AAPOR 2023

Dennis, B., Bueno, X., Jozkowski, K., Crawford, B., Turner, R., Lo, WJ. (2023, May 10-May 12). Using hypotheticals in Interviews on Abortion. American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA.

Early on, researchers discovered that one’s talk about their attitudesand one’s behavioral manifestations of their attitudes often did notmatch up. In other words, asking directly about one’s attitudes doesnot necessarily reflect attitudes that might be implicit to one’s behavior.Because attitudes are not directly observable, researchers must figureout how to study them. Most of the research on attitudes has beenconducted with questionnaires through large-scale surveys. Whilesome survey research has posited scenarios through which respondents are given the opportunity to manifest their attitudes asresponses to those scenarios, typically hypotheticals are not used.The interviews analyzed for this paper were part of a large multi-method study of U.S. attitudes on abortion. In these interviews, weasked interviewees to imagine that they were making a documentaryor film about abortion. We used follow-up questions to dive moredeeply into the details of their imagined films. These imagined filmswere remarkably similar despite how interviewees located themselveson a spectrum of abortion attitudes from anti-abortion to pro-abortion.The authors analyzed the responses using a metaphorical analysis(Lakoff and Johnson 1987) and an analysis of meaning fields (Dennis 2020, Carspecken 1996) to reconstruct (Dennis 2020, 2018) the: structures of an imagined ideal woman, erasure of pregnancy/the body, and privileging of the imagined potential of the fetus over the pregnant person. These themes emerged implicitly through the hypothetical opportunity to talk about abortion, and were rarely made explicit themselves by interviewees.

Jozkowski, K.N., Crawford, B.L., Bueno, X., Mena-Melendez, L., Buyuker, B. Turner, R.C., & Lo, W. (2023, May 10- May 12). People's Knowledge of and Attitudes Toward Abortion Laws Before and After the Dobbs v. Jackson Women’s Health Organization Decision: An Overview of Three Studies. Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

Although the media response to the Dobbs v. Jackson Women’s Health Organization decision was widespread, the extent that people were aware of the Mississippi law leading to the decision, the Dobbs v. Jackson case, and the resulting effects of the decision on legal abortion is unclear. To examine public opinion toward the decision and people’s awareness of and attitudes toward abortion legality, we conducted three unique studies, each comprising multiple waves of data collection to examine change in awareness and attitudes over time during the Dobbs v. Jackson decision announcement.

Study 1 was first fielded on June 6, 2022, prior to the decision announcement and again at the end of October 2022, approximately four months later, using Ipsos’ KnowledgePanel (N=1,014). Data for Studies 2 and 3 were collected using NORC’s AmeriSpeak Omnibus. The first wave of Study 2 (N=1,002) was in the field when the decision was announced such that half the sample was collected prior to the announcement and the other half was collected immediately following the announcement, allowing us to make comparisons between the two sub-samples. The second wave of Study 2 (N=1,067) was collected at the end of October—the same time wave 2 of Study 1 was fielded. Finally, Study 3 comprised five waves (N1=1,018; N2=1,070; N3=1,005; N4=1,014; N5=1,003) of cross-sectional data collected in two-week intervals following the decision announcement to examine potential changes in attitude trends over this 10-week period. This presentation will be the first in a panel examining changes in awareness and attitudes over time and pre and post the decision announcement. In this presentation, we will provide an overview of the methodology and descriptive information of the samples from all three studies, laying the foundation for the remaining presentations which will provide more in-depth findings from different aspects of the overall study.

Bueno, X.; Mena-Melendez, L.; Jozkowski, K.N.; Crawford, B.L.; Turner, R.C.; Lo, W. (2023, May 10- May 12). Becoming More Supportive or More Opposed. Has the Dobbs v. Jackson Decision Changed People’s Abortion Attitudes? a Longitudinal Assessment.  American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

