Correction to “Capturing Adults' Familiarity With Financial Fraud: Relationships With Targeting, Engagement, and Victimization”

Abstract

Başaran Şahin, D., O. Valdes, G. R. Mottola, et al. 2026. “ Capturing Adults' Familiarity With Financial Fraud: Relationships With Targeting, Engagement, and Victimization.” Journal of Consumer Affairs 60, no. 2: e70051. https://doi.org/10.1111/joca.70051.

We noticed small formatting mistakes in our published paper. We apologize for these errors.

Error #1: In Table 6, the coefficients corresponding to consumer-based, ID-based and threat-based were mixed up. Our results and their interpretation have not changed; however, their presentation in Table 6 is erroneous. The coefficients for Hispanic/Latino were missing from the table as well.

Error #2: In the “Results” section paragraphs related to Table 4 and Table 6, some coefficients and their statistical significance are not accurate and appear to report findings from an earlier version of the paper. Our results in the paper have not changed. The changes we are making are very minimal.

Corrected Paragraphs

6.3 RQ2: What Demographic, Socioeconomic, and Psychosocial Factors Are Associated With Knowledge of Distinct Fraud Types?

Table 4 presents coefficients from logistic regressions predicting the odds of listing each fraud type. Adults aged 40–64 and 65–90 are significantly more likely (OR = 2.97, p < 0.01, and OR = 4.08, p < 0.001, respectively, Column 3) than adults aged 21–39 to list consumer-based fraud, after controlling for other factors in the model; however, we do not find age differences in the odds of listing other fraud types. Compared to men, women are more likely to list imposter-based fraud (OR = 1.9, p < 0.05, Column 5) but less likely to list ID-based scams (OR = 0.58, p < 0.01, Column 4). Hispanic adults are less likely to mention ID-based fraud compared to NH White adults (OR = 0.53, p < 0.05, Column 4). While adults in the NH American Indian, NH Alaskan Native, NH Asian, NH Pacific Islander, and NH Other group are less likely to mention threat-based fraud (OR = 0.21, p < 0.01, Column 1), they are more likely to mention opportunity-based fraud (OR = 4.22, p < 0.001, Column 2). It is worth noting that the sample size for this racial/ethnic group is the smallest (N = 88), and findings should be considered accordingly.

Among socioeconomic factors, we found that an increase in the number of correct financial literacy answers significantly increased the odds of mentioning most fraud types, including threat-based (OR = 1.87, p < 0.001, Column 1), ID-based (OR = 1.68, p < 0.001, Column 4), imposter-based (OR = 1.53, p < 0.01, Column 5). We conducted supplemental analysis to test whether the number of fraud types that respondents listed would affect the relation between financial literacy and the likelihood of mentioning each fraud type. We found that the association of financial literacy with the increased likelihood of mentioning threat-, ID- and imposter-based fraud was no longer statistically significant. One possible explanation is that individuals with higher financial literacy are also the ones who are more aware of or knowledgeable about financial fraud and therefore more likely to report it. Results from this supplemental analysis did not yield meaningful changes in other variables (see Supporting Information Analyses Table S1). Adults with the most education (with a college degree or more) are more likely to list consumer-based fraud (OR = 1.72, p < 0.05, Column 3) compared to those who had high school education or less but less likely to list threat-based fraud (OR = 0.45, p < 0.05, Column 1). Family income is significantly related only to the likelihood of listing threat-based fraud for individuals whose family income was between $50,000 and $99,000 (OR = 2.79, p < 0.01, Column 1) and those with a family income of $100,000 or more compared to those who had less than $50,000 (OR = 2.49, p < 0.01, Column 1). We did not find a statistically significant association between fraud type mentions and loneliness.

Corrected Table

6.4 RQ3: How Does Knowledge of Financial Fraud Types Relate to Personal Fraud Experiences, Including Being Targeted by, Engaging With, and Losing Money to Fraudsters?

Table 5 provides a descriptive view of those reporting being targeted by, engaging with, and losing money to fraudsters. We found that individuals who reported being targeted by fraudsters (Column 1) do not generally differ from those who reported engaging with the fraudsters (Column 2). Furthermore, these two subsamples are similar to the full sample, as shown in Table 1, with the exception that the subsamples in Table 5 have a somewhat lower share of high school or less education than does the full sample (approximately 10.5% vs. 13.6%). That being said, individuals reporting losing money show some differences (Column 3, Table 5). They are slightly older (mean age of 63.4 vs. 60.2 in the full sample), slightly more female (59.5% vs. 57.8% in the full sample), and have a smaller share who are married or living with a partner (47.5% vs. 58.5% in the engaging and 62.1% in the targeted samples). Individuals reporting losing money to fraud have the highest share who answered none of the financial questions correctly and the lowest share who answered all of them correctly (55.0% in Column 3 compared to approximately 62% in Columns 1 and 2). Finally, as we move from being targeted to engaging to losing money to fraud, we see progressively higher rates of self-reported loneliness.

