Paternal Identity, Maternal Gatekeeping, and Father Involvement*

Paternal Identity, MaternalGatekeeping, and Father

Involvement*

Brent A. McBride Geoffrey L. Brown Kelly K. Bost Nana Shin Brian Vaughn Byran Korth**

Abstract: The purpose of this study was to examine whether mothers’ beliefs about the role of the father may con-tribute to mothers influencing the quantity of father involvement in their children’s lives. Participants were 30 two-parent families with children between the ages of 2 and 3 years. A combination of self-report and interview datawere collected from both mothers and fathers. Results from multiple regression analyses indicated that fathers’ per-ceived investments in their parental roles and actual levels of paternal involvement are moderated by mothers’ beliefsabout the role of the father. Findings are discussed in terms of implications for future research on parenting identityand maternal gatekeeping as well as the development of parenting programs for fathers.

Key Words: father involvement, maternal gatekeeping, mothers, paternal identity.

Background and Significance

Although children’s social, emotional, and academicoutcomes have long been of primary interest toresearchers, studies on the influence of parenting havefocused historically on the quality and quantity ofmaternal care. Fathers were identified traditionally asthe ‘‘breadwinners’’ of the family, whereas motherswere primarily responsible for caregiving. Thus, thetraditional family structure has not always encouragedfathers to become involved parents. Consequently,many early conceptualizations of father involvementcentered on the idea of ‘‘deficit’’ or ‘‘inadequate’’fathering (Hawkins & Dollahite, 1997).

However, in recent years, American society hasbegun to see a shift in the expectations for parental

roles. Some evidence indicates that this historicalchange in the ‘‘culture of fatherhood’’ (Larossa,1988) may also be reflected in the conduct of mod-ern fathers. More specifically, fathers are slowlybeginning to increase the amount of time they spendcaring for their children, and fathers today are morevisibly involved with their children than were fathersof past generations (Pleck & Masciadrelli, 2004).Accompanying this shift has been an increase inresearch attempting to measure, conceptualize, andexplain father involvement. Although the exacteffects of increased father involvement on familydynamics remains to be seen, the emerging evidencesuggests that fathers’ increased caregiving of childrenoften has positive outcomes for families (Pleck &Masciadrelli).

*This research was supported by grants from the National Science Foundation (BCS-0126427, K.K.B. and B.A.M. BCS-0126163, B.V. and B.K.). Opinions reflect

those of the authors and do not necessarily reflect those of the granting agency. The authors would like to express their appreciation to Cathy Inman, Paul King, and

Nikata Ligatum for their assistance with data collection.

**Brent A.McBride is a Professor in the Department of Human and Community Development, University of Illinois, 1105West Nevada, Urbana, IL 61801 (brentmcb@

uiuc.edu).Geoffrey L. Brown is a Research Assistant at theUniversity of Illinois atUrbana-Champaign. Kelly K. Bost is anAssociate Professor in theDepartment ofHuman

and Community Development at the University of Illinois at Urbana-Champaign. Nana Shin is a Research Assistant at the University of Illinois at Urbana-Champaign.

Brian Vaughn is a Professor in the Department of HumanDevelopment and Family Studies at the Auburn University. Byran Korth is an Assistant Professor in the Depart-

ment of Human Development and Family Studies at the Auburn University.

Family Relations, 54 (July 2005), 360–372. Blackwell Publishing. Printed in the USA.Copyright 2005 by the National Council on Family Relations.

Conceptualizations of FatherInvolvement

It is important to acknowledge that father involve-ment is multifaceted and thus can be defined andconceptualized in multiple ways (e.g., Day & Lamb,2004; Hawkins & Dollahite, 1997). For example,scholars have examined a wide range of facetsof father involvement from a variety of per-spectives including cognitive and affective aspects(Palkovitz, 1997), generative fathering (Hawkins &Dollahite), social constructionist (Marsiglio, Amato,Day, & Lamb, 2000), and social capital perspectives(Amato, 1998). In addition, there has been a recentcall for increased attention to the quality of fathers’involvement with their children as well as the quan-tity of involvement within the father involvementliterature (Pleck & Masciadrelli, 2004; Hawkins &Dollahite).

Lamb and his colleagues (Lamb, Pleck, Charnov,& Levine, 1987) have proposed a three-part modelof paternal involvement that encompasses the vari-ous forms of participation that fathers may take intheir children’s lives. The categories consist of (a)interaction—the father interacting one-on-one withhis child, (b) accessibility—the father being physi-cally and or psychologically available to his child,and (c) responsibility—the father assuming respon-sibility for his child’s welfare and care. This modelhas been influential in guiding much of the earlyresearch on father involvement and continues to bean important basis for scholarship on paternal roles(Pleck & Stueve, 2001). We utilize the Lamb et al.multidimensional model as a guide for the concep-tualization of father involvement in the presentstudy. The Lamb et al. conceptualization providesa useful way to examine aspects of father involve-ment that are pertinent to the present study andinforms our understanding relative to why somefathers choose to take a more active caregiving rolein raising their children rather than a traditionalapproach centered only on financial provision.

Antecedents of Father Involvement

Fathering Identity

Given changing societal expectations of paternal roles,it has become increasingly important to understand

the antecedents that may lead to variations in fatherinvolvement. One key influence on father involve-ment seems to be the emphasis that a father places onhis role as a parent. Identity theory posits that fatherinvolvement will vary as a function of the saliencewith which a father views his parenting role. Indeed,as Pleck (1997) notes, ‘‘parent-role identity’’ hasbecome one of the central constructs for defining theconcept of father involvement. Many researchers haveconceptualized father involvement as a direct result ofthe father’s investments in his fathering identity.

Parke (2002),Tamis-LeMondaandCabrera (1999),and others have argued that parental-role identity isa particularly important antecedent of father involve-ment because paternal behavior is somewhat discre-tionary and less scripted by societal norms thanmaternal behavior. Given this flexibility, it standsto reason that a father’s investment in his parentingidentity will be a particularly powerful determinantof elements of father involvement. In fact, a growingbody of research drawing from an identity theoryperspective suggests that fathers behave in ways thatreflect their role investments (e.g., Fox & Bruce,2001; Maurer, Pleck, & Rane, 2001; Rane &McBride, 2000).

