Fathers’ Role Model Perceptions Themes and Linkages with Involvement

Abstract

Qualitative interview data are used to explore fathers’( N = 24) perceptions of their own fathers and others as influential parental role models and associations between fathers’ role model perceptions and their involvement with their own children. In fathers’ descriptions of their parental role models, three themes emerged: types of models that fathers identified as role models for them as parents, affective evaluations the fathers ascribed to the models, and content that the fathers perceived learning from the models. Highly involved fathers were more likely to cite peer parents than to specifically cite their own fathers as influential role models for them and infrequently cited their spouses as models. Low-involvement fathers more often attributed positive affective evaluations to their models than did highly involved fathers.

Full text

10.1177/1097184X04270377MEN AND MASCULINITIES / Month XXXXMasciadrelli et al. / ROLE MODEL PERCEPTIONS
Fathers’ Role Model Perceptions
Themes and Linkages with Involvement
BRIAN P. MASCIADRELLI
JOSEPH H. PLECK
University of Illinois at Urbana-Champaign
JEFFREY L. STUEVE
University of New Mexico
Qualitative interview data are used to explore fathers’(N= 24) perceptions of their own
fathers and others as influential parental role models and associations between fathers’
role model perceptionsand their involvement with their own children.In fathers’ descrip-
tions of their parental role models, three themes emerged: types of models that fathers
identified as rolemodels for them as parents, affective evaluations the fathers ascribed to
the models, and content that the fathers perceived learning from the models. Highly
involved fathers were more likely to cite peer parents than to specifically cite their own
fathers as influential rolemodels for them and infrequently cited their spouses as models.
Low-involvement fathers more often attributed positive affective evaluations to their
models than did highly involved fathers.
Key words: fatherhood; father involvement; role models
According to social learning theory, fathers learn how to enact the paternal
role through interactions with significant role models (Parke 1995). Experi-
ences with models are considered to be of great importance in becoming gen-
erative, or productive and nurturing, in the paternal role (Hawkins et al.
1993). A possible route through which this learning takes place is the obser-
vation of other parents’ child rearing behaviors, using these other parents as
role models for enacting one’s own parenting behaviors, a process referred to
as observational learning (Bandura 1999). In general, this form of learning
takes place when an individual observes the behaviors of another person, or
model, and then enacts these behaviors himself or herself in an applicable
future situation. The learning that occurs through this process may take the
form of changes in the attitudes, values, emotions, and overtbehaviors of the
observer.
Authors’ Note: Please address all correspondence to Brian P. Masciadrelli, 167 Bevier Hall,
MC-180, 905 South Goodwin Avenue, Urbana, IL 61801; e-mail: masciadr@uiuc.edu.
Men and Masculinities, Vol. XX No. X, Month 2004 1-11
DOI: 10.1177/1097184X04270377
© 2004 Sage Publications
1

There has been interest in the influence of men’s own fathers as models of
the paternal role, and two major hypotheses have been advanced. The first is
the modeling hypothesis (Sagi 1982; Parke 1995). This hypothesis suggests
that men will be highly involved in child rearing and the paternal role if they
had fathers who were similarly (highly) involved in parenting them. The sec-
ond is the compensatory hypothesis. It asserts that high levels of involvement
in child rearing are a reaction by men whose own fathers were unavailable or
not supportive during their childhoods (Sagi 1982). It is seen as an attempt to
rework their childhood experiences through the involved parenting of their
own children, thus compensating for the deficiencies perceived in childhood
relationships with their own fathers (Parke 1995). In this circumstance, men
may look to other persons as positive role models.
Pleck (1997) suggests that men’s affective evaluations of their own
fathers’ parental involvement is the key moderating variable determining
whether modeling or compensation occurs. Specifically, when a man evalu-
ates the experience of his own father’s parenting as positive, he is more likely
to model the level of his father’s involvement as a parent himself. Likewise,
compensation will occur if a man negatively evaluates the experience of his
own father’s involvement during childhood.
It is important to note that the process of modeling may not be limited to
straightforward imitations of all behaviors observed from a single model.
The observer may take only selected behaviors and attitudes from a model
and then create new combinations or adaptations. These will differ from the
original, particularly in the specific content of actions and judgments
(Bandura 1999). Furthermore, a man’s own father is not necessarily the only
model he will use. For example, Hawkins et al. (1993) note that mothers
(referring to men’s partners) provide a source of modeling. Daly (1993)
found that rather than singling out one specific model, fathers use composites
of several models, or what he terms fragmented models, to develop paternal
identities.
Further investigation is needed to illuminate how models influence fathers
and fathering behavior. Toward this end, the present descriptive study uses
qualitative interview data to explore how fathers perceive their own fathers as
well as others as influential role models. In addition, the study examines the
association between fathers’ role model perceptions and fathers’involvement
with their own children.
METHOD
Participants
Interview data from fathers (N= 24) that completed the Parenting Narra-
tive Interview (PNI; Stueve and Pleck 2001) were used in the present study.
2 MEN AND MASCULINITIES / Month XXXX

