Alone in the Ivory Tower

374939 39Wolfinger et al.Journal of Family Issues © The Author(s) 2010 Reprints and permission: http://www. sagepub.com/journalsPermissions.nav JFI311210.1177/0192513X103749 Alone in the Ivory Tower Journal of Family Issues 31(12) 1652­–1670 © The Author(s) 2010 Reprints and permission: http://www. sagepub.com/journalsPermissions.nav DOI: 10.1177/0192513X10374939 http://jfi.sagepub.com Nicholas H. Wolfinger1, Marc Goulden2, and Mary Ann Mason2 Abstract The authors use data from the 2000 Census Public Use Microdata Sample to examine the likelihood of a birth event, defined as the household presence of a child younger than 2 years, for male and female professionals. Physicians have the highest rate of birth events, followed in order by attorneys and academics.Within each profession men have more birth events than women. For men, occupational variation in birth events can be explained by marital status, income, and spousal employment. These factors only partially account for occupational differences in birth events for women. Keywords fertility, fast-track professionals, higher education This article examines fertility in households containing male or female professionals using data from the 2000 Census Public Use Microdata Sample (PUMS). We consider the probability of birth events for individuals employed in three professions: doctors, lawyers, and professors. We attempt to explain variation in birth events on the basis of social and demographic differences, including age, working hours, race/ethnicity, marital status, income, and spousal characteristics. 1 University of Utah, Salt Lake City, UT, USA University of California, Berkeley, CA, USA 2 Corresponding Author: Nicholas H. Wolfinger, Department of Family and Consumer Studies, University of Utah, 225 South 1400 East, AEB 228, Salt Lake City, UT 84112-0080, USA Email: [email protected] Wolfinger et al. 1653 Background It has long been known that education and fertility are inversely related in the United States (Rindfuss & Sweet, 1977). Yang and Morgan (2004) recently showed that American women with a high school education have a total fertility rate about 0.5 higher than their counterparts who spent time in college. Over time educated women are also waiting longer to have their first child (Rindfuss, Morgan, & Offutt, 1996). These trends are well documented. How does fertility vary among professionals? Although there have been studies of gender differences in births among academics (Long, 2001; Mason & Goulden, 2004; Perna, 2001a) and doctors (Boulis, 2004), we know little about whether men and women in different professions have children at different rates. To the best of our knowledge, only two studies have considered this issue. Using data from the 1980 Census, Cooney and Uhlenberg (1989) show that female physicians are more likely to have children than female lawyers or professors. However, female physicians are less likely to have children than their male colleagues, or women in general (Boulis, 2004). Similarly, female tenure-track or tenured professors have fewer children than their male colleagues (Mason & Goulden, 2004). We know almost nothing about how male fertility varies by profession. Although few studies have considered how career choice affects fertility, many have examined the effects of childbirth on professional success. Several studies find that children incrementally decrease women’s wages (Avellar & Smock, 2003; Budig & England, 2001; Waldfogel, 1997). Others show that time out of the labor force lowers women’s incomes (Hewlett & Luce, 2005; Noonan, 2005; Noonan & Corcoran, 2004). Aspiring female academics with young children are less likely to get tenure-track positions than either childless women or men (Wolfinger, Mason, & Goulden, 2008). Implicit in these studies are arguments about causal order. Although all the studies used longitudinal data, the general presumption is that children are causing the professional outcomes in question. This raises a more general point regarding endogeneity in the relationship between family formation and professional success. Economists have long posited a relationship between fertility and women’s labor force participation, although there has been little consensus about the direction of causality (for an overview, see Macunovich, 1996). Recent evidence suggests it flows both ways: Budig (2003) found that young children increase women’s labor force departures whereas older children decrease them. Women with preschool-age children are less likely to work in the first place, but employed women are less likely to become pregnant. Many female academics report making joint decisions about career and childbirth (Van Anders, 2004). 