"Be Our Guest (or Not!): The Role of Public Opinion inProviding Services to Same-Sex Couples" by Lian Peach

Be Our Guest (or Not!): The Role of Public Opinion in

Providing Services to Same-Sex Couples

Lian Peach, Salisbury University

Abstract: “All we want is equality.” These are the cries and pleas of minoritized and marginalized communities all across America. In the past few decades, an immense spotlight has been on the LGBTQIA+ community, with same-sex marriage, same-sex couples’ rights, and transgender rights being front and center. This community faces intense and hateful opposition from individuals in a “traditional culture” context and community, who cite morals and personal beliefs as valid reasons for discrimination and hate. Recently, same-sex couples have been facing challenges from businesses that refuse to provide services to them, with the 2017 U.S. Supreme Court case Masterpiece Cakeshop v. Colorado Civil Rights Commission as primary evidence for the growing intolerance in society for LGBTQIA+ individuals. While these perspectives are part of the broader societal debate, they are often at odds with growing calls for equal rights and anti-discrimination policies, which seek to ensure fairness and inclusion for all citizens, regardless of sexual orientation. This paper aims to explore the factors behind why some Americans are against businesses providing their services to same-sex couples. The data set used for analysis is the American National Election Studies 2020 Time Series Study. Taking in the following results and findings, the implications of this research are far-reaching and stand as a strong jumping-off point for future research on the rights and discrimination that same-sex couples face in society, and especially in public accommodation settings. Especially in light of the upcoming second term of the Trump administration, many LGBTQIA+ individuals are fearful for what this might mean for the expansion of future legal rights and what restrictions may be placed on current legal rights. Many are affected by these rights, or the lack thereof, but as one can clearly see, all they want — and all they deserve — is equality.

---

Introduction

“All we want is equality.” These are the cries and pleas of minoritized and marginalized communities all across America. In the past couple of decades, immense focus and spotlight have been on the LGBTQIA+ community, with examples such as same-sex marriage, same-sex couples’ rights, and transgender rights. This community, much more than other historically marginalized communities, seems to face intense and even hateful opposition, often from individuals in a ‘traditional’ culture, context, and community, who go on to cite morals and personal beliefs as valid reasons for discrimination and hate. Recently, same-sex couples have been facing challenges from businesses that refuse to provide services to them – usually small, self-owned businesses that work in the wedding industry. The 2018 U.S. Supreme Court case Masterpiece Cakeshop Ltd. v. Colorado Civil Rights Commission, which set the precedent that certain rights seem to be weighted more than others, is primary evidence for the growing intolerance in society for LGBTQIA+ individuals. In this case, the right to religious freedom and expression prevailed, outweighing the importance of anti-discrimination laws and sexual orientation as a protected class.

From a public opinion standpoint, some Americans support this refusal to provide services to same-sex couples due to a combination of deeply rooted cultural, religious, and personal beliefs. For some, traditional views on marriage, often influenced by religious teachings, see marriage as a union exclusively between a man and a woman. They may perceive accommodating same-sex couples as conflicting with their values and beliefs about family and morality. Additionally, there is a fear that societal norms are shifting too quickly, leading to concerns about the erosion of what they consider to be time-honored traditions. While these perspectives are part of the broader societal debate, they are often at odds with growing calls for equal rights and anti-discrimination policies, which seek to ensure fairness and inclusion for all citizens, regardless of sexual orientation.

This paper aims to explore the factors behind why some Americans are against businesses providing their services to same-sex couples. Though it is a little explored factor, the author assumes that if Americans have more confidence in the Republican Party to better handle the national economy, they will be less likely to favor regulations requiring businesses to serve same-sex couples. The author predicts that this phenomenon is due to the fact that those who believe that the Republican Party is better at handling economic policy do not often have an accurate understanding of fiscal and monetary policies and either willingly or unwittingly listen to misinformation and disinformation, since the idea that the economy thrives much better under a Republican administration has been proven false (Blinder & Watson, 2016). Because of this, these individuals seem more likely to support discriminatory policies that would allow businesses to refuse to serve same-sex couples than others.

