10/30/00
ANALYSIS AND INTERPRETATION
Measures of central tendency
Mean
Median
Mode
Measures of variability
Range
Variance
Normal curve
2. Significance level, or "alpha"
By convention, p < .05
Type I error (saying your hypothesis is true when it’s really not)
Type II error (saying your hypothesis isn’t true when it is)
QUESTIONS OF COMPARISON:
chi-square (X2) are two nominal (categorical) variables associated?
t-test do two groups differ in mean score on a continuous variable?
analysis of variance (ANOVA) - do three or more groups differ in mean scores on a continuous variable?
QUESTIONS OF COVARIANCE:
correlation -- are two continuous variables associated?
linear regression what are the relative influences of two or more continuous or categorical variables on a continuous dependent variable?
Effect size
Statistical versus clinical significance
Power
11/6/00
SAMPLING & STUDIES WITH SMALL SAMPLES
Population
Sample
External validity (or generalizability)
Random sampling
Convenience sampling
Analogue samples
Characteristics of case studies
Case studies are a valuable tool
Source of ideas and hypotheses
Source for developing therapy techniques
Permit the study of rare phenomena
Provide counter-examples
Persuasive and motivational tool
The case of Freud
Limitations
No controls for internal validity
Subject to clinicians’ biases
Case may not generalize
Cannot be replicated
Impose more control
Begin with baseline assessment: ABAB
11/13/00
RESEARCH ETHICS
Above all, participants must be protected from psychological and physical harm
Planning research (6.06)
Responsibility (6.07)
Institutional approval (6.09)
Informed consent (6.11)
Competence
Knowledge
Volition
Anonymity
Confidentiality