Windsor’s Purgatory: State DOMA’s Can’t Stabilize Straight Marriages, but They Can Still Prevent Gay Marriage

[PDF]

Deirdre M. Bowen†

Introduction

Gareth used to prefer funerals to weddings. He said it was easier to get enthusiastic about a ceremony one had an outside chance of eventually being involved in. –from Four Weddings and a Funeral (1994).[1]

The United States Supreme Court recently struck down § 3 of the Defense of Marriage Act (DOMA).[2] The case rose on appeal from the Second Circuit, which determined that the statute contained an unconstitutional provision.[3] While the Supreme Court affirmed the lower court’s ruling, its analysis focused more on the animus behind DOMA’s enactment.[4] In doing so, it opened the door wide for state DOMAs to come under constitutional attack.[5] Supporters of DOMA consistently assert that DOMA is needed to protect our society—to strengthen and protect traditional marriage, the cornerstone of civilization.[6] In other words, banning same-sex marriage protects “traditional” marriage. But do DOMAs really protect the institution of marriage?

The Supreme Court ruling has substantial historical and social significance.[7] Given that nine states have statutes barring same-sex marriage and thirty states have constitutional amendments that prohibit the practice,[8] it is a worthy exercise to explore whether these states have reaped the benefits that they hoped to achieve by enacting mini-DOMAs.[9] This Essay offers a preview a forthcoming empirical investigation resolving that question and concludes that DOMA provided no measurable benefit to protecting families.[10] Regardless, in light of the Court’s decision, this discourse is not likely to disappear anytime soon.

The federal DOMA’s demise offers less relief for those gay and lesbian families who wish to wed but reside in states with statutes or constitutional amendments that bar same-sex marriage. Those states’ mini-DOMAs remain intact.[11]

Methodology and Results

Soon after the federal law went into effect,[12] states began enacting mini-DOMAs,[13] either by statute or state constitutional amendment, and sometimes both. However, not all states adopted their own version of DOMA, i.e. a mini- or super-DOMA, and, in the wake of the federal legislation, some states chose to find some parallel version of marriage in the form of civil unions[14] or domestic partnerships instead.[15] Furthermore, a select few states, mostly through court action, came to permit same-sex marriage or at least recognize same-sex marriages performed in other states even though, at one time, the state may have enacted a mini-DOMA.[16] Thus, differing state reactions to the conundrum of how to respond to same-sex marriage allows for a statistical analysis of how a state’s mini-DOMA legislation may have affected the culture of marriage and divorce in the United States.[17] The analysis is rather simple. Changes in marriage and divorces trends and rates are compared between states that enacted DOMAs and those that permit same-sex marriage. 

Analysis of State DOMA’s

Asked differently, the following question arises: did DOMA increase marriage and decrease divorce over time in states in which it was enacted compared with states that shunned DOMA legislation? One might predict that those states where voters have approved constitutional amendments would be associated with the greatest average rate of decline in divorce.[18] Accordingly, those states with just a DOMA statute might experience a less radical decline in divorce rates. Finally, those states without a DOMA statute or amendment would likely have the lowest decline in divorce.[19] As for marriage, one could hypothesize that in DOMA-amendment or amendment-plus-statute states, the greatest increases in marriage rates occurred, followed by lesser increases in DOMA-statute states and even fewer in no-DOMA states.

The analysis involves two separate statistical examinations—first, looking into the possibility of any statistically significant differences in the average marriage and divorce rates, and second, evaluating trends in the years before and after DOMA ratification while cross-referencing those states that enacted DOMAs with those that did not.

The determinative question is whether the mean difference in the trend of decline for divorce and marriage in the two time periods, pre- and post-DOMA amendment passage, is statistically different between the two groups of states: those that enacted an amendment or statute and those that did not. Put another way, this analysis seeks to identify whether states that passed DOMAs experienced greater marriage rates and reduced divorce rates compared to those states that remained DOMA-free.

