Load in the package lattice

library(lattice)

Chunk 1

Display the data set VADeaths.

VADeaths

      Rural Male Rural Female Urban Male Urban Female
50-54       11.7          8.7       15.4          8.4
55-59       18.1         11.7       24.3         13.6
60-64       26.9         20.3       37.0         19.3
65-69       41.0         30.9       54.6         35.1
70-74       66.0         54.3       71.1         50.0

Chunk 2

Find the object type of VADeaths.

class(VADeaths)

[1] "matrix"

Chunk 3

Find the methods in function dotplot{lattice}

methods("dotplot")

[1] dotplot.array*   dotplot.default* dotplot.formula* dotplot.matrix* 
[5] dotplot.numeric* dotplot.table*  
see '?methods' for accessing help and source code

Chunk 4

Use dotplot to draw four dot plots using data of four columns (groups) in VADeaths. Use groups=FALSE to not group data in the plots.

dotplot(VADeaths, groups=FALSE)

Chunk 5

Draw four dot plots using data of four columns (groups) in VADeaths.

1. Do not group data in the plots.
2. Assign the layout of plots to be one column and four rows.
3. Set tht physical aspect ratio of the panels.
4. Assign the original point.
5. Assign the types of plots to be “points” and “horizontal lines”.
6. Name the plot and x-axis.

dotplot(VADeaths, groups=FALSE,     #1
        layout=c(1, 4),             #2
        aspect=0.7,                 #3
        origin=0,                   #4
        type=c("p", "h"),           #5
        main="Death Rates in Virginia - 1940",  #6
        xlab="Rate (per 1000)")                 #6

Chunk 6

Draw a dot plots using data of four columns (groups) in VADeaths. 1. Use default groups=TRUE to ngroup data in a single plot. 2. Assign the type of plot to be “overplotted” point and polyline. 3. Display the legend of group names at the right part of the plot. Show both points and lines on the legend. 4. Name the plot and x-axis.

dotplot(VADeaths, type="o", #2
        auto.key=list(lines=TRUE, space="right"), #3
        main="Death Rates in Virginia - 1940",    #4
        xlab="Rate (per 1000)")                   #4

Chunk 7

Draw four barcharts using data of four columns (groups) in VADeaths.

1. Do not group data in the plots.
2. Assign the layout of plots to be one column and four rows.
3. Set tht physical aspect ratio of the panels.
4. Do not show the reference lines.
5. Name the plot and x-axis.

barchart(VADeaths, groups=FALSE,  #1
         layout=c(1, 4),          #2
         aspect=0.7,              #3
         reference=FALSE,         #4
         main="Death Rates in Virginia - 1940", #5
         xlab="Rate (per 100)")                 #5

Chunk 8

Load in the data set postdoc in package latticeExtra.

data(postdoc, package="latticeExtra")

Chunk 9

Draw a barchart using data of postdoc, a table with dimension of 8x5.

1. Compute the proportion of each group alone the rows (margin=1).
2. Name x-axis.
3. Display the legend of group (column in the table) with layout adjustment (adj=1).

barchart(prop.table(postdoc, margin=1), #1 
         xlab="Proportion",             #2
         auto.key=list(adj=1))          #3

Chunk 10

Draw 5 dot plots using data of postdoc, a table with dimension of 8x5.

1. Compute the proportion of each group alone the rows (margin=1).
2. Do not group data in the plots.
3. Name x-axis.
4. Abbreviate the texts with the maximum length of 10.

dotplot(prop.table(postdoc, margin=1), #1
        groups=FALSE,                  #2
        xlab="Proportion",             #3
        par.strip.text=list(abbreviate=TRUE, minlength=10)) #4

Chunk 11

Draw 5 dot plots using data of postdoc, a table with dimension of 8x5.

1. Compute the proportion of each group alone the rows (margin=1).
2. Do not group data in the plots.
3. List specifying the display order (use the descending order of the median of x, proportion, here) of the panels.
4. Name x-axis.
5. Assign the layout of plots to be one column and five rows.
6. Set tht physical aspect ratio of the panels.
7. Scale y-axis.
8. Order plots alone two axes

dotplot(prop.table(postdoc, margin=1), #1
        groups=FALSE,                  #2
        index.cond=function(x, y) median(x), #3
        xlab="Proportion",             #4
        layout=c(1, 5),                #5
        aspect=0.6,                    #6
        scales=list(y=list(relation="free", rot=0)), #7
        prepanel=function(x, y) {                    #8
            list(ylim=levels(reorder(y, x)))
        },
        panel=function(x, y, ...) {                  #8
            panel.dotplot(x, reorder(y, x), ...)
        })

Chunk 12

Load in the data set Chem97 in package mlmRev.

data(Chem97, package="mlmRev")

Chunk 13

Create a contingency table of gcsescore and gender with data of Chem97, and name it gcsescore.tab.

gcsescore.tab <- xtabs(~ gcsescore + gender, Chem97)

Chunk 14

Turn gcsescore.tab into a data frame and name it gcsescore.df.

gcsescore.df <- as.data.frame(gcsescore.tab)

Chunk 15

Using class(gcsescore.df$gcsescore) make us know that gcsescore, a variable in gcsescore.df is a factorial vector. Because a factorial vector uses serial integers (1, 2, …) to represent its different levels (factors), turning a factorial vector into a numeric vector directly would make a factorial vector become a meaningless vector with serial integers. Therefore, we turn a gcsescore.df$gcsescore into a character form and then turn it into a numeric form.

gcsescore.df$gcsescore <- as.numeric(as.character(gcsescore.df$gcsescore))

Chunk 16

Draw two plots to present the distribution of gcsescore with type of line and grouping of gender. Use layout=c(1, 2) to make the layout of plots to be one column and two rows.

xyplot(Freq ~ gcsescore | gender, 
       data = gcsescore.df, 
       type="h", 
       layout=c(1, 2), 
       xlab="Average GCSE Score")

Chunk 17

Create a contingency table of score and gender with data of Chem97, and name it score.tab.

score.tab <- xtabs(~score + gender, Chem97)

Chunk 18

Turn score.tab into a data frame and name it score.df.

score.df <- as.data.frame(score.tab)

Chunk 19

Draw a barchart the present the frequency of each score with grouping of gender. Set the original point as zero.

barchart(Freq ~ score | gender, score.df, origin=0)

## The End