1654 Journal of Family Issues 31(12) With these findings in mind, we should speak of correlation, not causation, when discussing the relationship between career and children. Differences Between Professions Differences in the professional development of doctors, lawyers, and professors lead us to hypothesize that doctors will have the most children and professors the fewest. We anticipate that these occupational differences in fertility will hold for both sexes but should be stronger for women than for men. The unique career structure of academia offers no good opportunity to take time out for children. After 4 to 8 years in graduate school, assistant professors have about 6 years to publish or perish. Only after tenure and promotion from assistant to associate professor are faculty assured of job security. The median doctorate recipient is already 33 or 34 years old (Hoffer et al., 2006; Jacobs & Winslow, 2004); after a probationary assistant professorship, close to 40 years. In terms of career development this would be an ideal time for professors to start their families, but biologically female faculty are already past prime childbearing ages. For instance, women older than 35 have almost quadruple the likelihood of having Down’s syndrome babies in comparison with women in their 20s (California Birth Defects Monitoring Program, 2005). Graduate school may not be an optimal time to have children, both because of the work load and because of the probable lack of income. Another impediment to academic fertility is the paucity of part-time tenure-track positions (Leslie & Walke, 2001). Academics who want to work less than full-time generally must resort to the reduced pay and status of adjunct professorships. Both medicine and law presumably offer more opportunities for part-time employment. Male academics face similar barriers to fertility. Like their female colleagues, they may feel compelled to put graduate school behind them before having children. Prior to obtaining tenure-track jobs men may feel their lives are too unsettled to become fathers. Men are also more involved with parenting than in years gone by (Bianchi, Robinson, & Milkie, 2006), another reason they may be inclined to wait until their professional lives settle down before having children. Young lawyers face similar challenges, albeit not as severe as those confronting academics. The average entering law student is 25 years old (Masters, 2004). After 3 years of law school, attorneys who join firms can be expected to spend about 8 years before they make partner (AllBusiness.com, 2004). This means lawyers must also wait until their mid to late 30s in order to reach a career stage conducive to child rearing. However, not all lawyers aspire to high-powered Wolfinger et al. 1655 corporate careers. Note also that failure to make partner is not as catastrophic as failing to get tenure: one can simply move to another firm. Professors may have to relocate if they do not get tenure but wish to remain in academia. The average medical school student matriculates at 24 years (Association of American Medical Colleges, 2007). Medical school is followed by a residency of 3 to 7 years (with additional training for some specialties); the majority of doctors are internists and, therefore, have shorter residencies. The arduous nature of residencies is well known, so this does not seem to be a likely time for female physicians to have children. However, given a shorter residency, doctors will have completed professional probation in their early 30s, before the age 35 cutoff for biologically risky pregnancies. At this point female doctors should feel ready to start families if they have not already done so. Based on career structure, then, we would expect the highest fertility for physicians, the second highest for attorneys, and the lowest for academics. Within each profession, men should have more children than women. Another factor affecting fertility is the ability to pay for children, and in particular, child care. Doctors, lawyers, and professors have dramatically different salaries: about $120,000 for a beginning physician, $60,000 for a fledging lawyer, and $51,000 for a starting assistant professor (www.payscale.com). Although often saddled with heavy student loan debt (Jolly, 2007), physicians presumably have a greater ability to pay for child care than do attorneys or professors. Income may also facilitate other time-saving services such as housekeeping. This may increase doctors’ willingness to have children. Goals of Study We update Cooney and Uhlenberg’s (1989) study on occupational differences in fertility in several respects. First, we use comparatively recent data from the 2000 Census. Second, we contrast the incidence of birth events for male and female professionals. Third, we use multivariate analysis to explain differences in fertility by profession. There are several reasons why we choose doctors, lawyers, and professors as our bases of comparison. First, Cooney and Uhlenberg (1989) used these categories. Second, as professionals they are easily identified using Census data (see below). Third, all three represent traditionally male enclaves that have witnessed dramatic female penetration in recent years. Yet all three share to varying degrees a career model that is not conducive to female fertility (or to men who assume a larger role in the traditionally female domain of child rearing). Our analysis will determine which professions have best facilitated male fertility and female fertility. 1656 Journal of Family Issues 31(12) Method We analyze data from the 2000 5% Census PUMS (U.S. Census Bureau, 2003). Although the PUMS offers relatively little information on participants, it provides a sufficient sample of doctors, lawyers, and professors. Analysis is limited to individuals aged 25 years (the approximate lowest age at which people could have finished their professional training) to 44 years (few new parents are older). This provides a 20-year window for observing birth events. Summary statistics and sample sizes by sex and occupation are shown in Table 1. There are no missing data. Doctors, lawyers, and professors are identified using a combination of occupational codes and education. For all three occupational groups, individuals are required to be working 1 or more hours a week and have one of the three relevant job