Nonetheless, this fathering identity is neither staticnor immune to social and contextual forces. Clearly,a father’s role investments can change dramaticallyfollowing a major transition such as divorce (Ihinger-Tallman, Pasley, & Beuhler, 1993) or incarceration(Arditti, Acock, & Day, 2005). Indeed, some re-searchers have emphasized the dynamic, fluid natureof identity for fathers relative to life transitions andsocial context (e.g., Fox & Bruce, 2001).

Maternal Influences

One of the most obvious aspects of this social contextis the family itself. More specifically, a father’s iden-tity hierarchy (i.e., his relative commitment to thevarious roles in his life) is quite likely to be affectedby the views and attitudes of his partner. Researchsupports this claim by showing that marital andcoparental relationship can directly influence paternalidentity (Rane & McBride, 2000) and perhaps evenmoderate the link between identity and fatherinvolvement (Ihinger-Tallman et al., 1993). In par-ticular, fathering identity appears to be influencedheavily by perceptions of maternal appraisals (Maureret al., 2001). All of this evidence suggests thatone’s identity as a father is clearly affected by the

Maternal Influences on Father Involvement � McBride et al. 361

personality, attitudes, and behaviors of his partner.Thus, as Rane and McBride suggest, it is importantto examine the status and role hierarchies of the father’spartner when attempting to understand the nature anddevelopment of fathers’ own identities.

In addition to influencing paternal identities,mothers may exert direct control over fatheringbehavior. The idea that paternal involvement isregulated by maternal characteristics is referred toas ‘‘maternal gatekeeping’’ and can be defined as(a) a reluctance to relinquish family responsibilityby setting rigid standards, (b) a desire to validate amaternal identity, and (c) differentiated concep-tions of family roles (Allen & Hawkins, 1999).Moreover, the study of Allen and Hawkins revealeda reliable group of gatekeeping mothers that tendedto rate highly on all three dimensions and reporteddoing more domestic labor and less equitable ar-rangements. A number of other researchers havesought to theoretically and empirically validate thematernal gatekeeping hypothesis by examining thedirect and indirect effects of maternal variables onfather involvement. For example, Ihinger-Tallmanet al. (1993) proposed that maternal characteristics(including maternal preferences, beliefs, encourage-ment, and emotional stability) influenced fatherinvolvement after divorce. However, this propositionis not limited to divorced families, and its presencehas been demonstrated in several different contexts.Hoffman and Moon (1999), for example, demon-strated that women’s personal characteristics ac-counted for a significant amount of the variance infather involvement scores. Nontraditional genderrole attitudes, positive ratings of interpersonal trust,and low hostility toward men each contributedsignificantly to the prediction of women’s supportfor father involvement with children.

In addition towomen’s personal characteristics,DeLuccie (1995) found that two ‘‘maternal mediators,’’importance of and satisfaction with father involve-ment, explained mothers’ reports of the frequency offather involvement. McBride and Rane (1997) pro-vide corroborating evidence with the finding thatmothers’ perceptions of the paternal role were the bestpredictors of father involvement (even considerablybetter than fathers’ own perceptions of this paternalrole). It seems entirely feasible that individual differ-ences in mothers’ parenting attitudes may cause themto impede or augment fathers’ access to their children.

The findings of Grossman, Pollack, and Golding(1988) further support the idea that maternal

influences impact fathering by showing that differ-ent predictors accounted for different types of fatherinvolvement. Men’s psychological characteristicswere the best predictors of the amount of play-time and quality of father-child interaction, whereaswomen’s characteristics (autonomy, occupation, andage) were the best predictors of men’s overall quan-tity of involvement and time spent caregiving.Grossman et al. propose that two separate gatekeep-ing mechanisms are at work: (a) wives allowing (ornot allowing) their husbands to function indepen-dently as fathers to their children and (b) wivesallowing (or not allowing) their husbands to modelor learn from them how to become a good parent.Thus, a more differentiated view of maternal influ-ences on father involvement (e.g., making the dis-tinction between play and caregiving) suggests thatthe role of maternal gatekeeping varies across con-texts and may not be equally relevant for all domainsof fatherhood (Beitel & Parke, 1998).

In sum, research indicates that maternal influen-ces, particularly their attitudes and beliefs aboutfathering, are important antecedents of fatherinvolvement (e.g., Allen & Hawkins, 1999; Beitel &Parke, 1998; De Luccie, 1995; McBride & Rane,1997). Further exploration of maternal gatekeepingis needed to elaborate upon current evidence as thedetails of this linkage are not yet specified empiri-cally (Pleck, 1997).

The Present Study

The present study builds on existing research byexamining the ways in which mothers may beinstrumental in determining how fathers approachand fulfill their parenting role. We were particularlyinterested in learning more about the extent towhich maternal beliefs and perceptions, particularlymothers’ beliefs about the role of the father, maycontribute to mothers serving as gatekeepers tofather involvement. In doing so, we adopt thenotion of Allen and Hawkins (1999) that gatekeep-ing is ‘‘a collection of beliefs and behaviors that ulti-mately inhibit a collaborative effort between menand women in family work’’ (p. 200). That is,although the current study does not measure gate-keeping ‘‘behavior’’ per se, it instead focuses onmater-nal beliefs about the role of the father as a potentialgatekeeping mechanism. Allen and Hawkins argue

Family Relations � Volume 54, Number 3 � July 2005362

that, by definition, maternal beliefs are an essentialpiece of the gatekeeping construct.Within this frame-work, mothers’ attitudes regarding fathers’ role inchildrearing are considered particularly integral totheir regulation of father involvement. Indeed, empir-ical work conceptualizing gatekeeping as mothers’beliefs about fathering has found links between a vari-ety of nontraditional attitudes about the role of thefather and decreased father involvement (Beitel &Parke, 1998; De Luccie, 1995). Thus, for purposesof this study, gatekeeping serves as a useful theoreticalconstruct for exploring the relations betweenmothers’attitudes toward the fathering role and fathers’involvement in their children’s lives.

