Conceptualizing Financial Literacy: The Connection Among Knowledge, Attitude, and Behaviors

Numerous new financial products and services have emerged in recent decades, requiring individuals to equip with financial literacy in order to make sound financial decisions. There is an imperative call for examining and improving the financial literacy for all populations and the poor in particular. Conceptually, however, financial literacy lacks a solid theoretical framework, and there is a great variety in its definitions and operationalization. Often, financial literacy is conceptualized as ability to understand basic financial concepts (financial knowledge), to make informed financial decisions (financial behavior), to hold positive views of money management (financial attitude), or a combination of all three constructs. What remains unclear is whether and to what extent financial knowledge, behavior, and attitude measure the concept financial literacy. The current study attempts to explore the relationships between three constructs of financial literacy using a sample of 329 Chinese rural migrant workers. 

Financial knowledge was measured by 23 objective knowledge questions on saving, borrowing, and investment. Financial attitudes was measured by one question that asked respondents their attitude towards money management. Financial behavior was measured by five questions that asked respondents whether had a specific behavior.  Each response was dichotomously coded and then summed to indicate corresponding construct. Covariates included age (coded continuously), education (1=high school or more, 0=less than high school), sex (1=male, 0=female), marital status (1=married, 0=otherwise), and whether having child (1=yes, 0=no). Structural equation modeling was employed to explore the pathways between three constructs. Chi-square value, Comparative Fit Index (larger than .95 indicating a good fit), and root mean square error of approximation (RMSEA) (less than .06 with upper bound of 90% confidence interval (CI) smaller than .08 indicating a good fit) were used as the criteria for evaluating models.

Analysis showed the model with financial attitude partially mediated the path from financial knowledge to behaviors best fit the data (χ²(23) = 27.65, p = 0.23; CFI = 0.980; RMSEA = 0.026 (90% CI: 0.000, 0.058). Saving, borrowing, and investment were all significantly loaded on financial knowledge. Paths between three constructs were significantly positive but weak. By controlling the covariates, financial knowledge explained totally 6.36% of the variances of financial behavior (z =3.52, p < .001), and 0.38% was explained via financial attitude (z = 3.38, p<.01). Among the covariates, having child and marriage was negatively (z = -2.36, p <. 05) and positively (z = 1.92, p <. 1) associated with financial knowledge. Education was positively associated with financial attitude (z = 1.72, p < .1). Age is positively associated with financial behavior (z = 5.47, p <. 001). 

This study examined the association between financial knowledge, behavior, and attitude.  The significant, but weak associations among financial knowledge, attitude, and behavior indicate that they are three distinct concepts that should be measured separately. Our findings also suggest that financial literacy is a multidimensional concept that requires different measures of knowledge, behavior, and attitude. To ensure the criteria validity, financial literacy measurement should incorporate behavior measures.