titles: (a) postsecondary teacher, hereafter referred to as professor; (b) physician (including surgeons); or (c) lawyer. Professors must have PhDs; doctors and lawyers are required to have professional degrees (accordingly, professors of law and medicine are, respectively, treated as lawyers and doctors). We refer to these individuals as “focal persons.” The dependent variable in all analyses is a birth event, defined as the presence of a child age 0 or 1 in the household. We view the child’s exact relationship to the focal person—biological, step, or adopted—as irrelevant. Although most birth events reflect biological children (and for convenience we will speak of them accordingly), any infant in the household reflects a conscious decision on the part of the focal person. Birth events may reflect initial or higher order births. Data on fertility history and intentions would be helpful but are not available in the Census. Independent variables are used to explain the relationship between an occupation and a birth event. For focal persons, independent variables include age, race/ ethnicity, hours worked, urbanicity, union status (including marital status and cohabitation), household presence of someone 65 and older, and individual income. Age is a continuous variable; its square is included to account for curvilinearity in its relationship to the likelihood of a birth event (see Figure 1). As we observed in the Background section, people become doctors, lawyers, or professors at different ages. Race/ethnicity is a set of dummy variables measuring whether a focal person is White, African American, Asian American, Latino, or a member of any other population group; White is the reference category. There are profound racial and ethnic differences in the professions (U.S. Census Bureau, 2009) and fertility (Dye, 2008). Work hours are dummy coded into six categories: 1-19 (the reference category), 20-29, 30-39, 40-49, 50-59, and 60 or more. Urbanicity is a continuous variable measuring the percentage of an 1657 Wolfinger et al. Table 1. Means or Percentages by Sex and Profession Men Women Professors Doctors Lawyers Professors Doctors Lawyers Birth event (%) 16 19 18 Race/ethnicity White (%) 75 72 88 African 4 5 3 American (%) Asian 16 17 3 American (%) Hispanic (%) 5 7 5 Other (%) 0.3 0.3 0.2 Age (years) 38 36 36 Weekly hours worked 47 58 50 Percentage urban 83 87 91 Union status Never married (%) 17 15 15 5 4 5 Divorced/ separated/ widowed (%) Married (%) 74 77 71 Cohabiting (%) 4 4 5 Household resident 2 3 2 65 and older (%) Personal income 48,500 100,000 75,000 (median, $) Household income 69,400 130,001 107,000 (median, $) Spousal/partner employment status Professor (%) 12 0.5 0.4 Physician (%) 2 16 1 Lawyer (%) 1 2 15 Other 60 42 60 employment (%) Out of the labor 22 38 23 force (%) Not in house (%) 3 2 1 N 3,918 11,683 13,914 12 17 15 78 5 65 8 82 8 11 19 5 5 0.5 37 43 84 7 0.5 35 52 90 5 0.5 35 44 92 23 10 20 5 26 7 61 7 3 70 5 4 59 7 3 41,200 57,000 53,000 74,810 121,530 106,410 28 3 3 60 2 36 3 50 1 3 32 61 3 5 2 4 4 1 2,547 8,536 5,952 Note. Numbers are weighted. Percentages may not sum to 100 because of rounding error. 1658 Journal of Family Issues 31(12) 30% Chance of birth event 25% Male professors Male doctors 20% Male lawyers Female professors 15% Female doctors Female lawyers 10% All employed men 5% 0% All employed women 25-29 30-34 35-39 40-44 Age Figure 1. Birth events by sex, age, and profession Note. Differences by sex, age, and profession are statistically significant (p < .001). individual’s SMSA that is urban; female professors—and perhaps also doctors and lawyers—are more likely than their male colleagues to reside in large cities and other areas with multiple colleges and universities (Kulis & Sicotte, 2002). Union status is strongly correlated with fertility (Dye, 2008) and includes the following categories: never married (the reference category), previously married (including divorced, separated, and widowed individuals), married, and cohabiting (the reference category for analyses including only married and cohabiting individuals). Household presence of someone 65 and older, a potential babysitter, is a dichotomous variable. Individual income, related to profession and fertility (Dye, 2008), is measured in dollars and logged to account for right skew. We also analyze characteristics of the focal person’s spouse or cohabiting partner. Employment is a six-category variable measuring whether a spouse or partner is a physician, an academic, an attorney, is employed in another profession, is not in the labor force, or is not in the household; other employment is the reference category. Finally, select analyses include the natural logarithm of household income. No additional information is available for spouses not residing with focal persons. We begin by examining birth events by sex, profession, and age. Next we attempt to explain these differences via multivariate analysis. Birth event is a dichotomous variable, so we use logistic regression. Analyses are weighted. Separate regressions are conducted for male and female focal persons, as Wolfinger et al. 1659 preliminary analysis with a pooled sample revealed statistically significant gender differences. All regression models include occupation, age, race/ethnicity, and weekly hours worked as independent variables. For successive models we introduce additional variables to account for the differences in birth events between the three occupational groups. With the introduction of the spousal employment variable, we