вход по аккаунту



код для вставкиСкачать
Further comments on the papers by Marks and Regnerus
Cynthia Osborne
LBJ School of Public Affairs,University of Texas at Austin,Austin,TX 78713,United States
Marks (2012) and Regnerus (2012) push forward the field of family studies with a methodical critique of the current lit-
erature on children of same-sex relationships and rigorous descriptive analysis of the outcomes for children whose parents
have had a same sex relationship.Each author details the compelling limitations of many of the studies that find no disad-
vantages associated with being raised by same-sex parents,and Regnerus contradicts the ‘‘no differences’’ findings by using a
representative,population-based sample of young adults whose parents had a same-sex relationship.
Each study is solid and makes a valuable contribution to the field.In the end,however,we are left with little understand-
ing about why family scholars should conduct research on children of same-sex relationships,why we might expect differ-
ences in outcomes between children of same-sex relationships and other family structures,and what the study findings
imply.This commentary begins with a brief discussion of the two studies and then addresses these questions in turn.
2.A more rigorous standard of research
Marks (2012) carefully critiques the American Psychological Association’s Brief on Lesbian and Gay Parenting,and asserts
that the brief’s claims of ‘‘no differences’’ between children of same-sex parents and other family arrangements are not fully
supported by the extant research.The author addresses seven questions that outline standards for research on the children
of gay and lesbian parents.Studies that fall short of these research standards cannot make generalizable conclusions about
this population.
Regnerus (2012) examines differences in young adult outcomes associated with eight different childhood family forms.
He pays particular attention to the differences associated with being raised in an intact biological married-parent family as
compared to being raised by parents who experienced a same-sex relationship.To what extent does the Regnerus study
meet the research standards outlined by Marks?The Regnerus study is one of the most comprehensive and rigorous studies
that has been conducted in this field to date.Still,he is careful to articulate that his analysis identifies differences in out-
comes,but does not provide evidence as to why these differences exist.Moreover,his study cannot isolate the effect of hav-
ing a parent who had a same-sex relationship from the effect of experiencing multiple family forms.
2.1.Diverse populations
One of the major contributions of the Regnerus paper is to introduce the NewFamily Structure Study (NFSS).The NFSS is a
nationally representative sample of almost 3000 young adults.It provides the first moderately large,population-based sam-
ple of children whose parents experienced a same-sex relationship (163 respondents’ mothers had a same-sex relationship
and 73 respondents’ fathers had a same-sex relationship).Similar to prior research,the study provides more information on
children of lesbian relationships than children of fathers who have same-sex relationships,and it also shows that children
are much less likely to live with a gay father as compared to a lesbian mother.
0049-089X/$ - see front matter 2012 Elsevier Inc.All rights reserved.
Tel.:+1 512 471 9808.
Social Science Research 41 (2012) 779–783
Contents lists available at SciVerse ScienceDirect
Social Science Research
j ournal homepage:www.el sevi ocat e/ssresearch
The NFSS paints a new portrait of children whose parents have same-sex relationships.Prior research consists primarily
of small,select samples of children of middle-class,white,openly gay and lesbian parents.By contrast,the NFSS shows that
there is considerable diversity in this population with regard to race and ethnicity,socio-economic status,and the mode by
which children experience the same-sex relationship of their parent.Approximately 43% of the NFSS respondents whose
mothers had a same-sex relationship were Black or Hispanic.
Most importantly,the sample is diverse with regard to the children’s exposure to their parents’ same-sex relationships.
Some respondents lived in planned same-sex parent families.However,most of the respondents who are included in the
group of children of lesbian mothers (LMs) or gay fathers (GFs) spent some portion of their childhood living in heterosexual
and single-parent households as well as in same-sex unions.Today’s children of same-sex parents may be more likely to be
planned jointly by the same-sex parents;however this was less feasible 15 or 20 years ago.
2.2.Comparison groups
The NFSS provides rich family histories for all respondents,painting a portrait of all of the family structures a child expe-
riences.The Regnerus study compares the outcomes of young adults across eight childhood family forms.He primarily fo-
cuses on comparisons between children who live their entire childhood with both of their married,biological parents and
children of lesbian mothers (LMs).
