medpolish-survey.Rmd

---
title: "medpolish-survey"
author: "John D. Smith"
date: '2022-03-21'
output: md_document
---

## Libraries

```{r setup, include=FALSE}
library(tidyverse)
library(grid)
library(ggplotify)
library(cowplot)
library(mpx)
theme_set(theme_light())
```


```{r}
survey_responses <- read_csv(
"generation,n,response
1970s,32,Strongly Disagree
1970s,32,Disagree
1970s,111,Neutral
1970s,64,Agree
1970s,17,Strongly Agree
1980s,23,Strongly Disagree
1980s,19,Disagree
1980s,83,Neutral
1980s,54,Agree
1980s,11,Strongly Agree
1990s,16,Strongly Disagree
1990s,31,Disagree
1990s,104,Neutral
1990s,52,Agree
1990s,12,Strongly Agree
2000-2007,18,Strongly Disagree
2000-2007,20,Disagree
2000-2007,95,Neutral
2000-2007,61,Agree
2000-2007,21,Strongly Agree
2008-2012,16,Strongly Disagree
2008-2012,13,Disagree
2008-2012,90,Neutral
2008-2012,43,Agree
2008-2012,10,Strongly Agree
2013-2017,13,Strongly Disagree
2013-2017,18,Disagree
2013-2017,94,Neutral
2013-2017,44,Agree
2013-2017,11,Strongly Agree
Since 2018,1,Strongly Disagree
Since 2018,4,Disagree
Since 2018,40,Neutral
Since 2018,9,Agree
Since 2018,6,Strongly Agree") %>% 
  group_by(generation) %>% 
  mutate(percent = 100 * (n / sum(n))) %>% 
  ungroup()

```

pivot_wider does the trick...

```{r}
get_array <- function(df, measure, row_name, col_name) {
  my_df <- df
  my_measure <- my_df %>% select({{measure}}) %>% unlist()
  my_row_name <- my_df %>% select({{row_name}}) %>% unique() %>% unlist()
  my_col_name <- my_df %>% select({{col_name}}) %>% unique() %>% unlist()
  my_row_length <- my_row_name %>% unlist() %>% length()
  my_col_length <- my_col_name %>% unlist() %>% length()
  my_array <- array(my_measure, c(my_row_length, my_col_length),
                    dimnames = list(my_row_name, my_col_name))
}
x <- get_array(survey_responses, n, generation, response)
x
str(x)
```

var <- rlang::enquo(var)
data %>%
}
1 summarise(mean = mean(!!var))

```{r}
counts <- array(survey_responses$n, c(7,5), 
                dimnames =  list(unique(survey_responses$generation), 
                                 unique(survey_responses$response)))

percents <- array(survey_responses$percent, c(7,5), 
                dimnames =  list(unique(survey_responses$generation), 
                                 unique(survey_responses$response)))

```

## Techniques described by Hadley Wickham

Hadley Wickham, R for Data Science chapter on Exploratory Data Analysis: <https://r4ds.had.co.nz/exploratory-data-analysis.html> 

### look at COUNTS

#### Dot size

```{r}
survey_responses %>% ggplot(aes(generation, y = response)) +
  geom_point(aes(size = n)) +
  coord_flip()
```

#### Heat map

```{r}
survey_responses %>% ggplot(aes(generation, y = response)) +
  geom_tile(mapping = aes(fill = n)) +
  coord_flip()

```

### Look at PERCENTS 

#### Dot Size

```{r}
survey_responses %>% ggplot(aes(generation, y = response)) +
  geom_point(aes(size = percent)) +
  coord_flip()
```

#### Heatmap

```{r}
survey_responses %>% ggplot(aes(generation, y = response)) +
  geom_tile(mapping = aes(fill = percent)) +
  coord_flip()

```

## Save medpolish output in a list and demonstrate basic additivity plot

### Look at counts


## Augment *medpolish* output to produce an augmented plot

Aiming to reproduce the plot on p 176

```{r}

lpc_out <- mp_fit(long_df = survey_responses, 
                  measure = percent, 
                  row_name = generation, 
                  col_name = response)
```
### additivity plot

```{r}
plot(lpc_out)
```

### Graph the median polish fit

```{r}

percent_plot <- mp_plot(lpc_out)
percent_plot
```

### Rotate the fit plot

```{r}

vp <- mp_rotate_plot(percent_plot)
pushViewport(vp)

```

## Save medpolish output in a list and demonstrate basic additivity plot

### Look at percents

```{r}
lpp <- medpolish(percents)
lpp
plot(lpp)
```


## Augment *medpolish* output to produce an augmented plot

Aiming to reproduce the plot on p 176

```{r}

lpp_out <- mp_fit(lpp)
```

### Graph the median polish fit

```{r}


count_plot <- mp_plot(lpp_out)
count_plot
```

### Rotate the fit plot

```{r}

vp <- mp_rotate_plot(count_plot)
pushViewport(vp)

```