This Monte Carlo study investigated the effect of including auxiliary variables on model fit during estimation of structural equation modeling with multiple imputation. Specifically, the study examined the influence of sample size(200 or 500), missing rates (10% or 20%), missingness mechanism combinations(MCAR-MCAR, MCAR-MAR, MAR-MAR, MCAR-MNAR, MAR-MNAR, MNAR-MNAR), missingness types(linear or convex), and the absence/presence of the auxiliary variables on F-statistic, NFl, TLI, CFI, McDonald's centrality index, and RMSEA. Including auxiliary variables in the imputation model was found to improve the fit on F-statistic, particularly when the missingness type was convex. The ad hoc fit indices examined in this study showed quite good fit, and MI with auxiliary variables tended to improve the fit in the more severe conditions of sample size of 200, missing rate of 20%, and MNAR-included cases. Implications and directions to future research are discussed.