To increase the sample size of children who experienced a same-sex parent,Regnerus included respondents in either the
LMor GF comparison groups if they reported that their parent ever had a same-sex relationship.Although this decision has a
lot of merit,it makes comparisons across groups somewhat of a challenge.Because the LM group is comprised of young
adults who experienced multiple family forms and transitions,it is impossible to isolate the effects of living with a lesbian
mother from experiencing divorce,remarriage,or living with a single parent.
2.3.Scientific rigor
Marks (2012) discusses Sarantakos (1996) as one example of a ‘‘scientifically viable’’ study that contradicts the ‘‘no dif-
ferences’’ findings.Marks notes that the Sarantakos study is more rigorous than most studies in this field because it is one of
the largest studies of child outcomes;compares child outcomes across heterosexual married (n = 58),heterosexual cohabit-
ing (n = 58),and homosexual parents (n = 58);uses data frommultiple sources;and employs standard regression techniques
to account for differences across groups that might be associated with membership in the group and child outcomes (e.g.
educational attainment).
The Regnerus (2012) study is admittedly a broad descriptive analysis of the differences in young adult outcomes associ-
ated with experiencing various family structures in childhood.He rightly makes no claims of causation,and encourages fu-
ture studies of the NFSS data to delve deeper into explanations of the differences he highlights.That said,the Regnerus study
is more scientifically rigorous than most of the other studies in this area.
The NFSS sample size of respondents who experienced a same-sex parenting relationship in childhood is larger than the
other studies,althoughthe sample of youngadults withgay fathers is toolowfor extensive research.Regnerus estimates mean
differences across eight comparison groups and determines whether those differences persist after accounting for a host of
control variables.The controls include the standard variables of age,gender,race/ethnicity,maternal educationlevel,and per-
ceivedhouseholdincomeas achild;andtwoadditional controls that are not commonbut maybe associatedwithbothlivingin
a same-sex household and subsequent outcomes:being bullied as a youth and the state’s legislative gay-friendliness index.
Regnerus may be over-controlling his models if family structure during childhood caused differences in the factors that
are being controlled,such as household income or bullying in youth.However,given that he is trying to isolate the differ-
ences in adult outcomes by family structure in childhood,rather than explain why the differences exist,he may have con-
sidered including ‘‘ever touched sexually by a parent/adult’’ as a control variable,rather than an outcome for several reasons.
This outcome differs significantly across groups and it may cause many of the other outcomes (e.g.depression) in the study.
Further,Regnerus notes that it is not possible to know the timing of the sexual abuse relative to experiencing the parent’s
same-sex relationship.The same concern may apply to the outcome ‘‘ever forced to have sex against will,’’ because it is not
clear if the forced sex occurred in childhood or adulthood.Moreover,‘‘family received welfare growing up’’ may also be con-
sidered a control variable rather than a young adult outcome,to be consistent with the rest of the analysis.
Although Regnerus presents his findings in a very straightforward manner,which is difficult to do with 40 outcomes and
eight comparison groups,one limitation to his presentation is that only the gross means are shown.He indicates whether the
differences in means are significant after including a host of controls,but we no longer knowthe size of the outcome differ-
ences and therefore do not know if those differences are substantively significant as well.
2.4.Outcomes considered
According to Marks (2012),a limitation of the extant research on children of same-sex parents is that it does not focus on
the ‘‘societal concerns of intergenerational poverty,collegiate education and/or labor force contribution,serious criminality,
incarceration,early childbearing,drug/alcohol abuse,or suicide that are frequently the foci of national studies on children,
adolescents,and young adults,’’ (page 16 in manuscript).Moreover,the research provides little information on the long-term
780 C.Osborne/Social Science Research 41 (2012) 779–783
outcomes of children of lesbian or gay parents.He contrasts this dearth of information with the abundance of research con-
ducted on children of divorce,remarriage,and increasingly,cohabitation.
Regnerus (2012) largely addresses these concerns and the NFSS data provide an excellent resource to explore themfur-
ther.Regnerus examines 40 outcomes among young adults including demographic characteristics,employment and educa-
tional attainment,relationship quality,physical and emotional health,and risky behaviors.He finds that for 25 of the 40
outcomes,children whose mothers had a same-sex relationship (LM) are significantly different in young adulthood relative
to those who spent their entire childhood in an intact family with their married,biological parents.Most of these differences
persist after statistical controls.
