Quiz question 2 feedback

Module 1. The Predictive Modeling Pipeline Tabular data exploration
#1

Spoiler alert: This is supposed to be feedback to the course organizers and not to share answers. Please let me know, if I should move/tag this differently.

Maybe I was just skimming the content too quickly to remember the correct option. But I found the answer options overlapping and not clearly separated. Also from my understanding, calling the content at this point already “data manipulation” felt a bit like a strech, when we actually barely loaded and viewed some nicely formatted data.

All other quiz questions so far, were really well phrased in comparison and the options really clear and easy to distinguish.

Hope this feedback helps.

12 Likes
#2

I agree with your observation. I just happened to get it right although I wasn’t sure what the difference is between the two options. I think the course coordinators should look into this and correct it. Maybe they meant something else with the second option!

Sam

1 Like
#3

Hye,
Thanks for the feedback.
But sorry, I deleted the capture of your post because there was the answer of the quiz, which is a shame!
I let you modify your post to explain the problem without showing the answer of the quiz.

#4

Hi @lfarhi.
Thanks! I wasn’t sure how to submit this properly. The deleted screenshot showed two answer options in question 2 with nearly identical phrasing of two options and I was trying to recommend more clearly distinguishable choices.

2 Likes
#5

Hi
When I read the description of the Quiz Intro.01, I can’t understand that 97630 is post code!, because in my country post code is different!
so it’s done for me because I used 2 of 2 submissions in question 2.
I hope you fix this in the future.
thanks.

2 Likes
#6

I completely agree! I was confused by the second question too. The first and the second choice is very similar and not everybody thinks about loading a dataset and calculating statistical summaries as “data manipulation” anyway.

I think the answers should be changed into:

  • load the dataset
  • calculate statistical summaries
4 Likes
#7

I found that in the course material on github the question already has better answers:

In the previous notebook, we used:

  • a) pandas to manipulate data
  • b) pandas and seaborn to visually inspect the dataset
  • b) numpy and scipy to perform numerical inspection (for instance using scipy.optimize.minimize)
  • c) scikit-learn to fit some machine learning models
5 Likes
#8

As the solution mentions seaborn, I guess seaborn should be in one of the answers.

1 Like
#9

I also made that mistake, and I’m glad I’m not the only one (postcode was ok though, I’m french ^^).

#10

Hi, I found the wording of the question 2 unclear as an English-speaker. Manipulation would typically involve some treatment of the data, not simply retrieving it. Maybe Extract-Transform-Load are better terms here? Peter

#11

Hello,
Sorry, actually, there was a mistake in the question proposals.
I have reset it for everybody, you can answer again. You have 2 possible submissions.

4 Likes
#12

same here,
my first thought was “data manipulation” = “data modification”, after seeing my answer was wrong I had a second look at this and came up with “data manipulation” = “data calculation”

1 Like
#13

Hello, I also had this problem, i thought manipulation = to modify, first i selected second option then i selected first option so i lost both submissions. Is there any solution to answer the question again? thanks