Why cybercrime victims doubt the police can help: a 1,500-person US survey on trust in cyber policing

The question: as cybercrime has grown into one of the most commonly experienced forms of victimization in the United States, do people actually believe the police can help them if it happens? And separately, do they want the internet policed more proactively? Rachel McNealey, Camille Figueroa and Cooper Maher set out to measure both attitudes directly, and to identify what predicts them.

Why it matters: perceived police efficacy is one of the strongest known predictors of whether a victim bothers to report a crime at all. Maher's own prior analysis of identity theft found only about 10% of online identity theft was reported to police, compared with nearly 46% of offline identity theft. If a large share of the public already doubts police can help with cybercrime before they are ever victimized, that belief itself may be suppressing the reporting that would otherwise generate accurate crime statistics and justify better-resourced cybercrime units.

What the researchers did. The team used the Reactions to the Threat of Crime in Offline and Online Spaces dataset, a cross-sectional survey of 1,500 US adults fielded by YouGov in December 2021, deliberately oversampling rural and urban respondents (500 each from urban, suburban and rural areas) and weighted to match national demographics on vote history, gender, age, race, education and rurality. Two outcomes were measured with single survey items: agreement with "If I am the victim of a cybercrime, law enforcement will be able to help me," and agreement with "The internet should be policed more to prevent crime." Both were modelled using ordinal logistic regression against three types of prior cybervictimization (computer-focused, interpersonal, financial), emotional fear of cybervictimization, perceived risk of cybervictimization, and a set of sociodemographic and general police-attitude controls.

What they found. Respondents who had experienced computer-focused victimization (malware, account theft, phishing) were significantly less likely to believe police could help them (odds ratio 0.65), while respondents who had experienced interpersonal cybervictimization (harassment, stalking) were more likely to believe police could help (odds ratio 1.37) but less likely to support broader internet policing (odds ratio 0.66). Emotional fear of cybervictimization was associated with more positive views on both outcomes. General attitudes toward police mattered more than any cybercrime-specific factor tested: respondents who rated police as generally effective in their community were substantially more likely to believe police could help with cybercrime specifically (odds ratio 1.88, the largest effect in the model). Black respondents were significantly more likely than White respondents to believe police could help (odds ratio 1.51) and to support more internet policing (odds ratio 2.04); men were less likely than women and non-binary respondents to agree with either statement.

What this does NOT establish. The underlying survey data was collected in December 2021, nearly five years before this piece; cybercrime's prevalence, public awareness of it, and police cybercrime capacity have all continued to shift since, particularly with the rise of AI-assisted scams, and attitudes measured then may not describe attitudes today. Both outcome variables were measured with a single survey item each rather than a validated multi-item scale; the authors themselves note this was a deliberate trade-off given how little prior work exists to draw a validated scale from, not an oversight, but it does mean each measurement is less precise than a multi-item index would be. The design is cross-sectional and correlational: the study can show that fear of cybercrime and support for more policing move together, but it cannot establish that one causes the other. The sample is US-only, so international generalisation is not established by this data, though the authors note comparable single-item measures have been used in the EU's Eurobarometer survey. On the race findings specifically: the paper's own interpretation, that these patterns may reflect minority respondents' heightened perceived vulnerability to online abuse and fraud, is explicitly offered as the authors' interpretation rather than something the survey data itself can confirm, and this piece treats it the same way.

Bizarus interpretation. The standout finding is that cybercrime-specific trust in police appears to be inherited from general trust in police rather than built independently, general police efficacy was the single strongest predictor in the model. That has a direct practical implication: a police department cannot fix public confidence in its cybercrime response by cybercrime-specific messaging alone if the underlying general relationship between that department and the community is weak. It also matters that different types of cybervictimization pulled attitudes in different directions, computer-focused victims lost faith in police specifically, while interpersonal-victimization victims retained faith in police help but grew more skeptical of broader internet policing, which is consistent with computer-focused crimes being widely seen (by researchers and victims alike) as technically beyond what local police are equipped to investigate.

What remains unanswered. Whether this pattern holds with 2026-era data, given how much the cybercrime landscape has shifted since December 2021, is untested. Whether the race and gender differences observed here would replicate in a larger or more recent sample is unknown. And because this study measured attitudes rather than behaviour, whether higher trust in police actually translates into higher rates of reporting cybercrime, the outcome that matters most for improving crime statistics, was outside its scope and remains a separate empirical question.

Source record. McNealey, R. L., Figueroa, C. I., and Maher, C. A. "Police can't help you": Exploring influences on perceptions of policing cybercrime. Journal of Criminal Justice, November-December 2025 issue. https://www.sciencedirect.com/science/article/pii/S0047235225001916. Journal of Criminal Justice, Law category, Q1 (SCImago, 2025); also Q1 in Social Psychology and Sociology and Political Science, Q2 in Applied Psychology (SCImago, 2025). Full text was openly accessible and read in full by Bizarus.