Q2 Right to work

fL <- "http://www1.aucegypt.edu/faculty/hadi/RABE5/Data5/P005.txt"
dta2 <- read.table(fL, sep = "\t", as.is = TRUE, h = T)
head(dta2)
         City COL   PD URate     Pop Taxes Income RTWL
1     Atlanta 169  414  13.6 1790128  5128   2961    1
2      Austin 143  239  11.0  396891  4303   1711    1
3 Bakersfield 339   43  23.7  349874  4166   2122    0
4   Baltimore 173  951  21.0 2147850  5001   4654    0
5 Baton Rouge  99  255  16.0  411725  3965   1620    1
6      Boston 363 1257  24.4 3914071  4928   5634    0

Q3 Junior school project

讀入資料

fL <- paste0("http://", upw, "@140.116.183.121/~sheu/dataM/Data/juniorSchools.txt")
fD <- "juniorSchools.txt"
download.file(fL, destfile = fD, mode = "wb")

jsp <- read.table(fD, h = T)
  1. 改Column名稱，檢視
jsp <- jsp %>% rename(Gender = sex)
colnames(jsp)
[1] "school"  "class"   "Gender"  "soc"     "ravens"  "pupil"   "english"
[8] "math"    "year"
  1. 改水準名稱
jsp <- jsp %>% mutate(soc = factor(soc)) 
levels(jsp$soc) <- c("I", "II", "III_0man", "III_man", "IV", "V", "VI_Unemp_L", "VII_emp_NC", "VIII_Miss_Dad")
levels(jsp$soc)
[1] "I"             "II"            "III_0man"      "III_man"      
[5] "IV"            "V"             "VI_Unemp_L"    "VII_emp_NC"   
[9] "VIII_Miss_Dad"

作圖

boxplot(math ~ soc, data = jsp, xlab = "SOC", ylab = "math")

Q4 Import AAUP2

fL <- "https://ww2.amstat.org/publications/jse/datasets/aaup2.dat.txt"

readr::fwf_empty(fL)[1:2]
$begin
 [1]  0  6 40 45 49 53 57 61 66 70 74 79 83 87 92 95

$end
 [1]  5 39 43 48 52 56 60 65 69 73 78 82 86 90 94 NA
dta4 <- read_fwf(fL, fwf_cols(ID = 5, Uni = 32, var3 = 3, var4 = 4, var5 = 4,
                              var6 = 4, var7 = 4, var8 = 5, var9 = 4, var10 = 4, 
                              var11 = 5, var12 = 4, var13 = 4, var14 = 4, var15 = 4,
                              var16 = 4, var17 = 5), na = c("*", "NA"))
Parsed with column specification:
cols(
  ID = col_integer(),
  Uni = col_character(),
  var3 = col_character(),
  var4 = col_character(),
  var5 = col_integer(),
  var6 = col_integer(),
  var7 = col_integer(),
  var8 = col_integer(),
  var9 = col_integer(),
  var10 = col_integer(),
  var11 = col_integer(),
  var12 = col_integer(),
  var13 = col_integer(),
  var14 = col_integer(),
  var15 = col_integer(),
  var16 = col_integer(),
  var17 = col_integer()
)
head(dta4,3)
# A tibble: 3 x 17
     ID Uni    var3  var4   var5  var6  var7  var8  var9 var10 var11 var12
  <int> <chr>  <chr> <chr> <int> <int> <int> <int> <int> <int> <int> <int>
1  1061 Alask~ AK    IIB     454   382   362   382   567   485   471   487
2  1063 Univ.~ AK    I       686   560   432   508   914   753   572   677
3  1065 Univ.~ AK    IIA     533   494   329   415   716   663   442   559
# ... with 5 more variables: var13 <int>, var14 <int>, var15 <int>,
#   var16 <int>, var17 <int>

Q5 LEP graphical exploration

下載並讀入資料

tmp <- tempfile()

fL <- paste0("http://", upw, "@140.116.183.121/~sheu/dataM/Data/Subject1.zip")
download.file(fL, tmp, mode = "wb")
unzip(tmp, exdir = "data")

fLs <- paste0("data/Subject1/", 1:4, "w.dat")

dta5 <- lapply(fLs, read.table, h = T, skip = 1, sep = "\t") %>% 
            rbindlist
dta5　<- dta5[,1:30]

更改column名稱

names(dta5) <- gsub("X......", "", names(dta5))
names(dta5) <- gsub("[.]", "", names(dta5))    # "." replace any character, except \n or line terminator

整理資料

new_dta <- dta5 %>% mutate(ID = 1:1804) %>% 
              mutate(electric = paste0("w", rep(1:4, each = 451))) %>% 
              gather(key = "Brain", value = "LEP", 1:30)  #the combination is from col1 to col30
              
new_dta <- new_dta %>% mutate(Brain = as.factor(Brain)) %>%
              mutate(electric = as.factor(electric)) %>%
              mutate(time = rep(seq(-100,800,by = 2), 120))

作圖

ggplot(new_dta, aes(x = time, y = LEP, color = electric)) +
  geom_line() +
  facet_wrap( ~ Brain) +
  labs(x ="Time(ms)", y = "LEP") +
  theme_bw()