On June 24, 2022, the Supreme Court of the United States overturned Roe v. Wade, the 1973 court decision that protected the constitutional right of a woman to have an abortion for almost fifty years. Landmarked legislative events of this magnitude do not generally go unnoticed but rather have the potential to strengthen people’s opinions or prompt attitudinal changes in public opinion. We conducted a longitudinal survey on abortion attitudes through the Ipsos Knowledge panel. We fielded data for wave 1 with 1,045 US adults in June 2022, right before the Dobbs v. Jackson decision. Data from wave 2 were fielded with the same sample in October 2022. In this study, we first use wave 2 to examine people’s awareness and attitudes towards the Dobbs v. Jackson decision as well as participants’ reasoning and perceptions of attitudinal change using open-ended data. Second, using both waves, we assess whether participants underwent changes in their abortion views across waves, that is, before and after the Dobbs decision was announced. In testing attitudinal changes over time, we will explore changes in (1) support for abortion legality (e.g., (il)legal in all cases, in most cases) and (2) people’s identification with the pro-life/pro-choice abortion labels. Third, using common sociodemographic predictors of abortion attitudes, we will determine which factors are associated with changes of greater opposition to, or greater support for, abortion after the overturn of Roe v. Wade.

Mena-Melendez, L.; Bueno, X.; Jozkowski, K.N.; Crawford, B.L.; Turner, R.C.; Lo, W. (2023, May 10- May 12). Does the Supreme Court decision in Dobbs v. Jackson Women’s Health Organization influence people’s attitudes and sentiment about abortion? American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

Background: On June 24, 2022, the U.S. Supreme Court ruled on Dobbs v. Jackson Women’s Health Organization (i.e., Dobbs v. Jackson), which resulted in the overturn of Roe v. Wade. While one of the most sweeping legislative changes to abortion policy in 50 years, it is unclear how the ruling influenced people’s abortion attitudes and sentiment.

Study Aim: Using this ruling as a natural experiment, we examined changes in people’s attitudes and sentiment towards abortion before and after the Dobbs v. Jackson decision. Specifically, we tested for changes in support/opposition to abortion by examining attitudes towards abortion circumstance (e.g., fetal defect, women’s health, low income, rape ) and gestational timing (e.g., 6 weeks, 15 weeks, 22 weeks, 23+ weeks).

Data and Methods: We collected two waves of data via Ipsos’ KnowledgePanel. We assessed participants’ knowledge and awareness of and attitudes toward abortion and abortion laws. Wave 1 (N=945) was collected from June 9-24, 2022 (prior to the decision announcement) and Wave 2 (N=700) is currently underway, comprising a sub-sample of Wave 1 participants to examine potential changes in people’s awareness, knowledge, and attitudes after the decision. We will conduct pre-post within-group comparisons using data from the samples before (Wave 1) and after (Wave 2) the decision. Specifically, we will examine the effect of the decision on changes in people’s abortion attitudes regarding support/opposition for abortion by (1) abortion circumstance and (2) gestational timing. Using common sociodemographic predictors of abortion attitudes (e.g., education, religion, political affiliation), we will determine which factors are associated with changes in abortion attitudes after the decision.

Implications: Assessing people’s abortion attitudes pre-post the Dobbs v. Jackson decision can elucidate the extent that this major judicial decision is immediately salient to the general public and in which directions and for whom change may occur.

Buyuker, B.E.; Hoffacker, C.; Jozkowski, K.N.; Crawford, B.L.; Turner, R.C.; Lo, W. (2023, May 10- May 12). Linking Perceptions about the Role of Federal and State Government and the Support for Texas and Mississippi Laws in the US  American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

Individual-level characteristics such as gender, education, level of religiosity, party identification, and political ideology are key predictors of public support for legal abortion (e.g., Lizotte 2015; Smith, 2016: Jozkowski et al., 2018). However, this research does not directly investigate how individual attitudes toward government involvement influence the support for abortion legality or whether there are differential attitudes toward federal- or state-level involvement.
A second line of research documents that the U.S. public infamously mistrusts government (e.g., Lerman, 2019), and the public’s view of government involvement differs depending on the specific policy issue (Pew Research Center 2020). For instance, most U.S. adults indicate support for major government involvement in issues like public health and immigration (Pew Research Center 2020), while a libertarian minority favors less government involvement regardless of the issue (Fleurbey, 2018). Yet, this research does not extend its findings to the issue of how people perceive the government’s role in abortion.
By drawing on this research, we examine whether and to what extent public attitudes toward federal and state government involvement are associated with supporting abortion legality, in connection with Texas and Mississippi laws in the U.S. To address this question, we conducted a longitudinal study by using a nationally representative sample of English and Spanish-speaking U.S. adults via Ipsos’ Knowledge Panel. We fielded Wave 1 with 1,045 US adults in June 2022, right before the Dobbs decision, and wave 2 with the in October 2022. Findings from this study may deepen understanding of the public’s view on federal and state government’s role in regulating abortion in the context of increased legislative attempts at the state level to limit abortion access and the US Supreme Court’s decision on Dobbs v. Jackson.