Palkovitz (2002) noted that men engage in a vari-ety of roles that may encompass the notion of ‘‘goodfathering,’’ and that these roles are largely sociallyconstructed. Accordingly, beliefs about the father’srole (both from fathers themselves and external sour-ces) may be an important predictor of variability infather involvement across multiple contexts(Palkovitz, 2002). Indeed, Palkovitz (1984) foundthat mothers’ beliefs in the importance of thefather’s role predicted greater paternal involvementin both dyadic and triadic interactions. Thus,mothers’ attitudes toward the paternal role mayexert direct effects on some aspects of father involve-ment and indirect effects via gatekeeping.

Indeed, Fagan and Barnett (2003) make a distinc-tion between the constructs of gatekeeping andmothers’ perceptions of the paternal role and ulti-mately conclude that gatekeeping behavior mediatesthe relationship between father competence and fatherinvolvement. However, it may be that maternalbeliefs play an equally important role by regulatingthe association between fathers’ perceptions of theirparenting investment and their actual level of involve-ment. The present study extends previous gatekeepingresearch by exploring the associations between invest-ment in the fathering role and father involvement andby conceptualizing maternal beliefs as a potentialmoderator to this relationship. Maternal gatekeepingmay be manifested as phenomena that either encour-ages or discourages fathers from acting on their pater-nal identity. The current investigation explores thehypothesis that maternal beliefs regarding the role ofthe father may be one such form of interference.

This investigation also has the advantage of utiliz-ing a broad conceptualization of father involvementand data from multiple sources, thus overcominga major limitation of previous research (see for

example, Fagan & Barnett, 2003). Walker andMcGraw (2000) suggest that research examining thematernal gatekeeping phenomenon should includea measure of fathers’ interest in the paternal role. Ourstudy addresses this issue by looking at both mothers’and fathers’ perceptions of the fathers’ role as well astheir perceptions of how invested the fathers are inthe parenting role. In doing so, we hope to gobeyond using gatekeeping as a way of depictingmothers as solely responsible for controlling whatfathers do. Instead, our goal is to further examine thepotentially complex relations between the beliefs andperceptions of both mothers and fathers and howthese factors may influence paternal involvement.

Because previous research has shown relativelyfew straightforward associations between parentingidentity and parenting behavior (see Maurer et al.,2001, for a review), we did not believe that fathersin this study would necessarily show a link betweenparenting investment and involvement. On the con-trary, we expected maternal perceptions of fathers’investment in the parenting role to influence fathers’parental identities (e.g., Rane & McBride, 2000)directly and to be linked with greater father involve-ment (i.e., McBride & Rane, 1997). Thus, wehypothesized that mothers’ perceptions of their hus-bands’ investment in the parenting role will be posi-tively associated with father involvement, whereasfathers’ own perceived investment in this role willnot show these main effects. Instead, we hypothe-sized that the relation between fathers’ perceivedinvestment in parenting and their own levels ofinvolvement would be moderated by mothers’ be-liefs in the role of the father. Thus, the link betweenfathers’ investment in parenting and fathers’ involve-ment in parenting will depend upon mothers’ (butnot fathers’ own) attitudes toward the role of thefather. Results from this study have implications foridentity theory, the maternal gatekeeping hypothe-sis, and the refinement of parenting programs aimedat augmenting father involvement.

Method

Participants

Data for this study are being drawn from the firstyear of a longitudinal project focused on exam-ining the impact of parent-child relationships on

Maternal Influences on Father Involvement � McBride et al. 363

children’s development. Families were recruited pri-marily through local day care centers and also via fli-ers placed in local community agencies, newspapers,and grocery stores. Criteria for inclusion in the studyconsisted of the child being 2 years old, both parentsliving in the home with the child, and the childbeing enrolled in some day care/preschool pro-gram. The study consisted of 30 children and theirmarried parents. Twenty-eight families consisted ofboth biological parents, one family consisted of bothadoptive parents, and one family consisted of a bio-logical mother and a stepfather. Mean ages forthe fathers, mothers, and children at the time ofdata collection were 38.17 (SD ¼ 6.94) years, 34.10(SD ¼ 5.15) years, and 31.30 (SD ¼ 6.09) months,respectively. Fifty-seven percent of the target chil-dren were boys and 43% were girls. Fifty percent ofthe families had one child, 30% had two children,17% had three children, and 3% had more thanthree children. Families with combined incomes lessthan $10,000 comprised 3.3% of the sample, 3.3%of the families had incomes between $10,000 and$20,000, 3.3% had incomes between $20,000and $30,000, 20% had incomes between $30,000and $40,000, 3.3% had incomes between$40,000 and $50,000, 10% had incomes between$50,000 and $60,000, and 56.7% had incomesgreater than $60,000. Ninety percent of the fathersand 86.7% of the mothers had a bachelors degree orbeyond. All fathers and 94.4% of the mothers wereemployed outside the home, with both groups work-ing an average of 43.47 and 33.09 hr per week,respectively.

Procedures

A combination of self-report and interview data wascollected for this study. Upon agreeing to partici-pate, families were scheduled for a home visit bytwo research assistants. During this visit, a Time-Diary interview protocol was used to measure inter-action and accessibility (temporal) forms of parentalinvolvement. Mothers and fathers were interviewed(by a female and male research assistant, respec-tively) simultaneously in separate rooms. Both par-ents also then completed a penny-sort interview taskto assess how participants perceived their investmentin various adult roles.

A packet of questionnaires was left with the fam-ily to be completed and turned in at the next phaseof data collection (usually several weeks later).

Mothers and fathers each individually completedidentical measures designed to assess parental respon-sibility, involvement, perceptions of the parentingalliance, and the role of the father in childrearing.

Measures

Involvement variables. The Interaction/Accessi-bility Time-Diary interview protocol (McBride &Mills, 1993) was used to measure interaction andaccessibility forms of involvement. Data were col-lected during these interviews with each parent indi-vidually using a forced-recall technique for the mostrecent workday and nonworkday prior to the inter-view. For those parents who were not employed out-side the home, nonworkday data were collected ondays that matched their partners’ most recent non-workdays. This approach was used based on previ-ous research indicating that single-earner familiestend to adjust their daily routines on days when theemployed parent is not at work. For the target days,each parent was asked to recount his or her activitiesin great detail (15-min intervals) from the time theywoke up in the morning until the time they went tosleep at night. Prompts and cues from the inter-viewer allowed the parents to elaborate upon theexact length and nature of activities as well as tospecify who was engaged in these activities withthem or otherwise present at the time.