Literature Review

Same-sex couples and marriage have been an extremely controversial topic over the past decade or so. Even before the ruling in Obergefell v. Hodges was decreed, legalizing same-sex marriage, there was strong, vocal opposition to marriages that were viewed as ‘non-traditional,’ those that were not one cisgender man and one cisgender woman of the same race, from a multitude of standpoints. One of the most commonly cited reasons for opposing non-traditional marriages was due to conflict with an individual’s religious beliefs (Thomas, 2018). Even with increasingly favorable attitudes toward same-sex couples, there is still a large portion of the population against supporting these couples’ rights (Gallup, 2025; Lee & Mutz, 2019). Following Obergefell v. Hodges in 2015 was Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission in 2018, which debated public accommodation rights for same-sex couples in contrast to freedom of religion. Research shows that the main factors for a rise in public support include a decline in religiosity, increasing education levels, and polarization in cultural and morality policy viewpoints (Gaines & Garand, 2010; Lee & Mutz, 2019). 

In the same vein, since the legalization of same-sex marriage, requirements for providing services to same-sex couples, often for weddings or similar celebrations, have been greatly debated. A study found that a surprising 53% of respondents support a refusal of service by businesses to gay couples (Powell et al., 2017). Perhaps even more surprising, about 50% of same-sex marriage supporters favor service refusal by self-employed business owners, regardless of the owners’ reasons for refusing service (Powell et al., 2017). Clearly, whether or not an individual supports same-sex marriage is neither an accurate nor a trustworthy factor alone to determine if they will support the refusal of service to same-sex couples by corporate and self-employed businesses.

While there is substantial research and evidence exploring public opinion and attitudes towards same-sex marriage as a whole, and even then, public opinion and elite discourse around the LGBTQIA+ community seem to continuously fluctuate, few studies have discussed the implications of businesses refusing services to same-sex couples. In 2017, the U.S. Supreme Court (SCOTUS) heard arguments for Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission, which sparked this recent division between antidiscrimination laws and religious freedoms. The case focused on whether Colorado’s public accommodations and antidiscrimination law can require a baker to make a cake for a same-sex couple when same-sex marriage conflicted with his religious beliefs. The next year, with a 7-2 decision, the SCOTUS Justices sided with Masterpiece Cakeshop in favor of religious freedom (Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission, 2018; Thomas, 2018). This decision underscores the severe lack of acknowledgement and regard for the rights that same-sex couples have and should be afforded. In essence, the SCOTUS determined that the importance of freedom of religion outpaces the importance of adherence to anti-discrimination laws, the Equal Protection Clause granted in the 14^th Amendment, and the status of sexual orientation being among other identities that are protected classes.

In fact, through this elite discourse from the SCOTUS ruling and from conservative news media that covered this case, public attitudes shifted toward favoring religious liberty at the expense of the rights of same-sex couples. Proponents of staunch freedoms of religion often argue that exemptions should be granted for business owners who do not recognize same-sex marriages as legal and moral (Thomas, 2018). However, opponents to this, who believe such exemptions would go too far and infringe on the rights of same-sex couples, argue that businesses that are open to the public must comply with anti-discrimination laws. Much like racial discrimination would be illegal and consequences for the business would be upheld, the same protections should extend to individuals regardless of their sexual orientation or gender identity (Thomas, 2018).

Interestingly enough, a study found that when religious individuals heard religious elites speaking favorably on same-sex marriage, they, themselves, were more likely to support same-sex marriage (Harrison & Michelson, 2015). These findings highlight several illuminating key factors. The first being that individuals with high religiosity also have diverse ideas, opinions, and perspectives, so that religiosity is not always a reliable factor for predicting public opinion. The second being that belief in religious freedom and support for same-sex marriage and antidiscrimination policies are not mutually exclusive and do not have to be viewed that way either. The third being that religious elites and elite discourse overall have an immense impact on public opinion and public attitudes (Harrison & Michelson, 2015). Going even further, in-group elites in general are often responsible for shifting public attitudes, even for highly controversial and divisive issues. Elites in all groups seem to showcase this ability to reliably and efficiently control the public opinion of their communities (Zaller, 1992). Elites in all groups seem to showcase this ability to reliably and efficiently control the public opinion of their communities.

A theory that embraces this sentiment is elite mobilization. The theory of elite mobilization posits that backlash and policy adoption attempts originate from organized elite interests instead of through natural challenges due to the changing public opinion (Bishin et al., 2020). All in all, the elite groups tend to control and maneuver public opinion on gay rights, advancing their own specific agendas under the guise of mass public support. Elites are specially situated to have an immense impact on mobilizing and swaying the public to join their crusade. This is especially true when they are from an involved and polarizing group, such as the religious right. For example, conservative religious organizations will use mail, media, and even sermons to mobilize advocates against gay rights within the evangelical churches. These elites have policy adoption on the top of their minds, using their directed public opinion as a means to push policy through (Bishin et al., 2020). Elite mobilization theory, at its core, questions the actuality of organic grassroots public mobilization and instead argues that mobilization is rather the result of careful, strategic action and planning taken by the in-group elites.

