calc_results.R

# Executing this script prints results and creates diagrams as PDFs in
# the current directory. Note that this script may contain german labels 
# or notes, as the original questionnaire was in German as well.
# Code partially stems from the lecture "Softwaretechnik" by Rainer Koschke.
library(missMethods)
library(data.table)
library(readr)
library(ggplot2)
library(ggridges)
library(esquisse)
library(scales)
library(hrbrthemes)
library(ggstatsplot)
library(coin)
library(effectsize)
SIZE <- 20
WIDTH <- 9
HEIGHT <- 7
ourtheme <- theme(
   axis.ticks = element_blank(),
   axis.line = element_line(color = "grey50"),
   panel.grid = element_line(color = "#b4aea9"),
   panel.grid.minor = element_blank(),
   panel.grid.major.x = element_blank(),
   panel.grid.major.y = element_line(linetype = "dashed")
)
get_dist <- function(first, second) {
   transform(data.frame(first, second), count = ave(rep(1, nrow(data.frame(first, second))), first, second, FUN = seq_along))
}
violin <- function(data, label) {
    ggbetweenstats(data=data, x=x, y=y, xlab="Group", ylab=label, plot.type="boxviolin", type="np", package="ggsci", palette="uniform_startrek", ggtheme = theme_grey(base_size = SIZE), outlier.tagging = TRUE, outlier.label.args = list(size=6, color = "red", nudge_x=0.2), results.subtitle=F, point.args=list(size=5, alpha=0.4, position=position_jitter(w=0.1, h=0)), centrality.point.args=list(size=7, color="darkgreen", alpha=0.6), centrality.label.args=list(size=6, nudge_x=0.5, nudge_y=0.2, segment.linetype=1, min.segment.length=0)) + ourtheme + scale_y_continuous(breaks=pretty_breaks())
    #ggplot(data.frame(as.factor(group), as.numeric(data)), aes(x=as.factor(group), y=as.numeric(data))) + geom_violin(trim=T) + theme_grey(base_size = SIZE) + geom_boxplot(width=0.1, color="grey", alpha=0.4) + xlab("Group") + ylab(label)
}
bar <- function(data, label) {
    # NOTE: label currently unused
    ggplot(as.data.frame(table(data)), aes(x=data, y=Freq)) + geom_bar(stat="identity") + coord_flip() + labs(x = "", y="Count") + theme_grey(base_size = SIZE) + scale_y_continuous(breaks=pretty_breaks()) + ourtheme
}
report = function(data, name) {
    cat(name, "Average", mean(data), "Median", median(data), "Standard Deviation", sd(data), "\n")
}
df <- read_delim("Dashboard_SEE.csv", ";", escape_double = FALSE, trim_ws = TRUE)
age <- df$age
# Age
report(age, "Age")
group <- df$group
cairo_pdf("Age.pdf", width=WIDTH)
violin(data.frame(y=age, x=group), "Age")
dev.off()
first <- df[1:10,]
second <- df[11:20,]
report(first$age, "Age Group 1")
report(second$age, "Age Group 2")
wilcox.test(x=first$age, y=second$age, alternative="two.sided", paired=FALSE, exact=TRUE)
# Degree
getfirst = function(data) {
   return(data[1:10])
}
getsecond = function(data) {
   return(data[11:20])
}
abschluss <- factor(df$abschluss, order = TRUE, levels = c("Noch in der Schule","Ohne Abschluss","Hauptschulabschluss","Mittlerer Schulabschluss (z.B. Realschulabschluss)","Fachhochschulreife (Fachabitur)","Allgemeine Hochschulreife (Abitur)","Bachelor-Abschluss","Master-Abschluss","Doktorgrad"))
print("Degree: ")
median(abschluss)
wilcox.test(x=as.numeric(getfirst(abschluss)), y=as.numeric(getsecond(abschluss)), alternative="two.sided", paired=FALSE, exact=TRUE)
cairo_pdf("Degree.pdf", width=WIDTH)
violin(data.frame(y=as.numeric(abschluss), x=group), "Highest Degree")
dev.off()
cairo_pdf("DataDegree.pdf", width=WIDTH*1.2)
bar(abschluss, "Highest Degree")
dev.off()
# Programming experience
programming <- factor(df$programming, order = TRUE, levels = c("Gar nicht","Weniger als 3 Jahre","3–9 Jahre","10–19 Jahre","20 Jahre oder mehr"))
print("Programming experience:")
median(programming)
wilcox.test(x=as.numeric(getfirst(programming)), y=as.numeric(getsecond(programming)), alternative="two.sided", paired=FALSE, exact=TRUE)
cairo_pdf("Programming.pdf", width=WIDTH)
violin(data.frame(y=as.numeric(programming), x=group), "Programming experience")
dev.off()
cairo_pdf("DataProgramming.pdf", width=WIDTH, height=HEIGHT/1.3)
bar(programming, "Programming experience")
dev.off()
# Open-Source-Experience
opensource <- factor(df$opensource, order = TRUE, levels = c("Noch gar nicht","Weniger als 3 Jahre","3–9 Jahre","10–19 Jahre","20 Jahre oder mehr"))
print("Experience in bigger software projects:")
median(opensource)
wilcox.test(x=as.numeric(getfirst(opensource)), y=as.numeric(getsecond(opensource)), alternative="two.sided", paired=FALSE, exact=TRUE)
cairo_pdf("BiggerProjects.pdf")
violin(data.frame(y=as.numeric(opensource), x=group), "Experience in bigger software projects")
dev.off()
cairo_pdf("DataBiggerProjects.pdf", height=HEIGHT/1.3)
bar(opensource, "Experience in bigger software projects")
dev.off()
# Experience with SEE
knowsee <- factor(df$knowsee, order = TRUE, levels = c("Nein", "Ja, habe davon gehört","Ja, habe ich bereits genutzt", "Ja, habe daran entwickelt"))
print("SEE experience:")
median(knowsee)
wilcox.test(x=as.numeric(getfirst(knowsee)), y=as.numeric(getsecond(knowsee)), alternative="two.sided", paired=FALSE, exact=TRUE)
cairo_pdf("KnowSEE.pdf", width=WIDTH)