Again, using IBM’s SPSS, a predictive analytics software, I conducted an independent-sample-means T-test to determine whether statistically significant differences marked DOMA and non-DOMA states for pre- and post-DOMA marriage trends and pre- and post-DOMA divorce trends.[20] The first hypothesis is that DOMA impacts marriage and divorce rates differently in DOMA states than in non-DOMA states, discussed supra. The second hypothesis predicts that the average rate of decline for divorce would be greatest in those states that passed an amendment compared with those states that passed only a statute or nothing at all. The results demonstrate otherwise.

Table One indicates that no statistically significant difference separates the DOMA and non-DOMA states in the divorce and marriage trends prior to enactment of DOMA legislation. The average rate of decline of marriage was greater for the DOMA states than the non-DOMA states prior to the enactment of any amendments, -.25 and -.22 respectively.[21] And for divorce trends, the analysis reveals that DOMA states are virtually identical in their decline to non-DOMA states, -.1 versus -.09, respectively.[22] These extremely slight differences are not surprising and do not rise to the level of statistical significance or substantive significance. However, as desired, the results establish a pre-DOMA baseline indicating that both categories behaved fundamentally the same with regard to family formation and dissolution. 

Table One

Comparison of Marriage and Divorce Trends between DOMA and non-DOMA States Prior to Amendment Passage

 

 

Marriage Trend

Divorce Trend

n

DOMA

-0.22

-0.1

31

Non-DOMA

-0.25

-0.09

19

 

More importantly, though, no statistically significant difference marks the two groups of states after the passage of DOMA. The average decline in marriage or divorce after DOMA does not differ in any statistically meaningful way between those states that adopted an amendment and those that did not. Post-DOMA, the decline in marriage was greater for DOMA states than non-DOMA states, -.26 versus -.12. Moreover, the falling off of divorce rates was greater in non-DOMA states than DOMA states, -.011 and -.008, respectively.

Although surprising that DOMA states, after the passage of DOMA, appear to have a lower rate of decline in divorce and a higher rate of decline in marriage compared to non-DOMA states, these results are not statistically significant. Thus, the data can only hint that the institution of marriage might be slightly more vulnerable in DOMA states.

Although no statistically significant differences emerge between the two groups of states in the analysis, the substantive differences are worth noting. Post-DOMA, the decline of marriage varies quite a bit between the two groups of states. While in all other areas the trends are negligible, the average drop in marriages post-DOMA enactment for DOMA states is more than double that of non-DOMA states. Certainly, the results should be read with caution, but they do raise some skepticism about the power of DOMAs to create family stability. 

Table Two

Comparison of Marriage and Divorce Trends between DOMA and non-DOMA States after Amendment Passage

 

 

Marriage Trend

Divorce Trend

n

DOMA

-0.26

-0.008

31

Non-DOMA

-0.12

-0.011

19

 

This preliminary[23] analysis suggests that DOMA is not statistically associated with increases in marriage rates or decreases in divorce rates.[24] Specifically, DOMA does not appear relevant to the narrative of why marriage plays an increasingly less visible role in family formation in the United States, particularly in DOMA states. 

Conclusion

The analysis suggests that DOMA states do not fare any better than non-DOMA states in terms of strengthening the “bonds and benefits to society of heterosexual marriage.”[25] In fact, the analysis offers an alternative theory. DOMA states tend to have lower marriage rates, larger declines in the trend towards marriage, and higher divorce rates. Thus, the argument that same-sex marriage undercuts traditional marriage is not relevant. My full-length article, soon to be published in the Denver University Law Review, explores why the argument has such staying power and what factors could strengthen all families.

