Toward a New Measure of State-Level LGBT Interest Group Strength

Abstract

Attempts to assess the influence of lesbian, gay, bisexual, and transgender (LGBT) groups on LGBT-related policy are hampered by imprecise measurement of LGBT group strength and activity. This research note examines the problems with existing measures of state-level LGBT advocacy strength and it develops an alternative measure of LGBT advocacy group strength. We utilize revenue and asset data available from the National Center for Charitable Statistics to develop better and reproducible state-level measures of LGBT interest group strength on an annual basis. We compare our measures to existing measures and demonstrate their utility for the ongoing study of LGBT politics. The approach used in construction of our measure can be extended over time, is replicable in other issue areas, and thus has broad utility for the study of interest groups at the subnational level.

Keywords

state policy LGBT rights gay rights interest groups measurement

State policy researchers often need to include measures of interest group strength in their models given the importance of these groups in the policy process (e.g., Wright 2003). For example, Mintrom and Vergari (1998) include teachers’ union strength in their study of education reform, Pacheco (2017) includes the smoker population in her study of tobacco policy making, and Karch et al. (2016) include a measure of professional associations in their study of interstate compacts. Likewise, much of the work on lesbian, gay, bisexual, and transgender (LGBT) policy in the states utilizes at least one variable to assess the strength of LGBT advocacy groups (Allen, Pettus, and Haider-Markel 2004; Haider-Markel 2001; 2010; Haider-Markel and Meier 1996; Kane 2003; Lewis and Jacobsmeier 2017). However, there are significant limitations to the current measures of LGBT interest group strength. Additionally, one of the most prominent measures of LGBT interest group strength faces an uncertain future given the expansion of same-sex marriage and changes in national policy since United States v. Windsor (2013), Obergefell v. Hodges (2015), and policies of the U.S. Census under the Trump administration. Here we briefly discuss common measures of LGBT interest group strength, highlight their limitations, and offer alternative measures.

Measures of LGBT Interest Group Strength

Attempts to measure LGBT advocacy strength require validity and reliability. In short, the variables should measure what we think that they are measuring and they should be reliable and consistent over time (Pollock 2009). Ideally, interest groups measures should also incorporate key political resources, such as financial strength and membership that groups can mobilize for political support (e.g., Wright 2003). From a practical standpoint, the data must also be readily available. Walker (1983) argues that the ability to secure financial resources is the key to the origin and maintenance of interest groups. As such nearly all existing measures of LGBT interest group strength are problematic.

Direct measures of state-level interest group strength have historically been difficult to construct due to data availability issues. Researchers have often relied on various proxy measures for LGBT groups that have a variety of problems. These include counts of gay bars, gay businesses, and gay newspapers per capita (Haider-Markel and Meier 1996; Kane 2003; Wald, Button, and Rienzo 1996), the percentage of unmarried same-sex partner households as noted by the Census Bureau (Barclay and Fisher 2003; Haider-Markel 2010), and attempts to gauge group financial resources by inferences to other sources (Haider-Markel 1997; Haider-Markel and Meier 1996: 2003). These approaches are flawed because they often do not vary over time, draw on sources are difficult to obtain, suffer from instrumentation problems (Cohn 2011; O’Connell and Feliz 2011), or suffer from comparability issues over time (O’Connell and Lofquis 2009; U.S. Census Bureau 2013).¹ Perhaps most importantly, because they are proxy measures, they ignore the ability of interest groups to actually organize and mobilize latent political resources, such as membership (Hansen 1985) or finances (Walker 1983). One attempt to directly measure state-level LGBT group strength (Taylor et al. 2012) was plagued by data limitations such as aspirational rather than actual group budgets and a lack of comprehensiveness. It was also held constant over time. Clearly, a new state-level measure of LGBT political strength is needed. The following sections discuss development of a possible alternative measure that systematically addresses the strength of LGBT rights groups in each state over time.

Data and Methods

We operationalize group strength in financial terms given that “money remains the mother’s milk of politics” (Loomis and Cigler 2007, 26). Money is one of the most important resources that advocates must obtain in politics. While it is not the only resource, it pays for a host of politically useful items such as full-time lobbyists, policy research, voter mobilization efforts, advertising, and outreach. Without money, it is difficult to mobilize other important resources, such as constituents.² In light of this, our measures are based around the revenues and assets of LGBT organizations that we aggregate at the state level. Focusing on revenues and assets of the LGBT groups provides increased content validity over population-based approaches for two reasons. First, Olson (1965) notes that not all interests in society mobilize. There might be differences in group mobilization in states as disparate as Alabama and California. Looking at the population of potential constituents ignores this issue. Second, Loomis and Cigler (2007, 5) argue, “interests with more resources (money, access, information and so forth) usually will obtain better results than interests that possess fewer assets and employ them less effectively.” Our measures directly address group resources.³ In addition, because financial resources are measured in monetary terms rather than self-identification with a socially marginalized community, where public acceptance and its own inclusiveness has varied over time (e.g., the exclusion of transgender individuals, gender nonconforming, bisexuals, or people of color), it is also likely that these measures are more reliable and comparable over time than are other approaches.

