April 3, 2023

Opinion Diagnostics, a Boston-based polling and market research firm, conducted a poll of 475 likely Massachusetts Republican Presidential primary voters from March 31 – April 1, the two days following the announcement that former President Donald Trump had been indicted in New York. 

Media inquiries can be directed to Brian Wynne at brian@opiniondiagnostics.com

Topline Results: Click here to download

ClearTabs: Click here to download


Quick Toplines

Bar graph of poll results described above
Bar graph of poll results described above


Full Summary of Results

The poll indicates that Trump holds a significant lead in the 2024 Massachusetts Republican Primary election, garnering 45% of the electorate’s support compared to 21% for Florida Governor Ron DeSantis. 15% of respondents chose either former UN Ambassador Nikki Haley, former Vice President Mike Pence, or a different candidate, and 19% of respondents are undecided. Trump’s lead is driven by his support from registered Republican voters (+35% over DeSantis) and senior citizens (+37% over DeSantis). While the Massachusetts Republican Party has yet to set rules for awarding delegates to the Republican National Convention, under past party rules Trump, DeSantis, and Haley would meet the 5% support threshold to be allocated delegates.

DeSantis’ standing improves in a hypothetical two-way matchup against Trump. In that scenario, Trump receives 46% of the electorate’s support, compared to 32% for DeSantis and 22% undecided. DeSantis would earn the majority of those that would otherwise be inclined to support Haley. While still trailing Trump, DeSantis performs relatively well among unaffiliated voters (-5% behind Trump), among men (-10% behind Trump) and in the northern and eastern parts of the Boston media market (-5% behind Trump, consisting of Middlesex, Essex, and Suffolk counties).

While Trump’s lead is significant, his failure to reach a 50% majority support threshold in either the four-way or two-way primary matchup indicates that he is far more vulnerable than in the 2020 Presidential Primary, where as sitting President he earned 87% of the vote. These results suggest the 2024 primary is more likely to resemble the 2016 primary election, where then-candidate Trump received 49% of the vote in a race that featured 13 candidates listed on the ballot.

78% of Republican primary voters believe that Donald Trump’s indictment in New York City is not justified, and is instead a politically motivated attack by a Democratic district attorney, based on what they have seen, read, and heard about the indictment. 11% believe the indictment is justified and 11% are undecided. No demographic subgroup of Republican primary voters believes the indictment is justified, and opposition to the indictment is particularly strong among registered Republicans and senior citizens, with 84% of both groups believing the indictment is not justified.

A 46% plurality of the Republican Primary electorate say that the indictment will have no effect on who they plan to support in next year’s primary election. Among those whose vote is affected, 42% say they are more likely to support Trump after the indictment compared to 12% that say they are less likely to support him. However, while this increase in Trump support following the indictment may serve to firm up his existing base, it is unlikely to gain Trump any new supporters beyond the level measured in this poll: only 6% of those that have said they are “more likely” to support Trump following the indictment are still undecided.


Significance of the Massachusetts Republican Presidential Primary

The Massachusetts Republican Presidential Primary will be held in less than a year on March 5, 2024. Although Republicans are historically uncompetitive in Presidential general elections in Massachusetts, the MA primary is more significant in the context of the national Republican presidential primaries. Massachusetts awards the 26th most delegates to the Republican National Convention, slightly above the median of the combined 56 states and U.S. territories that send delegates. Additionally, Massachusetts awards more Republican delegates than any other state in New England, nearly double the number awarded in New Hampshire. The Massachusetts primary falls early on the calendar, with only five states scheduled to select delegates before March 5, 2024. Finally, in the first 32 state primary elections and caucuses held during the 2016 primary season, then-candidate Donald Trump received his highest share of the vote in Massachusetts, with 49% support.


Poll Methodology

Opinion Diagnostics conducted a poll of likely Republican Presidential Preference Primary voters in Massachusetts. The survey was fielded on March 31 and April 1, 2023, beginning the day after the indictment of former President Donald Trump was announced. The sample size of the poll was n=475 likely primary voters and the margin of error is ±4.5%. The poll was conducted via mixed mode methodology with outreach via live operator calls, IVR calls, and outbound SMS messages directing to an online form. The sample was screened for likelihood to vote in Massachusetts Republican Presidential Primary elections. Respondent data was processed with Multi-Electorate Weighting™ to match the demographics of those that voted in the 2012, 2016, and 2020 Massachusetts Republican Presidential Primaries, as well as the average demographic attributes of the electorate across those three elections. For convenience, results here are reported according to the average weighting model. Results are presented in a toplines document and in ClearTabs™, Opinion Diagnostics’ proprietary crosstabulation format.


Opinion Diagnostics’ advantage in polling this election

The MA Republican Presidential Primary has a small and highly variable electorate. During the competitive 2016 primary election, 638,000 voters participated, about 15% of all registered voters and 23% of eligible voters (registered Republicans and unaffiliated voters). In contrast, during the 2020 election where President Trump’s renomination was a foregone conclusion, only 277,000 voters participated, about 5% of all registered voters and 9% of eligible voters. Mixed-mode polling methodology can acquire data on such small samples of the population with cost-effectiveness that is unachievable by pollsters that rely exclusively on live operator calling. Pollsters that rely on online opt-in panels are also unlikely to obtain accurate results because there are too few overall panel participants to create a statistically valid subgroup of Republican primary voters.

The variability of the electorate in MA Republican Presidential Primaries is driven by the ability of unaffiliated voters to choose to participate in either the Republican or Democratic primary, but not both. In 2012 and 2016 where there was no Republican incumbent, more than two-thirds of Republican primary voters were unaffiliated, as opposed to registered Republicans. That share dropped significantly in 2020 when President Trump’s renomination was a foregone conclusion, and unaffiliated voters could instead participate in the competitive Democratic Presidential Primary.

Highly variable elections like these are best analyzed with Multi-Electorate Weighting™, which gives a side-by-side display of poll results weighted to multiple past representative elections. The individuals that respond to public opinion polls are never perfectly reflective of those that typically vote in elections. Many pollsters adjust results by giving a higher weight to the responses of harder-to-reach subgroups, and a lower weight to the responses of easier-to-reach subgroups. When deciding how much to increase or decrease the weight of each response, pollsters can use a single past election as a proxy, conduct a “registered voters” poll and use the characteristics of those registered to vote, or attempt to make an educated guess about the characteristics of those that will vote in a future election. All three methods are flawed. Using a single past election fails to capture the variability that can occur between high and low turnout elections; registered voter polls can be valuable for polling about public policy issues, but are inappropriate for gauging elections where a substantial proportion of voters will not turnout; educated guesses are often influenced by groupthink among pollsters and firms, for example when pollsters collectively predicted a massive ‘red wave’ in 2022 that failed to materialize. Multi-Electorate Weighting™ solves these problems by displaying multiple past proxy elections, which allows the pollster to consider the variability of the voting electorate without resorting to guesswork.

Finally, Opinion Diagnostics has a keen understanding of Republican primary elections and nationally. The Opinion Diagnostics methodology was initially developed during the 2012 Presidential cycle, and Opinion Diagnostics staff formerly served in senior positions in the Massachusetts Republican Party and are familiar with the dynamics of Republican Primary elections in the Commonwealth.