Journal of Family Psychology © 2011 American Psychological Association

2011, Vol. 25, No. 6, 986 –992 0893-3200/11/$12.00 DOI: 10.1037/a0025855

BRIEF REPORT

The Association Between Paternal Sensitivity and Infant–Father

Attachment Security: A Meta-Analysis of Three Decades of Research
Nicole Lucassen Anne Tharner
Erasmus University Medical Center, Rotterdam, Leiden University, The Netherlands
The Netherlands
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Marinus H. Van IJzendoorn Marian J. Bakermans-Kranenburg

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Leiden University and Erasmus University, Leiden University, The Netherlands

The Netherlands

Brenda L. Volling Frank C. Verhulst,

University of Michigan Mijke P. Lambregtse-Van den Berg, and
Henning Tiemeier
Erasmus University Medical Center, Rotterdam,
The Netherlands

For almost three decades, the association between paternal sensitivity and infant–father attach-
ment security has been studied. The first wave of studies on the correlates of infant–father
attachment showed a weak association between paternal sensitivity and infant–father attachment
security (r ⫽ .13, p ⬍ .001, k ⫽ 8, N ⫽ 546). In the current paper, a meta-analysis of the
association between paternal sensitivity and infant–father attachment based on all studies currently
available is presented, and the change over time of the association between paternal sensitivity and
infant–father attachment is investigated. Studies using an observational measure of paternal
interactive behavior with the infant, and the Strange Situation Procedure to observe the attachment
relationship were included. Paternal sensitivity is differentiated from paternal sensitivity combined
with stimulation in the interaction with the infant. Higher levels of paternal sensitivity were
associated with more infant–father attachment security (r ⫽ .12, p ⬍ .001, k ⫽ 16, N ⫽ 1,355).
Fathers’ sensitive play combined with stimulation was not more strongly associated with attach-
ment security than sensitive interactions without stimulation of play. Despite possible changes in
paternal role patterns, we did not find stronger associations between paternal sensitivity and infant
attachment in more recent years.

Keywords: father, sensitivity, stimulation, attachment, meta-analysis

Attachment security represents the child’s trust in his or environment (Bowlby, 1969). Maternal sensitivity plays a
her caregiver, and is evident from the child’s preferential crucial role in shaping the mother–infant attachment rela-
desire for contact with the caregiver in times of stress and tionship. Sensitivity is defined as the ability to perceive and
use of the caregiver as a “secure base” to explore the to interpret accurately the signals and communications im-

This article was published Online First October 17, 2011. igan; Frank C. Verhulst, Mijke P. Lambregtse-Van den Berg,
Nicole Lucassen, The Generation R Study Group and Depart- Department of Child and Adolescent Psychiatry/Psychology, Eras-
ment of Child and Adolescent Psychiatry/Psychology, Erasmus mus University Medical Center, The Netherlands; Henning
University Medical Center, Rotterdam, The Netherlands; Anne Tiemeier, Department of Child and Adolescent Psychiatry/
Tharner, Centre for Child and Family Studies, Leiden University, Psychology and Department of Epidemiology, Erasmus University
The Netherlands; Marinus H. Van IJzendoorn, Centre for Child and Medical Center, The Netherlands.
Family Studies, Leiden University and Erasmus School of Peda- Correspondence concerning this article should be addressed to
gogical and Educational Sciences (ESPES), Erasmus University, Henning Tiemeier, Department of Child and Adolescent
The Netherlands; Marian J. Bakermans-Kranenburg, Centre for Psychiatry/Psychology and Department of Epidemiology, Erasmus
Child and Family Studies, Leiden University, The Netherlands; University Medical Center, PO Box 2060, 3000 CB Rotterdam, The
Brenda L. Volling, Department of Psychology, University of Mich- Netherlands. E-mail: [email protected]

