Setting Up the Research

1
Setting Up the Research

• Session 13

2
Tables, Charts, and Diagrams (continued)

• Relationships Between Interval Level Variables
• Scatterplot
• Positive
• Negative
• No Relationship
• Curvilinear
• Outliers

3

• Scatterplots and Regression Lines
• Regression line is a good predictor
• Regression line is a poor predictor

4
(No Transcript)
5
Looking for Explanations

• Independent and Dependent Variables
• Direction of relationship
• Strength of relationship
• Association versus Causation
• Criteria for Causality
• They must covary
• Temporal order the independent variable must
  come first
• Linkage must be plausible
• Relationship cannot be explained away though the
  influence of other competing independent
  variables

6

• A Causal Hypothesis Should Satisfy the Conditions
  of
• Necessity
• Sufficiency
• Research Design should maximize the conclusions
  that can be drawn about causality
• Null Hypothesis
• Alternative Hypothesis
• Null Hypothesis means the first criterion for
  causality has been met

7

• Control Variables and Causality
• Should
• Elaborating a Causal Hypothesis
• Elaboration
• Causal Chain
• Example Miller and Stokes model of
  representation
• H1
• H2
• H3

8
Analyzing More Than Two Variables

• Sorting Out Multiple Influences
• Multivariate analysis
• Testing a bivariate relationship requires the
  introduction of control variables to falsify or
  disprove the hypothesis for the bivariate
  relationship
• Multivariate statistical techniques will allow
  you to examine both the separate simultaneous
  effect of each independent variable and the
  combined effects of all the independent variables
  directly on the dependent variable at the same
  time

9

• Possible Effects of Control Variables on Original
  Relationships
• Refining
• Replicating
• Reducing
• Revealing
• Reversal
• Combined Effect of Control and Independent on the
  Dependent

10
More Research Designs

• Extensive versus Intensive Approaches
• Extensive
• Ordinary Lest Squares Regression
• Logit or Probit
• Two-Stage Least Squares
• Time Series (nonexperimental)
• Intensive
• Case Studies
• Experiments
• Q-Techniques or Q-Sorts
• Example Representation
• Focus Groups

11
Progress in the Research Process

• Circle
• Tentative Generalization
• Conceptualiztion
• Operationalization
• Research Design and Data Collection
• Data Analysis and Decisions to Accept or Reject
  Hypotheses
• Revising Original Generalization Based on Results
• Back to Generalization

12
Stages in Each Study

• Formulate a research idea
• Review literature to find out what is known about
  the topic
• Formulate general hypotheses and clarify concepts
• Operationalize and measure variables
• Formulate a strategy to test hypotheses
• Collect data
• Organize, Analyze, and Interpret the Data
• Draw Conclusions and Present Results