Crawford, B.L., Jozkowski, K.N., Turner, R.C., & Lo, W. (2023, May 10- May 12). Before and After Dobbs v. Jackson: The Importance of Abortion as a Social and Political Issue. Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

Abortion attitudes are complex. Understanding the importance of abortion as a social, political, and voting issue can be difficult. People frequently indicate abortion is a significant social issue; however, compared with other social issues, people typically rank abortion near the bottom. As such, people may recognize abortion as an important issue in the abstract but give it little weight relative to other issues. However, on June 24, 2022, the Supreme Court's decision in Dobbs v. Jackson Women's Health Organization resulted in the overturning of Roe v. Wade and the subsequent removal of the constitutional protection for a pregnant person to choose to have an abortion until about 22-24 weeks. Within 100 days of the decision, 66 clinics across 15 states stopped offering abortions with further restrictions likely to be passed soon.
 
Given this unprecedented change to the abortion policy landscape in the United States, there is may be a shift in people’s perceptions of the importance of abortion compared with other issues. We examined shifts in the importance of abortion across seven waves of cross-sectional data collected via NORC's Amerispeak Omnibus Survey (n=1,005 – 1,070). Across all waves, we asked participants to rank the importance of six social issues, including abortion. Data were collected directly preceding the Dobbs v. Jackson decision and, in the months, leading up to the 2022 midterm elections. Overall, we observed an immediate increase in people’s ranking of abortion as important after the decision and a further increase in the three waves leading into the midterms. In this presentation, we will (a) highlight changes in the importance of abortion overall, (b) explore changes in importance over time across subgroups such as gender and political affiliation, and (c) discuss what our findings may signal regarding the importance of abortion moving forward.

Hawbaker, A., Mena-Meléndez, L., Jozkowski, K.N., Turner, R.C. (2023, May 10- May 12). Varying Interpretations of the term "Any" in Abortion-Specific Survey Instrument Items: A Qualitative Examination Using Cognitive Interviews. American Association for Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA.

Background: Since 1977 the General Social Survey (GSS) has asked whether it should be possible for a pregnant woman to obtain a legal abortion if “the woman wants it for any reason.” The term “any” in attitudinal questions may have multiple meanings. Some may interpret “any” as all-inclusive whereas others interpret it as any single reason (reason specific). Although often conceptualized as all-inclusive, in 2016 and 2018, the GSS “any reason” item was endorsed at a rate equal to or higher than three other items (Smith et al., 2018), indicating the all-inclusive conceptualization may not be universal.

Purpose: The purpose of this study was to determine whether respondents interpret the “any reason” abortion item as all-inclusive or reason specific. Doing so will help establish validity and clarity about findings for researchers who use this item.

Methods: We administered a survey comprising abortion attitude items, including the “any reason” question to US adults via NORC’s AmeriSpeak panel (N= 1,025). From these participants, we conducted 51 cognitive interviews with English- (n=36) and Spanish-speaking (n=15) respondents to assess interpretation of items, including “any reason.” Interviews were analyzed using Applied Thematic Analysis.

Results: We found that a majority of respondents interpret “any reason” as reason specific rather than all-inclusive. Results show that there is not widespread understanding of “for any reason” as all-inclusive. Rather, participants opt for a more nuanced interpretation of “any” as a single reason that may not include abortion legality conditions.

