ndl (version 0.2.18)

anova.ndlClassify: Analysis of Model Fit for Naive Discriminatory Reader Models

Description

Compute an analysis of individual variable contributions or model comparisons for one or more Naive Discriminatory Reader model fits.

Usage

# S3 method for ndlClassify
anova(object, …, statistic = "deviance", test = "Chisq")

Arguments

object, …

Object(s) of class "ndlClassify", typically the result of a call to ndlClassify, or a list of objects for the ndlClassifylist method.

statistic

A character string specifying the statistic describing the fit that is to be compared, by default deviance, which is obtained from the object(s).

test

A character string, determining the statistical method by which the significance of the comparison are done, by default the Chi-squared test (Chisq).

Value

An object of class "anova" inheriting from class "data.frame".

Details

Currently, comparison of the terms of a single model or multiple models is only implemented based on the deviance statistic.

Specifying a single object gives a sequential analysis of deviance table for that fit. That is, the reductions in the residual deviance as each term of the formula is added in turn are given in as the rows of a table, plus the residual deviances themselves.

If more than one object is specified, the table has a row for the residual degrees of freedom and deviance for each model. For all but the first model, the change in degrees of freedom and deviance is also given. (This only makes statistical sense if the models are nested.) It is conventional to list the models from smallest to largest, but this is up to the user.

The table will contain test statistics (and P values) comparing the reduction in deviance for the row to the residuals. Only a comparison of models or contributions of their components by the chi-squared test has been implemented.

The comparison between two or more models by anova or anova.ndlClassifylist will only be valid if they are fitted to the same dataset. If anova.ndlClassifylist detects this, it will stop and report an error.

References

Arppe, A. and Baayen, R. H. (in prep.) Statistical classification and principles of human learning.

See Also

ndlClassify

Examples

# NOT RUN {
data(think)
set.seed(314)
think <- think[sample(1:nrow(think),500),]

think.ndl1 <- ndlClassify(Lexeme ~ Agent * Person, data=think)
anova(think.ndl1)

think.ndl2 <- ndlClassify(Lexeme ~ Agent * Person + Patient, data=think)
anova(think.ndl1, think.ndl2)

# }