The data set Abortion in vcdExtra gives a 2 × 2 × 2 table of opinions regarding abortion in relation to sex and status of the respondent. This table has the following structure:
library(vcdExtra)
#fourfold(Abortion,c(3,1,2))
str(Abortion)
## table [1:2, 1:2, 1:2] 171 152 138 167 79 148 112 133
## - attr(*, "dimnames")=List of 3
## ..$ Sex : chr [1:2] "Female" "Male"
## ..$ Status : chr [1:2] "Lo" "Hi"
## ..$ Support_Abortion: chr [1:2] "Yes" "No"
fourfold(Abortion)
fourfold(aperm(Abortion,c(3,1,2)))
oddsratio(Abortion,log="FALSE")
## odds ratios for Sex and Status by Support_Abortion
##
## Yes No
## 1.3614130 0.6338682
oddsratio(aperm(Abortion,log = F))
## log odds ratios for Support_Abortion and Status by Sex
##
## Female Male
## 0.5634609 -0.2009764
The graphs above tell us that more female will be in favor of abortion when the status is low, whereas more men will be in favor of abortion when status is high
Exercise 4.4 The Hospital data in vcd gives a 3 × 3 table relating the length of stay (in years) of 132 long-term schizophrenic patients in two London mental hospitals with the frequency of visits by family and friends.
(a)Carry out a χ2 test for association between the two variables.
chisq.test(Hospital)
##
## Pearson's Chi-squared test
##
## data: Hospital
## X-squared = 35.171, df = 4, p-value = 4.284e-07
(b)Use assocstats () to compute association statistics. How would you describe the strength of association here?
assocstats(Hospital)
## X^2 df P(> X^2)
## Likelihood Ratio 38.353 4 9.4755e-08
## Pearson 35.171 4 4.2842e-07
##
## Phi-Coefficient : NA
## Contingency Coeff.: 0.459
## Cramer's V : 0.365
(c)Produce an association plot for these data, with visit frequency as the vertical variable. Describe the pattern of the relation you see here.
x <- t(Hospital)
assocplot(x)
(d)Both variables can be considered ordinal, so CMHtest () may be useful here. Carry out that analysis. Do any of the tests lead to different conclusions?
CMHtest(Hospital)
## Cochran-Mantel-Haenszel Statistics for Visit frequency by Length of stay
##
## AltHypothesis Chisq Df Prob
## cor Nonzero correlation 29.138 1 6.7393e-08
## rmeans Row mean scores differ 34.391 2 3.4044e-08
## cmeans Col mean scores differ 29.607 2 3.7233e-07
## general General association 34.905 4 4.8596e-07
Exercise 4.6 The two-way table Mammograms in vcdExtra gives ratings on the severity of diagnosis of 110 mammograms by two raters.
(a)Assess the strength of agreement between the raters using Cohen’s κ, both unweighted and weighted.
data(Mammograms)
Kappa(Mammograms)
## value ASE z Pr(>|z|)
## Unweighted 0.3713 0.06033 6.154 7.560e-10
## Weighted 0.5964 0.04923 12.114 8.901e-34
(b)Use agreementplot () for a graphical display of agreement here.
agreementplot(Mammograms)
(c)Compare the Kappa measures with the results from assocstats (). What is a reasonable interpretation of each of these measures
assocstats(Mammograms)
## X^2 df P(> X^2)
## Likelihood Ratio 92.619 9 4.4409e-16
## Pearson 83.516 9 3.2307e-14
##
## Phi-Coefficient : NA
## Contingency Coeff.: 0.657
## Cramer's V : 0.503
Exercise 4.7 Agresti and Winner (1997) gave the data in below on the ratings of 160 movies by the reviewers Gene Siskel and Roger Ebert for the period from April 1995 through September 1996. The rating categories were Con (“thumbs down”), Mixed, and Pro (“thumbs up”).
dt <- matrix(c(24, 8, 13, 8, 13, 11, 10, 9, 64),
nrow = 3, ncol = 3, byrow = TRUE)
dt <- as.table(dt)
rownames(dt) <- c('Con', 'Mixed', 'Pro')
colnames(dt) <- c('Con', 'Mixed', 'Pro')
addmargins(dt)
## Con Mixed Pro Sum
## Con 24 8 13 45
## Mixed 8 13 11 32
## Pro 10 9 64 83
## Sum 42 30 88 160
Kappa(dt)
## value ASE z Pr(>|z|)
## Unweighted 0.3888 0.05979 6.503 7.870e-11
## Weighted 0.4269 0.06350 6.723 1.781e-11
agreementplot(dt, main="Weighted")
agreementplot(dt, main="Unweighted")