This study analyses the relationship between subjective and objective measures of well-being in selected European countries using the data of the European Community Household Panel (ECHP). In the first part of the paper, we develop a random-effect ordered probit model, separately for each country, relating the subjective measure of income satisfaction to actual income, and controlling for some individual and household socio-demographic fixed effects. In the second part of the paper, we fit a Bayesian cross-classified multilevel model, in order to control for intra-family correlation in subjective well-being, which actually appears to be present.