This brief report discusses the impact of different estimation methods on the fit indices RMSEA, CFI, and TLI in structural equation modeling (SEM) with ordered categorical data. The study highlights that traditional cutoff values for these indices, developed under normal-theory maximum likelihood (ML) with continuous data, may not be appropriate for ordered categorical data. When using unweighted least squares (ULS) or diagonally weighted least squares (DWLS) with polychoric correlation matrices, the RMSEA values are smaller, and CFI and TLI values are larger compared to ML results. This suggests that applying conventional cutoffs to ULS and DWLS may lead to an underestimation of model-data misfit. The study used both simulated and empirical polychoric correlation matrices to investigate these effects. The results show that ULS and DWLS produce different fit indices compared to ML, and that the conventional cutoffs may not accurately reflect model fit when applied to these methods. The study concludes that the use of RMSEA, CFI, and TLI for ordered categorical data requires careful consideration of the estimation method and the associated cutoff values. The findings suggest that the conventional cutoffs may not be suitable for ULS and DWLS, and that alternative approaches may be needed for accurate model fit evaluation.This brief report discusses the impact of different estimation methods on the fit indices RMSEA, CFI, and TLI in structural equation modeling (SEM) with ordered categorical data. The study highlights that traditional cutoff values for these indices, developed under normal-theory maximum likelihood (ML) with continuous data, may not be appropriate for ordered categorical data. When using unweighted least squares (ULS) or diagonally weighted least squares (DWLS) with polychoric correlation matrices, the RMSEA values are smaller, and CFI and TLI values are larger compared to ML results. This suggests that applying conventional cutoffs to ULS and DWLS may lead to an underestimation of model-data misfit. The study used both simulated and empirical polychoric correlation matrices to investigate these effects. The results show that ULS and DWLS produce different fit indices compared to ML, and that the conventional cutoffs may not accurately reflect model fit when applied to these methods. The study concludes that the use of RMSEA, CFI, and TLI for ordered categorical data requires careful consideration of the estimation method and the associated cutoff values. The findings suggest that the conventional cutoffs may not be suitable for ULS and DWLS, and that alternative approaches may be needed for accurate model fit evaluation.