In recent years, as this controversy has increased in intensity, anti-LBGTQIA+ elites have been successful in their attempts to introduce and adopt anti-gay policies and legislation, especially at the local and state levels. The factors that determine support for these policies vary greatly across the different aspects of anti-gay rights laws (Taylor et al., 2012). The research shows that a single factor, such as ideology or education, will not determine whether individuals support or oppose anti-gay policies. There is immense variation among these different internal factors, which made support for same-sex couples and the LGBTQIA+ community complex and uncertain (Taylor et al., 2012). More research, especially more recent research, is needed to truly assess the many aspects and factors that can determine public acceptance and policy adoption for nondiscrimination laws.

To make LGBTQIA+ rights issues even more complex, there are often varying viewpoints and acceptance rates for different groups under the LGBTQIA+ umbrella. Studies show that public opinions toward transgender individuals and related policies are often more negative than those of differing sexual orientations (Lewis et al., 2017). Some of the factors that may explain these differences in acceptance for nondiscrimination laws include interpersonal contact, ideology, and religious affiliation (Lewis et al., 2017). If respondents are close to several members of both the transgender community and those of differing sexual orientations, then they are more likely to support public accommodation nondiscrimination laws that aim to protect the rights of those with differing sexual orientations due to more knowledge, education, and familiarity surrounding gay rights issues.

Ideology is a large determinant, with those who are conservative or right-leaning often opposing nondiscrimination laws, especially those dealing with transgender rights, while those who are liberal or left-leaning are consistently in support of nondiscrimination laws across all groups in the LGBTQIA+ community. Interestingly, religious attendance is not a determinant of differences in support for transgender and gay rights policies, but religious affiliation is, with Christian respondents – Evangelical/Protestant and other denominations – reporting a larger difference in acceptance rates than their nonreligious counterparts (Lewis et al. 2017). Obtaining and supporting rights for the LGBTQIA+ community is often a fraught, uphill battle with a multitude of complexities and differing factors in determining acceptance rates. In keeping with such complexity, it has been difficult in the past to assess the factors that would determine support or opposition for public accommodation laws in favor of same-sex marriage rights.

Data and Methods

The data set used by the author for the following analysis is the American National Election Studies (ANES) 2020 Time Series Study. The ANES study conducts pre- and post-election surveys and interviews with participants, spanning topics such as voting, public opinion, political efficacy, political participation, media, and more, in order to provide prevalent research data to the political science field. For the ANES 2020 Time Series Study, data collection began in August 2020 and continued through December 2020 (ANES 2020, 2021). According to the website, the ANES 2020 Time Series Study features “a fresh cross-sectional sample, with respondents randomly assigned to one of three sequential mode groups: web only, mixed web, and mixed video” and additionally features “re-interviews with 2016 ANES respondents, and post-election surveys with respondents from the General Social Survey (GSS)” (ANES 2020, 2021).

The dependent variable used in this research is public opinion on the ability of businesses to choose to provide services to same-sex couples or not, with the variable title being “V201406.” This dependent variable is a dichotomous variable with one category being “1. Should be allowed to refuse” and the other being “2. Should be required to provide services.” The dependent variable will be cross-tabulated with eight different independent variables in order to determine which is the strongest factor affecting the dependent variable.

The first independent variable is five categories of generations, titled “AGE.” The categories are Gen Z, Millennials, Gen X, Boomers, and Greatest generations. The second independent variable is education levels in three categories, titled “ED3.” The categories are no college, bachelor’s degree, and graduate degree. The third independent variable is public opinion on which party handles the economy better, titled “V201239.” The categories for this variable are “1. Democrats would do a much better job 2. Democrats would do a somewhat better job, 3. Not much difference between them, 4. Republicans would do a somewhat better job, and 5. Republicans would do a much better job.” The fourth independent variable is three levels of political knowledge, titled “KNOW3.” The categories are low knowledge, medium knowledge, and high knowledge. The fifth independent variable is an individual’s identified party identification, titled “V201231x.” The categories are “1. Strong Democrat, 2. Not very strong Democrat, 3. Independent-Democrat, 4. Independent, 5. Independent-Republican, 6. Not very strong Republican, and 7. Strong Republican.” The sixth independent variable is ideologies among respondents in three categories, titled “IDEOLOGY.” The three ideological categories are liberal, moderate, and conservative. The seventh independent variable is respondent's self-identified race and ethnicity, titled “V201549x.” The six categories for this variable are “1. White, non-Hispanic, 2. Black, non-Hispanic, 3. Hispanic, 4. Asian or Native Hawaiian/other Pacific Island, non-Hispanic alone, 5. Native American/Alaska Native or other race, non-Hispanic alone, and 6. Multiple races, non-Hispanic.” The final and eighth independent variable is a dichotomous variable on whether or not the respondent typically follows conservative media sources, titled “RTMEDIA.” The one category is “does not follow conservative media,” with the other category being “follows conservative media.” This research chose to exclude the type of religion or religiosity as a dependent variable, as it has been shown to be a major influential factor in public opinion of the gay community already. The research aims to explore additional factors that seem to contribute to the support or opposition of requiring businesses to provide services to same-sex couples.