We obtained our measures of state LGBT group strength from aggregated revenue and asset data for LGBT nonprofit organizations collected from the National Center for Charitable Statistics (NCCS). This organization, in conjunction with the Internal Revenue Service (IRS), developed the dominant nonprofit organization classification system, the National Taxonomy of Exempt Entities (NTEE) (Barman 2013). The NTEE classification system rates nonprofit organizations on the primary aspect of their mission and this system is commonly used in research on nonprofit organizations (Tinkleman and Neely 2011).

Our revenue and asset data come from the NCCS collection of IRS Business Master Files (BMF) 1995–2015. These BMF files contain information about all active nonprofit organizations that have registered for tax-exempt status with the IRS and our data include organizations covered by tax code sections 501(c)3, 501(c)4 and other 501(c) types (National Center for Charitable Statistics n.d.). These groups are allowed to engage in varying types and varying amounts of political activity (e.g., lobbying or election-related activities) depending on which section of the tax code that they are organized under. For instance, 501(c)3 organizations face more restrictions on lobbying than do 501(c)4 organizations. Unfortunately, political action committees are not included in these data and that is a limitation.

When downloading the data, we obtained employment identification number, name, state, income, asset, revenue code subsection, and the National Taxonomy of Exempt Entities (NTEE) classification code. Because NCCS holdings are inconsistent as to whether data were collected annually, quarterly, or monthly, we queried the last record available in each calendar year. Our search terms included the following NTEE codes:

G81: AIDS

H81: AIDS Research

P88: LGBT Centers

R26: Lesbian and Gay Rights

We elected to include AIDS coded organizations because of the historic connection, in the United States context, between gay rights advocacy and the fight against HIV/AIDS. Because organizations are only afforded one NTEE code that is based on its major organizational purpose (National Center for Charitable Statistics 2008), we chose to expand our search for LGBT groups by looking for keywords in the organization name field. This included a variety of common LGBT rights–related terms like gay, lesbian, bisexual, and transgender. We consulted various editions of the Encyclopedia of Associations, Encyclopedia of Associations: Regional, State and Local Organizations, groups affiliated with the Equality Federation, groups that are members of Centerlink, and various other online sources to identify organizations with names not addressed by common keywords. Supplemental Appendix A further discusses this search and data cleaning. We chose not to include organizations that sometimes advocate on LGBT issues, such as the American Civil Liberties Union, because they do not focus primarily on the LGBT community. In total, our data cover 5,216 LGBT organizations that appeared in the BMF files. Because we look at annual data over time (1995–2015), these 5,216 organizations collectively produce a data set with 53,402 observations over 21 years. We categorized these organizations into a purpose based classification scheme that emerged from those data. Supplemental Appendix A provides a detailed description. Supplemental Appendix B provides a tutorial so that future researchers can apply our technique to develop other statewide measures of interest group strength and apply them to other policy domains.

Data collapse

To create annual state-level income and asset measures, the 53,402 observations and their income, asset, and organization type data (e.g., 501c3) were collapsed in Stata v.13 by state and year. This yielded a data set with 1,071 observations (50 states + DC × 21 years = 1,071). Because some types of LGBT-related groups (see Supplemental Appendix A for further discussion), such as sports leagues, LGBT friendly religious congregations, and AIDS groups might not be specifically or narrowly involved in LGBT rights advocacy, a separate collapse on the same variables but without the organization types—culture and sport, AIDS-related, and religious congregations—was performed. After dropping these types of groups, this left 20,682 observations for 2,243 organizations. We collapsed the data by state and year, yielding a separate data set of 1,071 observations. We then adjusted the revenue and asset data in both data sets to 2009 dollars and put them in per capita terms as noted in Supplemental Appendix A.⁴

The data in Table 1 show that in 2015, the wealthiest state in terms of real assets per capita is Colorado. This was true for both data sets. This is due to the outsize impact of the extremely wealthy Gill Foundation. Other outliers include the Arcus Foundation in NY/MI, and Pride Foundation in Washington. Otherwise, as would be expected, the wealthy states include many of the usual suspects, such as New York and California. Some states have no assets reported by LGBT groups. These include South Dakota in both data sets and Idaho, Kansas, Mississippi, Montana, and Arkansas in the smaller data set that lacks HIV/AIDS groups, cultural organizations, and religious groups. With respect to revenues, New York based LGBT groups bring in the most income. This is true in both the large data set and the small data set. Because of the impact of outliers on the combined state LGBT organization revenues and assets, we attempt to mitigate this issue by using a square root transformation of our real state totals for revenue and assets per capita in subsequent analyses.⁵

Table 1.

