steven — Sep 1, 2013, 10:35 PM
# HW 1 Chapter 3, this script assumes that data files are in your working directory
# Import file as serum_zinc in order to work with file and not change anything with
# chapter 2 homework serzinc files
serum_zinc <- read.csv("~/bio-informatics/Bio-stats HW 1/serzinc.csv")
# Summary of the serum_zinc file
summary(serum_zinc)
zinc
Min. : 50.0
1st Qu.: 76.0
Median : 86.0
Mean : 87.9
3rd Qu.: 98.0
Max. :153.0
# (a) find mean, median, mode, sd, range and IRQ
mean(serum_zinc$zinc)
[1] 87.94
median(serum_zinc$zinc)
[1] 86
# There is not a simple way to get the mode of a data set in R so we have to
# use a little work around in order to get it
mode <- table(as.vector(serum_zinc))
names(mode)[mode == max(mode)]
[1] "75"
sd(serum_zinc$zinc)
[1] 16
rng = range(serum_zinc$zinc)
IQR = IQR(serum_zinc$zinc)
IQR
[1] 22
rng
[1] 50 153
# Chebychev description of data
# 2 sd's
ci2 <- range((mean(serum_zinc$zinc)+2*sd(serum_zinc$zinc)),
((mean(serum_zinc$zinc)-2*sd(serum_zinc$zinc))))
ci2
[1] 55.93 119.95
# 3 sd's
ci3 <- range((mean(serum_zinc$zinc)+3*sd(serum_zinc$zinc)),
((mean(serum_zinc$zinc)-3*sd(serum_zinc$zinc))))
ci3
[1] 39.92 135.95
# IQRm ci2 and ci3 together
IQR
[1] 22
rng
[1] 50 153
ci2
[1] 55.93 119.95
ci3
[1] 39.92 135.95
# 12
birth_rate <- read.csv("~/bio-informatics/Bio-stats HW 1/brate.csv")
meanbrt <- mean(birth_rate$birthrt)
meanbrt
[1] 23.27
medianbrt <- median(birth_rate$birthrt)
medianbrt
[1] 23.4
# 5% trimmed mean
mean(birth_rate$birthrt, trim=0.05)
[1] 23.02
# 13
cigarette_data <- read.csv("~/bio-informatics/Bio-stats HW 1/cigarett.csv")
mean_nicotine <- mean(cigarette_data$nicotine)
mean_nicotine
[1] 0.9909
median_nicotine <- median(cigarette_data$nicotine)
median_nicotine
[1] 1.1
# histogram of nicotine levels
hist_nicotine <- hist(cigarette_data$nicotine, main="Histogram of Nicotine Data")
# 14 use the file ischemic that came with the text and look at var sbp
# systolic blood pressure
sbp <- read.csv("~/bio-informatics/Bio-stats HW 1/ischemic.csv")
mean(sbp$sbp)
[1] 150.3
sd(sbp$sbp)
[1] 18.66
# 15 lowbwt sex is binary 1 = male and 0 = female
lowbwt_data <- read.csv("~/bio-informatics/Bio-stats HW 1/lowbwt_data.csv")
boxplot(lowbwt_data$sbp ~ lowbwt_data$sex, xlab="0 = Female, 1 = Male",
ylab="Systolic Blood Pressure",
main="Box plot of Female vs. Male Systolic Blood Pressure")