Findings

There are some statistically and substantively significant findings when analyzing cross-tabulations of the eight independent variables individually with the dependent variable for bivariate regression analysis. Each of the cross-tabulations was found to be statistically significant with a p-value of 0.01, with the absolute value of gamma scores ranging from 0.057, a very weak relationship, to 0.671, a strong relationship.

As seen in Table 1 below, public opinion about providing services to same sex couples is clearly a very divided and controversial topic, with 49.31% of respondents believing that businesses should be allowed to refuse services and an incredibly close 49.34% of respondents believing that businesses should be required to provide services. There is a difference of only two respondents between the opposing, polarized opinions on this topic, demonstrating just how controversial public opinions are surrounding this topic.

Table 1    

The first independent variable of age divides respondents into five generations. In Table 2 below, one can see that Boomers are the largest generation represented in this survey at 33.26% of total respondents, with Millennials (27.32%) and Gen X (24.92%) trailing behind. As seen in the cross-tabulation below (Table 2), the younger the respondent, the more likely they are to believe that businesses should be required to provide services to same-sex couples and vice versa, though there is not a huge difference. The data is statistically significant with a negative weak gamma score of -0.161, proving the previous statement that there may be some negative relationship between the variables, but it is not very strong at all. This result, most especially the low gamma strength score, is surprising, as typically one would assume that the younger an individual, the more likely they are to support same-sex couple rights in many or all aspects, including public accommodations. As mentioned above, while this is slightly true, there is not much of a difference in public opinion across the generations, at least, much less than one might assume there would be.

Table 2

The second independent variable is education, divided into three categories. In Table 3, one can see that “No College” is the largest level represented by far at 55.25% of total respondents. As seen in the cross-tabulation (Table 3), education does not seem to have much of an effect on the dependent variable. The data is statistically significant but has a very weak gamma score of 0.100, proving little to no relationship between the variables. This result is unexpected, as studies have typically shown that the more education an individual receives, the more likely they are to have progressive ideals and support the LGBTQIA+ community. While there is a greater difference for those who have obtained graduate degrees, favoring the requirement of businesses to provide services, the other categories – no college and bachelor's degrees – are much closer in percentage, especially among those with bachelor’s degrees.

Table 3

The third independent variable is public opinion on which political party is better at handling the nation’s economy. The largest group of respondents believes that the Republican party would do a much better job handling the nation’s economy, at 28.20%, with not much difference between them following behind in second at 22.20%. The data is statistically significant with a relatively strong negative gamma score of -0.554, meaning that the more confident an individual is that Republicans are better at handling the economy, the more likely they believe that businesses should be allowed to refuse services. This can be seen in the cross tabulation (Table 4), where 78.9% of respondents who said that the Republican party would do a much better job handling the nation’s economy believe that businesses should be allowed to refuse services while 77.7% of respondents who said that the Democratic party would do a much better job handling the nation’s economy believe that businesses should be required to provide services. This finding is more expected and predicted, as those who believe that the Republican Party is better at handling the nation’s economy are more likely to be Republican or right-leaning themselves and may also be less politically knowledgeable, as it has been proven that across history, the economy is not, in fact, much better under a Republican administration. These individuals are then also more likely to not be in favor of government regulations, nondiscrimination laws, over freedom of religion arguments, or supporting the gay community.

