Artículo de revista: Ophir, Ariane; Boertien, Diederik; Vidal, Sergi (2023). Sexuality and Demographic Change: Documenting Family Formation Trajectories and Cohort Change in the LGB Population. Demography, 60 (5): 1581-1605. (ISSN 1533-7790) https://doi.org/10.1215/00703370-10968468 Demography (2023) 60(5):1581–1605 Published online: 20 September 2023 DOI 10.1215/00703370-10968468 © 2023 The Authors This is an open access arti­cle dis­trib­uted under the terms of a Creative Commons license (CC BY-NC-ND 4.0). Sexuality and Demographic Change: Documenting Family Formation Trajectories and Cohort Change in the LGB Population Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 Ariane Ophir, Diederik Boertien, and Sergi Vidal ABSTRACT Narratives of demo­graphic shifts over­look how soci­e­tal changes shape the fam­ily tra­jec­to­ries of sex­ual minor­i­ties. Using sequence anal­y­sis, we describe how part- nering and par­ent­hood evolve over the life course of les­bian, gay, and bisex­ual (LGB) women and men in the United Kingdom (N = 455) and how the types of these fam­ily tra­jec­to­ries changed across two birth cohorts (born before 1965 and in 1965–1979). We find five dis­tinct tra­jec­to­ries between ages 18 and 40, wherein two thirds of the sam­ple belonged to a fam­ily tra­jec­tory that did not involve liv­ing with chil­dren. Partnership- cen­tered tra­jec­to­ries became more com­mon across cohorts, and this increase came at the expense of tra­jec­to­ries char­ac­ter­ized by singlehood among gay men and les­ bian women. However, par­ent­hood tra­jec­to­ries became less com­mon among all­LGB groups. Furthermore, fam­ily tra­jec­to­ries became more com­plex across cohorts, includ­ ing more tran­si­tions, which coin­cides with trends in the gen­eral pop­ul­ a­tion. Yet we also find that fam­ily tra­jec­to­ries became less diverse among les­bian women and bisex­ual men, in con­trast to the trend among gay men and the gen­eral pop­u­la­tion. The results dem­on­strate the dynamic, com­plex, and diverse nature of LGB indi­vid­u­als’ fam­ily lives and why existing nar­ra­tives of fam­ily-related demo­graphic change should explic­itly con­sider sex­ual minor­i­ties in demo­graphic nar­ra­tives. KEYWORDS Sexual minor­ i­ties • Partnering • Parenthood • Life course • Sequence anal­y­sis Introduction What are the partnering and par­ent­hood tra­jec­to­ries of sex­ual minor­i­ties? Have these life course tra­jec­to­ries changed across cohorts? Existing the­o­ret­i­cal and empir­i­cal schol­ar­ship pro­vi­des lit­tle guid­ance in answer­ing these ques­tions. Theories of demo­ graphic change pri­mar­ily focus on explaining declines in rela­tion­ship dura­tion, mar­ riage, and par­ent­hood in the gen­eral pop­u­la­tion (Lesthaeghe 2010; Mills and Blossfeld 2003; van de Kaa 1987; Zaidi and Mor­gan 2017). Such per­spec­tives are not designed to describe cohort change among the les­bian, gay, and bisex­ual (LGB) pop­u­la­tion, which was his­tor­ic­ ally excluded from mar­riage and var­i­ous routes into par­ent­hood ELECTRONIC SUPPLEMENTARY MATERIAL The online ver­sion of this arti­cle (https:​­/​­/doi​­.org​­/10​­.1215​­/00703370​ ­-10968468) con­tains sup­ple­men­tary mate­rial. 1582 A. Ophir et al. (Adamczyk and Liao 2019; Moore and Stambolis-Ruhstorfer 2013; Trandafir 2015). Empirically, most rep­re­sen­ta­tive data sources avail­­able to study sex­ual minor­i­ties are lim­ited to cores­i­dent cou­ples consisting of two per­sons of the same sex or gen­der (Gates 2012; Kolk and Andersson 2020; Lau 2012; Manning and Payne 2021; Russell et al. 2020). Although these data are infor­ma­tive in under­stand­ing par­ent­hood and part­ner­ship out­comes of same-sex cou­ples at a given time, they hin­der our under­ Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 stand­ing of union for­ma­tion and par­ent­hood as dynamic pro­cesses that evolve over the life course. Moreover, these data rep­re­sent only sex­ual minor­i­ties that selected into (same-sex) partnering. Recently, large-scale nation­ally rep­re­sen­ta­tive sur­veys started incor­po­rat­ing ques­tions about sex­ual iden­tity. Nonetheless, the oppor­tu­ni­ties these data sources pro­vide to study the life course dynam­ics of partnering and par­ ent­hood among sex­ual minor­i­ties have remained underexplored (Badgett et al. 2021; Perales et al. 2020; Reczek 2020). In this arti­cle, we use infor­ma­tion on self-reported sex­ual iden­tity and ret­ro­spec­ tive and pro­spec­tive coresidence his­to­ries with part­ners and chil­dren to describe cohort change in the fam­ily for­ma­tion tra­jec­to­ries of sex­ual minor­i­ties. We answer two main ques­tions: (1) What are the fam­ily for­ma­tion tra­jec­to­ries of les­bian, gay, and bisex­ual indi­vid­u­als (LGBs here­af­ter)?1 (2) Was there a cohort change in the type of tra­jec­to­ries? Using data from the Understanding Society sur­vey (UKHLS), we con­struct the life courses of a rel­a­tively large rep­re­sen­ta­tive sam­ple of LGB women and men between ages 18 and 40. To exam­ine cohort change, we com­pare a cohort that reached mid-adult­hood (age 40) before most major legal changes took place in the United Kingdom (UK) (born before 1965) to a cohort that expe­ri­enced those changes dur­ing early adult­hood (born between 1965 and 1979). We employ a sequence anal­y­sis approach to answer our ques­tions. Previous research focused pri­mar­ily on par­ent­hood in same-sex unions (Goldberg 2006; Kolk and Andersson 2020; Manning and Payne 2021; Moore and Stambolis-Ruhstorfer 2013) or on union for­ma­tion and dis­so­lu­tion (Carpenter 2020; Joyner et al. 2017; Kalmijn et al. 2007; Lau 2012; Manning et al. 2016; Ruiz-Vallejo and Boertien 2021). The sequenc­ing of part­ner­ships and par­ent­hood over the life course pro­vi­des a rich descrip­tive account of cores­i­dent famil­ial expe­ri­ences by documenting the occur­ rence, order, tim­ing, and dura­tion of mul­ti­ple fam­ily-related events. We find descrip­ tive evi­dence that the like­li­hood of belong­ing to fam­ily tra­jec­to­ries char­ac­ter­ized by singlehood and tra­jec­to­ries that include par­ent­hood has declined. In con­trast, the like­li­hood of belong­ing to tra­jec­to­ries in which indi­vid­u­als live with a part­ner but not with chil­dren has increased across cohorts. Family Trajectories in the LGB Population We start the lit­er­a­ture review with an over­view of existing knowl­edge about union for­ma­tion, dis­so­lu­tion, and child­bear­ing among LGBs that guides sev­eral expec­ta­ tions regard­ing fam­ily life course tra­jec­to­ries. Subsequently, we dis­cuss how LGBs’ 1 Unfortunately, we do not have data on other sex­ual iden­ti­ties, such as asex­ual, pan­sex­ual, or queer, nor on gen­der minor­it­ies, such as trans­gen­der and non­bi­nary peo­ple, thus pre­vent­ing us from study­ing other groups within the LGBTQ+ pop­u­la­tion. Family Formation Trajectories and LGB Cohort Change 1583 tra­jec­to­ries fit within existing the­o­ret­i­cal frame­works and gen­eral nar­ra­tives of demo­ graphic change. Union Formation and Dissolution Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 Cross-sec­tional research in the United States sug­gests that LGBs are less likely to be in a cores­i­den­tial rela­tion­ship than het­ero­sex­u­als (Black et al. 2000; Carpenter and Gates 2008). Lesbian women’s part­ner­ship rates are sim­i­lar to those of het­ero­sex­ual women, while part­ner­ship rates are con­sid­er­ably lower among gay men and among bisex­ual men and women (Badgett et al. 2021). Moreover, stud­ies on atti­tudes toward partnering have found that les­bian, gay, and bisex­ual women and men desire to be in rela­tion­ships and to marry at sim­i­lar rates as het­ero­sex­ual women and men across var­i­ous coun- tries (Barrios and Lundquist 2012; D’Augelli et al. 2007; Riskind and Tornello 2017; Tate et al. 2019). Similarly, Meier et al. (2009) found mod­est dif­fer­ences across sex­ual iden­ti­ties in rela­tion­ship val­ues among young men and women in the United States. There are sev­eral fac­tors hin­der­ing LGBs’ union for­ma­tion that could shape dif­fer­ent partnering tra­jec­to­ries across cohorts. First, fewer oppor­tu­ni­ties exist to meet suit­able part­ners because fewer peo­ple iden­tify as LGB than as het­ero­sex­ual (Ellingson et al. 2004). Moreover, in con­trast to het­ero­sex­u­als, it is rel­a­tively uncom­ mon for same-sex cou­ples to have met their roman­tic part­ners through fam­ily or in school (Rosenfeld and Thomas 2012). However, these bar­ri­ers have less­ened over time; same-sex rela­tion­ships are more prev­a­lent, and nonheterosexual iden­ti­ties are more vis­i­ble (England et al. 2016; Gilroy and Kashyap 2021). Also, decreased paren­ tal con­trol and increased geo­graphic mobil­ity allow peo­ple more inde­pen­dence to form unions (Rosenfeld and Kim 2005). Furthermore, the emer­gence of online dat­ing helped to facil­i­tate con­nec­tions between LGBs more eas­ily. Second, minor­ity stress can lower the “returns” from being in a rela­tion­ship. Minority stress is caused by dis­crim­i­na­tion and microaggressions related to a spe­cific minor­ity sta­tus (Frost and Gola 2015). LGBs expe­ri­ence many of these stress­ors when they enter a same-sex union, which could deter peo­ple from enter­ing rela­tion­ships (Frost et al. 2017). Studies in Germany and the United States have found that peo­ple in same-sex rela­tion­ships are more likely to expe­ri­ence paren­tal and social dis­ap­proval and express more con­cerns about their part­ner’s accep­tance by fam­ily and friends than are peo­ple in dif­fer­ent-sex rela­tion­ships (Hank and Wetzel 2018; Holmberg and Blair 2016; Reczek 2014). In an anal­ys­ is of pref­er­ences among online dat­ers across eight countries, Potârcă and col­leagues (2015) found that resid­ing in a sup­port­ive envi­ ron­ment or region with for­mal rec­og­ni­tion of same-sex unions was asso­ci­ated with an increase in long-term dat­ing inten­tions among indi­vid­ua­ ls looking for same-sex part­ners. Given the rapid increases in accep­tance of same-sex cou­ples over the last decades (Rosenfeld 2017; Trandafir 2015), we would expect that youn­ger cohorts of LGBs are more likely to seek long-term unions than older cohorts of LGBs. Third, insti­tu­tional dis­crim­i­na­tion, such as the legal exclu­sion from mar­riage, can fur­ther increase the costs and reduce the ben­efi ­ ts of being in a cores­i­den­tial union for older LGBs. Same-sex mar­riage was not legal for large parts of older LGBs’ lives. Marriage, unlike cohab­i­ta­tion, is related to insti­tu­tional