Introduction

Since its inception in 2008, blockchain technology has led to the rise of digital currencies, commonly known as cryptocurrencies. In general, cryptocurrencies are digital currencies that utilize blockchain technologies (Buchmann, 2004). These digital assets have gained a lot of attention as a new investment vehicle in the financial market, along with other developments in financial technology (FinTech) (Liu & Serletis, 2019; Woebbeking, 2021; Yen & Cheng, 2021). For instance, the leading cryptocurrency, Bitcoin, has experienced exponential growth since its inception in 2008 and has become a novel form of tradable digital asset worldwide (Nakamoto, 2008).

The rise of cryptocurrencies has sparked a broad array of public perceptions, ranging from enthusiastic adoption to skepticism and concern. On one hand, cryptocurrencies adopt a peer-to-peer scheme under which the transaction could be completed instantly (Hashemi Joo et al., 2020), which provides a more efficient payment channel to conduct business. Such efficiency and the ability to bypass bank transaction fees underscore the revolution of the financial system transitioning into a decentralized form (Makarov & Schoar, 2022). On the other hand, anonymity afforded by cryptocurrencies has raised regulatory concerns as it complicates regulatory efforts to oversee the large volume of transactions, including potential illegal operations (Choi, 2018). This multifaceted public perception has been shaped by various sociopolitical, economic, and psychological factors, contributing to the complex narrative surrounding cryptocurrency adoption.

Despite the extensive discussion around public perceptions, there remains a critical gap in understanding the specific characteristics that drive individuals to become cryptocurrency investors. Understanding the characteristics of cryptocurrency investors is important, as the value of cryptocurrencies is primarily determined by supply and demand as well as investors’ perceptions of the value of cryptocurrencies (Almeida & Gonçalves, 2023; Dunbar & Owusu-Amoako, 2023; Sayim & My, 2023; Zhu et al., 2021). In addition, the high volatility of cryptocurrencies, coupled with unforeseeable market turbulence events (Demiralay & Golitsis, 2021; Mohammed et al., 2024), and the increase in criminal activities associated with cryptocurrencies (Foley et al., 2019) indicate the need for a deeper understanding of cryptocurrency investors.

Therefore, this study shifts the focus from the broad spectrum of public perceptions to a more targeted investigation of individual cryptocurrency investors by examining the characteristics of an investor that might be associated with cryptocurrency investment. This focus differs from previous studies like Gupta et al. (2021) and Fujiki (2021), which primarily examined how financial literacy or general attitudes shape intention to invest. This emphasis also differs from existing studies that investigate the how behavioral aspects, such as perceived risk and investor attitudes (Veerasingam & Teoh, 2023; Zhao & Zhang, 2021) affect cryptocurrencies investments.

Using data from the most recent cohort of the National Financial Capability Study 2021 (NFCS 2021), we find that people with a higher level of investment knowledge and investment experience are more likely to invest in cryptocurrency. Furthermore, age moderates the relationship between cryptocurrency investment and investment knowledge, which extends findings from earlier behavioral studies by highlighting how this link varies by age group. Specifically, the moderating effect reveals that younger investors are more likely to invest in cryptocurrency despite having lower levels of traditional investment knowledge, suggesting greater risk tolerance or different decision-making processes among younger cohorts. Meanwhile, the present study also shows that the commonly known determinants related to investment behaviors, such as financial knowledge and financial education, are not correlated with cryptocurrency investment, which contrasts with some prior evidence that suggests a positive link (Panos et al., 2020; Zhao & Zhang, 2021). Nevertheless, this finding needs to be interpreted with caution and merits further investigation.

The contribution of the present study is twofold. First, this paper extends prior literature by showing that investment experience and knowledge might be more significant factors that contribute to cryptocurrency investment than financial knowledge and financial education, and that the effect of investment knowledge varies by age. Therefore, the time an individual acquires investment education might be another important element that determines the likelihood of investing in cryptocurrencies. Second, by providing a clearer profile of cryptocurrency investors’ characteristics, this study helps open the “black box” of who is most likely to invest. This deeper understanding can support more effective risk management and regulatory changes related to cryptocurrency markets, while also offering practical guidance for identifying who are more likely or capable of investing in cryptocurrencies in the financial marketplace.

