PurposeThis work uses survey data from the Portuguese Securities Market Commission (Comissão de Mercado de Valores Mobiliários – CMVM) to examine financial literacy and literacy bias. The main objective of this study is to shed light on this issue by identifying the individual characteristics that are associated with financial literacy, namely overconfidence and underconfidence, which in turn might help explain individuals' financial decisions. The study distinguishes two groups, i.e. students and nonstudents, and considers several characteristics that are usually employed in this stream of research.Design/methodology/approachThe data are based on a survey conducted by a partnership between the CMVM and a consortium of Portuguese universities. This paper has a three-fold aim. First, it studies the main individual features associated with objective financial literacy. Second, it analyzes the relationship between those variables and the bias between self-perceived and objective literacy, distinguishing overconfidence and underconfidence. Third, and most originally, this framework was also used to examine the differences between students and nonstudents. Those aims are pursued using cross-sectional ordinary least squares (OLS) regressions, except for the study of the literacy bias, for which the authors use an ordered probit.FindingsLiteracy is higher in individuals of the male gender who are older, have higher incomes, live in metropolitan areas, are highly educated, have a field of study related to finance and have high self-perceived literacy. Younger people are more overconfident. Unconditionally, women are less overconfident than men, but conditionally, they overestimate their knowledge. People holding securities and with a field of study related to finance are more overconfident. The gender effect is mainly driven by students, and the impact of a field of study and of holding securities on overconfidence decreases and increases, respectively, for students. The results highlight the importance of financial education.Research limitations/implicationsDue to the way that the questionnaire was made available, there is no guarantee that the sample is representative of the Portuguese general population, or, for that matter, representative of the typical Portuguese retail investors or households. Also, there is no guarantee that the same individual did not answer the questionnaire more than once, although this is highly improbable. The link to the online questionnaire was only transmitted within e-mail databases owned by the CMVM and Portuguese universities, so the authors cannot guarantee its unbiasedness.Practical implicationsThe authors' results may help the National Plan for Financial Education (the acronym in Portuguese is PNFF) fine-tune the required actions towards different target groups and, most importantly, highlight that different groups may require different approaches aiming to narrow the gap between objective and perceived literacy. The first step should be creating procedures to provide feedback on the objective and perceived literacy of those who enroll in financial formation programs.Social implicationsThe study distinguishes two groups, students and nonstudents, providing additional insights that might guide policymakers on how to structure financial education to enhance individual financial behavior. This is especially important in a country such as Portugal which has the lowest objective financial literacy in the Eurozone.Originality/valueThis study contributes to the financial literacy literature, in particular to the stream of research that focuses on psychological biases, by shedding light on the factors associated with both individual overconfidence and underconfidence. Differentiating between students and nonstudents provides additional insights, which might guide policymakers on how to structure financial education to enhance individual financial behavior.