11 Steps to Data Analysis

Here is a list of Steps to Data Analysis from the Data Analysis Coursera course.

  1. Define Question
  2. Define Ideal Dataset
  3. Define what data you can access
  4. Obtain the data
  5. Clean the data
  6. Exploratory Data analysis
  7. Statistical prediction
  8. Interpret results
  9. Challenge Results
  10. Writeup results
  11. Create reproducible code for others to recreate

Update: A couple of comments have been made indicating the following 2 steps be added.

  1. Missing Value Analysis
  2. Outlier management

What do you think? Is anything missing?

About Ryan Swanstrom

Creator of Data Science 101

View all posts by Ryan Swanstrom

8 Comments on “11 Steps to Data Analysis”

    1. I would say model validation falls under statistical prediction, but I could also see it being under challenge the results as well. Either way, it is important and needs to occur somewhere.

      Thanks for commenting.

  1. Ryan,
    How would one correlate the earlier post “Levels of Data Analysis” with this particular post? I guess my question is : Are the steps mentioned above valid for one or more levels? Just trying to get the bigger complete picture

    1. Realized I haven’t responded to this yet. I don’t think it is worth its own post, so I will just leave my thoughts here. I would say the “Levels of Data Analysis” map into steps 6,7,8, and possibly step 9 above. How does that sound?

Leave a Reply

Your email address will not be published. Required fields are marked *