TIMOTHY J. BIBLARZ
University of Southern California
JUDITH STACEY
New York University*
Ideal Families and Social Science Ideals
The fundamental conviction that children need
both a mother and a father in the home dominates bipartisan family discourse and influences
weighty social policy in the United States.
What’s more, proponents of this view, including
some social scientists, assert social science
legitimacy for this claim. The preamble to
the 1996 Welfare Reform Act (Personal
Responsibility and Work Opportunity Reconciliation Act of 1996, Pub. L. No. 104 – 193, 110
Stat. 2105, 1996) asserts just that. The Federal
Marriage Initiative that diverts money from welfare to promoting heterosexual marriage rests on
this premise. On these grounds, the New York
Court of Appeals rejected a suit for same-sex
marriage (Hernandez v. Robles, 2006), proponents of Proposition 8 convinced California
voters to overturn their state supreme court’s
ruling in favor of same-sex marriage (McKinley
& Goodstein, 2008), and the state of Florida
successfully defended its ban on gay adoption
rights (Lofton v. Kearney, 2005). Some family
court judges still deny child custody to divorced
lesbian parents on these grounds. Although such
family values are generally identified with the
Republican Party and the Bush administration,
former President Clinton signed the welfare bill
and the Defense of Marriage Act, and President
Obama has repeated similar claims and statistics about children’s needs for fathers and has
Department of Sociology, University of Southern
California, Los Angeles, CA 90089-2539
([email protected]).
*Department of Sociology, New York University, 295
Lafayette St., 4th floor, New York, NY 10012.
continued many Bush-era marriage promotion
policies.
The social science research that is routinely
cited, however, does not actually speak to the
question of whether or not children need both a
mother and a father at home. Instead, proponents
generally cite research that compares such families with single parents, thus conflating the number with the gender of parents. At the same time,
recurrent claims about the risks of fatherlessness
routinely ignore research on same-gender parents that actually can speak directly to the issue.
This state of affairs was the launch pad for
our article. Much more than hoping to revive
an academic debate about the significance of
fathers for their children’s welfare, as commentator Strohschein believes, or deciding to
conduct another review of research on the effects
of lesbian parenthood, as Tasker seems to have
construed our project, we set out to ask, ‘‘What is
the best that social science research can do with
the question posed in the public domain, rather
than as a nuanced social scientist might wish
to frame it, of whether or not the mother-andfather family is best for children?’’ As one of us
has discussed elsewhere (Stacey, 1997, 2004b),
and although we certainly wish it were otherwise, courts and policymakers overwhelmingly
consider only positivist, quantitative research
evidence to qualify as true social science. And
so we set out to determine how close existing
family research of that sort can come to addressing whether children need both a mother and
father in the home. Although we largely agree
with Goldberg that gender and sexual orientation are ultimately inextricable, we attempted
to review all of the studies we could locate
Journal of Marriage and Family 72 (February 2010): 41 – 44
DOI:10.1111/j.1741-3737.2009.00682.x
41
42
that mitigated the widespread error of conflating
family structure variables that are extricable.
Unlike prior reviews of the research on lesbian and gay parenthood (e.g., Patterson, 2000;
Stacey & Biblarz 2001; Tasker, 2005) or the
vast literature that discusses the effects of marriage, divorce, and single parenthood on children
(e.g., McLanahan & Sandefur, 1994; Waite &
Gallagher, 2000), we examined two distinct bodies of research that hold the number of parents
constant—one that compares gay and straight
coparenting couples, the other that compares
female and male single parents.
Limiting our project in this way allowed us
to identify only a few provocative findings that
go beyond the stereotype that women, on average, are more involved with children than men,
but nothing to support the predominant view
that the ideal gender mix of parents is a man
and a woman. In the end, we concluded (like
two of our commentators) that massive selection effects, unequal access to legal marriage,
and the ultimate inextricability of our five core
parental variables (gender, sexuality, number,
and biogenetic and marital status) make the task
of isolating the unique effect of parental gender
on children near insurmountable. Put otherwise,
social science research does not and cannot support the claim that children need both a mother
and a father parenting together.
We would like to thank the commentators for
their careful, detailed feedback on our article.
Although none of them addressed the central
purpose of our methodological strategy, we
agree with many of the individual points raised.
