languageR-package {languageR}

Data sets and functions for 'Analyzing Linguistic Data'

Description

Data sets and functions accompanying 'Analyzing Linguistic Data: A practical introduction to statistics', Cambridge University Press, 2007.

Details

Package:	languageR
Type:	Package
Version:	1.0
Date:	2007-01-15
License:	GNU public license

The main function of this package is to make available the data sets discussed and analyzed in 'Analyzing Linguistic Data: A practical introduction to statistics using R', to appear with Cambridge University Press. The following packages should be installed, as ancillary functions in this package depend on them.

zipfR

for word frequency distributions

lme4

for mixed-effects models

coda

for Markov-Chain Monte Carlo estimation

lattice

for trellis graphics

Matrix

for mixed-effects modeling

The following packages need to be installed for working through specific examples.

rms

for regression modeling

rpart

for CART trees

e1071

for support vector machines

MASS

for many useful functions

ape

for phylogenetic clustering

The main convenience functions in this library are, by category:

correspondence analysis

(extending code by Murtagh, 2005)

corres.fnc

correspondence analysis

corsup.fnc

supplementary data

vocabulary richness

(supplementing current zipfR functionality)

compare.richness.fnc

for two texts, compare richness

growth.fnc

empirical vocabulary growth data for text

growth2vgc

conversion to vgc object of zipfR

spectrum.fnc

creates frequency spectrum

text2spc.fnc

conversion to spc object of zipfR

lmer functions

(p-values for mixed-effects models with lme4)

pvals.fnc

p-values for table of coefficients including MCMC

aovlmer.fnc

p-values for anova tables and/or MCMC p-value for specified factor

simulation functions

(for comparing mixed models with traditional techniques including F1, F2, and F1+F2)

simulateRegression.fnc

simulate simple regression design

simulateQuasif.fnc

simulate data for Quasi-F ratios

simulateLatinsquare.fnc

simulating simple Latin-square design

miscellaneous

(convenience functions)

pairscor.fnc

scatterplot matrix with correlation tests

collin.fnc

collinearity diagnostics

pvals.fnc

p-values and MCMC confidence intervals for mixed models

plot.logistic.fit.fnc

diagnostic visualization for logistic models

xylowess.fnc

trellis scatterplots with smoother

mvrnormplot.fnc

scatterplot for bivariate standard normal random numbers with regression line

lmerPlotInt.fnc

offers choice of four ways to visualize an interaction between two numeric predictors in an lmer model

Author(s)

R. H. Baayen

University of Alberta, Edmonton, Canada

[e-mail]

Maintainer: [e-mail]

References

R. H. Baayen (2007) Analyzing Linguistic Data: A practical introduction to statistics using R, Cambridge: Cambridge University Press.

Examples

## Not run: 
  library(languageR)
  data(package="languageR")

## End(Not run)