All interviews were audiotaped and later ana-lyzed. Data collected were categorized as (a) interac-tion, (b) accessibility, or (c) no involvement at all.Interaction consisted of activities in which both theparent and child were directly engaged. The Accessi-bility category additionally encompassed activities inwhich parents were available to the target child, eventhough not necessarily engaged with them. Thismay include being in the same room as one’s child,but also such things as being available to respond tocalls from the child’s teacher or doctor. By defini-tion, interaction forms of involvement were codedas accessibility as well (i.e., a parent must be accessi-ble to a child in order to interact with him or her).The final interaction score was the total number ofminutes the parent interacted with the child on theworkday and nonworkday combined. Interactionsubscale scores for the workday and nonworkdayalso were computed. An accessibility total score andsubscale scores were computed in a similar fashion.

An adapted version of the Parental ResponsibilityScale (PRS) developed by McBride and Mills (1993)

Family Relations � Volume 54, Number 3 � July 2005364

was used to measure responsibility forms of parentalinvolvement. This scale lists 14 common childcaretasks in which parents of 2 year olds typically partici-pate (e.g., supervising the child’s personal hygiene,making baby-sitting arrangements, and selectingappropriate clothes for the child to wear).Mothers andfathers completed this instrument together (a ¼ .73)and responded by designating who had primaryresponsibility for each task using a 5-point scale rangingfrom 1¼mother always responsible to 5¼ father alwaysresponsible. Responsibility was defined for the parentsas remembering, planning, and scheduling the task.

Z scores were computed on the Time-Diaryinterview data and the Parental Responsibility Scalefor fathers so that each component of father involve-ment would be equally weighted. These Z scoreswere combined to provide a composite measure oftotal father involvement for participant families andthen used in subsequent analyses. Such an approachis in response to recent calls by scholars for examin-ing father involvement as both a global constructand one consisting of several dimensions, each witha separate set of antecedents and consequences (Day& Lamb, 2004; Pleck & Masciadrelli, 2004;Schoppe, McBride & Ho, 2004).

Perceived role investment. The Role InvestmentsPenny-Sort Task interview protocol (McBride &Rane, 1997) was used to assess how participantsperceive their own, and their spouse’s, investmentsin five specific adult roles (parent, spouse, worker,social, and other). Because identity theory posits thata father’s identity is composed of multiple roles, thismeasure (for fathers) serves as a proxy for paternalidentity by tapping into their perceived investmentin several domains (e.g., Rane & McBride, 2000).During this interview protocol, parents were indi-vidually asked to reflect on their experiences asadults. As part of this reflective process, the fiveadult roles of interest for this study were presentedon cards and defined. Social roles were defined asrelationships with close personal friends and ormembership in social or community-based organiza-tions (e.g., Knights of Columbus, Kiwanis, JuniorLeague, and bowling leagues). Participants werethen asked to sort 15 pennies into the five categoriesbased on how they perceived to invest themselves ineach role. Before participants sorted the pennies, theinterviewer defined role investments as psychologicaland emotional investments in the roles (e.g., fromwhere do you draw your identity?) as opposed toactual time spent functioning in the roles. After

sorting the pennies, the interviewer probed partici-pants to confirm their responses and to identify rea-sons for using the ‘‘other’’ category (if used). Again,it is important to note that the participants were in-structed to divide the pennies based on how theyinvest themselves, not necessarily in relation to time,but based on how they make commitments to eachof the five areas. For the current study, fathers’ per-ceived investment in the parental role was used asa predictor variable in the analyses. This type of pro-cedure has been used in other studies of family inter-action (e.g., Cowan & Cowan, 2000). Moreover,this particular measure has shown predictive validityin previous investigations of parenting identity andfather involvement (e.g., McBride & Rane).

Beliefs about the role of the father. An adapted ver-sion of the Role of the Father Questionnaire (ROFQ)was used to measure parental attitudes toward thepaternal role. The modified ROFQ (Palkovitz,1984) is a 14-item scale designed to determine theextent to which a parent believes the fathers’ role isimportant in child development. Both mothers (a ¼.83) and fathers (a ¼ .58) indicated a level of agree-ment or disagreement along a 5-point scale foreach item (1 ¼ strongly agree, 5 ¼ strongly disagree).Responses were coded such that higher total scoreson this measure are reflective of attitudes that fathersare capable of and should demonstrate involvementwith their children. Mothers’ responses to the ROFQwere believed to be reflective of ‘‘gatekeeping’’ interms of maternal attitudes that might encourage ordiscourage father involvement. Originally intendedfor use with parents of infants, the ROFQwas adaptedto be reflective of attitudes about father involvementwith preschool-aged children. Sample items fromthe ROFQ such as, ‘‘It is as important for a fatherto meet a young child’s psychological needs as it isfor the mother to do so’’ and ‘‘it is essential for thechild’s well-being that fathers spend time interactingand playing with their children’’ highlight the utilityof this measure for tapping into beliefs about thepaternal role.

Results

Descriptive Analyses

Means, standard deviations, and bivariate correla-tions for all measures are reported in Table 1. The

Maternal Influences on Father Involvement � McBride et al. 365

Table1.Descriptive

StatisticsandCorrelations

Variables

12

34

56

78

910

11

12

1.PaternalRolePerception

(theRoleof

the

FatherQuestion

naireforfathers)

—

2.PaternalRolePerception

(theRoleof

the

FatherQuestion

naireformothers)

.41*

—

3.Perception

ofSelfAsaParent

(Penny-SortTaskforthefathers)

.31

.24

—

4.Perception

ofSpouse’srole

commitmentasaparent(Penny-SortTask

forthemothers)

.07

.19

.20

—

5.FatherInvolvem

ent(interaction

+accessibility+respon

sibility)

.07

.18

.15

.42*

—

6.Interaction

(fathers’Tim

e-Diary)

.05

.23

.17

.29

.87***

—

7.Interactionworkingday

(fathers’Tim

e-Diary)

2.21

.35

.26

.36

.59**

.65***

—

8.Interactionnon

workingday

(fathers’Tim

e-Diary)

.21

.06

.04

.13

.73***

.85***

.16

—

9.Accessibility(fathers’Tim

e-Diary)

2.14

.272.14

.01

.82***

.66***

.48**

.53**

—

10.Accessibilityworkingday

(fathers’Tim

e-Diary)

2.20

.28

.18

.15

.55**

.44*

.78***

.04

.65***

—

11.Accessibilitynon

workingday

(fathers’Tim

e-Diary)

2.06

.192.29

2.08

.66***

.57**

.13

.65***

.88***

.21

—

12.Respon

sibility(fathers’Parental

Respon

sibilityScale—

part1)

.20

.23

.48**

.25

.77***

.47*

.41*

.33

.40*

.35

.26

—

M4.28

4.39

4.55

4.63

.05

314.42

184.00

444.83

391.50

230.00

553.00

2.51

SD.36

.53

.48

1.16

.78

84.36

88.78

130.41

118.22

113.37

184.63

.35

*p,

.05.**p,

.01.***p,

.001.

