choosing the right design Flashcards
What is error variance?
• All the variance in the dependent variable caused by factors other than the independent variable
Example: Comparison of 2 depression treatments. Other influential factors could be: family support, age, diet, season, time of day, life events, personality traits…etc
How to cope with error variance?
Reducing error variance: • Holding extraneous variables as constant as possible (eg experimental protocol) • Match subjects on influential characteristics (see Matched-Group or Within-Group Design)
Other strategies:
• Choosing appropriate levels for the independent variable (Pilot-Study)
• Random assignment of subjects
• Statistical analysis to determine the
probability that error variance alone could cause differences between groups
What is confounding?
• Biased sample: select random sample
• Biased experimenter: blind study
• Extraneous variable( which changes with
independent variable): careful planning of how confounding could play a role.
Between-Subject Design - randomised two-group design
Sample random assignment: Group A (Treatment 1) Group B (treatment 2) –> Mean 1 (Group A) Mean 2 (Group B)
Pro and cons of randomized two group design?
Pro:
• Requires few participants
• No need for categorization/pretesting • Simple statistical analysis
Con:
• No information about type of relationship
• Limited sensitivity
What is randomized multi group design?
Like randomised 2 group design but more groups
Possibilities to add a group: • Quantitative change of the independent variable (Parametric Design) • Example: Cups of coffee 1, 2 or 4 • Qualitative change of independent variable (Nonparametric Design) • Example Cups of coffee 1, 2 or tea
What are pro and cons of matched-group design?
Pro: matching characteristic is meaningful • Individual differences controlled for
→ Higher sensitivity (less error variance) • Fewer subjects needed to show effect
Con: matching characteristic not meaningful
• power of the statistical test is lower compared
to completely randomized
What is matched-pair design?
≈ Randomized two-group design Pro: • Requires few participants • Simple statistical analysis Con: • No information about type of relationship
What is matched multigroup design?
≈ Randomized Multigroup Design
• can give information about the type of
relationship
Problem: If there are more than 3 groups a within-subjects design might be better
What is within-subject design?
- Each subject undergoes all conditions
- No random assignment
- Also called “repeated-measures” design
What are pro and cons of within-subject design?
Pro: • Individual differences problem solved • Very powerful • Few participants needed Con: • More demanding on subject • Carryover effects
What are sources of the carryover effect?
- Learning: performance on task improves simply by repetition
- Fatigue: performance on repeated task worsens, because of fatigue
- Habituation: repeated exposure leads to reduced responsiveness (eg to stimulus)
- Sensitization: repeated exposure leads to heightened responsiveness
- Contrast: comparison of elements within the experiment can change their individual evaluation
- Adaption: a period of adaptation can either increase or decrease responsiveness
How can you deal with the carryover effect? Counterbalancing?
Complete counterbalancing:
• subjects are distributed over all different
sequencing options (eg ABC, ACB,BAC..etc) Partial:
• a random group of possible treatment orders is selected
• Latin Square: each treatment is on each position equally often. As many options are chose as
there are treatments
What are problems of counterbalancing?
Differential carryover effects: treatments have differently strong carry over effects
Irreversible changes: lasting improvement or damage
How can you minimize carryover?
- Practice sessions: to eliminate carryover effects like Irreversible changes, Habituation or Sensitization
- Breaks: reducing short lasting Habituation effects, Adaptation or Fatigue
