DataM: Inclass Exercise 0323

In-class exercise 1.

The following student ID file is missing an ID U76067010 at the third to the last position. Find a way to fix it? Dowload and display the data contents in R.

[Solution and Answer]

studentID <- read.table("../data/student_ID.txt", header = T)
studentID$ID <- as.character(studentID$ID)
library(DataCombine)
studentID <- InsertRow(studentID, NewRow = rep('U76067010', 1),
                       RowNum = nrow(studentID)+1-3+1)
studentID

tail(studentID, 3)

In-class exercise 2.

A classmate of yours used data.entry() to change the first woman’s height to 50 in the women{datasets}. She then closed the editor and issued plot(women). To her surprise, she got this message:

[Solution and Answer]

data.entry(women)
class(women)

[1] "list"

Since using data.entry(women) will make the data frame women turn into a list, the command plot(women) can not recognize what variable in women x and y. Hence, we should use edit(women) instead of data.entry(women). We also can spicify x and y when calling plot():

plot(women$height, women$weight)

In-class exercise 3.

Make a Google citations plot (for the last 5 years) of one of NCKU faculty members you know.

[Solution and Answer]

library(scholar)
citation <- get_citation_history('39ghs30AAAAJ')
plot(citation, xlab="Year", ylab="Citations", main='Google citations plot of Prof. Cheng-Te Li', 
     type='h', lwd=2, xaxt="n", xlim = c(2009, 2020))
axis(side=1, at=seq(2007, 2019, by=1), cex.axis=0.7)
abline(h=seq(0, 400, by=200), lty=3, col="gray")

In-class exercise 4.

Data on body temperature, gender, and heart rate. are taken from Mackowiak et al. (1992). “A Critical Appraisal of 98.6 Degrees F …,” in the Journal of the American Medical Association (268), 1578-80. Import the file. Find the correlation between body temperature and heart rate and investigate if there is a gender difference in mean temperature.

[Solution and Answer]

library(readxl)
dta4 <- read_xls('../data/normtemp.xls', sheet='Data')
head(dta4)

str(dta4)

Classes 'tbl_df', 'tbl' and 'data.frame':   130 obs. of  3 variables:
 $ Temp : num  96.3 96.7 96.9 97 97.1 97.1 97.1 97.2 97.3 97.4 ...
 $ Sex  : num  1 1 1 1 1 1 1 1 1 1 ...
 $ Beats: num  70 71 74 80 73 75 82 64 69 70 ...

plot(dta4$Temp, dta4$Beats, pch=19)
grid()

cor(dta4$Temp, dta4$Beats)

[1] 0.2536564

There is a slightly positively correlation between body temperature and heart rate.

For the investigation of gender difference in mean temperature, we conduct an independent two sample t-test.

t.test(Temp ~ Sex, data=dta4)

    Welch Two Sample t-test

data:  Temp by Sex
t = -2.2854, df = 127.51, p-value = 0.02394
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 -0.53964856 -0.03881298
sample estimates:
mean in group 1 mean in group 2 
       98.10462        98.39385 

Since \(p<\alpha=.05\), we reject the null hypothesis. There is a significant gender difference in body temperature.

In-class exercise 5.

The AAUP2 data set is a comma-delimited fixed column format text file with '*' for missing value. Import the file into R and indicate missing values by 'NA'. Hint: ?read.csv

[Solution and Answer]

library(readr)
AAUP2 <- read_fwf('../data/AAUP2.txt', na = '*',
                  col_positions = fwf_widths(c(6, 31, 3, 5, 4, 4, 4, 5, 4, 4, 5, 4, 4, 4, 4, 4, 5)))

Parsed with column specification:
cols(
  X1 = col_double(),
  X2 = col_character(),
  X3 = col_character(),
  X4 = col_character(),
  X5 = col_character(),
  X6 = col_double(),
  X7 = col_double(),
  X8 = col_character(),
  X9 = col_character(),
  X10 = col_double(),
  X11 = col_character(),
  X12 = col_character(),
  X13 = col_double(),
  X14 = col_double(),
  X15 = col_double(),
  X16 = col_double(),
  X17 = col_double()
)

head(AAUP2)

AAUP2 is not comma-delimited actually. It is delimited without a regular form thus we should use read_fwf{readr} to load it correctly. Also, the arguement na = '*' can work to make missing values diplay in NA.