Discussion & Conclusion: Our findings complicate the common “all-inclusive” interpretation of the “any reason” item in the GSS Core Abortion Questions. These findings elucidate why “any reason” sometimes receives endorsement at rates equal or higher than other reasons in the same question battery. Researchers may consider updating their use of the “any reason” condition to account for multiple interpretations.

Paiz, J.Y., LaRoche, K.J., Jackson, F., Crawford, B.L., Turner, R.C., Lo, W., & Jozkowski, K.N. (2023, May 10- May 12). Betwixt and between: A Mixed Methods Study of People Who Simultaneously Identify As Pro-Life and Pro-Choice. American Association for Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA.

Introduction: The terms “pro-life” and “pro-choice” (which we refer to as abortion identity labels) are closely associated with abortion. Researchers commonly assess abortion identity via presenting “pro-life” and “pro-choice” as mutually exclusive response options on surveys. However, such dichotomous response options may not adequately represent people’s abortion identity. Thus, the purpose of this study was to better understand people who identify as both pro-life and pro-choice to some degree (i.e., dual endorsers).

Method: We administered a survey to US adults (n=580), assessing abortion identity via two different response formats: (1) a single bidirectional categorical item with response options ranging from strongly pro-choice to strongly pro-life (and “both” and “neither”) and (2) two separate slider items which assessed the extent people identified with each term on a 0 (not at all pro-life/pro-choice) to 6 (completely pro-life/pro-choice). Participants with dual endorsement on the slider items (i.e., participants who identified as both pro-life and pro-choice to some degree) were asked via an open-ended item to explain their response We used logistic regression to examine the relationship between participants’ socio-demographic characteristics and the likelihood of dual endorsement and thematic analysis for the open-ended data.
Results: 64% of participants indicated dual endorsement by identifying as both pro-life and pro-choice to some extent. We found conceptual discrepancies between people’s abortion identity endorsement on the close-ended items and their response to the open-ended questions. Dual endorsers described abortion as a serious and undesirable option but reflected positively on notions of personal choice and bodily autonomy. Men were more likely than women (OR = 0.40, p<.01), and Black/African American participants were more likely than white participants (OR = 2.70, p<.05), to report dual abortion identities.

Implication: Our findings demonstrate complexity in measuring people’s endorsement of abortion identity labels. We recommend a more nuanced approach to assessing abortion identity.

Asamoah, N.A., Turner, R.C., Lo, W.J., Crawford, B.L., & Jozkowski, K.N. (2023, May 10- May 12) Do Demographic Predictors of Neutral Response Options Differ by Topic?  American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

To authentically measure attitudes about social issues, researchers have implemented and refined several measurement techniques over the years. An example is the “neutral response, no response, or no opinion” option. Select research (e.g., Lizotte & Sidman, 2009; Krosnick et al., 2001) discourage the use of neutral response options with knowledge questions, or recommends incorporating neutral responses in the measurement models, as there are gender and education differences for those selecting a “no response” option even when knowledge level may not differ. Respondents may also select a neutral option as a way to satisfice or avoid socially undesirable responses on sensitive social issues (Krosnick & Presser, 2009). However, there are also cases where it can be important to provide respondents with a “no response” option when asking opinions about sensitive topics or when respondents may not have sufficiently developed an opinion (Montagni et al., 2019; Tourangeau, 1984).
In this study, we investigate what demographic characteristics of a respondent, if any, predict who is most likely to choose a neutral response option on scales measuring attitudes about abortion -- a topic sensitive for some respondents. We will compare these results to responses to a gun control/gun rights attitudinal scale. It is hypothesized that topics potentially subject to gendered attitudes may differ in their “no response” distributions. We plan to study how the choice of the neutral option is affected by the contents of the scale, their relevance to respondents, and the position of the neutral option among the other response options. Data are from a GfK panel of over 1500 people. The primary purpose of the study is to examine if prior results to knowledge and political attitudes items are generalizable to the topics in this study.