only include individuals who are married or cohabiting. Results Figure 1 shows how rates of birth events vary by occupation, sex, and age. Male physicians and lawyers are the most likely to have babies in the household whereas female professors are the least likely. Female physicians, male professors, and female lawyers are in the middle. Although female physicians and lawyers have more birth events than male professors from ages 30 to 39, male professors from ages 25 to 29 and from ages 40 to 44 have more birth events than women. Both male and female professors have fewer babies than do members of other professions. In addition, all groups except female professors have the most birth events in their early 30s. For female professors the peak years are the late 30s. Note also that these patterns of birth events diverge substantially from those of employed Americans in general, whose birth timing is depicted by the heavy lines. Compared with fast-track professionals, the average American has more birth events when young and fewer when older. Fertility Among Professional Men Table 2 shows regression results for the likelihood of a birth event for male professionals. Model 1 of Table 2 is consistent with Figure 1. Among men, professors are the least likely to have babies in the household. After controlling for race/ethnicity, age, and weekly hours worked, male professors are 21% [100 * (1 − exp(−0.24))] less likely than male physicians to report a birth event. Male lawyers are a little less likely than male doctors to have a baby in the household, with a 10% difference in odds [100 * (1 − exp(−0.10))]. The statistically significant coefficients for age and age squared confirm the quadratic pattern of birth timing shown in Figure 1. Note also that men with recent birth events work more, whereas Whites have more birth events than do members of other population groups. Model 2 adds marital status to the analysis. Male professors are still less likely than male physicians to have birth events, but the difference in odds is reduced from 21% to 15% [100 * (1 − exp(−0.15))]. Furthermore, marital status completely accounts for the baby gap between male attorneys and 1660 Journal of Family Issues 31(12) Table 2. Logistic Regressions of Birth Events for Male Professionals All Professionals Model 1 Model 2 Married/Cohabiting Model 3 Model 4 Model 5 Employment status Physician — — — — — Lawyer −0.10** −0.04 −0.02 −0.02 0.04 −0.03 Professor −0.24*** −0.15** −0.12* −0.11+ Race/ethnicity White — — — — — African American −0.21* −0.01 0.01 −0.02 0.05 Asian American −0.13* −0.18** −0.17** −0.17** −0.16** Hispanic −0.22** −0.17* −0.16* −0.16* −0.17* Other −1.03* −0.74 −0.72 −0.71 −0.62 Age 1.60*** 1.34*** 1.32*** 1.33*** 1.29*** Age squared −0.02*** −0.02*** −0.02*** −0.02*** −0.02*** Weekly hours 1-19 — — — — — 20-29 0.52* 0.60* 0.59* 0.60* 0.59* 30-39 0.41+ 0.46* 0.41+ 0.39+ 0.39+ 40-49 0.59** 0.52** 0.45* 0.44* 0.44* 50-59 0.67*** 0.49* 0.41* 0.40* 0.38+ 60 or more 0.62** 0.45* 0.37+ 0.36+ 0.35+ Percentage urban −0.20* 0.19* 0.20* 0.20* 0.22* Union status Never married — — — — — Divorced/separated/ — 1.68** 1.68** — — widowed Married — 5.69*** 5.70*** — — Cohabiting — 3.05*** 3.07*** −2.63*** −2.63*** Household resident 65 and — −0.08 −0.04 −0.05 −0.07 older Log of personal income — — 0.11*** 0.11*** 0.14*** Log of household income — — −0.03* −0.08** −0.16*** Spousal/partner employment status Professor — — — — −0.04 Physician — — — — 0.34*** Lawyer — — — — 0.18*** Other employment — — — — — Out of the labor force — — — — 0.37*** Not in house — — — — −2.32*** Constant −28.93*** −29.16*** −29.04*** −23.63*** −22.25*** Log-likelihood 12785.94 11197.78 11189.56 11023.97 10868.26 Note. Analyses are weighted. N is 29,515 for Models 1 to 3 and 22,281 for Models 4 and 5. + p < .10. *p < .05. **p < .01. ***p < .001, two-tailed. Wolfinger et al. 1661 physicians. This result can probably be explained by the fact that 77% of male doctors are married, compared with 71% of male attorneys (see Table 1). The large positive coefficients in Model 2 show that being married, cohabiting, or even having been married in the past is strongly associated with the household presence of babies. Furthermore, controlling for marital status eliminates the disparity in birth events between Whites and African Americans but does not substantially affect the lower rate at which Asian American and Hispanic men have children. The effect of marital status on African American fertility is understandable given the disproportionately low marriage rates and high divorce rates for Blacks (Kreider & Fields, 2002). Model 3 adds measures of personal and household income. This further reduces the baby gap between male professors and physicians. After controlling for income, male professors are 11% [100 * (1 − exp(−0.12))] less likely than male physicians to have birth events. Men’s personal income increases the likelihood of a baby in the household, whereas household income makes a baby less likely. On average, male professors make less money than either doctors or lawyers (according to Table 1, male professors have a median personal income of $48,500, compared with $75,000 for attorneys and $100,000 for physicians); this is apparently another reason why the former have fewer children than the latter. Model 4 is identical to Model 3 except that it includes only married and cohabiting male professionals. This model serves as a baseline for the introduction of additional variables associated with spouses or cohabiting partners. According