ben­efi ­ ts in many ­con­texts. For instance, legal­iz­ing same-sex mar­riage improved health out­comes among ­indi­vid­u­als 1584 A. Ophir et al. in same-sex unions by increas­ing access to health insur­ance (Gonzales and Blewett 2014). Moreover, reduc­tions in insti­tu­tional dis­crim­i­na­tion are gen­er­ally related to increased well-being for same-sex cou­ples (Gonzales and Blewett 2014; Hatzenbuehler et al. 2009; Kail et al. 2015). The devel­op­ments discussed so far would pre­dict an increase in the prev­a­lence of LGBs’ cores­i­den­tial rela­tion­ships across cohorts. However, reduced stress­ors and dis­ Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 crim­in­ a­tion might not directly increase part­ner­ship rates. For exam­ple, the legal­i­za­tion of same-sex mar­riage did not increase the shares of LGBs in a union in Mas­sa­chu­setts (Carpenter 2020). Furthermore, some peo­ple in same-sex mar­riages express ambiv­a­ lent feel­ings toward mar­riage and view it as a patri­ar­chal and heteronormative insti­ tu­tion, whereas oth­ers express their con­cerns that mar­riage could fur­ther assim­i­late sex­ual minor­i­ties into main­stream cul­ture (Bosley-Smith and Reczek 2018; Stambolis- Ruhstorfer and Descoutures 2020). Singlehood could be more appeal­ing to LGBs than to het­ero­sex­u­als thanks to strong friend­ship net­works (Dewaele et al. 2011; Lau 2012), and youn­ger LGBs might opt to reject the heteronormative script of fam­ily for­ma­tion that is embed­ded in cores­i­den­tial part­ner­ships (and child­bear­ing). Taken together, these trends could the­o­ret­i­cally imply cohort sta­bil­ity or even a decline, rather than an increase, in the prev­a­lence of partnering among LGBs. The fore­go­ing dis­cus­sion has cen­tered on union for­ma­tion pro­cesses, but the same fac­tors are also rel­e­vant to under­stand­ing union dis­so­lu­tion among LGBs. Same-sex unions are more likely to end in sep­a­ra­tion than dif­fer­ent-sex unions in some coun- tries (Kalmijn et al. 2007; Kolk and Andersson 2020; Lau 2012; Manning et al. 2016), pos­si­bly because of the neg­a­tive effects of stigma, dis­crim­i­na­tion, and lack of fam­ily sup­port on rela­tion­ship qual­ity. Obstacles toward par­ent­hood and mar­riage can pre­ vent LGBs from investing in cores­i­den­tial roman­tic rela­tion­ships (Lau 2012). From this per­spec­tive, minor­ity stress and stigma are likely to lead to rel­a­tively high sep­a­ ra­tion rates, but these could have declined across cohorts as invest­ments in rela­tion­ ships became more avail­­able to LGBs (Lau 2012; Ruiz-Vallejo and Boertien 2021). There are also rea­sons to expect sep­a­ra­tion rates to increase or remain unchanged across cohorts. For instance, higher sep­a­ra­tion rates can stem from LGBs giv­ing less impor­tance to life­long com­mit­ment in rela­tion­ships than do het­ero­sex­ual women (Meier et al. 2009). Similarly, Lau (2012) did not find changes in the sta­bil­ity of same-sex unions across cohorts in the UK and suggested that, among older cohorts, only com­mit­ted same-sex cou­ples might have decided to start liv­ing together. Finally, sep­a­ra­tion and serial cohab­i­ta­tion have become more com­mon across cohorts in the gen­eral pop­u­la­tion (Bukodi 2012; Eickmeyer and Manning 2018; Hiekel and Fulda 2018), which could have made sep­a­ra­tion and repartnering among youn­ger cohorts of LGBs prev­a­lent as well. In sum­mary, mul­ti­ple social and demo­graphic forces sug­gest that the prev­a­lence of tra­jec­to­ries involv­ing part­ner­ships should increase across cohorts, but expec­ta­tions regard­ing changes in the prev­a­lence of union insta­bil­ity are less clear. Pathways Into Parenthood Most evi­dence on par­ent­hood among LGBs is lim­ited to child­bear­ing and par­ent­ing in same-sex cou­ples. Overall, there is con­sis­tent evi­dence that women, and ­espe­cially Family Formation Trajectories and LGB Cohort Change 1585 men, in same-sex cou­ples are less likely to coreside with chil­dren than indi­vid­u­als in dif­fer­ent-sex cou­ples (Baumle and Compton 2011; Gates 2012; Kolk and Andersson 2020). However, there is var­i­a­tion within the LGB pop­u­la­tion. Recent research in the United States showed that bisex­ual women are as likely as het­ero­sex­ual women to coreside with chil­dren. In con­trast, bisex­ual and gay men are less likely than het­ero­sex­ual men to coreside with chil­dren (Badgett et al. 2021). Trends over time Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 are mixed. Kolk and Andersson (2020) found that child­bear­ing within same-sex mar­riages increased con­sid­er­ably over time for Swed­ish women. In con­trast, Gates (2015) reported that the share of same-sex cou­ples liv­ing with chil­dren in the United States has declined. There are numer­ous path­ways into par­ent­hood for LGBs. A prominent path­way among older LGB cohorts was through child­bear­ing in dif­fer­ent-sex rela­tion­ships (Andersson et al. 2006; Gates 2012). Studies have suggested that gay men con­tinue to expe­ri­ence pres­sures to form a dif­fer­ent-sex union (e.g., Schacher et al. 2005). Sexual minor­ity women also con­tinue to have higher rates of unwanted births (Ela and Budnick 2017; Everett et al. 2021). Nonetheless, par­ent­hood through a pre­vi­ous dif­fer­ent-sex rela­tion­ship is becom­ing less com­mon among same-sex cou­ples in the United States (Gates 2015). Alternative path­ways to par­ent­hood, which have become more acces­ si­ble for LGBs over time, include adop­tion, assisted repro­duc­tion tech­niques (ARTs), and sur­ro­gacy (Moore and Stambolis-Ruhstorfer 2013; Rozental and Malmquist 2015). However, these options require con­sid­er­able eco­nomic resources and plan­ning (Park et al. 2020). Therefore, expanding access to routes into par­ent­hood could have increased the num­ber of same-sex cou­ples with chil­dren, but these increases are lim­ited to those with eco­nomic resources. Relatively low lev­els of par­ent­hood among LGBs can also be related to inten­ tions and desires to become a par­ent. Studies on par­ent­hood desires have documented that most gay and les­bian peo­ple would like to have chil­dren, but this share is lower than for het­ero­sex­u­als and bisex­u­als (Hank and Wetzel 2018; Riskind and Patterson 2010; Romeu et al. 2015; Tate et al. 2019). Nonetheless, the gap between desired and observed par­ent­hood is greater for gay men and les­bian women than for het­ero­sex­ual men and women (Riskind and Tornello 2017). Several stud­ies have documented how minor­ity stress and the expe­ri­ence of dis­crim­i­na­tion reduce par­ent­hood inten­tions among gay men and les­bian women, while favor­able pol­icy envi­ron­ments and involve­ ment in the LGBT com­mu­nity increase par­ent­hood desires (Costa and Bidell 2017; Scandurra et al. 2019). This would imply that the gap in par­ent­hood desires between gay/les­bian and het­ero­sex­ual indi­vid­u­als is smaller in more favor­able con­texts. Furthermore, soci­e­tal pres­sures to fol­low main­stream fam­ily tra­jec­to­ries, includ­ing the for­ma­tion of “nuclear” fam­i­lies, could increase for LGB indi­vid­u­als as pos­si­bil­i­ ties to access insti­tu­tions like mar­riage and par­ent­hood expand (Duggan 2002). Because of the focus on same-sex cou­ples in the lit­er­a­ture, we know very lit­tle about LGBs’ expe­ri­ences with sin­gle par­ent­hood. However, stud­ies have con­sis­tently found that sex­ual minor­ity women are more likely to expe­ri­ence unwanted births (Ela and Budnick 2017; Everett et al. 2021). Moreover, given that many same-sex par­ent fam­il­ ies are formed after the dis­so­lu­tion of a dif­fer­ent-sex union (Gates 2015), it is likely that sin­gle par­ent­hood is prev­a­lent for some part of LGBs’ life course, espe­cially among older birth cohorts. However, it is unclear to what extent this was com­mon and whether it changed across cohorts. 1586 A. Ophir et al. In sum­mary, evi­dence sug­gests that social, legal, and demo­graphic forces have vary­ing impli­ca­tions for youn­ger LGBs’ expe­ri­ence with par­ent­hood over the life course. The increas­ing accep­tance of same-sex part­ner­ships could have reduced the prev­a­lence of par­ent­hood within dif­fer­ent-sex rela­tion­ships among LGBs, whereas alter­na­tive routes toward par­ent­hood have become more avail­­able. How the rel­a­tive weight of both trends trans­lates into cross-cohort trends in life course tra­jec­to­ries that Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 include par­ent­hood is unclear. Demographic Change and Sexual Minorities Existing nar­ra­tives of demo­graphic change are not constructed to under­stand fam­ily- related changes among the LGB pop­u­la­tion. These nar­ra­tives focus on a retreat from heg­e­monic fam­ily tra­jec­to­ries driven by ide­a­tional change in the gen­eral pop­u­la­tion (Lesthaeghe 2010; van de Kaa 1987; Zaidi and Mor­gan 2017), eco­nomic inse­cu­ rity (Mills and Blossfeld 2003), and (het­ero­sex­ual) gen­der rela­tions as driv­ers of demo­graphic change (Esping-Andersen and Billari 2015; Goldscheider et al. 2015). Among these driv­ers, only ide­a­tional shifts help to under­stand demo­graphic change among LGBs. The increas­ing prev­a­lence of same-sex unions and fam­i­lies can be per­ ceived as soci­e­tal changes toward self-real­iz­ a­tion (Rosenfeld and Kim 2005). How- ever, this gen­eral under­stand­ing does not pre­dict what type of fam­ily tra­jec­to­ries exist among LGBs and how dif­fer­ent fam­ily tra­jec­to­ries have changed over time as oppor­tu­ni­ties to form a fam­ily expanded. To illus­trate, rather than retreating from insti­tu­tions like mar­riage and par­ent­hood, the LGB pop­u­la­tion has made advance­ ments toward accessing these insti­tu­tions in recent decades in many countries. As a result, demo­graphic change within the LGB pop­u­la­tion will likely look very dif­fer­ent from that of the het­ero­sex­ual pop­u­la­tion. To empir­i­cally doc­u­ment demo­graphic change through life courses, past research has quan­ti­fied change in the type, com­plex­ity, and diver­sity of fam­ily tra­jec­to­ries in the gen­eral pop­u­la­tion (Brückner and Mayer 2005; Elzinga and Liefbroer 2007; Van Winkle 2018). Such empir­i­cal approaches could be more eas­ily appli­ca­ble to LGBs than the­o­ries of demo­graphic change. Past stud­ies found that fam­ily tra­jec­to­ ries in the gen­eral pop­u­la­tion became more diverse across indi­vid­u­als. The heg­e­monic tra­jec­tory of early life­long mar­riage and par­ent­hood gave way to alter­na­tive tra­jec­to­ ries, includ­ing cohab­i­ta­tion, sep­a­ra­tion, repartnering, and