All participants provided informed consent and received a monetary in-
centive for their participation in the PNI. The sample was drawn from a
university-based preschool setting in the midwestern United States. Most
participants were between the ages of 31 and 40 (66%), Caucasian (79%),
and completed 20 or more years of education (44%). All participants were
married.
Procedures and Data Collection
Role model influences. Data concerning the influences of models on
fathers were obtained from the PNI (Stueve and Pleck 2001). One subsection
of the PNI includes questions concerning role models that are influential in
determining an individual’s approach to parenting. Its two questions are as
follows:
•Are there any role models who have influenced you as a parent? If so, who are
they?
•How have they influenced you?
The initial data analysis used techniques from grounded theory methodology
(Strauss and Corbin 1998). Specifically, the first author performed “open
coding” to identify and categorize dominant themes in the interviews.
Parental involvement level. Along with the PNI, a self-report measure of
behavioral parental involvement was administered. The Parental Involve-
ment Scale (PIS; Bruce and Fox 1997, 1999) is a 21-item measure of in-
volvement in child rearing and caregiving with an alpha of .95 reported by its
original authors. An alpha coefficient of .88 was calculated with the present
sample. The instrument covers four domains of parental involvement in
childrearing. First, custodial caretaking functions are covered (i.e., assisting
or supervising child in bathing or personal hygiene). The second domain cov-
ers socioemotional functions (i.e., joining the child in his or her favorite
activities). The third domain involves teaching functions (i.e., sharing values
with the child). The last domain concerns executive functions involved in
parenting (i.e., making decisions that pertain to the child or assisting the child
in making decisions). Items were rated regarding level of involvement in the
tasks on a 4-point scale with 1 = never or hardly ever and 4 = almost daily.
Additional data analyses included comparisons between fathers based on
their level of parental involvement, using scores from this parental involve-
ment measure to form a median split between high and low involvement
fathers. Fathers’self-reported involvement is potentially valuable to consider
in relation to the themes emerging in fathers’ descriptions of influential role
models.
Masciadrelli et al. / ROLE MODEL PERCEPTIONS 3