Both Tasker and Goldberg, for example, call
for more research on diversity within gay- and
lesbian-parent families. We concur wholeheartedly. Elsewhere, for example, Stacey (e.g., 2006)
has unpacked the diversity of pathways to (and
away from) parenthood among a diverse sample
of gay men in Los Angeles and has documented
the astonishing complexity and creativity of varied families they formed (sometimes evolving
out of the practice of cruising; Stacey 2004a,
2005, 2006). Goldberg seems to be calling
for just this kind of research. Likewise, how
could we disagree with Goldberg that we might
have attained richer insights had we reviewed
much more of the qualitative research as well?
Indeed, our very decision to coauthor represents
an attempt to bridge the quantitative-qualitative
divide, and we often apply to our analyses
of quantitative findings insights gleaned from
Journal of Marriage and Family
qualitative research. In fact, Biblarz and Savci
(in press) identified new qualitative research on
diversity among lesbian comother families (e.g.,
Hequembourg & Farrell, 1999; Moore, 2008) as
among the more significant advances in LGBT
studies in the past decade and suggested ways
in which queer theory could fruitfully inform
empirical research.
Similarly, we are well aware (in part from
our own past research) that, as Goldberg points
out, ‘‘fatherless’’ families, whether comother
or single-mother families, often include active
male parents, such as noncustodial divorced
fathers, committed sperm donors and their partners, grandfathers, uncles, fictive kin, and so on.
We also take as a given that gender is never
fixed but shaped by social context (e.g., note our
discussion of gay fathers who self-identify as
mothers and of Risman’s 1987 study of single
men who mother). We wish public discourse
were at the stage where such academic insights
were widely understood, or even at the stage
where all family researchers embraced them.
Unfortunately, this is far from the reality. In
short, the quantitative bias in our article is not
ours but the bias of the political, policy, and
legal arenas we set out to address.
Unsurprisingly, we do not agree with all
of the criticisms raised by the commentators.
Our greatest differences are with Strohschein.
She claims, for example, that our article
focuses on family structure (in her terms,
‘‘social address’’) variables while ignoring
family process variables. Family process variables (concerning couple relationships, divisions of family labor, parental control, support,
attachment, involvement, values, activities with
children, and so on) dominate our tables and
our entire narrative, however. In fact, we composed our tables to feature precisely the links
between structure and process that Stohschein
claimed that we missed. She also faults us for
presuming that all of the heterosexual twoparent couples in the studies we review were
legally married even when their marital statuses
were not reported. Although this is technically
accurate, data on marital status of parents in
most of the studies reviewed are available.
For example, counts from Census Survey of
Income and Program Participation (SIPP) or
National Survey of America’s Families (NSAF)
applied to the representative sample of the
Panel Study of Income Dynamics (PSID) suggest that over 90% of the children who lived
Ideal Families and Social Science Ideals
with both of their biological (or adoptive)
parents in Yeung, Sandberg, Davis-Kean, and
Hofferth’s (2005) study were in fact living
with married parents (e.g., Fields, 2001). Moreover, Strohschein’s emphasis on the distinction
between married and cohabiting statuses stresses
precisely the kind of ‘‘social address’’ variables that her critique claimed to challenge.
More troubling, she surveys the cohabitation
‘‘address’’ through a heterosexist lens and delivers the (stereo) typical dysfunctional family message. We wonder whether Strohschein realizes
that most of the coparenting couples in the
research we reviewed are cohabiting couples,
and their family processes seem to be functioning pretty well. They just happen to be lesbian
couples.
We have smaller differences with our other
two interlocutors. Perhaps the most important
issue concerns our decision not to conduct
a statistical meta-analysis of the studies we
reviewed. Tasker faults us for this decision
and devotes much of her attention to favorable
discussion of two meta-analyses of research
on lesbian parenting (Allen & Burrell, 2002;
Crowl, Ahn, & Baker, 2008). We have
reservations about these analyses. It concerns
us that neither study considered the shift from
unplanned to planned lesbian parenthood nor
explored whether outcomes varied by path to
parenthood. Further, Allen and Burrell do not
address the crucial question of whether effect
sizes varied by study characteristics (moderator
analyses are half the reason for conducting a
meta-analysis). Crowl et al.’s study is a very
careful, useful meta-analysis, but its power
is limited by the need to pool effect sizes
derived from very different measurements and
to restrict its analysis to only 19 studies. This
inflates rather than minimizes bias. Because
the samples in this literature tend to skew to
particular demographics, Crowl et al. might have
considered relaxing their criteria to include as
many studies as possible that feature diverse
samples. We concluded, however, that metaanalysis demands a level of aggregating that
would obfuscate rather than illuminate the
handful of preliminary research findings that we
found most intriguing and potentially important
for future research, such as dissolution rates
in comother families. We chose to broadly
summarize findings rather than conduct a
meta-analysis because we believe that there are
not yet enough studies targeted to the same
43
outcome for a meta-analysis to provide much
added value.
