The existing methodology for treating certain kinds of quality control data assumes the existence of normality and independence in the data. Under these conditions the data can be treated simply through the use of available tables and simple calculations. When either independence and/or normality are not present, as is usually the case, application of the existing methodology introduces large errors in the analysis of the data and renders conclusions based on them dubious. This paper modifies and extends the existing standard methodology by utilizing the time series analysis approach and by introducing dependence via a second order autoregressive process (AR(2) Model). Curves of the modified auxiliary quality control factors are presented, showing the substantial effect of dependence on the classical quality control factors.