 

Appendix 1

       

Domestic Partnership

 
 

Amendment

Statutory

Civil

Unions

Same-Sex

AK

1996

1996

     

AL

2006

1998

     

AR

2004

2005

     

AZ

2008

1996

     

CA

2008

2000

 

2007

 

CO

2006

2006

2013

2009

 

CT

       

2009

DC

       

2010

DE

 

2009

2011

 

2013

FL

2008

1997

     

GA

2004

1996

     

HI

1998

 

2012

   

IA

       

2009

ID

2006

1996

     

IL

 

2006

2011

   

IN

 

1997

     

KS

2005

       

KY

2004

1998

     

LA

2004

1999

     

MA

       

2003

MD

 

2006

   

2012

ME

 

1997

 

2004

2012

MI

2004

1996

     

MN

 

1997

     

MO

2004

2001

     

MS

2004

       

MT

2004

1997

     

NC

2012

1996

     

ND

2004

1997

     

NE

2000

       

NH

       

2010

NJ

   

2007

   

NM

       

Other Jdx

NV

2002

   

2009

2011

NY

         

OH

2004

2004

     

OK

2004

1997

     

OR

2004

   

2011

 

PA

 

1996

     

RI

   

2011

 

2013

SC

2006

1996

     

SD

2006

2000

     

TN

2006

1996

     

TX

2005

2003

     

UT

2004

2004

     

VA

2006

2004

     

VT

   

2000

 

2009

WA

 

1998

 

2009

2012

WI

2007

2009

 

2013

 

WV

 

2001

     

WY

 

1977

     

 


        †     Associate Professor of Law, Seattle University School of Law. Many people deserve thanks for their contributions to this article: Professor Richard Delgado for his inspiration; Professors Jane Stover and Holning Lau for reading an earlier version of the article; the Emerging Family Law Scholars for their guidance at the incubator stage; my outstanding research assistants Sarah Albertson, Stacie Naczelnik, Valerie Queseda, Peter Rudolf, and Carl Schremp; and my co-number cruncher, Hayley A.B. Pippin.

       [1].     Four Weddings and a Funeral (PolyGram Filmed Entertainment 1994).

       [2].     Defense of Marriage Act, Pub. L. No. 104-199, 110 Stat. 2419 (codified as amended at 1 U.S.C. § 7 (1996)).

       [3].     On December 7, 2012, the Supreme Court granted a petition for a writ of certiorari for the Second Circuit cases combined under Windsor. United States v. Windsor, 133 S. Ct. 786, 787 (2012). In addition, the Court heard California’s Prop 8 case. Hollingsworth v. Perry, 671 F.3d 1052 (9th Cir. 2012). Argument was heard on March 27, 2013. Oral Argument, United States v. Windsor, 133 S. Ct. 786 (2012) (No. 12-307), available at http://www.supremecourt.gov/oral_arguments/argument_audio_detail.aspx?argument=12-307&TY=2012. Prior to this ruling, other U.S. district courts have ruled on the constitutionality of DOMA. E.g. Pedersen v. Office of Pers. Mgmt., 881 F. Supp. 2d 294 (2012) (finding the provision violates equal protection principles and granting the plaintiff’s motion for summary judgment). The court found that the plaintiffs deserved heightened scrutiny as a suspect class, but also found DOMA’s articulated goals do not pass even the most deferential rational basis review. Windsor v. United States, 699 F.3d. 169, 181–88 (2012) (finding a state’s interests behind DOMA not based in reality, and thus, plaintiff’s motion for summary judgment was granted).

       [4].     United States v. Windsor, 133 S. Ct. 2675, 2693–96 (2013).

       [5].     Id. at 2710 (Scalia, J., dissenting).

       [6].     See e.g., Bishops’ Committee For Defense of Marriage Disappointed Over DOMA Ruling, U.S. Conference of Catholic Bishops (June 1, 2012), http://www.usccb.org/news/2012/12-096.cfm; Karla Dial, 1st Circuit declares part of federal marriage law unconstitutional, Christian Civil League of Maine (May 31, 2012), http://www.cclmaine.org/1st-circuit-declares-part-of-federal-marriage-law-unconstitutional. Bishop Cordileone of the United States Conference of Catholic Bishops declared:

The federal appeals court in Boston did a grave injustice yesterday by striking down that part of the Defense of Marriage Act that reasonably recognizes the reality that marriage is the union of one man and one woman. DOMA is part of our nation’s long-established body of law rooted in the true meaning of marriage. Hopefully, this unjust ruling will be overturned by the U.S. Supreme Court, for the benefit of our nation’s children, and our nation as a whole.

