maximelaurenty / codecademy---introduction_to_data_analysis Goto Github PK

Criteria 1: Presentation/Analysis Document

• Score Level: 4 (Exceeds Expectations)
• Comment(s): Great job putting together the presentation slides and addressing each section of the project!

Criteria 2: Code Accuracy

• Score Level: 3 (Meets Expectations)

• Comment(s): Overall excellent work on the code and results! One small tip here to simplify your code - when your lambda function only involves 1 column of the dataframe, you could apply the function to just that column:

  species['is_sheep'] = species.common_names.apply(lambda x: 'Sheep' in x)

  And for the Foot and Mouth Disease section, something that may be beneficial to remember for the future is to try and be more precise in your input to the calculator. For example, using a Minimum Detectable Effect of 5/15 = 33.3% instead of 33% will actually give you a sample size of 870 instead of 890.

Criteria 3: Understanding hypothesis testing and interpreting the p-value

• Score Level: 3 (Meets Expectations)

• Comment(s): Great work on the chi-squared tests as well as correctly interpreting what the p-values show in terms of significance!

  An additional thing that would be good to incorporate into the discussion in your slides is the null hypothesis (ie. What is the null hypothesis? If the p-value falls below the significance threshold, would we reject or fail to reject the null hypothesis?)

Criteria 4: Ability to interpret the Objective & Outcome of the analysis in business context

• Score Level: 4 (Exceeds Expectations)
• Comment(s): Correct conclusions are formulated from the endangered species and sheep sighting analysis.

Overall Score: 14/16

Well done on this project! Nice work on the calculations and analysis. In addition, your presentation slides were thorough and efficient in presenting your findings! The colors and font used in your slides also work very well, and I liked how you used different text styles (ie. bold, italic) to highlight your points. Great job on the data visualizations as well, including making sure all graphs are properly titled/labeled!

As a further challenge, feel free to continue exploring these datasets and see if you could ask some more research questions about it. Try designing your own tests and continue to play around with different ways of data visualizations. Have fun! :)