Abstract
The increasing availability of population-level data on same-sex couples has given family demographers greater insight into the characteristics of same-sex couples in the United States as well as the implications of gender and sexuality for individual well-being across multiple outcomes. Whereas most nationally representative research on same-sex couples focuses on individuals, surveys such as the National Health Interview Survey (NHIS) can be used to construct couple-level data files, enabling researchers to consider how couple-level indicators and variables for both partners affect individual well-being. In this chapter, we detail the benefits of dyadic data methods for research on couples and demonstrate a dyadic data analysis of 1262 same-sex couples and 113,642 different-sex couples derived from the 2012 to 2016 NHIS person-level files. Specifically, we examine the effects of respondent gender × partner gender on: reported health status; partner education and reported health; and union status and reported health among those in same-sex versus different-sex couples. Our results suggest that analysis of same-sex couples using dyadic data methods yields more nuanced results than analyses of individuals. For instance, there is a significant positive association between partner college degree and reported health status that is greater in magnitude for women than men—but only if women are in different-sex relationships. In addition, cohabitation and marriage are similarly associated with health for men in same-sex relationships, but cohabitation is associated with poorer health for men and women in different-sex relationships. We conclude with suggestions for future research using dyadic data methods for research on couples.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
This also implies that standard panel-data regression approaches may be adapted for this purpose.
- 2.
The results for the ordered probit models are available upon request.
- 3.
Stata and R code for carrying out this analysis is available in the Appendix.
References
Allison, P. D. (2009). Fixed effects regression models (Vol. 160). Thousand Oaks: SAGE Publications.
Bell, A., & Jones, K. (2015). Explaining fixed effects: Random effects modeling of time-series cross-sectional and panel data. Political Science Research and Methods, 3(1), 133–153.
Bertakis, K. D., Azari, R., Helms, L. J., Callahan, E. J., & Robbins, J. A. (2000). Gender differences in the utilization of health care services. The Journal of Family Practice, 49(2), 147–152.
Bratter, J. L., & Eschbach, K. (2006). ‘What about the couple?’ Interracial marriage and psychological distress. Social Science Research, 35(4), 1025–1047. https://doi.org/10.1016/j.ssresearch.2005.09.001.
Brown, D. C., Hummer, R. A., & Hayward, M. D. (2014). The importance of spousal education for the self-rated health of married adults in the United States. Population Research and Policy Review, 33, 127–115. https://doi.org/10.1007/s11113-013-9305-6.
Case, A., & Paxson, C. (2005). Sex differences in morbidity and mortality. Demography, 42, 189–214. https://doi.org/10.1353/dem.2005.0011.
Cook, W. L., & Kenny, D. A. (2005). The actor-partner interdependence model: A model of bidirectional effects in developmental studies. International Journal of Behavioral Development, 29, 101–109. https://doi.org/10.1080/01650250444000405.
Donnelly, R., Umberson, D. J., & Kroeger, R. A. (2017). Life course stressors and marital strain in same-sex and different-sex marriages. Journal of Family Issues. https://doi.org/10.1177/2F0192513X17741177.
Eirich, G. M., & Robinson, J. H. (2017). Does earning more than your spouse increase your financial satisfaction? A comparison of men and women in the United States, 1982 to 2012. Journal of Family Issues, (17), 2371–2399. https://doi.org/10.1177/0192513X16638384.
Goldberg, A. E., Smith, J. Z., & Kashy, D. A. (2010). Preadoptive factors predicting lesbian, gay, and heterosexual couples’ relationship quality across the transition to adoptive parenthood. Journal of Family Psychology, 24, 221–232.
Kashy, D. A., Donnellan, M. B., Burt, S. A., & McGue, M. (2008). Growth curve models for indistinguishable dyads using multilevel modeling and structural equation modeling: The case of adolescent twins’ conflict with their mothers. Developmental Psychology, 44(2), 316–329.
Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data analysis. New York: Guilford Press.
Kroeger, R. A., & Williams, K. (2011). Consequences of black exceptionalism? Interracial unions with blacks, depressive symptoms, and relationship satisfaction. The Sociological Quarterly, 52, 400–420. https://doi.org/10.1111/j.1533-8525.2011.01212.x.
Liu, H., Reczek, C., Mindes, S. C. H., & Shen, S. (2016). The health disparities of same-sex cohabitors at the intersection of race-ethnicity and gender. Sociol Perspect, 60(3), 620–639.
