.Rhistory

knitr::opts_chunk$set(warning = FALSE, message = FALSE)
# Data Manipulation and Analysis
library(tidyverse)   # Data manipulation and visualization
library(dplyr)       # Data manipulation
library(broom)       # Statistical analysis utilities
library(reshape2)    # Data reshaping
library(glmnet)      # Regularized regression models
library(readr)       # Data import
# Graphics
library(ggplot2)     # Data visualization
library(ggridges)    # Ridgeline plots
library(ggthemes)    # Additional ggplot2 themes
library(ggrepel)     # Label placement in ggplot2
library(ppcor)       # Partial and semi-partial correlation
library(ggExtra)     # Additional ggplot2 functions
library(ggsci)       # Scientific journal-inspired color palettes
library(viridis)     # Color palettes
library(scales)      # Scale functions
library(patchwork)   # Multi-plot alignment
library(ggcorrplot)  # Correlation plot
library(gapminder)   # Dataset used to make the box plot connected by lines
library(RColorBrewer)# Color palettes
library(plotly)      # Interactive graphs
library(lubridate)   # Date functions
library(stringr)     # Text functions
library(extrafont)   # Font customization on graphs
library(htmlwidgets) # Interactive exports
library(cowplot)     # Plot arrangement
library(png)         # PNG image format support
library(apyramid)    # Create age pyramids
# Themes: "default", "bootstrap", "cerulean", "cosmo", "darkly", "flatly", "journal", "lumen", "paper", "readable", "sandstone", "simplex", "spacelab", "united", "yeti"
# Tables
library(kableExtra)  # Table formatting
library(xtable)      # Table formatting
library(reactable)   # Interactive tables
library(htmltools)   # HTML table formatting
# Date functions
library(anytime)     # Date conversion
library(lubridate)   # Date manipulation
# SlimStampen
library(SlimStampeRData)  # SlimStampen-specific functions
# Dashboard
library(flexdashboard)  # Dashboard creation
library(lme4)
library(sjPlot)
#prior_knowledge_analysis <- read_csv("prior_knowledge.csv", show_col_types = FALSE)
#prior_knowledge_analysis$userId <- as.factor(prior_knowledge_analysis$userId)
prior_knowledge_alldata <- read_csv("prior_knowledge_all.csv", show_col_types = FALSE)
ggplot(prior_knowledge_alldata, aes(x=answer, y=alpha, col=prior_knowledge))+
#geom_boxplot()+
geom_point(size=3, position=position_jitter(0.2)) +
#geom_line(aes(group=prior_knowledge), size=0.1, position=position_dodge(0)) +
xlab("Fact") +
ylab("SoF") +
scale_color_manual(values=c("darkblue", "lightgreen")) +
scale_fill_manual(values=c("darkblue", "lightgreen")) +
ggtitle(paste("Prior Knowledge and SoF across Facts")) +
labs(col="prior_knowledge") +
theme_hc() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
ggsave("figures/Prior_Knowledge_Across_Facts.png")
#prior_knowledge_analysis <- prior_knowledge_analysis %>%
#  left_join(cleandata_avg %>% select(userId, MeanAlpha))
prior_knowledge_analysis1 <- read_csv("prior_knowledge_analysis1.csv", show_col_types = FALSE)
ggplot(prior_knowledge_analysis1, aes(y = alpha, x = as.factor(userId), col = prior_knowledge, fill = prior_knowledge)) +
stat_summary(geom="point", fun.data = "mean_sdl", size=2.5) +
stat_summary(geom="ribbon", fun.data = "mean_se",
aes(group = prior_knowledge),
col = NA,
alpha = 0.25) +
stat_summary(geom="line", fun = "mean", aes(group = prior_knowledge)) +
xlab("Participant ID") +
scale_color_manual(values=c("darkblue", "lightgreen", "orange")) +
scale_fill_manual(values=c("darkblue", "lightgreen", "orange")) +
scale_x_discrete() +
ylab(expression(paste(italic("SoF")))) +
ylim(0.1, 0.5) +
ggtitle(expression(paste ("Effect of Prior Knowledge Across Participants"))) +
