Abstract
This study examines knowledge of and attitudes toward Roe v. Wade among a sample of 779 US Latinx adults. Survey response patterns were examined in relation to generational status and choice of survey language as well as to several demographic variables previously shown to influence abortion attitudes (e.g., age, religiosity, political affiliation). Differences were found in knowledge of Roe v. Wade by generational status and survey language, with those with higher generational statuses and those taking the survey in English exhibiting greater knowledge. Finally, greater knowledge of Roe v. Wade and choosing to take the survey in English predicted more positive attitudes toward Roe v. Wade controlling for other demographic variables; no effect on attitudes of generational status was observed. These findings contribute to our understanding of abortion attitudes among US Latinxs as well as the relationship between political socialization, knowledge, and attitudes toward social issues.
Given the large and growing Latinx population in the US (Noe-Bustamante et al., 2020), understanding Latinx public opinion and the factors that contribute to differences within Latinx attitudes toward various issues is important for understanding US public opinion more broadly. Attitudes toward abortion in the US are often assessed in national polls via items eliciting opinions about the legality of abortion (e.g., the General Social Survey, Smith & Son, 2013; Pew Research Center, 2021). National polls using items such as these have suggested that US Latinxs tend to hold less supportive views toward abortion overall (e.g., responding that abortion should be illegal in all or most cases in higher numbers) than the general US population (Hartig, 2018), and several studies suggest differences in attitudes toward abortion among US Latinxs according to factors such as country of origin, level of acculturation, language preference, nativity, and/or generational status (Amaro, 1988; Bolks et al., 2000; Branton et al., 2010, 2014; Ellison et al., 2005; Sanchez, 2006).
Another approach to gauging attitudes toward abortion in the US has been to ask about support for Roe v. Wade, the landmark 1973 Supreme Court decision that established a constitutional right to abortion prior to fetal viability (Watson, 2018). Items measuring attitudes toward Roe v. Wade have asked, for example, whether respondents favor or oppose the decision or whether respondents would like to see Roe v. Wade completely overturned (Brenan, 2018; Bowman & Sims, 2017, Dimock et al., 2013; Doherty et al., 2019). Adamek (1994) notes that polling regarding attitudes toward Roe v. Wade may be challenged by respondents’ (lack of) knowledge about the decision. Research has long established that political knowledge can help explain a range of political behaviors (e.g., voting; Delli Carpini & Keeter, 1996) and that policy-specific knowledge can affect people’s attitudes toward that policy (Gilens, 2001). Research has also shown that US Latinx tend to be less politically informed than their White counterparts (Abrajano, 2010; Abrajano & Alvarez, 2010) and that political socialization differs for recent immigrants as compared with nonimmigrants or those who have been in the US for a long time (Abrajano & Alvarez, 2010). Relatively little is known about attitudes toward Roe v. Wade specifically among US Latinxs or about the relationship between knowledge of Roe v. Wade among US Latinx respondents and their sentiments toward the decision. To better understand Latinx abortion attitudes in general as well as the relationship between knowledge and opinion among US Latinxs, this study examines knowledge of and attitudes toward Roe v. Wade among a sample of US Latinx respondents and considers the relationship between Roe v. Wade knowledge and attitudes and respondent generational status and choice of survey language.
Background
Knowledge of and Attitudes Toward Roe v. Wade
US public opinion on abortion has been relatively stable over several decades, with the majority of US adults indicating that abortion should be legal in at least some circumstances (Bowman & Sims, 2017; Smith & Son, 2013). Periodic polling on attitudes toward Roe v. Wade has, similarly, shown that consistent majorities of US adults believe that Roe v. Wade should be upheld (Bowman & Sims, 2017; Brenan, 2018), with 70% of US adults in 2019 reporting not wanting to see the case overturned (Doherty et al., 2019). Nevertheless, attitudes toward Roe v. Wade show sharp partisan and religious divides. Democrats and independents oppose overturning Roe v. Wade by wide margins, whereas a higher percentage of Republicans would like to see the decision overturned than not (Brenan, 2018). White evangelical Protestants favor overturning Roe v. Wade, whereas White mainline Protestants, Black Protestants, and White Catholics tend to oppose overturning the decision (Dimock et al., 2013). However, some scholars have pointed to weaknesses of previous polling related to Roe v. Wade, including a tendency to assess attitudes toward the decision via a single item (Adamek, 1994) and the lack of detail provided by dichotomous response options (e.g., overturn vs. not; Jozkowski et al., 2020), especially given the complexity of the Roe v. Wade decision.