Mejia, R., Braun, S., Peña, L., Gregorich, S., & Pérez-Stable, E. (2017). Validation of non-smoking status by spouse following a cessation intervention. Journal of Smoking Cessation, 12(1), 38–42. https://doi.org/10.1017/jsc.2015.11.
Quintana, H. K., Vikström, M., Andersson, T., Hallqvist, J., & Leander, K. (2015). Agreement between myocardial infarction patients and their spouses on reporting of data on 82 cardiovascular risk exposures. PLoS ONE, 10(7), e0132601. https://doi.org/10.1371/journal.pone.0132601.
Reczek, C., Spiker, R., Liu, H., & Crosnoe, R. (2016). Family structure and child health: Does the sex composition of parents matter? Demography, 53(5), 1605–1630.
Ross, C. E., Masters, R. K., & Hummer, R. A. (2012). Education and the gender gaps in health and mortality. Demography, 49, 1157–1183. https://doi.org/10.1007/s13524-012-0130-z.
Schoen, R., Astone, N. M., Kim, Y. J., Rothert, K., & Standish, N. J. (2002). Women’s employment, marital happiness, and divorce. Social Forces, 81(2), 643–662. https://doi.org/10.1353/sof.2003.0019.
Simon Pickard, A., & Knight, S. J. (2005). Proxy evaluation of health-related quality of life: A conceptual framework for understanding multiple proxy perspectives. Medical Care, 43(5), 493–499.
Snijders, T., & Bosker, R. (2012). Multilevel analysis: An introduction to basic and advanced multilevel modelling (2nd ed.). London: Sage.
Umberson, D. (1992). Gender, marital status and the social control of health behavior. Social Science & Medicine, 8, 907–917.
Umberson, D., & Kroeger, R. A. (2016). Gender, marriage, and health for same-sex and different-sex couples: The future keeps arriving. In A. Booth, V. King, S. McHale, & J. Van Hook (Eds.), Gender and couple relationships. New York: Springer.
Umberson, D., Thomeer, M. B., Kroeger, R. A., Lodge, A., & Xu, M. (2015). Challenges and opportunities for research on same-sex relationships. Journal of Marriage and Family, 77(1), 96–111.
Umberson, D., Thomeer, M. B., Reczek, C., & Donnelly, R. (2016). Physical illness in gay, lesbian, and heterosexual marriages: Gendered dyadic experiences. Journal of Health and Social Behavior, 57(4), 517–553.
Umberson, D., Thomeer, M. B., Kroeger, R. A., Reczek, C., & Donnelly, R. (2017). Instrumental- and emotion-focused care work during physical health events: Comparing gay, lesbian, and heterosexual marriages. The Journal of Gerontology Series B: Social Sciences, 72(3), 498–509.
West, C., & Zimmerman, D. H. (2009). Accounting for doing gender. Gender and Society, 23, 112–122.
West, T. V., Popp, D., & Kenny, D. A. (2008). A guide for the estimation of gender and sexual orientation effects in dyadic data: An actor-partner interdependence model approach. Personality and Social Psychology Bulletin, 34, 321–336. https://doi.org/10.1177/0146167207311199.