Table 4

The fourth independent variable is three levels of political knowledge. In the frequency table below (Table 5), one can see that the medium knowledge category is the largest group with 54.89% of the total respondents. In the cross-tabulation (Table 5), one can see that knowledge is another factor that does not really affect the public opinion on providing services to same-sex couples, even though the data is statistically significant, as the percentages are pretty even across the board. This is also shown in the gamma score, which is very weak and negative at -0.057. This result was initially surprising, but upon further reflection, it is not as surprising as it might seem at first glance. Those who are more politically knowledgeable are likely to be attuned to the 2018 Masterpiece Cakeshop Supreme Court case, and as the court sided with Masterpiece, they may be basing their public opinion on the overall decision of that Supreme Court case ruling.

Table 5

The fifth independent variable is self-determined party identification, and the largest number of respondents are considered Strong Democrats at 23.68%, and the next biggest group is Strong Republicans at 20.89%. The cross-tabulation (Table 6) shows that there is a strong correlation between party ID and opinions on providing services. Respondents who are Democrats or Democrat-leaning are much more likely to support the requirement of businesses to provide services to same-sex couples, while Republicans or Republican-leaning are much more likely to favor the ability of businesses to refuse services to same-sex couples. This can also be shown through the statistically significant data and relatively strong negative gamma score of -0.523. This result is not surprising and appeared as expected. Those who self-identify as Democrats are often much more in support of the gay community and the expansion of the community’s rights, while those who identify as Republicans are often less supportive of the gay community and are less likely to favor government regulation of businesses.

Table 6

The sixth independent variable is about the respondent’s overall ideologies – liberal, moderate, or conservative. About 39% of the respondents are conservative, whereas 35% of respondents are liberal, leaving the remaining 26% as moderate, which can be seen in the frequency table below (Table 7). Similar to the last independent variable, liberal respondents are extremely likely (75%) to support businesses being required to provide services to same-sex couples, while conservative respondents are much more likely (79%) to favor businesses refusing services to same-sex couples. Moderate respondents are considerably split down the middle – 54% vs. 45%. This relationship is statistically significant and has a strong negative correlation, shown through the gamma score of -0.671. This finding is also unsurprising, as individuals who identify as liberal are typically supportive of progressive policies for same-sex couples and believe that government regulation is an effective and efficient method of expanding these rights, with the opposite often being true of conservative individuals. Moderates are unsurprisingly fairly evenly split down the middle, with a slight preference towards requiring businesses to provide services.

Table 7

The seventh independent variable is the respondent’s self-identified race and ethnicity. The overwhelming majority of respondents are White non-Hispanic, at 72.02% of the total respondents, whereas the next largest group is Hispanic at only 9.20%. This reveals an immense gap in the data set (Table 8), as these percentages are not nearly reflective of the current racial and ethnic demographics of America, meaning race could be a more salient factor, given more data. There is no clear relationship shown in the cross-tabulation or through the positive weak gamma score of 0.179, though the relationship is statistically significant. This result, though, does not have an overall strong gamma score, but still indicates some interesting implications and findings. Those who identify as White non-Hispanic are pretty evenly split down the middle, with a slight preference to allow businesses to refuse services, which is not a surprising result, but those who identify as Black non-Hispanic, Hispanic, and Asian or Native Hawaiian/Pacific Islander non-Hispanic alone are fairly overwhelmingly more in support of requiring businesses to provide services. This is likely due to the phenomenon of solidarity and advocacy of other minoritized and historically marginalized communities with each other. Native American/Alaska Native or other non-Hispanic alone and respondents of Multiple races non-Hispanic are also evenly split down the middle with a slight lean towards requiring services to be provided.

Table 8

The eighth and final independent variable is whether respondents follow conservative media sources. The large majority, at 83.30%, of respondents do not regularly follow conservative media, while only 16.70% of total respondents do follow conservative media. In this cross-tabulation (Table 9), it is evident that a relationship exists between these variables. While opinions on providing services are more split among the respondents who do not follow conservative media, there is a clear correlation for those who do follow conservative media to then support businesses in refusing services to same-sex couples (80.8%). This relationship is statistically significant and is confirmed by the strong but negative gamma score of -0.655. This finding was not surprising, as those who frequently follow conservative media sources, such as Fox News, often have an alternative outlook on issues surrounding the gay community and are, of course, incredibly likely to identify as Republican and conservative themselves, which supports what we have seen in the previous cross-tabulations. Those who do not follow conservative media sources may still identify as conservative or Republican and instead follow more reputable and unbiased news sources, so their public opinion may not be as misconstrued, which may explain the closer to even split.