As Adamek (1994) points out, polling related to Roe v. Wade is also complicated by the fact that “the American public is quite ill-informed about the decision” (p. 412). For example, Blake (1977) reported that, on a 1975 survey, 28% of men and 22% of women indicated never having heard of Supreme Court decisions involving abortion in the preceding 3 years (i.e., Roe v. Wade and related decisions); only 45% of men and 48% of women indicated having heard of the decisions and knowing that their effect was to make it easier to have a legal abortion. More recently, according to findings from a 2013 Pew Research Center poll, 62% of respondents knew that Roe v. Wade dealt with abortion, whereas 17% thought that it dealt with another issue (e.g., death penalty, school desegregation), and 20% did not know. Among respondents younger than 30, only 44% answered correctly that Roe v. Wade dealt with abortion (Dimock et al., 2013).
To address some of the weakness of previous studies, Jozkowski et al. (2020) used a tiered design to assess knowledge of and attitudes toward Roe v. Wade among US adults. They first asked a series of questions to assess knowledge of Roe v. Wade. Then, they informed participants about Roe v. Wade, and, following the explanation of the decision, they assessed sentiments toward Roe v. Wade. Knowledge of and sentiments toward the decision were each measured via multiple items, with scaled response options. Jozkowski et al. (2020) found that their respondents generally knew that Roe v. Wade pertained to abortion and that abortion was legal at the time of the study 1 but were less knowledgeable about the implications of overturning Roe v. Wade. With regard to sentiments, a majority of their sample supported upholding Roe v. Wade, although support was lower compared with previous research. Jozkowski et al. (2020) suggest this lower level of support may have been influenced by respondents being informed about the decision and the implications of overturning it prior to responding to the sentiment items.
A notable trend observed in Jozkowski et al.’s (2020) results was that knowledge about Roe v. Wade was lower for Latinx respondents (as compared with all other race/ethnicity categories). Latinx respondents also reported among the lowest levels of support on all questions related to attitudes toward Roe v. Wade. With these findings in mind, the present study returns to the data examined in Jozkowski et al. (2020) to investigate knowledge of and attitudes toward Roe v. Wade among Latinx respondents more closely.
Latinx Political Knowledge and Attitudes Toward Abortion
Social issues, including and especially abortion, are often viewed as areas where US Latinx political alliances may be particularly malleable (e.g., Abrajano & Alvarez, 2010; Nicholson et al., 2006). Given the established link between political knowledge and attitudes as well as behavior (Delli Carpini & Keeter, 1996; Gilens, 2001), understanding knowledge of and attitudes toward Roe v. Wade among US Latinxs is an important next step in understanding abortion attitudes more generally. Aside from Jozkowski et al. (2020), relatively little is known specifically about Latinx knowledge of and attitudes toward Roe v. Wade. Nevertheless, previous research has suggested that Latinxs tend to oppose abortion at higher rates than the general US population (Bolks et al., 2000; Hartig, 2018). For example, using data from the Latino National Political Survey, Bolks et al. (2000) found that Latinx respondents endorsed the response “Abortion should never be permitted” at a rate 6% higher than the general population. Specifically, 20% of respondents indicated support for a full abortion ban, 51% responded that abortion should be allowed under certain circumstances, and 29% indicated support for the unrestricted right to choose to have an abortion. Further, Bolks et al. (2000) also found that Latinx attitudes toward abortion can be predicted by many of the same variables that predict non-Latinx abortion attitudes, including religiosity, feminist attitudes, income, age, and level of education. Other studies have corroborated the association among these variables, emphasizing, in particular, the importance of religious factors (e.g., denomination, beliefs, church attendance) in predicting Latinx abortion attitudes (e.g., Bartkowski et al., 2012; Ellison et al., 2005; Holman et al., 2020).
