceptable).

Another concern is measurement of father involvement. In this study, father involvement included five items that were measured on a frequency basis. How many times per week a father engages in these activities with his child may very well differ from how much time and quality his involvement entails. Fagan and Barnett (2003) found it useful to study involvement in the context of ideology, quality or satisfaction with the involvement given, and the amount of father involvement given. Palkovitz (1997) also suggests 118 ways fathers can be involved in the lives of their children.

CONCLUSION

In our study, how involved fathers become is a balancing act between mothers' expectations (that sometimes encourage traditional roles) and fathers' expectations (that appear to promote egalitarian forms of involvement). The results support Walker and McGraw's (2000) assertion that fathers also play a significant role in their own involvement; future studies should include both parents' reports when examining factors that limit father involvement. Measuring personal expectations instead of (or in addition to) gender-role ideology is likely to provide unique insight into how thought processes can influence father involvement. The practical utility of these findings is supported by the classification functions listed in the results section of this paper. Each of the four functions (or models) accurately predicts low paternal involvement between 70% and 81% of the time. This is particularly important because intervention and support services are typically geared toward situations where father involvement is less than adequate.

Table 1
Descriptive Statistics for Father Involvement

                          Wives'             Husbands'
                       expectations        expectations

                      M         SD        M         SD

Feeds child           7.14      4.85     10.11      7.94
Changes clothes       5.58      2.79      7.77      4.93
Bathes child          2.78      1.81      4.00      3.38
Plays with child     14.30      8.67     16.40     13.57
Reads to child        4.86      3.27      6.39      4.58

                          Wives              Husbands
                          report              report
                        involvement         involvement

                      M         SD        M         SD

Feeds child           6.15      6.94      6.50      7.42
Changes clothes       4.40      2.69      5.20      2.85
Bathes child          1.79      1.61      1.58      1.27
Plays with child     14.30      9.44     17.46     14.15
Reads to child        2.25      2.67      2.32      2.50

Table 2
Factor Analysis--Mothers' Reports of Fathers' Involvement

Mothers' reports         Factor I           Factor II
                      (Instrumental)       (Affective)

Bathes child                .82               -.17
Changes clothes             .74                .09
Feeds child                 .56                .32
Plays with child            .13                .83
Reads to child             -.01                .80

Table 3
Factor Analysis--Fathers' Reports of Fathers' Involvement

Mothers' reports         Factor I           Factor II
                      (Instrumental)       (Affective)

Bathes child                .84               -.13
Changes clothes             .66                .30
Feeds child                 .34                .72
Plays with child            .07                .82
Reads to child             -.07                .57

Table 4
Fathers Report their Instrumental Involvement with Infants

                                    Low involvement   High involvement
                                       (n = 37)           (n = 22)

                            SDC       M       SD         M       SD

Mother's affective exp      -.73    20.68    10.82     16.86     8.25
Father's affective exp       .53    18.11    10.51     28.14    18.34
Father's instrumental exp    .51    18.38     7.93     25.41    14.20
Mother's instrumental exp    .35    14.68     7.10     16.18     7.80
Father's work hours         -.26     5.86     1.40      5.45     1.65
Father's age                 .21    27.11     6.24     28.41     7.24

Notes. "Exp" is an abbreviation for "Expectations."

Canonical Correlation = .50.

Wilk's = .748.

Chi Square (N = 59, 6) = 15.66, p < .05.

Table 5
Fathers Report their Affective Involvement with Infants

                                    Low involvement   High involvement
                                        (n = 37)          (n = 20)

                             SDC      M       SD        M        SD

Father's affective exp       .54    16.86    8.02      31.25    19.66
Father's instrumental exp    .52    17.89    7.70      27.25    14.15
Father's work hours         -.48     6.05    1.25       5.10     1.74
Mother's affective exp       .25    17.92    7.52      22.05    13.38
Father's age                -.08    27.70    6.87      26.35     5.49
Mother's instrumental exp   -.07    14.70    6.04      27.25    14.15

Note. "Exp" is an abbreviation for "Expectations."

Canonical Correlation = .58.

Wilk's = .660.

Chi Square (N = 57, 6) = 21.60, p < .05.

Table 6
Mothers Report Fathers' Instrumental Involvement with Infants

                                    Low involvement   High involvement
                                        (n = 35)          (n = 26)

                            SDC       M       SD         M       SD

Mother's affective exp      -.59    19.94     9.33     18.07    10.68
Mother's instrumental exp    .53    14.03     6.67     17.50     7.92
Father's instrumental exp    .48    17.77     7.16     25.73    14.21
Father's affective exp       .47    17.94    10.08     27.73    18.02
Mother's work hours          .23     3.40     2.43      4.31     2.11
Mother's age                 .12    24.40     4.60     25.65     5.68

Notes. "Exp" is an abbreviation for "Expectations."

Canonical Correlation = .51.

Wilk's = .742.

Chi Square (N = 61, 6) = 16.71, p < .05.