RESULTS AND DISCUSSION
Themes in Fathers’ Role Model Perceptions
In general, three different themes emerged from the PNI data. Appearing
first was a broad category containing the various types of models cited.
Model type refers to particular individuals that the respondents perceived as
persons they learned something of value from in terms of parenting. A sec-
ond theme to emerge was the affective evaluation for many of the models
cited (i.e., whether the model was perceived as positive, negative, or other-
wise). The third thematic category to emerge in the data was the modeled
content learned. This category contained three subthemes within it, consist-
ing of responsibility, good values, and patience.
Model type. Only two fathers reported having no role models. The remain-
ing described between one and five models. The sample of twenty-four
fathers reported forty-three models in all. In the following analyses, the unit
of analysis is the model described by the father, rather than the responding
father. Using models as the unit of analysis provides more insight into the role
of models in the fathering experience.
The participants described a wide variety of model types. These consisted
of the father’s own parents (both mothers and fathers), the father’s own
grandparents (both grandmothers and grandfathers), the father’s spouse, the
spouse’s parents, and peers who are parents. Table 1 (panel A) includes the
frequencies of models types cited. For model type, a man’s own father was
cited relatively infrequently (16%), even if general responses of “own par-
ents” are included with this number (37% [16% + 21%]). Also, spouse was
not frequently cited as a model (5%), but her parents were cited more fre-
quently (14%).
Affective evaluation. Four affective evaluations appeared in the data: posi-
tive, negative, ambivalent, and unspecified. It should be noted that fathers
volunteered this information without a prompting question. An example of a
positive evaluation is seen in this father’s describing his wife’s aunt and
uncle, saying, “They were just really great. Her aunt and uncle raised three
great kids and I hope our kids turn out to be like them. I plan to go to them for
advice.” One father exemplifies the negative evaluation in his simple state-
ment that “my own parents were, in my opinion, an object lesson on how not
to do things.” Ambivalent evaluations are exemplified by this man who, in
response to the question of who his role models were as a parent, said, “I
would say my own parents, but I’d be hard pressed whether to say that was
positive or negative.” Those fathers who did not indicate any affective evalua-
tion of their models were grouped under the unspecified category. Table 1
4 MEN AND MASCULINITIES / Month XXXX

(panel B) includes the frequencies of affective evaluations associated with
the models. About half the models were evaluated positively (49%). One-
fourth of the models were evaluated in negative or ambivalent terms (25%
[16% + 9%]). Affective evaluations for the remainder of the models cited
were unspecified.
Modeled content learned. Three prominent subthemes emerged here: re-
sponsibility, good values, and patience. Responsibility emerged as the first
subtheme in the content these fathers learned from their models. Many
fathers felt that the role models they had observed promoted the importance
of responsibility in their development as parents. Responsibility concerned
the importance of fulfilling obligations, including those to other family mem-
bers. One father, when discussing learning the value of responsibility in
family life, recounted how this came across to him. He said, “I got the sense
from the bigger family that hard work was important, family was important,
and that we all hung together.” Another father, who brought up responsibil-
ity themes when describing how his own father influenced him, said the
following:
Masciadrelli et al. / ROLE MODEL PERCEPTIONS 5
Table 1:
Models Cited by Three Themes
Model Themes Models Cited (%)
Model type (A)
Father’s own family 62
Father’s own father 16
Father’s own mother 14
Father’s parents (nonspecific) 21
Father’s grandparents 11
Spouse and her family 19
Spouse 5
Spouse’s parents 14
Peer parents 19
Affective evaluation (B)
Positive 49
Negative 16
Ambivalent 9
Unspecified 26
Content learned (C)
Responsibility 7
Good values 7
Patience 5
Unspecified 81
NOTE: Percentages refer to number of models cited within groupings of participants by involvement level.
The total number of models cited overall is forty-three. N= 24 fathers responding.

He was very strong on responsibility both by example and telling us what we
had to do and when we had to do things, especially like if we had committed to
something and we forgot about it, he would see to it that we saw it through,
whatever it was, and got it finished.
In saying that their models influenced them to be more responsible them-
selves as parents, fathers seemed to imply that they also learned it is impor-
tant to promote responsibility in their children.
The second subtheme emerging in this category was good values as a par-
ent. Good values concerned how to be a morally correct parent by maintain-
ing integrity and honesty. Fathers reported learning good values from the var-
ious models they observed. One man described how influential his
grandfather, a military chaplain, was as a model for him in promoting the
importance of good values in the parenting process. He had this to say in
response to the question of how his model influenced him:
[His influence was] through his behavior. Watching and listening to his lessons
on life. How important it was to be honest and not to tell lies. Always to use the
golden rule and treat others as you want them to treat you.
Fathers seemed to perceive good values as something their models taught
them to have, as well as taught them to promote in their children (paralleling
fathers’ descriptions of responsibility). One father’s description of his late
grandfather as a role model made this connection explicit:
One thing that always struck me about him was his integrity,his quiet fortitude,
work ethic, and those sorts of things . . . definitely influence me as a man and
how I respond to the process of parenting. And that is the model, those sorts of
characteristics of men in my life, that I want to impart upon my children.
Patience was perceived as a skill that some of these fathers took away from
their experiences observing role models. The following excerpt from one
interview catches the meaning of patience being a skill that the fathers often
saw themselves as initially lacking:
My wife’s dad. Probably the nicest guy you’ll ever see. Probably the most
patient person you’ll ever see. If anything, he’s taught me the virtue of patience.
When we first had our children, I didn’t have a lot of that. When I saw how he
interacted with my children, saw how he dealt with them, I picked up some, you
know, some helpful hints.
Table 1 (panel C) includes frequencies of modeled content learned asso-
ciated with the models. Only a small proportion of the models were cited
as providing modeling of responsibility (7%), good values (7%), and pa-
tience (5%).
It should be noted that the three subthemes in the modeled content learned
category all emerged from models described in positive terms by the fathers.
6 MEN AND MASCULINITIES / Month XXXX

