(Sex) Crime and Punishment in the #MeToo Era: How the Public Views Rape

Certain legally irrelevant features of rape (such as details relating to the victim's gender and criminal record) decrease the likelihood of reporting an incident to the police or its perception as severe and worthy of punishment.

• Cases containing details that are construed as victim blaming (that is, it contained details that could be used to blame the incident on the victim such as the victim’s choice of clothing or location of the crime) are less likely to be recommended for reporting or severe punishment.
  • Incidents occurring at parties are 6 percentage points less likely to be recommended for reporting, and 17.6 percentage points less likely to be recommended for severe punishment, than incidents occurring in the victim’s home.
  • Incidents occurring in parks are 10.4 percentage points less likely to be recommended for severe punishment than incidents occurring in the victim’s home.
  • Incidents with victims wearing night club attire at the time of the incident are 4.8 percentage points less likely to be reported than incidents with victims wearing work attire.
• Cases with details understood to imply consent (previous relationship with the perpetrator; past sexual history) are less likely to be recommended for reporting or for severe punishment.
  • Incidents in which victim and perpetrator are acquaintances are 11.8 percentage points less likely to be recommended for reporting, compared to cases where the victim and perpetrator are strangers.
• Incidents in which victims have multiple sexual partners are 9 percentage points less likely to be recommended for reporting and 9 percentage points less likely to be recommended for severe punishment, compared to incidents with married victims. Similarly, incidents with victims who are single are 5.4 percentage points less likely to be reported, compared to incidents with married victims.
• Cases with details calling into question victims’ credibility (such as if victims have a minor felony record) are 6.4 percentage points less likely to be recommended for reporting.

Additional findings relate to the identity of the victim:

• The victims’ gender is one of the most salient factors influencing people’s decision-making.
  • People are 16.4 percentage points less likely to report and 5.6 percentage points less likely to recommend for severe punishment in cases involving male rather than female victims.
• People exhibit racial sympathy by being more sympathetic to both Black victims and Black perpetrators.
  • People are 4.7 percentage points more likely to recommend for severe punishment if the victim is Black rather than white.
  • Similarly, people are 5.0 and 6.2 percentage points, respectively, less likely to recommend for reporting  or punishment when the perpetrator is Black than when they are white.

In this study, researchers investigated how numerous legally-irrelevant details about a rape incident affects public opinion on reporting and punishment of the incident. The four empirically measurable dimensions of rape culture that the study focused on were: victim blaming, empathizing with perpetrators, assuming the victim’s consent, and questioning the victim’s credibility.

The study’s 5400 participants were recruited from a nationally representative survey from NORC’s Ameri-Speak panel and from convenience samples from Amazon Mechanical Turk (MTurk). The participants were randomly sorted to receive prompts about a rape incident or an armed robbery incident. The case also had randomly assigned legally-irrelevant characteristics such as: the victim’s sex, race, past relationship history, criminal record, the perpetrator’s race, socioeconomic status (as signaled by profession), and prior relationship to the victim, along with other circumstantial details such as the incident location and victim’s attire at the time of the incident. To more closely mirror rape cases in reality, all the perpetrators were designated to be male, 70% of the victims were designated to be female and 30% as male.

The combination of the nine attributes used in the study generated 1728 different profiles that were tested. The participants were all given five pairs of cases. In the first study, they were asked to choose which case they would be more likely to report, and in the second study, they were asked which case to recommend for a more severe punishment had they been a jury member and the perpetrator was found guilty. A similar procedure was repeated with the MTurk participants, who were also then asked to explain their choices in an open-ended question section, yielding detailed qualitative data on the justifications for decisions.

The advantage of the conjoint study design is that by forcing participants to choose between two cases that were only distinguishable in their legally irrelevant factors, it could illustrate evidence of systematic bias based on any statistically significant differences that emerged in which cases were chosen more frequently.