986
 META-ANALYSIS SENSITIVITY INFANT-FATHER ATTACHMENT 987

plicit in the infant’s behavior, and given this understanding, shown positive correlates with a secure father– child attach-
to respond to them appropriately and promptly (Ainsworth, ment relationship. For example, children with a secure
Blehar, Waters, & Wall, 1978). Examples of sensitivity attachment relationship with their father have fewer behav-
include contingent vocalizations, encouragement of the ior problems (Verschueren & Marcoen, 1999) and show
child’s efforts, and soothing the infant in times of distress. higher levels of sociability (Lamb, Hwang, Frodi, & Frodi,
The association between maternal sensitivity and the 1982). In this paper, a meta-analysis of the association
mother–infant attachment relationship is a frequently repli- between paternal sensitivity and infant–father attachment is
cated finding, not only in correlational studies (see De Wolff presented, elaborating on the meta-analysis conducted by
& Van IJzendoorn, 1997, for a meta-analysis), but also in Van IJzendoorn and De Wolff in 1997. Through cumulative
randomized experimental intervention studies. Interventions meta-analysis (Borenstein, Hedges, Higgings, & Rothstein,
that most successfully manage to enhance the level of 2009), we document the change in effect sizes across the
maternal sensitivity also create the largest increase in at- past three decades of research. This cumulative meta-
tachment security (Bakermans-Kranenburg, Van IJzen- analysis is important for the field of father studies, as fathers
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

doorn, & Juffer, 2003), which suggests a causal relation seem to have become gradually more involved in positive
This document is copyrighted by the American Psychological Association or one of its allied publishers.

between sensitivity and attachment. engagement activities with their children over the last few
At the core of attachment theory is the claim that infants decades, which might affect the relationship with their child
not only become attached to their biological mother but also (Pleck, 2010). In a cumulative meta-analysis, new research
to other caregivers who interact regularly with them. Al- findings are added to what was previously known. Each
though biological mothers were the target of most studies on time a new study is added, a separate meta-analysis is
sensitivity and infant attachment, several studies focused on conducted (Muellerleile & Mullen, 2006). Typically, each
the role of paternal sensitivity in infant–father attachment. new study reduces the confidence intervals around an in-
The first wave of studies on the correlates of infant–father creasingly precise estimate (Hanson & Broom, 2005). A
attachment showed a weak association between paternal cumulative meta-analysis has more statistical power, which
sensitivity and infant–father attachment security. Van IJzen- is especially relevant to a set of studies in which small effect
doorn and De Wolff (1997) reported that the combined sizes are observed (Mulrow, 1994). Special attention in this
effect size across eight studies amounted to a correlation of analysis is given to the potentially different roles of paternal
.13, whereas the effect size for maternal sensitivity and sensitivity and paternal sensitivity combined with stimula-
infant–mother attachment was estimated to be r ⫽ .24 in a tion. We expect that higher levels of paternal sensitive
much larger set of 21 studies (De Wolff & Van IJzendoorn, warmth, and in particular when co-occurring with sensitive
1997). More recent studies on fathers show mixed results, stimulation, are associated with more infant–father attach-
with some failing to find a significant relation between ment security.
paternal sensitivity and attachment security (Braungart-
Rieker, Garwood, Powers, & Wang, 2001; Volling, Method
McElwain, Notaro, & Herrera, 2002), and others report-
ing a significant but modest association (Eiden, Edwards, Dataset
& Leonard, 2002; Lucassen et al., 2010).
At least two reasons for the rather weak association between For our meta-analysis, we systematically searched the
paternal sensitivity and attachment may be mentioned. First, databases PsycInfo, Social Sciences Citation Index, Educa-
fathers spend less time with their child compared to mothers, tional Resources Information Center, and ProQuest Disser-
who are usually the primary caregiver. The link between their tations and Theses with the keywords father, fathering,
parenting behavior and the attachment relationship may be paternal, sensitivity, responsivity, synchrony, warmth, par-
more tenuous than in the case of mothers. Research on infant enting, and attachment in the title or abstract. We searched
attachment and sensitivity of highly involved fathers is scarce the online available integrative statements of the paper
and equivocal (Cox, Owen, Henderson, & Margand, 1992; symposia organized at the biennial meetings of the Society
Lucassen et al., 2010). Second, fathers may interact with their for Research in Child Development (2009 –2011). Further-
infants in somewhat different ways compared to mothers. more, we used the search engine Google Scholar with the
Traditionally, fathers have been described as focused on stim- keywords father, sensitivity, and attachment to identify rel-
ulating and exploratory play interactions with their children, evant studies. In addition to the electronic searches, we
with less emphasis on emotional support and warmth (Gross- searched recently published books on attachment, fathering,
mann, Grossmann, Kindler, & Zimmermann, 2008). Mothers or parenting with contributions from international experts
might relate to their infants with sensitive warmth, whereas on these topics (Cummings & Cummings, 2002; Grossmann
fathers might choose sensitive stimulation as a way to promote et al., 2008; Lamb & Lewis, 2010; Parke, 2002). Infant–
feelings of security in their infants. Stimulation can be de- father attachment studies conducted before 1996 were de-
scribed as any activating interaction on the part of the parent rived from the first meta-analysis on this topic by Van
directed toward the infant in order to promote his or her IJzendoorn and De Wolff (1997). Lastly, the reference lists
exploration or playful behavior (De Wolff & Van IJzendoorn, of the collected papers were searched for relevant studies.
1997). We finished the search in May, 2011.
Examining determinants of the attachment relationship We detected 1,613 studies with our search terms. Our
with the infant is important, since several studies have search was restricted to studies using an observational mea-
 988 LUCASSEN ET AL.

