The Basic, Correlation/Reliability, Regression 1, Regression 2, Time Series and Multivariate examples workbooks have also been revised for compatibility with the new release. These are available for free download at Download Examples Workbooks. The Non-parametric examples workbook has been divided into two workbooks: Non-parametric 1 and Non-parametric 2.
Over the course of the next several days, the website will be updated for compatibility with the new release.
If you are getting value from the Real Statistics website or software, I would appreciate your donations to help offset the costs of the website by going to Please Donate.
The following is an overview of the new features in Release 6.7.
Permutational MANOVA
A new Permutational MANOVA data analysis tool has been added that carries out a non-parametric version of MANOVA.
In addition, the following new functions have been added: PERMANOVA and PseudoF. The first of these functions executes the Permutational MANOVA procedure using the PseudoF function multiple times.
Logistic Regression statistical power and sample size capabilities
Support for calculating the statistical power and minimum sample size requirements for binary logistic regression have been added. The new capabilities cover logistic regression models with continuous, normally distributed independent variables and binary, binomially distributed independent variables.
In addition to the new Logistic Regression (Normal) Power, Logistic Regression (Normal) Size, Logistic Regression (Binomial) Power and Logistic Regression (Binomial) Sample Size data analysis tools, the following new functions are provided: LOGIT_POWER and LOGIT_SIZE (for normal variables) and LOGIT_POWER0 and LOGIT_SIZE0 (for binomial variables).
Cohen’s Kappa statistical power and sample size capabilities
Support for calculating the statistical power and minimum sample size requirements for Cohen’s kappa in the case where there are two categories has been added.
In addition to the new Cohen’s Kappa Power and Cohen’s Kappa Sample Size data analysis tools, the following new functions are provided: BKAPPA_POWER and BKAPPA_SIZE. These functions, in turn, call BKAPPA_SD(κ, p1, q1) which returns the standard deviation when κ = Cohen’s kappa, p1 = the marginal probability that rater 1 chooses category 1 and q1 = the marginal probability that rater 2 chooses category 1.
Trinomial Test
The trinomial test provides similar functionality to the sign test, but has the advantage of better utilizing data pairs of equal value, thereby achieving better statistical power than the sign test.
This test is carried out via the new TRINOM_TEST and TrinomTest functions.
Spline Fitting
A new Spline Fit data analysis tool has been added that fits a spline curve to a set of data points. This data analysis tool uses the existing SPLINE function.
Minor Enhancements
A new RANKCOL function has been added that calculates the ranks of the elements in each column of an array or cell range (the data in each column is ranked separately). Using this function you can create a Spearman’s correlation matrix for the data in range R1 via the array formula =CORR(RANKCOL(R1)).
The QCRIT and QPROB functions have been enhanced to support more than the k = 20 variables shown in the Studentized Range q table. For values of k > 20 these functions call the existing QINV and QDIST functions.
The MCONST array function has been added. MCONST(c, m, n) returns an m × n array filled with the value c. If the second parameter is omitted then the currently highlighted range is filled with the value c.
Bug Fix
A bug in the validation procedure for the Poisson Regression data analysis tool has been corrected.
A bug has been fixed in the eVECT function (for finding eigenvalues and functions) when the order argument is set to FALSE