Turner, R.C., Jozkowski, K.N., Lo, W., Crawford, B.L., & Buyuker, B.E. (2023, May 10- May 12) Impacts of Pregnancy and Abortion Timelines, Safety, and Fetal Milestone Questions on Recommended Abortion Legality Timeframes. Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

Does having participants answer questions about 1) when people commonly find out they are pregnant, 2) when certain milestones occur for a fetus, and 3) how safe abortion and pregnancy are for a woman, inform them about [or influence] when, in a pregnancy, they think abortion should be legal? We conducted an experiment using a nationally representative sample [IPSOS KnowledgePanel, 2022, n = 1014] to compare attitudes about how early into a pregnancy the government [US or state] should be able to ban abortion. Group 1 answered the pregnancy, abortion, and fetal milestone questions prior to receiving attitudinal questions about when abortion should be banned. Group 2 received the abortion ban questions prior to the pregnancy and abortion perception questions.
Comparing Responses to Pregnancy and Abortion, Fetal Milestone, and Safety Variables: There were no significant differences between the two experimental groups on perceptions of how far into a pregnancy women typically learn they are pregnant, the most common pregnancy timeframes for abortion, occurrences of fetal milestones, and the safety of abortion and pregnancy.
Comparing Responses to Abortion Bans: There were significant differences between the two experimental groups on when, in a pregnancy, federal and state governments should be able to ban abortion. The group asked about pregnancy and abortion safety, fetal milestone, and pregnancy awareness prior to abortion ban questions indicated that abortion bans should begin later in a pregnancy than participants who were asked after the abortion ban questions. The results indicate that priming individuals to think about pregnancy and fetal timelines may impact responses to when government banning of abortions should be allowed. Additional information about US adults' beliefs about pregnancy and abortion safety, fetal milestone occurrences, and pregnancy timelines will be presented, along with a comparison of missing responses by item groups across experimental conditions.

Buyuker, B.E.; Hoffacker, C.; Jozkowski, K.N.; Crawford, B.L.; Turner, R.C.; Lo, W. (2023, May 10- May 12). The Public’s Perceptions about the Role of Federal and State Government and Support for Legal Abortion in the U.S. American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA.

Individual-level characteristics such as gender, education, level of religiosity, party identification, and political ideology are key predictors of public support for legal abortion (e.g., Lizotte 2015; Smith, 2016: Jozkowski et al., 2018). However, this research does not directly investigate how individual attitudes toward government involvement influence the support for abortion legality or whether there are differential attitudes toward federal- or state-level involvement.
A second line of research documents that the U.S. public infamously mistrusts government (e.g., Lerman, 2019), and the public’s view of government involvement differs depending on the specific policy issue (Pew Research Center 2020). For instance, most U.S. adults indicate support for major government involvement in issues like public health and immigration (Pew Research Center 2020), while a libertarian minority favors less government involvement regardless of the issue (Fleurbey, 2018). Yet, this research does not extend its findings to the issue of how people perceive the government’s role in abortion.
By drawing on this research, we examine whether and to what extent public attitudes toward federal and state government involvement are associated with supporting abortion legality, in connection with Texas and Mississippi laws in the U.S. To address this question, we conducted a longitudinal study by using a nationally representative sample of English and Spanish-speaking U.S. adults via Ipsos’ Knowledge Panel. We fielded Wave 1 with 1,045 US adults in June 2022, right before the Dobbs decision, and wave 2 with the in October 2022. Findings from this study may deepen understanding of the public’s view on federal and state government’s role in regulating abortion in the context of increased legislative attempts at the state level to limit abortion access and the US Supreme Court’s decision on Dobbs v. Jackson.

Jozkowski, K.N., Mena-Melendez, L., Crawford, B.L., Turner, R.C., & Lo, W.J. What happens when abortion is illegal? Attitudes toward illegal abortion, responsibility for illegal abortion, and perceived punishments of illegal abortion. American Association of Public Opinion Research, Philadelphia, PA.

