languageR (version 1.5.0)

dative: Dative Alternation

Description

Data describing the realization of the dative as NP or PP in the Switchboard corpus and the Treebank Wall Street Journal collection.

Usage

data(dative)

Format

A data frame with 3263 observations on the following 15 variables.

Speaker

a factor coding speaker; available only for the subset of spoken English.

Modality

a factor with levels spoken, written.

Verb

a factor with the verbs as levels.

SemanticClass

a factor with levels a (abstract: 'give it some thought'), c (communication: 'tell, give me your name'), f (future transfer of possession: 'owe, promise'), p (prevention of possession: 'cost, deny'), and t (transfer of possession: 'give an armband, send').

LengthOfRecipient

a numeric vector coding the number of words comprising the recipient.

AnimacyOfRec

a factor with levels animate and inanimate for the animacy of the recipient.

DefinOfRec

a factor with levels definite and indefinite coding the definiteness of the recipient.

PronomOfRec

a factor with levels nonpronominal and pronominal coding the pronominality of the recipient.

LengthOfTheme

a numeric vector coding the number of words comprising the theme.

AnimacyOfTheme

a factor with levels animate and inanimate coding the animacy of the theme.

DefinOfTheme

a factor with levels definite and indefinite coding the definiteness of the theme.

PronomOfTheme

a factor with levels nonpronominal and pronominal coding the pronominality of the theme.

RealizationOfRecipient

a factor with levels NP and PP coding the realization of the dative.

AccessOfRec

a factor with levels accessible, given, and new coding the accessibility of the recipient.

AccessOfTheme

a factor with levels accessible, given, and new coding the accessibility of the theme.

References

Bresnan, J., Cueni, A., Nikitina, T. and Baayen, R. H. (2007) Predicting the dative alternation, in Bouma, G. and Kraemer, I. and Zwarts, J. (eds.), Cognitive Foundations of Interpretation, Royal Netherlands Academy of Sciences, 33 pages, in press.

Examples

# NOT RUN {
data(dative)

# analysis with CART tree

library(rpart)

# ---- initial tree

dative.rp = rpart(RealizationOfRecipient ~ ., 
 data = dative[ ,-c(1, 3)]) # exclude the columns with subjects, verbs
plot(dative.rp, compress = TRUE, branch = 1, margin = 0.1)
text(dative.rp, use.n = TRUE, pretty = 0)

# ---- pruning the initial tree

plotcp(dative.rp)
dative.rp1 = prune(dative.rp, cp = 0.041)
plot(dative.rp1, compress = TRUE, branch = 1, margin = 0.1)
text(dative.rp1, use.n = TRUE, pretty = 0)


# analysis with logistic regression

# ---- logistic regression with the rms package

library(rms)
dative.dd = datadist(dative)
options(datadist = 'dative.dd')
dative.lrm = lrm(RealizationOfRecipient ~ 
  AccessOfTheme + AccessOfRec + LengthOfRecipient + AnimacyOfRec +
  AnimacyOfTheme + PronomOfTheme + DefinOfTheme + LengthOfTheme+
  SemanticClass + Modality, data = dative)
anova(dative.lrm)
plot(Predict(dative.lrm))

# ---- mixed-effects logistic regression with the lme4 package

require(lme4) 
require(lmerTest)
require(optimx)

dative.lmer = glmer(RealizationOfRecipient ~ AccessOfTheme +
  AccessOfRec + LengthOfRecipient + AnimacyOfRec + 
  AnimacyOfTheme + PronomOfTheme + DefinOfTheme + LengthOfTheme + 
  SemanticClass + Modality + (1|Verb), 
  control=glmerControl(optimizer="optimx",optCtrl=list(method="nlminb")),
  data = dative, family = "binomial") 

summary(dative.lmer)

# multiple comparisons for Accessibility of Theme
require(multcomp)
par(mar=c(5,8,3,1))
AcOfTheme.glht <- glht(dative.lmer, linfct = mcp(AccessOfTheme = "Tukey"))
plot(AcOfTheme.glht)
abline(v=0)
summary(AcOfTheme.glht)


# }
# NOT RUN {
 
# }