Table 7
Mothers Report Fathers' Affective Involvement Infants

                                    Low involvement   High involvement
                                       (n = 33)           (n = 27)

                             SDC      M       SD         M       SD

Father's instrumental exp    .75    18.27     8.66     24.11    13.27
Mother's age                -.30    25.33     5.76     24.56     4.25
Father's affective exp       .29    19.42     9.51     24.74    18.98
Mother's work hours         -.21     3.82     2.53      3.67     2.09
Mother's affective exp       .20    18.06    10.92     20.41     8.68
Mother's instrumental exp    .03    14.94     7.13     16.03     7.82

Note. "Exp" is an abbreviation for "Expectations."

Canonical Correlation = .30.

Wilk's = .911.

Chi Square (N = 60, 6) = 5.14, p > .05.

Table 8
Classification Results

Reports of involvement              Model's prediction
                                        accuracy

                           Predicting          Predicting
                         low involvement    high involvement

Fathers' reports
  Instrumental                81.1%              77.3%
  Affective                   86.5%              80.0%

Mothers' reports
  Instrumental                77.1%              76.9%
  Affective                   69.7%              55.6%

REFERENCES

Allen, S.M., & Hawkins, A.J. (1999). Maternal gatekeeping: Mothers' beliefs and behaviors that inhibit greater father involvement in family work. Journal of Marriage and Family, 61, 199-212.

Barclay, L., & Lupton, D. (1999). The experiences of new fatherhood: A sociocultural analysis. Journal of Advanced Nursing, 29, 1013-1020.

Beitel, A.H., & Parke, R.D. (1998). Paternal involvement in infancy: The role of maternal and paternal attitudes. Journal of Family Psychology, 12(2), 268-288.

Bonney, J.F., Kelley, M.L., & Levant, R.F. (1999). A model of fathers' behavioral involvement in child care in dual-earner families. Journal of Family Psychology, 13(3), 401-415.

Bulanda, R.E. (2004). Paternal involvement with children: The influence of gender ideologies. Journal of Marriage and Family, 66, 40-45.

Cowan, C.P., & Cowan, P.A. (1992). When partners become parents: The big life change for couples. Mahwah, NJ: Erlbaum.

Cowan, P.A., & Cowan, C.P. (1998). New families: Modem couples as new pioneers. In M.A. Mason, A. Skolnick, & S.D. Sugarman (Eds.), All our families: New policies for a new century (pp. 169-192). New York: Oxford University Press.

Fagan, J., & Barnett, M. (2003). The relationship between maternal gatekeeping, paternal competence, mothers' attitudes about the father role, and father involvement. Journal of Family Issues, 24, 1020-1043.

Greenstein, T.N. (1996). Husbands' participation in domestic labor: Interactive effects of wives' and husbands' gender ideologies. Journal of Marriage and Family, 58, 585-595.

Hawkins, A.J., & Dollahite, D.C. (Eds). (1997). Generative fathering: Beyond deficit perspectives. Thousand Oaks, CA: Sage.

Hoffman, C., & Moon, M. (1999). Women's characteristics and gender role attitudes: Support for father involvement with children. Journal of Genetic Psychology, 160, 411-418.

Hooker, K., Fise, B.H., Jenkins, L., Morfei, M.Z., & Schagler, J. (1996). Possible selves among parents of infants and preschoolers. Developmental Psychology, 32, 542-550.

LaRossa, R. (1997). The modernization of fatherhood. Chicago: University of Chicago Press.

Litton-Fox, G., Bruce, C., & Combs-Orme, T. (2000). Parenting expectations and concerns of fathers and mothers of newborn infants. Family Relations, 49, 123-131.

Marsiglio, W. (1991). Paternal engagement activities with minor children. Journal of Marriage and Family, 53, 973-986.

McBride, B.A., & Darragh, J. (1995). Interpreting the data on father involvement: Implications for parenting programs for men. Families in Society, 76, 490-497.

McHale, S.M., & Huston, T.L. (1984). Men and women as parents: Sex role orientations, employment, and parental roles with infants. Child Development, 55, 1349-1361.

Palkovitz, R. (1997). Reconstructing "involvement": Expanding conceptualization of men's caring in contemporary families. In A.J. Hawkins & D.C. Dollahite (Eds.), Generative fathering: Beyond deficit perspectives (pp. 200-216). Thousand Oaks, CA: Sage.

Pleck, J.H. (1985). Working wives/working husbands. Beverly Hills, CA: Sage.

Sandelowski, M. (1994). Separate, but less unequal: Fetal ultrasonography and the transformation of expectant mother/fatherhood. Gender and Society, 8, 230-245.

Shapiro, J.L. (1987). The expectant father. Psychology Today, 21, 36-42.

Strauss, R., & Goldberg, W.A. (1999). Self and possible selves during the transition to fatherhood. Journal of Family Psychology, 13, 244-259.

Walker, A.J., & McGraw, L.A. (2000). Who is responsible for responsible fathering? Journal of Marriage and Family, 62, 563-569.

Wille, D.E. (1995). The 1990s: Gender differences in parenting roles. Sex Roles, 33, 803-817.

Correspondence concerning this article should be addressed to Jerry L. Cook, Department of Family and Consumer Sciences, California State University, Sacramento, 6000 J Street, Sacramento, CA 95819-6053. Electronic mail: [e-mail].