U.S. Conference of Catholic Bishops, supra. “Society should protect and strengthen marriage, not undermine it. The federal Defense of Marriage Act provides that type of protection, and we trust the U.S. Supreme Court will reverse the 1st Circuit's erroneous decision." Karla Dial, supra (quoting Dale Schowengerdt, Legal Counsel for the Alliance Defense Fund).

       [7].     Daniel Fisher, DOMA is Dead. So Where Does that Leave Gay Couples? Forbes.com (March 28, 2013, 11:04 AM), http://www.forbes.com/sites/danielfisher/2013/03/28/doma-is-dead-so-where-does-that-leave-gay-couples.

       [8].     See infra note 12. The Second Amendment to Hawaii’s Constitution did not ban same-sex marriage. Haw. Const. amend. II. It gave the legislature the authority to define marriage as it sees fit. Id.

       [9].     See Massachusetts v. U.S. Dept. of Health and Human Services, 682 F.3d 1 (1st Cir. 2012). The First Circuit opinion did not address § 2 of DOMA, which frees states that ban same-sex marriage from having to recognize same-sex marriages performed in states that do license homosexual matrimony. Id. at 6.

     [10].     I use the phrase at-risk families or families-in-crisis interchangeably to refer to the rhetoric employed around the weakening family structure—i.e. family formation without marriage or family cycles that include divorce and perhaps remarriage and perhaps divorce again.

     [11].     William Saletan, Gay Bells in Bondage: Most Americans now support gay marriage. But they can’t legalize it, thanks to the voters of 2004, Slate (June 28, 2011, 8:58 AM), http://www.slate.com/articles/news_and_politics/frame_game/2011/06/gay_bells_in_bondage.html. But see Anna Staver, Same-Sex Marriage Amendment in Ohio Gets Green Light, The Huffington Post (Apr. 3, 2012, 6:29 PM), http://www.huffingtonpost.com/2012/04/03/same-sex-marriage-amendment-ohio_n_1400714.html?ncid=edlinkusaolp00000009 (Ohio may vote to overturn its 2004 Constitutional Amendment banning same-sex marriage). Maine will vote on whether to overturn a same sex-marriage ban passed originally in 2009. Jillian Rayfield, The Fight for Marriage Equality Moves to State Ballots, Rolling Stone (June 8, 2012, 10:35 AM), http://www.rollingstone.com/politics/blogs/national-affairs/the-fight-for-marriage-equality-moves-to-state-ballots-20120608.

     [12].     Forty-one states have, at one time or another, enacted mini-DOMA legislation through their constitution or statutory law; many states overlap and have both statutory and constitutional mini-DOMAs.

               Currently, thirty-one states have constitutional mini-DOMAs. See Ala. Const. art. I, § 36.03 (2006); Alaska Const. art. I, § 25 (1996); Ariz. Const. art. XXX, § 1 (2008); Ark. Const. amend. 83, § 1 (2004); Cal. Const. art. I. § 7.5 (2008); Colo. Const. art. II, § 31 (2006); Fla. Const. art. I, § 27 (2008); Ga. Const. art. I, § IV (2004); Haw. Const. art. I, § 23 (1998)(amendment empowers legislature “to reserve marriage to opposite sex couples” but does not explicitly ban same-sex marriages); Idaho Const. art. III, § 28 (2006); Kan. Const. art. XV, § 16 (2005); Ky. Const. § 233a (2004); La. Const. art. XII, § 15 (2004); Mich. Const. art. I, § 25 (2004); Miss. Const. art. XIV, § 263-A (2004); Mo. Const. art. I, § 33 (2004); Mont. Const. art. XIII, § 7 (2004); Neb. Const. art. I, § 29 (2000); Nev. Const. art. I, § 21 (2002); N.C. Const. art. XIV, § 6 (2012); N.D. Const. art. XI, § 28 (2004); Ohio Const. art. XV, § 11 (2004); Okla. Const. art. II, § 24 (2004); Or. Const. art. XV, § 5a (2004); S.C. Const. art. XVII, § 15 (2006); S.D. Const. art. XXI, § 9 (2006); Tenn. Const. art. XI, § 18 (2006); Tex. Const. art. I, § 32 (2005); Utah Const. art. I, § 29 (2004); Va. Const. art. I, § 15-A (2006); Wis. Const. art. XIII, § 13 (2007).