violin(data.frame(y=as.numeric(knowsee), x=group), "Experience with SEE")
dev.off()
cairo_pdf("DataSEE.pdf", width=WIDTH, height=HEIGHT/1.3)
bar(knowsee, "Experience with SEE")
dev.off()
# Experience with the Dashboard
knowdash <- factor(df$knowdash, order = TRUE, levels = c("Nein", "Ja, habe davon gehört","Ja, habe ich bereits genutzt", "Ja, habe daran entwickelt"))
print("Dashboard experience:")
median(knowdash)
wilcox.test(x=as.numeric(getfirst(knowdash)), y=as.numeric(getsecond(knowdash)), alternative="two.sided", paired=FALSE, exact=TRUE)
cairo_pdf("KnowDash.pdf", width=WIDTH)
violin(data.frame(y=as.numeric(knowdash), x=group), "Experience with the Dashboard")
dev.off()
cairo_pdf("DataDash.pdf", width=WIDTH, height=HEIGHT/1.3)
bar(knowdash, "Experience with the Dashboard")
dev.off()
# Experience with video games
knowgames <- factor(df$knowvideogames, order = TRUE, levels = c("Ja, habe bereits viel gespielt","Ja, habe schon ein bisschen gespielt","Ja, aber habe kaum gespielt","Nein, noch nie"))
print("Video game experience: ")
median(knowgames)
wilcox.test(x=as.numeric(getfirst(knowgames)), y=as.numeric(getsecond(knowgames)), alternative="two.sided", paired=FALSE, exact=FALSE)
cairo_pdf("KnowGame.pdf", width=WIDTH)
violin(data.frame(y=as.numeric(knowgames), x=group), "Experience with video games")
dev.off()
cairo_pdf("DataGame.pdf", width=WIDTH, height=HEIGHT/1.3)
bar(knowgames, "Experience with video games")
dev.off()
# Now the exciting part
# Correctness
d1f <- c(sum(getfirst(df$a1_1_file_c), na.rm = TRUE), sum(getfirst(df$a1_2_file_c), na.rm = TRUE), sum(getfirst(df$a1_3_file_c), na.rm = TRUE), sum(getfirst(df$a1_4_file_c), na.rm = TRUE), sum(getfirst(df$a1_5_file_c), na.rm = TRUE))/10
d1n <- c(sum(getfirst(df$a1_1_num_c), na.rm = TRUE), sum(getfirst(df$a1_2_num_c), na.rm = TRUE), sum(getfirst(df$a1_3_num_c), na.rm = TRUE), sum(getfirst(df$a1_4_num_c), na.rm = TRUE), sum(getfirst(df$a1_5_num_c), na.rm = TRUE))/10
d1 <- c(d1f, d1n)
d2f <- c(sum(getfirst(df$a2_1_file_c), na.rm = TRUE), sum(getfirst(df$a2_2_file_c), na.rm = TRUE), sum(getfirst(df$a2_3_file_c), na.rm = TRUE), sum(getfirst(df$a2_4_file_c), na.rm = TRUE), sum(getfirst(df$a2_5_file_c), na.rm = TRUE))/10
d2n <- c(sum(getfirst(df$a2_1_num_c), na.rm = TRUE), sum(getfirst(df$a2_2_num_c), na.rm = TRUE), sum(getfirst(df$a2_3_num_c), na.rm = TRUE), sum(getfirst(df$a2_4_num_c), na.rm = TRUE), sum(getfirst(df$a2_5_num_c), na.rm = TRUE))/10
d2 <- c(d2f, d2n)
d3 <- c(sum(getfirst(df$a3_1_c), na.rm = TRUE), sum(getfirst(df$a3_2_c), na.rm = TRUE), sum(getfirst(df$a3_3_c), na.rm = TRUE), sum(getfirst(df$a3_4_c), na.rm = TRUE), sum(getfirst(df$a3_5_c), na.rm = TRUE))/10
d4f <- c(sum(getsecond(df$a4_1_file_c), na.rm = TRUE), sum(getsecond(df$a4_2_file_c), na.rm = TRUE), sum(getsecond(df$a4_3_file_c), na.rm = TRUE), sum(getsecond(df$a4_4_file_c), na.rm = TRUE), sum(getsecond(df$a4_5_file_c), na.rm = TRUE))/10
d4n <- c(sum(getsecond(df$a4_1_num_c), na.rm = TRUE), sum(getsecond(df$a4_2_num_c), na.rm = TRUE), sum(getsecond(df$a4_3_num_c), na.rm = TRUE), sum(getsecond(df$a4_4_num_c), na.rm = TRUE), sum(getsecond(df$a4_5_num_c), na.rm = TRUE))/10
d4 <- c(d4f, d4n)
d5f <- c(sum(getsecond(df$a5_1_file_c), na.rm = TRUE), sum(getsecond(df$a5_2_file_c), na.rm = TRUE), sum(getsecond(df$a5_3_file_c), na.rm = TRUE), sum(getsecond(df$a5_4_file_c), na.rm = TRUE), sum(getsecond(df$a5_5_file_c), na.rm = TRUE))/10
d5n <- c(sum(getsecond(df$a5_1_num_c), na.rm = TRUE), sum(getsecond(df$a5_2_num_c), na.rm = TRUE), sum(getsecond(df$a5_3_num_c), na.rm = TRUE), sum(getsecond(df$a5_4_num_c), na.rm = TRUE), sum(getsecond(df$a5_5_num_c), na.rm = TRUE))/10
d5 <- c(d5f, d5n)
d6 <- c(sum(getsecond(df$a6_1_c), na.rm = TRUE), sum(getsecond(df$a6_2_c), na.rm = TRUE), sum(getsecond(df$a6_3_c), na.rm = TRUE), sum(getsecond(df$a6_4_c), na.rm = TRUE), sum(getsecond(df$a6_5_c), na.rm = TRUE))/10
dc_all <- c(d1, d2, d3, d4, d5, d6)
s1f <- c(sum(getsecond(df$a1_1_file_c), na.rm = TRUE), sum(getsecond(df$a1_2_file_c), na.rm = TRUE), sum(getsecond(df$a1_3_file_c), na.rm = TRUE), sum(getsecond(df$a1_4_file_c), na.rm = TRUE), sum(getsecond(df$a1_5_file_c), na.rm = TRUE))/10
s1n <- c(sum(getsecond(df$a1_1_num_c), na.rm = TRUE), sum(getsecond(df$a1_2_num_c), na.rm = TRUE), sum(getsecond(df$a1_3_num_c), na.rm = TRUE), sum(getsecond(df$a1_4_num_c), na.rm = TRUE), sum(getsecond(df$a1_5_num_c), na.rm = TRUE))/10
s1 <- c(s1f, s1n)
s2f <- c(sum(getsecond(df$a2_1_file_c), na.rm = TRUE), sum(getsecond(df$a2_2_file_c), na.rm = TRUE), sum(getsecond(df$a2_3_file_c), na.rm = TRUE), sum(getsecond(df$a2_4_file_c), na.rm = TRUE), sum(getsecond(df$a2_5_file_c), na.rm = TRUE))/10
s2n <- c(sum(getsecond(df$a2_1_num_c), na.rm = TRUE), sum(getsecond(df$a2_2_num_c), na.rm = TRUE), sum(getsecond(df$a2_3_num_c), na.rm = TRUE), sum(getsecond(df$a2_4_num_c), na.rm = TRUE), sum(getsecond(df$a2_5_num_c), na.rm = TRUE))/10
s2 <- c(s2f, s2n)
s3 <- c(sum(getsecond(df$a3_1_c), na.rm = TRUE), sum(getsecond(df$a3_2_c), na.rm = TRUE), sum(getsecond(df$a3_3_c), na.rm = TRUE), sum(getsecond(df$a3_4_c), na.rm = TRUE), sum(getsecond(df$a3_5_c), na.rm = TRUE))/10
