Results - Charts Tab

Select the Charts tab of the Results dialog box to access the options described here.

Note: X-Bar and R (S) Charts Computed from Aggregated Data

If the current chart was computed from aggregated data (i.e., the input file contained means and ranges or standard deviations, and raw data or measurements are not available), then all options in this dialog box that require access to individual measurements will not be available. See also, the documentation for the Defining Variables for X-bar and Variability Chart dialog - Quick tab.

SixGraph. Click this button to display a compound graph. The graphs displayed are determined by the selections made after clicking the SixGraph options button. This option is not available if the input data file contains aggregated data (sample means and ranges or standard deviations; in that case plots of individual measurements or observations as well as process capability histograms of observations cannot be created).

SixGraph options

Click this button to display the Select the contents and layout for the six graph summary dialog box. The selections you make here determine which graphs will be displayed when you click the SixGraph button.

Standard charts. The next three buttons will have different results depending on the type of analysis chosen. Note that in all of the options described below, if an R -type chart is selected (e.g., X-bar & R) on the Quality Control Charts Startup Panel, the R/S chart produced will be a range chart for the variability of the process. If an S-type chart is selected (e.g., X-bar & S), the R/S chart produced will be a standard deviation chart for the variability of the process. (Note also that histograms accompanying the charts described can be eliminated by clearing the Include histograms checkbox on the Options dialog - Layout tab, accessible by clicking the Options button in the Results dialog box.)

Variable Control Charts.

X(MA..) & R/S. Click this button to create a compound graph that is made up of four individual graphs: two standard quality control (Shewhart control) charts (X-bar and R/S charts) and two histograms (means and ranges or standard deviations).

X

Click this button to produce a compound graph that is made up of two individual graphs: an X-bar chart and a histogram of means.

R/S

Click this button to create a compound graph that is made up of two individual graphs: an R/S chart and a histogram of ranges or standard deviations.

Moving Average (MA) and Exponentially Weighted Moving Average (EWMA) Charts.

X(MA..) & R/S. Click this button to produce a compound graph that is made up of four individual graphs: a moving average (MA) or exponentially weighted moving average (EWMA) chart accompanied by a histogram of those sample values, and an R/S chart accompanied by a histogram of ranges or standard deviations.  

X. Click this button to create a compound graph containing the  (exponentially weighted) moving average chart plus a histogram.

R/S

Click this button to produce a compound graph that is made up of two individual graphs: an R/S chart and a histogram of ranges or standard deviations.

Cusum Charts.

X(MA..) & R/S. Click this button to create a compound graph that is made up of four individual graphs: a cumulative sum (CUSUM) chart for observations (sample size 1) accompanied by a histogram of the individual observations and a moving range chart accompanied by a histogram of the moving ranges. See Computational Details for details on how the CUSUM chart is constructed, and how control limits are established for this chart.

X

Click this button to produce a compound graph containing the cumulative sum (CUSUM) chart and histogram.

R/S

Click this button to create a compound graph containing a moving range chart plus a histogram of moving ranges

X (Individual and Moving Range) Charts.

X(MA..) & R/S. Click this button to produce a compound graph that is made up of four individual graphs: an individual (sample size 1) Shewhart control chart accompanied by a histogram of the individual observations and a moving range chart accompanied by a histogram of the moving ranges.  

X. Click this button to create a compound graph containing an individual (sample size 1) chart plus a histogram of individual observations.

R/S. Click this button to produce a compound graph containing a moving range chart plus a histogram of moving ranges.

Descriptive Stats (Chart Summary). Click this button to create two spreadsheets of descriptive statistics (means, ranges, control limits, etc.). Samples that fall outside the control limits will be highlighted.

Outliers

Click this button to produce two spreadsheets for outliers.

Check Alarms

Click this button to create a spreadsheet with the current summary of possible control chart violations (quality problems) that may trigger alarms. STATISTICA will check and report all (potential) alarms as described on the Options - Alarm tab regardless of whether checking of specific violations (e.g., a specific alarm) is requested. The results spreadsheet will indicate in the column headers which violations are currently being monitored:

The label Not Applicable means that the respective type of violation is not applicable to the current type of chart or that the specifications for the current chart are such that the respective type of violation is not applicable (e.g., no warning limits were specified);

The label Not Checked means that the respective type of violation is not currently being monitored by the program (see also the Options - Alarm tab for details).

This spreadsheet provides a useful quick summary of the types of problems that may be evident in the current control chart.

Histogram X

Click this button to create a histogram of the sample means (or moving averages, for MA charts). The histogram will also show the normal curve superimposed over the observed frequencies (indicated by bars).  Note that for the X-bar chart it is assumed that the sample means follow the normal distribution.

Histogram R/S

Click this button to produce a histogram of the sample ranges or standard deviations (depending on the type of variability chart which was created). Note that for moving average charts, the histogram will be made using the sample moving ranges/moving standard deviations. Note also, that this option is not available if you have specified a MA chart with sigma estimated from the within-sample standard deviations.

OC X

Click this button to create an Operating characteristic curve chart. In the OC curve for an X-bar chart, we can determine the beta (b) probability of not detecting a shift in the process mean from the center line to the respective values on the X axis. Also shown in the plot are the respective curves for larger and smaller sample sizes.

OC R

Click this button to produce an Operating characteristic curve chart for the R chart (based on the distribution of the studentized range statistic), which is a computationally complex procedure.

Runs Tests (Chart Summary)

Click this button to create two spreadsheets that report samples (from sample - to sample) which break one or more of the runs rules.

X non-normal

Click this button to produce a compound graph that is made up of two individual graphs: a non-normal X-bar chart and a histogram of means. See also the Non-Normal tab of the Results dialog. This option is not available if the input data file contains aggregated data (sample means and ranges or standard deviations; in that case skewness and kurtosis statistics are not available, and non-normal charts and related results cannot be computed).

Process capability.

Histogram

Click this button to create a histogram of observations which includes lines for + and - 3 sigma, upper and lower specification limits, and nominal. The graph will also have a normal distribution super-imposed on it. This option is not available if the input data file contains aggregated data (sample means and ranges or standard deviations; in that case plots of individual measurements or observations as well as process capability histograms of observations cannot be created).

Summary

Click this button to produce two spreadsheet summaries for process capability.

Detect shift at

This option is only available for the CuSum chart. The cumulative sum (CuSum) chart is constructed following the recommendations found in, for example, Montgomery (1996, Chapter 7; STATISTICA computes the recommended tabular or algorithmic CuSum chart, and not the "old-style" V-mask control limits). The shift value determines the shift in the mean, in terms of multiples of sigma, that you want to detect. The shift value affects the values of the plot points for the CuSum chart, as described in Computation Details. For further recommendations regarding the CuSum chart and how to establish control limits, see Montgomery (1996; note that STATISTICA uses the recommended default of 5*Sigma or H=5 for establishing control limits).