labs(col="Prior Knowledge", fill="Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
ggplot(prior_knowledge_alldata, aes(y = alpha, x = userId, col = prior_knowledge, fill = prior_knowledge)) +
stat_summary(geom="point", fun.data = "mean_sdl", size=2.5) +
stat_summary(geom="ribbon", fun.data = "mean_se",
aes(group = prior_knowledge),
col = NA,
alpha = 0.25) +
stat_summary(geom="line", fun = "mean", aes(group = prior_knowledge)) +
xlab("Participant ID") +
scale_color_manual(values=c("darkblue", "lightgreen")) +
scale_fill_manual(values=c("darkblue", "lightgreen")) +
scale_x_discrete() +
ylab(expression(paste(italic("SoF")))) +
ylim(0.1, 0.5) +
ggtitle(expression(paste ("Effect of Prior Knowledge Across Participants"))) +
labs(col="Prior Knowledge", fill="Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
#ggsave("figures/Prior_Knowledge_Across_Participants.png")
ggplot(prior_knowledge_alldata, aes(y = alpha, x = userId, col = prior_knowledge, fill = prior_knowledge)) +
stat_summary(geom="point", fun.data = "mean_se", size=5,
position = position_dodge()) +
stat_summary(geom="errorbar", size=0.8, color = "black", fun.data = "mean_se", aes(group = prior_knowledge)) +
xlab("Participant ID") +
scale_color_manual(values=c("darkblue", "lightgreen")) +
scale_fill_manual(values=c("darkblue", "lightgreen")) +
scale_x_discrete() +
ylab(expression(paste(italic("SoF")))) +
ylim(0, 0.5) +
ggtitle(expression(paste ("Effect of Prior Knowledge Across Participants"))) +
labs(col="Prior Knowledge", fill="Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
prior_knowledge_alldata$prior_knowledge <- factor(prior_knowledge_alldata$prior_knowledge,
levels = c("no", "yes"))
#Plot with added error bars
plot_with_stacked_bar_and_error_bars <- ggplot(prior_knowledge_alldata, aes(x = factor(userId), y = alpha, fill = prior_knowledge)) +
geom_bar(stat = "summary", fun = "mean", position = "dodge", alpha = 0.7) +
geom_errorbar(stat = "summary", fun.data = "mean_se", position = position_dodge(width = 0.75), width = 0.25, size = 0.7) +
xlab("Participant ID") +
ylab(expression(paste(italic("SoF")))) +
ylim(0.0, 0.7) +
ggtitle(expression(paste("Effect of Prior Knowledge Across Participants"))) +
scale_fill_manual(values = c("darkblue", "lightgreen")) +
labs(fill = "Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text(angle = 90),
panel.grid.major = element_blank())
print(plot_with_stacked_bar_and_error_bars)
ggsave("figures/Prior_Knowledge_Across_Participants_bargraph.png")
prior_agg <- prior_knowledge_analysis %>%
group_by(userId, prior_knowledge) %>%
summarize(alpha = mean(alpha)) %>%
pivot_wider(names_from = prior_knowledge, values_from = alpha)
prior_agg <- prior_knowledge_alldata %>%
group_by(userId, prior_knowledge) %>%
summarize(alpha = mean(alpha)) %>%
pivot_wider(names_from = prior_knowledge, values_from = alpha)
ggplot(prior_agg, aes(x=no, y=yes)) +
geom_smooth(method  ="lm") +
geom_point(size=4, col="blue", alpha=0.5) +
geom_text_repel(aes(label=userId)) +
xlab("SOF: Previously Unknown") +
ylab("SOF: Previously Known") +
ggtitle("SOF by Prior Knwledge") +
theme_minimal()
ggsave("figures/PriorKnowledgeCorrelations.png")
base <- lmer(alpha ~ prior_knowledge  + (1|userId) +(0+prior_knowledge|userId),
prior_knowledge_analysis)
base <- lmer(alpha ~ prior_knowledge  + (1|userId) +(0+prior_knowledge|userId),
prior_knowledge_alldata)
tab_model(base)
#write.csv(cleandata_lastRep, "prior_knowledge_all.csv", row.names = FALSE)
#cleandata_lastRep <- cleandata_lastRep %>%
#  mutate(BLC = -log(cleandata_firstRep$reactionTime - .300))
#new_dataframe <- cleandata_avg %>%