s4f <- c(sum(getfirst(df$a4_1_file_c), na.rm = TRUE), sum(getfirst(df$a4_2_file_c), na.rm = TRUE), sum(getfirst(df$a4_3_file_c), na.rm = TRUE), sum(getfirst(df$a4_4_file_c), na.rm = TRUE), sum(getfirst(df$a4_5_file_c), na.rm = TRUE))/10
s4n <- c(sum(getfirst(df$a4_1_num_c), na.rm = TRUE), sum(getfirst(df$a4_2_num_c), na.rm = TRUE), sum(getfirst(df$a4_3_num_c), na.rm = TRUE), sum(getfirst(df$a4_4_num_c), na.rm = TRUE), sum(getfirst(df$a4_5_num_c), na.rm = TRUE))/10
s4 <- c(s4f, s4n)
s5f <- c(sum(getfirst(df$a5_1_file_c), na.rm = TRUE), sum(getfirst(df$a5_2_file_c), na.rm = TRUE), sum(getfirst(df$a5_3_file_c), na.rm = TRUE), sum(getfirst(df$a5_4_file_c), na.rm = TRUE), sum(getfirst(df$a5_5_file_c), na.rm = TRUE))/10
s5n <- c(sum(getfirst(df$a5_1_num_c), na.rm = TRUE), sum(getfirst(df$a5_2_num_c), na.rm = TRUE), sum(getfirst(df$a5_3_num_c), na.rm = TRUE), sum(getfirst(df$a5_4_num_c), na.rm = TRUE), sum(getfirst(df$a5_5_num_c), na.rm = TRUE))/10
s5 <- c(s5f, s5n)
s6 <- c(sum(getfirst(df$a6_1_c), na.rm = TRUE), sum(getfirst(df$a6_2_c), na.rm = TRUE), sum(getfirst(df$a6_3_c), na.rm = TRUE), sum(getfirst(df$a6_4_c), na.rm = TRUE), sum(getfirst(df$a6_5_c), na.rm = TRUE))/10
sc_all <- c(s1, s2, s3, s4, s5, s6)
bothc <- merge(data.frame(percent=sc_all, type=factor("SEE")), data.frame(percent=dc_all, type=factor("Dashboard")), all=T)
# Previously summarized by column, now by row per exercise
assignment1 <- data.frame(group=c("α","β"), type=c(factor("Dashboard"), factor("SEE")))
assignment2 <- data.frame(group=c("β","α"), type=c(factor("Dashboard"), factor("SEE")))
a1c <- as.numeric(df$a1_1_file_c) + as.numeric(df$a1_2_file_c) + as.numeric(df$a1_3_file_c) + as.numeric(df$a1_4_file_c) + as.numeric(df$a1_5_file_c) + as.numeric(df$a1_1_num_c) + as.numeric(df$a1_2_num_c) + as.numeric(df$a1_3_num_c) + as.numeric(df$a1_4_num_c) + as.numeric(df$a1_5_num_c)
a1cu <- as.numeric(df$a1_1_file_c_uo) + as.numeric(df$a1_2_file_c_uo) + as.numeric(df$a1_3_file_c_uo) + as.numeric(df$a1_4_file_c_uo) + as.numeric(df$a1_5_file_c_uo) + as.numeric(df$a1_1_num_c) + as.numeric(df$a1_2_num_c) + as.numeric(df$a1_3_num_c) + as.numeric(df$a1_4_num_c) + as.numeric(df$a1_5_num_c)
a1cp <- merge(x=data.frame(p=a1c/10, group), y=assignment1, by="group")
a1cup <- merge(x=data.frame(p=a1cu/10, group), y=assignment1, by="group")
a2c <- as.numeric(df$a2_1_file_c) + as.numeric(df$a2_2_file_c) + as.numeric(df$a2_3_file_c) + as.numeric(df$a2_4_file_c) + as.numeric(df$a2_5_file_c) + as.numeric(df$a2_1_num_c) + as.numeric(df$a2_2_num_c) + as.numeric(df$a2_3_num_c) + as.numeric(df$a2_4_num_c) + as.numeric(df$a2_5_num_c)
a2cu <- as.numeric(df$a2_1_file_c_uo) + as.numeric(df$a2_2_file_c_uo) + as.numeric(df$a2_3_file_c_uo) + as.numeric(df$a2_4_file_c_uo) + as.numeric(df$a2_5_file_c_uo) + as.numeric(df$a2_1_num_c) + as.numeric(df$a2_2_num_c) + as.numeric(df$a2_3_num_c) + as.numeric(df$a2_4_num_c) + as.numeric(df$a2_5_num_c)
a2cp <- merge(x=data.frame(p=a2c/10, group), y=assignment1, by="group")
a2cup <- merge(x=data.frame(p=a2cu/10, group), y=assignment1, by="group")
a3c <- as.numeric(df$a3_1_c) + as.numeric(df$a3_2_c) + as.numeric(df$a3_3_c) + as.numeric(df$a3_4_c) + as.numeric(df$a3_5_c)
a3cp <- merge(x=data.frame(p=a3c/5, group), y=assignment1, by="group")
a4c <- as.numeric(df$a4_1_file_c) + as.numeric(df$a4_2_file_c) + as.numeric(df$a4_3_file_c) + as.numeric(df$a4_4_file_c) + as.numeric(df$a4_5_file_c) + as.numeric(df$a4_1_num_c) + as.numeric(df$a4_2_num_c) + as.numeric(df$a4_3_num_c) + as.numeric(df$a4_4_num_c) + as.numeric(df$a4_5_num_c)
a4cu <- as.numeric(df$a4_1_file_c_uo) + as.numeric(df$a4_2_file_c_uo) + as.numeric(df$a4_3_file_c_uo) + as.numeric(df$a4_4_file_c_uo) + as.numeric(df$a4_5_file_c_uo) + as.numeric(df$a4_1_num_c) + as.numeric(df$a4_2_num_c) + as.numeric(df$a4_3_num_c) + as.numeric(df$a4_4_num_c) + as.numeric(df$a4_5_num_c)
a4cp <- merge(x=data.frame(p=a4c/10, group), y=assignment2, by="group")
a4cup <- merge(x=data.frame(p=a4cu/10, group), y=assignment2, by="group")
a5c <- as.numeric(df$a5_1_file_c) + as.numeric(df$a5_2_file_c) + as.numeric(df$a5_3_file_c) + as.numeric(df$a5_4_file_c) + as.numeric(df$a5_5_file_c) + as.numeric(df$a5_1_num_c) + as.numeric(df$a5_2_num_c) + as.numeric(df$a5_3_num_c) + as.numeric(df$a5_4_num_c) + as.numeric(df$a5_5_num_c)
a5cu <- as.numeric(df$a5_1_file_c_uo) + as.numeric(df$a5_2_file_c_uo) + as.numeric(df$a5_3_file_c_uo) + as.numeric(df$a5_4_file_c_uo) + as.numeric(df$a5_5_file_c_uo) + as.numeric(df$a5_1_num_c) + as.numeric(df$a5_2_num_c) + as.numeric(df$a5_3_num_c) + as.numeric(df$a5_4_num_c) + as.numeric(df$a5_5_num_c)
a5cp <- merge(x=data.frame(p=a5c/10, group), y=assignment2, by="group")
a5cup <- merge(x=data.frame(p=a5cu/10, group), y=assignment2, by="group")
a6c <- as.numeric(df$a6_1_c) + as.numeric(df$a6_2_c) + as.numeric(df$a6_3_c) + as.numeric(df$a6_4_c) + as.numeric(df$a6_5_c)
a6cp <- merge(x=data.frame(p=a6c/5, group), y=assignment2, by="group")
aacp <- data.frame(p=(a1c+a2c+a3c+a4c+a5c+a6c)/50, group)
aacup <- data.frame(p=(a1cu+a2cu+a3c+a4cu+a5cu+a6c)/50, group)
report(a1cp$p, "Correctness A1")