Family Relations � Volume 54, Number 3 � July 2005366

correlational data yielded several significant associa-tions, one of which was particularly relevant to theexploration of maternal influences on fathering. Spe-cifically, mothers’ Penny-Sort Task indicating theirperceptions of fathers’ relative investment in parent-ing was significantly related to the father involve-ment composite score (r ¼ .42, p , .05). Incontrast, fathers’ Penny-Sort Task was not signifi-cantly related to the father involvement compositescore, although this variable was significantly associ-ated with fathers’ scores on the PRS (r ¼ .48, p ,

.01). This suggests that fathers who are moreinvolved in their children’s lives have wives who seethem as more invested in the parenting role, eventhough these involved fathers do not necessarily seethemselves as highly invested in their fathering roles.Although these results indicated associations be-tween father involvement and maternal percep-tions of their husbands’ investment in the fatheringrole, there were no significant relationships be-tween father’s own commitment to parenting andhis actual level of interaction, accessibility, or com-posite forms of involvement.

Moderating Effects

The thrust of our study was testing the hypothesizedmoderating effect of maternal perceptions on fatherinvolvement. A series of hierarchical linear regressionswas conducted on the dependent variables (accessibil-ity, interaction, responsibility, and the compositefather involvement variable) to examine the extent towhich fathers’ perceptions of their investment in theparental role and mothers’ beliefs about the role ofthe father independently and conjointly were associ-ated with father involvement.

For interaction terms, variables were centered toreduce multicollinearity. Here, centering involvedsubtracting the mean from a variable, leaving devia-tion scores. Interaction terms were then computedfrom those centered variables. Working hours ofmothers and fathers were included in the first blockas control variables. Hours of maternal employmentwas entered as a control in light of previous research,suggesting a positive relationship between mothers’work hours, attitudes toward fathering, and fatherinvolvement (Pleck, 1997), and that workingmothers with less traditional perspectives towardparental roles are more likely to encourage fatherinvolvement (Barnett & Baruch, 1988). Hours ofpaternal employment was also controlled because

identity is proposed to influence individual behavior‘‘given freedom of choice’’ (Stryker, 1987). Thenumber of hours a father works determines, to animportant degree, the amount of time he will haveavailable to be involved directly with his children.Further, both hours of maternal and paternalemployment were significantly correlated with totalfather involvement in the present sample. No otherdemographic variables (including mother or fathereducation or child gender) were significantly corre-lated with total father involvement; hence, they werenot controlled in the analyses.

The second step included mothers’ and fathers’perceptions of the fathers’ investment in the parent-ing role (Penny-Sort task) and attitudes toward thepaternal role (ROFQ). Interaction terms wereentered in the final step of the regression. At eachstep, the significant change in R 2 was assessed to de-termine the contribution of each block of variables.Interaction terms were further analyzed if they weresignificant and if they added a significant incrementto the variance accounted for by the equation.

We looked for the presence of interaction termsusing the composite father involvement score as thedependent variable, as well as its component partsof interaction, accessibility, and responsibility. Aftercontrolling for maternal and paternal workinghours, there was one significant interaction (seeTable 2). Father’s perception of himself as a parent(Penny-Sort task) and mother’s beliefs about the roleof the father (ROFQ) did not combine to accountfor a significant portion of the variance in fatheraccessibility. However, the interaction between thesevariables was indeed significant (DR2 ¼ .18; F ¼7.49, p , .05).

To clarify the significant interactions betweenfather’s perception of his parenting commitmentand mother’s beliefs about the role of the father, thepost hoc plotting procedure described in Aiken andWest (1991) was used. Following this procedure, wetested the effects on the father accessibility of father’sperception of his parenting commitment at differentlevels of mother’s beliefs about the role of the father(Z). Separate regression lines were computed andplotted for individuals 1 standard deviation belowthe mean on predictor Z, at the mean of predictorZ, and 1 standard deviation above the mean of thepredictor Z.

Results from interactions highlighted in theseanalyses indicated that father’s perception of himselfas a highly committed parent was associated with

Maternal Influences on Father Involvement � McBride et al. 367

higher father accessibility whenmother’s beliefs aboutthe role of the father were greater. However, whenfathers perceived themselves as a less committed par-ent, mother’s beliefs about the role of the father didnot play a significant role in father involvement. ThisFather’s Self-Perceived Commitment to Parenting �Mother’s Beliefs About the Role of the Father interac-tion indicates that father’s perception of himself asa highly committed parent is associated with higherfather accessibility only when his wife believes thatfathers should play a significant role in parenting.On the other hand, father’s perception of his parent-ing commitment is not significantly related to fatherinvolvement when mothers report more traditionalbeliefs about the role of the father.

We next looked for the presence of this interac-tion when breaking up the accessibility variable intoworking and nonworking days. After controlling forparental working hours, the interaction effect re-mained when parenting accessibility during non-working days was the dependent variable (DR2 ¼.22; F ¼ 8.67, p , .01) (see Table 2). This indicates

that the significant interaction between father’s per-ception of himself as a parent and mother’s beliefsabout the role of the father is present when mea-suring accessibility forms of father involvement,especially on nonworking days. However, this mod-erating effect does not exist when using measuresthat assess only workday accessibility or father’slevels of interaction and responsibility.

Discussion

Findings from the current study suggest thatmothers may play an active role in influencing howfathers approach parenting. Specifically, it seemsthat the relationship between fathers’ perceivedinvestments in their parental roles and actual levelsof paternal involvement are moderated by mothers’beliefs about the role of the father. These findingscannot be viewed as conclusive, yet they do raise sev-eral interesting questions and implications for futureresearch that explore the roles played by mothers ininfluencing father involvement.