Table 9

Conclusion

Taking in the above results and findings, some stick out and need to be highlighted for their statistical significance or their gamma strength. One of the most surprising findings from this research is that age does not have a larger effect on individuals’ public opinion on providing services to same-sex couples. Typically, it is assumed that those who are younger are more tolerant and accepting of progressive and anti-discriminatory laws, especially for members of the LGBTQIA+ community; however, there appears to be an incredibly weak relationship between these variables. Another interesting finding is the low gamma score for racial and ethnic identity; however, a disproportionate number of respondents self-identify as White non-Hispanic, far among the other groups, and this discrepancy warrants further study and exploration.

All in all, from this research, ideology (liberal, moderate, and conservative) appears to be the strongest and most prominent factor in determining public opinion surrounding whether or not businesses should be required to provide services to same-sex couples, as it has the largest gamma strength score of -0.671, which somewhat contradicts the earlier hypothesis. Though there does appear to be a strong, negative relationship between the feelings on the ability to choose to provide services to same-sex couples or not and the perception of which political party is better at handling national economic policies, it was not found to be the strongest factor among the data. Additionally, instead of the initial reasoning that this relationship is due to a lack of political, historical, and economic knowledge, the reasoning seems to be more closely linked to the party identification or ideology aspects of the perception of which political party handles the economy better.

The implications of this research are far-reaching and stand as a strong jumping-off point for further research into the rights and discrimination that same-sex couples face in society, and especially in public accommodation settings. Public opinion seems to differ widely when it comes to rights for different segments of the LGBTQIA+ community and even for different types of rights for individuals within one segment, such as marriage rights, child service rights, education and employment rights, and public accommodation rights. Especially in light of the current second term of the Trump administration, many from the LGBTQIA+ community are fearful and pessimistic about what this means for the expansion of future legal rights and what restrictions may be placed on current legal rights for the community, and these fears have already been proven to have substance (Gallup, 2025; Wolfe, 2025). Additional research delving into overall public acceptance of the different sectors of the LGBTQIA+ community, as well as public opinion on several distinct types of rights (e.g., marriage, adoption, public accommodations, sports, career opportunities), would be beneficial to this study and future studies that focus on LGBTQIA+ rights. Without knowing these intricacies, the expectation of support or opposition is too often thrown into one big bucket and cannot be efficiently utilized or addressed.

References

American National Election Studies. (2021). ANES 2020 time series study full release (February 10, 2022 version) [Data Set]. https://www.electionstudies.org.

Bishin, B. G., Incantalupo, M. B., Hayes, T. J., & Smith, C. A. (2020). Elite mobilization: A theory explaining opposition to gay rights. Law & Society Review, 54(1), 233–264. http://www.jstor.org/stable/45284553.

Gaines, N. S., & Garand, J. C. (2010). Morality, equality, or locality: Analyzing the determinants of support for same-sex marriage. Political Research Quarterly, 63(3), 553–567. http://www.jstor.org/stable/25747958.

Gallup. (2025). LGBTQ+ rights. Gallup News. https://news.gallup.com/poll/1651/gay-lesbian-rights.aspx

Harrison, B. F., & Michelson, M. R. (2015). God and marriage: The impact of religious identity priming on attitudes toward same-sex marriage. Social Science Quarterly, 96(5), 1411–1423. https://www.jstor.org/stable/26612284.

Lee, H.-Y., & Mutz, D. C. (2019). Changing attitudes toward same-sex marriage: A three-wave panel study. Political Behavior, 41(3), 701–722. https://www.jstor.org/stable/48688526.

Lewis, D. C., Flores, A. R., Haider-Markel, D. P., Miller, P. R., Tadlock, B. L., & Taylor, J. K. (2017). Degrees of acceptance: Variation in public attitudes toward segments of the LGBT community. Political Research Quarterly, 70(4), 861–875. http://www.jstor.org/stable/26384822.

Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission, 584 U.S. 617 (2018). https://www.supremecourt.gov/opinions/17pdf/16-111_j4el.pdf.

Powell, B., Schnabel, L., & Apgar, L. (2017). Denial of service to same-sex and interracial couples: Evidence from a national survey experiment. Science advances, 3(12), eaao5834. https://doi.org/10.1126/sciadv.aao5834.

Taylor, J. K., Lewis, D. C., Jacobsmeier, M. L., & DiSarro, B. (2012). Content and complexity in policy reinvention and diffusion: Gay and transgender-inclusive laws against discrimination. State Politics & Policy Quarterly, 12(1), 75–98. http://www.jstor.org/stable/41575844.

Thomas, G. (2018). Religious liberty, same-sex marriage and public accommodations. Perspectives on Politics, 16(1), 58–72. https://www.jstor.org/stable/26615360.