US Latinxs have repeatedly been shown to have systematically lower levels of political knowledge than their White counterparts (Abrajano, 2010; Abrajano & Alvarez, 2010), including specifically about the Supreme Court (Pedraza & Ura, 2021). Still, the ethnic and social diversity of the US Latinx population highlights the lack of a distinct Latinx political identity. And the high (though declining) proportion of US Latinxs who are foreign-born means that what we know generally about political socialization processes may not always apply (Abrajano & Alvarez, 2010). Thus, understanding the factors that may play a role in the formation of political knowledge and attitudes toward social issues specifically among US Latinxs is important for understanding the public opinion and political behaviors of this heterogeneous group.
Generation, Language, and Latinx Abortion Attitudes
A sizeable and rapidly growing body of literature exists on the relationship between acculturation—defined by Abraído-Lanza et al. (2006) as “the process by which individuals adopt the attitudes, values, customs, beliefs, and behaviors of another culture” (p. 1342)—and various health indicators among US Latinxs (Abraído-Lanza et al., 2016; Schwartz et al., 2010). Likewise, public opinion researchers have explored the relationship between acculturation and related variables (e.g., language preference, nativity, generational status) and US Latinx opinion toward topics such as immigration, education, marriage equality, and abortion (Abrajano & Alvarez, 2011; Branton et al., 2010, 2014). With regard to attitudes toward abortion, overall, a positive relationship between acculturation—assessed in various ways—and support for abortion rights has been found across several studies. For example, Amaro (1988) found that higher acculturation scores on the Acculturation Rating Scale for Mexican Americans (Cuellar et al., 1980) were significantly correlated with more positive attitudes about abortion for Mexican American women. Ellison et al. (2005) found that foreign-born Latinos were more likely to favor more restrictive abortion policies than US-born Latinos. Jones and Cox (2012) similarly reported differences in attitudes toward abortion between foreign- and US-born Hispanic Americans, with a majority of second (55%) and third (57%) generation Hispanic Americans supporting legal abortion in all or most cases, whereas only 41% of first-generation Hispanic Americans held the same view. Finally, using data from the 2006 Latino National Survey, Branton et al. (2010) found a significant, positive relationship between acculturation (measured on a 6-point additive scale based on generational status, language preference, and one’s ability to carry out a conversation in one’s non-dominant language) and support for abortion rights, and Branton et al. (2014) found that English-dominant Latinxs were in favor of less restrictive abortion policy options than Spanish-dominant Latinxs. In contrast to these findings, Sanchez (2006) found that foreign-born Latinxs were more likely to support the legality of abortion than Latinxs born in the US, a finding that Sanchez hypothesizes may be due to the “hard-line approach of most Latin American countries” to abortion (p. 439). To better understand the role that contexts of socialization may play in the relationship between knowledge and attitudes toward abortion and abortion-specific policy among US Latinxs, the present study considers language of survey (Spanish versus English) and generational status in relation to Latinx knowledge of and attitudes toward Roe v. Wade. The issues explored are especially timely given the recent decision in Dobbs v. Jackson Women’s Health Organization, which essentially overturned the decision in Roe v. Wade (Liptak, 2022). US Latinx women have nearly twice the abortion rate of non-Hispanic White women in the US (Kortsmit et al., 2021) and, thus, are likely to be disproportionately affected by changes to the decision.
Present Study
The present study examines US Latinx knowledge of and attitudes toward Roe v. Wade and explores the relationship between knowledge, attitudes, and respondents’ generational status and choice of survey language. The study is guided by three research questions:
To what extent do US Latinxs know about Roe v. Wade, and what are their attitudes toward the decision?
Does knowledge of Roe v. Wade differ by generational status or choice of survey language?
Do knowledge of Roe v. Wade, generational status, and language of survey predict US Latinx attitudes toward Roe v. Wade, controlling for various demographic characteristics?
Method
Procedure
A national, cross-sectional survey, administered by Qualtrics, was fielded in Fall 2018 to assess knowledge of and attitudes toward Roe v. Wade among English- and Spanish-speaking US adults. Eligibility criteria included being 18 years of age or older, residing in the US, and having English and/or Spanish reading fluency. Quotas were set to increase sample diversity according to age, gender, political affiliation, race/ethnicity, and language; in particular, Spanish-speaking participants identifying as Latina/o, Hispanic, or Latinx were oversampled. Those interested in the study read a brief informed consent page that provided information about the study and informed participants that continuing in the survey signified their consent to participate. The study protocol was approved by the institutional review board prior to data collection.