In our correlational analyses, mothers’ percep-tions of their husbands’ investment in the paternalrole—unlike these fathers’ own perceptions—wererelated to fathers’ overall involvement. Althoughthese findings provide support for previous researchindicating that mothers’ perceptions of the pater-nal role are better predictors of father involvementthan fathers’ own perceptions of the paternal role(De Luccie, 1995; McBride & Rane, 1997; Pleck,1997), they underscore the need for future researchthat identifies and examines how these perceptionsare formed. Fathers’ own perceptions of their in-vestment in parenting were only related to the re-sponsibility dimension of father involvement. Suchfindings raise an interesting question: Why is therea lack of connection between fathers’ perceptions oftheir parenting role and their actual behavior in alldimensions of father involvement?

Results from our regression analyses providepotential insights on possible answers to this ques-tion. These analyses suggest that a mother’s beliefabout the role of fathers moderates the link betweena father’s perception of himself as a parent and thedegree to which he is accessible to his child. Closerexamination of these findings suggests that a father’sperception of himself as a parent is only positivelyrelated to accessibility when the mother believes that

Table 2. Hierarchical Regression Predicting FatheringAccessibility

Accessibility

(Total)

Accessibility

(Nonworking)

B b B b

Step 1

Father’s working

hours

.87 .03 22.03 2.11

Mother’s working

hours

10.70** .64 7.14** .64

F 5.07* 4.11*

R2 .32 .27

Step 2

Father’s perception of

self as a parent (A)

68.99 .32 41.55 .29

Mother’s role

questionnaires (B)

60.74 .17 57.36 .25

F change .56 .22

R2 .04 .02

Step 3

A � B 191.83** .62 114.82* .56

F change 7.49* 8.67**

R2 .18 .22

Overall F 4.37* 3.98*

*p ,.05. **p ,.01.

Family Relations � Volume 54, Number 3 � July 2005368

fathers should be involved parents. Likewise, norelationship was revealed between fathers’ perceivedinvestment in their parental roles and actual paternalaccessibility when mothers did not believe thatfathers should be involved. This moderating effectof maternal beliefs about the role of fathers providesone possibility of how mothers might play a centralrole in regulating father involvement. These findingsprovide some support for previous research suggest-ing that mothers’ beliefs play a gatekeeping role ininfluencing father involvement. These findings alsohighlight an important question for future researchon this topic: What are the different mechanismsand pathways through which maternal beliefs influ-ence actual fathering behaviors? For instance, it maybe that mothers who believe in the importance offathering make themselves less accessible to theirchildren, thus forcing fathers to be more promi-nently involved. Likewise, it could be that workingmothers are not only less available, but may also beless traditional and gendered with regard to parent-ing, thus preferring greater father involvement.

These results help elaborate upon the mecha-nisms that may underlie the gatekeeping phenome-non. Although some previous research has examinedthe degree to which maternal behaviors (e.g., Fagan& Barnett, 2003) may regulate father involvement,it is striking that mothers’ beliefs about the fatheringrole seem to be so influential in determiningwhether fathers do or do not act on their feelings ofinvestment in the paternal role. It does seem likelythat maternal beliefs may be manifested in some sortof behavior that regulates fathers’ parenting, even ifthis influence is very understated (e.g., Pleck &Stueve, 2001). Although this particular study doesnot measure gatekeeping behavior, parenting beliefsand parenting behaviors are often related (see Sigel& McGillicuddy-DeLisi, 2002, for a review). Like-wise, it is likely that ‘‘maternal attitudes may leadto behavior that in turn limits father involvementand constitutes a form of gatekeeping’’ (Parke,2002, p. 40). Maternal beliefs could initiate a patternof interactions that may increase or decrease thestrength of association between fathering identityand father involvement.

This study also provides insight on the contextsin which maternal influences may occur. Day andLamb (2004), Pleck & Masciadrelli (2004),Schoppe et al. (2004), and others have argued thatfather involvement is a multidimensional construct.These researchers also argue that each of the

components making up this construct may havea different set of determinants. Likewise, we wereinterested in the specific elements of involvement,and whether these components showed similar ordifferent patterns of associations than did fatherinvolvement as a whole. When individually examin-ing each of the components used to operationalizeour measures of father involvement (i.e., interaction,accessibility, and responsibility), the moderatingimpact of maternal beliefs about the role of fathersemerged only for the accessibility measure. Thisfinding supports previous research suggesting thatmaternal gatekeeping varies across contexts (e.g.,Beitel & Parke, 1998; De Luccie, 1995; Grossman,et al., 1988) and underscores the importance forresearchers to examine each of the components offathering separately as they explore the antecedentsand consequences of father involvement.

Likewise, the fact that this moderating effectholds only for nonworkdays suggests that employ-ment may also influence the contexts in which fatherinvolvement—and maternal influence on it—mayoccur. Fathers would seemingly have more choiceand control over how they spend their time duringthese nonworking days, and nonworkday accessibil-ity should be particularly reflective of whetherfathers are choosing to spend time with their chil-dren. It is compelling that maternal influences aremost strong in this particular context. The influenceof maternal beliefs on fathers’ nonworkday accessi-bility indicates both the strength of this effect andthe general need for researchers to examine the con-texts in which it may be found.

Accessibility forms of involvement may be lesssalient to fathers’ perceptions of themselves as par-ents than either interaction or responsibility compo-nents. Fathers’ identity and societal expectations forthe fathering role may call for increased father-childplay, direct interaction, or a higher burden of par-enting responsibility, but expected involvement isnot normally geared toward accessibility. As a result,mothers will have more control in determiningfather involvement in this domain. Whereas fathers’intrapersonal identities and societal expectations arefocused on certain forms of involvement, mothersmay still play a potentially important role in regulat-ing fathers’ accessibility to their children.