sure of paternal interactive behavior with the infant, and the formed through random effects models (Borenstein et al.,
gold standard assessment of attachment security, the 2009). We computed the 95% confidence interval around
Strange Situation Procedure (SSP, Ainsworth et al., 1978). the mean effect size. Q-statistics and their p values were
After examining the abstracts of all studies, 22 studies were also computed to assess differences between combined ef-
relevant for our meta-analysis. Six studies presented data on fect sizes for specific subsets of study effect sizes grouped
(partly) overlapping samples. When samples of studies were by moderators. Contrasts were only tested when at least two
overlapping, the papers that reported on the largest groups of the subsets consisted of at least four studies (Bakermans-
of participants were included in our meta-analysis. We Kranenburg et al., 2003). Again, random effects model tests
found one dissertation abstract (ProQuest document ID: were used. For continuous moderators, Fisher’s Z-scores
2105027951) with measurements of sensitivity and attach- were used in weighted least squares metaregression analy-
ment in “parents”, but it was unclear whether fathers were ses. The “trim and fill” method was used to test the influence
included in the study, and this graduate work was not of possible adjustments of the sets of studies for publication
available for purchase at the request of the author, thus we bias (Duval & Tweedie, 2000a, 2000b). No outliers (stan-
dardized z-values smaller than ⫺3.29 or larger than 3.29;
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

could not determine what type of measurements were used.

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Finally, we identified 16 pertinent effect sizes on 1,355 Tabachnick & Fidell, 2001) were found for study effect
subjects, which are presented in Table 1. sizes.