Participants
A total of 3,030 participants completed the survey. Participants were excluded from the analysis if they did not meet a minimum response time criterion (n = 330) or for missing data or incorrect responses to a data quality item embedded in the survey (n = 140). For the current analyses, we limited our sample to include only people who identified as Hispanic or Latino/a/x and provided responses to all items included in the present analysis (n = 779). Participants could complete the survey in English (n = 376) or Spanish (n = 403). Table 1 presents demographic information about the analytic sample.
Descriptive Statistics by Generational Status and Survey Language.
Measures
The survey and survey items were designed and written by a team of monolingual (English only) and bilingual (English-Spanish) researchers using a parallel development approach (Harkness et al., 2003) with English as the source language. Survey creation and translation was guided by the Translate, Review, Adjudicate, Pretest, and Document (TRAPD) framework (Harkness et al., 2003), which involves a multi-step, team-based process with several levels of review and reconciliation. When conflicts with translation from English to Spanish arose, an iterative process was used to make changes to both the source (English) and target (Spanish) documents. We also worked with a legal scholar with specialization in constitutional law and abortion when writing the Roe v. Wade knowledge and attitude items as well as the corresponding narrative embedded in the survey between these items, described below. Relevant to the current study, we assessed knowledge of and attitudes toward Roe v. Wade, generational status, survey language, and additional demographic items.
Roe v. Wade knowledge
We assessed knowledge of Roe v. Wade via five items. These items were divided into two sets, each of which measured a unique knowledge dimension. The first set examined knowledge regarding to what Roe v. Wade pertains as well as knowledge about abortion legality. Specifically, this set of items asked: (1) “The Supreme Court ruling in Roe v. Wade deals with constitutional rights related to—(a) Gun ownership, (b) Abortion, (c) De-segregation of public schools, (d) Corporate political donations” (correct answer: abortion); (2) “Based on what you know or have heard, is it legal to get an abortion in [insert home state]?—(a) Yes, it is legal here, (b) No abortion is not legal here, (c) Not sure/Don’t know” (correct answer: yes); and (3) “Based on what you know or have heard, is it legal to get an abortion in the United States?—(a) Yes, it is legal here, (b) No abortion is not legal here, (c) Not sure/Don’t know” (correct answer: yes). Following this set, participants were informed that the Roe v. Wade decision pertains to abortion and that it established the constitutional right to abortion until about 24 weeks of pregnancy. The second set of items examined whether participants were knowledgeable about the implications of overturning Roe v. Wade. Participants were asked two true/false items: (1) “If the Supreme Court overturned (got rid of) Roe v. Wade, it would be illegal to get an abortion everywhere in the United States” (correct answer: false) and (2) “If the Supreme Court overturned (got rid of) Roe v. Wade, states could pass more laws restricting abortion than they can now” (correct answer: true). Correct responses were scored as one, and incorrect responses were scored as zero. Item scores were summed to create a knowledge score ranging from zero (lowest level of knowledge of Roe v. Wade) to five (highest level of knowledge of Roe v. Wade).
Roe v. Wade attitudes
Attitudes toward Roe v. Wade were measured using four items with a bipolar, 7-point snapped-to-grid slider format including a neutral middle option. These items included (1) “How would you feel if the Supreme Court overturned (got rid of) Roe v. Wade and removed the constitutional right for a woman to choose to have an abortion?” (response options ranging from “Very Happy” to “Very Unhappy”); (2) “Do you think that Roe v. Wade should be overturned (gotten rid of)?” (response options ranging from “Definitely Should” to “Definitely Should Not”); (3) “Which best represents your feelings toward Roe v. Wade?” (response options ranging from “Strong desire to see Roe v. Wade upheld (kept)” to “Strong desire to see Roe v. Wade overturned (gotten rid of)”); and (4) “Which best represents your feelings toward the constitutional right of a woman to choose to have an abortion until about 24 weeks into her pregnancy?” (response options ranging from “Strong desire to keep the right to have an abortion” to “Strong desire to remove the right to have an abortion”). Following Jozkowski et al. (2020), these items were individually scored (with items 1, 2, and 4 being reverse scored) from 0 to 6 and then were summed to create a Roe v. Wade sentiment variable, ranging from 0 (indicating lowest level of support for Roe v. Wade) to 24 (indicating highest level of support for Roe v. Wade).