Since the Dobbs v. Jackson Women’s Health Organization decision resulted in overturning Roe v. Wade (i.e., the Supreme Court decision establishing abortion prior to viability as constitutionally protected), many states have and will continue to pass legislation criminalizing most abortion. These laws articulate punishments applied to certain people (i.e., actors) involved in illegal abortion with differential severity. These actors can include (1) pregnant women, (2) healthcare providers, (3) men involved in the pregnancy, (4) those who provided information or assistance to the pregnant woman, and (5) parents of a minor seeking abortion. Certain laws stipulate specific punishments that may be applied to some or all of these actors with differential degrees of severity. Given current and expected future increases in legislation criminalizing abortion occurring in different circumstances (e.g., at different weeks’ gestation) and different actors involved in the abortion (e.g., pregnant woman, healthcare provider), the purpose of this study was to assess public opinion regarding punishments for illegal abortion. Specifically, we examined public opinion regarding who should be held most responsible for illegal abortion, which abortion circumstances people believe should be illegal, and the type of punishment that should be enacted for illegal abortions.
An online survey was administered to 2,489 English and Spanish-speaking US adults, using quota-based sampling, with post-stratification weights applied to match US benchmarks. People indicated healthcare providers and women were more responsible than other actors. Although the plurality did not support abortion being illegal for most circumstances, a substantial proportion were unsure. Additionally, the plurality of participants did not endorse a punishment for the woman or healthcare provider. However, those who endorsed “any punishment,” compared with “no punishment,” typically indicated the woman and healthcare provider were most responsible. Our findings suggest that current laws criminalizing abortion are misaligned with public opinion regarding punishments for illegal abortion.

Bueno, X.; Mena-Melendez, L.; Jozkowski, K.N.; Crawford, B.L.; Turner, R.C.; Lo, W. (2023, May 10- May 12). How Do People Interpret Pro-Life and Pro-Choice Terms in Spanish? a Mixed Method Approach. American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

In the US, abortion attitudes are often categorized using the terms “pro-life” and “pro-choice”. However, with a growing Spanish-speaking population in the country and increasing multilanguage -English and Spanish- survey research, it is important to explore whether these US-centric abortion-related terms and their common translations as pro-vida and pro-elección are interpreted similarly by Spanish-speaking –and sometimes foreign-born and monolingual– populations in the US. To explore the interpretation of pro-vida and pro-elección in Spanish as compared to pro-life and pro-choice in English, we analyze data from an online survey to N=1,025 US adults in English (n=821) and Spanish (n=204) followed by 15 cognitive-like personal interviews to a subsample of the Spanish-speaking respondents. This mixed method approach allowed us to (1) quantitatively examine patterns of response to abortion labels-related survey items across English and Spanish respondents; and (2) qualitatively assess how a sub-sample of Spanish-speaking adults in the US define the terms pro-vida and pro-elección, their familiarity with the terms, provide definitions and interpretations of the terms, perceive people who self-identify with the terms, and propose alternative Spanish terms related to abortion. Preliminary results suggest that English and Spanish-speaking respondents identify with the abortion-identity labels differently. Findings suggest that both labels entail a degree of ambiguity for some respondents. However, misinterpretations and a lack of awareness of the term pro-elección were more common among Spanish-speaking respondents. This study provides important insight on how the terms pro-vida and pro-elección are understood by Spanish-speakers in the US and has significant implications for survey research methodology related to multilingual polling on the topic of abortion.

Turner, R.C., Crawford, B.L., Lo, W., Hadfield, J., & Jozkowski, K.N. (2023, May 10- May 12) Do Attitudes about Government Involvement and Personal Freedom Account for Variation in Abortion Attitudes Beyond Political Affiliation?  Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

Research indicates that demographics such as gender, age, educational level, religiosity, and political affiliation are significantly related to attitudes about abortion legality with additional research branching into personality characteristics related to abortion support vs opposition (e.g., Osborne et al., 2022). Research has also indicated that people's attitudes about the morality of abortion do not always align with their attitudes about whether abortion should be legal (e.g., Jones & Cox, 2012; Montenegro et al., 2022). In this study, we investigate two constructs that may inform about differences in attitudes about the morality and legality of abortion. These include attitudes about the personal rights and freedoms of others (regardless of whether we agree with the behavior or belief) and attitudes about the degree government should be involved in social issues in the US.
Using a nationally representative sample of 1,025 US adults (NORG Amerispeak© panel), we investigate the constructs of belief in personal freedom and government involvement in social issues, as predictors beyond the commonly applied demographic and political affiliation variables, in predicting attitudes about the morality and legality of abortion. It is hypothesized that participants who believe in the rights of others to engage in behaviors that they may not agree with (e.g., offensive religious speech, offensive racial speech, assisted suicide, stem cell research) will have higher abortion legality scores than those who do not, controlling for other demographic differences. Similarly, participants who are against increased government involvement in social issues (e.g., gun ownership, poverty, racism, police reform) are hypothesized to have higher abortion legality scores than those in favor of increased government involvement. Conversely, these two variables are hypothesized to be less predictive of attitudes about the morality of abortion. This study is intended to provide additional insight into the complexity of belief systems that influence abortion attitudes.