to Model 4, married or cohabiting professors are 10% less likely than married or cohabiting physicians to have a birth event [100 * (1 − exp(−0.11))]. The effect sizes and significance levels of the other covariates are similar to those in Model 3. Model 5 introduces a measure of spousal employment. This variable accounts for the remaining difference between male physicians and academics in the chances of a birth event. According to Table 1, male doctors are almost twice as likely to have spouses who are out of the labor force in comparison with male academics (38% vs. 22%). Therefore, married doctors probably have more birth events in part because they have wives available for child care. Also, male professionals whose wives are physicians and lawyers are disproportionately likely to have birth events. So too are men whose wives are out of the labor force and, therefore, more available for child care. However, male professionals whose wives are academics do not have an elevated probability of a birth event. Given the relatively high rate at which academics marry other academics (Table 1; Jacobs, 2004), it appears likely that the low fertility of female professors, described below, can help account for the relative paucity of birth events among male professors. 1662 Journal of Family Issues 31(12) The results shown in Table 2 confirm that male lawyers and, especially, male professors are less likely to have babies than male physicians. For lawyers, this disparity can be explained by marital status. They are less likely to be married than doctors, and married people have more birth events. Finally, differences in marital status and spousal employment can account for the low rate at which male professors have birth events. Fertility Among Professional Women Table 3 shows multivariate differences in fertility between female professors, lawyers, and physicians. Controlling for age, weekly hours worked, and race/ ethnicity, Model 1 confirms that professors are less likely than physicians to have a baby in the household. The disparity is larger for women than men, with female professors 41% [100 * (1 − exp(−0.53))] less likely than female physicians to have a birth event. Female lawyers are also less likely than female physicians to have a baby in the household, with 23% lower odds [100 * (1 − exp(−0.26))]. As is the case for men, age is strongly correlated with fertility. In contrast to their male counterparts, working long hours is correlated with decreased fertility among female professionals. Most notably, putting in 40 to 49 hours a week is associated with 38% [100 * (1 − exp(−0.48))] lower odds of a birth event compared with professional women who work less than 20 hours a week; working 50 to 59 hours a week produces 55% lower odds [100 * (1 − exp(−0.79))] and working 60 or more hours yields 64% lower odds [100 * (1 − exp(−1.01))]. Among female professionals, professors work the least, with an average 43-hour work week; physicians work the most, with a 52-hour work week; and lawyers are in the middle, with a 44-hour work week (see Table 1). As for men, Whites are significantly more likely to have birth events than the members of the other population groups. Model 2 adds marital status to the analysis. Marital status is associated with fertility, though not as strongly as for men. Among male professionals (see Table 2), being married increases the odds of having a baby by 296 times [100 * exp(5.69)], in comparison with 35 times [100 * exp(3.55)] for women professionals. Cohabiting and previously married women also have higher rates of birth events than do never-married women. Adjusting for marital status markedly reduces the baby gap between female professors, lawyers, and physicians. After controlling for marital status, female professors are 28% [100 * (1 − exp(−0.33))] and female lawyers are 10% [100 * (1 − exp(−0.11))] less likely to have babies than are female physicians; the latter effect is only significant at the .10 level. The corresponding figures 1663 Wolfinger et al. Table 3. Logistic Regressions of Birth Events for Female Professionals All Professionals Model 1 Model 2 Married/Cohabiting Model 3 Model 4 Model 5 Employment status Physician — — — — — −0.12* −0.13* −0.08 Lawyer −0.26*** −0.11+ Professor −0.53*** −0.33*** −0.29*** −0.28*** −0.23* Race/ethnicity White — — — — — African American −0.42*** −0.06 −0.03 −0.13 −0.12 Asian American −0.16* −0.23** −0.22** −0.22** −0.20* Hispanic −0.35*** −0.28* −0.27* −0.23* −0.23* Other −0.41 −0.42 −0.42 −0.37 −0.38 Age 1.92*** 1.76*** 1.74*** 1.84*** 1.84*** Age squared −0.03*** −0.03*** −0.03*** −0.03*** −0.03*** Weekly hours 1-19 — — — — — 20-29 0.04 0.03 0.09 0.11 0.10 30-39 −0.19 −0.12 −0.04 −0.04 −0.04 40-49 −0.48*** −0.30** −0.20 −0.19 −0.19 50-59 −0.79*** −0.55*** −0.46*** −0.42*** −0.42** 60 or more −1.01*** −0.73*** −0.63*** −0.60*** −0.61*** Percentage urban −0.14 −0.13 0.08 0.18 0.23+ Union status Never married — — — — — Divorced/separated/ — 1.51*** 1.53*** — — widowed Married — 3.55*** 3.45*** — — Cohabiting — 2.00*** 1.93*** −1.52*** −1.55*** Household resident 65 — −0.04 −0.06 −0.14 −0.06 and older Log of personal income — — −0.07** −0.08** −0.07* Log of household income — — 0.15*** 0.16*** 0.12** Spousal/partner employment status Professor — — — — −0.05 Physician — — — — 0.14+ Lawyer — — — — 0.06 Other employment — — — — — Out of labor force — — — — 0.21 Not in house — — — — −1.06*** Constant 32.82*** 33.68*** 33.04*** 32.09*** 31.69*** Log-likelihood −6435.97 −5675.47 −5667.53 −5337.19 −5319.18 Note. Analyses