It should be noted that after responding to the five knowledge items and immediately prior to responding to the four sentiment items, respondents received additional information about Roe v. Wade to provide information that would allow them to respond to the sentiment items with knowledge of the potential impacts of changes to Roe v. Wade. Specifically, the survey read: “If Roe v. Wade was overturned (gotten rid of), the right to choose to have an abortion would no longer be constitutionally guaranteed. Instead, individual states could choose to prohibit abortion if they wanted. If Roe v. Wade was overturned by the Supreme Court, it is likely that abortion would be legal in some states, and illegal in other states in all or most situations. Individual states would get to decide when a woman could have an abortion, if at all, and how far into a pregnancy a woman could obtain an abortion.”
Generational status and survey language
We also assessed respondents’ generational status (i.e., the number of generations removed from immigration) and choice of survey language (English vs. Spanish). Both variables have been used as indicators of or proxies for acculturation in previous studies, including studies related to abortion attitudes (Branton et al., 2010, 2014). Generational status was assessed using data from two items: (1) “What best describes you?—(a) U.S. Born Latino/a or (b) Foreign Born Latino/a” and (2) “Where were your parents born?—(a) Both in the U.S., (b) Both outside the U.S., (c) One in the U.S. and one outside the U.S., (d) One in the U.S. and the other unknown, (e) One outside the U.S. and the other unknown, (f) I do not know.” Responses from these two items were combined to categorize participants as first, second, or third+ generation. Participants were categorized as first generation if they and their parents were born outside the United States, as second generation if they were born in the United States and at least one of their parents was born outside the United States, and as third+ generation if they and their parents were born in the United States. Language choice was assessed via the language in which the respondent elected to complete the survey (i.e., English or Spanish).
Demographic characteristics
Information was gathered via individual survey items regarding participants’ gender, age, bible literalism (i.e., the extent to which they believe the bible is the literal word of God), political affiliation, and abortion identity label (e.g., prolife, prochoice, neither, both), as represented in Table 1.
Analysis
Respondents’ knowledge of and sentiments toward Roe v. Wade were assessed using summary statistics of responses to each knowledge and sentiment survey item as well as via mean overall knowledge and sentiment scores. Data were analyzed using SPSS 25. After testing statistical assumptions without any violations, chi-square tests (for categorical variables) and analysis of variance (ANOVA; for continuous variables) were used to examine for statistical differences among acculturation-related variables on knowledge of Roe v. Wade. An ordinary least squares (OLS) regression was conducted to predict attitudes toward Roe v. Wade according to knowledge of Roe v. Wade, generational status, and choice of survey language, with the following demographics as control variables: age, gender, bible literalism, political affiliation, and abortion identity.
Results
We first assessed Latinx knowledge of Roe v. Wade and whether level of knowledge differed on the basis of generational status or survey language. Table 2 provides a summary of the number and percentage of correct responses to individual knowledge items across generational status and survey language as well as chi-square tests of significance across these variables for individual items; total mean knowledge scores are presented in the bottom row.
Correct Responses to Roe v. Wade Knowledge Items.
As shown in Table 2, for four of the five Roe v. Wade knowledge items, a majority of the sample answered correctly, with average correct responses to each of those four items ranging from 58.0% to 77.9%. However, on one item (“If the Supreme Court overturned [got rid of] Roe v. Wade, it would be illegal to get an abortion everywhere in the United States”), only 31% of the sample correctly selected “false” as the answer. Responses to the first three knowledge questions (Roe v. Wade content, abortion legality in respondent’s state, abortion legality in US) showed clear trends with regard to generational status, with a higher percentage of correct responses for those from second and third+ generations; differences in the percentage of correct responses across generations were statistically significant for these questions (Table 2). The fourth knowledge question (“If the Supreme Court overturned (got rid of) Roe v. Wade, it would be illegal to get an abortion everywhere in the United States” [correct answer: false]) showed no statistical differences by generational status, with fewer than one-third of each generation group answering the item correctly. There was not a clear trend for generational status on the question pertaining to the implication of overturning Roe v. Wade, with the second generation group having the lowest percent answering the item correctly. With regard to survey language, respondents answering in English had a significantly higher percentage of correct responses on all knowledge questions (Table 2).