wilcox.test(x=getfirst(a1cp$p), y=getsecond(a1cp$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a1cp$p), x=getsecond(a1cp$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(y=getfirst(a1cp$p), x=getsecond(a1cp$p), alternative="less", paired=FALSE, exact=TRUE)
report(a2cp$p, "Correctness A2")
wilcox.test(x=getfirst(a2cp$p), y=getsecond(a2cp$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a2cp$p), x=getsecond(a2cp$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(y=getfirst(a2cp$p), x=getsecond(a2cp$p), alternative="less", paired=FALSE, exact=TRUE)
report(a3cp$p, "Correctness A3")
wilcox.test(x=getfirst(a3cp$p), y=getsecond(a3cp$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a3cp$p), x=getsecond(a3cp$p), alternative="less", paired=FALSE, exact=TRUE)
report(a4cp$p, "Correctness A4")
wilcox.test(x=getsecond(a4cp$p), y=getfirst(a4cp$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(x=getsecond(a4cp$p), y=getfirst(a4cp$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getsecond(a4cp$p), x=getfirst(a4cp$p), alternative="less", paired=FALSE, exact=TRUE)
# We need to correct ties here and use a permutation test
DV <- c(getfirst(a4cp$p), getsecond(a4cp$p))
IV <- factor(rep(c("A", "B"), c(10, 10)))
oneway_test(DV ~ IV, alternative="greater", exact=T)
report(a5cp$p, "Correctness A5")
wilcox.test(x=getsecond(a5cp$p), y=getfirst(a5cp$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getsecond(a5cp$p), x=getfirst(a5cp$p), alternative="less", paired=FALSE, exact=TRUE)
report(a6cp$p, "Correctness A6")
wilcox.test(x=getsecond(a6cp$p), y=getfirst(a6cp$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getsecond(a6cp$p), x=getfirst(a6cp$p), alternative="less", paired=FALSE, exact=TRUE)
cairo_pdf("A1C.pdf", width=WIDTH)
violin(data.frame(y=a1cp$p, x=a1cp$type), "Correctness")
dev.off()
cairo_pdf("A2C.pdf", width=WIDTH)
violin(data.frame(y=a2cp$p, x=a2cp$type), "Correctness")
dev.off()
cairo_pdf("A3C.pdf", width=WIDTH)
violin(data.frame(y=a3cp$p, x=a3cp$type), "Correctness")
dev.off()
cairo_pdf("A4C.pdf", width=WIDTH)
violin(data.frame(y=a4cp$p, x=a4cp$type), "Correctness")
dev.off()
cairo_pdf("A5C.pdf", width=WIDTH)
violin(data.frame(y=a5cp$p, x=a5cp$type), "Correctness")
dev.off()
cairo_pdf("A6C.pdf", width=WIDTH)
violin(data.frame(y=a6cp$p, x=a6cp$type), "Correctness")
dev.off()
cairo_pdf("ACA.pdf", width=WIDTH)
violin(data.frame(y=aacp$p, x=aacp$group), "Correctness")
dev.off()
report(aacp$p, "Correctness total")
report(getfirst(aacp$p), "Correctness G1")
report(getsecond(aacp$p), "Correctness G2")
print("Comparing Correctness zwischen Group 1 und 2:")
wilcox.test(x=getfirst(aacp$p), y=getsecond(aacp$p), alternative="two.sided", paired=FALSE, exact=TRUE)
rank_biserial(x=getfirst(aacp$p), y=getsecond(aacp$p), alternative="two.sided", paired=FALSE, exact=TRUE)
acp <- merge(x=data.frame(p=a1c+a2c+a3c, group), y=assignment2, by="group")
print("Comparing Correctness zwischen SEE/Dashboard")
seedashc <- merge(x=data.frame(p=c(getfirst(a1c + a2c + a3c) + getsecond(a4c + a5c + a6c), getsecond(a1c + a2c + a3c) + getfirst(a4c + a5c + a6c))/50, group), y=assignment1, by="group")
report(getfirst(seedashc$p), "Correctness Dashboard")
report(getsecond(seedashc$p), "Correctness SEE")
cairo_pdf("AC.pdf", width=WIDTH)
violin(data.frame(y=seedashc$p, x=seedashc$type), "Correctness")
dev.off()
wilcox.test(x=getfirst(seedashc$p), y=getsecond(seedashc$p), alternative="less", paired=FALSE, exact=TRUE)
# Now the time
# Protect against "cheating" by determining maximum
maxtime1 <- do.call(pmax, c(data.frame(df$done1_update, 
                    df$a1_av1b_filename_update, df$a1_av1b_av_update,
                    df$a1_av2b_filename_update, df$a1_av2b_av_update,
                    df$a1_av3b_filename_update, df$a1_av3b_av_update,
                    df$a1_av4b_filename_update, df$a1_av4b_av_update,
                    df$a1_av5b_filename_update, df$a1_av5b_av_update), list(na.rm=T)))
maxtime2 <- do.call(pmax, c(data.frame(df$done2_update, 
                    df$a2_sv1b_filename_update, df$a2_sv1b_sv_update,
                    df$a2_sv2b_filename_update, df$a2_sv2b_sv_update,
                    df$a2_sv3b_filename_update, df$a2_sv3b_sv_update,
                    df$a2_sv4b_filename_update, df$a2_sv4b_sv_update,
                    df$a2_sv5b_filename_update, df$a2_sv5b_sv_update), list(na.rm=T)))
maxtime3 <- do.call(pmax, c(data.frame(df$done3_update, 
                    df$a3_1_update,
                    df$a3_2_update,
                    df$a3_3_update,
                    df$a3_4_update,
                    df$a3_5_update), list(na.rm=T)))
maxtime4 <- do.call(pmax, c(data.frame(df$done4_update, 
                    df$a4_av1b_filename_update, df$a4_av1b_av_update,
                    df$a4_av2b_filename_update, df$a4_av2b_av_update,
                    df$a4_av3b_filename_update, df$a4_av3b_av_update,
                    df$a4_av4b_filename_update, df$a4_av4b_av_update,
                    df$a4_av5b_filename_update, df$a4_av5b_av_update), list(na.rm=T)))
maxtime5 <- do.call(pmax, c(data.frame(df$done5_update, 
                    df$a5_mv1b_filename_update, df$a5_mv1b_mv_update,
