/
2015_06_11.R
90 lines (71 loc) · 3.38 KB
/
2015_06_11.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
setwd("J:/R/data")
outreach <- read.csv("outreach.csv")
ratr <- read.csv("ratr.csv")
#### This tells us the values of attd "A" and "L" = attended = 1; "I" = "Incomplete" we will drop; "R" does not count in denominator we will drop
table(ratr$attd)
ratr <-ratr[which(ratr$attd!='R' & ratr$attd!='I'),] ##EDIT
###### Why does I and R still appear even if they don't have any values.
table(droplevels(ratr$attd)) ##EDIT
ratr$present <- ifelse(ratr$attd != "A", 1,0)
table(droplevels(ratr$attd), ratr$present) ##EDIT
names(outreach)
class(ratr$school.day)
head(ratr$school.day)
#install.packages("lubridate")
library(lubridate)
ratr$date <- mdy(ratr$school.day) ##EDIT
class(ratr$date)
head(ratr$date)
#### move to other data set
class(outreach$date.of.absence)
head(outreach$date.of.absence)
outreach$date <- ymd(outreach$date.of.absence) ##EDIT
class(outreach$date)
head(outreach$date)
class(ratr$student.id)
class(outreach$student.id)
# mergedset <- merge(x = ratr, y = outreach, by=c("student.id","date"), all.x=TRUE)
mergedset2 <- merge(x = ratr, y = outreach, by=c("student.id","date"))
# mergedset$hasoutreach <- ifelse(!is.na(mergedset$username), 1,0)
mergedset2$hasoutreach <- ifelse(!is.na(mergedset2$username), 1,0)
table(mergedset$username, mergedset$hasoutreach, useNA="ifany") ##GroupEDIT
table(mergedset2$username, mergedset2$hasoutreach)
sum(grepl("[Ss][Ii][Cc][Kk]", mergedset$notes)) ##EDIT
sum(grepl("[Ss][Ii][Cc][Kk]", mergedset2$notes)) ##EDIT
mergedset$sick <- grepl("[Ss][Ii][Cc][Kk]", mergedset$notes)
#mergedset$sick2 <- grep("[Ss][Ii][Cc][Kk]", mergedset$notes) ##GroupEDIT Not actually useful
sicklist <- grep("[Ss][Ii][Cc][Kk]", mergedset$notes)
sick2list <- grep("[Ss][Ii][Cc][Kk]", mergedset2$notes)
#mergedset$notes <- tolower(mergedset$notes) how would this code translate.
table(mergedset$date)
library(dplyr)
### create a before and after intervention YTD attendance.
beforeint <- mergedset[which(mergedset$date <"2015-02-09"), ]
afterint <- mergedset[which(mergedset$date >="2015-02-09"), ]
# ints <- c("beforeint", "afterint") ## not used
beforeint <- tbl_df(beforeint)
before <- group_by(beforeint, student.id) %>%
summarise(attd = mean(present), outreach = sum(hasoutreach))
afterint <- tbl_df(afterint)
after <- group_by(afterint, student.id) %>%
summarise(attd = mean(present), outreach = sum(hasoutreach), outbin = ceiling(sum(hasoutreach)/20))
# before <- tapply(beforeint$present, beforeint$student.id, mean)
# before <- data.frame(key=names(before), value=before)
# after <- tapply(afterint$present, afterint$student.id, mean)
# after <- data.frame(key=names(after), value=after)
mergeapply <- merge(x = before, y = after, by="student.id", all.x=TRUE)
mergeapply$change = (mergeapply$attd.y - mergeapply$attd.x) / mergeapply$attd.x
## These are poorly done summary stats
mean(beforeint$present, na.rm=T)
mean(afterint$present, na.rm=T)
summary(before$attd)
summary(after$attd)
## These are good summary stats
by(mergeapply$attd.x, mergeapply$outbin, function(x) summary(x))
by(mergeapply$attd.y, mergeapply$outbin, function(x) summary(x))
table(after$outbin)
table(as.factor(after$outreach))
library(ggplot2)
ggplot(data = mergeapply, aes(x = factor(outreach.y), y = change)) + geom_boxplot() + scale_y_continuous(limits=c(-1,1))
ggplot(data = mergeapply, aes(x = factor(outbin), y = change)) + geom_boxplot() + scale_y_continuous(limits=c(-1,1))
head(arrange(mergeapply, desc(change)))