Limitations of the conceptualization of fatherinvolvement as reflected in the Lamb et al. (1987)model must be acknowledged. Although useful inguiding research on father involvement by capturing

Maternal Influences on Father Involvement � McBride et al. 369

a range of activities that influence children’s devel-opment (Marsiglio et al., 2000), this model fails toacknowledge other important dimensions of father-ing (e.g., providing financially for children, cogni-tive monitoring) that may also influence children’sdevelopment (Palkovitz, 1997). There is also grow-ing acknowledgment that the quality of interactionbetween fathers and their children may be moreimportant for child outcomes than the quantity ofinteraction (Hawkins & Dollahite, 1997; Pleck &Masciadrelli, 2004). Although we assert that it is stillimportant to understand why some fathers choose tobecome more actively involved in childrearing thanothers, future investigations of father involvementshould consider a variety of operationalizations ofinvolvement and consider quality as well as quantityof involvement. Despite the multidimensional frame-work of father involvement that was employed in thisstudy, these measures still yield a view of involvementthat is temporally and conceptually constrained.Fathers were not asked to define the nature of involve-ment in their children’s lives. In order to gain a morecomplete picture, fatherhood research should work todevelop context-sensitive measures that assess themeaning of involvement patterns to both parents.

Caution must be used when trying to generalizefindings from the current study, especially given oursmall sample size. It is clearly difficult to draw causalconclusions given these results and the reciprocalnature of family interactions. Mothers may very wellbase their beliefs about the role of the father onfathers’ own patterns of behavior. Accordingly,fathers whose perception of their parental identitymatches their level of involvement may increasetheir partners’ beliefs in the importance and abilityof fathers. Although it is reasonable to assume thatsome maternal attitudes pre-exist before the birth ofa child, reciprocal influences over the course ofchildrearing must be acknowledged. Additional lim-itations associated with this investigation include thelack of family diversity (e.g., predominantly Whitemiddle-class participants) and our fairly narrowmeasures of father involvement, thus prohibiting theapplication of these findings to other populationsand or ecological contexts for fatherhood. In spite ofthese limitations, the findings are encouraging inthat they provide added support to the growingbody of literature suggesting that maternal factorsmay be important determinants of father involve-ment. These findings also suggest the possibility thatmaternal gatekeeping may be context specific, and

that the roles played by mothers in influencingfathering behaviors may vary based on the specificaspect of paternal involvement being examined.Future research is needed to further delineate thenature of the roles played by mothers in influencingfather involvement, how maternal influences mightvary based on the populations of fathers being stud-ied, and the contexts in which involvement is beingexamined.

Implications for Practice

Our findings indicate that maternal perceptions offathers’ investment in the parenting role are relatedto father involvement. In addition, mothers’ beliefsabout the role of the father moderate the linkbetween fathers’ own perceived investment in par-enting and father involvement. These findings canbe used to help practitioners and programs developeffective intervention strategies to increase paternalinvolvement. Although many programs explicitlyattempt to help fathers become more involved intheir children’s lives, relatively few have focused onthe way that mothers may influence this process (seeMcBride & Lutz, 2004, for a review).

This research suggests the possibility thatmothers may play a role in either encouraging ordiscouraging fathers to act on their feelings of invest-ment in the paternal role. If this is the case, then onepath to changing fathers’ behavior may involvechanging the way that mothers look at them. Ifmothers believe that fathers can and should be capa-ble parents, they are more likely to allow fathers intothe lives of their children. In changing women’sbeliefs about the paternal role, invested fathers maybe able to change their behavior by becoming moreaccessible to parent their sons and daughters. Itseems that maternal characteristics could becomea focus of programs designed to augment fatherinvolvement. Thus, we hope that this research canbe used to better understand the family influenceson father involvement and subsequently help fathersplay a more important role in their own families.

We are not suggesting that mothers are solelyresponsible for fathers’ involvement in their chil-dren’s lives. Indeed, many fathers may possess littleor no initiative to become involved at all. Addition-ally, many mothers may play active roles in encourag-ing their partners to become more involved parents,even when these fathers are not highly invested inthe parenting role. However, our results do suggest

Family Relations � Volume 54, Number 3 � July 2005370

that maternal beliefs may prevent some men who arecommitted to the parenting role from fully acting onthis commitment.

Within these families, there are a number ofpotential strategies that could focus on maternal be-liefs as a means for increasing paternal involvement.Mothers must be alerted to the fact that fathers areoften willing and able to participate in child care.For instance, participation in mixed-gender parent-ing classes could help to ease the restrictions thatsome gatekeeping mothers place on their partners’involvement. If mothers recognize that fatherspossess the initiative and knowledge to actively con-tribute to their children’s lives, then they mayencourage fathers to act on their feelings of paternalinvestment.

Further, parenting programs could be aimedtoward sensitizing mothers to paternal involvementbehavior. Many positive paternal behaviors may gounnoticed, which could lead some mothers to pos-sess skewed perceptions of their partners as ‘‘unin-volved’’ and consequently discourage further fatherinvolvement. Similarly, parenting programs shouldattempt to clearly demonstrate that men are success-fully able to perform a multitude of child-relatedtasks. A number of mothers may not be aware thatfathers are capable of being competent caregivers.Demonstrations of effective paternal childrearingcould serve to alter these mothers’ beliefs about therole of the father.

We must emphasize that incorporating these sug-gestions into parenting programs is certainly not theonly method for increasing paternal involvement.Child, family, and contextual variables interact incomplex ways to promote involved fathering, andintervention at one particular point is insufficientfor producing the desired behavioral effect. A num-ber of other strategies, particularly those that directlyfocus on fathers’ own beliefs, attitudes, and behav-ior, may be more efficient ways of increasing fathers’involvement. Moreover, it is not necessarily true thatfathers provide benefits for their children onlythrough temporal involvement. In fact, fathersassume a wide variety of roles (including breadwin-ning) that may have positive effects for children andfamilies but were not captured in our conceptualiza-tion of the involved father.

Nonetheless, this research suggests the possibilitythat maternal beliefs may be tied to fathers’ involve-ment in their children’s lives. Future research shouldattempt to determine if these beliefs actually play

a role in influencing the quantity and quality offather involvement. This knowledge would have thepotential to inform programs and policies intendedto encourage involved fathering.

References

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and inter-preting interactions. Newbury Park, CA: Sage.

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

Amato, P. (1998). More than money? Men’s contributions to their child-ren’s lives. In A. Booth & A. C. Crouter (Eds.), Men in families: Whendo they get involved? What difference does it make? (pp. 241–278).Mahwah, NJ: Erlbaum.

