A statistical test of the indirect effect of monitoring on adherence and hemoglobin A1c was performed using
bootstrapped standard errors as recommended by Shrout and Bolger (2002).
The statistical significance of the path coefficients and indirect effects were determined with
bootstrapped standard errors.
The indirect effects of gender on adherence and HbA1C through externalizing symptoms were significant when assessed by
bootstrapped standard errors, p =.02 and.03, respectively.
When mediation coefficients were tested,
bootstrapped standard error estimates, using 1000 draws, were computed.