State LGBT Group Revenue and Asset Data 2015.

State	Assets per capita (small)	Revenue per capita (small)	Assets per capita (large)	Income per capita (large)
Alabama	$0.00	$0.00	$2.06	$5.27
Alaska	$0.31	$0.56	$3.55	$3.95
Arizona	$0.06	$0.35	$2.02	$2.76
Arkansas	$0.00	$0.00	$0.10	$0.22
California	$3.39	$3.39	$12.85	$26.72
Colorado	$40.20	$8.31	$41.12	$10.88
Connecticut	$0.18	$0.40	$1.94	$3.08
Delaware	$4.05	$1.14	$5.50	$5.12
Florida	$0.46	$0.55	$3.30	$4.32
Georgia	$0.05	$0.11	$0.27	$1.21
Hawaii	$0.07	$0.10	$1.27	$4.07
Idaho	$0.00	$0.00	$0.46	$0.26
Illinois	$1.90	$0.64	$5.02	$3.53
Indiana	$0.06	$0.11	$0.74	$1.16
Iowa	$0.06	$0.29	$0.06	$0.31
Kansas	$0.00	$0.00	$0.43	$0.36
Kentucky	$0.28	$0.13	$0.68	$1.41
Louisiana	$0.06	$0.02	$8.28	$14.43
Maine	$0.09	$0.32	$0.48	$2.35
Maryland	$0.18	$0.17	$6.15	$12.95
Massachusetts	$0.91	$1.68	$3.14	$6.75
Michigan	$0.48	$0.32	$0.88	$1.16
Minnesota	$0.56	$0.82	$1.36	$2.18
Mississippi	$0.00	$0.00	$0.27	$0.29
Missouri	$0.03	$0.19	$1.67	$3.46
Montana	$0.00	$0.00	$0.19	$0.49
Nebraska	$0.02	$0.04	$0.29	$0.67
Nevada	$2.00	$0.88	$5.53	$5.50
New Hampshire	$0.00	$0.00	$0.74	$1.88
New Jersey	$0.03	$0.09	$2.25	$4.67
New Mexico	$0.09	$0.30	$2.06	$1.89
New York	$13.96	$14.37	$33.72	$36.78
North Carolina	$0.18	$0.31	$0.86	$1.64
North Dakota	$0.09	$0.10	$0.09	$0.10
Ohio	$0.21	$0.25	$0.80	$1.42
Oklahoma	$0.28	$0.23	$1.22	$1.39
Oregon	$1.16	$1.05	$1.79	$2.62
Pennsylvania	$0.33	$0.35	$1.88	$3.60
Rhode Island	$0.28	$1.04	$4.25	$4.15
South Carolina	$0.06	$0.14	$2.39	$3.26
South Dakota	$0.00	$0.00	$0.00	$0.00
Tennessee	$0.06	$0.07	$0.87	$2.17
Texas	$0.28	$0.31	$2.40	$3.87
Utah	$0.89	$1.14	$0.97	$1.41
Vermont	$0.17	$0.88	$0.65	$3.14
Virginia	$0.36	$0.45	$0.94	$2.36
Washington	$5.86	$1.43	$7.81	$6.79
West Virginia	$0.08	$0.26	$0.65	$0.77
Wisconsin	$0.35	$0.53	$2.68	$8.32
Wyoming	$0.82	$1.21	$1.17	$1.49
U.S. Average	$1.62	$0.90	$3.60	$4.37

Note. Compiled by the authors from Internal Revenue Service data. LGBT = lesbian, gay, bisexual, and trans.

Figure 1 presents the mean square root LGBT assets per capita from our smaller data set, Figure 2 presents the mean square root LGBT revenue per capita from our smaller data set, and Figure 3 presents same-sex partner household percentage by state from the 2010 Census in graphical form so that we can observe distributions across the country. Not surprisingly, the densest collections of assets, revenue, and same-sex households occur in the far West, upper Mid-West, and Northeastern states, with the lowest representations in the deep South.

Figure 1.

Mean state per capita LGBT non-profit group assets, square root.

Figure 2.

Mean state per capita LGBT non-profit group revenue, square root.