               Similarly, thirty-four states have statutory mini-DOMAs. See Ala Code § 30-1-19 (1998); Alaska Stat. Ann. §§ 25.05.011, .013 (West 1996); Ariz. Rev. Stat. Ann. §§ 25-101,-112 (1996); Ark. Code Ann. § 9-11-109 (West 2005); Cal. Fam. Code § 308.5 (West 2000); Colo. Rev. Stat. Ann. § 14-2-104 (West 2006); Del. Code. ann. tit. 13, § 101 (West 2009); Fla. Stat. Ann. § 741.212 (West 1997); Ga. Code. Ann. § 19-3-3.1 (West 1996); Idaho Code. Ann. § 32-201 (West 1996); 750 Ill. Comp. Stat. Ann. 5/212 (West 2006) (But see H.B. 5170, 97th Gen. Assemb., 2d Reg. Sess. (Ill. 2011) (proposed legislation H.B. 5170 provides that marriages between same-sex couples will be treated the same as marriages between opposite-sex couples); Ind. Code. Ann. § 31-11-1-1 (West 1997); Ky. Rev. Stat. Ann. §§ 402.005, .020, .040, .045 (West 1998); La. Civ. Code Ann. art. 89 (1999); Mich. Comp. Laws Ann. §§ 551.271, .271, .272 (West 1996); Minn. Stat. Ann. §§ 571.01, .03, 518.01 (West 1997) (But see H.F. 1761, 87th Leg. Reg. Sess. (Minn. 2011) (proposed legislation alters language, replacing the word “marriage” with “civil unions,” granting same-sex couples the same rights as opposite-sex couples); Mo. Ann. Stat. § 451.022 (West 2001); Mont. Code Ann. § 40-1-401 (1997); N.C. Gen. Stat. Ann. § 51-1.2 (West 1996); N.D. Cent. Code Ann. §§ 14-03-01, -08 (West 1997); Ohio Rev. Code Ann. § 3010.01(A) (West 2004); Okla. Stat. Ann. tit. 43, § 3.1 (West 1997); 23 Pa. Stat. Ann. § 1704 (West 1996) (But see H.B. 1835, 195th Gen. Assemb., Reg. Sess. (Pa. 2011) (proposed legislation providing for same-sex marriages and repeal laws that prohibit them); S.C. Code Ann. § 20-1-10, -15 (1996); S.D. Codified Laws §§ 25-1-1, -38 (West 1996 & 2000); Tenn. Code Ann. § 36-3-113(West 1996); Tex. Code Ann. §§ 2.001, 6.204 (West 1997 & 2003); Utah Code Ann. §§ 30-1-2. -4.1 (West 1999 & 2004); Va. Code Ann. §§ 20-45.2, -45.3 (West 1997 & 2004); W. Va. Code Ann. § 48-2-603 (West 2001); Wis. Stat. Ann. §§ 765.001(2), 765.01 (West 2009); Wyo. Stat. Ann. § 20-1-101 (West 1977).