sin­gle par­ent­hood (Elzinga and Liefbroer 2007; Van Winkle 2018). The evi­dence of cohort changes in the com­ plex­ity of indi­vid­ual tra­jec­to­ries (i.e., the num­ber of fam­ily events expe­ri­enced) in the gen­eral pop­u­la­tion is mixed. McMunn et al. (2015) found that partnering tra­jec­to­ries became more event­ful with increases in repartnering, while par­ent­hood tra­jec­to­ries became less com­plex over time in the UK. Van Winkle (2018) found that fam­ily tra­jec­to­ries became more com­plex across cohorts in var­i­ous countries, but cohort changes were rel­a­tively minor com­pared with cross-national dif­fer­ences. In the case of the LGB pop­u­la­tion, the option to marry and have chil­dren through path­ways such as adop­tion and ARTs could increase the diver­sity and com­plex­ity of fam­ily tra­jec­to­ries across cohorts. This can espe­cially be the case if some of the LGB pop­u­la­tion do not fol­low heteronormative fam­ily tra­jec­to­ries, whereas oth­ers embark on more heteronormative path­ways of fam­ily for­ma­tion that were pre­vi­ously Family Formation Trajectories and LGB Cohort Change 1587 unavail­able (Bosley-Smith and Reczek 2018; Hank and Wetzel 2018; Riskind and Patterson 2010). However, whether com­plex­ity and diver­sity in fam­ily tra­jec­to­ries between indi­vid­u­als increased depends on how diverse and com­plex they were in older cohorts. It is unclear how typ­i­cal or dom­i­nant fam­ily tra­jec­to­ries, such as life­ long singlehood, were among LGBs in the past. In short, pre­vi­ous the­o­ret­i­cal per­spec­tives of demo­graphic change have focused Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 on major­ity pop­u­la­tions and pop­u­la­tion aver­ages, while empir­i­cal accounts sum­ma­ riz­ing demo­graphic change have not yet stud­ied the LGB pop­u­la­tion. Therefore, in the remain­der of the arti­cle we pres­ent an anal­y­sis that directly acknowl­edges and focuses on the con­tex­tu­ally unique expe­ri­ence of minoritized groups (Johnson-Hanks et al. 2011; Zaidi and Mor­gan 2017). Study Context The UK’s fer­til­ity and union dis­so­lu­tion rates are higher than those of other E ­ uro­pean countries but lower than those of the United States (Lesthaeghe 2010; Musick and Michelmore 2018). The envi­ron­ment for LGB indi­vid­u­als has changed rap­idly and pos­i­tively over time. In 2005, civil part­ner­ships became avail­­able to same-sex cou­ ples. The same year, adop­tion became avail­­able to same-sex cou­ples and sin­gle indi­vid­u­als in England and Wales (2009 in Scotland and 2013 in Northern Ireland). Sexual ori­en­ta­tion was incor­po­rated into anti­dis­crim­i­na­tion laws in 2007 and 2010. In 2009, ARTs and the pos­si­bil­ity of hav­ing two moth­ers on a birth cer­tif­i­cate became avail­­able. In 2014, same-sex mar­riage was legal­ized in England, Wales, and Scotland (2020 in Northern Ireland). Attitudes toward sex­ual diver­sity are rel­a­tively pos­i­tive in the UK com­pared with other countries (Adamczyk and Liao 2019). We com­pare LGBs born before 1965 to LGBs born between 1965 and 1979. The older cohort reached age 40 before sig­nif­i­cant legal changes took place. The youn­ger cohort saw these changes unfold dur­ing early or mid-adult­hood and could, to some extent, have taken advan­tage of the chang­ing pos­si­bil­i­ties to form unions and tran­si­ tion into par­ent­hood. Younger cohorts born after 1979 are excluded from our anal­y­sis because they have yet to com­plete their prime fam­ily for­ma­tion years, albeit being the cohorts to expe­ri­ence the most exten­sive range of pos­si­bil­i­ties yet. Nonetheless, our anal­y­sis pro­vi­des the first bench­marks of fam­ily tra­jec­to­ries for LGB indi­vid­u­als. Methods Data and Sample We use the Understanding Society sur­vey data from 2009 to 2019. UKHLS is a rep­re­ sen­ta­tive house­hold panel sur­vey of the UK pop­ul­ a­tion (Buck and McFall 2011) and is one of the very few large-scale sur­veys to col­lect infor­ma­tion about sex­u­al­ity as well as part­ner­ship and par­ent­hood his­to­ries. These unique fea­tures allow us to iden­tify LGB indi­vid­u­als on the basis of their self-reported sex­ual iden­tity instead of infer­ring it from the gen­der of their part­ner and to recon­struct their his­to­ries of coresidence with part­ners and chil­dren. We include respon­dents who were pres­ent in at least one wave 1588 A. Ophir et al. in which com­plete ret­ro­spec­tive his­to­ries were col­lected (Waves 1 and 6)2 as well as one of the waves in which sex­ual iden­tity was reported on (Waves 3 and 9).3 Subse- quently, we exclude all­indi­vid­ua­ ls who had not reached age 40 by the date of the last inter­view, as our anal­ys­ is relies on ana­lyz­ing com­plete partnering and par­ent­hood his­ to­ries. We also exclude cases with more than six years of miss­ing part­ner­ship and par­ ent­hood his­to­ries or miss­ing infor­ma­tion about age and sex­u­al­ity. The main ana­lytic Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 sam­ple includes 455 LGB indi­vid­u­als (45% women) observed between ages 18 and 40 and born before 1979 (50% born by 1964 and 50% born between 1965 and 1979). Measures Sexual Identity The Understanding Society sur­vey asked respon­dents, “Which of the fol­low­ing options best describes how you think of your­self?” The answer options were het­ero­ sex­ual or straight, gay or les­bian, bisex­ual, other, pre­fer not to say, and don’t know. Sexual iden­tity is a time-vary­ing char­ac­ter­is­tic but is mea­sured only at two points in time in our data. Our anal­y­sis, there­fore, doc­u­ments dif­fer­ences in fam­ily tra­jec­to­ries based on sex­ual iden­tity mea­sured in the last wave with nonmissing infor­ma­tion.4 We restrict the anal­ys­ is to two mutu­ally exclu­sive groups: bisex­ual (n = 176) and gay/ les­bian (n = 279) indi­vid­u­als.5 Gender/sex was col­lected with the ques­tion, “And you are male/female?” thus pre­vent­ing us from con­sid­er­ing other gen­der iden­ti­ties or from com­par­ing cis- and trans­gen­der per­sons. Partnerships Union for­ma­tion, dis­so­lu­tion, and part­ner­ship spells are based on self-reported cores- idence with a part­ner of any gen­der.6 We do not have infor­ma­tion about non­res­i­ dent part­ner­ships. In Waves 1 and 6, respon­dents were asked to ret­ro­spec­tively report the starting and end­ing dates of up to 12 cores­i­den­tial rela­tion­ships (mar­riages, civil ­part­ner­ships, or cohabiting unions) that lasted at least three months. This ­infor­ma­tion 2 Even though new respon­dents in other waves were asked about their rela­tion­ship and par­ent­hood his­to­ ries, this infor­ma­tion is not suf­fic­ ient to recon­struct coresidence his­to­ries across the life course. 3 We also include new respon­dents who were asked about sex­ual iden­tity in Waves 5 and 7. 4 Note that peo­ple’s sex­ual iden­tity might have changed dur­ing the obser­va­tion period, but we are unable to incor­po­rate such changes in the main anal­y­sis. Table S1 (in the online Supplementary Materials A; all­ tables and fig­ures des­ig­nated with an “S” appear in the online sup­ple­ments) shows that results are robust when exclud­ing 13 respon­dents who reported dif­fer­ent sex­ual iden­ti­ties across waves. 5 The “other” cat­eg­ ory is com­pli­cated to inter­pret as pre­vi­ous research sug­gests that this cat­e­gory includes both nonheterosexual indi­vid­u­als and het­ero­sex­ual indi­vid­u­als who do not under­stand the ques­tion or response set (Elliott et al. 2019). In our data, 75% of indi­vid­u­als iden­ti­fy­ing as “other” in Wave 3 iden­ti­fied as het­ero­sex­ual in Wave 9. Eighty-seven per­cent of per­sons iden­ti­fy­ing as gay/les­bian in Wave 3 also iden­ ti­fied as such in Wave 9, whereas the cor­re­spond­ing num­ber for bisex­ual indi­vid­ua­ ls is 45%. 6 Note that we can­not dis­tin­guish dif­fer­ent-sex from same-sex part­ner­ships because this infor­ma­tion is not col­lected ret­ro­spec­tively. Therefore, we can­not esti­mate changes in the prev­a­lence of dif­fer­ent- ver­sus same-sex part­ner­ships across cohorts (Gates 2015). Family Formation Trajectories and LGB Cohort Change 1589 was updated and har­mo­nized with pro­spec­tive infor­ma­tion col­lected annu­ally in the sub­se­quent waves (Nandi et al. 2020). We use these ret­ro­spec­tive his­to­ries on cores­ i­dent part­ners to con­struct a per­son-month file with a dummy var­i­able indi­cat­ing whether, in that month, the per­son coresided with a part­ner.7 Parenthood is recorded from reports about coresidence with chil­dren youn­ger than 18. Respondents were asked in Waves 1 and 6 whether they were ever the par­ent of Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 an adopted, bio­log­i­cal, or step­child, and this infor­ma­tion is updated in each wave. For bio­log­i­cal chil­dren, the sur­vey col­lects the year of birth and the year the respon­dent last lived with the child. For adopted and non­bi­o­log­i­cal chil­dren, the sur­vey col­lects the year the respon­dent started and stopped liv­ing with the child (if appli­ca­ble). We assume respon­dents started liv­ing with bio­log­ic­ al chil­dren in the year of birth (if they ever lived with their bio­log­i­cal child) and cre­ate a dummy var­i­able of coresidence with a child in each year of the per­son-year file accord­ingly. Focusing on coresidence allows us to cap­ture spells of sin­gle par­ent­hood and res­i­dence with non­bi­o­log­i­cal chil­dren. However, it does pose some lim­i­ta­tions, namely, that it does not account for non­res­i­dent chil­dren. In an addi­tional anal­ys­ is, we show that par­ent­hood trends are sim­i­lar when also con­sid­er­ing non­res­i­dent bio­log­i­cal chil­dren (see Table S2 in online Supplementary Materials B). More gen­er­ally, we acknowl­edge that our fam­ily states, that is, coresidence with part­ners and chil­dren, do not cap­ture a wider range of fam­ily roles and struc­tures in the LGB pop­u­la­tion. Given sam­ple size lim­i­ta­tions, our fam­ily sequences can­not dif­ fer­en­ti­ate between mar­riage and cohab­i­ta­tion for those partnered or between bio­log­i­ cal and non­bi­o­log­i­cal chil­dren. We pro­vide the prev­a­lence of ever hav­ing res­i­dent or non­res­i­dent (bio­log­i­cal) chil­dren as part of the descrip­tive sta­tis­tics to offer a bet­ter char­ac­ter­iza­tion of fam­ily tra­jec­to­ries among sex­ual minor­i­ties. We keep indi­vid­u­ als with miss­ing infor­ma­tion on partnering or par­ent­hood for part of the obser­va­tion period and include “miss­ing” as an addi­tional state in their fam­ily