Figure 3.

Percentage of same-sex partner households by state, U.S. Census 2010.

Our analyses include (1) a Pearson correlation with existing measures of LGBT rights strength, (2) a series of OLS regressions where the dependent variable is a count of positive laws in the Human Rights Campaign’s 2015 State Equality Index (Human Rights Campaign 2016), and (3) a series of event history models (1995–2015) where the dependent variable is a dichotomous indicator of passage of a sexual orientation inclusive employment nondiscrimination law (National Gay and Lesbian Task Force 2014). The latter two sets of analyses deploy our measures to demonstrate their potential utility in state politics research.

In the bivariate analyses, we correlate each of our new measures with same-sex partner households to gauge the relationship between our real per capita asset and revenue variables and this commonly used variable in state-level LGBT policy research.⁶ We also correlate our 1995 to 2015 LGBT group income and asset per capita measures with a time varying LGBT population measure that was derived from the relationship between two measures of state LGBT population (Gates 2017; Gates and Newport 2013) and the proportion of same-sex partner households in a state.⁷

For the multivariate analyses, we include as controls the updated measures of citizen and government ideology (nominate) from Berry et al. (1998).⁸ For the cross-sectional analysis (as of 2015), we take the mean of these figures to determine the average citizen and government ideologies.⁹ In addition, we include controls for whether the state is in the South, the percentage of the population that is evangelical or Church of Jesus Christ of Latter Day Saints identified (Pew Research Center 2014), the presence of direct democracy (e.g., Lewis 2011), and a measure of public support for sexual orientation inclusive nondiscrimination laws (Lax and Phillips 2009) in both sets of models.¹⁰ In both analyses, we also include a real (1995 dollars) state per capita income variable to control for wealth disparities between states (Bureau of Economic Analysis 2018).¹¹ In the cross-sectional analyses, this was averaged over the years 1995 to 2015. In the event history models, we include a diffusion variable that addresses the percentage of the U.S. Census defined region that has passed a sexual orientation inclusive nondiscrimination law in the year in question.

Analyses

Table 2 displays the Pearson correlations between our revenue and asset per capita measures. There is a moderately strong and positive correlation between all of our revenue and asset per capita measures and the percentage of the state that is LGBT identified and the percentage of households with same-sex partners. One would expect there to be a positive correlation given that LGBT groups represent LGBT communities. However, the moderate level of the correlation could be indicative of varying abilities of the groups to mobilize the latent resources in the states. Unsurprisingly there are strong correlations between our revenue and asset per capita measures. In multivariate models, researchers would need to use a single revenue or asset per capita measure because of multicollinearity concerns.

Table 2.

Correlations Between Competing Measures of Interest Group Strength (N = 1,050).

	A-sm	R-sm	a-lg	R-lg	Same-sex partner households	LGBT
Assets per capita Square root (small)	1.00
Revenue per capita Square root (small)	.80	1.00
Assets per capita Square root (large)	.86	.78	1.00
Revenue per capita Square root (large)	.54	.74	.83	1.00
Same-sex partner households	.40	.52	.47	.53	1.00
LGBT percentage	.33	.44	.36	.40	.92	1.00

Note. LGBT = lesbian, gay, bisexual, and transgender.

In our cross-sectional OLS models, the dependent variable is a count of pro-LGBT policies from HRC’s 2015 State Equality Index.¹² We estimated all models with robust standard errors to control for heteroscedasticity. The results reported in Table 3 suggest that the asset per capita measures from both data sets (models 1 and 2) are statistically significant and positively related to number of pro-LGBT laws in the state. Turning to the revenue per capita measures, in model 3, we find that LGBT group revenue per capita in the smaller data set (that lacks HIV, religious, and cultural groups) is positively associated with the dependent variable. For the larger data set (model 4), LGBT revenue per capita is also statistically significant at a more charitable p <. 10 level. Across all models, government ideology is a statistically significant predictor of pro-LGBT laws. Legislatures that are more liberal are more likely to adopt more pro-LGBT policies. In addition, Southern states are far less likely to adopt these measures.

Table 3.

Predicting State Index of LGBT Rights Policies.

