I am pleased to announce **Release 3.0** of the **Real Statistics Resource Pack**. The new release is available for free download (Download Resource Pack) and is compatible with Excel 2002, 2003, 2007, 2010 and 2013. The versions for Excel 2007, 2010 and 2013 are available now; those for Excel 2002 and 2003 will be available tomorrow. I hope to make a Macintosh version available shortly.

The **Real Statistics Examples Workbook** has been updated to reflect the new release. You can also download this file for free (Download Examples). The website is being updated to reflect the new features. These changes will be available over the course of the next couple of days.

The release provides the following new capabilities:

**New Reliability data analysis tool**

This tool supports the following capabilities:

- Cronbach’s alpha (includes Kuder-Richardson and Split-half capabilities)
- Cohen’s kappa
- Cohen’s weighted kappa
- Fleiss’s kappa
- Intraclass correlation
- Kendall’s W
- Item analysis (item difficulty, item discrimination and point-serial correlation)

**New Reliability functions **

**SPLITHALF**(R1, *type*): calculates the split-half measure for or the scores in the first half of the items in R1 vs. the second half of the items if *type* = 0 and the odd items in R1 vs. the even items if *type* = 1.

**ITEMDIFF**(R1, *mx*): calculates the **item difficulty** for the scores in R1 where *mx* is the maximum score for the item.

**ITEMDISC**(R1, R2, *p, mx*) : calculates **item discrimination** where R1 contains the scores for each subject for a single item and R2 contains the corresponding scores for all items based on the top/bottom *p*% of total scores and *mx* is the maximum score for the item whose scores are contained in R1.

**Enhanced Reliability functions**

**KAPPA**(R1, *k, lab, alpha*): in addition to calculating **Fleiss’s kappa** (if *k* = 0) or kappa for item *k (if k ≠ *0

*)*the function now returns the standard error, z-stat, p-value and confidence interval.

**New one-sample correlation data analysis tool**

This tool calculates the **Pearson’s**, **Spearman’s** (rho) and **Kendall’s** (tau) correlation coefficients.

For the Pearson’s and Kendall’s coefficients, the standard error, p-value and confidence intervals are also calculated.

**New correlation functions**

**KCORREL**(R1, R2, *lab, tails, alpha*): calculates **Kendall’s correlation coefficient** (**tau**) plus the standard error, z-stat, z-crit, p-value and confidence interval.

**SCORREL**(R1, R2): calculates **Spearman’s correlation coefficient** (**rho**)

**Enhanced correlation functions**

A final argument has been added to the **CorrelTest** and **CorrelTTest** functions allowing the user to specify whether a one or two tail test is to be used.

**New Power and Sample Size data analysis tool**

This tool calculates the power of the following tests (post-hoc) as well as the sample size required for these tests (a priori):

- one-sample normal test
- two-sample normal test
- one-sample and paired-sample t test
- two-sample t test
- one-sample binomial test
- one-sample correlation tesrìt
- one-sample variance test
- two-sample variance test
- Chi-square test (goodness of fit and independence tests)
- one-way ANOVA

**New power and sample size functions**

**NORM1_POWER**(*d, n, tails, alpha*): calculates the power of a one-sample normal test where *d* = Cohen’s effect size and *n* = sample size.

**NORM1_SIZE**(*d, pow, tails, alpha*): calculates the minimum sample required to achieve power of *pow* for a one-sample normal test where *d* = Cohen’s effect size.

**NORM2_POWER**(*m, s 1, s2, n1, n2, tails, alpha*): calculates the power of a two-sample normal test where

*m*= difference between the sample means,

*n1*and

*n*2 are the standard deviations, and

*n1*and

*n*2 are the sample sizes.

**NORM2_SIZE**(*d, pow, tails, alpha, nratio*): calculates the minimum sample required to achieve power of *pow* for a two-sample normal test where *m, s 1, s2, tails, alpha *are as for NORM2_POWER and

*nratio*is as for VAR2_SIZE.

**Mood’s Median Test functions**

**MOODS_STAT**(R1): calculates the chi-square statistic for Mood’s Median Test for the data in range R1.

**MOODS_TEST**(R1): calculates the p-value for Mood’s Median Test for the data in range R1.

**Function which calculates a column label**

**ColLabel**(*n*): calculates the label of the *n*th column in a worksheet

E.g. the 5th column is E and so ColLabel(5) = “E”. The 200th column is GR and so ColLabel(200) = “GR”. ColLabel(20000) = “ACOF”

**Minor improvements**

**Two Factor ANOVA** data analysis tool: reports whether the model is **balanced** or not.

**Data Description and Normality** data analysis tool: the **outlier limit** can now be set by the user (previously it was set automatically to 2.5 or 3.0 standard deviations).

Improved error checking in data analysis tools when input data are in **standard format**.

**Bug fixes**

Bugs have been fixed in the following functions: TRIMDATA, COUNTAU (when the input range is a single cell) and some of the power and sample size functions.

Charles