are weighted. N is 17,035 for Models 1 to 3 and 10,919 for Models 4 and 5. p < .10. *p < .05. **p < .01. ***p < .001, two-tailed. + 1664 Journal of Family Issues 31(12) in Model 1, lacking the control for marital status, are 41% and 23%, respectively. Like their male counterparts, female professors and lawyers are less likely to be married or cohabiting than female physicians: According to Table 1, 68% of female professors and 66% of female lawyers are married or cohabiting, compared with 75% of female physicians. Female professors are the most likely (10%) of the three professions to be separated, divorced, or widowed. These differences in marital status apparently account for part of the disparity in birth events between female doctors, lawyers, and professors, as well as the difference between Whites and African Americans. Model 3 includes measures of personal and household income. As was the case with hours worked, income has the opposite effect for female professionals as it does for their male colleagues. Female doctors, lawyers, and professors with higher personal incomes are less likely to have birth events. Conversely, the more household income, the more babies women have. One explanation for these findings is that female professionals experiencing a birth event take time off work, producing a commensurate decline in income. This in turn leads to income generation by the woman’s spouse or cohabiting partner. Model 3 shows that income differentials help explain the lower fertility of professors but not attorneys. Controlling for income has a small effect on the likelihood a professor experiences a birth event: the corresponding odds ratios decline from −28% (exp[−0.33]) in Model 2 to −25% (exp[−0.29]). Perhaps one reason female academics do not have children is because of their comparably low incomes. By dint of their salaries, doctors are better able to pay for child care or provide their partners with the opportunity to be stay-at-home fathers. However, female attorneys’ fertility is not affected by controlling for income. Models 4 and 5 include only married and cohabiting women. Model 4 is otherwise similar to Model 3; Model 5 adds a measure of spousal employment. This accounts for a little more of the difference in fertility between women professors and women physicians: Based on Model 5, female professors are 21% [100 * (1 − exp(−0.23))] less likely to have a baby than women physicians. For women lawyers, spousal employment accounts for the rest of the baby gap: The regression coefficient measuring attorneys’ birth events loses statistical significance in Model 5. As is the case with male professionals, a physician spouse increases the likelihood of a birth event. Predictably, women professionals with spouses outside the house are much less likely to have a baby than those with spouses at home (although the difference is not as large as the corresponding gap for male professions). In contrast to men, women with a spouse outside the labor force do not have higher rates of birth events. Men “opt out” at far lower rates than do women. Wolfinger et al. 1665 In sum, female lawyers and professors have fewer birth events than do physicians. For lawyers, this disparity is the product of differences in marital status, income, and spousal employment. Together these variables account for approximately half the baby gap between female doctors and professors. The other half is attributable to factors that cannot be measured with Census data. Discussion Birth events vary dramatically by sex and profession. Male professionals are more likely to have a baby at home than women; physicians have the most babies, attorneys are in the middle, and professors have the fewest. Male professors have fewer children than doctors or lawyers, but this can be explained by differences in marital status, income, and spousal employment. Male professors make less money and marry at lower rates than male physicians; when they do wed, they choose spouses conducive to low fertility—women who are employed as professors, women in other jobs, or women who reside elsewhere. In particular, the propensity for college professors to marry each other may well contribute to low fertility among male faculty. Differences in income, marital, and spousal characteristics can also explain the lower fertility rate for female attorneys. However, these factors only account for part of the baby gap between female doctors and professors. Perhaps the remainder is rooted in the lengthy training and probationary periods characteristic of academia. With these challenges in mind, some scholars of higher education have called for flexible tenure clocks, temporary part-time options, and other family-friendly interventions on behalf of female professors (Frasch, Mason, Stacy, Goulden, & Hoffman, 2007; Smith & Waltman, 2006). Perhaps also there are occupational differences in personality: professors may simply be less interested in having children than doctors and lawyers. In any event, our results show that work–family balance issues are particularly salient for America’s colleges and universities. A shortcoming of this study is the lack of information on the type of employment other than broad job classification. We do not know if academics are tenured professors at large research universities or part-timers at community colleges. Lawyers may have high-powered corporate careers or low-paying jobs with nonprofit agencies. Physicians could be internists or neurosurgeons. This information would likely reveal larger occupational