#  select(MeanAlpha, userId, lessonTitle) %>%
#  mutate(prior_knowledge = "all")%>%
#  mutate(factId = 11111)%>%
#  mutate(answer = "ans")%>%
#  rename(alpha = MeanAlpha)
scatter_plot <- ggplot(prior_knowledge_alldata, aes(x = BLC, y = alpha)) +
geom_point(data = filter(prior_knowledge_alldata, prior_knowledge == "yes"),
color = "blue") +
geom_smooth(method  ="lm") +
labs(title = "Scatterplot of BLC vs. Alpha",
x = "BLC",
y = "Alpha")
correlation_result <- cor.test(prior_knowledge_alldata$BLC,
prior_knowledge_alldata$alpha)
# Add correlation coefficient to the plot
scatter_plot <- scatter_plot +
annotate("text", x = Inf, y = -Inf,
label = paste("r =", round(correlation_result$estimate, 2)),
hjust = 1, vjust = 0)
# Display the scatterplot with correlation
print(scatter_plot)
recovered_blc <- read_csv("example.csv", show_col_types = FALSE)
# Create a scatterplot
scatterplot <- ggplot(recovered_blc, aes(x = BLC, y = prior_knowledge)) +
geom_point(color = "darkblue") +
geom_smooth(method = "lm", color = "darkblue") +
labs(title = "Scatterplot of Recovered BLC vs. Prior Knowledge",
x = "BLC",
y = "Prior Knowledge")
# Calculate correlation and p-value
correlationResult <- cor.test(recovered_blc$BLC, recovered_blc$prior_knowledge)
# Add correlation coefficient and p-value to the plot
scatterplot <- scatterplot +
annotate("text", x = Inf, y = -Inf,
label = paste("r =", round(correlationResult$estimate, 2),
"\n",
"p < .0001"),
hjust = 1, vjust = -0.5)
# Display the scatterplot with correlation
print(scatterplot)
p.value <- format(correlationResult$p.value, digits = 2)
print(p.value)
ggsave("figures/recovered_blc_correlations.png")
knitr::opts_chunk$set(warning = FALSE, message = FALSE)
# Data Manipulation and Analysis
library(tidyverse)   # Data manipulation and visualization
library(dplyr)       # Data manipulation
library(broom)       # Statistical analysis utilities
library(reshape2)    # Data reshaping
library(glmnet)      # Regularized regression models
library(readr)       # Data import
# Graphics
library(ggplot2)     # Data visualization
library(ggridges)    # Ridgeline plots
library(ggthemes)    # Additional ggplot2 themes
library(ggrepel)     # Label placement in ggplot2
library(ppcor)       # Partial and semi-partial correlation
library(ggExtra)     # Additional ggplot2 functions
library(ggsci)       # Scientific journal-inspired color palettes
library(viridis)     # Color palettes
library(scales)      # Scale functions
library(patchwork)   # Multi-plot alignment
library(ggcorrplot)  # Correlation plot
library(gapminder)   # Dataset used to make the box plot connected by lines
library(RColorBrewer)# Color palettes
library(plotly)      # Interactive graphs
library(lubridate)   # Date functions
library(stringr)     # Text functions
library(extrafont)   # Font customization on graphs
library(htmlwidgets) # Interactive exports
library(cowplot)     # Plot arrangement
library(png)         # PNG image format support
library(apyramid)    # Create age pyramids
# Themes: "default", "bootstrap", "cerulean", "cosmo", "darkly", "flatly", "journal", "lumen", "paper", "readable", "sandstone", "simplex", "spacelab", "united", "yeti"
# Tables
library(kableExtra)  # Table formatting
library(xtable)      # Table formatting
library(reactable)   # Interactive tables
library(htmltools)   # HTML table formatting
# Date functions
library(anytime)     # Date conversion
library(lubridate)   # Date manipulation
# SlimStampen
library(SlimStampeRData)  # SlimStampen-specific functions
# Dashboard
library(flexdashboard)  # Dashboard creation
library(lme4)
library(sjPlot)
#prior_knowledge_analysis <- read_csv("prior_knowledge.csv", show_col_types = FALSE)