The remaining part of this paper is organized as follows: Section “Literature review” discusses relevant literature about cryptocurrency investment and the contribution of the present work. Section “Empirical method” introduces data, variable measurements, and the empirical model. Section “Results” presents the main results. Section “Robustness Check” shows the robustness check, and Section “Discussion and conclusion” concludes this paper.

Literature review

Many studies have attempted to identify factors that relate to cryptocurrency investment from different perspectives. For example, in the field of behavioral finance, studies have shown that perception of investment risk is the main predictor of cryptocurrency investment (Veerasingam & Teoh, 2023). This is because cryptocurrencies are often viewed as risky financial assets, and they frequently attract risk-seeking investors (Bouri et al., 2020). Furthermore, given the nature and design of cryptocurrencies, perceived profitability and security are important features that determine whether an investor would invest in cryptocurrencies (Yilmaz & Hazar, 2018). Likewise, the characteristics of investors could also relate to cryptocurrency investment. It is evident that financial literacy is correlated with investing in cryptocurrencies (Gupta et al., 2021). Nevertheless, the direction of the association between financial literacy and cryptocurrency investment is still controversial. A study focused on the financial marketplace in Japan has revealed that cryptocurrency owners have a higher level of financial literacy than non-owners (Fujiki, 2021). On the other hand, research examining investment attitudes toward cryptocurrency found that people who are more financially literate are less likely to invest in cryptocurrencies, and this is because more financially literate individuals can better assess investment risk and are often more cautious about risky assets than people who are less financially literate (Panos et al., 2020).

As an attempt to better understand the fast-growing trend of digital assets, more effort has been spent on constructing theoretical models that could help explain investment behaviors regarding cryptocurrency (Andrianto & Diputra, 2017; Jung & Seo, 2025; Lammer et al., 2020; Xi et al., 2019). In particular, Zhao and Zhang (2021) used the social cognitive theory to examine whether financial knowledge and investment experience could lead to more cryptocurrency investment. This theory posits that people’s behavior is influenced by the interaction between cognitive, behavioral, and environmental factors (Sen, 1986). Their study documented that while there is a positive association between both factors and cryptocurrency investment, investment knowledge and experience are more significant factors. Likewise, Fujiki (2021) also found that investment experience and financial literacy are connected to cryptocurrency investments.

In addition, conceptual frameworks such as the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT) offer a useful theoretical fundament for understanding how individuals decide to engage with new financial technologies or financial derivatives like cryptocurrencies. The TAM posits that perceived usefulness and perceived ease of use are central to individuals’ willingness to adopt new technologies (Davis, 1989). The UTAUT model further extends this idea by incorporating elements such as social influence, facilitating conditions, and performance expectancy to explain variation and disparity between the intention to use and actual usage (Venkatesh et al., 2003). When applied to the context of FinTech investment, these models highlight how trust, perceived risk, and self-efficacy interact with individual factors like financial literacy and prior investment experience (Chhatwani & Parija, 2023) in financial decision-making. For example, an investor’s willingness to trust a decentralized digital asset is shaped not only by their risk appetite but also by whether they perceive the technology to be reliable, secure, and beneficial compared to conventional financial investment options.

Despite these theoretical insights, much of the empirical literature on cryptocurrency investment has remained at the level of intention or perception, rather than examining which concrete investor characteristics determine actual participation. Some studies have found that higher general financial literacy can increase the likelihood of cryptocurrency investment (Fujiki, 2021), while others have suggested that more financially literate individuals may avoid speculative assets like cryptocurrency due to greater risk awareness (Panos et al., 2020). The present study bridges this gap by linking the technology acceptance perspective with behavioral finance evidence to show how investment knowledge, investment experience, and demographic factors together relate to actual crypto investment decisions. This approach helps explain why general financial literacy or perceived market conditions may not be significant on their own, but how specific experience and confidence can translate perceived usefulness into real investment behavior in an emerging and volatile market.