Mean Roe v. Wade knowledge scores (out of 5) for the current sample overall as well as by generational status and survey language are presented in the final row of Table 2. Overall, the current sample had a Roe v. Wade knowledge score of 2.95. Regarding group differences, separate one-way ANOVAs indicated statistical differences in mean knowledge scores by generational status, F(2, 776) = 18.08, p < .001, η2 = .05, and by survey language, F(1, 777) = 44.92, p < .001, η2 = .06. Bonferroni post hoc analyses revealed that first generation Latinx respondents had statistically lower mean knowledge scores than second generation (p = .012) and third+ generation (p < .001) respondents. With regard to survey language, respondents taking the survey in English had statistically higher scores on average (M = 3.26, SD = 1.29) than those taking the survey in Spanish (M = 2.65, SD = 1.25).
Regarding attitudes toward Roe v. Wade, summary results for each response option at the item level (not presented in detail here) indicated that a higher percentage of respondents support Roe v. Wade than oppose it. For each question, the most frequent response was that indicating the highest level of support; additionally, more than 50% of responses to each question indicated some level of support (i.e., a response of 4 or higher). Table 3 presents mean response scores, out of six, for each sentiment question overall and by generational status and survey language.
Mean and Standard Deviations for Roe v. Wade Attitude Items by Generational Status and Survey Language.
A look at mean response scores reveals that the question asking about abortion rights that does not include the words “Roe v. Wade” had the lowest scores overall (M = 3.25) and within each generation and language group. That is, as a whole, respondents reported more negative feelings toward the constitutional right of a woman to choose to have an abortion than they did toward Roe v. Wade, even after being informed that Roe v. Wade establishes the constitutional right of women to have an abortion. Mean overall Roe v. Wade sentiment scale scores by generation and language groups indicate that second and third+ generation respondents had numerically higher sentiment scores than first generation respondents as did participants responding to the English version (as compared to the Spanish version) of the survey.
OLS regression was used to examine whether attitudes toward Roe v. Wade were predicted by knowledge of Roe v. Wade and/or generational status or survey language, controlling for the demographic variables previously described (i.e., gender, age, bible literalism, political affiliation, and abortion identity). The multivariate model predicting attitudes toward Roe v. Wade was statistically significant, F(20, 758) = 27.55, p < .001, with an R2 of .42. Details of the regression model, which accounts for 42% of the variance in attitudes toward Roe v. Wade, are presented in Table 4.
OLS Regression Model Predicting Attitudes Toward Roe v. Wade.
With regard to control variables, gender, age, bible literalism, political affiliation, and abortion identity all significantly predicted Roe v. Wade sentiments for this sample. Specifically, identifying as a woman (β = 1.37, p = .003), being older than 65 (as compared with being 18–24; β = 2.80, p = .007), and believing the Bible is written by people (as opposed to believing it is the literal word of God; β = 2.91, p < .001) were all predictors of higher Roe v. Wade sentiment scores. Politically, identifying as a Democrat (β = 2.93, p < .001), an Independent (β = 1.97, p = .001), or having no political affiliation (β = 1.56, p = .024) were also significantly associated with higher Roe v. Wade sentiments as compared with identifying as Republican. Finally, abortion identity significantly predicted attitudes toward Roe v. Wade: Respondents identifying as pro-choice (β = 8.33, p < .001), as equally pro-life and pro-choice (β = 6.52, p < .001), or as neither pro-life nor pro-choice (β = 4.49, p < .001) were all more likely to have higher Roe v. Wade sentiment scores than those identifying as pro-life.
With regard to our variables of interest, Roe v. Wade knowledge significantly predicted Roe v. Wade attitudes for this sample (β = .36, p = .026), with each 1-point increase in initial Roe v. Wade knowledge (out of five) predicting a .36 increase in Roe v. Wade sentiment. Survey language was also statistically significant, with taking the survey in Spanish (as compared with English) associated with lower Roe v. Wade sentiments (β = -1.17, p = .011). Generational status did not significantly predict attitudes toward Roe v. Wade after controlling for sociodemographic variables.