Crawford, B.L., Jozkowski, K.N., Turner, R.C., & Lo, W. (2023, May 10- May 12). Do Perceptions of When Fetal Development Markers Occur during a Pregnancy Influence Their Attitudes Toward Fetal Development-Based Abortion Bans? Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

Since 2010, there has been a significant increase in state-level legislation limiting access to abortion across in the United States. With the 2022 ruling in Dobbs v. Jackson and subsequent overturning of Roe v. Wade, states are now able to pass even more restrictive legislation including near-total bans of abortion. The focus of many of these abortion laws and the associated discourse is the timing and presence of fetal milestones, such as when a heartbeat can be detected or when a fetus is thought to be capable of feeling pain. Yet, it is unclear when people think these milestones occur, if/how those perceptions may be related to abortion attitudes, and whether the relationship between perceptions of fetal development and attitudes toward fetal milestone-based legislation is moderated by ideology.

We assessed (a) people’s perceptions regarding when various fetal development milestones such as fetal heartbeat and pain occur, (b) the relationship between these perceptions and attitudes toward fetal development-based bans, and (c) how ideology moderates that relationship. We use survey data collected from IPSOS’s Knowledge Panel (N=860) in 2022 directly before the Dobbs v. Jackson decision. We assessed people’s attitudes toward abortion in the context of weeks’ gestation and fetal development milestones, and the extent people identified as pro-life, pro-choice, both, or neither. First, we will use crosstabs to highlight differences in perceptions of the timing of fetal development across abortion ideology. Next, we will use logistic regression to assess if perceptions of fetal development significantly predict support for laws banning abortion at 6 weeks and 15 weeks; we will also examine ideology as a moderator. As states continue to pass legislation restricting abortion using both weeks’ gestation and fetal milestones, these findings will be useful to highlight the extent that laws align with public opinion.

Mena-Melendez, L.; Jozkowski, K.; Crawford, B.; Turner, R.; Lo, W.J. (2023, May 10 - May 12): Public Opinion on Abortion Illegality: Attitudes Toward Abortion Being Illegal and Perceived Punishments of “Illegal Abortion." American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA.

Background: A bulk of research and polls assessing abortion attitudes has focused on whether or not abortion should be legal. However, research about attitudes toward whether—and how—abortion actors should be punished if abortion became illegal remains largely unexplored. Since the June 24, 2022, U.S. Supreme Court’s decision in Dobbs v. Jackson Women’s Health Organization and the subsequent overturning Roe v. Wade, states have and will continue to pass legislation criminalizing most abortions. Thus, it is most imperative to assess public opinion regarding endorsement of different punishments for “illegal abortions”.

Study Aim: We will assess public opinion regarding endorsement of different punishments (e.g., therapy/education, fines, jail/prison, death penalty) for seekers and providers of an illegal abortion.

Data and Methods: We administered an online survey to English (n = 2,204) and Spanish-speaking (n= 285) US adults via Qualtrics’ national panels in the spring of 2021. We used descriptive statistics to examine endorsement of punishments for the two actors and a series of multinomial logistic regression models to assess the likelihood of endorsing various punishments (versus no punishment) across gender, age, race, Hispanic ethnicity, education, political affiliation, bible literalism, abortion legality measure, living in a trigger law state, abortion hostility environment, abortion responsibility, Rule Conditionality Scale (RCS), and Perceived Obligation to Obey Laws (POOL). We also visualize the results using each model's marginal predictions.