Moderators Results
We coded the following study characteristics: year of Table 2 presents the combined effect size of the 16
publication, sample size, the age of the child at the time of studies as well as confidence intervals, homogeneity tests,
assessment of attachment security (for the purpose of mod- and contrast tests. The combined effect size for the associ-
erator analysis, age was split in two sets: younger than 16 ation between paternal sensitivity and infant–father attach-
months of age vs. 16 months and older), study design ment was r ⫽ .12 (95% CI .06, .17, k ⫽ 16, N ⫽ 1,355, p ⬍
(concurrent vs. predictive), percentage of secure infant– .001). The set of study outcomes was homogeneous, Q
father attachment relationships, and measurement of sensi- (15) ⫽ 12.80, p ⫽ .60. Using the trim and fill method
tivity (sensitivity or sensitivity combined with stimulation). (Duval & Tweedie, 2000a, 2000b), we did not find asym-
The selection of these moderator variables was based on metry in the funnel plot (precision plot, random effects),
theoretical and empirical reasons. We focused on the child’s which suggested the absence of a potential publication bias.
age as a moderator since the quality of the relationship Moderator analyses did not show significant contrasts for
between father and child may depend on the developmental measurement of sensitivity, the age of the child at the time
phase of the child. Also, the meta-analysis of infant–mother of assessment of attachment security, and study design (see
dyads (De Wolff & Van IJzendoorn, 1997) showed that the Table 2). The metaregressions with the continuous moder-
age of the infants at the time of the attachment assessment ators “year of publication,” “percentage secure,” “age of the
was a significant moderator, with older infants showing child at the time of assessment of attachment,” and “sample
stronger effect sizes. A concurrent study design may show size” did not show significant effects either (z ⫽ ⫺0.25, p ⫽
stronger effect sizes than a predictive design because a .80, z ⫽ 0.89, p ⫽ .38, z ⫽ ⫺1.44, p ⫽ .15, and z ⫽ 0.09,
possibly bidirectional association between sensitivity and p ⫽ .93, respectively).
attachment could contribute to the strength of the observed A cumulative meta-analysis confirmed the absence of an
effects in a concurrent design. Also, the meta-analysis of De association between year of publication and effect size. In
Wolff and Van IJzendoorn (1997) found that a shorter time Figure 1 the development of the combined effect size across
interval between the sensitivity and attachment assessments time is presented, and a trend toward stronger associations
showed stronger effect sizes. All of the included studies between paternal sensitivity and infant security did not
used either a predictive or a concurrent design; none of the appear to be present.
studies used a combination of predictive and concurrent
measurements. Lastly, as described in the introduction, we Discussion
distinguished studies in which the traditional concept of
sensitivity (warmth, emotional support, responsiveness) was In this meta-analysis, we showed that higher levels of
observed from studies in which the quality of stimulating paternal sensitivity were associated with more infant–father
and exploratory play was additionally observed. attachment security. The association between sensitivity
and attachment security was reliable, but weak. Yet, this
Data Analysis finding contributes to the field of father research in several
ways. First, this meta-analysis included twice as many stud-
The Comprehensive Meta-Analysis (CMA) program was ies with more than twice as many infant–father dyads com-
used to transform the results of the individual studies into pared to the meta-analysis conducted by Van IJzendoorn
the common metric of Pearson’s product-moment correla- and De Wolff in 1997. Thus, it provides an update of studies
tion coefficients (r) and to combine weighted effect sizes on paternal sensitivity and attachment. Second, we were
(Borenstein et al., 2009). Heterogeneity across studies was able to examine whether a trend toward stronger associa-
assessed using the Q-statistic. Significance tests were per- tions between paternal sensitivity and infant attachment
 This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Table 1
Infant–Father Attachment Studies: Descriptives and Effect Sizes
Measurement of sensitivity Design
Age observation
S ⫽ sensitivity sensitivity C ⫽ concurrent Age Strange Situation Secure Effect size
Author(s) N SS ⫽ sensitivity & stimulation (months) P ⫽ predictive Procedure (months) (%) (r)
ⴱ
Belsky (1983) 51 Overall engagement, response, stimulation, 1, 3, 9 P 13 63 .11
care giving, positive affection (SS)
ⴱ
Easterbrooks (1984) 75 Emotional supportiveness, quality of 20.5 C 20.5 66 .00
assistance (SS)
ⴱ
Goossens (1990) 75 Sensitivity (S) 15 C 15 64 .26
ⴱ
Cox (1992) 33 Sensitivity, warmth, level of activity, 3 P 12.5 63 .43
stimulation (SS)
ⴱ
Grossmann (1992) 46 Interactive smoothness, empathy (S) 2, 6, 10 P 18 41 .03
ⴱ
Volling (1992) 113 Responsiveness, stimulation (SS) 3, 9 P 13 80 .16
ⴱ
Schneider Rosen (1993) 62 Paternal acceptance, quality of assistance 21.5 C 21.5 69 .00
(SS)
ⴱ
Caldera (1995) 90 Positive regard, sensitivity to nondistress, 6 P 18 68 .16
intrusiveness (SS)
Braungart-Rieker (2001) 86 Sensitivity, intrusiveness (SS) 4 P 13 68 .11
Volling (2002) 62 Sensitivity, intrusiveness (SS) 12.5 C 12.5 60 .00
Eiden (2002) 220 Sensitive responding, negative affect, 13.5 C 13.5 61 .14
positive involvement (S)
Kochanska (2005) 100 Responsiveness (S) 7 P 15 66 .00
Schoppe-Sullivan (2006) 91 Sensitivity (S) 12 C 12 64 .17
Wong (2009) 62 Sensitivity (S) 3.5 P 13 64 .24
Hazen (2010) 105 Sensitivity (S) 8 P 13.5 59 .00
Lucassen (2010) 84 Sensitivity, cooperation (SS) 14 C 14 68 .20
META-ANALYSIS SENSITIVITY INFANT-FATHER ATTACHMENT

