Rscripts/WmGuidance_Journey_4.R at master · hbvndrzwt/Rscripts · GitHub

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
#This R script should be an indication on how to procede with analytics after experiments
#This script is specific for the Journey_4 experiment of the WmGuidance
# ctrl + shift + c for multiple line commenting :)

# Metrics:
# GuidanceFinished
# LectureCompleted
# WmActivated
# Activated
# Retained

# Groupings:
# CompanyType
# UserType
# ExperimentGroup
# GuidanceFinished
# BrokenEvents

## MISSING :


#Install packages if necessary
# install.packages("compute.es");
# install.packages("car");
# install.packages ("ggplot2");
# install.packages("multcomp");
# install.packages("pastecs");
# install.packages("WRS");
# install.packages("reshape");
# install.packages("lazyeval");
# install.packages("ggpubr");
# install.packages("mlogit");

library(compute.es);
library(car);
library(ggplot2);
library(multcomp);
library(pastecs);
library(WRS);
library(reshape);
library(lazyeval);
library(ggplot2);
library(ggpubr);
library(mlogit);
library(readxl)

#Set working directory for output files
setwd("C:\\Users\\hzw\\Dropbox (Mendix)\\Community\\Onboarding\\_Analytics\\R Scripts\\Experiment Analysis\\Output")

#Load Dataset
Journey_4_Output <- read_excel("C:/Users/hzw/Dropbox (Mendix)/Community/Onboarding/_Analytics/Experiment Output/Journey_4_Output.xlsx")
View(Journey_4_Output)
Output <- Journey_4_Output

#Disregard null values in dataframe
Output$UserType[Output$UserType == "NULL"] <- NA
Output$BrokenEvents[Output$BrokenEvents == "NULL"] <- 0
Output$UserType[Output$UserType == "Project management"] <- NA #This group is too small to take into consideration in this experiment

#Set the independent variables as factors
Output$CompanyType <- factor(Output$CompanyType, levels = c("Other", "Customer", "Partner", "University"))
Output$UserType <- factor(Output$UserType, levels = c("Low-code", "No-code", "Project management"))
Output$ExperimentGroup <- as.factor(Output$ExperimentGroup)

#Setting contrasts for the independent variables - IF NECESSARY
#contrasts(Output$Grouping1) <- cbind(c(-3, 1, 1, 1), c(0, -2, 1, 1), c(0, 0, -1, 1))
#contrasts(Output$Grouping2) <- cbind(c(-2, 1, 1), c(0, -1, 1))
#contrasts(Output$ExperimentGroup) <- c(-1, 1)


#Creating Subsets
Other <- subset(Output, CompanyType == "Other")
Customer <- subset(Output, CompanyType == "Customer")
Partner <- subset(Output, CompanyType == "Partner")
University <- subset(Output, CompanyType == "University")

NoCode <- subset(Output, UserType == "No-code")
LowCode <- subset(Output, UserType == "Low-code")
ProjectManagement <- subset(Output, UserType == "Project management")

A <- subset(Output, ExperimentGroup == "A")
B <- subset(Output, ExperimentGroup == "B")


#Set Metric & Grouping for below script
# colnames(Output)[which(names(Output) == "LectureCompleted")] <- "Metric"
# colnames(Output)[which(names(Output) == "Grouping")] <- "UserType"


#First get some Descriptive statistics per metric
by(Output$Metric, Output$Grouping, stat.desc)


#Creating graphs for Report and Identify metrics/groups to focus on
ggline(Journey_4_Output, x = "BrokenEvents", y = "LectureCompleted", color = "ExperimentGroup", plot_type = "l", add = c("mean_ci"),  palette = c("#00AFBB", "#E7B800"),
       xlab = "", ylab = "LectureCompleted")#+facet_grid(.~BrokenEvents)


## ONE-FACTOR ANALYSIS

#Create a Linear Model: Outcome = Group + error
LinearModel1 <- glm(WmActivated ~ CompanyType, data = Output, family = binomial())

#Create an ANOVA Model
AnovaModel <- aov(WmActivated ~ ExperimentGroup, data = Customer)
AnovaModel1 <- aov(WmActivated ~ ExperimentGroup, data = Other)
AnovaModel2 <- aov(WmActivated ~ ExperimentGroup, data = Partner)
AnovaModel3 <- aov(WmActivated ~ ExperimentGroup, data = University)


Output$GuidanceFinished <- as.factor(Output$GuidanceFinished)
Journey_4_Output$BrokenEvents <- as.factor(Journey_4_Output$BrokenEvents)
AnovaGroup <- aov(GuidanceFinished ~ BrokenEvents, data = Journey_4_Output)

#Summary of Models
summary(AnovaModel)
summary(AnovaModel1)
summary(AnovaModel2)
summary(AnovaModel3)
summary(AnovaGroup)

summary(LinearModel1)


out <- capture.output(summary(AnovaModel1))
cat("ANOVA Model LectureCompleted ~ ExperimentGroup (Other)", out, file = "testfile.txt", sep = "\t", append = TRUE)


#TWO-FACTOR ANOVA (Not sure if/how to set this up)

#Get descriptive statistics for multiple independent variables on the dependent variable
by(Output$Metric3, list(Output$Grouping2, Output$ExperimentGroup), stat.desc)


#Define the Factorial Anova models: Set up the models we want to look at (can be more than 1 model)
FactAnovaModel1 <- aov(Metric3 ~ ExperimentGroup*Grouping1, data = Output, family = binomial())
FactAnovaModel2 <- aov(Metric1 ~ ExperimentGroup*Grouping1*Grouping2, data = Output)

#Define Linear Models: Set up the models we want to look at (can be more than 1 model)
GLMModel <- glm(Metric3 ~ ExperimentGroup*Grouping2, data = Output, family = binomial())

#Summary of the Models
summary(GLMModel)
Anova(FactAnovaModel1, type = "III")


## Additional - Archived
#post hoc analysis
pairwise.t.test(Output$Metric1, Output$Grouping1, p.adjust.method = "bonferroni")
postHocs <- glht(FactAnovaModel1, linfct = mcp(Grouping1 = "Tukey"))
summary(postHocs)
confint(postHocs)


#NOT WORKING
interaction.plot(x.factor = Output$Grouping1, trace.factor = Output$ExperimentGroup, response = Output$Metric1)