Worksheet 18

library(tidyverse)
library(openintro)

Exercise 1

This exercise uses the rain gauge data from the city of Seattle.

rain <- read.csv(url("https://data.seattle.gov/api/views/rdtp-hzy3/rows.csv"))
# The next line of code tells the computer how the date is formatted. This is needed for the graph below. We will also filter just the first two gauges.
rain$Date <- as.Date(rain$Date, format = "%m/%d/%Y")

The code below makes a scatterplot for Rain Gauge 1 and Rain Gauge 2 over time.

ggplot(data=rain)+
  geom_point( aes(x=Date, y=RG01), color = "blue")+
  geom_point( aes(x=Date, y=RG02), color = "red")+
  scale_x_date(date_labels = "%b %Y")+
  theme_classic()+
  labs(x="Date", y= "Rain Fall Amount")

This graph is hard to read but it looks like the blue and red points are close to each other on each date. An easier way to see this is to remove the date and to graph each rain gauge on its own axis. Create this graph in the code chunk below.

ggplot(data=rain, aes(x=RG02, y=RG01))+
  geom_point( color = "blue")+
  theme_classic()+
  labs(x="RG02", y= "RG01")+
  geom_smooth(method = "lm", se = FALSE)

## `geom_smooth()` using formula = 'y ~ x'

Does there seem to be a correlation between the rain gauges? To look at the best fit line, add the code to the scatterplot to the code for the regression line below.

  geom_smooth(method = "lm", se = FALSE)

## geom_smooth: na.rm = FALSE, orientation = NA, se = FALSE
## stat_smooth: na.rm = FALSE, orientation = NA, se = FALSE, method = lm
## position_identity

To see how close a fit this line is to the data, we can find Pearson’s r with the code below.

  cor(rain$RG01, rain$RG02)

## [1] 0.9619801

What does this tell us? The data is not far from the line. ### Exercise 2

We will use the Human Freedom Index dataframe which is already loaded into R Studio. Use the glimpse command to look at the data.

#  glimpse(hfi)+

We only want to look at the year 2016 so use the code chunk below to create a new dataframe by filtering for the year 2016. Call it hfi_2016.

# filter for 2016 
hfi_2016 <- filter(hfi, year==2016)

We want to examine the relationship between the personal freedom score of a country and the pf_expression_control, or its score out of 10, with 0 being the most, of political pressures and controls on media content.

To start, make a scatterplot with pf_expression_control as the explanatory variable and pf_score the response variable.

ggplot(data= hfi_2016, aes(x=pf_expression_control , y=pf_score))+
  geom_point()+
  theme_classic()

Does there appear to be a linear relationship between these variables? Yes. Positive Linear relationship.

To measure the strength of this relationship, we look at Pearson’s coefficient of r by using the code chunk below.

  cor(hfi_2016$pf_expression_control, hfi_2016$pf_score)

## [1] 0.8450646

To look at the best fit line, add the code to the scatterplot to the code for the regression line below.

ggplot(data=hfi_2016, aes(x=pf_expression_control, y=pf_score))+
  geom_point()+
  theme_classic()+
  geom_smooth(method = "lm", se = FALSE)

What does this tell us about the relationship between personal freedom and political media control? The higher the personal freedom the higher the media is free to report what they do. This is a positive linear relationship.