               In November 2012, mini-DOMAs in Maine, Maryland, and Washington were usurped by popular referenda legalizing same-sex marriage. See A Festive Mood in Maine as Same-Sex Marriage Becomes Legal, NY Times, Dec. 30, 2012, at A20, available at http://www.nytimes.com/2012/12/30/us/same-sex-marriage-becomes-legal-in-maine.html?_r=0; Many Weddings as Gay Marriage Becomes Legal in Md., USA Today (Jan. 1, 2013, 5:41 PM), http://www.usatoday.com/story/news/nation/2013/01/01/same-sex-marriage-maryland/1801917/; Ashley Fantz, Washington Voters Pass Same-Sex Marriage, CNN Projects, CNN (Nov. 9, 2012, 3:21 PM), http://www.cnn.com/2012/11/09/us/washington-passes-same-sex-marriage/index.html. Prior to November’s vote, Maine and Maryland had the following statutory DOMAs in place: Me. Rev. Stat. Ann. tit. 19-A, § 701(5) (1997); Md. Code Ann., Fam. Law § 2-201 (But see S.B. 241, 430th Gen. Assemb., Reg. Sess. (Md. 2012) (Civil Marriage Protection Act that defines marriage as between “two individuals,” rather than between “a man and a woman” as previously stated, effective Oct. 1, 2012).

     [13].     Mini-DOMAs limit marriage to one man and one woman, “but not necessarily the attributes of civil unions,” while super-DOMAs deny all forms of relationship recognition—civil unions, domestic partnerships, and reciprocal benefits—to same-sex couples. Daniel R. Pinello, Location, Location, Location: Same-Sex Relationship Rights by State, L. Trends & News: Prac. Area Newsl. (Am. Bar Ass’n), Fall 2009, available at http://www.americanbar.org/newsletter/publications/law_trends_news_practice_area_e_newsletter_home/bl_feat5.html.

     [14].     The following states have not enacted mini-DOMA legislation: Connecticut, Iowa, Massachusetts, New Hampshire, New Jersey, New Mexico (will recognize out-of-state same-sex marriages), New York, Rhode Island, Vermont, and the District of Colombia. The following states recognize civil unions between same-sex partners: Colorado (S.B. 11, 69th Gen. Reg. Assemb., 1st Sess. (Co. 2013); Delaware (S.B. 30, 146th Gen. Assemb., 1st Sess. (Del. 2011) (Delaware’s Civil Marriage Equality and Religious Freedom Act of 2013 converts civil unions to same-sex marriages.)adopting same-sex marriage, effective on July 1, 2013); Hawaii (H.B. 2569, 26th Leg., Reg. Sess. (Haw. 2012)), Illinois (750 Ill. Comp. Stat. Ann. 75/10 (2011)), New Jersey (N.J. Stat. Ann. § 37:1-28 (2007)), Rhode Island (R.I. Gen. Laws 1956 § 15-3.1.2 (2011)) (Rhode Island’s Marriage Equality Act of 2013 converts civil unions to same-sex marriages.). Vermont, in 2000, was an early adopter of civil unions. An Act Relating to Civil Unions, H. 847, 123rd Gen. Assemb., Reg. Sess. (Vt. 2000). See generally Human Rights Campaign, http://www.hrc.org/state_laws (last visited July 29, 2012) (tracking same-sex marriage legislation).

     [15].     The following four states provide the equivalent of state-level spousal rights to same-sex couples in the form of domestic partnerships: California, Cal. Fam. Code §§ 297–297.5 (2007); Nevada, Nev. Rev. Stat. § 122A (2009); Oregon, Or. Rev. Stat § 106.305 (2011); and Washington, Wash. Rev. Code § 26.60 (2009) (Referendum 74, passed by voters Nov. 2012 legalized same-sex marriage. Most existing domestic partnerships will be converted to marriages, effective June 30, 2014.). Two states provide limited spousal rights to same-sex couples: Colorado, Co. Rev. Stat. § 15-22-101 (2009) (calling partners “designated beneficiaries”)and Wisconsin, Wis. Stat. § 770 (2013) (State granted domestic partnerships are afforded inheritance and survivor protections, family and medical leave and hospital visitation rights, and exemptions from real estate transfer fees.).

