NCSS 2020 Update

Update Release Notes

This page lists the changes that have been made to NCSS 2020 since it was released.

Version: 20.0.8

Released: January 20, 2023

1. Corrected an issue with plotting the grid lines at stage breaks in Quality Control Charts when running NCSS under a language setting that uses comma as the decimal. The stage break grid lines would not appear.

2. Corrected various input options in multiple procedures when running NCSS under a language setting that uses comma as the decimal. Periods used in text were incorrectly replaced with commas.

3. Fixed an error displaying value labels in the Data Table after inserting or deleting columns. The value labels now display correctly.

4. Removed a bound in the Passing-Bablok Regression for Method Comparison and Robust Linear Regression (Passing- Bablok Median-Slope) procedures that restricted the regression calculations when the number of pairs was greater than just over 3000.

5. Corrected an “Overflow” error that occurred in procedures that calculate Kendall’s Tau correlation with a very large number of observations (> 40,000).

6. Corrected the estimation of Model #5, PolyRatio(2,2) in the Curve Fitting - General procedure. The procedure now estimates all parameters, including E.

7. Corrected an error with the calculation of Weighted COV in the Appraisal Ratio Studies procedure. Weighted COV is now calculated correctly.

8. Corrected an error in the Repeated Measures ANOVA and GLM ANOVA means output when values were equal to zero. In some rare cases, the software would report a very small mean (e.g., 1E-14) in place of zero because of computer rounding error.

9. This is the final update to NCSS 2020.

Version: 20.0.7

Released: August 4, 2022

1. Corrected the formula for the SD of the Uniform Distribution in the documentation for the Data Simulation procedure. The procedure calculations were not affected.

2. Updated four example data files (Fisher, LogLin1, Resale, Tree) so that the Value Labels work correctly when the computer language setting is set to other language settings that use the comma as the decimal separator.

3. Corrected an error in the Correlation Matrix procedure. When “Show the eigenvalue percentage in the eigenvector labels” was checked, the Spearman Eigenvector Plot was displaying the Pearson eigenvalue percentages in the labels. Furthermore, when “Show the eigenvalue percentage in the eigenvector labels” was checked with only Spearman Eigenvector Plot(s) checked, an “Object variable or With block variable not set” error would result.

4. Added support for importing/exporting Statistica version 7+ data files.

5. Corrected an error when exporting delimited text files with a blank delimiter without using double quotes. Double quotes were being added when not requested.

Version: 20.0.6

Released: January 20, 2022

1. Corrected an error that caused the text in some single-value dropdown inputs on the Procedure Window not to be read correctly after selecting an item from the dropdown list.

2. Corrected an error in the Data Unstack Tool. For large datasets, the process would run out of memory and terminate without unstacking the data. This has been fixed so that large datasets will unstack.

3. Corrected an issue with graphics dropdown buttons when using Windows 11. Dropdown buttons on plot format windows would display two images (one slightly offset from the other).

4. With this update, NCSS 2020 is considered to be fully compatible with Windows 11.

Version: 20.0.5

Released: August 30, 2021

1. Corrected an error that occurred when parsing user-entered TO BY lists (e.g., "-2 to 2 by 0.1"). In some cases, rounding errors were causing unexpected results.

2. Corrected a rare error in the Descriptive Statistics – Summary Tables procedure when a group variable with Data Type = Text had been specified. If the data contains leading special characters (e.g. \, ., /, etc.) or a mix of upper- and lower-case characters, the summary tables would not display correctly.

3. Corrected an error when using power transformations with negative data and negative exponents. NCSS now calculates results for more exponents and correct results for some integer exponents that were not being applied correctly (e.g., -2, -3).

4. Corrected an error in the Repeated Measures ANOVA and GLM ANOVA means output when values were equal to zero with Count = 1. The software would report a very small mean (1E-14) in place of zero because of computer rounding error.

5. Corrected an error in the Correlation Matrix procedure when creating a heat map with only 2 variables. The software would return a “Column1 is constant and must be removed” error when creating the heat map.

6. Corrected “Unhandled exception…” errors that occurred when opening PASS procedures in Windows 10 when the Microsoft Region language option Beta: Use Unicode UTF-8 for worldwide language support is active. (To determine if the option is active on your computer, see the steps below.)

Having the Beta: Use Unicode UTF-8 for worldwide language support option active leads to other errors when using NCSS (known to adversely affect data import and plot display in some cases, at the very least). We recommend that the option be disabled when using NCSS.

To disable this option in Windows 10, perform the following steps with administrator rights on your computer:

1. Close all open documents and programs on your computer.
2. Press the Windows key.
3. Type “Settings.”
4. Select Time & Language.
5. Click Language.
6. Under Related settings, click Administrative language settings.
7. On the Region window, click the Administrative tab.
8. Click Change system locale. (Here you will need administrator rights.)
9. Clear the Beta: Use Unicode UTF-8 for worldwide language support checkbox.
10. Click OK.
11. Restart your computer.

Version: 20.0.4

Released: February 17, 2021

1. Corrected a “Subscript out of range” error in the R & R Study procedure. The error would occur if Data Plots was selected with only one Measurement Variable specified.

2. Various minor documentation updates.

Version: 20.0.3

Released: August 28, 2020

1. Corrected the Multiple Comparison’s using the McNemar Test report in the Cochran’s Q procedure. In some cases the binomial exact prob level was lower than it should have been.

2. Corrected an issue with the Y-Axis title Two-Factor Chart procedures when only plotting one factor. An appropriate value is now substituted for {Y} when a two-factor chart is created with only one factor.

3. Corrected a rarely occurring error that affected scatterplots with line segments drawn. In some cases, the plot would generate an “overflow” error. This has been fixed.

4. Corrected the label on the selection popup when inserting columns. The message indicated that columns were to be deleted instead of inserted.

5. Corrected a “subscript out of range” error in the Hierarchical Clustering / Dendrograms and Clustered Heat Maps (Double Dendrograms) procedures that occurred in some cases when applying a filter.

Version: 20.0.2

Released: May 22, 2020

1. Corrected a rare issue that occurred with procedures that contain a checked list box input (e.g. Descriptive Statistics – Summary Tables) after changing the computer’s language setting. Changing the language setting would cause the program to crash after launching the procedure.

2. Corrected an error in the One-Sample, Paired, and Two-Sample T-Tests procedures. The output page title was not being printed when the bootstrap report was at the top of a new page. Calculations were not affected.

3. Corrected an error in the Michaelis-Menten Equation procedure that caused the Individual Function Plots not to appear in the output.

4. Corrected an issue when importing and exporting delimited files (.csv, .txt) with text enclosed in double quotes. The importer and exporter now correctly handles text within quotes.

5. Corrected an error when importing Fixed Width Text Files using the Import Wizard. In rare cases the data would not import correctly if the user clicked “Finish” before proceeding all the way through the wizard.

6. Corrected an issue in the Correlation Matrix procedure that caused the procedure to use Row-Wise missing value deletion when Pair-Wise was selected in some cases. Similar issues in the PCA and Factor Analysis procedures were also corrected.

7. Fixed a rare issue in the Binary Diagnostic Tests – Clustered Samples procedure. For the test of Sensitivity & Specificity, when the variance of the difference is negative or zero, the z-value and prob level were output as zero when they should have been blank.

8. Corrected an error in the Confounding/Alias Section of the Analysis of Two-Level Designs output. The Term No. was not listed correctly.

Version: 20.0.1

Released: February 10, 2020

1. Initial Release of NCSS 2020.