tra­jec­to­ries.8 Analytic Approach We use sequence anal­y­sis, in which a per­son’s fam­ily tra­jec­tory is con­ceived as a suc­ces­sion of fam­ily states dur­ing the prime fam­ily for­ma­tion period. In our anal­ y­sis, each respon­dent’s sequence con­sists of a series of monthly or yearly states of coresidence with part­ners and chil­dren starting from age 18. Four dif­fer­ent fam­ily states are observed: (1) sin­gle with­out cores­i­dent chil­dren, (2) sin­gle with cores­ i­dent chil­dren, (3) partnered with­out cores­i­dent chil­dren, and (4) partnered with cores­i­dent chil­dren. Our oper­a­tional def­i­ni­tion of a fam­ily sequence enables us to 7 For part of our anal­y­sis, we use per­son-year files based on the sta­tus at the start of each year. In par­tic­u­lar, our ana­ly­ses based on opti­mal matching (i.e., typol­ogy of tra­jec­tory path­ways and diver­sity index) deploy sequences of yearly states to obtain mean­ing­ful results (i.e., dis­tances between sequences) since there are few changes in fam­ily states in our sequence data, and multiple changes occur rarely within the same year. 8 This applies to miss­ing dates at the start or end of the obser­va­tion period (i.e., starting date of the first union/child or the end date of the last union/child). For other miss­ing dates, we inter­po­lated dates by tak­ ing the mid­dle point between the pre­vi­ous and sub­se­quent nonmissing dates. If two sub­se­quent dates were miss­ing, we inter­po­lated by divid­ing the period between the pre­vi­ous and sub­se­quent nonmissing dates into three. We exclude cases with more extended peri­ods of miss­ing data. 1590 A. Ophir et al. S1 S2 Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 S3 t1 t2 t3 t4 t5 t6 t7 t8 t9 t10 Time Single, no children Partnered, no children Single, with children Partnered, with children Fig. 1 Fictional sequences of family states. Sequences are chronological successions of categorical family states. assess mul­ti­ple part­ner­ship epi­sodes (i.e., union for­ma­tion, dis­so­lu­tion, and repart- nering), the tran­si­tion into par­ent­hood, and their tim­ing and sequenc­ing within a per­son’s fam­ily tra­jec­tory. Figure 1 shows three fic­tional sequences of fam­ily states over a 10-year period. In this exam­ple, S1 depicts a tra­jec­tory with no fam­ily tran­si­tions, wherein a per­son does not live with chil­dren or a part­ner through­out the obser­va­tion period; S2 fea­tures a tran­ si­tion to sin­gle par­ent­hood in the sixth obser­va­tion; and S3 fea­tures the start of a part­ner­ ship in the sec­ond obser­va­tion, and the pres­ence of a child since the seventh obser­va­tion. On the basis of these sequences, we gen­er­ate a typol­ogy of under­ly­ing fam­ily tra­jec­to­ries to answer the first research ques­tion.9 First, we com­pare each pair of sequences in our sam­ple and cal­cu­late their sim­il­ar­ity or opti­mal matching (OM) dis­tance.10 These pairwise OM dis­tances are based on cri­te­ria that account for dif­ fer­ences in tim­ing, sequenc­ing, and dura­tion of the dif­fer­ent states observed across sequences (Aisenbrey and Fasang 2010). Second, we use hier­ar­chi­cal clus­ter anal­y­ sis with the Ward link to clus­ter sim­il­ar sequences into a typol­ogy of tra­jec­to­ries or pro­files. We chose a five-clus­ter solu­tion among the avail­­able solu­tions given by the clus­ter anal­y­sis because of its empir­i­cal fit to the data at hand and because it ren­ders a set of the­o­ret­i­cally mean­ing­ful path­ways (see the cut­off cri­te­ria for dif­fer­ent clus­ter solu­tions in Figure S1 in online Supplementary Materials C). Even though sequences 9 We use the soft­ware R and the TraMineR pack­age for opti­mal matching and clus­ter ana­ly­ses (Gabadinho et al. 2011). 10 The OM dis­tance (or dis­sim­i­lar­ity) between two sequences is the min­i­mum cost of transforming one sequence into another sequence by means of sub­sti­tu­tions between ele­ments of the sequence, or their inser­ tions and dele­tions (Abbott and Tsay 2000). We use the OM-Future algo­rithm to empha­size that rel­e­vant pat­terns in fam­ily tra­jec­to­ries sub­stan­tively dif­fer in the tim­ing of key tran­si­tions across fam­ily states. The OM-Future var­i­ant con­sid­ers that states in a given posi­tion in the sequence are more sim­i­lar if they share a com­mon future state (Studer and Ritschard 2016). This ren­ders an OM dis­tance more sen­si­tive to the tim­ing when tran­si­tions occur. Since oper­a­tion costs are derived from state dis­tri­bu­tions, our dis­tance of choice is also less affected by the crit­i­cism that cost assign­ment is arbi­trary and has no socio­log­i­cal mean­ing. Family Formation Trajectories and LGB Cohort Change 1591 1.0 Single, no children Single, with children Partnered, no children .8 Partnered, with children Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 .6 Proportion .4 .2 0 18 22 26 30 34 38 Age Fig. 2 State distribution by age among LGB individuals ages 18 to 40 (N = 455). Source: Understanding Society (2009–2019). that belong to the same group are not iden­ti­cal, they con­form to a gen­eral tra­jec­ tory pat­tern, allowing us to bet­ter cat­e­go­rize fam­ily tra­jec­to­ries than what would be obtained by using sim­ple indi­ca­tors of the prev­a­lence of events. We explore cohort change through mul­ti­no­mial regres­sion anal­y­sis predicting clus­ter mem­ber­ship by birth cohort. We mea­sure the birth cohort as a dummy var­i­able that com­pares LGBs born before 1965 and between 1965 and 1979. We con­trol for gen­der, sex­ual iden­tity, edu­ca­tion, race-eth­nic­ity, and fam­ily struc­ture at age 16. We also pres­ent predicted prob­a­bil­i­ties of clus­ter mem­ber­ship by gen­der and sex­u­al­ity while con­trol­ling for the same covariates. In a sup­ple­men­tary anal­y­sis, we cal­cu­late com­pos­ite mea­sures of changes in the com­plex­ity and diver­sity of sequences used by pre­vi­ous empir­i­cal research (see online Supplementary Materials D for a more tech­ ni­cal expla­na­tion and results). We pres­ent results only for LGBs in the main text because they are the focus of our anal­y­sis. We refer to het­ero­sex­u­als when it is use­ful for com­par­a­tive con­text and pres­ent results for het­ero­sex­u­als in the online Supplementary Materials. Results Descriptive Family Formation Statistics Figure 2 shows the dis­tri­bu­tion of the four fam­ily states at every age between 18 and 40 for LGB women and men. The graph shows that singlehood, that is, not liv­ing with a part­ner or chil­dren, is a prominent state for LGBs across all­ages in our ­sam­ple. 1592 A. Ophir et al. Cluster 1: Cluster 2: Mostly single, no children (n = 172, 38%) Mostly partnered, no children (n = 128, 28%) Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 18 22 26 30 34 38 18 22 26 30 34 38 Cluster 3: Cluster 4: Early partnering & parenthood (n = 87, 19%) Delayed partnering & parenthood (n = 48, 11%) 18 22 26 30 34 38 18 22 26 30 34 38 Cluster 5: Age Single parenthood (n = 20, 4%) Single, no children Single, with children Partnered, no children Partnered, with children Missing 18 22 26 30 34 38 Age Fig. 3 Latent cluster typology of family formation among LGB individuals. Source: Understanding Soci- ety (2009–2019). About 40% of LGB sam­ple mem­bers were not liv­ing with a part­ner or chil­dren at age 40, com­pared with about 10% of the het­ero­sex­ual sam­ple mem­bers (see Figure S2 for het­ero­sex­u­als’ state dis­tri­bu­tion in online Supplementary Materials E). Being part- nered with chil­dren is also a prominent state of LGBs’ life courses, but being in a union with­out chil­dren is more com­mon (in con­trast to het­ero­sex­u­als, see Figure S2). More- over, hav­ing chil­dren within part­ner­ships occurs late in the life course. Finally, sin­gle par­ent­hood is the least com­mon state. When it does occur, it is later in the life course. Family Formation Trajectory Typology Figure 3 pres­ents results for the clus­ter anal­y­sis that pro­duced five clus­ters of fam­ily for­ma­tion tra­jec­to­ries among LGB women and men in the sam­ple: (1) mostly sin­gle, no chil­dren (n = 172, 38%); (2) mostly partnered, no chil­dren (n = 128, 28%); (3) early partnering and par­ent­hood (n = 87, 19%); (4) delayed partnering and par­ent­hood (n = 48, 11%); and (5) sin­gle par­ent­hood (n = 20, 4%). Family Formation Trajectories and LGB Cohort Change 1593 Table 1 Sample char­ac­ter­is­tics by clus­ter Singlehood: Partnership: Parenthood: Cluster 1 Cluster 2 Clusters 3–5 Overall (n  = 172) (n = 128) (n = 155) (N  = 455) Sociodemographic Characteristics Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 % women 34.9 43.0 59.4 45.5 % 1965–1979 birth cohort 45.9 58.6 47.1 49.9 % bisex­ual 25.0 22.7 67.1 38.7 % White 89.5 89.1 74.2 84.2 % par­ents together at age 16 70.3 86.6 76.8 77.1 Education (years, mean) 14 15 14 14 % ever had a res­i­dent part­ner 58 100 99 84 Number of part­ners (mean) 0.95 1.49 1.43 1.27 % ever had a non­res­i­dent bio­log­i­cal child 6.4 3.9 56.8 22.9 Mean Total Duration in Each State (months) Single, no chil­dren 240.4 120.2 72.8 149.7 Single, with chil­dren 2.1 0.4 28.4 10.6 Partnered, no chil­dren 27.9 148.7 50.7 69.7 Partnered, with chil­dren 1.3 2.5 120.2 42.0 Indices (mean) Dissimilarity index 0.1 0.4 0.6 0.4 Complexity index 1.7 2.4 3.3 2.4 Notes: Dissimilarity index ranges from 0 (least het­ero­ge­ne­ity) to 1 (most het­ero­ge­ne­ity). Complexity index is the num­ber of tran­si­tions between states over the life course. Source: Understanding Society (2009–2019). Two thirds of the sam­ple belonged to Clusters 1 and 2, which are char­ac­ter­ized by not resid­ing with chil­dren but dif­fer in partnering pat­terns. The larg­est clus­ter in the sam­ple is Cluster 1, with 38%; it is char­ac­ter­ized by sta­ble singlehood or short part­ ner­ship spells, mostly between ages 24 and 34. A sig­nif­i­cant share of the tra­jec­to­ries in this clus­ter (58%, see Table 1) had a part­ner­ship at least once by age 40, but these part­ner­ships were short-lived (28 months on aver­age). Twenty-eight per­cent of the sam­ple belonged to Cluster 2, which is char­ac­ter­ized by long part­ner­ship spells after age 30 or ear­lier partnering and repartnering through­out the observed life course. A third of the sam­ple was dis­trib­uted across Clusters 3, 4, and 5, char­ac­ter­ized by cores­i­dent par­ent­hood but dif­fer­ing partnering pat­terns. Nineteen per­cent belonged to Cluster 3, char­ac­ter­ized by early partnering and par­ent­hood tran­si­tions by the mid-20s. Another 11% belonged to Cluster 4, with delayed partnering