Note. References marked with an asterisk indicate studies included in the meta-analysis conducted by Van IJzendoorn and De Wolff (1997).
989
 990 LUCASSEN ET AL.

Table 2
Meta-Analytic Results on the Association Between Paternal Sensitivity and Infant–Father Attachment Security
Confidence interval Homogeneity Contrast Contrast
k N r 95% Q Q P
Total set 16 1,355 .12ⴱⴱ .06, .17 12.80
Sensitivity measure 0.00 .96
ⴱⴱ
sensitivity 7 569 .12 .03, .20 6.04
sensitivity ⫹ stimulation 9 786 .12ⴱⴱ .05, .19 6.77
Design 0.02 .89
concurrent 7 669 .12ⴱⴱ .05, .20 4.85
predictive 9 686 .11ⴱⴱ .04, .19 7.94
Age SSP 1.15 .28
12 to 15 months 12 1,082 .13ⴱⴱ .07, .19 10.25
18 months and older 4 273 .06 ⫺.06, .18 1.40
ⴱ ⴱⴱ
p ⬍ .05. p ⬍ .01.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

security would emerge across the past three decades. Al- sitive interactions without stimulation; nor did other mod-
though family sociologists report on increased participation erator analyses (year of publication, sample size, the age of
in positive engagement activities of the father with their the child at the time of assessment of attachment security,
child (Pleck, 2010), we failed to find evidence for its in- percentage secure, and study design) show significant ef-
creasing effect on infant attachment. This might, however, fects. It should be noted that these moderator analyses need
be influenced by selective participation of fathers in family to be interpreted with caution because of the small number
research. Fathers who participate in research tend to be of studies. The relatively small number of studies is a
more involved in family life than nonparticipating fathers general limitation of this meta-analysis.
(Costigan & Cox, 2001), which might result in less vari- The quality of parenting behavior accounts for a small
ability in parenting behaviors like sensitivity. Third, be- portion of the explained variance in individual differences
cause of the larger number of studies on fathers, we were in attachment security. The small effect size of the associ-
able to differentiate the traditional assessment of sensitivity ation between paternal sensitivity and infant–father attach-
from sensitivity combined with stimulation which is sug- ment seems to imply that little can be gained from preven-
gested to be a paternal “specialization” (Grossmann et al., tion or intervention strategies directed at increasing paternal
2008; Naber, Van IJzendoorn, Deschamps, Van Engeland, sensitivity. Nevertheless, interventions based on small ef-
& Bakermans-Kranenburg, 2010). However, we failed to fect sizes can yield substantial benefits. A striking example
find evidence for the hypothesis that fathers’ sensitivity was shown in a major biomedical research on the associa-
combined with stimulation during play and other challeng- tion between aspirin intake and risk of myocardial infarction
ing situations promotes attachment security more than sen- (heart attack). Although the effect size of .034 was consid-

Figure 1. Forest plot of the studies included in the cumulative meta-analysis with the effect sizes
and 95% confidence intervals. The combined effect size is r ⫽ .12; k ⫽ 16; N ⫽ 1,355; p ⬍ .001.
 META-ANALYSIS SENSITIVITY INFANT-FATHER ATTACHMENT 991