In the end, however, our goal was not to persuade all social scientists of the wisdom of any
of our particular methodological choices or our
interpretations of particular studies. Rather, our
article attempted to persuade more researchers,
policymakers, and citizens that, as our concluding lines put it, ‘‘to ascertain whether any
particular form of family is ideal would demand
sorting a formidable array of often inextricable family and social variables. We predict that
even ‘ideal’ research designs will find instead
that ideal parenting comes in many different
genres and genders.’’
REFERENCES
Allen, M., & Burrell, N. (2002). Sexual orientation
of the parent: The impact on the child. In
M. Allen & R. W. Preiss (Eds.), Interpersonal
communication research: Advances through metaanalysis (pp. 125 – 143). Mahwah, NJ: Usum
Associates.
Biblarz, T. J., & Savci, E. (in press). Lesbian, gay,
bisexual, and transgender families. Journal of
Marriage and Family.
Crowl, A. L., Ahn, S., & Baker, J. (2008). A metaanalysis of developmental outcomes for children
of same-sex and heterosexual parents. Journal of
GLBT Family Studies, 4, 385 – 407.
Fields, J. (2001). Living arrangements of children:
Fall 1996. Current population reports, P70 – 74.
Washington, DC: U.S. Census Bureau.
Hequembourg, A. L., & Farrell, M. P. (1999).
Lesbian motherhood: Negotiating marginalmainstream identities. Gender and Society, 13,
540 – 557.
Hernandez v. Robles. 2006. 7 N.Y.3d 338, 821
N.Y.S.2d 770 2006 N.Y. Slip Op. 05239. New
York Court of Appeals (Justice Robert S. Smith).
July 6.
Lofton v. Kearney. (2005). 93 F. Supp. 2d. 1343
(S.D. Fla. 2000); 157 F. Supp. 2d. 1372 (S.D. Fla.
2001), aff’d., 358 F.3d 804 (11th Cir. 2004), reh’g
en banc denied, 377 F.3d 1275 (11th Cir. 2005),
cert. denied, 125 S. Ct. 869 (2005).
McKinley, J., & Goodstein, L. (2008, November 6).
Bans in 3 states on gay marriage. New York Times,
p. A1.
McLanahan, S., & Sandefur, G. (1994). Growing up
with a single parent. Cambridge, MA: Harvard
University Press.
Moore, M. R. (2008). Gendered power relations
among women: A study of household decision
making in Black, lesbian stepfamilies. American
Sociological Review, 73, 335 – 356.
44
Patterson, C. J. (2000). Family relationships of
lesbians and gay men. Journal of Marriage and
the Family, 62, 1052 – 1069.
Personal Responsibility and Work Opportunity
Reconciliation Act Pub. L. 104 – 193, 110 Stat.
2105 (1996).
Risman, B.J. (1987). Intimate relationships from a
microstructural perspective: Men who mother.
Gender and Society, 1, 6 – 32.
Stacey, J. (1997). In the name of the family. Boston:
Beacon Press.
Stacey, J. (2004a). Cruising to familyland: Gay hypergamy and rainbow kinship. Current Sociology, 52,
181 – 197.
Stacey, J. (2004b). Marital suitors court social science
spin-sters: The unwittingly conservative effects of
public sociology. Social Problems, 51, 131 – 145.
Journal of Marriage and Family
Stacey, J. (2005). The families of man: Gay male
intimacy and kinship in a global metropolis. Signs,
30, 1911 – 1933.
Stacey, J. (2006). Gay parenthood and the decline of
paternity as we knew it. Sexualities, 9, 27 – 55.
Stacey, J., & Biblarz, T. J. (2001). (How) does the
sexual orientation of parents matter? American
Sociological Review, 66, 159 – 183.
Tasker, F. (2005). Lesbian mothers, gay fathers,
and their children: A review. Developmental and
Behavioral Pediatrics, 26, 224 – 240.
Waite, L., & Gallagher, M. (2000). The case for
marriage. New York: Doubleday.
Yeung, W. J., Sandberg, J. F., Davis-Kean, P. E., &
Hofferth, S. L. (2001). Children’s time with fathers
in intact families. Journal of Marriage and Family,
63, 136 – 154.