This chart illustrates the most important results from the study’s principal regression analysis. The analysis took each of the study’s independent variables (listed in the left-hand column of this chart) and tested to determine whether or not they had any significant impact upon our dependent variable—the likelihood that a broad-based pro-democracy civic movement would emerge prior to a successful democratic transition.
The values in the middle column, the P> |t| column, indicate whether or not the relationship between an independent variable and our dependent variable isstatistically significant. In other words, they indicate whether or not any meaningful, sound conclusions can be drawn from the analysis. In order for the relationship to be statistically significant the value in this column must be 0.1 or less. Federalism is therefore the only independent variable that has a statistically significant result.
The right-hand column, the coefficient column, illuminates the nature of the relationship between the independent variable and the dependent variable. If the number in that column is positive, then the independent variable has a positive impact upon the likelihood of a pro-democracy civic movement emerging. If the coefficient is negative, then the nature of the relationship is reversed.
In this case, the middle column shows that federalism is the only independent variable with a statistically significant result, while the right-hand column shows that the existence of a federated form of governance in a country has a negative impact upon the likelihood that a broad-based civic movement will emerge prior to a democratic transition.
|Regime Type 1||0.395||0.8202083|
|Regime Type 2||0.329||1.54553|
|Regime Type 3||0.11||1.516408|
|Regime Type 4||0.358||1.315632|
|Regime Type 5||0.74||-0.5469561|
|4||Single Party Hybrids with either Military or Personalist|
|5||Military/Personalist/Single Party Amalgam|