                    df$a5_mv2b_filename_update, df$a5_mv2b_mv_update,
                    df$a5_mv3b_filename_update, df$a5_mv3b_mv_update,
                    df$a5_mv4b_filename_update, df$a5_mv4b_mv_update,
                    df$a5_mv5b_filename_update, df$a5_mv5b_mv_update), list(na.rm=T)))
maxtime6 <- do.call(pmax, c(data.frame(df$done6_update, 
                    df$a6_1_update,
                    df$a6_2_update,
                    df$a6_3_update,
                    df$a6_4_update,
                    df$a6_5_update), list(na.rm=T)))
a1t <- signif(as.numeric(maxtime1 - df$ready1_update), digits=3)
a1ct <- merge(x=data.frame(p=a1t, group), y=assignment1, by="group")
a2t <- signif(as.numeric(maxtime2 - df$ready2_update), digits=3)
a2ct <- merge(x=data.frame(p=a2t, group), y=assignment1, by="group")
a3t <- signif(as.numeric(maxtime3 - df$ready3_update), digits=3)
a3ct <- merge(x=data.frame(p=a3t, group), y=assignment1, by="group")
a4t <- signif(as.numeric(maxtime4 - df$ready4_update), digits=3)
a4ct <- merge(x=data.frame(p=a4t, group), y=assignment2, by="group")
a5t <- signif(as.numeric(maxtime5 - df$ready5_update), digits=3)
a5ct <- merge(x=data.frame(p=a5t, group), y=assignment2, by="group")
a6t <- signif(as.numeric(maxtime6 - df$ready6_update), digits=3)
a6ct <- merge(x=data.frame(p=a6t, group), y=assignment2, by="group")
aact <- data.frame(p=a1t+a2t+a3t+a4t+a5t+a6t, group)
# Divide by 60 here, otherwise these are seconds
report(a1ct$p, "Time A1")
wilcox.test(y=getfirst(a1ct$p), x=getsecond(a1ct$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(y=getfirst(a1ct$p), x=getsecond(a1ct$p), alternative="less", paired=FALSE, exact=TRUE)
report(a2ct$p, "Time A2")
wilcox.test(y=getfirst(a2ct$p), x=getsecond(a2ct$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(y=getfirst(a2ct$p), x=getsecond(a2ct$p), alternative="less", paired=FALSE, exact=TRUE)
report(a3ct$p, "Time A3")
wilcox.test(y=getfirst(a3ct$p), x=getsecond(a3ct$p), alternative="less", paired=FALSE, exact=TRUE)
report(a4ct$p, "Time A4")
wilcox.test(x=getfirst(a4ct$p), y=getsecond(a4ct$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(x=getfirst(a4ct$p), y=getsecond(a4ct$p), alternative="less", paired=FALSE, exact=TRUE)
report(a5ct$p, "Time A5")
wilcox.test(x=getfirst(a5ct$p), y=getsecond(a5ct$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(x=getfirst(a5ct$p), y=getsecond(a5ct$p), alternative="less", paired=FALSE, exact=TRUE)
report(a6ct$p, "Time A6")
wilcox.test(x=getfirst(a6ct$p), y=getsecond(a6ct$p), alternative="less", paired=FALSE, exact=TRUE)
cairo_pdf("A1T.pdf", width=WIDTH)
violin(data.frame(y=a1ct$p, x=a1ct$type), "Time (minutes)")
dev.off()
cairo_pdf("A2T.pdf", width=WIDTH)
violin(data.frame(y=a2ct$p, x=a2ct$type), "Time (minutes)")
dev.off()
cairo_pdf("A3T.pdf", width=WIDTH)
violin(data.frame(y=a3ct$p, x=a3ct$type), "Time (minutes)")
dev.off()
cairo_pdf("A4T.pdf", width=WIDTH)
violin(data.frame(y=a4ct$p, x=a4ct$type), "Time (minutes)")
dev.off()
cairo_pdf("A5T.pdf", width=WIDTH)
violin(data.frame(y=a5ct$p, x=a5ct$type), "Time (minutes)")
dev.off()
cairo_pdf("A6T.pdf", width=WIDTH)
violin(data.frame(y=a6ct$p, x=a6ct$type), "Time (minutes)")
dev.off()
cairo_pdf("ATA.pdf", width=WIDTH)
violin(data.frame(y=aact$p, x=aact$group), "Time (minutes)")
dev.off()
report(aact$p, "Time total")
report(getfirst(aact$p), "Time G1")
report(getsecond(aact$p), "Time G2")
print("Comparing Correctness between SEE and Dashboard:")
seedasht <- merge(x=data.frame(p=c(getfirst(a1t + a2t + a3t) + getsecond(a4t + a5t + a6t), getsecond(a1t + a2t + a3t) + getfirst(a4t + a5t + a6t)), group), y=assignment1, by="group")
report(getfirst(seedasht$p), "Time Dashboard")
report(getsecond(seedasht$p), "Time SEE")
cairo_pdf("AT.pdf", width=WIDTH)
violin(data.frame(y=seedasht$p, x=seedasht$type), "Time (minutes)")
dev.off()
wilcox.test(x=getfirst(seedasht$p), y=getsecond(seedasht$p), alternative="less", paired=FALSE, exact=TRUE)
# SUS
dsus <- 2.5 * (20 + as.numeric(strtrim(df$dsus1, 1)) + as.numeric(strtrim(df$dsus3, 1)) + as.numeric(strtrim(df$dsus5, 1)) + as.numeric(strtrim(df$dsus7, 1)) + as.numeric(strtrim(df$dsus9, 1)) - as.numeric(strtrim(df$dsus2, 1)) - as.numeric(strtrim(df$dsus4, 1)) - as.numeric(strtrim(df$dsus6, 1)) - as.numeric(strtrim(df$dsus8, 1)) - as.numeric(strtrim(df$dsus10, 1)))
ssus <- 2.5 * (20 + as.numeric(strtrim(df$ssus1, 1)) + as.numeric(strtrim(df$ssus3, 1)) + as.numeric(strtrim(df$ssus5, 1)) + as.numeric(strtrim(df$ssus7, 1)) + as.numeric(strtrim(df$ssus9, 1)) - as.numeric(strtrim(df$ssus2, 1)) - as.numeric(strtrim(df$ssus4, 1)) - as.numeric(strtrim(df$ssus6, 1)) - as.numeric(strtrim(df$ssus8, 1)) - as.numeric(strtrim(df$ssus10, 1)))
dsus_u <- 3.125 * (10 + as.numeric(strtrim(df$dsus1, 1)) + as.numeric(strtrim(df$dsus3, 1)) + as.numeric(strtrim(df$dsus5, 1)) + as.numeric(strtrim(df$dsus7, 1)) + as.numeric(strtrim(df$dsus9, 1)) - as.numeric(strtrim(df$dsus2, 1)) - as.numeric(strtrim(df$dsus6, 1)) - as.numeric(strtrim(df$dsus8, 1)))
ssus_u <- 3.125 * (10 + as.numeric(strtrim(df$ssus1, 1)) + as.numeric(strtrim(df$ssus3, 1)) + as.numeric(strtrim(df$ssus5, 1)) + as.numeric(strtrim(df$ssus7, 1)) + as.numeric(strtrim(df$ssus9, 1)) - as.numeric(strtrim(df$ssus2, 1)) - as.numeric(strtrim(df$ssus6, 1)) - as.numeric(strtrim(df$ssus8, 1)))