Discussion
This study assessed knowledge of and attitudes toward Roe v. Wade and abortion legality among a sample of US Latinxs and explored the relationship between these variables and respondents’ generational status and choice of survey language. Results indicated that a majority of the current sample was knowledgeable about Roe v. Wade, with 61.0% knowing that Roe v. Wade deals with abortion; a rate quite similar to the 62% reported by Pew Research Center for the general US population in 2013 (Dimock et al., 2013). More than 65% of the sample also responded correctly about the legal status of abortion in the US, and 58% responded correctly about the legal status of abortion in their own state. Similar to the overall findings (Crawford et al., 2021; Jozkowski et al., 2020), however, respondents were less knowledgeable about the nuances of the decision, especially as related to the implications of overturning Roe v. Wade. Although nearly 78% of the current sample knew that overturning Roe v. Wade would result in states being able to pass more restrictions on abortion, fewer than 32% knew that overturning Roe v. Wade would not result in making abortion illegal everywhere in the US.
Overall knowledge of Roe v. Wade was significantly higher among second and third+ generation respondents than first generation and among respondents who took the survey in English as compared with Spanish. Such findings provide clues to potentially help understand overall lower rates of political knowledge among US Latinxs (Abrajano & Alvarez, 2010) as well as important details specifically related to knowledge about US abortion policy. Given the importance of early socialization experiences in the formation of political knowledge (Pedraza & Ura, 2021; Valentino & Sears, 1998), it makes sense that US Latinxs who themselves were born and potentially spent many of their formative years outside of the US may exhibit lower levels of US-specific political knowledge than those who were born and raised in the US. Likewise, such differences may occur or be amplified based on language preferences or barriers, as we know that that access to political information via sources such as news media and campaign advertisements is lower for Latinxs and especially for those who are Spanish-speaking than for English-speaking non-Latinxs (Abrajano, 2010), which may help explain the lower levels of Roe v. Wade knowledge for the present sample’s Spanish survey takers as compared with those who took the survey in English. That is, differences in structural opportunities related to linguistic proficiency and cultural preferences may result in differences in political and social knowledge, even within Latinx populations.
With regard to Latinx attitudes toward Roe v. Wade, a majority of the present study’s sample expressed support for Roe v. Wade across all sentiment questions. While the scaled response options of the present study do not permit a direct comparison to dichotomous (e.g., overturn vs. not) items common on previous polling related to Roe v. Wade, 54.7% of the present sample indicated some preference for not overturning Roe v. Wade. Likewise, 55.4% of the sample indicated support for upholding Roe v. Wade. Thus, in general, attitudes toward Roe v. Wade in the current sample of US Latinxs appear to fall within the range of support toward Roe v. Wade among the general US population reported by various polls (i.e., Gallup, Pew, Harris) over the last several decades (i.e., 51% to 69% of US adults did not want to see Roe v. Wade overturned; Bowman & Sims, 2017). However, our sample’s support for upholding Roe v. Wade was substantially below recent levels of support for upholding Roe v. Wade, which have topped 70% (Brenan, 2018; Doherty et al., 2019). Such a finding corroborates previous reports of generally lower support for abortion among US Latinxs than among the general US population (Bolks et al., 2000; Hartig, 2018).
Our regression analysis tested whether knowledge of Roe v. Wade, generational status, or survey language predicted attitudes toward Roe v. Wade after controlling for sociodemographic variables. The control variables of gender, age, Bible literalism, political affiliation, and abortion identity all significantly predicted Roe v. Wade sentiments for this sample largely in ways consistent with previous literature (Bolks et al., 2000; Fingerhut, 2017; Jelen & Wilcox, 2003; Lizotte, 2015). Higher initial knowledge of Roe v. Wade was found to be a significant predictor of more supportive attitudes about abortion and Roe v. Wade, as was taking the survey in English. Generational status did not significantly predict Roe v. Wade sentiment after controlling for sociodemographic variables. This differs from some studies (e.g., Ellison et al., 2005; Jones & Cox, 2012) that have found significant relationships between generational status and abortion attitudes. It is noteworthy that knowledge of Roe v. Wade was a significant predictor of attitudes toward Roe v. Wade despite the present study’s tiered design. That is, even though all participants received information about Roe v. Wade following the knowledge items and preceding the sentiment items, respondents’ existing knowledge of this abortion-related decision prior to any information provided by the survey predicted attitudes toward the decision overall, and this level of knowledge was higher among second and third+ generation Latinxs as compared with first generation and among those taking the survey in English as compared with Spanish. Policy- and issue-specific campaigns aiming to sway attitudes and/or voting behavior of US Latinxs may be well served by understanding linguistic and cultural variation within the US Latinx population as well as how such variation relates to political knowledge and, subsequently, attitudes.