     [16].     The following states issue marriage licenses to same-sex couples: Connecticut, Delaware, District of Columbia, Iowa, Maryland, Massachusetts, New Hampshire, New York, Rhode Island, and Vermont. See S.B. 899, 2009 Gen. Assemb., Reg. Sess. (Conn. 2009); Civil Marriage Equality and Religious Freedom Act of 2013, 147th Gen. Assemb., 1st Leg. Sess. (Del. 2013); Jury and Marriage Amendment Act of 2009, 56 D.C. Reg. 3797 (Aug. 7, 2009); Varnum v. Brien, 763 N.W.2d 862 (Iowa 2009); Civil Marriage Protection Act of 2012, 2012 Md. Laws ch. 2 (H.B. 438); Goodridge v. Dept. of Pub. Health, 440 Mass. 309 (Mass. 2003); 2009 N.H. Laws ch. 0059 (H.B. 436); Marriage Equality Act, 2011 N.Y. Laws A08354 ; 2013-S38, Gen. Assembly, Jan. Sess. (R.I. 2013); An Act Relating to Civil Marriage, S. 115, 2009–2010 Leg. Sess.(Vt. 2009). In response to stagnant legislatures—as seen in New Jersey, where the assembly passed a same-sex marriage bill in 2009, but the bill failed to pass in the state senate in 2010—voters began turning to referenda legalizing same-sex marriage. Recent referenda passed by narrow margins on the Nov. 2012 ballot in Maine (51.5% approve Question 1), Maryland (52.4% approve Question 6), and Washington (53.7% approve Referendum 74). See Marriage and Family Law on the Ballot, Ballotpedia, http://ballotpedia.org/wiki/index.php/Marriage_and_family_on_the_ballot (last modified June 29, 2013) (listing same-sex ballot measures by state, including Maine, Maryland, and Washington).

     [17].     The theory behind the analysis is that while federal legislation may have some effect on a state’s cultural consciousness, a state’s decision to enact DOMA would play a greater role in expressing the cultural values and desires of that state’s collective conscience and perhaps influence marital behavior. Likewise, a state’s close proximity to other states that have taken action may influence state behavior. Some states respond in kind to a neighboring state, or one in close proximity. For example, Massachusetts’s neighboring states—Vermont, New Hampshire, Connecticut, and New York—followed suit in permitting same-sex marriage. Supra note 16. Maine attempted to do so, but a referendum quickly overturned the legislation. See An Act to End Discrimination in Civil Marriage and Affirm Religious Freedom, 2009 Me. Laws 1020; November 3, 2009 General Election Tabulations, Bureau of Corps., Elections & Comm’ns, http://www.maine.gov/sos/cec/elec/2009/referendumbycounty.html (last visited Jul. 29, 2012). New Hampshire is debating overturning its licensing of same-sex marriage. See H.R. 437, 163rd Leg., 2d Reg. Sess. (N.H. 2011) (voted down 211-116 on Mar 21, 2012); Michael K. Lavers, N.H. Lawmakers Reject Marriage Equality Repeal Bill, Edge (Mar. 21, 2012), http://www.edgeboston.com/news/national/news/131180/nh_lawmakers_reject_marriage_equality_repeal_bill. On the other hand, in the days before Washington, D.C. permitted same-sex marriage, Maryland’s Attorney General Douglas Gansler issued an opinion forecasting that marriages valid in other jurisdictions would soon be recognized. Aaron C. Davis & John Wagner, Md. To Recognize Same-Sex Marriages From Other Places, Wash. Post (Feb. 25, 2010),http://articles.washingtonpost.com/2010-02-25/news/36919745_1_gansler-legal-opinion-couples. Two years later, Maryland would allow same-sex marriage with the Civil Marriage Act of 2012, but within months, a ballot referendum was certified for November to overturn the legislation. 2012 Petition Filings, Md. State Bd. of Elections, http://www.elections.state.md.us/petitions/index.html (last visited Aug. 15, 2012).