ered to be very small, it suggested that 3.4% fewer persons *Belsky, J. (1983). Father-infant interaction and security of at-
who would probably experience a heart attack will not tachment: No relationship. Unpublished manuscript, The Penn-
experience it if they follow the regimen as prescribed in the sylvania State University, University Park, Pennsylvania.
aspirin treatment condition (Rosnow & Rosenthal, 1989). Belsky, J. (1996). Parent, infant, and social-contextual antecen-
As Rose described (in his discussion of the “prevention dents of father-son attachment security. Developmental Psychol-
ogy, 32, 905–913. doi:10.1037//0012-1649.32.5.905
paradox”; 1981), even moderate alterations of modest risk Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R.
factors can achieve major public health benefits. (2009). Introduction to meta-analysis. Chichester, England: Wi-
The significant and replicable effect of modest size may ley. doi:10.1002/9780470743386
be relevant as few father determinants of infant–father at- Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New
tachment are established and because a secure attachment York: Basic Books.
relationship is predictive for positive developmental out- *Braungart-Rieker, J. M., Garwood, M. M., Powers, B. P., &
comes of the child. Moreover, sensitivity is amenable to Wang, X. (2001). Parental sensitivity, infant affect, and affect
interventions. Intervention studies aimed at increasing sen- regulation: Predictors of later attachment. Child Development,
72, 252–270. doi:10.1111/1467-8624.00277
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

sitivity and attachment security in mothers and their infants

This document is copyrighted by the American Psychological Association or one of its allied publishers.

appeared to be effective, and more successful sensitivity Brown, G. L., Schoppe-Sullivan, S. J., Mangelsdorf, S. C., & Neff,
interventions were also more effective in enhancing infant C. (2010). Observed and reported supportive coparenting as
attachment security (Bakermans-Kranenburg et al., 2003). predictors of infant-mother and infant-father attachment security.
Early Child Development and Care, 180, 121–137. doi:10.1080/
Interventions involving fathers appeared to be more effec- 03004430903415015
tive than interventions focusing on mothers only. The num- *Caldera, Y. M., Huston, A., & O’Brien, M. (1995). Antecedents
ber of intervention studies on fathers is too small to be of father-infant attachment: A longitudinal study. Paper pre-
included in a meta-analysis in a way similar to the meta- sented at the meetings of the Society for Research in Child
analysis on mother–infant interventions (Bakermans- Development, Indianapolis, IN.
Kranenburg et al., 2003). Although most studies in the Costigan, C. L., & Cox, M. J. (2001). Fathers’ participation in
current meta-analysis had a predictive design with the as- family research: Is there a self-selection bias? Journal of Family
sessment of sensitivity conducted several months before the Psychology, 15, 706 –720. doi:10.1037/0893-3200.15.4.706
Strange Situation Procedure, only experimental intervention Cowan, P. A. (1997). Beyond meta-analysis: A plea for a family
studies can establish a causal relation. systems view of attachment. Child Development, 68, 601– 603.
Besides sensitivity and stimulating play, other determi- doi:10.2307/1132111
*Cox, M. J., Owen, M. T., Henderson, V. K., & Margand, N. A.
nants may facilitate a secure infant–father attachment rela-
(1992). Prediction of infant-father and infant-mother attachment.
tionship. Previous studies have shown that fathering is in- Developmental Psychology, 28, 474 – 483. doi:10.1037//0012-
fluenced, more than mothering, by contextual factors in the 1649.28.3.474
family such as marital satisfaction (Belsky, 1996) and co- Cummings, E. M., & Cummings, J. S. (2002). Parenting and
parental relationship quality (Brown, Schoppe-Sullivan, attachment. In M. H. Bornstein (Ed.), Handbook of parenting:
Mangelsdorf, & Neff, 2010). Studies focusing on the family Vol. 5. Practical issues in parenting, 2nd ed., (pp. 35–58).
system (e.g., Schoppe-Sullivan et al., 2006) might give Mahwah, NJ: Erlbaum.
more insight into the specific role of the father in the De Wolff, M. S., & Van IJzendoorn, M. H. (1997). Sensitivity and
development of the child. Such studies could also examine attachment: A meta-analysis on parental antecedents of infant
whether the father has a more direct or a more buffering attachment. Child Development, 68, 571–591. doi:10.1111/
effect on child development. Future research should focus j.1467-8624.1997.tb04218.x
on the causal relation between paternal sensitivity and Duval, S., & Tweedie, R. (2000a). Trim and fill: A simple funnel-
plot– based method of testing and adjusting for publication bias
infant–father attachment, as well as other possible determi-
in meta-analysis. Biometrics, 56, 455– 463. doi:10.1002/
nants of infant–father attachment security. Given the few 0470870168
extant studies looking at the infant–father attachment rela- Duval, S., & Tweedie, R. (2000b). A nonparametric “trim and fill”
tionship, we recommend that future studies utilize a family method of accounting for publication bias in meta-analysis.
systems perspective (Cowan, 1997) in the continuing search American Psychologist, 95, 89 –98. doi:10.2307/2669529
for the multiple predictors of infants’ secure attachments to *Easterbrooks, M. A., & Goldberg, W. A. (1984). Toddler devel-
their fathers. opment in the family: Impact of father involvement and parent-
ing characteristics. Child Development, 55, 740 –752. doi:
10.2307/1130126
References *Eiden, R. D., Edwards, E. P., & Leonard, K. E. (2002). Mother-
infant and father-infant attachment among alcoholic families.
References marked with an asterisk indicate studies included in Development and Psychopathology, 14, 253–278. doi:10.1017/
the meta-analysis. S0954579402002043
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). *Goossens, F. A., & Van IJzendoorn, M. H. (1990). Quality of
Patterns of attachment: A psychological study of the Strange infants’ attachments to professional caregivers: Relation to infant-
Situation. Hillsdale, NJ: Erlbaum. parent attachment and day-care characteristics. Child Development,
Bakermans-Kranenburg, M. J., Van IJzendoorn, M. H., & Juffer, 61, 832– 837. doi:10.1111/j.1467-8624.1990.tb02825.x
F. (2003). Less is more: Meta-analyses of sensitivity and attach- *Grossmann, K., & Grossmann, K. E. (1992). Newborn behavior,
ment interventions in early childhood. Psychological Bulletin, the quality of early parenting and later toddler-parent relation-
129, 195–215. doi:10.1037/0033-2909.129.2.195 ships in a group of German infants. In J. K. Nugent, B. M. Lester,
 992 LUCASSEN ET AL.