and par­ent­hood tran­si­tions. Specifically, this clus­ter includes fam­ily for­ma­tion tra­jec­to­ries that start with early union for­ma­tion and later par­ent­hood tran­si­tion or delayed union for­ma­ tion with a tran­si­tion to par­ent­hood soon after. Finally, a small group of LGB sam­ple mem­bers (4%) belonged to Cluster 5, char­ac­ter­ized by long sin­gle par­ent­hood spells. Because of the small sam­ple size, we com­bined Clusters 3, 4, and 5 into one group of par­ent­hood clus­ters for the rest of the anal­y­sis. What sociodemographic char­ac­ter­is­tics are asso­ci­ated with belong­ing to each clus­ ter? Table 1 shows bivar­i­ate descrip­tive sociodemographic char­ac­ter­is­tics for each of the three (groups of) clus­ters: mostly singlehood (labeled “singlehood” h­ ere­af­ter 1594 A. Ophir et al. for brev­ity), mostly part­ner­ship (“part­ner­ship” here­af­ter), and cores­i­dent par­ent­hood (“par­ent­hood” here­af­ter). Overall, women and bisex­ual peo­ple were under­rep­re­sented in the singlehood clus­ter but over­rep­re­sented in the par­ent­hood clus­ters. White LGBs were rel­a­tively less likely to belong to a par­ent­hood clus­ter. LGBs whose par­ents were not together when they were 16 years old were more likely to be in the singlehood clus­ter, whereas Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 LGBs whose par­ents were together were more likely to be in the part­ner­ship clus­ ter. Overall, there are no nota­ble dif­fer­ences in edu­ca­tion across the clus­ters in our sam­ple. Table 1 also shows that more than half of the peo­ple in the singlehood clus­ter have had a cores­i­dent part­ner, but as Figure 3 showed, most of these part­ner­ships were short-lived. On aver­age, the num­ber of part­ners was sim­i­lar among LGBs in the part­ner­ship and par­ent­hood clus­ters (1.5 and 1.4 part­ners, respec­tively). More than half (57%) of the peo­ple in the par­ent­hood clus­ters have ever had non­res­i­dent minor chil­dren, in con­trast to 6% and 4% among LGBs in the singlehood and part­ner­ship clus­ters, respec­tively. This strik­ing con­trast implies that a very small pro­por­tion of LGBs in these clus­ters were ever par­ents. Cohort Change in Family Formation Trajectories Cluster Membership Across Cohorts We first exam­ine how the two LGB cohorts dis­trib­ute across clus­ters to explore cohort change in fam­ily tra­jec­to­ries. Table 1 shows that LGBs born between 1965 and 1979 were more likely to belong to the part­ner­ship clus­ter than LGBs born before 1965. Table 2 shows results for a mul­ti­no­mial regres­sion anal­y­sis predicting clus­ter mem­ ber­ship. These results con­firm that the youn­ger LGB cohort was more likely than the older cohort to expe­ri­ence part­ner­ship tra­jec­to­ries than par­ent­hood or singlehood tra­jec­to­ries. Figure 4 gives bet­ter insight into abso­lute changes in the prev­a­lence of fam­ily tra­jec­to­ries and how these changes dif­fered by gen­der and sex­ual iden­tity (using a three-way inter­ac­tion of gen­der, sex­ual iden­tity, and cohort to pre­dict the prob­a­bil­ity of belong­ing to each of the three groups of clus­ters). We observe that both gay men and les­bian women have become con­sid­er­ably more likely to fol­low part­ner­ship tra­ jec­to­ries across cohorts. This has come at the expense of tra­jec­to­ries char­ac­ter­ized by par­ent­hood and singlehood, although few gay men followed par­ent­hood tra­jec­to­ries in the older cohort. Among gay men, singlehood was clearly the most com­mon fam­ily tra­jec­tory type among the older cohort (60%), but part­ner­ship tra­jec­to­ries were most com­mon (46%) among the youn­ger cohort. For les­bian women, both ­par­ent­hood and singlehood tra­jec­to­ries were prev­a­lent among the older cohort (39% and 40%, respec­tively), yet part­ner­ship tra­jec­to­ries became the most com­mon tra­jec­tory among the youn­ger cohort (51%). These results align with the expec­ta­tion that part­ner­ship tra­jec­to­ries will become more prev­a­lent across cohorts as the oppor­tu­ni­ties to meet same-sex part­ners increase and the stigma and dis­crim­i­na­tion decline over time. However, our results regard­ing par­ent­hood sug­gest that only rel­a­tively small shares of the cohorts stud­ied have taken Family Formation Trajectories and LGB Cohort Change 1595 Table 2 Relative risk ratios from mul­ti­no­mial regres­sion ana­ly­ses for clus­ter mem­ber­ship Parenthood Singlehood Parenthood vs. Partnership vs. Partnership vs. Singlehood Birth Cohort 1965–1979 0.43** 0.52** 0.82 (0.12) (0.13) (0.21) Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 Women 2.16** 0.75 2.88** (0.58) (0.18) (0.73) Bisexual 6.48** 1.18 5.50** (1.89) (0.34) (1.44) Education (years) 0.97 1.00 0.97 (0.03) (0.03) (0.03) Parents Together at Age 16 0.43* 0.32** 1.35 (0.15) (0.10) (0.39) White 0.51 0.99 0.51 (0.20) (0.41) (0.18) Constant 3.17 5.22** 0.61 (2.13) (3.30) (0.35) Notes: Standard errors are shown in paren­the­ses. n = 453. Source: Understanding Society (2009–2019). *p < .05; **p < .01 Women Men 1.0 .17 .14 .15 .20 .22 .26 .8 .22 .46 .51 .18 Predicted Probability .25 .38 .6 .40 .60 .4 .30 .65 .64 .44 .52 .47 .2 .39 .19 .13 .10 0 Before 1965– Before 1965– Before 1965– Before 1965– 1965 1979 1965 1979 1965 1979 1965 1979 Lesbian Bisexual Gay Bisexual Parenthood Singlehood Partnership Fig. 4 Predicted probability for cluster membership by birth cohort, gender, and sexuality. Multinomial regression analyses include interactions between sexual identity, gender, and cohort, and control for race-ethnicity, education, and family structure at 16. Source: Understanding Society (2009–2019). 1596 A. Ophir et al. Women Men 1.0 .8 Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 Dissimilarity .6 .4 .2 0 Lesbian & Lesbian Bisexual Gay & Gay Bisexual bisexual bisexual Before 1965 1965–1979 Fig. 5 Dissimilarity across birth cohorts, by gender and sexuality. Dissimilarity index ranges from 0 (least heterogeneity) to 1 (most heterogeneity). Source: Understanding Society (2009–2019). advan­tage of increas­ing access to alter­na­tive routes into par­ent­hood, such as ARTs and adop­tion. Results for bisex­ual women and men show dif­fer­ent pat­terns of change than those observed among gay/les­bian indi­vid­u­als. Although the sam­ple size calls for cau­tion in inter­pre­ta­tion, the descrip­tive pat­terns among bisex­u­als align with gen­eral nar­ra­ tives of demo­graphic change. Bisexuals from the youn­ger birth cohort in our sam­ple have expe­ri­enced a decline in par­ent­hood and part­ner­ship tra­jec­to­ries but an increase in singlehood tra­jec­to­ries. This change was more sub­stan­tial among bisex­ual men than bisex­ual women. Additional Analysis: Measures of Diversity and Complexity Empirical stud­ies on demo­graphic change often pro­vide sum­mary mea­sures of how event­ful fam­ily tra­jec­to­ries are (com­plex­ity) and how diverse tra­jec­to­ries are across indi­vid­u­als (diver­sity) to quan­tify other forms of change in life tra­jec­to­ries. Figures 5 and 6 illus­trate such mea­sures for our sam­ple, and we briefly sum­ma­rize the main takeaways here (see Supplementary Material D for more meth­od­o­log­i­cal infor­ma­tion and elab­o­rated results). Overall, we find that fam­ily tra­jec­to­ries of les­bian women and bisex­ual men became less diverse over time, mean­ing their fam­ily path­ways are becom­ing more sim­i­lar to each group’s rep­re­sen­ta­tive tra­jec­tory (Figure 5). This result does not imply that all­les­bian women and bisex­ual men fol­low the same fam­ily for­ma­tion path­ways, Family Formation Trajectories and LGB Cohort Change 1597 Women Men 4 3 Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 Complexity 2 1 0 Lesbian & Lesbian Bisexual Gay & Gay Bisexual bisexual bisexual Before 1965 1965–1979 Fig. 6 Complexity across birth cohorts, by gender and sexuality. Complexity index is the number of transi- tions between states over the life course. Source: Understanding Society (2009–2019). nor that the sub­jec­tive expe­ri­ence of sim­il­ar path­ways is alike. However, regard­ ing cores­i­dent part­ners and chil­dren, the sequences of youn­ger les­bian women and bisex­ual men in our sam­ple were more likely to be sim­i­lar to one another than the sequences of their peers born before 1965. Results for com­plex­ity—that is, the num­ber of events (tran­si­tion from one state to another) peo­ple expe­ri­ence—show a rel­a­tively uni­form increase across cohorts for all­groups (Figure 6). This prob­a­bly reflects the replace­ment of low-com­plex­ity tra­ jec­to­ries, such as those char­ac­ter­ized by singlehood, with tra­jec­to­ries char­ac­ter­ized by part­ner­ship. Discussion Research on demo­graphic change has mostly overlooked the fam­ily tra­jec­to­ries of sex­ual minor­i­ties (Badgett et al. 2021; Perales 2016; Russell et al. 2020). Existing empir­i­cal research has yet to doc­u­ment how part­ner­ships, par­ent­hood, and single- hood evolve over the life course of sex­ual minor­i­ties. Similarly, existing the­o­ret­i­cal nar­ra­tives of demo­graphic change focus on the gen­eral pop­u­la­tion and can, at best, explain why same-sex unions have become more vis­i­ble in recent decades. These nar­ra­tives pro­vide lit­tle under­stand­ing of demo­graphic change among sex­ual minor­ i­ties in a con­text where pos­si­bil­i­ties to pur­sue part­ner­ships, marry, and have chil­dren within same-sex rela­tion­ships have expanded. In this arti­cle, we pro­vide what is, to our knowl­edge, the first quan­ti­ta­tive descrip­tion of fam­ily life courses among LGB 1598 A. Ophir et al. p­ er­sons and the first assess­ment of how fam­ily tra­jec­to­ries changed across two cohorts of LGBs in the UK. Our results pro­vided sev­eral novel empir­i­cal obser­va­tions. First, we iden­ti­fied five dis­tinct pro­files of fam­ily tra­jec­to­ries between ages 18 and 40. Trajectories char­ac­ter­ized by singlehood or part­ner­ship with­out chil­dren were the most prev­a­lent. Previous cross-sec­tional research in the United States dem­on­strated that LGBs were less likely to have a part­ner at a given time than het­ero­sex­u­als, except Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 for les­bian women (Badgett et al. 2021). We add to this body of work that a con­sid­er­ able share of LGBs born before 1979 did not coreside with a part­ner before age 40 (or expe­ri­enced very short partnership spells) in the UK. For some groups, these shares were sub­stan­tial. For instance, 60% of gay men born before 1965 followed a tra­jec­ tory that mainly consisted of singlehood. This