1 / 24

Specifying a Research Question

Specifying a Research Question. Searching for “What causes y to vary” Facts are interesting but research into facts is usually limiting. Usually they are a building block in a more significant research activity. Questions. Causal. Factual.

menora
Download Presentation

Specifying a Research Question

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. Specifying a Research Question • Searching for “What causes y to vary” • Facts are interesting but research into facts is usually limiting. Usually they are a building block in a more significant research activity.

  2. Questions Causal Factual What percentage of registered voters cast a ballot in 2008? How much trash did citizens of Austin and Houston recycle last year? What did Obama do to respond to the Arab Spring in Tunisia, Egypt & Libya? • Why do citizens turn out to vote for President? For other political offices? • Why is the rate of recycling higher in some cities than others? • Why do some Presidents respond to international crises militarily while others rely on diplomacy?

  3. Proposing Explanations • Proposing an explanation involves indentifying phenomena that could account for the dependent variable of interest. • We’re looking for causes or independent variables. • The dependent variable (Y) is dependent upon the values of the independent variable (X). • The consequences of the independent variable (X) are its effect on the dependent variable (Y).

  4. HYPOTHESIS (Definitions) • “A proposed scientific explanation for a set of observations.” • “An assertion of a causal relationship between two variables.”

  5. HYPOTHESIS (Definitions) • “An educated guess of what you believe will happen.”

  6. Formulating Hypotheses “Rules” for Hypotheses 1. Identify 2 variables and how they are (plausibly) related 2. Clear, concise, unambiguous 3. Communicate the population of interest 4. Be as general as possible 5. Should not include value statements 6. Should be empirical (about the real world) 7. Should be testable

  7. Types of Variables: • Dependent Variable (“Effect”) • Endogenous Variable • Independent Variable (“Cause”) • Exogenous Variable • Antecedent Variable • Intervening/Mediating Variable • Control Variable

  8. Types of Hypotheses/Terminology 1. Null (A statement of no relationship. Typically used in hypothesis testing as a baseline to be disproven.) 2. Correlative (variables are related but the causal status of each is unknown) 3. Directional/Causal A. Direct B. Inverse C. Curvilinear D. Logarithmic

  9. Types of Hypotheses Type 1. As X increases, Y increases (or decreases) Type 2. (Category A of X) tends to show more of Y than (Category B of X). Type 3. (Category A of X) tends to show more of (Category C of Y) than (Category B of X). Type 4. (Category A of X) tends to show more of (Category C of Y) while (Category B of X) tends to show (Category D of Y).

  10. Examples of Hypotheses Type 1. As (education) increases, (tolerance) increases. Type 2. (Poor citizens) tend to show more (alienation from the political system) than (wealthy citizens). Type 3. (Older Americans) are more likely to (vote) than (younger Americans). Type 4. (Older Americans) tend to (vote) while (younger Americans) tend to (abstain from voting).

  11. Common Problems w/ Hypotheses 1. Statements with only one variable. e.g., Citizens are alienated from the political system. More tolerant people are less prejudiced than less tolerant people. 2. Unclear relationships. e.g., Urban families have low incomes. Urbanization is related to literacy.

  12. Common Problems w/ Hypotheses 3. Assertions lack generality. a. Avoid Predictions e.g. More Incumbents will be elected this Fall than non-incumbents. b. Avoid Personal names e.g., Abraham Lincoln was a better President than Grover Cleveland. 4. Value judgments. e.g., Conservatives are better than liberals.

  13. Criteria for Causal Inference • Language & Intent: “Cause” = “Producing” If X is a cause of Y, then a change in X “produces” or “forces” (rather than is simply “followed by”) a change in Y. • The relationship is “asymmetrical”: If X causes Y, then a change in X produces a change in Y. This doesn’t mean a change is Y produces a change in X.

  14. Cause & Effect Three Conditions: • Temporal Sequence • Constant Conjunction • Nonspuriousness

  15. In a Little More Detail…(3 Types of Empirical Evidence are Required to Demonstrate Causality) • I. The Cause (X) Must Precede The Effect (Y) in Time. • II. The Cause and Effect Must Be Empirically Related. (A “Constant Conjunction” Must Exist.) • III. The Observed Relationship Cannot Be Due To The Influence Of A Third Factor That Causes Both Variables. (No Spurious Correlations.)

  16. Arrow diagrams are often used to represent (sets of) causal hypotheses Two Types of Arrows: • Straight arrows (to indicate causal relationships) • Curved, two headed arrows (to indicate correlational, non-causal or unanalyzed relationships)

  17. “Constituency Influence in Congress” (Miller & Stokes, 1963)

  18. Arrow Diagram for a spurious (non-causal) relationship between B & C

  19. Public Opinion & Public Policy • Republic, not Democracy • Foundering Fathers’ concern about the “Tyranny of the Majority” • “Majority Rule” tradition

  20. Lead or Respond?

  21. Cause and Effect?Democracy or Demagoguery • What’s our theory of democracy? Public Opinion Public Policy/ Political Leaders

More Related