summary.BEST {BEST} R Documentation

## Extract summary statistics from an object of class BEST.

### Description

Provides summary statistics for each of the parameters (mean and standard deviation) of the group(s) of observations and their differences.

### Usage

```## S3 method for class 'BEST'
summary(object, credMass = 0.95,
ROPEm = NULL, ROPEsd = NULL, ROPEeff = NULL,
compValm = 0, compValsd = NULL, compValeff = 0, ...)

```

### Arguments

 `object` an object of class `BEST`, as produced by the function `BESTmcmc`. `credMass` the probability mass to include in credible intervals. `ROPEm` a two element vector, such as `c(-1, 1)`, specifying the limit of the ROPE on the difference of means (for 2 groups) or the mean (for 1 group). See `plot.BEST` for an explanation of ROPE. `ROPEsd` a two element vector, such as `c(-1, 1)`, specifying the limit of the ROPE on the (difference of) standard deviations. `ROPEeff` a two element vector, such as `c(-1, 1)`, specifying the limit of the ROPE on the effect size. `compValm` a value for comparison with the (difference of) means. `compValsd` a value for comparison with the (difference of) standard deviations. `compValeff` a value for comparison with the effect size. `...` additional arguments for the summary or print function.

### Value

Returns a matrix with the parameters in rows and the following columns:

 `mean, median, mode` the mean, median and mode of the MCMC samples for the corresponding parameter. `hdi%, hdiLow, hdiHigh` the percentage of posterior probability mass included in the highest density interval and the lower and upper limits. `compVal, %>compVal` the value for comparison and the percentage of the posterior probability mass above that value. `ROPElow, ROPEhigh, %InROPE` the lower and upper limits of the Region Of Practical Equivalence (ROPE) and the percentage of the posterior probability mass within the region.

If the analysis concerns a comparison of two groups, the matrix will have rows for:

 `mu1, mu2, muDiff` the means of each group and the difference in means `sigma1, sigma2, sigmaDiff` the standard deviations of each group and the difference in standard deviations `nu, log10nu` the normality parameter and its log `effSz` the effect size; d[a] from Macmillan & Creelman (1991).

For a single group, the rows will be:

 `mu` the mean `sigma` the standard deviation `nu, log10nu` the normality parameter and its log `effSz` the effect size.

Many of the elements of the matrix will be NA. The print method for the summary attempts to print this nicely.

### Author(s)

Mike Meredith, based on code by John K. Kruschke.

### References

Kruschke, J K. 2013. Bayesian estimation supersedes the t test. Journal of Experimental Psychology: General 142(2):573-603. doi: 10.1037/a0029146

Macmillan, N. A., & Creelman, C. D. (1991). Detection Theory: A User's Guide. New York, Cambridge University Press

Use the `plotAll` function for a graphical display of these same values.
```## see "BEST-package"