The present study is therefore based on literature that emphasizes building a conceptual framework of financial literacy, investment knowledge, and cryptocurrency investment (Ante et al., 2022; Chhatwani & Parija, 2023; Hackethal et al., 2022). Taking a step further, this work includes perceived market confidence as an additional covariate and tests how these factors interact with age. Prior research has shown that perceived market confidence can determine whether an investor engages in risky investments (Doran et al., 2010; Meier, 2018). By integrating this with established technology acceptance models, the study provides a more comprehensive understanding of the behavioral, cognitive, and experiential factors that shape who actually becomes a cryptocurrency investor.

Empirical method

To study the potential connection between cryptocurrency investment and variables related to financial literacy, investment knowledge, and investment experience, a logistic regression model was carried out on data extracted from the National Financial Capability Study 2021. This section introduces the data, the empirical model for analysis, and variable measurements.

Data

Data used in this study were extracted from the investor survey in the National Financial Capability Study 2021 (NFCS 2021). The NFCS is a large, nationally representative survey conducted across the United States to measure the financial knowledge, behaviors, and attitudes of American adults. Funded by the FINRA Investor Education Foundation, the survey uses a stratified random sampling design to ensure balanced representation across states, age groups, income levels, and other key demographics. The broad geographic and demographic coverage of the survey enhances the generalizability of findings related to investor characteristics nationwide. The NFCS has been widely used in recent research to investigate cryptocurrency investment behavior (e.g., Todd et al. (2024), Bai et al. (2024), and Kim and Fan (2025)). In the present study, we used the 2021 cohort, which is the latest version of the survey. Respondents with “prefer not to say” or “do not know” responses to key questions were excluded from the sample, resulting in a final analytic sample of 1810 observations.

Empirical model and variable measurements

The present study used the logistic regression model to investigate the correlation between cryptocurrency investment and variables related to investment knowledge, investment experience, financial literacy, and market confidence. The empirical mode can be described by the following functional forms:

$$Y={\beta }_{0}+{\beta }_{1}\,\cdot\, T+{\beta }_{2}\,\cdot\, X+\varepsilon$$

The variable “crypto investment” is the dependent variable \(Y\), which measures whether a participant has made investments in cryptocurrencies, either through direct means or by utilizing a fund that contains cryptocurrency investments. The dependent variable was coded as “1” if an investor has such investments and “0” otherwise.

In the equation above, \(T\) stands for a vector of covariates that are related to financial literacy, investment knowledge, investment experience, and perceived market condition.

Following the existing literature, the present study adopts variables that measure financial literacy from both subjective and objective perspectives (Bai, 2021). Table 1 summarizes how financial literacy is defined and its construct based on subjective and objective measures.

Table 1 Summary of the definition of financial literacy.
Full size table

Specifically, subjective financial literacy is measured by two variables “math skills” and “financial knowledge”. In the NFCS 2021, respondents were provided with the following two statements: “I am good at dealing with day-to-day financial matters, such as checking accounts, credit and debit cards, and tracking expenses.” and “How would you assess your overall financial knowledge?” Responses for the two statements are recorded on a scale from “1 = strongly disagree” to “7 = strongly agree” and “1 = very low” and “7 = very high”, respectively. Objective financial literacy was measured by the variable “financial education”, based on the following question: “Was financial education offered by a school or college you attended, or a workplace where you were employed?”. Answers to the question were coded as “1” if the answer was “Yes, and I did participate in the financial education” and “0” if the response was “Yes, but I did not participate in the financial education offered” or “No”.

Following Zhao and Zhang (2021), the variable “investment experience” is a binary one, which takes the value of “1” if the investor has invested in stocks, commodities, features, or options in their non-retirement account and “0” otherwise. The variable “investment knowledge” is a numerical variable based on the survey question: “On a scale from 1 to 7, where 1 means very low and 7 means very high, how would you assess your overall knowledge about investing?”. It takes values between “1 = very low” and “10 = very high”.

Perceived market conditions were measured by two variables: “market (good for investor)” and “market (well-regulated)” which were related to the following questions: “How confident are you that the U.S. financial markets offer good long-term opportunities for investors?” and “How confident are you that the US financial markets are effectively regulated to protect investors from fraud and abusive sales practices?”. The answers to those questions are recorded on a scale of “1 = not at all confident” to “10 = extremely confident.”

