1 / 20

MAX 201 Quantitative Methods for the Social Sciences

MAX 201 Quantitative Methods for the Social Sciences. Controlling for a Third Variable; Developing a Research Topic. Univariate, bivariate, and multivariate analysis. Most basic data description : statistics applied to one variable: mean, median, percentages.

Download Presentation

MAX 201 Quantitative Methods for the Social Sciences

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. MAX 201Quantitative Methods for the Social Sciences Controlling for a Third Variable; Developing a Research Topic

  2. Univariate, bivariate, and multivariate analysis • Most basic data description: statistics applied to one variable: mean, median, percentages. • Bivariate analysis uses contingency tables to look at two variables at a time. • Introducing a third variable is called controlling or specifying the original relationship.

  3. Controlling for a third variable: sex and auto accidents • Men have more accidents than women (gender is related to accident involvement) • Introduce a third variable: miles driven per year • Gender is related to miles driven (men drive more) • Miles driven is related to accidents (the more you drive, the more likely you are to have an accident).

  4. SEX ACCIDENT Frequency MILES driven SEX ACCIDENT Frequency Sex & accidents: spurious relationship

  5. Control Variable as an Intervening Variable Ind. Var. Dep. Var. Control Variable

  6. Caffeine and Heart Attacks • One study linked caffeine consumption with an increased of heart disease. • Study did not control for a third variable: whether patients in the smoked. • Heavy caffeine consumers also smoke. • Further study revealed that smoking not caffeine was the cause of the heart disease. • Source: Ian Ayres, Super Crunchers. NY: Bantam: 2007: pp 90-91.

  7. Wealthier is Healthier • For countries, a graph of mortality against income reveals that populations of richer countries live longer. • For individuals within a country, higher- income individuals live longer than lower-income individuals. • The relationships are much more complex, however.

  8. Wealthier is Healthier • For countries, public health –sanitation and removal of waste – plays an important role. Big Medicine – vaccination and antibiotics – play a role. Individual care plays a larger role as income grows. • In poor countries, mosquitoes, AIDS, unsafe drinking water and other public health issues play a large role in mortality. • Insecticide will work to some degree on mosquitoes, but eventually mosquitoes develop a resistance. Need to remove open, stagnant pools where mosquitoes breed.

  9. Wealthier is Healthier • For individuals, the relationship between health and income is more complex. • Better utilization of knowledge about health (smoking) and more education may cause health, rather than income directly causing health. • Certain groups experience more stress from discrimination, low-income and that harms health. • Source: David Cutler, Angus Deaton, Adriana Lleras-Muney. “”The Determinants of Mortality.” Journal of Economic Perspectives. 20 (3) 2006: 97-120.

  10. Controlling for a third variable: health care and party id • Controlling for a third variable discussed in the chapter from the Cole book, “Sorting Out Relationships” • Work through that example: • original relationship is support of health care and party identification (10.1) • control variable is income • different possibilities are considered

  11. Party ID and Health Care Reform

  12. Party ID and Health Care Reform

  13. Interpreting the controls • Controlling for a third variable can help you understand the bivariate relationship you’re interested in • Controlling for a third variable can tell you that the original relationship • Is spurious • Is maintained • Differs within categories of the 3rd variable

  14. GSS data and 3-variable relationships • May find spurious relationships • More likely to find specified relationships • Very likely to look at joint impact of two independent variables on a dependent variable

  15. Example • Hypothesis: People with more children will have more conservative or traditional views on childrearing. • These views may differ by gender.

  16. Assignment 8 • Control for third variable and interpret • Find and summarize an article related to your topic • Briefly describe your projected research project.

  17. Citations of Scholarly Articles • Research-based papers very important • In “Refereed” or Peer-reviewed journals • Other sources (Newsweek, Washington Post, NYT, blogs, on-line informational sites) may be used as supplements, but be careful here. Blogs and wikpedia might be simply uninformed opinion. Do they belong in your paper? • Library data bases (e-journals)

More Related