-
Notifications
You must be signed in to change notification settings - Fork 0
/
CS3DP_mgmt_charts.Rmd
192 lines (132 loc) · 5.66 KB
/
CS3DP_mgmt_charts.Rmd
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
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
---
title: "CS3DP Management Survey"
output: github_document
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```
## Summary of Results
These are quick summaries of the initial results to the CS3DP Data Management survey.
1. Raw response data is in the [CS3DP Management group folder](https://drive.google.com/drive/u/0/folders/1fc-wqiG6J3lPyPC7WoWyV0GP1bJ6A2BQ)
2. Intermediate step: Made sure column headers & values were aligned
3. R-Scripts to generate or edit this doc are in [this github repo](https://github.com/magpiedin/CS3DP-management-survey)
```{r echo=FALSE, warning=FALSE, message=FALSE}
# Prep & Sumamry stats for CS3DP survey data
library("tidyr")
library("data.table")
library("ggplot2")
library("stringr")
library("stargazer")
library("plotly")
# library("gridExtra")
surveyBU <- read.csv(file = "Survey2-Data-20190422-CS3DP.csv",
stringsAsFactors = FALSE)
# drop last two columns of survey data
survey <- surveyBU[,1:(NROW(surveyBU)-2)]
# some extra data cleanup:
for (i in 15:ncol(survey)) {
# survey[,i] <- gsub(" \\(please explain below\\)\\:", "", survey[,i])
survey[,i] <- gsub(" \\(please .+", "", survey[,i])
survey[,i] <- gsub("\\. Please .+", "", survey[,i])
}
colnames(survey)[grepl("Q2$|Q2_4", colnames(survey))] <- c("Group","Group_TEXT")
survey$Group <- gsub("Both a Creator and a Repository Manager", "Both",
survey$Group)
survey$Group <- gsub("Repository manager", "Repository",
survey$Group)
# split explanatory-rows from response-rows
questionKey <- survey[1:2,]
responses <- survey[-c(1:2),]
# prep for summary-charts
questionKey2 <- questionKey[,grepl("ResponseId|Group$|Q[0-9]+$", colnames(questionKey)) > 0]
questionKey2 <- questionKey2[,c(1,3,2,4:ncol(questionKey2))]
questionKey2[3,] <- gsub(" \\(Select all that apply\\) \\- Selected Choice", "", questionKey2[1,])
responses2 <- responses[,grepl("ResponseId|Group$|Q[0-9]+$", colnames(responses)) > 0]
responses2 <- responses2[,c(1,3,2,4:ncol(responses2))]
# order variables
responses2$Group <- factor(responses2$Group, ordered = TRUE,
levels = c("Creator", "Repository",
"Both", "Other"))
ggSumList <- list()
# Response Summaries ungrouped ####
for (i in 3:ncol(responses2)) {
# split out each main response [excluding supplementary "TEXT" columns]
temp <- responses2[,c(1:2,i)]
# max number of commas in a response
ncomma <- max(str_count(temp[,3], ","))
# rename columns and split to new columns by commas
colnames(temp) <- c("ResponseId", "Group", "Qresponse")
if (ncomma > 0) {
temp2 <- separate(temp,
Qresponse,
into = paste0(colnames(responses2)[i],
"_",
seq(1:ncomma)),
sep = ",")
} else {temp2 <- temp}
# gather/transform to 2-column table
if (NCOL(temp2) > 3) {
temp3 <- gather(temp2, key = "RespNum", value = "Qresp",
3:ncol(temp2), na.rm = TRUE)
temp3 <- temp3[,-3]
} else {temp3 <- temp2}
temp3 <- temp3[nchar(temp3[,3]) > 0 & is.na(temp3[,3])==F,]
colnames(temp3) <- c("ResponseId",
"Group",
"Resp")
# colnames(responses2)[i])
assign(paste0(colnames(responses2)[i]), temp3)
ggSumList[[i-2]] <- get(paste0(colnames(responses2)[i]))
}
# # Chart setup ####
# Clean values in responses
Q6$Resp <- gsub("1990 -1994", "1990-1994", Q6$Resp)
Q10$Resp <- gsub("5-Jan", "1-5", Q10$Resp)
Q10$Resp <- gsub("10-Jun", "6-10", Q10$Resp)
Q10$Resp <- gsub("25-Nov", "11-25", Q10$Resp)
Q14$Resp <- gsub(" \n", ",", Q14$Resp)
Q18$Resp <- gsub("3-Jan", "1-3", Q18$Resp)
Q18$Resp <- gsub("9-Apr", "4-9", Q18$Resp)
Q19$Resp <- gsub("3-Jan", "1-3", Q19$Resp)
Q19$Resp <- gsub("9-Apr", "4-9", Q19$Resp)
# Re-order values in responses
Q6$Resp <- factor(Q6$Resp, ordered = TRUE,
levels = c("Before 1990", "1990-1994", "1995-1999", "2000-2004",
"2005-2009", "2010-2014", "2015-2019"))
Q10$Resp <- factor(Q10$Resp, ordered = TRUE,
levels = c("0", "1-5", "6-10", "11-25",
"26-75", "76-150", "151+"))
Q18$Resp <- factor(Q18$Resp, ordered = TRUE,
levels = c("<1", "1-3", "4-9", "10+"))
Q19$Resp <- factor(Q19$Resp, ordered = TRUE,
levels = c("<1", "1-3", "4-9", "10+"))
```
## Charts {#chartsanchor}
```{r echo=FALSE}
# Generate graphs
# # If need to cut out columns with non-categ. data
# # https://stackoverflow.com/questions/33962630/knitr-r-grouping-multiple-plots-in-to-html
for (i in 1:NROW(ggSumList)) {
Qresp <- get(paste0(colnames(responses2)[i+2]))
# Only generate a chart if number of rows > 0
if (NROW(Qresp) > 0) {
if (grepl("Q14", colnames(responses2)[i+2]) < 1) {
print(
ggplotly(
ggplot(data = Qresp,
aes(x = Resp,
fill = Resp)) +
geom_bar(position = "dodge") +
facet_grid(Qresp[,2] ~ .) +
labs(title = colnames(responses2)[i+2],
subtitle = str_wrap(questionKey2[3,c(colnames(responses2)[i+2])]),
x = "", y = "count") +
theme(axis.text.x = element_text(angle = 45, hjust = 1, size = 10),
plot.title=element_text(size=20),
legend.position="none")
)
)
}
}
}
```