That is, descriptions of positively evaluated models all included some sort of
concrete or specific discussion of what the father learned from them and how
it was meaningful. For models evaluated negatively, what fathers learned was
vague or nonspecific.
Differences Between
High- and Low-Involvement Fathers
In light of the themes identified in fathers’ role model perceptions, three
questions of interest are posed. First, how do fathers with high levels of
involvement compare to those with low levels of involvement in whom they
identify as role models for parenting? Second, what differences exist
between these two groups of fathers in terms of affective evaluation? Finally,
what differences exist between these two groups of fathers in the content
learned from their models? A median split on the PIS (M= 66.6, Mdn = 65.5,
SD = 9.3) was employed to divide the cases. There were no differences
between the two groups in demographic composition.
Table 2 shows the association between involvement level and the three
model themes. The first section (panel A) shows the association between
involvement level and general model type. Both high- and low-involvement
fathers report about the same frequency for the citation of models from their
own families of origin (including men’s own fathers). Spouses and their fam-
ilies are cited less frequently as models among the high-involvement fathers
than for the low-involvement fathers. In the high-involvement group, though,
there is a greater availability of peer parent models reported compared to the
low-involvement group.
Panel B of Table 2 shows the association between involvement levels and
more specific model types (excluding peers since there are no subcatego-
ries). High-involvement fathers differ little from low-involvement fathers in
their citation of any of the specific model types within the broader father’s
own family category, consistent with the results in panel A. Likewise, high-
involvement fathers less often report their spouse and their spouse’s parents
as models, in accord with panel A. It is notable that the high-involvement
fathers are about equally likely to cite their own father as a role model (14%)
as they are to cite their own mother (14%) and own grandparents (13%).
These findings are interesting in terms of the modeling and compensatory
hypotheses (Sagi 1982; Parke 1995) because both emphasize men’s own
fathers as the determining role model for becoming highly involved in
parenting their children. Only 14% of the models cited by the highly involved
fathers in this study were their own fathers, while 27% of those models
cited by them were peers who were parents. On the other hand, the low-
involvement fathers showed the opposite pattern, with 19% citing their own
fathers and only 10% citing peers. These results call the modeling and com-
Masciadrelli et al. / ROLE MODEL PERCEPTIONS 7

pensatory hypotheses into some question because highly involved fathers are
less likely than other fathers to describe their own fathers as role models of
any sort, to either imitate or compensate for. However, this interpretation is
limited because we do not have data about all fathers’ perceptions of their
own fathers, only of those perceiving their fathers as influential role models.
Highly involved fathers’ greater availability of peer parents as models com-
pared to the low-involvement group is consistent with the findings of Daly
(1993), which suggest that fathers incorporate peers that are parents as mod-
els in their developing paternal identities.
These findings are also surprising in light of the suggestion that children’s
mothers may be one source of modeling for fathers to learn how to become
highly involved, generative parents (Hawkins et al. 1993). The evidence
available in the present study, however, appears to go against this suggestion.
None of the models cited by the high-involvement group were the spouse.
Next shown in Table 2 (panel C) is the comparison of high- and low-
involvement fathers on the affective evaluation that the men ascribe to their
8 MEN AND MASCULINITIES / Month XXXX
Table 2:
Models Cited by Three Themes for High- and Low-Involvement Fathers
Fathers’ Level of Involvement
Model Theme High Involvement (%) Low Involvement (%)
Model type—General (A)
Father’s own family 64 61
Spouse and her family 9 29
Peer parents 27 10
Model type—Specific (B)
Father’s own father 14 19
Father’s own mother 14 14
Father’s parents (nonspecific) 23 19
Father’s grandparents 13 9
Spouse 0 10
Spouse’s parents 9 19
Affective evaluation (C)
Positive 36 62
Negative 9 23
Ambivalent 14 5
Unspecified 41 10
Content learned (D)
Responsibility 13 0
Good values 5 10
Patience 5 5
Unspecified 77 85
NOTE: Percentages refer to number of models cited within groupings of participants by involvement level.
The total number of models cited by fathers was forty-three (high-involvement fathers group = twenty-two
models cited; low-involvement fathers group = tweny-one models cited). N= 12 fathers responding in each
group.