Finally, following the literature studying FinTech and cryptocurrency investment utilizing data from the National Financial Capability Studies (Bai et al., 2025), this paper also included the following covariates that measure other demographic information of respondents: age (18–34, 35–44, and above 44), gender (males and females), degree obtained (association degree or higher and below associate degree), marital status (married and others), income, and employment status (self-employed or work full-time and others). \(X\) in the empirical model represents a vector of those variables. For all logistic regression models, the values of the coefficients are converted into marginal effects for easier interpretation.

Results

Table 2 provides summary statistics of the probability of having cryptocurrency investments by different classes of categorical variables adopted in the regression model.

Table 2 Summary statistics results.
Full size table

For variables related to the demographic information of respondents, people with an age between 18 and 34 have the highest likelihood of 57.1% of investing in cryptocurrencies, in contrast with the respondents between 35 and 44 (37.9%) and those above 44 (9.4%). Married individuals have a 19.9% chance of investing in cryptocurrency, while the probability of investing in cryptocurrency is 29.2% for unmarried individuals. Regarding education level and employment status, people with an associate degree or higher are about 5.4% less likely to invest in cryptocurrencies, and employed individuals are around 22.4% more likely to participate in cryptocurrency investment than their unemployed counterparts.

With regard to categorical covariates that measure financial literacy and investment experience and knowledge, the summary statistics show that people who have acquired financial education either from the workplace or school have a probability of 23.2% of investing in cryptocurrencies. This probability is about 1% higher than that of individuals without a financial education background, with a 22.2% likelihood of engaging in cryptocurrency investment. On the other hand, the probability of cryptocurrency investments varies significantly based on one’s investment experience. Specifically, respondents with investment experience in risky assets such as stocks, commodities, features, or options have a 24.8% chance of investing in cryptocurrency. This probability is about 15.2% higher than the probability of those without investment experience (9.6%).

Table 3 reports the main results from the logistic regression model.

Table 3 Logistic regression results.
Full size table

Table 3 presents the results from the logistic regression models examining the relationship between cryptocurrency investment and various investor characteristics. Across all specifications, investment experience and investment knowledge consistently show significant positive associations with the likelihood of investing in cryptocurrency. Specifically, in the baseline model reported in column 1, individuals with investment experience are about 8.5% more likely to invest in cryptocurrency than those without investment experience. Likewise, investment knowledge is also positively associated with the propensity for cryptocurrency investment. Specifically, each one-point increase in self-assessed investment knowledge corresponds to a ~5% higher probability of investing in cryptocurrency.

Regarding age, compared to individuals aged between 18 and 34, older investors (especially those above age 44) are significantly less likely to invest in cryptocurrency. Specifically, as reported in the baseline model in column 1, people between 35 and 44 and those above 44 are 16.3 and 40. 3% less likely to invest in cryptocurrency than those between 18 and 34, respectively. The interaction terms in Columns 2 and 3 further reveal that the positive effect of investment knowledge on cryptocurrency investment weakens as age increases. For example, the negative and significant coefficient on the interaction between investment knowledge and the “above 44” age group suggests that higher investment knowledge does not translate into the same likelihood of crypto investment for older investors as it does for younger ones.

Figure 1 demonstrates the interaction effect between age and investment knowledge.

Fig. 1: Interaction effect on cryptocurrency investment by age.
figure 1

This figure illustrates the interaction effect of age on the association between investment knowledge and cryptocurrency investment.

Full size image

As shown in Fig. 1, while the probability of investing in cryptocurrencies varies significantly based on different levels of investment knowledge when looking at the age group 18–34, the gaps between each line reduced dramatically as we approached the age group above 44. This confirms the negative interaction term between age and investment knowledge from the regression model, indicating that the association between investment knowledge and cryptocurrency becomes less pronounced as age increases to above 44. In other words, age negatively moderates the positive linkage, where more investment-knowledgeable respondents are more likely to invest in cryptocurrency.

Meanwhile, variables traditionally linked to general investment behaviors, such as financial knowledge and financial education, do not show statistically significant relationships with cryptocurrency investment in any of the models. This finding highlights the distinct roles of investment experience and investment knowledge in affecting individuals’ propensity to invest in cryptocurrency.