& T. B. Brazelton (Eds.), The cultural context of infancy, Vol. 2 roendocrinology, 35, 1583–1586. doi:10.1016/j.psyneuen.2010
(pp. 3–38). Norwood, NJ: Ablex. .04.007
Grossmann, K., Grossmann, K. E., Kindler, H., & Zimmermann, Parke, R. D. (2002). Fathers and families. In M. H. Bornstein
P. (2008). A wider view of attachment and exploration: The (Ed.), Handbook of parenting: Vol. 3. Being and becoming a
influence of mothers and fathers on the development of psycho- parent (2nd ed., pp. 27–73). Mahwah, NJ: Erlbaum.
logical security from infancy to young adulthood. In J. Cassidy & Pleck, J. H. (2010). Paternal involvement: Revised conceptualiza-
P. R. Shaver (Eds.), Handbook of attachment: Theory, research, tion and theoretical linkages with child outcomes. In M. E. Lamb
and clinical applications (pp. 857– 879). New York: The Guil- (Ed.), The role of the father in child development (5th ed., pp.
ford Press. 58 –93). Hoboken, NJ: Wiley.
Hanson, R. K., & Broom, I. (2005). The utility of cumulative Rose, G. (1981). Strategy of prevention: Lessons from cardiovas-
meta-analysis: Application to programs for reducing sexual vio- cular disease. British Medical Journal, 282, 1847–1851. doi:
lence. Sexual Abuse: A Journal of Research and Treatment, 17, 10.1136/bmj.282.6279.1847
357–373. doi:10.1177/107906320501700402 Rosnow, R. L., & Rosenthal, R. (1989). Statistical procedures and
the justification of knowledge in psychological science. Ameri-
*Hazen, N. L., McFarland, L., Jacobvitz, D., & Boyd-Soisson, E.
can Psychologist, 44, 1276 –1284. doi:10.1037//0003-066X.44
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

(2010). Fathers’ frightening behaviours and sensitivity with

.10.1276
This document is copyrighted by the American Psychological Association or one of its allied publishers.

