1 / 13

Hello and Welcome to… Data analysis

Hello and Welcome to… Data analysis. with your hosts: Erin Sills and Jerry Shively. Description vs. Explanation. Describing a situation is good Explaining why a situation exists is better Example of description: poor households are more reliant on forests Example of explanation:

warren
Download Presentation

Hello and Welcome to… Data analysis

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Hello and Welcome to…Data analysis with your hosts: Erin Sills and Jerry Shively

  2. Description vs. Explanation Describing a situation is good Explaining why a situation exists is better Example of description: poor households are more reliant on forests Example of explanation: poor household have low agricultural capacity, and therefore must rely on forests

  3. Description vs. Explanation • Answering interesting questions: • which radio program? • Look at your data: • listening to the radio – what station are looking for? • Unconditional means vs. conditional means • Statistics vs. Economics • Signal vs. Noise • Is that static I hear?

  4. Explaining variation • Variance is your friend (up to a point) • Variance in data = underlying variation in either behavior or constraints • Without variance, there is nothing to explain • But, love is like oxygen…

  5. Tuning in • What is the relationship between an outcome and a key variable of interest (for policy or theory)? • What are the determinants of outcomes(as suggested by theory, literature, field experience, patterns in the data) • Causation vs. correlation

  6. What is your story? • Find a story → try to change or undermine your story→ new & potentially more interesting story • Subject your story to robustness checks • Embrace parsimony • What is the simplest story that is consistent with your data? • Simple stories are more appealing

  7. Developing your story Example: Income from Fishing Mean = $310/hh/ year St. dev. Median • 175 300 village 1 • 343 85 village 2 (+ outliers?) • 530 0 village 3 (few fish?)

  8. Village 1 Mean = 310, Stdev = 175, Median = 300

  9. Village 2 Mean = 310, Stdev = 343, Median = 85

  10. Village 3 Mean = 310, Stdev = 530, Median = 0

  11. Bonus round

  12. Back to fishing Hypothesis: more educated HHs fish more Estimate a bivariate regression Y = income from fishing X = yrs education Y = 5 + 0.65*X Tuning in • Parsimonious → Challenge the story

  13. Back to fishing IncF = 5 + 0.65 yrs educ Source: NIST

More Related