Arditti, J., Acock, A., & Day, R. (2005). Incarceration and reentry offathers into the lives of their families. In V. Bengston, A. Acock,K. Allen, P. Dilworth Anderson, & D. Klein (Eds.), Sourcebook of fam-ily theory and research (pp. 352–354). Thousand Oaks, CA: Sage.

Barnett, R. C., & Baruch, G. K. (1988). Correlates of fathers’ participationin family work. In P. Bronstein & C. P. Cowan (Eds.), Fatherhoodtoday: Men’s changing role in the family (pp. 66–78). New York: Wiley.

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

Cowan, C. P., & Cowan, P. A. (2000). When partners become parents: Thebig life change for couples. Mahwah, NJ: Erlbaum.

Day, R., & Lamb, M. E. (2004). Conceptualizing and measuring fatherinvolvement. Mahway, NJ: Erlbaum.

De Luccie, M. F. (1995). Mothers as gatekeepers: A model of maternalmediators of father involvement. Journal of Genetic Psychology, 156,115–131.

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

Fox, G. L., & Bruce, C. (2001). Conditional fatherhood: Identity theoryand parental investment theory as alternative sources of explanation offathering. Journal of Marriage & the Family, 63, 394–403.

Grossman, F. K., Pollack, W. S., & Golding, E. (1988). Fathers and chil-dren: Predicting the quality and quantity of fathering. DevelopmentalPsychology, 24, 82–91.

Hawkins, A. J. & Dollahite, D. C. (1997). Beyond the role-inadequacyperspective of fathering. In A. J. Hawkins & D. C. Dollahite (Eds.),Generative fathering: Beyond deficit perspectives (pp. 3–16). ThousandOaks, CA: Sage.

Hoffman, C. D., & Moon, M. (1999). Women’s characteristics and genderrole attitudes: Support for father involvement with children. Journal ofGenetic Psychology, 160, 411–418.

Ihinger-Tallman, M., Pasley, K., Beuhler, C. (1993). Developing a middle-range theory of father involvement postdivorce. Journal of Family Issues,14, 550–571.

Lamb, M. E., Pleck, J. H., Charnov, E. L., & Levine, J. A. (1987). A bio-social perspective on paternal behavior and involvement. In J. B.Lancaster, J. Altman, A. Rossi, & L. R. Sherrod (Eds.), Parenting acrossthe lifespan: Biosocial perspectives (pp. 111–142). New York: Academic.

Larossa, R. (1988). Fatherhood and social change. Family Relations, 37,451–457.

Marsiglio, W., Amato, P., Day, R. D, & Lamb, M. E. (2000). Scholarshipon fatherhood in the 1990s and beyond. Journal of Marriage and theFamily, 62, 1173–1191.

Maurer, T. W., Pleck, J. H., & Rane, T. R. (2001). Parental identity andreflected-appraisals: Measurement and gender dynamics. Journal ofMarriage and the Family, 63, 309–321.

McBride, B. A., & Lutz, M. M. (2004). Intervention: Changing the natureand extent of father involvement. In M. E. Lamb (Ed.), The role of thefather in child development (4th ed., pp. 446–475). New York: Wiley.

Maternal Influences on Father Involvement � McBride et al. 371

McBride, B. A., & Mills, G. (1993). A comparison of mother and fatherinvolvement with their preschool age children. Early Childhood ResearchQuarterly, 8, 457–477.

McBride, B. A., & Rane, T. R. (1997). Role identity, role investments, andpaternal involvement: Implications for parenting programs for me.Early Childhood Research Quarterly, 12, 173–197.

Palkovitz, R. (1984). Parental attitudes and fathers’ interactions withtheir 5-month-old infants. Developmental Psychology , 20 , 1054–1060.

Palkovitz, R. (1997). Reconstructing ‘‘involvement’’: Expanding conceptu-alizations of men’s caring in contemporary families. In A. J. Hawkins& D. C. Dollahite (Eds.), Generative fathering: Beyond deficit perspec-tives (pp. 200–216). Thousand Oaks, CA: Sage.

Palkovitz, R. (2002). Involved fathering and men’s adult development: Provi-sional balances. Mahwah, NJ: Erlbaum.

Parke, R. D. (2002). Fathers and families. In M. Bornstein (Ed.), Hand-book of parenting: Vol. 3. Being and becoming a parent (2nd ed., pp. 27–73). Hillsdale, NJ: Erlbaum.

Pleck, J. H. (1997). Paternal involvement: Levels, sources, and consequen-ces. In M. E. Lamb (Ed.), The role of the father in child development(3rd ed., pp. 66–103). New York: Wiley.

Pleck, J. H., & Masciadrelli, B. P. (2004). Paternal involvement by U.S.residential fathers: Levels, sources, and consequences. In M. E. Lamb

(Ed.), The role of the father in child development (4th ed., pp. 222–271).New York: Wiley.

Pleck, J. H., & Stueve, J. L. (2001). Time and paternal involvement. InK. Daly (Ed.), Minding the time in family experience: Emerging perspec-tives and issues (pp. 205–226). Oxford: Elsevier Science.

Rane, T. R., & McBride, B. A. (2000). Identity theory as a guide to under-standing fathers’ involvement with their children. Journal of FamilyIssues, 21, 347–366.

Schoppe, S. J., McBride, B. A., & Ho, M. (2004). Unidimensional versusmultidimensional perspectives on father involvement. Fathering: A Jour-nal of Theory, Research, and Practice, 2, 147–164.

Sigel, I. E., & McGillicuddy-DeLisi, A. V. (2002). Parent beliefs are cogni-tions: The dynamic belief systems model. In M. Bornstein (Ed.), Hand-book of parenting: Vol. 2. Being and becoming a parent (2nd ed., pp.485–508). Hillsdale, NJ: Erlbaum.

Stryker, S. (1987). Identity theory: Developments and extensions. InK. Yardley & T. Honess (Eds.), Self and identity: Psychosocial perspectives(pp. 89–103). New York: Wiley.

Tamis-LeMonda, C. S., & Cabrera, N. (1999). Perspectives on fatherinvolvement: Research and policy. Social Policy Report: Society forResearch in Child Development, 13(1), 1–32.

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

Family Relations � Volume 54, Number 3 � July 2005372