Inclass Exercise1_20211013
Ruei-Ning
2021/10/24
The math attainment page has a dataset and a script of R code chunks. Generate a markdown file from the script to push the output in HTML for posting to course Moodle site.
first R session using math attainment data set
input data
read in a plain text file with variable names and assign a name to it
把資料讀到dta <- read.table(“D:/User/Desktop/datamanagement/20211018/math_attainment.txt”, header = T) 檔案有title 的話，一定要加 header=T(true)
dta <- read.table("D:/User/Desktop/datamanagement/20211018/math_attainment.txt", header = T)
dta## math2 math1 cc
## 1 28 18 328.20
## 2 56 22 406.03
## 3 51 44 386.94
## 4 13 8 166.91
## 5 39 20 328.20
## 6 41 12 328.20
## 7 30 16 160.28
## 8 13 5 94.94
## 9 17 9 98.20
## 10 32 18 331.62
## 11 32 20 328.20
## 12 46 30 191.36
## 13 37 10 191.36
## 14 36 16 191.36
## 15 27 14 211.11
## 16 23 14 131.25
## 17 23 9 131.25
## 18 44 24 203.54
## 19 17 8 211.11
## 20 23 5 44.72
## 21 17 10 135.08
## 22 27 13 150.39
## 23 21 7 85.26
## 24 18 14 129.82
## 25 47 21 187.43
## 26 21 4 129.82
## 27 34 16 144.25
## 28 26 12 141.15
## 29 16 10 115.47
## 30 27 16 129.82
## 31 23 11 155.67
## 32 35 16 187.43
## 33 31 16 141.15
## 34 20 14 189.14
## 35 23 12 187.43
## 36 37 29 187.43
## 37 31 25 189.68
## 38 19 16 183.61
## 39 21 15 129.82checking data
structure of data
str(dta)## 'data.frame': 39 obs. of 3 variables:
## $ math2: int 28 56 51 13 39 41 30 13 17 32 ...
## $ math1: int 18 22 44 8 20 12 16 5 9 18 ...
## $ cc : num 328 406 387 167 328 ...dta裡面有3個變數、39個觀察值
first 6 rows
head(dta)## math2 math1 cc
## 1 28 18 328.20
## 2 56 22 406.03
## 3 51 44 386.94
## 4 13 8 166.91
## 5 39 20 328.20
## 6 41 12 328.20列出前六個觀察值
summary(dta$math1)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 4.00 10.00 14.00 15.36 18.00 44.00summary(dta$math2)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 13.00 21.00 27.00 28.77 35.50 56.00summary(dta$cc)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 44.72 131.25 183.61 188.84 197.45 406.03做變項(math1, math2, cc)的總覽
descriptive statistics
variable mean
colMeans(dta)## math2 math1 cc
## 28.76923 15.35897 188.83667計算column的平均值
variable sd
apply(dta, 2, sd)## math2 math1 cc
## 10.720029 7.744224 84.842513在上面的 apply() 中，dta 是我們的資料 (矩陣)，1 代表 apply 的計算是 by row，sd 是告訴 apply 我們要算每個列的sd。
apply(dta, 1, sum)## [1] 374.20 484.03 481.94 187.91 387.20 381.20 206.28 112.94 124.20 381.62
## [11] 380.20 267.36 238.36 243.36 252.11 168.25 163.25 271.54 236.11 72.72
## [21] 162.08 190.39 113.26 161.82 255.43 154.82 194.25 179.15 141.47 172.82
## [31] 189.67 238.43 188.15 223.14 222.43 253.43 245.68 218.61 165.822 告訴 apply 這次的計算要 by column ，sum代表加總，應該會有39筆資料(39個觀察值)。 這部分是嘗試apply中1, 2的差別。
correlation matrix
cor(dta)## math2 math1 cc
## math2 1.0000000 0.7443604 0.6570098
## math1 0.7443604 1.0000000 0.5956771
## cc 0.6570098 0.5956771 1.0000000做dta資料變項的相關性。
plot data
specify square plot region
par(pty="s")scatter plot of math2 by math1
plot(math2 ~ math1, data=dta, xlim=c(0, 60), ylim=c(0, 60),
xlab="Math score at Year 1", ylab="Math score at Year 2")
add grid lines
grid()
plot(math2 ~ math1, data=dta, xlim=c(0, 60), ylim=c(0, 60),
xlab="Math score at Year 1", ylab="Math score at Year 2")
grid() #加上背景參考線
abline(h=mean(dta$math2), lty=2) #加上math2的平均值當橫線
regression analysis
regress math2 by math1
dta.lm <- lm(math2 ~ math1, data=dta)做math2和math1的迴歸分析，命名為dta,lm # show results
summary(dta.lm)##
## Call:
## lm(formula = math2 ~ math1, data = dta)
##
## Residuals:
## Min 1Q Median 3Q Max
## -10.430 -5.521 -0.369 4.253 20.388
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 12.944 2.607 4.965 1.57e-05 ***
## math1 1.030 0.152 6.780 5.57e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 7.255 on 37 degrees of freedom
## Multiple R-squared: 0.5541, Adjusted R-squared: 0.542
## F-statistic: 45.97 on 1 and 37 DF, p-value: 5.571e-08show anova table
anova(dta.lm)## Analysis of Variance Table
##
## Response: math2
## Df Sum Sq Mean Sq F value Pr(>F)
## math1 1 2419.6 2419.59 45.973 5.571e-08 ***
## Residuals 37 1947.3 52.63
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1add regression line
plot(math2 ~ math1, data=dta, xlim=c(0, 60), ylim=c(0, 60),
xlab="Math score at Year 1", ylab="Math score at Year 2")
abline(dta.lm, lty=2)
做Math1和Math2的plot，並加上dta.lm這條迴歸線。 # add plot title
plot(math2 ~ math1, data=dta, xlim=c(0, 60), ylim=c(0, 60),
xlab="Math score at Year 1", ylab="Math score at Year 2")
abline(dta.lm, lty=2)
title("Mathematics Attainment with the Regression Line")
加上圖的標題title(“Mathematics Attainment”)
diagnostics
specify maximum plot region
par(pty=“m”)
#
par(pty="m")
plot(scale(resid(dta.lm)) ~ fitted(dta.lm),
ylim=c(-3.5, 3.5), type="n",
xlab="Fitted values", ylab="Standardized residuals")
#give the value and lable
text(fitted(dta.lm), scale(resid(dta.lm)), labels=rownames(dta), cex=0.5)
grid()#背景參考線
# add a horizontal red dash line
abline(h=0, lty=2, col="red")
## normality check
qqnorm(scale(resid(dta.lm))) #殘差的qqplot
qqline(scale(resid(dta.lm))) #殘差的迴歸線
grid()