infants: Relations with fathers’ attachment representations,

*Schneider Rosen, K., & Rothbaum, F. (1993). Quality of parental
father-infant attachment, and children’s later outcomes. Early caregiving and security of attachment. Developmental Psychol-
Child Development and Care, 180, 51– 69. doi:10.1080/ ogy, 29, 358 –367. doi:10.1037//0012-1649.29.2.358
03004430903414703 *Schoppe-Sullivan, S. J., Diener, M. L., Mangelsdorf, S. C.,
*Kochanska, G., Aksan, N., & Carlson, J. J. (2005). Temperament, Brown, G. L., McHale, J. L., & Frosch, C. A. (2006). Attachment
relationships, and young children’s receptive cooperation with and sensitivity in family context: The roles of parent and infant
their parents. Developmental Psychology, 41, 648 – 660. doi: gender. Infant and Child Development, 15, 367–385. doi:
10.1037/0012-1649.41.4.648 10.1002/icd.449
Lamb, M. E., Hwang, C. P., Frodi, A. M., & Frodi, M. (1982). Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate
Security of mother- and father-infant attachment and its relation statistics. New York: Harper & Row.
to sociability with strangers in traditional and non-traditional Van IJzendoorn, M. H., & De Wolff, M. S. (1997). In search of the
Swedish families. Infant Behavior and Development, 5, 355–367. absent father—Meta-analyses of infant–father attachment: A re-
doi:10.1016/S0163-6383(82)80046-5 joinder to our discussants. Child Development, 68, 604 – 609.
Lamb, M. E., & Lewis, C. (2010). The development and signifi- doi:10.1111/j.1467-8624.1997.tb04223.x
cance of father-child relationships in two-parent families. In Verschueren, K., & Marcoen, A. (1999). Representation of self
M. E. Lamb (Ed.), The role of the father in child development and socioemotional competence in kindergartners: Differential
(5th ed., pp. 94 –153). Hoboken, NJ: Wiley. and combined effects of attachment to mother and to father.
*Lucassen, N., Tharner, A., Van IJzendoorn, M. H., Bakermans- Child Development, 70, 183–201. doi:10.1111/1467-8624.00014
Kranenburg, M. J., Volling, B. L., Hofman, A., . . . Tiemeier, H. *Volling, B. L., & Belsky, J. (1992). Infant, father, and marital
(2010). Paternal sensitivity and history of depression/anxiety antecedents of infant-father attachment security in dual-earner
predict infant-father attachment security. Unpublished manu- and single-earner families. International Journal of Behavioral
script, Department of Child and Adolescent Psychiatry, Erasmus Development, 15, 83–100. doi:10.1177/016502549201500105
University Medical Center, Rotterdam, the Netherlands/ *Volling, B. L., McElwain, N. L., Notaro, P. C., & Herrera, C.
Department of Child and Family Studies, Leiden University, the (2002). Parents’ emotional availability and infant emotional
Netherlands. competence: Predictors of parent-infant attachment and emerg-
Muellerleile, P., & Mullen, B. (2006). Sufficiency and stability of ing self-regulation. Journal of Family Psychology, 16, 447– 465.
doi:10.1037//0893-3200.16.4.447
evidence for public health interventions using cumulative meta-
*Wong, M. S., Mangelsdorf, S. C., Brown, G. L., Neff, C., &
analysis. Public Health Matters, 96, 515–522. doi:10.2105/
Schoppe-Sullivan, S. J. (2009). Parental beliefs, infant tempera-
AJPH.2003.036343
ment, and marital quality: Associations with infant-mother and
Mulrow, C. D. (1994). Systematic reviews: Rationale for system- infant-father attachment. Journal of Family Psychology, 23,
atic reviews. British Medical Journal, 309, 597–599. 828 – 838. doi:10.1037/a0016491
Naber, F., Van IJzendoorn, M. H., Deschamps, P., Van Engeland,
H., & Bakermans-Kranenburg, M. J. (2010). Intranasal oxytocin Received December 31, 2010
increases fathers’ observed responsiveness during play with their Revision received August 29, 2011
children: A double-blind within-subject experiment. Psychoneu- Accepted September 1, 2011 䡲