Week 7: Apply it to your data 6

Import data

data <- read_excel("../00_data/myData.xlsx")
data

## # A tibble: 21,560 × 12
##     year gnis_id geometry   metad…¹ numbe…² parkn…³ region state unit_…⁴ unit_…⁵
##    <dbl>   <dbl> <chr>      <chr>     <dbl> <chr>   <chr>  <chr> <chr>   <chr>  
##  1  1904 1163670 POLYGON    NA            1 Crater… PW     OR    CRLA    Crater…
##  2  1941 1531834 MULTIPOLY… NA            1 Lake R… PW     WA    LARO    Lake R…
##  3  1961 2055170 MULTIPOLY… NA            1 Lewis … PW     WA    LEWI    Lewis …
##  4  1935 1530459 MULTIPOLY… NA            1 Olympic PW     WA    OLYM    Olympi…
##  5  1982  277263 POLYGON    NA            1 Santa … PW     CA    SAMO    Santa …
##  6  1919  578853 MULTIPOLY… NA            1 NA      NE     ME    ACAD    Acadia…
##  7  1969 1329499 MULTIPOLY… NA            1 NA      IM     TX    AMIS    Amista…
##  8  1967  589056 POLYGON    NA            1 NA      NE     MD    ASIS    Assate…
##  9  1944 1377082 POLYGON    NA            1 NA      IM     TX    BIBE    Big Be…
## 10  1989  302659 POLYGON    NA            1 NA      SE     FL    BICY    Big Cy…
## # … with 21,550 more rows, 2 more variables: unit_type <chr>, visitors <dbl>,
## #   and abbreviated variable names ¹​metadata, ²​number_of_records, ³​parkname,
## #   ⁴​unit_code, ⁵​unit_name

Introduction

Questions

Variation

Visualizing distributions

data %>%
    ggplot(aes(x = region)) +
    geom_bar()

data %>%
    ggplot(mapping = aes(x = year)) +
    geom_histogram(binwidth = 10)

## Warning: Removed 386 rows containing non-finite values (stat_bin).

data %>%
    filter(year > 1920) %>%
    
    ggplot(aes(x = year)) +
    geom_histogram(binwidth = 10)

data %>%
    ggplot(aes (x = year)) +
    geom_freqpoly()

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

## Warning: Removed 386 rows containing non-finite values (stat_bin).

Typical values

data %>%
    
    # Filter out early years
    filter(year > 1920) %>%
    
    # Plot
    ggplot(aes(x = year)) +
    geom_histogram(binwidth = 10)

Unusual values

Missing Values

Covariation

A categorical and continuous variable

data %>%
    
    ggplot(aes(x = year, y = number_of_records)) +
    geom_boxplot()

## Warning: Removed 386 rows containing non-finite values (stat_boxplot).

Two categorical variables

data %>%
    
    count(region, unit_type) %>%
    
    ggplot(aes(x = region, y = unit_type, fill = n)) +
    geom_tile()

Two continous variables

library(hexbin)
data %>%
    ggplot(aes(x = region, y = year)) +
    geom_hex()

## Warning: Removed 386 rows containing non-finite values (stat_binhex).

data %>%
    filter(year > 1920) %>%
    ggplot(aes(x = year, y = region)) +
    geom_boxplot(aes(group = cut_width(year, 5)))

Patterns and models

library(modelr)
mod <- lm(log(gnis_id) ~ log(year), data = data)

data2 <- data %>%
    modelr::add_residuals(mod) %>%
    mutate(resid = exp(resid)) 

data2 %>%
    ggplot(aes(gnis_id, resid)) +
    geom_point()

## Warning: Removed 387 rows containing missing values (geom_point).

data2 %>%
    ggplot(aes(year, resid)) +
    geom_boxplot()

## Warning: Continuous x aesthetic -- did you forget aes(group=...)?

## Warning: Removed 386 rows containing missing values (stat_boxplot).

## Warning: Removed 1 rows containing non-finite values (stat_boxplot).