Zajacova, A., Huzurbazar, S., & Todd, M. (2017). Gender and the structure of self-rated health across the adult life span. Social Science & Medicine, 187, 58–66.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix 1: Syntax for Regression Analyses
Appendix 1: Syntax for Regression Analyses
1.1 Stata Syntax
*Empty Model for ICC mixed health || cid:, cov(exc) reml estat icc ****************************************************************** *Table 3 sort cid pid *M0 empty mixed health || cid:, nocon res(exc) reml
*M1 No Controls- No Interaction mixed health i.fem i.pfem || cid:, nocon res(exc) reml outreg2 using "table3", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) replace
*M2 No Controls- Add Interaction mixed health i.fem##i.pfem || cid:, nocon res(exc) reml outreg2 using "table3", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append
*WM lincom _cons+1.fem *MW lincom _cons+1.pfem *WW lincom _cons+1.fem+1.pfem+1.fem#1.pfem
*M3 Add dyad level control vars mixed health i.fem##i.pfem i.cohab i.numkids || cid:, nocon res(exc) reml outreg2 using "table3", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append
*M4 Add Respondent Controls mixed health i.fem##i.pfem i.cohab i.numkids i.coldeg i.race c.cage_p || cid:, nocon res(exc) reml outreg2 using "table3", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append
*M5 Add Partner Controls mixed health i.fem##i.pfem i.cohab i.numkids i.coldeg i.race c.cage_p /// i.coldeg_pid i.race_pid c.cage_p_pid || cid:, nocon res(exc) reml outreg2 using "table3", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append ***************************************************************** *Table 4 *M1- no interactions: mixed health i.fem i.pfem i.coldeg_pid i.coldeg i.race i.race_pid /// c.cage_p c.cage_p_pid i.cohab i.numkids || cid:, nocon res(exc)reml outreg2 using "table4", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) replace *M2- fem##coldeg mixed health i.fem i.pfem i.coldeg_pid i.fem##i.coldeg_pid i.coldeg i.race i.race_pid /// c.cage_p c.cage_p_pid i.cohab i.numkids || cid:, nocon res(exc) reml outreg2 using "table4", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append *M3- pfem##coldeg mixed health i.fem i.pfem i.coldeg_pid i.pfem##i.coldeg_pid i.coldeg i.race i.race_pid /// c.cage_p c.cage_p_pid i.cohab i.numkids || cid:, nocon res(exc) reml outreg2 using "table4", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append *M4- fem##pfem##coldeg mixed health i.fem##i.pfem##i.coldeg_pid i.coldeg i.race i.race_pid /// c.cage_p c.cage_p_pid i.cohab i.numkids || cid:, nocon res(exc) reml outreg2 using "table4", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append *effect for MM is 1.coldeg_pid *effect for WM lincom 1.coldeg_pid+1.fem#1.coldeg_pid *effect for MW lincom 1.coldeg_pid+1.pfem#1.coldeg_pid *effect for WW lincom 1.coldeg_pid+1.fem#1.coldeg_pid+1.pfem#1. coldeg_pid+1.fem#1.pfem#1.coldeg_pid *WW vs MM lincom (1.coldeg_pid)-(1.coldeg_pid+1.fem#1.coldeg_pid+1.pfem#1. coldeg_pid+1.fem#1.pfem#1.coldeg_pid) *WW vs MW lincom (1.coldeg_pid+1.pfem#1.coldeg_pid)-(1.coldeg_pid+1.fem#1. coldeg_pid+1.pfem#1.coldeg_pid+1.fem#1.pfem#1.coldeg_pid) *WW vs WM lincom (1.coldeg_pid+1.fem#1.coldeg_pid+1.pfem#1.coldeg_pid+1. fem#1.pfem#1.coldeg_pid)-(1.coldeg_pid+1.fem#1.coldeg_pid) *MW vs WM lincom (1.coldeg_pid+1.pfem#1.coldeg_pid)-(1.coldeg_pid+1. fem#1.coldeg_pid) ***************************************************************** *Table 5 *M1- no interactions: mixed health i.fem i.pfem i.cohab /// i.coldeg i.coldeg_pid i.race i.race_pid c.cage_p c.cage_p_pid i. numkids /// || cid:, nocon res(exc) reml outreg2 using "table5", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) replace *M2- i.fem##i.cohab mixed health i.fem i.pfem i.cohab i.fem#i.cohab /// i.coldeg i.coldeg_pid i.race i.race_pid c.cage_p c.cage_p_pid i. numkids /// || cid:, nocon res(exc) reml outreg2 using "table5", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append *M3- i.pfem##i.cohab mixed health i.fem i.pfem i.cohab i.pfem#i.cohab /// i.coldeg i.coldeg_pid i.race i.race_pid c.cage_p c.cage_p_pid i. numkids /// || cid:, nocon res(exc) reml outreg2 