models. For the high-involvement fathers, 36% of the models cited by them
are described positively, while the low-involvement fathers describe 62% of
the models they cite as positive. The models cited by high involvement
fathers are also less often negative (9% vs. 23%). Finally, of the models cited
by the high-involvement fathers, 55% (14% + 41%) are described with
ambivalent or no evaluations, while only 15% (5% + 10%) of those cited by
the low involvement fathers are described this way. The general pattern
that emerges shows high-involvement fathers describing less affect, either
positive or negative, in evaluating their models than do the low-involvement
fathers (45% vs. 85%).
Connected to the modeling and compensatory hypotheses is the sugges-
tion that the affective evaluation of models tends to influence the involvement
level of the fathers with their own children (Pleck 1997). Specifically, posi-
tively or negatively evaluated models may lead to high levels of father
involvement in child rearing because of modeling or compensatory pro-
cesses, respectively. (These hypotheses can apply, of course, not only to
men’s own fathers but to any modeling source.) The evidence from the pres-
ent sample, however, goes against this notion. For either or both of these
hypotheses to apply, the models cited by high-involvement fathers should
more often be either positive or negative. However, this is the case for only
45% (36% + 9%) of the models cited by high-involvement fathers, compared
to 85% (62% + 23%) of the models cited by low involvement fathers. For
low-involvement fathers’ models to be more often positively or negatively
valued, and for the modeling and compensatory hypotheses to apply, the
role models that low-involvement fathers feel positively about modeled low
levels of parental involvement, and the models they evaluated negatively
showed high levels of involvement. If true, these possibilities go beyond the
modeling and compensatory hypotheses as currently formulated.
Finally, Table 2 (panel D) compares high- and low-involvement fathers on
how they differ in modeled content learned. Responsibility was only reported
as being modeled from fathers in the high involvement group. Thirteen per-
cent of the models in this group were cited as modeling responsibility. Good
values were slightly more frequently reported as the modeled content learned
among low-involvement fathers than among high-involvement fathers (10%
vs. 5%). There was no difference in frequency of models cited for modeling
patience between the two groups (5% in each).
SUMMARY AND CONCLUSIONS
In this qualitative exploration of fathers’ descriptions of their role models,
three themes emerged. The first involved the types of models that fathers
identified as influential role models for them as parents, such as their own
parents, spouses, and peer parents. The second involved theaffective evalua-
Masciadrelli et al. / ROLE MODEL PERCEPTIONS 9

