sariabod/codeacademy

intro to data analysis

codeacademy's Issues

Minor Suggestion: Write out your contingency table

codeacademy/musclehub_project/musclehub.py

Lines 34 to 38 in 56a6d84

	app_counts = df.groupby(['is_application','ab_test_group'])['email'].count().reset_index()
	app_pivot = app_counts.pivot('ab_test_group', 'is_application', 'email').reset_index()
	app_pivot['Total'] = app_pivot['Application'] + app_pivot['No Application']
	app_pivot['Percent with Application'] = app_pivot['Application'] / app_pivot['Total']
	chi2, pval, dof, expected = chi2_contingency([ [app_pivot.iloc[0]['Application'],app_pivot.iloc[1]['Application']], [app_pivot.iloc[0]['No Application'],app_pivot.iloc[1]['No Application']]])

Your data transformations and chi-square test are all correct. Nice job! I would just suggest that you possibly write out in comments your null hypothesis and the pval you got. It may be helpful to save your contingency table to a variable that you can easily call whenever needed. It may also be helpful to readers to write out in comments the format of your contingency table and the associated values.
For example:

Nice! Your plots are great

codeacademy/musclehub_project/musclehub.py

Lines 90 to 99 in 56a6d84

	ax4 = plt.subplot()
	plt.title('Percent of Visitors that Purchase')
	plt.bar(range(len(final_member_pivot)),final_member_pivot['Percent Purchase'].values)
	ax4.set_xticks(range(len(app_pivot)))
	ax4.set_xticklabels(['Fitness Test', 'No Fitness Test'])
	ax4.set_yticks(yticks[0:5])
	ax4.set_yticklabels(ytickl[0:5])
	plt.savefig('percent_visitor_apply_purchase.png')
	#plt.show()
	plt.clf()

Your plots are correct and use good titles and axis labeling.

Remember to use reset_index() with your groupbys

codeacademy/musclehub_project/musclehub.py

Line 32 in 56a6d84

ab_counts = df.groupby('ab_test_group')['email'].count()

reset_index() will give you index numbers for each row of your data frame. This allows for easy access and retrieval of specific rows in your data frame.
Click here for more information on reset_index().

Suggestion: Use Comments

codeacademy/musclehub_project/musclehub.py

Line 24 in 56a6d84

query = "select v.first_name, v.last_name,v.email,v.gender,v.visit_date,f.fitness_test_date, a.application_date, p.purchase_date from visits v left join fitness_tests f on f.first_name=v.first_name and f.last_name=v.last_name and f.email=v.email left join purchases p on p.first_name=v.first_name and p.last_name=v.last_name and p.email=v.email left join applications a on a.first_name=v.first_name and a.last_name=v.last_name and a.email=v.email where v.visit_date >= '7-1-17'"

Use comments liberally to describe what each code chunk is trying to accomplish. This will help future readers to more easily understand your code and critique it.

Also, how did you get the solutions to this project?

Summary: Meets Expectations

Overall, nice project! Your code is functionally correct and performs the correct analysis of the data.
You use both space and time efficient methods. I would just suggest organizing your code and including comments for easier reading.

Your slide presentation looks great! Your introduction and context explanations are clear and concise. You used one overarching hypothesis. However, this project actually attempts to test 3 separate hypothesis:

1. Does the fitness test have an effect on whether or not someone picks up an application?
2. Does the fitness test have an effect on whether or not someone signs up for a membership?
3. Of those who picked up an application, how many purchased a membership?
   You should make note of these questions/motivations in your slides.