pat­tern means that empir­i­cal stud­ies based on same-sex cou­ples omit a large part of the LGB pop­u­la­tion, espe­cially among older cohorts of gay men. We also found some dif­fer­ences com­pared with stud­ies from the United States (Badgett et al. 2021). For exam­ple, tra­jec­to­ries char­ac­ter­ized by singlehood were more prev­a­lent among les­bian women than among bisex­ual women in the UK. We also found that gay men were more likely to fol­low singlehood tra­jec­to­ries than were bisex­ual men. Both results dif­fer from pat­terns observed in the United States, where bisex­ual indi­vid­u­als were more likely to be sin­gle than were gay/les­bian indi­vid­ u­als (Badgett et al. 2021). This diver­gence could be because we take a life course rather than a cross-sec­tional approach. This diver­gence high­lights the impor­tance of a life course per­spec­tive in under­stand­ing fam­ily for­ma­tion pro­cesses among LGBs.11 However, we also observed that singlehood tra­jec­to­ries became more com­mon among bisex­u­als across cohorts, an issue we elab­o­rate on later. Our sec­ond novel obser­va­tion is the increas­ing prev­a­lence of tra­jec­to­ries char­ac­ ter­ized by cores­i­den­tial part­ner­ships among the youn­ger gay and les­bian indi­vid­u­als in our sam­ple (born between 1965 and 1979). This trend was expected amid the increas­ing avail­abil­ity and vis­i­bil­ity of poten­tial LGB part­ners, the decreas­ing stigma and dis­crim­i­na­tion toward same-sex cou­ples (Rosenfeld 2017; Trandafir 2015), and LGBs’ expressed inter­est in hav­ing a part­ner (Barrios and Lundquist 2012; D’Augelli et al. 2007; Riskind and Tornello 2017; Tate et al. 2019). Which mech­a­nism spe­cif­ i­cally drives this trend is a ques­tion for future research. However, if the under­ly­ing mech­a­nism is the decreas­ing stigma and insti­tu­tional dis­crim­i­na­tion, we expect that singlehood tra­jec­to­ries will con­tinue to be highly prev­a­lent among gay/les­bian indi­ vid­u­als in con­texts where stigma and dis­crim­i­na­tion are higher than in the UK. None- theless, our results for gay and les­bian indi­vid­u­als con­tra­dict pre­dic­tions derived from the­o­ries of demo­graphic change. The quest for inde­pen­dence is cen­tral to the sec­ond demo­graphic tran­si­tion frame­work, which should man­i­fest in more singlehood tra­ jec­to­ries across cohorts rather than a decrease. Therefore, to under­stand demo­graphic change among LGBs, demog­ra­phers should also con­sider the diver­sity of fam­ily for­ ma­tion expe­ri­ences within soci­e­ties. 11 To illus­trate, if we cross-tab­u­late part­ner­ship sta­tus with sex­ual iden­tity using the 2017/2018 wave, dif­ fer­ences between het­ero­sex­ual and les­bian women dis­ap­pear: 54% of les­bian women are in a cores­i­den­tial rela­tion­ship com­pared with 59% of bisex­ual and 57% of het­ero­sex­ual women. However, for bisex­ual men, the rel­a­tively high rates of partnering per­sist, with 37% of gay men in a cores­i­den­tial rela­tion­ship com­ pared with 49% of bisex­ual men and 64% of het­ero­sex­ual men. Family Formation Trajectories and LGB Cohort Change 1599 The results for bisex­u­als, how­ever, were con­sis­tent with the trend in the gen­eral pop­u­la­tion. Bisexuals in the youn­ger birth cohort, espe­cially bisex­ual men, were more likely to expe­ri­ence a singlehood tra­jec­tory than the older birth cohort. Barriers to same-sex partnering restricted options for both bisex­ual and gay/les­bian peo­ple. However, the avail­abil­ity of dif­fer­ent-sex partnering could explain why bisex­u­als in the older cohort were more likely to fol­low a partnering or par­ent­hood tra­jec­ Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 tory than gay and les­bian peo­ple. As some bar­ri­ers to same-sex partnering have been reduced, youn­ger bisex­ual men could opt out of dif­fer­ent-sex rela­tion­ships as a form of inde­pen­dence. However, the decline in bisex­u­als’ part­ner­ship tra­jec­to­ries could also reflect var­ i­a­tion in stig­mas that dif­fer­ent sex­ual minor­ity groups expe­ri­ence amid the sec­ond demo­graphic tran­si­tion. For instance, bisex­u­als expe­ri­ence dou­ble era­sure by both het­ero­sex­u­als and other sex­ual minor­i­ties, which reduces their partnering oppor­tu­ni­ ties (Morgenroth et al. 2022). This biphobia impacts bisex­ual men more than bisex­ual women, which could explain the more pro­nounced increase in singlehood tra­jec­to­ries among bisex­ual men. In other words, although bisex­u­als’ trends align with trends in the gen­eral pop­u­la­tion, they could be driven by dif­fer­ent soci­e­tal forces. Hence, our results under­score the impor­tance of under­stand­ing the diver­sity of fam­ily tra­jec­to­ ries across dif­fer­ent sex­ual minor­ity groups in future research. A third novel empir­i­cal obser­va­tion was that a nonnegligible but rel­a­tively small share of UK LGBs expe­ri­enced life courses that included prolonged spells of cores- idence with chil­dren. This share was smallest among gay men and larg­est among bisex­ual men and women. Parenthood tra­jec­to­ries became slightly less com­mon across cohorts for gay men, bisex­ual men, and bisex­ual women, but a pro­nounced decline was observed among les­bian women. This trend aligns with evi­dence of a decline in same-sex cou­ples’ like­li­hood to coreside with chil­dren born in pre­vi­ous dif­fer­ent-sex rela­tion­ships (Gates 2015). This expla­na­tion is plau­si­ble because the most nota­ble decline in our data was among les­bian women, who are more likely to live with chil­dren from past dif­fer­ent-sex rela­tion­ships than are gay men.12 Nonetheless, the decline in par­ent­hood tra­jec­to­ries is par­a­dox­i­cal amid rel­a­tive expan­sion in access to planned path­ways to par­ent­hood for same-sex cou­ples, includ­ ing adop­tion and ARTs. What could explain this par­a­dox? First, youn­ger LGBs might be more likely to reject heteronormative fam­ily tra­jec­to­ries (Hank and Wetzel 2018; Riskind and Patterson 2010). Second, planned path­ways to par­ent­hood might be out of reach for youn­ger LGBs. Despite sig­nif­i­cant prog­ress in ARTs, the pro­ce­dures are costly (Cozzani et al. 2021; Park et al. 2020). Moreover, sub­sidy eli­gi­bil­ity is deter­ mined by heteronormative cri­te­ria, such as not con­ceiv­ing after hav­ing had unpro­ tected inter­course for at least a year. Consequently, women in same-sex cou­ples are often denied subsidies unless they try to con­ceive through pri­vately funded insem­i­na­ tion donors or have under­gone an infer­til­ity test (Warnes 2019). Similarly, although adop­tion became avail­­able to same-sex cou­ples in the 2000s, rates are still low and same-sex cou­ples are over­rep­re­sented among the cou­ples who adopt hard-to-place 12 In online Supplementary Materials, we show the per­cent­ages of ever hav­ing a bio­log­i­cal child regard­ less of coresidence. These per­cent­ages are higher, espe­cially among gay men in the older cohort. None- theless, the sup­ple­men­tary anal­y­sis also shows a decline in the prev­al­ence of par­ent­hood across cohorts when using this mea­sure. 1600 A. Ophir et al. chil­dren in the UK (Tasker and Bellamy 2019). Future research should inves­ti­gate cohort change in LGBs’ path­ways to par­ent­hood across more and less favor­able con­ texts than the UK. The decline in par­ent­hood tra­jec­to­ries also raises ques­tions about LGBs’ future care needs as they age. Sexual and gen­der minor­i­ties have his­tor­i­cally relied on cho­ sen fam­i­lies and their com­mu­ni­ties for care sup­port owing to indi­vid­ual and insti­tu­ Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 tional dis­crim­i­na­tion (Allen and Lavender-Stott 2020; Boyd and Wei forth­com­ing; Jackson Levin et al. 2020; Reczek 2020). However, amid pop­u­la­tion aging and increas­ing demand for fam­ily-pro­vided care, an addi­tional unmet need for care could emerge among LGBs of the youn­ger cohorts if they con­tinue to be less likely to expe­ ri­ence par­ent­hood tra­jec­to­ries. Future research should inves­ti­gate how fam­ily-related changes affect dif­fer­ent sex­ual minor­ity groups’ access to care sup­port over time. Besides describ­ing fam­ily tra­jec­tory types, we also explored the com­plex­ity and diver­sity of LGBs’ fam­ily tra­jec­to­ries, which is another pil­lar of demo­graphic change research. Previous research argued that fam­ily tra­jec­to­ries have become more event­ ful (i.e., more com­plex) and diverse (i.e., more het­ero­ge­neous) among the gen­eral pop­u­la­tion (Van Winkle 2018). However, our results were mixed. On the one hand, LGBs’ fam­ily tra­jec­to­ries became more com­plex across cohorts owing to a decrease in singlehood tra­jec­to­ries. On the other hand, the trend in diver­sity, that is, the extent to which peo­ple’s tra­jec­to­ries are dif­fer­ent from a sin­gle dom­i­nant tra­jec­tory, was dif­fer­ent across groups. Lesbian women’s tra­jec­to­ries became slightly less diverse, while gay men’s tra­jec­to­ries became slightly more diverse across cohorts. Both groups expe­ri­enced increases in part­ner­ship tra­jec­to­ries that increased diver­sity, and both groups expe­ri­enced declines in par­ent­hood tra­jec­to­ries that decreased diver­sity. However, for les­bian women, the decline in par­ent­hood tra­jec­to­ries was much more sub­stan­tial, lead­ing to an over­all decline in diver­sity. Our study has sev­eral lim­i­ta­tions that offer direc­tions for future research. First, our small sam­ple lim­ited our abil­ity to break down trends by edu­ca­tion, race, eth­nic­ity, or more detailed cohorts. In addi­tion, our LGB sam­ple could only iden­tify as women or men, and sex­ual iden­ti­ties beyond LGB were not avail­­able as answer options. None- theless, using sex­ual iden­tity mea­sures is one step for­ward beyond existing prac­tices in fam­ily schol­ar­ship that focus on same-sex cou­ples. The sex­ual iden­tity mea­sure allowed us, for instance, to uncover dif­fer­ences between bisex­ual and gay/les­bian indi­vid­u­als and to look at LGB indi­vid­u­als who are sin­gle or sin­gle par­ents. Second, our con­cep­tu­al­i­za­tion of fam­ily for­ma­tion is lim­ited to coresidence with part­ners and chil­dren, representing a heteronormative def­i­ni­tion of fam­ily life (Reczek 2020; van Eeden-Moorefield et al. 2011). Our approach does not include non­res­i­dent part­ners and other forms of part­ner­ships and rela­tion­ships that could be more prominent in LGBs’ per­cep­tion of a fam­ily (Allen and Lavender-Stott 2020; Reczek 2020; van Eeden-Moorefield et al. 2011). However, we