tions the fathers ascribed to the models, such as seeing them as positive or
negative. The third theme consisted of the content that the fathers perceived
as having been learned from the models. This category included three
subthemes that emerged as most prominent for the fathers interviewed:
responsibility, good values, and patience.
Many of this study’s findings are surprising. Highly involved fathers were
more likely to cite peer parents than to specifically cite their own fathers as
influential role models and did not cite their spouses as models, while low-
involvement fathers more often attributed a positive affective evaluation to
their models than the highly involved fathers. These results seem inconsistent
with the modeling and compensatory hypotheses. Applying the hypotheses
to the results necessitates the assumption that the models observed and evalu-
ated positively by the low-involvement fathers engaged in low levels of
involvement and/or that the models they evaluated negatively showed high
involvement. The processes are certainly possible, but they are not the ways
the modeling and compensatory processes have been assumed to operate.
Thus, these results may suggest that modeling and compensation need to be
conceptualized in a broader way. That is, modeling may occur with low-
involvement models, not only high-involvement ones. Likewise, compen-
sation may occur for high-involvement models, not only low-involvement
ones.
The findings of the present study are limited in a number of respects. First,
the participants were not randomly selected. This limits the generalizability
of the findings derived from this data set. In addition, the sample is very
homogenous demographically, adding to limitations in generalizability.
The use of an existing data set also places limitations on the findings of
this research. The questions asked may not be in complete alignment with the
research goals of the present study. However, the qualitative methods used in
the present analyses reduce this limitation to some degree, as the methodol-
ogy from the grounded theory model is well suited to exploring existing
interview data such as that in the PNI. A second limitation of the findings
derived from the existing data set used here is that the interviewers did not
elicit extra detail from participants who provided only very brief responses to
the questions (e.g., not eliciting which specific parent might be a role model
when only the brief response of “my parents” was provided). However, the
finding that positively evaluated models tended to receive greater attention
in the participants’ responses than those evaluated negatively may in fact
explain this pattern in the interviews.
Future directions for research include investigating the stories and mean-
ings that are attached to negatively evaluated role models since they received
descriptions that were much different from those attached to positively evalu-
ated role models. Also, in light of the findings concerning affective evalua-
tions, it would be of interested to collect data from a larger, randomly selected
sample of fathers to see if the patterns uncovered in these data hold across
10 MEN AND MASCULINITIES / Month XXXX

larger and more diverse samples. In addition, examining the role model inter-
view questions from this study in a longitudinal manner beginning with men
before parenthood might allow us to see if the role models men select remain
consistent throughout various parenthood experiences. Finally, a similar
study of mothers would provide interesting comparisons of role model influ-
ences between genders.
REFERENCES
Bandura, A. 1999. Social cognitive theory of personality. In Handbook of personality: Theory
and practice, 2nd ed., edited by L. A. Pervin and O. P. John, 154–96. New York: Guilford.
Bruce, C., and G. L. Fox. 1997. Measuring father involvement among lower-income white and
African American populations. Poster presented at the annual meeting of the National Coun-
cil on Family Relations, Arlington, VA.
———. 1999. Accounting for patterns of father involvement: Age of child, father-child
coresidence, and father role salience. Sociological Inquiry 69:458–76.
Daly, K. J. 1993. Reshaping fatherhood: Finding the models. Journal of Family Issues 14:
510–30.
Hawkins, A. J., S. L. Christiansen, K. P. Sargent, and E. J. Hill. 1993. Rethinking fathers’involve-
ment in child care: A developmental perspective. Journal of Family Issues 14:531–49.
Parke, R. D. 1995. Fathers and families. Status and social conditions of parenting. Vol. 3 of
Handbook of parenting, edited by M. Borenstein, 27–63. Mahwah, NJ: Lawrence Erlbaum.
Pleck, J. H. 1997. Paternal involvement: Levels, sources, and consequences. In The role of the
father in child development, 3rd ed., edited by M. E. Lamb, 66–103. New York:John Wiley.
Sagi, A. 1982. Antecedents and consequences of various degreesof paternal involvement in child
rearing: The Israeli project. In Nontraditional families: Parenting and child development,
edited by M. E. Lamb, 205–32. Hillsdale, NJ: Lawrence Erlbaum.
Strauss, A., and J. Corbin. 1998. Basics of qualitative research: Techniques and procedures for
developing grounded theory. 2nd ed. Thousand Oaks, CA: Sage.
Stueve, J. L., and J. H. Pleck. 2001. Parenting voices:Solo parent identity and co-parent identities
in married parents’ narrative of meaningful parent experiences. Journal of Social and Per-
sonal Relationships 18:691–708.
PLS PROVIDE.
PLS PROVIDE.
PLS PROVIDE.
Masciadrelli et al. / ROLE MODEL PERCEPTIONS 11