Results: We find significant differences in the endorsement of different punishments by the abortion legality measure and living in a trigger law state. Generally, participants were more likely to endorse punishments for the abortion seeker compared to the provider. In stark contrast with current legislation in many states—where abortion is now illegal, and punishments include significant jail/prison time and fines—we found limited support for these forms of severe punishment.

Lo, W. J., Turner, R. C., Crawford, B. L., & Jozkowski, K.N. (2023, May 10 - May 12). Investigating design effect on public opinion on abortion using the traditional and structural equation modeling approach. Poster for American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA 

The survey data are commonly presented in frequency or analyzed using traditional approaches, such as a t-test or an analysis of variance to test of differences in means. However, these traditional approaches have their limitations, such as using composite scores instead of item level scores and needing to meet model assumptions. Therefore, the purpose of this study is twofold. First, we aim to compare two analysis approaches – the traditional approach and the more flexible structural equation modeling (SEM) approach – and to evaluate the potential SEM advantages that test mean difference on latent variables. Second, we conducted an experiment to examine whether presented gestational age could impact people’s attitudes toward abortion using six General Social Survey (GSS) abortion items.

A total of 1,006 participants were recruited through the Prolific panel and randomly assigned to one of three sets of online surveys. Each set had 12 GSS abortion items – 6 items in the original wording and 6 items that added “…in 4 month pregnant” at the end. The first set had dichotomous options of Agree/Disagree. The second set had 4-point response options by adding “strongly agree” and “strongly disagree”. The third set was on the 6-point Likert type scale with three adverbs (i.e., strongly, moderately, and slightly) added to both agree and disagree categories. For analyses, we computed the composite scores underlying Muthen’s (1981) two-factor structure for the original GSS wording and questions that specified the gestational age. Then, we calculated the mean differences between questions with and without gestational age wording. Second, we used the latent path modeling underlying the SEM framework to estimate the predicting differences between latent structures (i.e., with and without presented gestational age). The full paper presents a discussion of the findings and common practice.

Buyuker, B.E.; Mena-Melendez, L.; Bueno, X.; Jozkowski, K.N.; Crawford, B.L.; Turner, R.C.; Lo, W. (2023, May 10- May 12). Cross-Culturalism and Abortion Attitudes amongst Latinxs in the U.S. American Association of Public Opinion Research (AAPOR) Annual Meeting, Philadelphia, PA.

Previous studies documented differences in abortion attitudes across racial/ethnic groups, with U.S. Latinxs showing overall lower endorsement for legal abortion compared with other groups (Carter et al., 2009). The growing and diverse Latinx population in the U.S. requires a more nuanced understanding of inner-group differences in abortion attitudes. However, only a handful of studies have explored the relationship between cross-culturalism and abortion attitudes (e.g., Amaro, 1988). A more nuanced understanding of how cross-culturalism interacts with abortion attitudes is important to (1) identify whether (and how) the different forms of cultural identity may elucidate another layer of complexity in abortion attitudes among U.S. Latinx, (2) achieve a better understanding of how to assess abortion attitudes in the U.S. from a more inclusive perspective, sensitive to cultural diversity; and (3) develop cross-language survey measures which take into account the complexity inherent to culturally diverse populations.
We explored the relationship between cross-culturalism and support for legal abortion among U.S. Latinx respondents by comparing the effects of four previously validated acculturation-related scales—BASH, PAS, AMAS-ZABB, and ARSMA-II on attitudes toward legal abortion considering specific circumstances. We fielded a survey with 1,183 self-identified Latinx respondents in 2021 using a non-probability, quota-based sample provided by Qualtrics. We found that increased engagement with Anglo-culture was associated with greater support for legal abortion among Latinx. Disregarding the circumstance for seeking an abortion, the BASH scale—based on English/Spanish language usage—represents the most consistent predictor of support for legal abortion compared with scales oriented to measure cultural identity—i.e., AMAS-ZABB or ARSMA-II. However, the latter are positively correlated with support for legal abortion in the most endorsed circumstances such as fetal anomalies and rape/incest. These findings provide helpful insights when conducting multi-cultural public opinion research on abortion attitudes among the U.S. Latinx population.