dem­on­strated a cohort change within the scope of cores­id­ en­tial famil­ial rela­tion­ships. These cores­i­ den­tial trends are essen­tial to doc­u­ment amid existing social and insti­tu­tional bar­ri­ers. Moreover, our sequence anal­y­sis allowed us to focus on the diver­sity and com­plex­ity of fam­ily tra­jec­to­ries amid a small sam­ple. Future stud­ies should inves­ti­gate fur­ther the fam­ily dynam­ics of sex­ual minor­i­ties over the life course, namely, the life courses of youn­ger LGB indi­vid­u­als born after 1980, who are in the pro­cess of fam­ily for­ma­ tion within a rap­idly chang­ing con­text. Family Formation Trajectories and LGB Cohort Change 1601 Despite these lim­i­ta­tions, we expanded existing research about fam­ily for­ma­tion among sex­ual minor­it­ies by tak­ing a novel life course approach that cen­ters on the dynamic, com­plex, and diverse ways that part­ner­ship and par­ent­hood evolve over the life course of LGB women and men. This study con­trib­uted by documenting cohort change in fam­ily for­ma­tion among LGBs and how these changes are prolonged and unequal despite legal expan­sions and social changes. Our results have con­se­quences Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 for gen­eral nar­ra­tives of demo­graphic change over time. Whereas chang­ing social norms have made it eas­ier for sig­nif­i­cant parts of the gen­eral pop­u­la­tion to “devi­ate” from the nor­ma­tive tra­jec­tory of early fam­ily for­ma­tion, the same social changes might have facil­i­tated cores­i­den­tial partnering tra­jec­to­ries for the LGB pop­u­la­tion. We also expected par­ent­hood tra­jec­to­ries to become more com­mon across cohorts for the LGB pop­u­la­tion but found the oppo­site for the cohorts stud­ied here. This seems to imply that sex­ual minor­i­ties con­tinue to face obsta­cles to access alter­na­tive routes into par­ent­hood, such as adop­tion and ARTs. These results illus­trate that demo­ graphic change man­i­fests dif­fer­ently for dif­fer­ent groups within a soci­ety. Hence, existing nar­ra­tives of fam­ily-related demo­graphic change should explic­itly con­sider how dif­fer­ent groups within soci­e­ties, such as sex­ual minor­i­ties, expe­ri­ence and react to social and demo­graphic changes. ■ Acknowledgments Earlier ver­sions of this man­u­script were presented at the 2022 annual meet­ing of the Population Association of America (Atlanta, GA) and the 2022 Euro­pean Population Conference (Groningen, the Netherlands). The research has benefited from funding from the Euro­pean Research Council (ERC-2020-STG-948557-MINEQ; PI: Diederik Boertien). Ariane Ophir (FJC2021-048042-I) and Diederik Boertien (PID2021-124267OB-I00; RYC2019-026510-I) also acknowl­edge funding from the Span­ish Ministry of Education. References Abbott, A., & Tsay, A. (2000). Sequence anal­y­sis and opti­mal matching meth­ods in soci­ol­ogy: Review and pros­pect. Sociological Methods & Research, 29, 3–33. Adamczyk, A., & Liao, Y.-C. (2019). Examining pub­lic opin­ion about LGBTQ-related issues in the United States and across mul­ti­ple nations. Annual Review of Sociology, 45, 401–423. Aisenbrey, S., & Fasang, A. E. (2010). New life for old ideas: The “sec­ond wave” of sequence anal­y­sis bring­ing the “course” back into the life course. Sociological Methods & Research, 38, 420–462. Allen, K. R., & Lavender-Stott, E. S. (2020). The fam­i­lies of LGBTQ older adults: Theoretical approaches to cre­a­tive fam­ily con­nec­tions in the con­text of mar­gin­al­i­za­tion, social-his­tor­i­cal change, and resil- ience. Journal of Family Theory & Review, 12, 200–219. Andersson, G., Noack, T., Seierstad, A., & Weedon-Fekjær, H. (2006). The demo­graph­ics of same-sex mar­riages in Norway and Sweden. Demography, 43, 79–98. Badgett, M. V. L., Carpenter, C. S., & Sansone, D. (2021). LGBTQ eco­nom­ics. Journal of Economic Perspectives, 35(2), 141–170. Barrios, R. J., & Lundquist, J. H. (2012). Boys just want to have fun? Masculinity, sex­ual behav­iors, and roman­tic inten­tions of gay and straight males in col­lege. Journal of LGBT Youth, 9, 271–296. Baumle, A. K., & Compton, D. R. (2011). Legislating the fam­ily: The effect of state fam­ily laws on the pres­ence of chil­dren in same-sex house­holds. Law & Policy, 33, 82–115. Black, D., Gates, G. J., Sanders, S., & Taylor, L. (2000). Demographics of the gay and les­bian pop­u­la­tion in the United States: Evidence from avail­­able sys­tem­atic data sources. Demography, 37, 139–154. Bosley-Smith, E. R., & Reczek, C. (2018). Before and after “I do”: Marriage pro­cesses for mid-life gay and les­bian mar­ried cou­ples. Journal of Homosexuality, 65, 1985–2004. 1602 A. Ophir et al. Boyd, A., & Wei, J. (forth­com­ing). Stretched kin­ship: Queer female uni­ver­sity stu­dents nego­ti­at­ing fam­ ily and iden­tity. Sexualities. Advance online pub­li­ca­tion. https:​­/​­/doi​­.org​­/10​­.1177​­/13634607231152598 Brückner, H., & Mayer, K. U. (2005). De-stan­dard­i­za­tion of the life course: What it might mean? And if it means any­thing, whether it actu­ally took place? Advances in Life Course Research, 9, 27–53. Buck, N., & McFall, S. (2011). Understanding soci­ety: Design over­view. Longitudinal and Life Course Studies, 3, 5–17. Bukodi, E. (2012). Serial cohab­i­ta­tion among men in Britain: Does work his­tory mat­ter? Euro­pean Journal Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 of Population / Revue Européenne de Démographie, 28, 441–466. Carpenter, C. S. (2020). The direct effects of legal same-sex mar­riage in the United States: Evidence from Mas­sa­chu­setts. Demography, 57, 1787–1808. Carpenter, C. S., & Gates, G. J. (2008). Gay and les­bian part­ner­ship: Evidence from California. Demography, 45, 573–590. Costa, P. A., & Bidell, M. (2017). Modern fam­i­lies: Parenting desire, inten­tion, and expe­ri­ence among Por­tu­guese les­bian, gay, and bisex­ual indi­vid­u­als. Journal of Family Issues, 38, 500–521. Cozzani, M., Aradhya, S., & Goisis, A. (2021). The cog­ni­tive devel­op­ment from child­hood to ado­les­cence of low birthweight chil­dren born after med­i­cally assisted repro­duc­tion—A UK lon­gi­tu­di­nal cohort study. International Journal of Epidemiology, 50, 1514–1523. D’Augelli, A. R., Rendina, H. J., Sinclair, K. O., & Grossman, A. H. (2007). Lesbian and gay youth’s aspi­ ra­tions for mar­riage and rais­ing chil­dren. Journal of LGBT Issues in Counseling, 1(4), 77–98. Dewaele, A., Cox, N., Van den Berghe, W., & Vincke, J. (2011). Families of choice? Exploring the sup­port­ ive net­works of les­bi­ans, gay men, and bisex­u­als. Journal of Applied Social Psychology, 41, 312–331. Duggan, L. (2002). The new homonormativity: The sex­ual pol­i­tics of neo­lib­er­al­ism. In R. Castronovo & D. D. Nelson (Eds.), Materializing democ­racy: Toward a revi­tal­ized cul­tural pol­i­tics (pp. 175–194). Dur­ham, NC: Duke University Press. Eickmeyer, K. J., & Manning, W. D. (2018). Serial cohab­i­ta­tion in young adult­hood: Baby boom­ers to mil­len­ni­als. Journal of Marriage and Family, 80, 826–840. Ela, E. J., & Budnick, J. (2017). Non-het­ero­sex­u­al­ity, rela­tion­ships, and young women’s con­tra­cep­tive behav­ior. Demography, 54, 887–909. Ellingson, S., Laumann, E. O., Paik, A., & Mahay, J. (2004). The the­ory of sex mar­kets. In E. O. Laumann, S. Ellingson, J. Mahay, A. Paik, & Y. Youm (Eds.), The sex­ual orga­ni­za­tion of the city (pp. 3–38). Chicago, IL: Chicago University Press. Elliott, M. N., Dahlhamer, J. M., MacCarthy, S., Beckett, M. K., Orr, N., Guerino, P., . . . ​Martino, S. C. (2019). Using ancil­lary sociodemographic data to iden­tify sex­ual minor­ity adults among those responding “some­thing else” or “don’t know” to sex­ual ori­en­ta­tion ques­tions. Medical Care, 57, e87–e95. https:​­/​­/doi​­.org​­/10​­.1097​­/MLR​­.0000000000001190 Elzinga, C. H., & Liefbroer, A. C. (2007). De-stan­dard­i­za­tion of fam­ily-life tra­jec­to­ries of young adults: A cross-national com­par­i­son using sequence anal­y­sis. Euro­pean Journal of Population / Revue Européenne de Démographie, 23, 225–250. England, P., Mishel, E., & Caudillo, M. L. (2016). Increases in sex with same-sex part­ners and bisex­ual iden­tity across cohorts of women (but not men). Sociological Science, 3, 951–970. Esping-Andersen, G., & Billari, F. C. (2015). Re-the­o­riz­ing fam­ily demo­graph­ics. Population and Development Review, 41, 1–31. Everett, B. G., Limburg, A., Charlton, B. M., Downing, J. M., & Mat­thews, P. A. (2021). Sexual iden­tity and birth out­comes: A focus on the mod­er­at­ing role of race–eth­nic­ity. Journal of Health and Social Behavior, 62, 183–201. Frost, D. M., & Gola, K. A. (2015). Meanings of inti­macy: A com­par­is­ on of mem­bers of het­ero­sex­ual and same-sex cou­ples. Analyses of Social Issues and Public Policy, 15, 382–400. Frost, D. M., LeBlanc, A. J., de Vries, B., Alston-Stepnitz, E., Stephenson, R., & Woodyatt, C. (2017). Couple-level minor­ity stress: An exam­i­na­tion of same-sex cou­ples’ unique expe­ri­ences. Journal of Health and Social Behavior, 58, 455–472. Gabadinho, A., Ritschard, G., Müller, N. S., & Studer, M. (2011). Analyzing and visu­al­iz­ing state sequences in R with TraMineR. Journal of Statistical Software, 40(4), 1–37. Gates, G. J. (2012). Family for­ma­tion and rais­ing chil­dren among same-sex cou­ples (Family Focus Report, No. FF51). Minneapolis, MN: National Council on Family Relations. Gates, G. J. (2015). Marriage and fam­ily: LGBT indi­vid­u­als and same-sex cou­ples. Future of Children, 25(2), 67–87. Family Formation Trajectories and LGB Cohort Change 1603 Gilroy, C., & Kashyap, R. (2021). Digital traces of sexualities: Understanding the salience of sex­ual iden­ tity through dis­clo­sure on social media. Socius, 7. https:​­/​­/doi​­.org​­/10​­.1177​­/23780231211029499 Goldberg, A. E. (2006). The tran­si­tion to par­ent­hood for les­bian cou­ples. Journal of GLBT Family Studies, 2, 13–42. Goldscheider, F., Bernhardt, E., & Lappegård, T. (2015). The gen­der rev­o­lu­tion: A frame­work for under­ stand­ing chang­ing fam­ily and demo­graphic behav­ior. Population and Development Review, 41, 207–239. Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 Gonzales, G., & Blewett, L. A. (2014). National and state-spe­cific health insur­ance disparities for adults in same-sex rela­tion­ships. Amer­i­can Journal of Public Health, 104(2), e95–e104. https:​­/​­/doi​­.org​­/10​ ­.2105​­/AJPH​­.2013​­.301577 Hank, K., & Wetzel, M. (2018). Same-sex rela­tion­ship expe­ri­ences and expec­ta­tions regard­ing part­ner­ship and par­ent­hood. Demographic Research, 39, 701–718. https:​­/​­/doi​­.org​­/10​­.4054​­/DemRes​­.2018​­.39​­.25 Hatzenbuehler, M. L., Keyes, K. M., & Hasin, D. S. (2009). State-level pol­ic­ ies and psy­chi­at­ric mor­bid­ity in les­bian, gay, and bisex­ual pop­u­la­tions. Amer­i­can Journal of Public Health, 99, 2275–2281. Hiekel, N., & Fulda, B. E. (2018). Love. Break up. Repeat: The prev­al­ence and sta­bil­ity of serial cohab­ i­ta­tion among West Ger­man women and men born in the early 1970s. Demographic Research, 39, 855–870. https:​­/​­/doi​­.org​­/10​­.4054​­/DemRes​­.2018​­.39​­.30 Holmberg, D., & Blair, K. L. (2016). Dynamics of per­ceived social net­work sup­port for same-sex ver­sus mixed-sex rela­tion­ships. Personal Relationships, 23, 62–83. Jackson Levin, N., Kattari, S. K., Piellusch, E. K., & Watson, E. (2020). “We just take care of each other”: Navigating ‘cho­sen fam­ily’ in the con­text of health, ill­ness, and the mutual pro­vi­sion of care amongst queer and trans­gen­der young adults. International Journal of Environmental Research and Public Health, 17, 7346. https:​­/​­/doi​­.org​­/10​­.3390​­/ijerph17197346 Johnson-Hanks, J. A., Bachrach, C. A., Mor­gan, S. P., & Kohler, H.-P. (2011). Understanding pop­u­la­ tion trends and pro­cesses: Vol. 5. Understanding fam­ily change and var­ia­ ­tion: Toward a the­ory of conjunctural action. Dordrecht, the Netherlands: Springer Science+Business Media. Joyner, K., Manning, W. D., & Bogle, R. (2017). Gender and the sta­bil­ity of same-sex and dif­fer­ent-sex rela­tion­ships among young adults. Demography, 54, 2351–2374. Kail, B. L., Acosta, K. L., & Wright, E. R. (2015). State-level mar­riage equal­ity and the health of same-sex cou­ples. Amer­i­can Journal of Public Health, 105, 1101–1105. Kalmijn, M., Loeve, A., & Manting, D. (2007). Income dynam­ics in cou­ples and the dis­so­lu­tion of mar­ riage and cohab­i­ta­tion. Demography, 44, 159–179. Kolk, M., & Andersson, G. (2020). Two decades of same-sex mar­riage in Sweden: A demo­graphic account of devel­op­ments in mar­riage, child­bear­ing, and divorce. Demography, 57, 147–169. Lau, C. Q. (2012). The sta­bil­ity of same-sex cohab­i­ta­tion, dif­fer­ent-sex cohab­i­ta­tion, and mar­riage. Journal of Marriage and Family, 74, 973–988. Lesthaeghe, R. (2010). The unfolding story of the sec­ond demo­graphic tran­si­tion. Population and Development Review, 36, 211–251. Manning, W. D., Brown, S. L., & Stykes, J. B. (2016). Same-sex and dif­fer­ent-sex cohabiting cou­ple rela­ tion­ship sta­bil­ity. Demography, 53, 937–953. Manning, W. D., & Payne, K. K. (2021). Same-sex mar­ried and cohabiting cou­ples rais­ing chil­dren (Family Profile, No. 8). Bowling Green: OH: National Center for Family and Marriage Research. Retrieved from https:​­/​­/doi​­.org​­/10​­.25035​­/ncfmr​­/fp​­-21​­-08 McMunn, A., Lacey, R., Worts, D., McDonough, P., Stafford, M., Booker, C., . . . ​Sacker, A. (2015). De- stan­dard­i­za­tion and gen­der con­ver­gence in work–fam­ily life courses in Great Britain: A multi-chan­nel sequence anal­y­sis. Advances in Life Course Research, 26, 60–75. Meier, A., Hull, K. E., & Ortyl, T. A. (2009). Young adult rela­tion­ship val­ues at the inter­sec­tion of gen­der and sex­u­al­ity. Journal of Marriage and Family, 71, 510–525. Mills, M., & Blossfeld, H.-P. (2003). Globalization, uncer­tainty and changes in early life courses. Zeitschrift für Erziehungswissenschaft, 6, 188–218. Moore, M. R., & Stambolis-Ruhstorfer, M. (2013). LGBT sex­u­al­ity and fam­i­lies at the start of the twenty-first cen­tury. Annual Review of Sociology, 39, 491–507. Morgenroth, T., Kirby, T. A., Cuthbert, M. J., Evje, J., & Anderson, A. E. (2022). Bisexual era­sure: Per- ceived attrac­tion pat­terns of bisex­ual women and men. Euro­pean Journal of Social Psychology, 52, 249–259. 1604 A. Ophir et al. Musick, K., & Michelmore, K. (2018). Cross-national com­par­i­sons of union sta­bil­ity in cohabiting and mar­ried fam­i­lies with chil­dren. Demography, 55, 1389–1421. Nandi, A., Menon, S., & Smith, R. (2020). Understanding soci­ety: Marital and cohab­i­ta­tion his­to­ries, 1991–2018 user guide. Colchester, UK: Institute for Social and Economic Research, University of Essex. Park, N., Schmitz, R. M., & Slauson-Blevins, K. (2020). “It takes a lot of plan­ning”: Sexual minor­ity young adult per­cep­tions of gay and les­bian par­ent­hood. Journal of Family Issues, 41, 1785–1809. Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 Perales, F. (2016). The costs of being “dif­fer­ent”: Sexual iden­tity and sub­jec­tive wellbeing over the life course. Social Indicators Research, 127, 827–849. Perales, F., Simpson Reeves, L., Plage, S., & Baxter, J. (2020). The fam­ily lives of Aus­tra­lian les­bian, gay and bisex­ual peo­ple: A review of the lit­er­a­ture and a research agenda. Sexuality Research and Social Policy, 17, 43–60. Potârcă, G., Mills, M., & Neberich, W. (2015). Relationship pref­er­ences among gay and les­bian online dat­ ers: Individual and con­tex­tual influ­ences. Journal of Marriage and Family, 77, 523–541. Reczek, C. (2014). The inter­gen­er­a­tional rela­tion­ships of gay men and les­bian women. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 69, 909–919. Reczek, C. (2020). Sexual- and gen­der-minor­ity fam­i­lies: A 2010 to 2020 decade in review. Journal of Marriage and Family, 82, 300–325. Riskind, R. G., & Patterson, C. J. (2010). Parenting inten­tions and desires among child­less les­bian, gay, and het­ero­sex­ual indi­vid­u­als. Journal of Family Psychology, 24, 78–81. Riskind, R. G., & Tornello, S. L. (2017). Sexual ori­en­ta­tion and future par­ent­hood in a 2011–2013 nation­ ally rep­re­sen­ta­tive United States sam­ple. Journal of Family Psychology, 31, 792–798. Romeu, A., Monzó, A., López Villaverde, V., Duque, C. C., Lucas, V., Gilabert-Estellés, J., . . . ​Arnott, I. (2015). Orientación sex­ual y deseo reproductivo [Sexual ori­en­ta­tion and repro­duc­tive desire]. Revista Iberoamericana de Fertilidad, 32(2), 27–41. Rosenfeld, M. J. (2017). Moving a moun­tain: The extraor­di­nary tra­jec­tory of same-sex mar­riage approval in the United States. Socius, 3. https:​­/​­/doi​­.org​­/10​­.1177​­/2378023117727658 Rosenfeld, M. J., & Kim, B.-S. (2005). The inde­pen­dence of young adults and the rise of inter­ra­cial and same-sex Unions. Amer­i­can Sociological Review, 70, 541–562. Rosenfeld, M. J., & Thomas, R. J. (2012). Searching for a mate: The rise of the internet as a social inter­ me­di­ary. Amer­i­can Sociological Review, 77, 523–547. Rozental, A., & Malmquist, A. (2015). Vulnerability and accep­tance: Lesbian women’s fam­ily-mak­ing through assisted repro­duc­tion in Swed­ish pub­lic health care. Journal of GLBT Family Studies, 11, 127–150. Ruiz-Vallejo, F. R., & Boertien, D. (2021). Do same-sex unions dis­solve more often than dif­fer­ent-sex unions? Methodological insights from Colombian data on sex­ual behav­ior. Demographic Research, 44, 1149–1164. https:​­/​­/doi​­.org​­/10​­.4054​­/DemRes​­.2021​­.44​­.48 Russell, S. T., Bishop, M. D., Mallory, A. B., & Muraco, J. A. (2020). The use of rep­re­sen­ta­tive datasets to study LGBTQ-par­ent fam­i­lies: Challenges, advan­tages, and oppor­tu­ni­ties. In A. E. Goldberg & K. R. Allen (Eds.), LGBTQ-par­ent fam­i­lies: Innovations in research and impli­ca­tions for prac­tice (2nd ed., pp. 491–506). Cham, Switzerland: Springer Nature. https:​­/​­/doi​­.org​­/10​­.1007​­/978​­-3​­-030​­-35610​­-1_29 Scandurra, C., Bacchini, D., Esposito, C., Bochicchio, V., Valerio, P., & Amodeo, A. L. (2019). The influ­ ence of minor­ity stress, gen­der, and legal­i­za­tion of civil unions on par­ent­ing desire and inten­tion in les­bian women and gay men: Implications for social pol­icy and clin­i­cal prac­tice. Journal of GLBT Family Studies, 15, 76–100. Schacher, S. J., Auerbach, C. F., & Silverstein, L. B. (2005). Gay fathers expanding the pos­si­bil­i­ties for us all. Journal of GLBT Family Studies, 1, 31–52. Stambolis-Ruhstorfer, M., & Descoutures, V. (2020). Licence required: French les­bian par­ents con­front the obli­ga­tion to marry in order to estab­lish kin­ship. International Social Science Journal, 70, 79–97. Studer, M., & Ritschard, G. (2016). What mat­ters in dif­fer­ences between life tra­jec­to­ries: A com­par­a­tive review of sequence dis­sim­i­lar­ity mea­sures. Journal of the Royal Statistical Society: Series A (Statistics in Society), 179, 481–511. Tasker, F., & Bellamy, C. (2019). Adoption by same-sex cou­ples—Reaffirming evi­dence: Could more chil­dren be placed? Family Law Journal, 49(Feb­ru­ary). Available from https:​­/​­/www​­.familylaw​­.co​­.uk ​­/news_and_comment​­/publication​­-alert​­-family​­-law​­-february​­-2019 Family Formation Trajectories and LGB Cohort Change 1605 Tate, D. P., Patterson, C. J., & Levy, A. J. (2019). Predictors of par­ent­ing inten­tions among child­less les­ bian, gay, and het­ero­sex­ual adults. Journal of Family Psychology, 33, 194–202. Trandafir, M. (2015). Legal rec­og­ni­tion of same-sex cou­ples and fam­ily for­ma­tion. Demography, 52, 113–151. van de Kaa, D. J. (1987). Europe’s sec­ond demo­graphic tran­si­tion (Population Bulletin, Vol. 42 No. 1). Washington, DC: Population Reference Bureau. van Eeden-Moorefield, B., Martell, C. R., Williams, M., & Preston, M. (2011). Same-sex rela­tion­ships Downloaded from http://read.dukeupress.edu/demography/article-pdf/60/5/1581/2013403/1581ophir.pdf by UAB/SALA DE REVISTES user on 18 October 2023 and dis­so­lu­tion: The con­nec­tion between heteronormativity and homonormativity. Family Relations, 60, 562–571. Van Winkle, Z. (2018). Family tra­jec­to­ries across time and space: Increasing com­plex­ity in fam­ily life courses in Europe? Demography, 55, 135–164. Warnes, A. (2019). Exploring pronatalism and assisted repro­duc­tion in UK med­i­cine. Journal of International Women’s Studies, 20(4), 8. Retrieved from https:​­/​­/vc​­.bridgew​­.edu​­/jiws​­/vol20​­/iss4​­/8 Zaidi, B., & Mor­gan, S. P. (2017). The sec­ond demo­graphic tran­si­tion the­ory: A review and appraisal. Annual Review of Sociology, 43, 473–492. Ariane Ophir (cor­re­spond­ing author) aophir@ced​­.uab​­.es Ophir • Centre d’Estudis Demogràfics (CED-CERCA), Universitat Autònoma de Barcelona, Barcelona, Spain; https:​­/​­/orcid​­.org​­/0000​­-0002​­-0706​­-2322 Boertien • Centre d’Estudis Demogràfics (CED-CERCA), Universitat Autònoma de Barcelona, Barcelona, Spain; https:​­/​­/orcid​­.org​­/0000​­-0002​­-4105​­-8001 Vidal • Centre d’Estudis Demogràfics (CED-CERCA), Universitat Autònoma de Barcelona, Barcelona, Spain; https:​­/​­/orcid​­.org​­/0000​­-0003​­-4011​­-2077