Finally, we also examined the VIF of independent variables to determine whether there is an issue of multicollinearity in the baseline logistic regression model in column 1 of Table 2. As presented in Table 4, we find no independent variable has a VIF of higher than 5, indicating no multicollinearity issue in the regression analysis.

Table 4 VIF of independent variables.
Full size table

Robustness check

To examine if the interaction effect between investment knowledge and age from the main analysis is robust, we carry out the regression analysis for the three age groups. Table 5 shows the robustness check result.

Table 5 Regression by age groups.
Full size table

In this robustness check, we focus on the magnitude of the coefficients of investment knowledge from each regression model. As reported in Table 5, investment knowledge is statistically significant in all three regression models applied to different age groups. Despite this board tendency, the value of the coefficient decreased from 10.7 to 3.4% when we moved from the age group 18–34 to the age group above 44. This provides evidence that the correlation between investment knowledge and cryptocurrency investment is weaker when looking at respondents above 44 than younger age groups. As a result, we conclude that the interaction effect between age and investment knowledge identified in the main analysis is robust.

Discussion and conclusion

The primary goal of this study is to investigate the relationship between the characteristics of financial investors and the likelihood of being cryptocurrency investors, aiming to open the black box of cryptocurrencies. The inherent anonymity associated with cryptocurrencies limits our knowledge of understanding the characteristics of cryptocurrency investors. By examining the association between factors related to financial literacy, investment knowledge and experience, and the likelihood of an investor investing in cryptocurrencies, we aim to shed the on the profile of cryptocurrency investors. These findings contribute both to academic literature on behavioral finance and fintech adoption, and to the development of more targeted and effective practices in investor education and financial advising.

Our study contributes to the growing body of cryptocurrency research by offering new insights into the characteristics of cryptocurrency investors. Understanding who invests in cryptocurrencies is critical, as the traditional investors may be hesitant to enter the cryptocurrencies market, limiting the opportunity for them to reap the financial benefits of cryptocurrencies rapid growth. This hesitation can be attributed to individual herding behavior, as the existing research has pointed out (Li & Wu, 2018; Sun, 2013), where individuals are influenced by the success stories of others. Our findings show that investment knowledge and experiences play an important role in becoming a cryptocurrency investor. The findings suggest that experienced and knowledgeable investors may feel more comfortable venturing into the relatively new and volatile realm of cryptocurrencies. We found that having investment experience in risky assets such as stocks, commodities, features, and options is related to a higher likelihood of investing in cryptocurrencies. Moreover, investment experience is also the most pronounced covariate with a connection to cryptocurrencies compared to other variables examined in this paper. Indeed, cryptocurrencies, in general, are viewed as a high-risk investment asset. This aligns with the general view that cryptocurrencies are high-risk investment vehicles, and therefore tend to attract individuals who are already comfortable managing financial risk (Hala et al., 2020; Wärneryd, 1996).

Furthermore, cognitive research indicates that experienced investors are typically more confident (Hoffmann & Post, 2016) and are more willing to explore new investment options such as cryptocurrencies (Nguyen, 2019). These findings contribute to the literature by challenging the common perception that cryptocurrency markets are dominated primarily by speculative or inexperienced traders (Fujiki, 2021; Hadan et al., 2024). Instead, our evidence suggests that the market is also drawing seasoned investors with diverse investment portfolios.

Also, the results from regression analysis show that age negatively moderates the association between investment knowledge and cryptocurrency investment, especially noting that individuals older than 44 are less inclined toward cryptocurrency investments than those younger than 34. Surprisingly, our findings also indicate that while a higher level of investment knowledge relates to a greater chance of investing in cryptocurrency, investment knowledge does not play the same role in becoming a cryptocurrency investor for different age groups, highlighting the possible generational divide in investment preferences, as the younger investors are more likely to embrace new and riskier investment forms. This finding ties well with a mix of literature studying the influence of age on investment behaviors (Beck et al., 2023). For example, Charles and Kasilingam (2013) found that investors gradually reduce their investment portion in the equity market and tend to choose less risky investment options as they grow older. This could be explained by their increasing family responsibility as age increases, which makes them more cautious when making investment decisions (Lee et al., 2010). Likewise, prior literature has found that investment risk tolerance levels often vary based on the age of investors and that younger investors are usually more inclined to choose riskier assets in contrast with older investors when making investment decisions (Dickason & Ferreira, 2018). Since cryptocurrencies are high-risk financial instruments, it is possible that older investors are less likely to invest in them due to their investment risk preferences. Therefore, we call for future research on understanding the broader shifts in attitude toward financial technologies across various age groups.