	Model 1	Model 2	Model 3	Model 4
Assets per cap—small	4.361*** (1.132)
Assets per cap—large		3.896** (1.169)
Revenue per cap—small			8.225** (3.108)
Revenue per cap—large				4.345^† (2.179)
Citizen ideology	.016(0.171)	.031(0.168)	−.048(0.161)	−.013(0.172)
Legislature ideology	.323** (0.096)	.301** (0.093)	.335*** (0.090)	.294** (0.094)
Direct democracy	−1.983(2.892)	−1.249(2.826)	−2.102(2.703)	−.718(2.813)
South	−8.593** (2.957)	−9.661** (2.929)	−8.513** (2.872)	−10.885*** (3.151)
Evangelical pop	−.124(0.211)	−.006(0.211)	−.061(0.196)	.124(0.237)
Jobs Lax and Phillips	−.131(0.497)	.010(0.478)	.003(0.443)	.274(0.494)
Per cap income	.001^† (0.001)	.001(0.001)	.001(0.001)	.001(0.001)
Constant	−6.127(33.89)	−16.659(33.109)	−13.478(30.749)	−35.141(33.543)
F	20.47**	16.31**	13.87**	13.53**
R ²	.72	.71	.73	.70

Note. All models include 50 cases. ordinary least squares standard errors are robust standard errors in parentheses.

†

p < .1. *p < .05. **p < .01. ***p < .001.

To assess the relative impact of the independent variables, we calculated the marginal effect of a one standard deviation change in the interest group measures on the index score. For comparison, we also calculated the marginal effect of a one standard deviation change in the legislature ideology measure. For example, a one standard deviation change in the small assets measure is associated with a 3.8-point change on the policy index while a one standard deviation change in the legislature ideology variable is associated with a 6.1-point change on the index. Clearly, legislature ideology has a greater substantive impact on LGBT friendly policy, but interest group assets play a meaningful role.¹³

To check for the possibility that our results are sensitive to the presence of an outlier, despite the transformation of the revenue and asset measures via a square root procedure, we dropped Colorado from the analysis (see Table A1 in the Supplemental Online Appendix). In this set of models, the per capita asset measure the small data set remained statistically significant and positively associated with the adoption of pro-LGBT policies. The assets per capita measure from the large data set and the per capita revenue measure from the small data set did as well if one uses a more charitable p < .10 standard. The per capita revenue from the large data set was not statistically significant at even the more relaxed standard.

Table 4 presents the findings from our series of event history analyses. The dependent variable is the adoption of a sexual orientation employment discrimination law. We use the same controls as in the cross-sectional analysis but add a time varying regional diffusion variable, the percentage of the region that has adopted a gay inclusive employment nondiscrimination law. We also allow citizen ideology, institutional ideology, and per capita income to vary over time. Because our data cover 1995 to 2015, states that adopted such laws prior to 1995 are removed from the data set leaving 41 subjects with 13 failure events. Model 1 shows the real assets per capita measure from the smaller data set is positively related to the adoption of gay inclusive employment nondiscrimination laws. The only other statistically significant variable is legislature ideology. The magnitude of the hazard ratios for each variable is not directly comparable in a Cox model.

Table 4.

Event History Analysis: State Adoption of Sexual Orientation Antidiscrimination in Employment.

	Model 1	Model 2	Model 3	Model 4
Assets per cap—small	1.878* (0.582)
Assets per cap—large		1.754^† (0.571)
Revenue per cap—small			1.123(0.566)
Revenue per cap—large				1.009(0.543)
Citizen ideology	1.014(0.048)	1.012(0.048)	1.010(0.045)	1.010(0.045)
Legislature ideology	1.080** (0.031)	1.073** (0.028)	1.059* (0.025)	1.059* (0.025)
Direct democracy	0.440(0.360)	0.499(0.391)	0.698(0.527)	0.724(0.540)
South	0.240(0.291)	0.243(0.291)	0.331(0.389)	0.330(0.398)
Evangelical pop	0.936(0.093)	0.966(0.092)	0.976(0.095)	0.975(0.097)
Jobs Lax and Phillips	0.781(0.160)	0.838(0.165)	0.897(0.187)	0.898(0.190)
Per cap income	1.000(0.0002)	1.000(0.0002)	1.000^† (0.0002)	1.000^† (0.0002)
Regional diffusion	10.032(17.336)	9.755(17.313)	8.370(14.239)	8.236(14.034)
Log likelihood	−30.029	−30.600	−31.968	−31.994
Chi-square	32.47**	31.33**	28.60**	28.54**

Note. All models include 689 cases. Cox regression for event history with hazard ratios; hazard ratios greater than 1 indicate an increase in the likelihood of adoption and less than 1 indicates a decrease in the likelihood of adoption; standard errors are in parentheses.

†

p < .1. *p < .05. **p < .01. ***p < .001.

Our interest group variables do little to improve the other models. Model 2 shows that the per capita assets measure from the large data set is not significant at traditional levels but it is if one uses a more charitable standard (p < .1). Models 3 and 4 show that real per capita LGBT group revenues are not statistically significant from either data set. As with the cross-sectional analyses, government ideology is consistently associated with of the adoption of a gay inclusive employment nondiscrimination law across our models.

