In recent years, an abnormal combustion phenomenon called low-speed pre-ignition (LSPI) has arisen from the downsizing of gasoline engines in order to improve fuel economy and comply with global CO2 legislation. The type and quality of the fuel and lubricant has been found to influence LSPI occurrence rates. A methodology for studying LSPI has been implemented, and a rigorous statistical approach for studying the data from a stationary engine test can provide consistent results as shown in Part 1 of the series.LSPI events can be determined by an iterative statistical procedure based on calculating the mean and standard deviation of peak pressure (PP) and crank angle location of 2% mass fraction burned (MFB02) data, determining cycles with parameters which exceeded n standard deviations from the mean and identifying outliers. Outliers for the PP and MFB02 metrics are identified as possible LSPI events.Further investigation was conducted to refine the approaches and methods used for analyzing data. This paper expands on some of the methodology described in Part 1 and explores the assumption of normal distribution that is used to determine the number of standard deviations beyond which correspond to outliers. The application of a method for adjusting departure from normality in terms of skewness and kurtosis is explored to reduce the frequency of false positives and negatives.