This research attempts to assess the relationship between democracy, social policy and welfare variables. The main question of this research is “does democracy makes difference in the effectiveness of social policy in delivering welfare outcomes?” Thus, this research is assigning social policy as mediating variable between democracy and welfare outcome. One of the methodological challenges of this research is the fact that all of the three variables cannot be measured only by one indicator. Democracy includes wide array of aspects, such as civic organization, freedom of expression, civil liberty, regime type, and the quality of election. Social policy can be expressed by spending on several sectors, such as education, healthcare, social security, or/and general public services. As well as for welfare outcomes; it can be measured by GDP growth, infant mortality, poverty rate, and/or employment rate. Quantifying the relationship of each indicator separately would be complicated and time consuming. Structural equation modelling will be used in this research in order to overcome this kind of problem. The three main variables will be treated as unobserved variables (constructs), while the indicators will be used as observed variables. Once the confirmatory factors (between construct and its indicators) well set up, the causal relationship analysis of the three main variables can be done simultaneously.