Meeting you was a fake: investigating the increase in romance fraud during COVID-19
Buil-Gil, D. ; Zeng, Y. (2021) — Journal of Financial Crime
Type:
Journal Article
Country:
United Kingdom
Synopsis (AI-Generated)
This study modeled monthly U.K. Action Fraud romance-fraud reports from April 2014 through December 2020 and Understanding Society estimates of loneliness and daily internet use. ARIMA forecasts based on pre-pandemic trends showed that reported romance fraud rose sharply after March 2020 and remained above the upper prediction interval through December; the increase was more abrupt among younger adults. Only people aged 16-29 showed loneliness above forecast levels, while adults 50 and older, especially those over 70, showed unusually large increases in daily internet use. These parallel time trends do not establish that loneliness or internet use caused fraud growth. Underreporting, limited subgroup periods, a single loneliness item, and sparse internet-use waves constrain interpretation.
Identified Gaps (AI-Generated)
Before this study, official UK sources indicated increased romance fraud after March 2020, but no evaluation had established its extent or whether it reflected COVID-related restrictions rather than pre-existing growth. The study also identifies a need for individual-level evidence on relationships among loneliness, internet use, and fraud victimisation, because its aggregate time-series approach cannot test causal or correlational associations. More detailed demographic analyses and cross-national replication are needed.
Methods (AI-Generated)
The study analysed monthly UK Action Fraud romance-fraud reports from April 2014–December 2020 and Understanding Society survey estimates of loneliness and daily internet use from January 2017–November 2020. Univariate ARIMA models, fitted to pre-COVID trends, generated 95% prediction intervals to assess whether pandemic-period values exceeded expected variation. It compared detailed Action Fraud records for November–December 2019 versus the same months in 2020 using Poisson mean tests. Analyses were conducted in R using the forecast package.
Limitations (AI-Generated)
Police-recorded fraud data are affected by victim underreporting and police underrecording, creating measurement error. Detailed victim characteristics were available only for the final 13 months, limiting subgroup comparisons to November–December 2019 and 2020. The design cannot establish causal or correlational relationships between loneliness, internet use, and romance fraud. Internet use could only be compared with September 2020 because the measure appeared in only COVID survey wave 5. Loneliness was measured by one frequency item rather than the 20-item UCLA scale.
Future Work (AI-Generated)
Use survey data and individual-level survey microdata to assess COVID-19’s effect on fraud and test associations between loneliness, internet use, and victimisation. Use the 20-item UCLA loneliness scale where feasible. Replicate the study in other countries and examine variation by victims’ sex, ethnicity, and language skills. The authors also recommend awareness campaigns, cross-jurisdictional investigative collaboration, partnerships with NGOs and educational institutions, and platform flagging systems.
AI-Generated Content Notice
The synopsis and research notes on this page were generated with AI from available publication information and, when available, the uploaded paper text. They may contain errors, omissions, or interpretation issues. Readers should follow the DOI or source link, review the original publication, and make their own judgment about the content.
Found a possible error? Request a correction.