     [18].     Supra note 12. States with both constitutional amendments and statutes currently banning same-sex marriage are: Alabama, Alaska, Arizona, Arkansas, California, Colorado, Florida, Georgia, Idaho, Kentucky, Louisiana, Michigan, Mississippi, Missouri, Montana, North Carolina, North Dakota, Ohio, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, and Virginia. See Appendix 1.

               States with a constitutional amendment currently banning same-sex marriage, but having no matching statute, are Hawaii, Kansas, Mississippi, Nevada, and Oregon. Supra note 12.

               States with statutes currently banning same-sex marriage, but no constitutional provisions, are Delaware, Illinois, Indiana, Maine, Maryland, Minnesota, Pennsylvania, West Virginia, and Wyoming. Supra note 12; see also Appendix 1.

               States not banning same-sex marriage by either statute or constitutional amendment are Connecticut, District of Columbia, Iowa, Massachusetts, New Hampshire, New Jersey (Gov. Christie vetoed the Feb. 2012 same-sex marriage bill; the matter is stayed pending a public referendum on the Nov. 2012 ballot), New Mexico, Rhode Island, and Vermont. Supra note 12; see also Appendix 1.

     [19].     To create a meaningful “before” and “after” comparison to the states that enacted amendments, states with a statute or no statute had their marriage and divorce trends grouped between 1999–2004 and 2005–10.

     [20].     An independent sample mean t test compares two independent groups to determine whether the average measurement for a particular characteristic differs for these two groups. Tests of Means Statistics Workshops, Wadsworth Cengage Learning, http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/stat_workshp/test_means/test_means_15.html (last visited Oct. 12, 2013). Because the sample sizes are different and the variance differs from state to state in the distance the slope is from the mean of the slopes, I used the Welch-Satterthwaite equation to perform the t test.

     [21].     While all measurements discussed infra represent average or mean rates of decline, for ease of reading, the text uses the shorthand decline to represent this measurement.

     [22].     However, despite a trend of smaller declines in divorce rates prior to DOMA, amendment ratifications for non-DOMA states, on average, have lower divorce rates to begin with than DOMA states. It is important not to confuse the average divorce and marriage rates with the average rate of change in the divorce and marriage rates. Put another way, DOMA states prior to the passage of DOMA amendments had a greater rate of decline in divorce than non-DOMA states, but these states also have lower marriage rates and greater divorce rates than non-DOMA states. DOMA states start from a place of greater marital instability than non-DOMA states.

     [23].     I use the word preliminary because this trend data contains a maximum of ten years of analysis. More data is always ideal to truly capture whether trends are emerging. L. Langbein & M. Yost, Same-Sex Marriage & Negative Externalities, 90 Social Science Q. 292 (2009).

     [24].     Given how large the standard deviations were for each group of slopes, and the relatively small but inflexible sample size, achieving enough statistical power to find statistical significance would be incredibly challenging. I ran alternative analyses eliminating outliers in an attempt to decrease the standard deviation and increase the chance of detecting an effect should one exist. But even under the most conservative testing, the sample size must also decrease to accommodate eliminating outliers. Thus, the more compelling story is one of substantive significance rather than statistical significance. Statistical power refers to the possibility of making a Type II error, in which we conclude that no difference exists between the means of the two groups when one does. Social science, by convention, recommends no more than a .2 chance of this occurring. Statistical Power, Social Research Methods, http://www.socialresearchmethods.net/kb/power.php (last modified Oct. 20, 2006).

     [25].     See Massachusetts v. U.S. Dept. of Health and Human Services, 682 F.3d 1 (1st Cir. 2012) (articulating goals of enacting DOMA).