Maps and the Geospatial Revolution: Lesson 4, Lecture 2

2,076 views

Published on

These are the slides to accompany the second lecture from Lesson 4 of Maps and the Geospatial Revolution on Coursera.

www.coursera.org/course/maps/

Published in: Education, Technology
1 Comment
1 Like
Statistics
Notes
  • With regards to the number of vacant homes/lots in the north Midwest, I believe is not only related to second homes, but also the weather. Is the land even viable to be developed given the (possible) brutal weather conditions.
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
No Downloads
Views
Total views
2,076
On SlideShare
0
From Embeds
0
Number of Embeds
6
Actions
Shares
0
Downloads
117
Comments
1
Likes
1
Embeds 0
No embeds

No notes for slide

Maps and the Geospatial Revolution: Lesson 4, Lecture 2

  1. 1. Maps and the Geospatial RevolutionLesson 4 – Lecture 2Anthony C. Robinson, Ph.DLead Faculty for Online Geospatial EducationJohnA. Dutton e-Education InstituteAssistant Director, GeoVISTA CenterDepartment of GeographyThe Pennsylvania State UniversityThis content is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License
  2. 2. Analysis Pitfalls• Good spatial analysis can tell you a lot• Bad spatial analysis can be very misleading(and hard to spot by the untrained eye)• Maps tend to come across as objective andfactual to most normal people
  3. 3. Correlation (is not Causation)• Just because two things co-occur, it doesn’tmean that they are causally related• Examples– Centre County: 18.9% living in poverty, thereforeUniversities are terrible for the economy– Lung cancer mortality and rainfall
  4. 4. Lung Cancer & Rainfall
  5. 5. Lung Cancer & Rainfall
  6. 6. Scale Matters• Depending on the scale at which you analyzethings, you may be able to derive verydifferent results• This is called the Modifiable Areal UnitProblem (MAUP)– Also the toughest Pictionary prompt ever and thesound a cat makes before launching a hairball
  7. 7. Scale Matters
  8. 8. Scale Matters
  9. 9. Scale Matters
  10. 10. Map Rates, NotTotals• Almost anything you can imagine measuring about people andsociety will be population dependent• This means a map that isn’t normalized will just highlightpopulated places• Normalization calculates rates of occurence as a proportion ofoverall population– If you have data for every US county with the # of smelly dogsyou’d need to divide those # by the total population of dogs ineach county– Then you’d have a stink-rate for each place
  11. 11. Normalization
  12. 12. Normalization
  13. 13. Maps and the Geospatial Revolution www.coursera.org/course/mapsTwitter @MapRevolutionOnline Geospatial Education @ Penn State www.pennstategis.comThis content is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License

×