Week 3 Notes - Course Introduction

Data visualization & exploratory

why visualization matters

bias in visualization summary can hide some important details example: Anscombe’s Quartet

• can not get ACS for census block, only decenial
• census block groups have big margin of error (ACS), T island problem
• census tracts are better

the smaller of the sample, the bigger margin of error

Grammar of Graphics

ggplot ( data = your_data ) + aes ( x = variable1, y = variable2 ) + geom_something ( ) + additional_layers (color… )

aes: - X,y - color - fill - size - shape - alpha _ transparency

exploratory data analysis

• distribution

join

• left join is preserve the table at first (often option)
• right join …. The second table
• full join is to preserve all the tables (necessary sometimes)
• inner joint is to find the result held by both tables