Support for abortion access is broadly shared and firmly held across a wide range of demographics. Opinions are not easily moved — however, there is opportunity to influence civic action. 

This is according to recent research conducted by Supermajority Education Fund and Incite Studios on April 21 and 22, 2022 — prior to any news of the leaked SCOTUS opinion to overturn the constitutional right to an abortion. 

We asked a national sample of 8,135 people to tell us how much they agreed or disagreed with the statements below. We then tested six different videos to learn if we could shift opinion. Answers reflect agreement/disagreement scores on a scale of 0 to 100. 

Do you agree or disagree: All women should have access to safe and legal abortion care.

Baseline agreement that all women should have access to safe and legal abortion care scores 67.9 overall. 

Do you agree or disagree: The government has the right to limit or restrict women’s access to abortion. 

Baseline disagreement that the government has the right to limit or restrict access to abortion scores 60.1 overall. 

These baselines are consistent with other recent research on abortion, including Navigator’s finding that 3 in 5 people in our country describe themselves as “pro-choice,” and Pew Research’s recent finding that at least 61% of people in our country believe abortion should be legal in all/most cases. 

And importantly, none of the six video messages we tested produced statistically significant change in agreement/disagreement with the above statements in any group — an indication of how firmly views on abortion are held and how little room for persuasion there is on this subject. However, one video did produce meaningful shifts in civic behavior metrics. (More information on that video to come later.)

In our findings, women show slightly more agreement (+4.7) with abortion access than men (scoring 70.2 compared to 65.5), but there is broadly shared support across all major demographic traits like age, race, geography, income, and educational attainment. 

In fact, the only demographic for whom agreement scores fall below 50 are people who describe themselves as “Very Conservative,” and even then, support for this group is at 48. Even self-reported 2020 Trump voters score their support for abortion care at 54.8. 

Support for access to abortion care is particularly high among women of color and young women under the age of 35, whose scores come in at 74.2 and 74.6 respectively.

People in the U.S. disagree with government restrictions on abortion.

The same trends hold true for the government’s right to limit or restrict abortion — women show more disagreement (+8.6) with government restrictions on abortion than men (scoring 64.2 compared to 55.6), with minor difference across other demographics. Once again, scores only dip below 50 for “Very Conservative” folks and Trump 2020 voters. 

And once again, disagreement with government restrictions is particularly high among women of color and young women under the age of 35, whose scores come in at 64.2 and 63.3 respectively.

While opinions on abortion access and government restrictions held relatively firm, we learned there is opportunity to influence people’s civic behaviors on the issue of abortion access. 

Importantly, this first round of testing was conducted prior to the news of SCOTUS’s potential decision to overturn Roe v. Wade, which would remove the constitutional right to an abortion. Now that this issue has more often been discussed in the news, we suspect future rounds of testing are likely to yield different results. Other research, including from Navigator, indicates that the leaked SCOTUS opinion has increased motivation to vote for many demographics, and we would expect to see similar shifts in future rounds of testing.  

Even still, in the pre-leak environment, we found that one asset in particular produced statistically significant results on metrics like: 

• Likelihood to volunteer or donate to a cause or nonprofit
• Likelihood to vote in the 2022 midterm elections
• Agreement with the statement “Individuals can affect change”

The asset, titled At Least She Has Support, leveraged creative storytelling to approach this seemingly fixed topic in a new and fresh way. 

First, we don’t reveal that the video is about abortion until the very end, leveraging the element of surprise. Second, our video centers on a male storyteller, with a script that focuses on his support for a woman seeking an abortion. We suspect that the reveal and unexpected main character helped this video break through in ways other, perhaps more expected content did not.  

This video increased the scores for likelihood to volunteer or donate among Young Women (<35 years old) by +4.0 points overall. Importantly, this increase was statistically significant across virtually every demographic variable (like age, race or ethnicity, income, region, education, etc.), although it appears to have been particularly effective in motivating:

• Young Asian women (+5.9)
• Young Hispanic or Latino women (+4.7)
• Young White women (+3.6) 
• Young women who identified their race/ethnicity as “Other” (+5.7)

Interestingly, this same video increased agreement that an individual can affect change for Women of Color (+2.1) and Young Women (+3.0). The largest increases in agreement are seen with:

• WOC under age 35 (+3.6) 
• WOC in the Northeast (+3.5) and West (+3.6) 
• Women earning >$100k (+3.0) 
• Asian women (+3.1) 
• Hispanic or Latino women (+3.5)
• WOC with college degrees (+3.3)
• Atheist WOC (+3.3) and Catholic WOC (+3.1) 
• WOC who did not vote in 2020 (+2.7) 
• WOC who are Unlikely & Potential voters (+2.2 each)

However, this video decreased likelihood to vote scores for some women of color. The score for our overall Women of Color category decreased by 2.9 points; however, this drop is not shared across all women of color. The only statistically significant drops in vote likelihood are seen among the following subsets of women of color:

• Black women (-4.9) and Native American/Indigenous women (-2.9)
• WOC over age 55 (-3.6)
• Rural WOC (-5.7) 
• WOC in the South (-4.4)
• WOC who describe themselves as “Very Conservative” (-4.4)
• Evangelical Protestant or Other Protestant WOC (-4.6 and -4.1)
• Likely Voter WOC (-4.1) 
• WOC who voted for Donald Trump (-5.3)

This suggests that a focus on abortion messaging may impact older, more conservative, more rural, and more religious women women of color differently than other women of color. More research is needed to better understand this.

We are interested to re-test in the new post-SCOTUS leak environment to see if results significantly shift. Stay tuned for results from future testing.