#prior_knowledge_analysis$userId <- as.factor(prior_knowledge_analysis$userId)
prior_knowledge_alldata <- read_csv("prior_knowledge_all.csv", show_col_types = FALSE)
ggplot(prior_knowledge_alldata, aes(x=answer, y=alpha, col=prior_knowledge))+
#geom_boxplot()+
geom_point(size=3, position=position_jitter(0.2)) +
#geom_line(aes(group=prior_knowledge), size=0.1, position=position_dodge(0)) +
xlab("Fact") +
ylab("SoF") +
scale_color_manual(values=c("darkblue", "lightgreen")) +
scale_fill_manual(values=c("darkblue", "lightgreen")) +
ggtitle(paste("Prior Knowledge and SoF across Facts")) +
labs(col="prior_knowledge") +
theme_hc() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
ggsave("figures/Prior_Knowledge_Across_Facts.png")
#prior_knowledge_analysis <- prior_knowledge_analysis %>%
#  left_join(cleandata_avg %>% select(userId, MeanAlpha))
prior_knowledge_analysis1 <- read_csv("prior_knowledge_analysis1.csv", show_col_types = FALSE)
ggplot(prior_knowledge_analysis1, aes(y = alpha, x = as.factor(userId), col = prior_knowledge, fill = prior_knowledge)) +
stat_summary(geom="point", fun.data = "mean_sdl", size=2.5) +
stat_summary(geom="ribbon", fun.data = "mean_se",
aes(group = prior_knowledge),
col = NA,
alpha = 0.25) +
stat_summary(geom="line", fun = "mean", aes(group = prior_knowledge)) +
xlab("Participant ID") +
scale_color_manual(values=c("darkblue", "lightgreen", "orange")) +
scale_fill_manual(values=c("darkblue", "lightgreen", "orange")) +
scale_x_discrete() +
ylab(expression(paste(italic("SoF")))) +
ylim(0.1, 0.5) +
ggtitle(expression(paste ("Effect of Prior Knowledge Across Participants"))) +
labs(col="Prior Knowledge", fill="Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
ggplot(prior_knowledge_alldata, aes(y = alpha, x = userId, col = prior_knowledge, fill = prior_knowledge)) +
stat_summary(geom="point", fun.data = "mean_sdl", size=2.5) +
stat_summary(geom="ribbon", fun.data = "mean_se",
aes(group = prior_knowledge),
col = NA,
alpha = 0.25) +
stat_summary(geom="line", fun = "mean", aes(group = prior_knowledge)) +
xlab("Participant ID") +
scale_color_manual(values=c("darkblue", "lightgreen")) +
scale_fill_manual(values=c("darkblue", "lightgreen")) +
scale_x_discrete() +
ylab(expression(paste(italic("SoF")))) +
ylim(0.1, 0.5) +
ggtitle(expression(paste ("Effect of Prior Knowledge Across Participants"))) +
labs(col="Prior Knowledge", fill="Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
#ggsave("figures/Prior_Knowledge_Across_Participants.png")
ggplot(prior_knowledge_alldata, aes(y = alpha, x = userId, col = prior_knowledge, fill = prior_knowledge)) +
stat_summary(geom="point", fun.data = "mean_se", size=5,
position = position_dodge()) +
stat_summary(geom="errorbar", size=0.8, color = "black", fun.data = "mean_se", aes(group = prior_knowledge)) +
xlab("Participant ID") +
scale_color_manual(values=c("darkblue", "lightgreen")) +
scale_fill_manual(values=c("darkblue", "lightgreen")) +
scale_x_discrete() +
ylab(expression(paste(italic("SoF")))) +
ylim(0, 0.5) +
ggtitle(expression(paste ("Effect of Prior Knowledge Across Participants"))) +
labs(col="Prior Knowledge", fill="Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text (angle=90),
panel.grid.major = element_blank())
prior_knowledge_alldata$prior_knowledge <- factor(prior_knowledge_alldata$prior_knowledge,
levels = c("no", "yes"))
#Plot with added error bars
plot_with_stacked_bar_and_error_bars <- ggplot(prior_knowledge_alldata, aes(x = factor(userId), y = alpha, fill = prior_knowledge)) +
geom_bar(stat = "summary", fun = "mean", position = "dodge", alpha = 0.7) +