Some of these differences may easily be considered disadvantages,such as lower levels of income,more receipt of public
assistance,lower levels of employment,poorer mental and physical health,poorer relationship quality with current partner,
more unresolved issues with regard to family of origin,and higher levels of smoking and criminality.The direction of the dif-
ference is left for interpretation on some of the other outcomes,such as receiving therapy,sexual orientation,and sexual
Importantly,one cannot clearly link having a lesbian mother (or gay father) with any of these outcomes.As stated earlier,
the group is comprised of young adults who experienced multiple family structures,not only a same-sex parent household
(indeed,some of the respondents never lived with the mother’s same-sex partner).It is quite possible,for example,that
many or most of the negative outcomes result from the divorce of the young adult’s biological parents that preceded the
mother’s same-sex relationship.
An interesting finding is that the outcomes for young adults of lesbian mothers are considerably different than those for
young adults of single parents.Single-parent households are typically considered the most disadvantaged households;thus
it is interesting that LMhouseholds are more disadvantaged than this group in the NFSS.Again,is it likely that the LMgroup
also experienced single-parenting at one time,and that experiencing multiple changes in family types may predict the poor-
est outcomes.
2.5.Detectable differences
Marks (2012) claims that samples with fewer than 393 respondents lack the necessary statistical power to detect small
differences in outcomes.Regnerus (2012) has fewer than 393 respondents in both the LMand GF groups;however,all of the
mean differences are quite substantial.The concern for Regnerus is not Type II errors (saying something is NOT significant
when it is),but the possible attribution of differences to living in a same-sex household rather than to experiencing multiple
family structures in childhood,one of which happened to be a same-sex parenting relationship.
Marks’ (2012) discussion of type II errors raises two important questions:why do we seek to find differences in outcomes
between children raised by same-sex and heterosexual parents,and why might we expect these differences to exist?Taking
these up in order:Why the interest in children of same-sex relationships?
I am not a scholar of same-sex relationships or GLB identity.I am a family demographer interested in the well-being of
children and howit is associated with family structure and stability.Generally,scholars in my field study children in various
family forms for two often-related reasons.One reason is that theory asserts that a particular family structure or dynamic
causes harmto child well-being (e.g.divorce),and we seek to understand the mechanisms through which the harmoccurs.
A second,related reason is that a particular family form is increasing rapidly in the population and that particular family
structure is associated with negative parent and child outcomes (e.g.cohabitation).The increase in the type of family struc-
ture in question foretells anticipated declines in overall child well-being.
It seems implausible that we study children of same-sex relationships because of a growing number of children in these
family arrangements.Regnerus (2012) estimates that approximately 1.7% of young adults have a parent who had a same-sex
relationship during their offspring’s childhood.Even fewer respondents ever lived with their parent and the same-sex part-
ner,and most of these same-sex relationships were short-lived.
The proportion of children who will have a parent who has a same-sex relationship is likely to increase given that the
stigma associated with same-sex relationships is waning,cohabitation is increasing,and the legal restrictions to same-sex
marriage are beginning to loosen.However,the proportion of children experiencing a same-sex parenting relationship is
likely to remain quite small relative to other family forms such as heterosexual cohabitation or single-parenthood.Almost
25% of children are born to cohabiting parents,and nearly half of all children will spend some portion of their childhood out-
side of an intact,married,biological family.
The focus on children of same-sex parents seems,then,to be driven more by the sensitive political and social issues sur-
rounding same-sex relationships than by evidence that this family structure is increasing rapidly or,for that matter,harmful
to children.Because the topic is so politicized,scholars must pay even more careful attention to the presentation and inter-
pretation of their findings.Although scholars are trained to use great care to disentangle the causal versus selection effects of
family structure and child well-being,we understand that true causation can never be determined because we cannot ran-
domly assign children to various family structures.Consumers of research on children of same-sex relationships,by contrast,
may not always have the same training or be so careful in their interpretations.The results of scholarly studies are often
scrutinized by pundits and legislators to support their pre-existing ideas of differences or ‘‘no differences’’ across groups.
The papers in question provide two excellent examples of howresearch in this area can easily be misinterpreted,and why
scholars need to be extremely clear about what the study can and cannot claim.Marks (2012) implies that the American
C.Osborne/Social Science Research 41 (2012) 779–783
Psychological Association did not uphold its own recommended standards of research in its brief and made assertions of ‘‘no
differences’’ that the research cannot fully support.Regnerus (2012) finds substantial differences across groups and uses
great care to note that his descriptive analysis does not imply causation and that the LMrespondents may have lived in many
different family structures.Still,the rigor of the study may lead some advocates to claim that growing up with a same-sex
parent causes harm and should,therefore,be illegal.