Discussion and Conclusion

Existing measures of LGBT advocacy group strength suffer from problems associated with validity, reliability, and availability. These problems persist with population-based methods of measuring LGBT advocacy strength, such as those from the U.S. Census and American Community Survey. These measures suffer from reliability issues due to changes in instrumentation, time invariance of the census data, and the rise of same-sex marriage. They also only measure the latent group rather than actual advocacy organizations and their resources. For this reason, we propose new measures. Our new measures are based in interest groups’ financial assets or revenue per capita. These measures vary across states and over time, sidestepping problems with existing measures that are often held constant in time series analysis.

In our analyses, we consistently found that assets per capita in our data set that lacked HIV, cultural, and religious groups was a statistically significant predictor of LGBT rights laws. Revenue per capita did not perform as well. Perhaps the surprising performance of assets over revenues in our models speaks to stronger organizational management capacity and greater organizational stability in advocacy groups with more assets than those with less. While raising money is important, groups must also take care not to waste their resources. Assets might offer protection against unforeseen events. In addition, revenues might be far more sensitive to rare events like ballot initiatives that fill the coffers, large grants, or economic downturns. We examined the mean state asset per capita and state revenue per capita data to see how they were affected during the recent economic downturn. The mean state assets per capita were far less affected (loss of 7.6% of value) than were mean state revenues per capita (loss of 14% of value) between 2010 and 2011. Regardless of whether one uses assets per capita or revenues per capita, our measures have greater content validity than population-based approaches of capturing interest group strength because we assess the ability of organizations to extract and conserve a politically relevant resource (money) from their environment. Money is vital in politics.

Measuring the strength of interest groups is a difficult task and researchers have relied upon a host of proxy measures. Our analysis suggests that researchers should consider the use of publicly available IRS business master file data on nonprofit organizations as a proxy for interest group strength. These data contain annual revenue and asset data for nonprofit organizations that file tax form 990s. The NCCS provides access to these data and they also classify by organization type. The group data can be examined at the organization level, aggregated to the city, state, metropolitan statistical area, and zip code levels, as well as be explored over time. Although the specific measures discussed in this article are perhaps only of interest to scholars examining LGBT politics and policy, the applicability of similarly constructed measures could easily extend to other types of interest groups. This is important because while LGBT rights scholars have long had a potential data source to approximate LGBT advocacy strength, not all types of interest groups are based around identities directly or indirectly tracked by the Census Bureau. The census does not track the number of environmental groups for instance. The IRS data contain information on all interest groups that file under this part of the tax code. This accessible data include environmental groups, women’s groups, religious organizations, trade groups and so forth. We believe these types of revenue or asset per capita measures are useful in other issue areas or types of studies and we have demonstrated that there is potential value in this data source for state politics scholars.

Supplemental Material

Supplemental_Appendix_A-Online_(1) – Supplemental material for Toward a New Measure of State-Level LGBT Interest Group Strength

Supplemental material, Supplemental_Appendix_A-Online_(1) for Toward a New Measure of State-Level LGBT Interest Group Strength by Jami K. Taylor, Donald P. Haider-Markel and Benjamin Rogers in State Politics & Policy Quarterly

Supplemental Material

tutorial_Appendix_B-Online_(1) – Supplemental material for Toward a New Measure of State-Level LGBT Interest Group Strength

Supplemental material, tutorial_Appendix_B-Online_(1) for Toward a New Measure of State-Level LGBT Interest Group Strength by Jami K. Taylor, Donald P. Haider-Markel and Benjamin Rogers in State Politics & Policy Quarterly

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received $1,000 from the University of Toledo in support for the research, authorship, and/or publication of this article.

Notes

Supplemental Material

Supplemental material for this article is available online.

ORCID iD

Donald P. Haider-Markel

Author Biographies

Jami Taylor is a professor of political science and public administration at the University of Toledo.

Donald Haider-Markel is a professor of political science at the University of Kansas. Along with Jami Taylor, he edited Transgender Rights and Politics: Groups, Issue Framing, and Policy Adoption (University of Michigan Press, 2014). With Daniel Lewis, they are the authors of The Remarkable Rise of Transgender Rights (University of Michigan Press, 2018).

Benjamin Rogers is assistant professor of political science at Norfolk State University.

References

Allen

Mahalley

Pettus

Carrie

Haider-Markel

Donald P.

2004. “Making the National Local: Specifying the Conditions for National Government Influence on State Policymaking.” State Politics & Policy Quarterly 4 (3): 318–44.

