17.2. Tables of Records

Say you are conducting research about bike sharing for an internship in Washington D.C. You are given the Capital Bike Sharing dataset, contains information on Washington D.C.’s bike share program for 2011. The database has hundreds of thousands of records about every ride that anyone took and you would like to look for patterns in the data to understand questions like: “How long is the average ride?” and “Where is the most popular place to start a ride?”

In a database, the data is arranged in tables like the one shown below. This trip_data table has all of the information about individual trips that were made. Each row in a table is a record - here, each record is an individual trip someone made.

The trip_data table from the bike dataset.

The columns are the pieces of data associated with each record. Here is a description of the columns available in the trip_data:

Column Name

Description

index

A unique number that identifies the trip

duration

The time in seconds of the trip

start_date

Date and time the bike was checked out

end_date

Date and time the bike was returned

start_station

Station number the bike was checked out at

end_station

Station number the bike was returned at

bike_number

Which bike was used

member_type

Type of user - Casual (one time), or Member

Note

Don’t worry about memorizing all of the columns, you can check back here as needed.

Information about the stations is stored in a separate table called the bikeshare_stations table. In that table, each record is the information about one particular station.

The bikeshare_stations table from the bike dataset.

It has these columns:

Column Name

Description

index

A number that identifies the record (1, 2, 3…)

station_id

A unique number that identifies the station

name

The name of the station

status

If the station is open or closed

latitude

Latitude (North-South position) of station

longitude

Longitude (East-West position) of station

You have attempted of activities on this page