Different from the existing studies that have documented a significant association between financial literacy and cryptocurrency investment (Fujiki, 2021; Gignac et al., 2023; Zhao & Zhang, 2021), this paper shows that financial literacy is not related to investing in cryptocurrencies. However, this does not mean that the findings from the present study contradict those from prior literature. Instead, the surprising finding of this study is the negative relationship between education level and cryptocurrency investment, indicating that individuals with lower levels of formal education are more likely to invest in cryptocurrencies. These surprising findings challenge the traditional view that higher education correlates with a greater propensity for investment in fintech markets. This suggests that financial literacy, typically acquired through formal education, may no longer be a critical factor in becoming a new fintech investor. In addition to that, it is worth noting that the insignificant linkage between cryptocurrency investment and financial literacy could be attributed to the nature of the sample used for analysis. Our study utilizes data from the National Financial Capability Study 2021, which includes responses from 1810 financial investors nationwide. Therefore, it is possible that investors in this sample share a similar level of financial literacy, and the small variation in financial literacy among investors in the sample is not enough to yield statistically significant results. Therefore, further investigation is needed when more data is available in the future, and we call for future research to conduct a more in-depth investigation into how financial literacy may influence individual financial investment decisions.

From a practical standpoint, our research also makes several contributions. First, our findings can guide financial advisors, service providers, and fintech platforms in identifying which segments of investors are most likely to engage with crypto products. Specifically, investors with a background in managing risky assets may be more receptive to targeted crypto offerings and stand to benefit from tailored education on managing crypto-related volatility and risks. These insights can help bridge the gap between traditional finance and emerging digital assets by equipping institutions to engage the right audience more effectively.

Second, since investment knowledge is more influential for younger investors than for older ones, crypto education programs should be tailored to different age groups. For instance, while there is no doubt that cryptocurrencies have become popular investment assets in recent years, they are significantly more volatile compared to conventional financial instruments. Since the positive association between cryptocurrency investment and investment knowledge diminishes as age increases (Kim & Fan, 2025; Luo et al., 2025), investment education in early life may be more effective in helping develop a healthy and beneficial investment habit regarding cryptocurrency investment. Therefore, educating young investors about the potential risks of cryptocurrency investment is essential. Especially, our study found that financial literacy has no impact on becoming a cryptocurrency investor, and this non-significant association may imply that existing financial literacy initiatives, especially those grounded in formal education, may be insufficient to influence crypto behaviors. This finding indicates the need for universities to consider promoting applied, hands-on financial education that focuses on simulation-based learning rather than purely conceptual financial knowledge. For instance, financial educators and advisors can develop age-specific content, such as using gamified and exploratory formats for younger users, as young adults are more likely to be engaged by gamified learning approaches rather than traditional lecture-based instruction (Buckley & Doyle, 2016; Li et al., 2025).

The present study is not without limitations. First of all, since the data adopted in the present study is cross-sectional, this paper focuses on identifying associations between variables rather than making a causal argument. Future research may examine the causal effect of variables on cryptocurrency investment when panel data is available (Heise, 1970). It is also worth noting that the data collected from the NFCS 2021 are self-reported, which means that responses could be either over- or under-reporting. This limitation may introduce social desirability bias, where respondents tend to provide responses in ways that are more acceptable or favorable to themselves, resulting in understating risky behaviors or overreporting positive financial habits. In addition, non-response bias may arise if individuals who choose not to participate in the survey differ systematically from those who do, which could limit the generalizability of the findings. Future research using a panel dataset, which allows between-time comparisons, could help validate these results and reduce such biases.