I amnot arguing that scholars should avoid studying children of same-sex relationships.Indeed,if substantial differences
between groups exist,scholars should provide this evidence,even if it is contradictory to one’s personal beliefs.When schol-
ars began studying children of divorce and remarriage,many entered the field hoping to disprove the differences that had
been identified in early studies.Ever more rigorous research showed,however,that there was a clear link between negative
outcomes and childhood experience of divorce.Even more careful study helped us to better understand the mechanisms and
the role that selection plays in these negative outcomes.
Although scholarship on child outcomes in gay and lesbian families is to be encouraged,moving beyond basic analysis of
gross depictions of outcome differences will require stronger,more cogent theoretical frameworks to explain why outcome
differences exist.Better theory in this area is clearly needed.
3.Why might differences in child well-being exist?
Often,research on the well-being of children of same-sex parents,including Regnerus (2012),seems motivated by a fam-
ily studies conundrum.As Marks (2012) shows,almost all of the prior research in this area finds no differences in outcomes
between children raised by married,biological parents and children raised by same-sex parents.However,almost all of the
research that compares children of married,biological parents to any other family formfinds differences (the differences are
often moderate and accounted for by selection,but differences in means are clear).
Given that children of same-sex parents cannot be the biological offspring of both of their parents,then these children
must have experienced another family form as well.For example,they must have experienced their biological parents’ di-
vorce,the adoption by at least one of their parents at birth,or their parents’ cohabitation,given that same-sex marriage is not
universally legal.Indeed,Regnerus shows that children of same-sex parents are likely to have experienced several family
forms in childhood.
Because negative associations between child well-being and these other family forms are well-documented,it seems
implausible that no differences exist for children of same-sex relationships net of the effects of having experienced these
other family forms.Regnerus (2012) confirms,in contrast,that significant differences do exist between groups after appro-
priate controls,but he does not provide a theoretical rationale for why these differences exist.
A theoretical framework generally implies a causal pathway between a particular input and some outcome.Over time,a
solid theoretical framework has been developed to explain why children of married parents are advantaged relative to other
groups.However,no research can confirm a causal pathway (only render it more plausible than competing hypothesized
Theory asserts that children born and raised by both of their biological parents are advantaged for several reasons.Nota-
bly,the legal and social sanctioning of marriage leads to a stronger commitment between the partners,greater sharing of
resources and pooling of risk,and stronger support from kin and friends.Moreover,the legal and social bond makes sepa-
ration more costly and therefore less likely.In turn,these factors will cause married parents to have greater relationship
quality,more financial resources,better physical and mental health,and greater relationship stability.Each of these factors
leads to higher quality parenting,which leads to enhanced child outcomes.
In simple terms,theory claims that children of divorce are disadvantaged because their parents experience a loss in the
sharing of resources and commitment,a decline in parenting quality,and the transition itself (that is,the divorce) is asso-
ciated with stress that leads to negative outcomes for the mother and child.Children of remarriage may not experience the
gains associated with a newmarriage because they lack the biological relatedness to each parent,and biological relatedness
is associated with higher levels of investment in children.
Cohabitation as a family structure is identical to marriage but lacks the legal and social sanctioning of marriage.There-
fore,according to theory,the parents and children do not benefit from that family form as compared to marriage.Finally,
single parents,particularly single mothers,are much less likely to have the financial resources of married parents,and
the single parent must do the job of both parents with regard to affection,monitoring,supervision,and discipline.Lower
levels of resources and poorer parenting are associated with poor outcomes for children of single parents.
Same-sex relationships may encompass any one of these family forms,and children are likely to experience several differ-
ent structures over the course of their childhood.There is little theoretical guidance,however,as to why the effects onchildren
of two mothers (or two fathers) as compared to a mother and father would be different within any of these family
Women earn less than men,so it is possible that children of lesbian parents would have fewer resources than children of
gay men or heterosexual parents.Moreover,parents may provide important gender role models to their children;thus,chil-
dren who do not have a parent from each gender may be negatively affected.However,each of these factors may apply to
children of single parents as well.
