06. Data Analysis Process Quiz

Data Analysis Process Quiz

Let's build some intuition for the steps in the data analysis process with some questions you might have at each step. We'll use data from Kaggle's Bike Sharing Demand competition. In this dataset, you are given hourly rental data spanning two years from the Capital Bikeshare program in Washington, D.C. Below is a screenshot of the first 5 rows of this dataset. Note: This dataset has been slightly modified.

Note: Solutions to all quiz questions are provided in downloadable link at bottom of page.

*Click on images to make them larger!*

Click on images to make them larger!

Use the table below to help you answer the first question. Each feature is a column in the dataset.

Feature Description
datetime hourly date + timestamp
season 1 = spring, 2 = summer, 3 = fall, 4 = winter
holiday whether the day is considered a holiday
workingday whether the day is neither a weekend nor holiday
weather * 1, 2, 3, 4 (see descriptions below)
temp temperature in Celsius
atemp "feels like" temperature in Celsius
humidity relative humidity
windspeed wind speed
casual number of non-registered user rentals initiated
registered number of registered user rentals initiated
count number of total rentals
* Keys for Weather Feature

1 = clear, few clouds, partly cloudy, partly cloudy
2 = mist + cloudy, mist + broken clouds, mist + few clouds, mist
3 = light snow, light rain + thunderstorm + scattered clouds, light rain + scattered clouds
4 = heavy rain + ice pallets + thunderstorm + mist, snow + fog

Question Step

Question Step

Given the above data on variables that potentially influence the number of bikes rented each hour, what questions would be relevant to ask? (You may select more than one.)

SOLUTION:
  • Which attributes are most important in predicting the number of bikes rented?
  • For which day of the week should the bikesharing company run promotions if the goal is to smooth out the number of rentals across the week?

*Helpful stats on each column for the next question*

Helpful stats on each column for the next question

*\# Non-null values and datatype for each column*

# Non-null values and datatype for each column

Wrangle Step

Wrangle Step

What potential problems do you see with this Kaggle bike sharing dataset that would need to be fixed before continuing with analysis? (You may select more than one.)

SOLUTION:
  • Dates are not in date format
  • Some values are missing
  • Temperature values are far outside a realistic range on Earth

*Scatterplots on the number of bikes rented vs. day of week, temperature, and humidity*

Scatterplots on the number of bikes rented vs. day of week, temperature, and humidity

Explore Step

Explore Step

Based on these scatterplots, which of these three features seems most helpful in predicting count?

SOLUTION: Temperature

*Scatter plot of count vs. temperature with a line of best fit for the next question*

Scatter plot of count vs. temperature with a line of best fit for the next question

Draw Conclusions Step

Draw Conclusions Step

Based on this graph of regressing bike rental count on temperature, how many additional bikes do you think would be checked out if the temperature rose from 2 degrees celsius to 30 degrees celsius?

SOLUTION: 250 bikes

Communicate Step

Communicate Step

What would be valid methods of communicating your conclusions from the Bike Sharing data?

SOLUTION:
  • A written report detailing the most important variables to consider when predicting the number of bike rentals

Great job! Next, let's learn about Python packages, which were used to prepare this quiz!

Next Concept