The Display Options sub-menu contains the following options:
Moves the Count, Prob, Density, and Normal Quantile Plot axes to the left instead of the right.
This option is applicable only if Horizontal Layout is selected.
The Histogram Options sub-menu contains the following options:
 • Std Error Bars
 • Show Counts
 • Show Percents
The density is the length of the bars in the histogram. Both the count and probability are based on the following calculations:
Use the Normal Quantile Plot option to visualize the extent to which the variable is normally distributed. If a variable is normally distributed, the normal quantile plot approximates a diagonal straight line. This type of plot is also called a quantile-quantile plot, or Q-Q plot.
Normal Quantile Plot
 • The y-axis shows the column values.
 • The x-axis shows the empirical cumulative probability for each value.
Related Information
 • Statistical Details for the Normal Quantile Plot
Outlier Box Plot
 • The horizontal line within the box represents the median sample value.
 • The ends of the box represent the 25th and 75th quantiles, also expressed as the 1st and 3rd quartile, respectively.
 • The difference between the 1st and 3rd quartiles is called the interquartile range.
 • The box has lines that extend from each end, sometimes called whiskers. The whiskers extend from the ends of the box to the outermost data point that falls within the distances computed as follows:
 • The bracket outside of the box identifies the shortest half, which is the most dense 50% of the observations (Rousseuw and Leroy 1987).
 1
 2 Click Box Plot.
 3 Deselect the check box next to Confidence Diamond or Shortest Half.
Quantile Box Plot
Quantiles are values.where the pth quantile is larger than p% of the values. For example, 10% of the data lies below the 10th quantile, and 90% of the data lies below the 90th quantile.
Each line of the plot has a Stem value that is the leading digit of a range of column values. The Leaf values are made from the next-in-line digits of the values. You can see the data point by joining the stem and leaf. In some cases, the numbers on the stem and leaf plot are rounded versions of the actual data in the table. The stem-and-leaf plot actively responds to clicking and the brush tool.
CDF Plot
Use the Test Mean window to specify options for and perform a one-sample test for the mean. If you specify a value for the standard deviation, a z-test is performed. Otherwise, the sample standard deviation is used to perform a t-test. You can also request the nonparametric Wilcoxon Signed-Rank test.
Use the Test Mean option repeatedly to test different values. Each time you test the mean, a new Test Mean report appears.
 Statistics that are calculated for Test Mean: t Test (or z Test) Lists the value of the test statistic and the p-values for the two-sided and one-sided alternatives. Signed-Rank (Only appears for the Wilcoxon Signed-Rank test) Lists the value of the Wilcoxon signed-rank statistic followed by the p-values for the two-sided and one-sided alternatives. The test assumes only that the distribution is symmetric. See Statistical Details for the Wilcoxon Signed Rank Test. Probability values: Prob>|t| The probability of obtaining an absolute t-value by chance alone that is greater than the observed t-value when the population mean is equal to the hypothesized value. This is the p-value for observed significance of the two-tailed t-test. Prob>t The probability of obtaining a t-value greater than the computed sample t ratio by chance alone when the population mean is not different from the hypothesized value. This is the p-value for an upper-tailed test. Prob
 PValue animation Starts an interactive visual representation of the p-value. Enables you to change the hypothesized mean value while watching how the change affects the p-value. Power animation Starts an interactive visual representation of power and beta. You can change the hypothesized mean and sample mean while watching how the changes affect power and beta. Remove Test Removes the mean test.
Use the Test Std Dev option to perform a one-sample test for the standard deviation (details in Neter, Wasserman, and Kutner 1990). Use the Test Std Dev option repeatedly to test different values. Each time you test the standard deviation, a new Test Standard Deviation report appears.