ssus_l <- 12.5 * (10 - as.numeric(strtrim(df$ssus4, 1)) - as.numeric(strtrim(df$ssus10, 1)))
dsus_l <- 12.5 * (10 - as.numeric(strtrim(df$dsus4, 1)) - as.numeric(strtrim(df$dsus10, 1)))
cat("Dashboard SUS values: ", dsus, "\n")
cat("SEE SUS values: ", ssus, "\n")
report(dsus, "Dashboard")
report(ssus, "SEE")
report(dsus_u, "Dashboard-Usability")
report(ssus_u, "SEE-Usability")
report(dsus_l, "Dashboard-Learnability")
report(ssus_l, "SEE-Learnability")
sus <- data.frame(sus=c(dsus, ssus), type=c(rep(factor("Dashboard"), 20), rep(factor("SEE"), 20)))
cairo_pdf("SUS.pdf", width=WIDTH)
violin(data.frame(y=sus$sus, x=sus$type), "SUS-Score")
dev.off()
wilcox.test(ssus, dsus, alternative="greater", paired=FALSE)
wilcox.test(ssus, dsus, alternative="less", paired=FALSE)
rank_biserial(ssus, dsus, alternative="less", paired=FALSE)
wilcox.test(ssus_u, dsus_u, alternative="greater", paired=FALSE)
wilcox.test(ssus_u, dsus_u, alternative="less", paired=FALSE)
rank_biserial(ssus_u, dsus_u, alternative="less", paired=FALSE)
wilcox.test(ssus_l, dsus_l, alternative="greater", paired=FALSE)
wilcox.test(ssus_l, dsus_l, alternative="less", paired=FALSE)
rank_biserial(ssus_l, dsus_l, alternative="less", paired=FALSE)
# Finally, the ASQ, analyzing "easy" and "time"
a1qe <- as.numeric(factor(df$a1easy, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a1cqe <- merge(x=data.frame(p=a1qe, group), y=assignment1, by="group")
a2qe <- as.numeric(factor(df$a2easy, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a2cqe <- merge(x=data.frame(p=a2qe, group), y=assignment1, by="group")
a3qe <- as.numeric(factor(df$a3easy, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a3cqe <- merge(x=data.frame(p=a3qe, group), y=assignment1, by="group")
a4qe <- as.numeric(factor(df$a4easy, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a4cqe <- merge(x=data.frame(p=a4qe, group), y=assignment2, by="group")
a5qe <- as.numeric(factor(df$a5easy, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a5cqe <- merge(x=data.frame(p=a5qe, group), y=assignment2, by="group")
a6qe <- as.numeric(factor(df$a6easy, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a6cqe <- merge(x=data.frame(p=a6qe, group), y=assignment2, by="group")
aacqe <- data.frame(p=a1qe+a2qe+a3qe+a4qe+a5qe+a6qe, group)

a1qt <- as.numeric(factor(df$a1time, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a1cqt <- merge(x=data.frame(p=a1qt, group), y=assignment1, by="group")
a2qt <- as.numeric(factor(df$a2time, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a2cqt <- merge(x=data.frame(p=a2qt, group), y=assignment1, by="group")
a3qt <- as.numeric(factor(df$a3time, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a3cqt <- merge(x=data.frame(p=a3qt, group), y=assignment1, by="group")
a4qt <- as.numeric(factor(df$a4time, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a4cqt <- merge(x=data.frame(p=a4qt, group), y=assignment2, by="group")
a5qt <- as.numeric(factor(df$a5time, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a5cqt <- merge(x=data.frame(p=a5qt, group), y=assignment2, by="group")
a6qt <- as.numeric(factor(df$a6time, order = TRUE, levels = rev(c("1: Sehr schwer", "2", "3", "4", "5", "6", "7: Sehr leicht"))))
a6cqt <- merge(x=data.frame(p=a6qt, group), y=assignment2, by="group")
aacqt <- data.frame(p=a1qt+a2qt+a3qt+a4qt+a5qt+a6qt, group)
print("Comparison between ASQs")
wilcox.test(x=getfirst(a1cqe$p), y=getsecond(a1cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a1cqe$p), x=getsecond(a1cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a1cqt$p), y=getsecond(a1cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a1cqt$p), x=getsecond(a1cqt$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(y=getfirst(a1cqt$p), x=getsecond(a1cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a2cqe$p), y=getsecond(a2cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a2cqe$p), x=getsecond(a2cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a2cqt$p), y=getsecond(a2cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a2cqt$p), x=getsecond(a2cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a3cqe$p), y=getsecond(a3cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a3cqe$p), x=getsecond(a3cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a3cqt$p), y=getsecond(a3cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a3cqt$p), x=getsecond(a3cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a4cqe$p), y=getsecond(a4cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a4cqe$p), x=getsecond(a4cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a4cqt$p), y=getsecond(a4cqt$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(x=getfirst(a4cqt$p), y=getsecond(a4cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a4cqt$p), x=getsecond(a4cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a5cqe$p), y=getsecond(a5cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a5cqe$p), x=getsecond(a5cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a5cqt$p), y=getsecond(a5cqt$p), alternative="less", paired=FALSE, exact=TRUE)
