Gender Composition of Children and the Third Birth in the United States

• Author: S. Philip Morgan,Felicia F. Tian
• DOI: http://doi.org/10.1111/jomf.12218
• Published date: 01 October 2015
• Date: 01 October 2015

F F. T Fudan University

S. P M University of North Carolina at Chapel Hill∗

Gender Composition of Children and the Third Birth

in the United States

Pollard and Morgan (2002) argued that the

parental mixed-gender preference (i.e., parents’

preference to have at least one son and one

daughter) will weaken in the United States

as aspects of gender become increasingly

deinstitutionalized. They presented evidence

that mixed-gender preference weakened in the

1986–1995 period compared to earlier and

coined this change as emerging gender indiffer-

ence. On the other hand, credible claims and

evidence suggest that after 1985, the “gender

revolution” has stalled. Such argumentssuggest

that weakened mixed-gender preference will

persist. In this study, the authors replicated and

extended Pollard and Morgan using 4 waves of

data from the National Survey of Family Growth

that allow examination of the 1966–2009

period. The results show that the effect of

same-gender children on intending/having

another child declined in the 1986–1995

period; however, no evidence that this effect

has continued to weaken was found. Thus, these

data show evidence consistent with a “stalled

revolution.”

Department of Sociology, Fudan University, 220 Handan

Rd., Shanghai, 200433 China (ftian@fudan.edu.cn).

∗Department of Sociology and Carolina Population Center,

University of North Carolina at Chapel Hill, 206 W.

Franklin St., USQ East Tower,Oice #221, Campus Box

8120, Chapel Hill, NC 27516.

Key Words: children, fertility,gender, gender roles, inequal-

ity, U.S. population.

Parental gender preferences are structural and

cultural forces that reify gender difference

(Lorber, 1995). In many societies, the dominant

preference is for sons. This son preference

reects institutionalized patterns of both inher-

ent difference and hierarchy (e.g., patriarchy).

The strength and change in sex ratios and excess

fertility due to patriarchy are an active research

area in demography (see Bongaarts, 2013). In

many Western societies, parents have a mixed

gender preference: a desire for at least one son

and one daughter. This preference is signaled

when those with two sons or two daughters are

more likely to intend and to have a third birth

compared to parents of one daughter and one son

(Hank & Kohler, 2000; Yamaguchi & Ferguson,

1995). This preference indicates that sons and

daughters are fundamentally different rather

than substitutable. As Williamson (1976) put it,

parents “desire variety, based on the notion that

the sexes will have different traits, strengths,

leisure activities, and interests” (p. 22). Thus,

this common gender preference reects the

institutionalized notion of inherent difference

(gender essentialism; see Ridgeway, 2009), but

not of inherent hierarchy.

Our focus, the mixed-gender preference,

has been observed for decades in the United

States (see Sloane & Lee, 1983). Pollard

and Morgan (2002) presented evidence that

this mixed-gender preferences weakened in

the 1985–1995 period compared to earlier

(1955–1985) and claimed that a gender indif-

ference toward offspring was emerging. They

predicted a continuation of this trend toward

Journal of Marriage and Family 77 (October 2015): 1157–1165 1157

DOI:10.1111/jomf.12218