Ballotpedia. 2018. “States.” https://ballotpedia.org/States#Initiative_.26_referendum_in_the_U.S (accessed January 29, 2018).

Barclay

Scott

Fisher

Shauna

. 2003. “The States and the Differing Impetus for Divergent Paths on Same-Sex Marriage, 1990-2001.” Policy Studies Journal 31 (3): 331–52. doi:10.1111/1541-0072.00025.

Barman

Emily

. 2013. “Classification Struggles in the Nonprofit Sector: The Formation of the National Taxonomy of Exempt Entities, 1969–1987.” Social Science History 37 (1): 103–41.

Berry

William

Ringquist

Evan

Fording

Richard

Hanson

Russell

. 1998. “Measuring Citizen and Government Ideology in the American States, 1960-93.” American Journal of Political Science 42 (1): 327–48.

Bureau of Economic Analysis. 2018. “SA1 Personal Income Summary: Personal Income, Population, per Capita Personal Income.” https://www.bea.gov/iTable/index_regional.cfm (accessed January 29, 2018).

Cohn

D’Vera

. 2011. “Census Bureau Flaws in Same-Sex Couple Count.” Social & Demographic Trends. Pew Research Center. http://www.pewsocialtrends.org/2011/09/27/census-bureau-flaws-in-same-sex-couple-data/ (accessed March 21, 2014).

Erikson

Robert S.

Wright

Gerald C.

McIver

John P.

1993. Statehouse Democracy: Public Opinion and Policy in the American States. Cambridge: Cambridge University Press.

Fording

Richard

. 2015. “Updated Measures of Citizen and Government Ideology.” https://rcfording.wordpress.com/state-ideology-data/ (accessed January 24, 2018).

10.

Gates

Gary

. 2017. “Vermont Leads States in LGBT Identification.” Gallup News, February. http://news.gallup.com/poll/203513/vermont-leads-states-lgbt-identification.aspx (accessed January 24, 2018).

11.

Gates

Gary

Newport

Frank

. 2013. “Gallup Special Report: New Estimates of the LGBT Population in the United States.” The Williams Institute.” http://williamsinstitute.law.ucla.edu/research/census-lgbt-demographics-studies/gallup-lgbt-pop-feb-2013/ (accessed February 15, 2014).

12.

Haider-Markel

Donald P

. 1997. “Interest Group Survival: Shared Interests versus Competition for Resources.” Journal of Politics 59 (3): 903–12.

13.

Haider-Markel

Donald P.

2001. “Policy Diffusion as a Geographical Expansion of the Scope of Political Conflict: Same-Sex Marriage Bans in the 1990s.” State Politics & Policy Quarterly 1 (1): 5–26.

14.

Haider-Markel

Donald P.

2010. Out and Running: Gay and Lesbian Candidates, Elections, and Policy Representation. Washington, DC: Georgetown University Press.

15.

Haider-Markel

Donald P.

Meier

Kenneth J.

1996. “The Politics of Gay and Lesbian Rights: Expanding the Scope of the Conflict.” Journal of Politics 58 (2): 332–49. doi:10.2307/2960229.

16.

Haider-Markel

Donald P.

Meier

Kenneth J.

2003. “Legislative Victory, Electoral Uncertainty: Explaining Outcomes in the Battles over Lesbian and Gay Civil Rights.” Review of Policy Research 20 (4): 671–90.

17.

Hansen

John M.

1985. “The Political Economy of Group Membership.” The American Political Science Review 79 (1): 79–96.

18.

Human Rights Campaign. 2016. “2015 State Equality Index.” https://assets2.hrc.org/files/assets/resources/SEI-2015-FullReport.pdf (accessed January 24, 2018).

19.

Kane

Melinda D.

2003. “Social Movement Policy Success: Decriminalizing State Sodomy Laws, 1969-1998.” Mobilization 8 (3): 313–34.

20.

Karch

Andrew

Nicholson-Crotty

Sean C.

Woods

Neal D.

Bowman

Ann O’M.

2016. “Policy Diffusion and the Pro-innovation Bias.” Political Research Quarterly 69 (1): 83–95.

21.

Lax

Jeffrey

Phillips

Justin

. 2009. “Gay Rights in the States: Public Opinion and Policy Responsiveness.” American Political Science Review 103 (3): 367–86.

22.

Lewis

Daniel C.

2011. “Direct Democracy and Minority Rights: Same-Sex Marriage Bans in the U.S. States.” Social Science Quarterly 92 (2): 364–83.

23.

Lewis

Daniel C.

Jacobsmeier

Matthew L.