Next, we examined each of these associations in a multivariate sense, controlling for all other factors. Table 6 shows the odds ratios when predicting being targeted by (Column 1), engaging with (Column 2), and losing money to fraudsters (Column 3) within logistic regressions. Our primary goal in this second set of regression models was to examine whether knowledge of specific types of fraud was associated with individuals’ experiences with fraudsters after controlling for demographic and socioeconomic factors and loneliness. Specifically, knowledge of consumer-based and ID-based fraud was both significantly associated with reporting being targeted by a fraudster (OR = 1.67, p < 0.05 and OR = 2.98, p < 0.001, respectively, Column 1). Perhaps surprisingly, knowledge of any of the five fraud types was generally unrelated to reporting engaging with a fraudster. Knowledge of threat-based fraud, on the other hand, was negatively associated with reporting losing money to a fraud (OR = 0.54, p < 0.05, Column 3).

Table 6 also provides a multivariate view of demographic factors associated with being targeted by, interacting with, or losing money to a fraudster. For example, adults aged 65–90 were almost twice as likely as those aged 21–39 to report ever having been targeted by a fraudster (OR = 2.03, p < 0.05, Column 1), perhaps not surprising given having lived longer, and even more likely to report losing money to fraud (OR = 3.2, p < 0.05, Column 3). Whereas NH Black adults were more likely than NH White adults to report being targeted by a fraudster (OR = 2.49, p < 0.05, Column 1), there were no differences between NH Black and NH White adults in terms of their odds of engaging with a fraudster or losing money to fraud. And whereas Hispanic adults tended to report engaging more with fraudsters relative to their NH White counterparts (OR = 2.30, p < 0.01, Column 2), they had lower odds of reporting losing money to fraud (OR = 0.33, p < 0.01, Column 3). We did not find any differences by gender or marital status.

Higher levels of education were positively associated with odds of being targeted by fraudsters (OR = 4.0, p < 0.001 and OR = 2.31, p < 0.01, respectively for adults with some college education or college or more, compared to adults with high school or less, Column 1). In contrast, there were no differences by education in reporting engaging with fraudsters or losing money to scams. Family income was not associated at all with fraud experiences after controlling for other factors. While having higher levels of financial literacy was related to higher odds of reporting engaging with fraudsters (OR = 1.42, p < 0.01, Column 2), it had the opposite association with odds of reporting losing money to fraud (OR = 0.65, p < 0.05, Column 3). Finally, adults who reported feeling lonely were significantly more likely to engage with a fraudster (OR = 2.16, p < 0.05, Column 2), controlling for all other factors. However, there was no association between feeling lonely and losing money.

Error #3: We noticed that in the “Introduction” section, three in-text citations and the corresponding full citations under “References” are not accurate.

i. Corrected in-text citation: In 2024 alone, over 2.6 million fraud complaints were reported to the Federal Trade Commission (FTC), totaling $12.8 billion in losses (Federal Trade Commission 2025).

ii. Corrected in-text citation: In fact, the FTC estimates that after adjusting for under-reporting the true cost of fraud may be as high as $196 billion (Federal Trade Commission 2025).

Corrected full citation: Federal Trade Commission. 2025. “Protecting Older Consumers 2024-2025: A Report of the Federal Trade Commission.” December. https://www.ftc.gov/system/files/ftc_gov/pdf/P144400-OlderAdultsReportDec2025.pdf

iii. Corrected in-text citation: Several studies have reported that knowledge of financial fraud, and in particular, the schemes and tactics used by fraudsters, can be an important protective mechanism against victimization (Burke et al. 2022; Federal Trade Commission 2024).

Corrected full citation: Federal Trade Commission. 2024. “A Review of Scam Prevention Messaging Research: Takeaways and Recommendations.” Federal Trade Commission. April. https://consumer.ftc.gov/system/files/consumer_ftc_gov/pdf/A%20Review%20of%20Scam%20Prevention%20Messaging%20Research.pdf.