To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.
 The data should be continuous

Continuous data are measurements that may potentially take on any numeric value within a range of values along a continuous scale, including fractional or decimal values. Common examples include measurements such as length, weight, and temperature.
Usually the measurements are of the project CTQs (critical to quality) characteristics, which are the product or process measurements that have standards that must be met.
 Data should be collected in subgroups, if possible
 You can collect data as individual observations or in subgroups. Observations for each subgroup should be collected under the same inputs and conditions, such as personnel, environment, or equipment.
 The process must be stable and in control
 If the current process is not stable, then the capability indices cannot be reliably used to assess the future, ongoing capability of the process.
 Collect enough data to characterize the process
 If the process is stable and you have collected enough data, the cumulative standard deviation of the process will stabilize on the plots. If the lines on the plots continue to oscillate, either you have not collected enough data or the process variation is unstable.
 The data should follow a normal distribution
 Estimates of the probability of a defect (such as DPMO) will not be accurate if the data are non normal. In most cases, these estimates tend to be lower than the actual values. So, check the normal plot and the two histograms to see whether the data are at least reasonably normal before using any estimates such as DPMO.