Political Science 5 Lecture 3, 232004

1
Political Science 5Lecture 3, 2/3/2004

• 1. Homework 1 due next Tuesday!
• 2. Read Chapter 3 and article for Thursday
• 3. Office Hours
• 4. Hypothesis Examples now on Website

2
Review antecedent and intervening variables

• Hypothesis a person who favors national health
  insurance is more likely to have voted for
  Clinton in 1992.
• What is the independent variable?
• What is the dependent variable?
• What could be an antecedent variable?
• What could be an intervening variable?
• What could be a that drove or resulted from this
  research?

3
One more example(to make sure we really got it)

• Hypothesis the number of years of formal
  education had by a voter will affect that voters
  propensity to vote
• Dependent variable? Independent variable?
• Intervening variables
• Formal education causes a sense of civic duty
• Formal education lets one understand the issues
• Use arrow diagram to tell causal story

4
A Note on Causality

• If we say that something (X) is the cause of
  something else (Y), we are making three claims
• X and Y covary. A change in one variable is
  associated with a change in the other.
• A change in X precedes the change in Y.
• The covariation between X and Y is not spurrious
  (not due to a change in some other variable) and
  not just a coincidence.

5
What is a concept?

• A concept is an abstraction based on
  characteristics of perceived reality.
• Concepts should be accurate, precise, and
  informative.
• In fact, a scientific discipline maintains its
  identity because different researchers within it
  share a concern for the same concepts.
• How is this different from a variable?

6
(No Transcript)
7
What is the difference between a concept and a
variable?
8
What is the difference between a concept and a
variable?

• A variable is a measured concept.
• From the text how can the concept of political
  participation become a variable?
• There can be more than one variable for a single
  concept.
• VARIATION refers to differences within a set of
  measurements of a variable.

9
Measurement I. Knowing a Good Measure When You
See It.

• Conceptual and operational definitions of a
  variable
• An example voter turnout in the U.S.
• Measurement error and validity

10
Conceptual Definitions

• A conceptual definition of a variable states
  precisely what you mean when you use a particular
  term.
• A society that supports democracy is one that
  emphasizes tolerance, trust, political activism,
  and Post-Materialist values, not just one that
  pays lip service to the ideals of democracy. --
  Inglehart

11
Conceptual Definitions

• A good conceptual definition
• Looks at how other scholars have defined a term,
  and goes with the consensus unless there is a
  good reason to deviate.
• Allows you to test the theory that you want to
  test.
• Is not circular (for example, does not say that
  support for democracy is when the public voices
  support for democratic values)

12
Operational Definitions

• An operational definition of a variable is a
  complete recipe for going out into the world and
  measuring a variable.
• It helps us make the leap from our subjective
  impressions to an objective measure.
• Like a recipe in any good cook book, a good
  operational definition can be followed by any
  other chef.

13
Operational Definitions

• To measure support for democracy, Inglehart
  uses the World Values/ European Values Survey
  from 1995-97 and 1999-2001 and looks at average
  levels of agreement with nine statements such as
  
• Is it good to have a strong leader who does not
  have to bother with parliament and elections?

14
Voter Turnout in the U.S.The Traditional Measure

• Conceptual Turnout is the percentage of those
  who can vote who actually do.
• The traditional operational definition is
• Where Voting Age Population (VAP) includes
  noncitizens and convicted felons, but not
  citizens living overseas.

15
Voter Turnout in the U.S.The Traditional Measure
16
Voter Turnout in the U.S.The McDonald Popkin
Measure

• Conceptual Turnout is the percentage of those
  who can vote who actually do.
• Where Voting Eligible Population is purged of
  noncitizens and convicted felons but includes
  citizens living overseas.

17
Voter Turnout in the U.S.The McDonald Popkin
Measure
18
Measurement Validity

• Measurement Validity is highest when the gap
  between your conceptual definition and your
  operational definition is smallest.
• Validity means the extent to which a measurement
  procedure measures what it is intended to measure

19
Measurement Reliability

• Reliability means the extent to which a
  measurement procedure consistently measures
  whatever it is supposed to measures

20
Example

• Hypothesis courses taught by political
  scientists are more worthwhile than those taught
  by psychologists.
• You and a friend enroll in poli sci and psych
  courses, respectively
• After two weeks, turns out that your poli sci
  course is dreadful
• Does this disprove the hypothesis?
• What if the person teaching your course was not a
  political scientist after all?
• Measurement was inaccurate, so the comparison you
  made with your friend is irrelevant to the
  hypothesis

21
How to test for Reliability

• Test for same results on repeated trials
• E.g. count ballots several times, ask other
  people to count them
• Get several people to read articles (if testing
  orientation of a newspaper)
• Problem something may be wrong
• Use two different measures of the same concept
• See if you get similar results
• Two different questions about liberalism

22
How to test for validity

• Again correspondence between measure and the
  concept it is thought to measure
• Hypothesis the larger a citys police force is,
  the less crime the city will have.
• Use police data on crime. (invalid?)
• Hypothesis the more productive a scholar a
  faculty member is, the better teacher they are.
• Use student evaluations. (invalid?)

23
How to test for validity

• Face validity
• IQ tests Measuring intelligence and exposure to
  middle class white culture? Not valid. Matter
  of judgment.
• Content validity
• Determine the full domain or meaning of a
  particular concept
• Construct validity
• GRE example

24
One more example

• From an actual survey
• Please look at this page and tell me the letter
  of the income group that includes all the members
  of your family living here in 2002 before taxes.
  This should include salaries, wages, pensions,
  dividends, interest, and all other income.

25
Inglehart article Group Activity

• Answer the following questions
• What is the hypothesis?
• What is the dependent variable?
• Conceptual definition and operational definitions
• What is the independent variable?
• Conceptual and operational definitions
• What is the causal story?
• Are issues of validity or reliability mentioned?