We provide the first direct quasi-experimental evidence of the effects of legalized same-sex marriage in the United States on marriage and partnership rates by studying Massachusetts, which in May 2004 became the first state in the US to issue marriage licenses to gay and lesbian couples after the unexpected state Supreme Court ruling in Goodridge v. Department of Health in November 2003.  Using confidential public health data from the state’s Behavioral Risk Factor Surveillance System from 2001 to 2010 that asked respondents direct questions about sexual orientation, marital status, and household composition, we show that Goodridge led to immediate, large, and statistically significant increases in the probability that gay men, lesbians, and bisexual women report being married by 14, 31, and 12 percentage points, respectively, relative to the associated rates for heterosexual men and women.  These changes are entirely driven by increases in marriage among sexual minorities as opposed to decreases in marriage among heterosexuals.  We estimate that about half of gay men and lesbians in cohabiting relationships married.  These increases in marriage were mostly offset by statistically significant decreases in the probability of being ‘a member of an unmarried couple’, suggesting that legalized gay marriage had relatively modest overall effects on union formation, particularly among lesbians.  The presence of children in the household is the strongest correlate of marriage among lesbians in the post-Goodridge era, suggesting the importance of economic and legal factors in the decision of same sex couples to marry.