geom_errorbar(stat = "summary", fun.data = "mean_se", position = position_dodge(width = 0.75), width = 0.25, size = 0.7) +
xlab("Participant ID") +
ylab(expression(paste(italic("SoF")))) +
ylim(0.0, 0.7) +
ggtitle(expression(paste("Effect of Prior Knowledge Across Participants"))) +
scale_fill_manual(values = c("darkblue", "lightgreen")) +
labs(fill = "Prior Knowledge") +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text(angle = 90),
panel.grid.major = element_blank())
print(plot_with_stacked_bar_and_error_bars)
ggsave("figures/Prior_Knowledge_Across_Participants_bargraph.png")
prior_agg <- prior_knowledge_alldata %>%
group_by(userId, prior_knowledge) %>%
summarize(alpha = mean(alpha)) %>%
pivot_wider(names_from = prior_knowledge, values_from = alpha)
ggplot(prior_agg, aes(x=no, y=yes)) +
geom_smooth(method  ="lm") +
geom_point(size=4, col="blue", alpha=0.5) +
geom_text_repel(aes(label=userId)) +
xlab("SOF: Previously Unknown") +
ylab("SOF: Previously Known") +
ggtitle("SOF by Prior Knwledge") +
theme_minimal()
ggsave("figures/PriorKnowledgeCorrelations.png")
base <- lmer(alpha ~ prior_knowledge  + (1|userId) +(0+prior_knowledge|userId),
prior_knowledge_alldata)
tab_model(base)
#write.csv(cleandata_lastRep, "prior_knowledge_all.csv", row.names = FALSE)
#cleandata_lastRep <- cleandata_lastRep %>%
#  mutate(BLC = -log(cleandata_firstRep$reactionTime - .300))
#new_dataframe <- cleandata_avg %>%
#  select(MeanAlpha, userId, lessonTitle) %>%
#  mutate(prior_knowledge = "all")%>%
#  mutate(factId = 11111)%>%
#  mutate(answer = "ans")%>%
#  rename(alpha = MeanAlpha)
scatter_plot <- ggplot(prior_knowledge_alldata, aes(x = BLC, y = alpha)) +
geom_point(data = filter(prior_knowledge_alldata, prior_knowledge == "yes"),
color = "blue") +
geom_smooth(method  ="lm") +
labs(title = "Scatterplot of BLC vs. Alpha",
x = "BLC",
y = "Alpha")
correlation_result <- cor.test(prior_knowledge_alldata$BLC,
prior_knowledge_alldata$alpha)
# Add correlation coefficient to the plot
scatter_plot <- scatter_plot +
annotate("text", x = Inf, y = -Inf,
label = paste("r =", round(correlation_result$estimate, 2)),
hjust = 1, vjust = 0)
# Display the scatterplot with correlation
print(scatter_plot)
recovered_blc <- read_csv("example.csv", show_col_types = FALSE)
# Create a scatterplot
scatterplot <- ggplot(recovered_blc, aes(x = BLC, y = prior_knowledge)) +
geom_point(color = "darkblue") +
geom_smooth(method = "lm", color = "darkblue") +
labs(title = "Scatterplot of Recovered BLC vs. Prior Knowledge",
x = "BLC",
y = "Prior Knowledge")
# Calculate correlation and p-value
correlationResult <- cor.test(recovered_blc$BLC, recovered_blc$prior_knowledge)
# Add correlation coefficient and p-value to the plot
scatterplot <- scatterplot +
annotate("text", x = Inf, y = -Inf,
label = paste("r =", round(correlationResult$estimate, 2),
"\n",
"p < .0001"),
hjust = 1, vjust = -0.5)
# Display the scatterplot with correlation
print(scatterplot)
p.value <- format(correlationResult$p.value, digits = 2)
print(p.value)
ggsave("figures/recovered_blc_correlations.png")
user_id_to_keep <- 83678
recovered_data_83678 <- recovered_blc[recovered_blc$Subject == user_id_to_keep, ]
ScatterPlot <- ggplot(recovered_data_83678, aes(x = Fact, y = BLC, color = factor(prior_knowledge))) +
geom_point(size = 3) +  # Adjust the size of the dots
labs(title = "Participant 83678 Recovered BLC and Prior Knowledge",
x = "Fact",
y = "BLC") +
scale_color_manual(values = c("darkblue", "lightgreen")) +
theme_minimal() +
theme(legend.position = "right",
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1)) +  # Rotate x-axis labels
ylim(-2, 1)  # Set y-axis limits from -2 to +1
print(ScatterPlot)