differences in birth events: We suspect that some women choose undemanding career paths to avoid work–family conflict. We know, for instance, that adjunct faculty are disproportionately female (Curtis, 2004; see also Perna, 2001b); this is especially 1666 Journal of Family Issues 31(12) true for women with young children (Wolfinger, Mason, & Goulden, 2009). Moreover, only 26% of full professors are women (American Association of University Professors, 2001), as are 17% of partners at law firms (O’Brien, 2006). Of the approximately 26,300 practicing general surgeons in the United States, just 12% are women (American College of Surgeons, 2007). In addition, our data do not speak to men and women who leave the paid labor force altogether after a birth event. For years traditional liberal feminism has been partly framed as women’s struggle to achieve economic and vocational parity with men (National Organization for Women, 2007; Roth, 2003; Tong, 1989). The majority of women now work, even if they have children (U.S. Census Bureau, 2009). The gender gap in wages has gradually narrowed (U.S. Census Bureau, 1999). As Suzanne Bianchi (1997) remarked 13 years ago at a Consortium of Social Science Associations Congressional Breakfast seminar, “Men and women are not equal, but when it comes to market work, to earnings, to the jobs they hold, the changes are all in the direction of greater equality.” Becoming a doctor, a lawyer, or a professor—all high-status jobs—can be viewed as a triumph of the women’s movement. But there is another way to measure women’s progress relative to men: whether their professional gains have been offset by familial losses (Mason, 1988/2002). Most Americans want children (Thornton & Young-DeMarco, 2001), so it may undercut women’s professional accomplishments if they have to forego motherhood in the process. If this metric is applied to our findings, female fast-track professionals have not attained equality. Moreover, professors make greater familial sacrifices than do doctors or lawyers. In a survey of faculty at all nine (now 10) branches of the University of California, 20% of men and 40% of women said they had fewer children than they would have liked (Mason & Goulden, 2004). Like Cooney and Uhlenberg (1989), our results show that female doctors are most likely to have children, while female professors are least likely. Also in accordance with previous research (Avellar & Smock, 2003; Budig & England, 2001; Waldfogel, 1997), our study finds that women’s wages are negatively correlated with birth events. We offer two new findings. First, female professionals have fewer children than their male counterparts. Second, fertility varies dramatically by occupation. If women are sacrificing families for careers, the Women’s Movement has not come nearly as far as we might like. Wolfinger et al. 1667 Author’s Note A previous version of this article was presented at the 2008 annual meeting of the Population Association of America in New Orleans. Acknowledgments Phil Morgan, Benita Roth, Sharon Sassler, and the Utah Demography Research Network provided useful advice on this project. Sonya Anderson and Alta Williams furnished able research assistance. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article. Funding The authors thank the Alfred P. Sloan Foundation (2004-5-25 DLC) for its generous support. References AllBusiness.com. (2003, December). Rising to the top. Retrieved from http://www .allbusiness.com/legal/law-firms/982729-1.html?yahss=114-3470923-982729 American Association of University Professors. (2001). Statement of principles on family responsibilities and academic work. Retrieved from http://www.aaup.org/ AAUP/pubsres/policydocs/contents/workfam-stmt.htm American College of Surgeons. (2007). Survey shows gender differences are a factor when surgeons in training choose a subspecialty. Retrieved from http://www.facs .org/news/jacs/genderdiffsub1107.html Association of American Medical Colleges. (2007). FACTS: Table 15: Applicant age at anticipated matriculation, 2003-2006. Retrieved from http://www.aamc.org/data/ facts/2006/2006age.htm Avellar, S., & Smock, P. J. (2003). Has the price of motherhood declined over time? A cross-cohort comparison of the motherhood wage penalty. Journal of Marriage and Family,65, 597-607. Bianchi, S. M. (1997). What the Census tells us about women in the workforce. In Using the Census (pp. 18-31). Washington, DC: Consortium of Social Science Associations. Bianchi, S. M., Robinson, J. P., & Milkie, M. A. (2006). Changing rhythms of American family life. New York, NY: Russell Sage Foundation. Boulis, A. (2004). The evolution of gender and motherhood in contemporary medicine. Annals of the American Academy of Political and Social Science, 296, 172-206. 