We created a dummy variable, MainComp, that takes the value of 1 if a firm is in the main connected component and 0 if not.
Third, there are two dependent variables that are constrained to the values of 0 and 1: MainComp and public.
The third group arises from the various measures we used to locate a firm within industry networks: Closeness centrality and MainComp are highly correlated.
Membership in the main component (MainComp) is also steadily increasing (p = .065).
Predictor variables Degree Closeness (at time t) centrality centrality MainComp
Central connectedness was measured variously by degree and closeness centrality and membership in the main component (MainComp).