Take Five

Duration: ~30 Minutes

Learning Objectives

What is it?
How to download?
How to open?
Structure:
- Data Management
- Data Analysis

What is it?

Take Five: A nightly game from the New York State Lottery.
Nightly, five numbers are chosen "at random".
Each is between 1 and 39.
The data for all winning numbers, starting in 1992, are available on-line at the New York State Open Data site.

Objectives

Download the data from Open Data.
Use the Chi Squared test to see if the distribution of winning numbers is evenly distributed.
With over 36,000 winning numbers chosen since 1992, we have very high power.
Project Code
Download to your local computer & unzip on Desktop.

New Project

Let's make a new project!

File -> New Project
Three options
- New Directory
- Existing Directory
- Version Control
Select the second option.
Select the folder you just created on your desktop.
Let us know now if you don't get this to work.

Documentation

I keep saying R is an ecosystem.
This is a good example.
Open the README.md file with RStudio.
Remember, you can open files in the bottom right pane.
See how that Markdown renders to this: LINK!
Proficient R users are (almost) universally polyglots.
We use lots of languages and tools.
The learning curve _is_ steep. But it _is_ worth it.

Working Directory

Because RStudio sets the current working directory to the root folder of the project, we can use relative urls in our code.
I'll show you what that means in just a minute.

App Tokens

New York State Open Data runs on a platform called Socrata.
To download data, you need a APP TOKEN.
To get one, you must go: here.
You will need an account. Sorry.
Let's walk through this together.

Research Question

Are the winning Take Five numbers selected randomly?

Your Turn!

Team up with a neighbor (or two). Can you run this code?

You don't have to understand all of it.
READMEs are there to help!
You WILL need your APP TOKEN.
Code will not run without it.
Download Code Here!
Can you answer our research question from the last slide?

Where did you start?

Did you create take_five.Rda?

Did you answer Research Question?

We will walk through this thing together.

Where should we start?

Why would we want to separate data management?

Start: data-raw/take_five.R

APP TOKEN

Insert YOUR APP TOKEN into the code.


## GET DATA ====================================================================
## To learn more about using th Socrata API with this data set:
## https://dev.socrata.com/foundry/data.ny.gov/hh4x-xmbw
take_five <- read.socrata("https://data.ny.gov/resource/hh4x-xmbw.json?$$app_token=YOUR_APP_TOKEN")

Data Munging

Can anyone explain this part of take_five.R?


## Transforms " " separated values into a single column, called winning_numbers:
take_five <-
    take_five %>%
    transform(winning_numbers=strsplit(winning_numbers, split=" ")) %>%
    unnest(winning_numbers) %>%
    select(draw_date, winning_numbers=winning_numbers, bonus) %>%
    distinct()