rank_biserial(x=getfirst(a5cqt$p), y=getsecond(a5cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a5cqt$p), x=getsecond(a5cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a6cqe$p), y=getsecond(a6cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a6cqe$p), x=getsecond(a6cqe$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(x=getfirst(a6cqt$p), y=getsecond(a6cqt$p), alternative="less", paired=FALSE, exact=TRUE)
wilcox.test(y=getfirst(a6cqt$p), x=getsecond(a6cqt$p), alternative="less", paired=FALSE, exact=TRUE)

cairo_pdf("A1QE.pdf", width=WIDTH)
violin(data.frame(y=a1cqe$p, x=a1cqe$type), "Complexity")
dev.off()
cairo_pdf("A2QE.pdf", width=WIDTH)
violin(data.frame(y=a2cqe$p, x=a2cqe$type), "Complexity")
dev.off()
cairo_pdf("A3QE.pdf", width=WIDTH)
violin(data.frame(y=a3cqe$p, x=a3cqe$type), "Complexity")
dev.off()
cairo_pdf("A4QE.pdf", width=WIDTH)
violin(data.frame(y=a4cqe$p, x=a4cqe$type), "Complexity")
dev.off()
cairo_pdf("A5QE.pdf", width=WIDTH)
violin(data.frame(y=a5cqe$p, x=a5cqe$type), "Complexity")
dev.off()
cairo_pdf("A6QE.pdf", width=WIDTH)
violin(data.frame(y=a6cqe$p, x=a6cqe$type), "Complexity")
dev.off()
cairo_pdf("AAQE.pdf", width=WIDTH)
violin(data.frame(y=aacqe$p, x=aacqe$group), "Complexity")
dev.off()

cairo_pdf("A1QT.pdf", width=WIDTH)
violin(data.frame(y=a1cqt$p, x=a1cqt$type), "Effort")
dev.off()
cairo_pdf("A2QT.pdf", width=WIDTH)
violin(data.frame(y=a2cqt$p, x=a2cqt$type), "Effort")
dev.off()
cairo_pdf("A3QT.pdf", width=WIDTH)
violin(data.frame(y=a3cqt$p, x=a3cqt$type), "Effort")
dev.off()
cairo_pdf("A4QT.pdf", width=WIDTH)
violin(data.frame(y=a4cqt$p, x=a4cqt$type), "Effort")
dev.off()
cairo_pdf("A5QT.pdf", width=WIDTH)
violin(data.frame(y=a5cqt$p, x=a5cqt$type), "Effort")
dev.off()
cairo_pdf("A6QT.pdf", width=WIDTH)
violin(data.frame(y=a6cqt$p, x=a6cqt$type), "Effort")
dev.off()
cairo_pdf("AAQT.pdf", width=WIDTH)
violin(data.frame(y=aacqt$p, x=aacqt$group), "Effort")
dev.off()
report(aacqe$p, "Complexity total")
report(getfirst(aacqe$p), "Complexity G1")
report(getsecond(aacqe$p), "Complexity G2")
report(aacqt$p, "Effort total")
report(getfirst(aacqt$p), "Effort G1")
report(getsecond(aacqt$p), "Effort G2")
starthour <- data.frame(y=c( 09, 15, 01, 10, 21, 18, 09, 11, 14, 18, 20, 08, 11, 18, 20, 09, 11, 17, 15, 12), x=group)
cairo_pdf("StartHour.pdf", width=WIDTH)
violin(starthour, "Time of day (hour)")
dev.off()
repcor = function(data) {
    cat("$\\glssymbol{tau} = ", round(data$estimate, digits=2), "$ ($p \\approx ", round(data$p.value, digits=2), "$)\n")
}
# Correlation to Correctness
acp_dash <- c(getfirst(a1cp$p + a2cp$p + a3cp$p), getsecond(a4cp$p + a5cp$p + a6cp$p))
acp_see <- c(getsecond(a1cp$p + a2cp$p + a3cp$p), getfirst(a4cp$p + a5cp$p + a6cp$p))
print("Correlation between SEE experience and SEE-Correctness")
repcor(cor.test(as.numeric(knowsee), acp_see, method="kendall", exact=F))
print("Correlation between SEE experience and Dashboard-Correctness")
repcor(cor.test(as.numeric(knowsee), acp_dash, method="kendall", exact=F))
print("Correlation between Dashboard experience and SEE-Correctness")
repcor(cor.test(as.numeric(knowdash), acp_see, method="kendall", exact=F))
print("Correlation between Dashboard experience and Dashboard-Correctness")
repcor(cor.test(as.numeric(knowdash), acp_dash, method="kendall", exact=F))
print("Correlation between Degree and SEE-Correctness")
repcor(cor.test(as.numeric(abschluss), acp_see, method="kendall", exact=F))
print("Correlation between Degree and Dashboard-Correctness")
repcor(cor.test(as.numeric(abschluss), acp_dash, method="kendall", exact=F))
print("Correlation between OpenSource and SEE-Correctness")
repcor(cor.test(as.numeric(opensource), acp_see, method="kendall", exact=F))
print("Correlation between OpenSource and Dashboard-Correctness")
repcor(cor.test(as.numeric(opensource), acp_dash, method="kendall", exact=F))
print("Correlation between Age and SEE-Correctness")
repcor(cor.test(as.numeric(age), acp_see, method="pearson", exact=F))
print("Correlation between Age and Dashboard-Correctness")
repcor(cor.test(as.numeric(age), acp_dash, method="pearson", exact=F))
# Correlation to task completion time
act_dash <- c(getfirst(a1ct$p + a2ct$p + a3ct$p), getsecond(a4ct$p + a5ct$p + a6ct$p))
act_see <- c(getsecond(a1ct$p + a2ct$p + a3ct$p), getfirst(a4ct$p + a5ct$p + a6ct$p))
print("Correlation between SEE experience and SEE-Time")
repcor(cor.test(as.numeric(knowsee), act_see, method="kendall", exact=F))
print("Correlation between SEE experience and Dashboard-Time")
repcor(cor.test(as.numeric(knowsee), act_dash, method="kendall", exact=F))
print("Correlation between Dashboard experience and SEE-Time")
repcor(cor.test(as.numeric(knowdash), act_see, method="kendall", exact=F))
print("Correlation between Dashboard experience and Dashboard-Time")
repcor(cor.test(as.numeric(knowdash), act_dash, method="kendall", exact=F))
print("Correlation between Degree and SEE-Time")
repcor(cor.test(as.numeric(abschluss), act_see, method="kendall", exact=F))
print("Correlation between Degree and Dashboard-Time")
