Describing Variables

1
Describing Variables Hypotheses TestingMay
26, 2008

• Ivan Katchanovski, Ph.D.
• POL 242Y-Y

2
Research Hypothesis

• Hypothesis A basic statement that is tested in
  research
• Based on a political science theory and previous
  research
• States expected relationship between dependent
  and independent variables
• This relationship can be tested with empirical
  data
• Hypothesis is either supported or not supported
  by research

3
Dependent and independent variables

• Dependent variable

• Independent variable

• A variable assumed to depend on or be affected by
  the independent variable
• A variable which we are interested in explaining
• Income level depends on education level
• The level of democracy is affected the level of
  economic development

• A variable which presumed to affect the dependent
  variable
• A variable which we use to explain the dependent
  variable
• Education level increases personal income
• The level of economic development has a positive
  effect on the odds that a country is democratic

4
Examples of Research Hypotheses

• People with higher income are more likely to vote
  for the Conservative Party
• Pocket book theory of voting behaviour
• Americans are more likely to join civic
  associations than Canadians are
• Political culture theory
• Democratic countries do not wage wars with other
  democratic countries
• Democratic peace theory

5
Exercise
Formulate a hypothesis describing relationship
between political views and religiosity. Identify
the dependent variable and the independent
variable.
6
Outline of Research Paper

• Political science theory
• previous research
• Your research hypothesis
• Data
• One of datasets from POL 242Y
• Research Method

7
Quantitative Analysis

• Univariate - describe a case in terms of a single
  variable.
• Frequency distribution tables and charts
• Bivariate - describe a case in terms of two
  variables simultaneously.
• Multivariate - analysis of two or more variables
  simultaneously.

8
Frequency distributions

• Tables or charts that summarize the distribution
  of a variable
• Economic development level (World dataset)

Frequency Percent Cumulative Percent
Least 22 20.8 20.8
Middle 54 50.9 71.7
Most 30 28.3 100.0
Total 106 100.0
9
Collapsing categories
Table 1a. How is the government doing in solving
the budget crisis?
Table 1b. Evaluation of government performance
in solving the budget crisis?
Response Percent
Very good 5
Good 42
Poor 21
Very poor 6
Dont Know 26
Total, percent 100
Response Percent
Good or better 64
Poor or worse 36
Total, percent 100
10
Graphic description of variables

• Pie chart

• Bar chart

11
Graphic description of variables

• Line chart

• Histogram
• Figure1. Age distribution of respondents

12
Measures Of Central Tendency

• The Mode Value that occurs most frequently
• useful for nominal variables
• The Median middle value
• useful for ordinal and interval-ratio variables
• Other Measures of Position
• Percentiles, deciles and quartiles _
• The Mean average score X?(Xi)/N
• useful for interval-ratio variables

13
(No Transcript)
14
Skewed Distributions

• Skew The extent to which a distribution of
  scores has a few scores that are extremely high
  or extremely low.
• 
  
• Positively skewed distribution Unskewed
  distribution