Finally, it is interesting to note that among Roe v. Wade attitude items in the current study, support was lowest for the item that did not include the phrase “Roe v. Wade” in its wording; this item asked directly about the constitutional right to abortion. Specifically, 51.1% of the present study’s sample indicated some support for the question “Which best represents your feelings toward the constitutional right of a woman to choose to have an abortion until about 24 weeks into her pregnancy?”, whereas 55.4% of the sample responded with a 4 or higher to the question “Which best represents your feelings toward Roe v. Wade?”. This difference echoes slightly lower rates of support for Roe v. Wade found in Harris polls as compared to Pew surveys (Taylor, 2009), in that the Harris Poll item does not include the phrase “Roe v. Wade.” Their item asks: In 1973, the U.S. Supreme Court decided that states laws which made it illegal for a woman to have an abortion up to three months of pregnancy were unconstitutional, and that the decision on whether a woman should have an abortion up to three months of pregnancy should be left to the woman and her doctor to decide. In general, do you favor or oppose this part of the U.S. Supreme Court decision making abortions up to three months of pregnancy legal? (Taylor, 2009, p. 2).
Rates of US adults indicating they favor this part of the decision have not exceeded 60% since 1992 and dipped below 50% in 2006 (Taylor, 2009). In contrast, in 2019, 70% of US adults responded “Not overturn” to Pew’s question “Would you like to see the Supreme Court completely overturn its Roe versus Wade decision, or not?” (Doherty et al., 2019). Thus, differences in Roe v. Wade items especially related to specific reference to Supreme Court cases and amount of information provided appear to affect levels of support expressed, both for the US population more broadly and for the present sample of US Latinxs.
The implications of the present findings are, of course, tempered by acknowledgement of the study’s limitations. In particular, this study adopted a quota-based sampling design to purposefully increase sample diversity and, specifically, to obtain a larger Latinx sample; thus, the present results cannot be interpreted as nationally representative. Additionally, the use of an online sample aggregator and web-based survey limits our sample to those with access to and relative facility with technology and the internet. Given the focus on variables related to acculturation, this is a notable weakness, as it is possible the present sample differs substantially along acculturation-related variables from a non-web-based sample. Additionally, given the variability of abortion-related policy across Latin America, future studies that explore the role of ethnic origin among US Latinxs may provide additional nuance to the discussion of knowledge, attitudes, and contexts of socialization. Finally, it should be noted that recent research on political knowledge “gaps” between US Latinxs and Whites has found that question biases may be inflating actual knowledge differences (Pérez, 2015). Future research on abortion-related knowledge may do well to explore additional types of questions to examine whether the inter- and intra-group differences observed here and in Jozkowski et al. (2020) remain.
Conclusion
The present study investigated knowledge of and attitudes toward Roe v. Wade among US Latinxs and the relationship among these measures and generational status as well as survey language. Results indicated that the majority of the Latinx participants answered four of five Roe v. Wade knowledge questions correctly, and overall attitudes about abortion were more supportive than unsupportive. Differences were found both in knowledge of and attitudes toward Roe v. Wade based on respondents’ generational status and choice of survey language, with choice of survey language being a stronger predictor of abortion attitudes than generational status. This study adds to prior findings by indicating that knowledge of Roe v. Wade, which differs statistically across generational status and survey language, significantly predicted abortion attitudes for this Latinx sample even though the number of generations one has been in the US did not. The fact that the present study used a tiered design, ensuring that all participants were informed about Roe v. Wade prior to assessing sentiments toward the decision also enhances our understanding of the relationship between existing knowledge and sentiments toward Roe v. Wade among US Latinxs. The present study contributes to our understanding of US abortion attitudes as well as of US Latinx public opinion more broadly and encourages continued detailed investigation of the attitudes of this growing and increasingly politically powerful population.
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a confidential foundation. The funder did not have any involvement in the study design, data collection, analysis, interpretation of the data, writing the report, and the decision to submit the article for publication.