repcor(cor.test(as.numeric(abschluss), act_dash, method="kendall", exact=F))
print("Correlation between OpenSource and SEE-Time")
repcor(cor.test(as.numeric(opensource), act_see, method="kendall", exact=F))
print("Correlation between OpenSource and Dashboard-Time")
repcor(cor.test(as.numeric(opensource), act_dash, method="kendall", exact=F))
print("Correlation between Age and SEE-Time")
repcor(cor.test(as.numeric(age), act_see, method="pearson", exact=F))
print("Correlation between Age and Dashboard-Time")
repcor(cor.test(as.numeric(age), act_dash, method="pearson", exact=F))
# Correlation to SUS
print("Correlation between SEE experience and SEE-SUS")
repcor(cor.test(as.numeric(knowsee), ssus, method="kendall", exact=F))
print("Correlation between SEE experience and Dashboard-SUS")
repcor(cor.test(as.numeric(knowsee), dsus, method="kendall", exact=F))
print("Correlation between Dashboard experience and SEE-SUS")
repcor(cor.test(as.numeric(knowdash), ssus, method="kendall", exact=F))
print("Correlation between Dashboard experience and Dashboard-SUS")
repcor(cor.test(as.numeric(knowdash), dsus, method="kendall", exact=F))
print("Correlation between Degree and SEE-SUS")
repcor(cor.test(as.numeric(abschluss), ssus, method="kendall", exact=F))
print("Correlation between Degree and Dashboard-SUS")
repcor(cor.test(as.numeric(abschluss), dsus, method="kendall", exact=F))
print("Correlation between OpenSource and SEE-SUS")
repcor(cor.test(as.numeric(opensource), ssus, method="kendall", exact=F))
print("Correlation between OpenSource and Dashboard-SUS")
repcor(cor.test(as.numeric(opensource), dsus, method="kendall", exact=F))
print("Correlation between Age and SEE-SUS")
repcor(cor.test(as.numeric(age), ssus, method="pearson", exact=F))
print("Correlation between Age and Dashboard-SUS")
repcor(cor.test(as.numeric(age), dsus, method="pearson", exact=F))
# Correlation to ASQ-T (Effort)
asqt_dash <- c(getfirst(a1cqt$p + a2cqt$p + a3cqt$p), getsecond(a4cqt$p + a5cqt$p + a6cqt$p))
asqt_see <- c(getsecond(a1cqt$p + a2cqt$p + a3cqt$p), getfirst(a4cqt$p + a5cqt$p + a6cqt$p))
print("Correlation between SEE experience and SEE-ASQ-T")
repcor(cor.test(as.numeric(knowsee), asqt_see, method="kendall", exact=F))
print("Correlation between SEE experience and Dashboard-ASQ-T")
repcor(cor.test(as.numeric(knowsee), asqt_dash, method="kendall", exact=F))
print("Correlation between Dashboard experience and SEE-ASQ-T")
repcor(cor.test(as.numeric(knowdash), asqt_see, method="kendall", exact=F))
print("Correlation between Dashboard experience and Dashboard-ASQ-T")
repcor(cor.test(as.numeric(knowdash), asqt_dash, method="kendall", exact=F))
print("Correlation between Degree and SEE-ASQ-T")
repcor(cor.test(as.numeric(abschluss), asqt_see, method="kendall", exact=F))
print("Correlation between Degree and Dashboard-ASQ-T")
repcor(cor.test(as.numeric(abschluss), asqt_dash, method="kendall", exact=F))
print("Correlation between OpenSource and SEE-ASQ-T")
repcor(cor.test(as.numeric(opensource), asqt_see, method="kendall", exact=F))
print("Correlation between OpenSource and Dashboard-ASQ-T")
repcor(cor.test(as.numeric(opensource), asqt_dash, method="kendall", exact=F))
print("Correlation between Age and SEE-ASQ-T")
repcor(cor.test(as.numeric(age), asqt_see, method="pearson", exact=F))
print("Correlation between Age and Dashboard-ASQ-T")
repcor(cor.test(as.numeric(age), asqt_dash, method="pearson", exact=F))
# Correlation to ASQ-E (complexity)
asqe_dash <- c(getfirst(a1cqe$p + a2cqe$p + a3cqe$p), getsecond(a4cqe$p + a5cqe$p + a6cqe$p))
asqe_see <- c(getsecond(a1cqe$p + a2cqe$p + a3cqe$p), getfirst(a4cqe$p + a5cqe$p + a6cqe$p))
print("Correlation between SEE experience and SEE-ASQ-E")
repcor(cor.test(as.numeric(knowsee), asqe_see, method="kendall", exact=F))
print("Correlation between SEE experience and Dashboard-ASQ-E")
repcor(cor.test(as.numeric(knowsee), asqe_dash, method="kendall", exact=F))
print("Correlation between Dashboard experience and SEE-ASQ-E")
repcor(cor.test(as.numeric(knowdash), asqe_see, method="kendall", exact=F))
print("Correlation between Dashboard experience and Dashboard-ASQ-E")
repcor(cor.test(as.numeric(knowdash), asqe_dash, method="kendall", exact=F))
print("Correlation between Degree and SEE-ASQ-E")
repcor(cor.test(as.numeric(abschluss), asqe_see, method="kendall", exact=F))
print("Correlation between Degree and Dashboard-ASQ-E")
repcor(cor.test(as.numeric(abschluss), asqe_dash, method="kendall", exact=F))
print("Correlation between OpenSource and SEE-ASQ-E")
repcor(cor.test(as.numeric(opensource), asqe_see, method="kendall", exact=F))
print("Correlation between OpenSource and Dashboard-ASQ-E")
repcor(cor.test(as.numeric(opensource), asqe_dash, method="kendall", exact=F))
print("Correlation between Age and SEE-ASQ-E")
repcor(cor.test(as.numeric(age), asqe_see, method="pearson", exact=F))
print("Correlation between Age and Dashboard-ASQ-E")
repcor(cor.test(as.numeric(age), asqe_dash, method="pearson", exact=F))
print("Correlation between Age and Dashboard experience")
repcor(cor.test(as.numeric(knowdash), age, method="kendall", exact=F))