 Test Statistic Min PValue Prob>ChiSq The probability of obtaining a Chi-square value greater than the computed sample Chi-square by chance alone when the population standard deviation is not different from the hypothesized value. This is the p-value for observed significance of a one-tailed t-test. Prob
The Confidence Interval options display confidence intervals for the mean and standard deviation. The 0.90, 0.95, and 0.99 options compute two-sided confidence intervals for the mean and standard deviation. Use the Confidence Interval > Other option to select a confidence level, and select one-sided or two-sided confidence intervals. You can also type a known sigma. If you use a known sigma, the confidence interval for the mean is based on z-values rather than t-values.
 Command Column Added to Data Table Description Level Numbers Level The level number of each observation corresponds to the histogram bar that contains the observation. The histogram bars are numbered from low to high, beginning with 1. Level Midpoints Midpoint The midpoint value for each observation is computed by adding half the level width to the lower level bound. Ranks Ranked Provides a ranking for each of the corresponding column’s values starting at 1. Duplicate response values are assigned consecutive ranks in order of their occurrence in the data table. Ranks Averaged RankAvgd If a value is unique, then the averaged rank is the same as the rank. If a value occurs k times, the average rank is computed as the sum of the value’s ranks divided by k. Prob Scores Prob For N nonmissing scores, the probability score of a value is computed as the averaged rank of that value divided by N + 1. This column is similar to the empirical cumulative distribution function. Normal Quantiles N-Quantile Standardized Std Centered Centered Saves values for centering on zero. Spec Limits (none) Stores the specification limits applied in a capability analysis as a column property of the corresponding column in the current data table. Automatically retrieves and displays the specification limits when you repeat the capability analysis. Script to Log (none)
When you select the Prediction Interval option for a variable, the Prediction Intervals window appears. Use the window to specify the confidence level, the number of future samples, and either a one-sided or two-sided limit.
Related Information
 • Statistical Details for Prediction Intervals
 • Example of Prediction Intervals
When you select the Tolerance Interval option for a variable, the Tolerance Intervals window appears. Use the window to specify the confidence level, the proportion to cover, and either a one-sided or two-sided limit. The calculations are based on the assumption that the given sample is selected randomly from a normal distribution.
Related Information
 • Statistical Details for Tolerance Intervals
 • Example of Tolerance Intervals
The Capability Analysis option measures the conformance of a process to given specification limits. When you select the Capability Analysis option for a variable, the Capability Analysis window appears. Use the window to enter specification limits, distribution type, and information about sigma.
Note: To save the specification limits to the data table as a column property, select Save > Spec Limits. When you repeat the capability analysis, the saved specification limits are automatically retrieved.
The Capability Analysis window, report, and options are described in the following tables.
 Estimates sigma (σ) using the selected methods. See Statistical Details for Capability Analysis.
 • Below LSL gives the percentage of the data that is below the lower specification limit.
 • Above USL gives the percentage of the data that is above the upper specification limit.
 • Total Outside gives the total percentage of the data that is either below LSL or above USL.
Note: There is a preference for Capability called Ppk Capability Labeling that labels the long-term capability output with Ppk labels. Open the Preference window (File > Preferences), then select Platforms > Distribution to see this preference.
 • Below LSL gives the percentage of the fitted distribution that is below the lower specification limit.
 • Above USL gives the percentage of the fitted distribution that is above the upper specification limit.
 • Total Outside gives the total percentage of the fitted distribution that is either below LSL or above USL.
The PPM value is the Percent column multiplied by 10,000.
 Z Bench Shows the values (represented by Index) of the Benchmark Z statistics. According to the AIAG Statistical Process Control manual, Z represents the number of standard deviation units from the process average to a value of interest such as an engineering specification. When used in capability assessment, Z USL is the distance to the upper specification limit and Z LSL is the distance to the lower specification limit. See Statistical Details for Capability Analysis. Capability Animation Interactively change the specification limits and the process mean to see the effects on the capability statistics. This option is available only for capability analyses based on the Normal distribution.
Related Information
 • Statistical Details for Capability Analysis
 • Example of Capability Analysis