Real Statistics Release 6.4

I am pleased to announce Release 6.4 of the Real Statistics Resource Pack. The new release is now available for free download at Download Resource Pack for Excel 2010, 2013, 2016, 2019, 365 in Windows environments. The Mac version will be available shortly.

The Distribution, Multivariate, Correlation and Reliability and Goodness of Fit examples workbooks are also being revised for compatibility with the new release. These will be available by tomorrow for free download at Download Examples Workbooks.

Over the course of the next several days, the website will be updated for compatibility with the new releases.

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 6.4.

Rasch Item Response Analysis

A Rasch Item Response data analysis tool has been added (accessible from the Corr tab). This approach to testing is based on the interaction between items’ difficulty and  subjects’ abilities.

The following array functions have been added to support this data analysis tool. Here, R1 is an array or cell range whose rows represent the subjects and whose columns represent the items. Values are one (correct response) or zero (incorrect response).

RASCH(R1): returns a table containing the expected scores and expected subject abilities and item difficulty

RASCHFIT(R1): returns a table which determines the fit of the Rasch model

Lin’s Concordance Correlation Coefficient

Lin’s Concordance Correlation Coefficient (CCC) is a method for comparing two measurements of the same variable. This is especially important if you are trying to introduce a new measurement capability which has some advantages (e.g. it is less expensive or safer to use) over an existing measurement technique (the “gold standard”).

The following array function has been added to support this capability:

LINCCC(R1, R2, lab, alpha): returns a column array with Lin’s CCC plus the lower and upper limits of the 1–alpha confidence interval. If lab = TRUE then a column of labels is appended to the output (default FALSE)

Kolmogorov-Smirnov Test for Normality

The following two functions have been added that support the KS test for normality (including the Lilliefors version of the test).

KSSTAT(R1, avg, sd) = the KS statistics for the data in R1

KSTEST(R1, avg, sd) = the p-value of the KS test for normality for the data in R1

If avg and sd are specified then the KS table of critical values is used in KSTEST. If avg and sd are not specified, then the mean (avg) and standard deviation (sd) are determined from the data, in which case the Lilliefors table of critical values is used in KSTEST.

R1 is an array or cell range containing one or two columns. If it contains two columns, then this is interpreted as a frequency table.

New Distribution Functions

The following new functions have been added which support the PERT and Triangular distributions:

PERT_DIST(x, a, b, c, cum) = the pdf/cdf of the PERT distribution at x based on the parameters a (minimum), b (mode) and c (maximum). If cum = TRUE then the cdf is returned, while if cum = FALSE then the pdf is returned.

PERT_INV(p, a, b, c) = the inverse of the PERT distribution at p where a, b and c are as for PERT_DIST.

TRIANG_DIST(x, a, b, c, cum) = the pdf/cdf of the triangular distribution at x based on the parameters a (minimum), b (mode) and c (maximum). If cum = TRUE then the cdf is returned, while if cum = FALSE then the pdf is returned.

TRIANG_INV(p, a, b, c) = the inverse of the triangular distribution at p where a, b and c are as for TRIANG_DIST.

In addition, the following new functions have been added which return the mean and variance for various distributions

MEAN_DIST(dist, param1, param2, param3) = the mean of the distribution defined by the text string dist based on up to three distribution parameters.

VAR_DIST(dist, param1, param2, param3) = the variance of the distribution defined by the text string dist based on up to three distribution parameters.

The distributions that are currently supported are the normal, lognormal, chi-square, t, F, binomial, Poisson, Skellam, beta, gamma, uniform, PERT, triangular, Weibull, exponential, geometric, hypergeometric and negative binomial.

For example, the formula =MEAN_DIST(“beta”,3,4) returns the mean of a beta distribution with alpha parameter 3 and beta parameter 4. The text string is the one used by Excel or Real Statistics to return the cdf/pdf of the distribution (e.g. “chisq” for chi-square, “triang” for triangular, etc.). The order of the parameters is the same as that found in the Excel or Real Statistics function that returns the cdf/pdf.

Bug Fixes

  • Fixed a bug in the CVTEST and CV2TEST functions.
  • Fixed a bug in SSRInt function. This bug also affected the Regression option of Repeated Measures ANOVA with a Between Subjects Factor data analysis tool 

 

2 thoughts on “Real Statistics Release 6.4”

  1. Hi Charles,

    I tried LINCCC, after successfully installing the newest version (ctrl-m works!).
    However, I am given: Can’t find project or library. If I press Ok (which is the only option), I am asked for RealStats Password.

    Thank you for your help!

    Best,
    Yasser

Comments are closed.