This course will become read-only in the near future. Tell us at if that is a problem.


Stepping back for a moment to reflect on what we learned (and what comes next)

Congratulations. You took the first steps on your way to becoming a master data wrangler. You tackled some of the most important tasks that are part of the data wrangling pipeline [link]. 

You will work through similar steps in almost every data project. Often you will try to answer harder questions, it will be difficult to find the data to answer it, and you will have to spend more time analyzing and representing it to make sure your audience understand the point you are making. But that's icing - we've got the cake. 
Your task:
  • Post comments below to reflect on the experience. 
  • What were key things you learned? (about data wrangling or about other countries)

Task Discussion

  • Lee Meng said:

    I think the biggest problem is not in finding data, but the data's correctness,
    sometime it's difficult to verify the data you found is right or wrong, should be care about that.

    on May 22, 2013, 10:15 p.m.
  • Pelle said:

    What struck me is that a lot of data is scattered around the place, in different formats, hard to compare, and never seems to be complete. 

    Getting the right data to 'prove your point' (or to discover similarities or differences) still takes a lot of work. (echoing Joris, more or less.)



    on Feb. 21, 2013, 6:49 a.m.
  • Leonie McGlashan said:

    I think data is useful when you want somewhere to start and help inform your practice. I think both qualitative and quantitative data are useful. There were many interesting sites that people have shared, and which have led me to some excellent sites in Australia including the Australian Data Archive

    on Nov. 9, 2012, 5:57 a.m.
  • Mark Senn said:

    Data can be wrong.  Never depend on other people's data if you can avoid it.  The cost of doing one's own data collection can be prohibitively expensive so in real life data is used that may be wrong.

    on Sept. 15, 2012, 9:18 a.m.
  • Raul said:

    It's a simple and fun challenge but it is precious and needed. There are lot of data on the web but sometimes is hard to find the correct.

    on Sept. 6, 2012, 1:36 p.m.
  • Jon Mason said:

    hmmm, yes, a nice & new experience. P2PU looks like it's worth pursuing. 

    Can't help myself questioning, though -- reflection here seems to be something that is relegated to the finish of an activity. Reflective practice, on the other hand, integrates it into in situ learning. No doubt, many other learning designs will be used in P2PU

    on Aug. 9, 2012, 1:53 a.m.
  • Pinky said:

    This challenge is very interesting. Nice seeing different perspectives from the natives (or not) of countries.

    on July 18, 2012, 12:11 a.m.
  • Joris Pekel said:

    As noted in the previous comment, the European dataset was not useful because of all the provinces were included. It would become possible if you can take the average for each country which is probably possible with a little bit more time. For now, the worldbank dataset was enough to see that the Netherland has the highest population density in the European Union

    on May 24, 2012, 9:21 a.m.