1 / 0

Arguing with Data: Introduction to Descriptive Data Analysis

Arguing with Data: Introduction to Descriptive Data Analysis. Professor Sarah Reber Lecture 5. Today. Part-to-whole graphs (Pies and alt-pies) Finding and describing the problem High-Achieving, Low-Income High Schoolers The Uninsured Entry and exit Getting IPUMS extracts.

leif
Download Presentation

Arguing with Data: Introduction to Descriptive 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. Arguing with Data: Introduction to Descriptive Data Analysis

    Professor Sarah Reber Lecture 5
  2. Today Part-to-whole graphs (Pies and alt-pies) Finding and describing the problem High-Achieving, Low-Income High Schoolers The Uninsured Entry and exit Getting IPUMS extracts
  3. Part to Whole Graphs Distribution that adds to 100% Pie graph Visualization people don’t like this WSJ Guide says OK People are used to it Clear the pieces add to 100 Donut – better than a pie Bar chart Easier to perceive differences in sizes of the slices Not obvious adds to 100 Can compare 2 distributions
  4. Part to Whole (cont) Stacked bar Clear it adds to 100 Eye perceives area better than angle (pie), less well than length (bar) One looks silly, two looks fat Easier to compare more than 2 Stacked area Continuous version of stacked bar Can use for over-time data Use stacked bar unless you have a lot of years
  5. Pie Chart Rules Direct label both the percent and category if you can Put the largest slices on top Keep the shading simple Highlight one interesting slice if necessary Don’t explode a piece to a pie Use a stacked bar Don’t explode pieces at all Don’t use 3D Consider a donut
  6. The Missing One-Offs Highly selective colleges would like to attract more low-income students to their campuses Would rather not lower admissions standards High-Achieving, Low-Income students can go to highly selective colleges for cheap Don’t apply much Is there a supply of high-achieving, low-income students If so, where, what are there characteristics, how can colleges attract them?
  7. Avery-Hoxby Use College Board and ACT data to find the high-achieving, low-income (HALI) students Top 10% of SAT/ACT distribution Bottom 25% of (estimated) income distribution Who are they? Where are they?
  8. Racial Composition
  9. Racial Composition
  10. Asian 15.2% Hispanic 7.6% Black 5.7% White 69.4% Other 1.4% Native Am 0.7%
  11. Distribution of High Achievers by Race and Ethnicity 75.8% White 69.4% 15.0% Asian 15.2% 4.7% Hispanic All Incomes 7.6% Low Incomes 1.5% Black 5.7% 3.0% Other 2.1%
  12. Who are the uninsured? Income Family Status Age Health Status All this may affect how you design policies to cover them
  13. IPUMS Extract for HW http://www.ipums.org/
  14. Entry, Exit, and Duration Is reduction in welfare caseload because Fewer people are entering welfare (entry) More people are leaving (exit) How will duration be affected? Why might we care? http://aspe.hhs.gov/hsp/indicators04/ch2.htm#IND7
  15. Distribution of Spell Duration by Program, 1996 AFDC 47% 29% 11% 13% Food Stamps 43% 28% 9% 20% SSI 34% 19% 9% 38% 0-4 Months 4-12 Months 13-20 Months 21+ Months
  16. Distribution of Spell Duration by Year 1992 30% 25% 11% 34% 1993 31% 25% 13% 31% 1996 47% 29% 12% 13% 0-4 Months 4-12 Months 13-20 Months 21+ Months
More Related