782 C.Osborne/Social Science Research 41 (2012) 779–783
The social stigma associated with same-sex relationships may lead to discrimination and lower levels of family support,
and these may cause negative child outcomes.However,it is not clear that this factor would drive all of the differences in
outcomes.Current theories linking family structure to child well-being offer fewclues of pathways that are unique to same-
sex parents as a separate family structure.
Most of the extant research finds limited support for potential causal mechanisms between family structure and child
well-being,and more support for selection.Studies on family structure find that the characteristics of married parents
are advantaged compared to any other family structure,and that these antecedent characteristics explain most of the differ-
ences in child outcomes.For example,on average,cohabiting parents are systematically less educated and have lower levels
of income than married parents.These systematic differences in socio-economic status are largely responsible for the differ-
ences found in children’s outcomes,rather than cohabitation per se.
It is not clear that there are systematic differences between married parents and same-sex parents.Indeed,same-sex par-
ents are not a monolithic group;same-sex parents include married parents,cohabiting parents,step-parents,and single par-
ents.To my knowledge,no studies have been able to adequately compare heterosexual- and same-sex parents within a given
family structure (e.g.married to married;cohabiting to cohabiting,unstable to unstable,etc.) Confounding the same-sex
relationship with a variety of family forms (and changes in family forms) is a limitation of the Regnerus study,but may
be a limitation of all studies because of the small sample size associated with children of same-sex parents.
Nevertheless,if a study is able to compare within-family structure differences between children of heterosexual and
same-sex parents,a clear theoretical and testable argument should be developed as to why differences should exist between
the groups.
Regnerus (2012) provides convincing evidence that various young adult outcomes are associated with having a parent
who had a same-sex relationship.The findings contradict the ‘‘no differences’’ claim of the American Psychological Associ-
ation’s Brief on Lesbian and Gay Parenting,and the study is one of the first studies to employ a larger sample and more rig-
orous analytic techniques.Whether same-sex parenting causes the observed differences is,however,quite another matter.
We lack theory to tell us exactly why gay and lesbian parenting would produce negative child outcomes.For all we know,the
effect derives entirely from the stigma attached to such relationships and to the legal prohibitions that prevent same-sex
couples from entering and maintaining ‘‘normal’’ marital relationships.
Moreover,research being done today on young adults who experienced same-sex parenting in childhood cannot speak to
what will eventually happen to today’s children in same-sex households.Same-sex parenting relationships are,as Regnerus
shows,quite diverse and include many family forms.Today’s children of same-sex parents may be more likely to be planned
and deliberate,as social stigma declines and fertility options increase;these children’s outcomes may be more akin to chil-
dren of married,biological parents.Grouping all children of same-sex relationships into one group provides limited and
potentially misleading information on ‘‘differences.’’
Policy makers who may feel obligated to act on these findings would do well to note,for example,that some researchers
have found a strong negative impact of being raised in very large families on various child outcomes (see Desai,1995,for a
review).But no one has ever suggested (outside of China) that large families should therefore be illegal.Given the sensitive
political and cultural environment surrounding same-sex relationships,preliminary findings of ‘‘differences’’ such as those
reported by Regnerus should not be used to support punitive legislation aimed at limiting the family formation and fertility
choices of gays and lesbians.Advocates who wish to create mischief with these studies will surely do so,no matter howcau-
tious and circumspect their authors have been.Asking everyone to be as careful in their causal conclusions as Marks and
Regnerus have been is no doubt wishful thinking.
Desai,S.,1995.When are children from large families disadvantaged?Evidence from cross-national analyses.Population Studies 49,195–210.
Marks,Loren,2012.Same-sex parenting and children’s outcomes:a closer examinations of the American Psychological Association’s brief on lesbian and
gay parenting.Social Science Research 41 (4),735–751.
Regnerus,Mark,2012.How different are the adult children of parents who have same-sex relationships?Findings from the new family structures study.
Social Science Research 41 (4),752–770.
Sarantakos,W.R.,1996.Children in three contexts:family,education,and social development.Children Australia 21,23–31.
C.Osborne/Social Science Research 41 (2012) 779–783
47   документов Отправить письмо
Без категории
Размер файла
145 Кб
002, ssresearch, 2012
Пожаловаться на содержимое документа