using "table5", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append *M4- i.fem##i.pfem##i.cohab mixed health i.fem##i.pfem##i.cohab /// i.coldeg i.coldeg_pid i.race i.race_pid c.cage_p c.cage_p_pid i.numkids /// || cid:, nocon res(exc) reml outreg2 using "table5", excel symbol(***, **, *, +) alpha(0.001, 0.01, 0.05, 0.10) dec(2) append *effect for MM is 1.cohab *effect for WM lincom 1.cohab+1.fem#1.cohab *effect for MW lincom 1.cohab+1.pfem#1.cohab *effect for WW lincom 1.cohab+1.fem#1.cohab+1.pfem#1.cohab+1.fem#1.pfem#1.cohab *WW vs MM lincom (1.cohab)-(1.cohab+1.fem#1.cohab+1.pfem#1.cohab+ 1.fem#1.pfem#1.cohab) *WW vs MW lincom (1.cohab+1.pfem#1.cohab)-(1.cohab+1.fem#1.cohab+1.pfem#1. cohab+1.fem#1.pfem#1.cohab) *WW vs WM lincom (1.cohab+1.fem#1.cohab+1.pfem#1.cohab+1.fem#1.pfem#1. cohab)-(1.cohab+1.fem#1.cohab) *MW vs WM lincom (1.cohab+1.pfem#1.cohab)-(1.cohab+1.fem#1.cohab)
1.2 R Syntax
# TABLE 3 # M1.a <- lmer(health ~ 1 + (1 | cid), REML=TRUE, dat) summary(M1.a) # takes a while on this M1.b <- gls(health ~ 1, correlation = corCompSymm(form = ~1 | cid), dat) summary(M1.b) # faster M1.c <- geeglm(health ~ 1, id=cid, corstr="exchangeable", family='gaussian', data=dat ) summary(M1.c) M1 <- gls(health ~ fem + pfem, correlation = corCompSymm(form = ~1 | cid), dat) summary(M1) M2 <- gls(health ~ fem*pfem, correlation = corCompSymm(form = ~1 | cid), dat) summary(M2) M3 <- gls(health ~ fem*pfem + cohab + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M3) M4 <- gls(health ~ fem*pfem + cohab + numkids + race + cage_p, correlation = corCompSymm(form = ~1 | cid), dat) summary(M4) M5 <- gls(health ~ fem*pfem + cohab + numkids + race + cage_p + coldeg_pid + race_pid + cage_p_pid, correlation = corCompSymm(form = ~1 | cid), dat) summary(M5) # # TABLE 4 # #M3.1- no interactions: M3.1 <- gls(health ~ fem + pfem + coldeg + coldeg_pid + race + race_pid + cage_p + cage_p_pid + cohab + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M3.1) #M3.2- femXcoldeg M3.2<- gls(health ~ fem + pfem + coldeg + coldeg_pid + fem*coldeg_pid + ace + race_pid + cage_p + rcage_p_pid + cohab + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M3.2) #M3- pfemXcoldeg M3.3<- gls(health ~ fem + pfem + coldeg + coldeg_pid + pfem*coldeg_pid + race + race_pid + cage_p + age_p_pid + cohab + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M3.3) #M4- femXpfemXcoldeg M3.4<- gls(health ~ pfem + coldeg + coldeg_pid + fem*pfem*coldeg_pid + race + race_pid + cage_p + cage_p_pid + cohab + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M3.4) # # TABLE 5 # #M1- All Controls- no interactions: M4.1 <- gls(health ~ fem + pfem + cohab + coldeg + coldeg_pid + race + race_pid + cage_p + cage_p_pid + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M4.1) #M2- femXcohab M4.2<- gls(health ~ fem + pfem + cohab + fem*cohab + coldeg + coldeg_pid + race + race_pid + cage_p + cage_p_pid + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M4.2) #M3- pfemXi.cohab M4.3<- gls(health ~ fem + pfem + cohab + pfem*cohab + coldeg + coldeg_pid + race + race_pid + cage_p + cage_p_pid + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M4.3) #M4- femXpfemXcohab M4.4<- gls(health ~ fem*pfem*cohab + coldeg + coldeg_pid + race + race_pid + cage_p + cage_p_pid + numkids, correlation = corCompSymm(form = ~1 | cid), dat) summary(M4.4) # equiv M4.4a <- geeglm(health ~ fem*pfem*cohab + coldeg + coldeg_pid + race + race_pid + cage_p + cage_p_pid + numkids, id=cid, corstr="exchangeable", family='gaussian', data=dat ) summary(M4.5a)
Rights and permissions
Copyright information
© 2019 Springer International Publishing AG, part of Springer Nature
About this chapter
Cite this chapter
Kroeger, R.A., Powers, D.A. (2019). Examining Same-Sex Couples Using Dyadic Data Methods. In: Schoen, R. (eds) Analytical Family Demography. The Springer Series on Demographic Methods and Population Analysis, vol 47. Springer, Cham. https://doi.org/10.1007/978-3-319-93227-9_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-93227-9_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-93226-2
Online ISBN: 978-3-319-93227-9
eBook Packages: Social SciencesSocial Sciences (R0)