Following the United States Supreme Court’s reversal of Roe v. Wade in 2022, dozens of elected officials from more than half the states across the country made a joint decision. 

Ninety elected prosecutors – district attorneys, mainly – issued a statement promising that, despite laws going into effect, they would not prosecute abortion charges. 

“Prosecutors have so much discretion and power, and they can determine the criminal justice sanctions at the local level,” says Yu-Hsien Sung, a doctoral candidate in USC’s political science department. 

Sung researches the unique discretion these officials have in their roles, and her findings so far have garnered international attention.

“In the U.S. there are around 2,300 district attorneys, or local prosecutors. Most of them are elected and some of them are appointed,” Sung says. “I study what influences their policymaking and behavior in office.” 

One of the most striking things to Sung is that in the United States, citizens can vote for their district attorneys and other judicial positions.  

“In most similar democracies, these positions are appointed,” Sung says. “But if people can vote for them, then their behavior should be a response to the public and they may feel electoral pressure. That’s quite unique to the U.S.” 

Sung has a background in comparative and judicial politics, having earned a master’s in political science in her home country of Taiwan. In studying judicial politics at USC, she found that few political scientists have researched prosecutor elections in the U.S. 

Surprising survey results

Sung created a survey experiment to study how voters draw on the information available to them during an election to choose a prosecutor. 

In some elections, voters may get to hear from robust campaigns where they learn whether candidates are tough on crime or more reformed. In others, voters only learn their backgrounds and party affiliations. In still others, prosecutors’ parties may not appear on the ballots. 

Sung’s online survey recreated these election scenarios. One of the questions she wanted to answer was whether voters even pay attention to the positions held by judicial candidates. 

“I found that my survey respondents do care about issue positions,” Sung says. “They would try to pick the candidate that matched their policy preferences.” 

“When they don’t have that information, they use the candidate’s party label to guess which one is closest to their preference.” 

Sung says her respondents generally assumed that Republican candidates would be more tough on crime while Democratic candidates would be reformed and progressive in prosecutorial policy. 

However, voters were willing to vote opposite their usual partisan affiliation if they learned that the candidate’s policy stance aligned with their own. 

“Republican respondents who prefer reform-minded policy showed greater probability of voting for reform-minded candidates – even for Democratic candidates – compared to tough-on-crime respondents,” Sung says. 

“This is one of the results that surprised me.”

More information means more accountability

Another surprise in her findings, and in her broader research, is the growing public support and shifting political preference toward more progressive policies for policing and prosecution. 

Her findings defy the long-held view that the majority prefers a tough-on-crime stance. 

In fact, increasingly voters and prosecutors from both sides of the partisan divide are embracing reform in how the criminal justice system operates. 

“I think one thing influencing this shift is the effort of social activists who are working on criminal justice reform,” Sung says. “But also, over the decades, people have noticed that the U.S. has this high incarceration rate, and people find that this isn’t working to fix the issues.” 

Sung’s research also includes a database project funded in part by the National Science Foundation, the American Political Science Association and USC SPARC. In this research, Sung collects local prosecutor positions on drugs and incarceration – and whether they provide treatment programs or diversion programs – and classifies the data accordingly. 

“This new collection of data is a big contribution to the field,” Sung says.

Her research will help the political science field and the public have a better understanding of how these officials are making decisions, what the effects of those decisions are and who – namely, the voters – can hold them accountable. 

“When people know their actual positions, they will vote for them according to these issues. This in turn signals public preference to prosecutors and can aid in holding them accountable,” she says. 

“It has very great implications to the current political landscape for our criminal justice issues.”