A probably plot displays each data point versus the percentage of values in the sample that are less than or equal to that data point.
A probability plot includes the following components:
The expected percentile from the distribution based on maximum likelihood parameter estimates.
Confidence bound lines
The left curved line indicates the lower bounds of the confidence intervals for the percentiles. The right curved line indicates the upper bounds of the confidence intervals for the percentiles.
Anderson-Darling test statistic and p-value
The results of a test to determine whether your data follow the distribution.
Use the probability plots to assess the fit of the nonnormal distribution for each variable.
If the distribution is a good fit for the data, the points should form an approximately straight line. Departures from this straight line indicate that the fit is unacceptable. If the p-value is greater than 0.05, you can assume that the data follow the nonnormal distribution used in the analysis.
If the p-value is less than 0.05, your data do not follow the selected distribution and the capability analysis results may not be accurate. Use Individual Distribution Identification to determine which nonnormal distribution or data transformation is more effective for your data.
If the distributions differ for multiple variables, you should perform a separate capability analysis for each variable.
The capability histogram shows the distribution of your sample data. Each bar on the histogram represents the frequency of data within an interval.
The solid red curve represents the nonnormal distribution model that was selected for the analysis.
Use the capability histogram to view your sample data in relation to the distribution fit and the specification limits.
Look for evidence of lack-of-fit of your selected nonnormal data distribution
For each variable, compare the distribution curve to the bars of the histogram to assess whether your data seem to follow the distribution that you chose for the analysis. If the bars vary greatly from the curve, your data may not follow your chosen distribution and the capability estimates may not be reliable for your process. If you are unsure which distribution best fits your data, use Individual Distribution Identification to identify an appropriate distribution or transformation.
The histograms provide only a rough indication of the distribution fit. To more definitively assess the distribution fit, use the results on the probability plots. If the distributions differ for multiple variables, you should perform a separate capability analysis for each variable.
Examine the sample data in relation to the specification limits
For each variable, visually examine the data in the histogram in relation to the lower and upper specification limits. Ideally, the spread of the data is narrower than the specification spread, and all the data are inside the specification limits. Data that are outside the specification limits represent nonconforming items. Ideally, few or no parts are outside of the specification limits.
To determine the actual number of nonconforming items in your process, use the results for PPM < LSL, PPM > USL, and PPM Total. For more information, go to All statistics and graphs.
Assess the location of the process
For each variable, evaluate whether the process is centered between the specification limits or at the target value, if you have one. The peak of the distribution curve shows where most of the data are located.