2017. “Evaluating Policy Representation with Dynamic MRP Estimates: Direct Democracy and Same-Sex Relationship Policies in the United States.” State Politics & Policy Quarterly 17 (4): 441–64.

24.

Loomis

Burdett

Cigler

Allan

. 2007. “Introduction: The Changing Nature of Interest Group Politics.” In Interest Group Politics, eds. Cigler

Allan

Loomis

Burdett

, 7th ed. Washington, DC: CQ Press, 1–33.

25.

Mintrom

Michael

Vergari

Sandra

. 1998. “Policy Networks and Innovation Diffusion: The Case of State Education Reforms.” The Journal of Politics 60 (1): 126–48.

26.

National Center for Charitable Statistics. 2008. “How and by Whom Are NTEEs Assigned?” https://web.archive.org/web/20150526090551/http://nccsweb.urban.org/knowledgebase/detail.php?linkID=728 (accessed April 28, 2017).

27.

National Center for Charitable Statistics. n.d. “NCCS All Registered Nonprofits Table Wizard.” http://nccs-data.urban.org/data.php?ds=bmf.

28.

National Gay and Lesbian Task Force. 2014. “State Nondiscrimination Laws in the U.S.” https://web.archive.org/web/20180310192030/http://www.thetaskforce.org/static_html/downloads/reports/issue_maps/non_discrimination_5_14_new.pdf (accessed January 11, 2018).

29.

Obergefell v. Hodges 576 U.S.____(2015).

30.

O’Connell

Martin

Feliz

Sarah

. 2011. “Same-Sex Couple Household Statistics from the 2010 Census.” Fertility and Family Statistics Branch, Social, Economic and Housing Statistics Division U.S. Bureau of the Census. https://www.census.gov/library/working-papers/2011/demo/SEHSD-WP2011-26.html (accessed March 21, 2014).

31.

O’Connell

Martin

Lofquis

Daphne

. 2009. “Changes to the American Community Survey between 2007 and 2008 and Their Potential Effect on the Estimates of Same-Sex Couple Households.” U.S. Census Bureau. https://www.census.gov/population/www/socdemo/files/changes-to-acs-2007-to-2008.pdf (accessed March 21, 2014).

32.

Olson

Mancur

. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge: Harvard University Press.

33.

Pacheco

Julianna

. 2017. “Free-Riders or Competitive Races? Strategic Interaction across the American States on Tobacco Policy Making.” State Politics & Policy Quarterly 17 (3): 299–318.

34.

Pew Research Center. 2014. “Religious Landscape Study.” http://www.pewforum.org/religious-landscape-study/ (accessed January 24, 2018).

35.

Pollock

Philip

. 2009. The Essentials of Political Analysis. 3rd ed. Washington, DC: CQ Press.

36.

Sullivan

Andrew

. 2005. “The End of Gay Culture: Assimilation and its Meaning.” The New Republic, October 24. http://www.newrepublic.com/article/politics/the-end-gay-culture (accessed February 15, 2014).

37.

Taylor

Jami

Lewis

Daniel

Jacobsmeier

Matthew

DiSarro

Brian

. 2012. “Content and Complexity in Policy Reinvention and Diffusion: Gay and Transgender-Inclusive Laws against Discrimination.” State Politics & Policy Quarterly 12 (1): 75–98. doi:10.1177/1532440011433589.

38.

Thomas

June

. 2011. “The Gay Bar: Can It Survive?” Slate, July 1. http://www.slate.com/articles/life/the_gay_bar/2011/06/the_gay_bar.html (accessed February 15, 2014).

39.

Tinkleman

Daniel

Neely

Daniel

. 2011. “Some Econometric Issues in Studying Nonprofit Revenue Interactions Using NCCS Data.” Nonprofit and Voluntary Sector Quarterly 40 (4): 751–61.

40.

United States v. Windsor 570 U.S. _____(2013).

41.

U.S. Census Bureau. 2013. “Frequently Asked Questions about Same-Sex Partner Households.” https://www2.census.gov/topics/families/same-sex-couples/faq/sscplfactsheet-final.pdf (accessed April 28, 2017).

42.

Wald

Kenneth

Button

James

Rienzo

Barbara

. 1996. “The Politics of Gay Rights in American Communities: Explaining Antidiscrimination Ordinances and Policies.” American Journal of Political Science 40 (4): 1152–78. doi:10.2307/2111746.

43.

Walker

Jack

. 1983. “The Origins and Maintenance of Interest Groups in America.” American Political Science Review 77 (2): 390–406.

44.

Wright

John

. 2003. Interest Groups and Congress: Lobbying, Contributions and Influence. New York: Longman.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB

0.02 MB