1668 Journal of Family Issues 31(12) Budig, M. J. (2003). Are women’s employment and fertility histories interdependent? An examination of causal order using event history analysis. Social Science Research, 32, 376-401. Budig, M. J., & England, P. (2001). The wage penalty for motherhood. American Sociological Review, 66, 204-225. California Birth Defects Monitoring Program. (2005). Registry data, 1997-2001. Retrieved from http://www.cbdmp.org/bd_down_syn.htm Cooney, T. M., & Uhlenberg, P. (1989). Family-building patterns of professional women: A comparison of lawyers, physicians and postsecondary teachers. Journal of Marriage and the Family, 51, 749-758. Curtis, J. W. (2004). Faculty salary and faculty distribution fact sheet 2003-04. Washington, DC: American Association of University Professors. Dye, J. L. (2008). Fertility of American women: 2006 (Current Population Reports, P20-558). Washington, DC: U.S. Census Bureau. Frasch, K., Mason, M. A., Stacy, A., Goulden, M., & Hoffman, C. (2007). Creating a family friendly department: Chairs and deans toolkit. Berkeley, CA: University of California, Berkeley. Retrieved from http://ucfamilyedge.berkeley.edu Hewlett, S. A., & Luce, C. B. (2005). Off-ramps and on-ramps: Keeping talented women on the road to success. Harvard Business Review, March, 1-11. Hoffer, T. B., Welch, V., Jr., Webber, K., Williams, K., Lisek, B., Hess, M., … GuzmanBarron, I. (2006). Doctorate recipients from United States universities: Summary report 2005. Chicago, IL: National Opinion Research Center. Retrieved from http:// www.norc.uchicago.edu/issues/docdata.htm Jacobs, J. A. (2004). The faculty time divide. Sociological Forum, 9, 3-27. Jacobs, J. A., & Winslow, S. E. (2004). The academic life course, time pressures, and gender inequality. Community, Work, & Family, 7, 143-161. Jolly, P. (2007). Medical school tuition and young physician indebtedness: An update to the 2004 report. Washington, DC: American Association of Medical Colleges. Kreider, R. M., & Fields, J. M. (2002). Number, timing, and duration of marriages and divorces: 1996 (Current Population Reports, P70-80). Washington, DC: U.S. Census Bureau. Kulis, S., & Sicotte, D. (2002). Women scientists in academia: Geographically constrained to big cities, college clusters, or the coasts? Research in Higher Education, 43, 1-30. Leslie, D. W., & Walke, J. T. (2001). Out of the ordinary: The anomalous academic. Unpublished manuscript, The College of William and Mary, Williamsburg, VA. Long, S. J. (Ed.). (2001). From scarcity to visibility: Gender differences in the careers of doctoral scientists and engineers. Washington, DC: National Academies Press. Macunovich, D. J. (1996). Relative income and price of time: Exploring their effects on U.S. fertility and female labor force participation. Population and Development Wolfinger et al. 1669 Review, Supplement: Fertility in the United States: New Patterns, New Theories, 22, 223-257. Mason, M. A. (2002). The equality trap. Somerset, NJ: Transaction. (Original work published 1988) Mason, M. A., & Goulden, M. (2004). Marriage and baby blues: Redefining gender equity in the academy. Annals of the American Academy of Political and Social Science, 596, 86-103. Masters, E. R. (2004). The importance of course management tools. Paper presented at the annual meeting of the Association of American Law Schools, Atlanta, GA. National Organization for Women. (2007). Retrieved from http://www.now.org Noonan, M. C. (2005). The long-term costs of women’s work interruptions. Unpublished manuscript, Department of Sociology, University of Iowa, Iowa City, IA. Noonan, M. C., & Corcoran, M. E. (2004). The mommy track and partnership: Temporary delay or dead end? Annals of the American Academy of Political and Social Science, 596, 130-150. O’Brien, T. L. (2006, March 19). Why do so few women reach the top of big law firms? The New York Times. Retrieved from http://www.nytimes.com/2006/03/19/business/ yourmoney/19law.html Perna, L. (2001a). Sex and race differences in faculty tenure and promotion. Research in Higher Education, 42, 541-567. Perna, L. (2001b). The relationship between family responsibilities and employment status among college and university faculty. Journal of Higher Education, 72, 584-611. Rindfuss, R. R., Morgan, S. P., & Offutt, K. (1996). Education and the changing age pattern of American fertility: 1963-89. Demography,33, 277-290. Rindfuss, R. R., & Sweet, J. A. (1977). Postwar fertility trends and differentials in the United States. New York, NY: Academic Press. Roth, B. (2003). Separate roads to feminism: Black, Chicana, and White feminist movements in America’s second wave. New York, NY: Cambridge University Press. Smith, G. C., & Waltman, J. A. (2006). Designing and implementing family friendly policies in higher education. Ann Arbor: Center for the Education of Women, University of Michigan. Thornton, A., & Young-DeMarco, L. (2001). Four decades of trends in attitudes toward family issues in the United States: The 1960s through the 1990s. Journal of Marriage and Family, 63, 1009-1037. Tong, R. (1989). Feminist thought. Boulder, CA: Westview Press. U.S. Census Bureau. (1999). Money income in the United States: 1999 (Current Population Reports, P60-209). Washington, DC: U.S. Government Printing Office. U.S. Census Bureau. (2003). Public Use Microdata Sample (PUMS). Retrieved from http://www.census.gov/main/www/pums.html 1670 Journal of Family Issues 31(12) U.S. Census Bureau. (2009). Statistical abstract of the United States: 2009 (128th ed.). Washington, DC: Government Printing Office. Van Anders, S. M. (2004). Why the academic pipeline leaks: Fewer men than women perceive barriers to becoming professors. Sex Roles, 51, 511-521. Waldfogel, J. (1997). The effect of children on women’s wages. American Sociological Review, 62, 209-217. Wolfinger, N. H., Mason, M. A., & Goulden, M. (2008). Problems in the pipeline: Gender, marriage, and fertility in the ivory tower. Journal of Higher Education, 79, 388-405. Wolfinger, N. H., Mason, M. A., & Goulden, M. (2009). “Stay in the game”: Gender, family formation, and alternative trajectories in the academic life course. Social Forces, 87, 1591-1621. Yang, Y., & Morgan, S. P. (2004). How big are educational and racial fertility differentials in the U.S.? Social Biology, 51, 167-187.