Control issues

Question 1

How can the other threats to internal validity be controlled for in an experimental design?

Answer 1

• Instrumentation effect is also controlled for unless the measurement/test varies significantly
  between measures
• Regression to the mean is controlled for if the two groups were drawn from the same
  sample
• Attrition can still be an issue.

Question 2

Does experimental design have the strongest internal validity?

Answer 2

Yes. It controls for history, maturation, testing effect, and selection bias

Question 3

What are the two basic experimental designs?

Answer 3

• Between-subject
• Within-subject

Question 4

What is the between-subject (independent measures) experimental design? And what is a problem associated with it?

Answer 4

• Each person takes part in only one condition of the research
• Problem: creating equivalent groups
• They are the most commonly used design type.

Question 5

What is the within-subject (repeated measures) experimental design? And what is the problem with it?

Answer 5

• Each person takes part in all conditions of the research
• Problem: participating in one condition might affect behaviour in another condition (sequencing effects)

Question 6

When do we use between-subject designs?

Answer 6

• Use when the IV is a subject variable (e.g. extrovert/introvert; marital status)
• Used when experience gained in one level would make it impossible to participate in another level

Question 7

What are two ways to tro have equivalent group in between-subject desgins?

Answer 7

• Random assignment
• Matching

Question 8

What is random assignment?

Answer 8

• The goal is take individuals factors that could bias the study and spread them evenly throughout different groups
• randomness isn’t guaranteed
• works best with large numbers)
• Random assignment+large sample controls for selection bias, as well as:
• Maturation (controlled for because both groups mature at same rate)
• Instrumentation
• Attrition can unbalance groups (this can only be countered statistically)

Question 9

What is matching?

Answer 9

• NB: this is not an experimental design
• Use only when you have a small number of participants available, and/or can’t use random assignment.
• Choose a matching variable that correlates with the DV (i.e. is expected to affect the outcome in some way)
• Make sure there is some reasonable way of measuring participants on the matching variable.
• Decidedly not random.

Question 10

What are the advantages of within-subject designs?

Answer 10

• Needs fewer people
• No problem with equivalent groups
• Reduces error variance, since you have no between-condition individual difference- so it gives more statistical power to find an effect if there is one.

Question 11

What are the problem with within-subjects design?

Answer 11

• Mainly sequence or order:
• Practice effects
• Fatigue effects
• Carryover effects (does it matter if condition A come before condition B)

Question 12

What is counter-balancing?

Answer 12

It solves the carryover effects.
- Ensure that half the participants do condition A first, and the other half do condition B first.

Question 13

How do you minimise practice effects in within-subject designs?

Answer 13

o randomizing or counterbalancing of the order in which participants take part in conditions
o randomizing the order that stimuli are presented over the course of an experimental session
o minimizing the overall effects of practice by allowing participants ‘practice trials’ that serve to get them performing at a desired level before testing takes place

Question 14

What are mixed designs?

Answer 14

• Sometimes you have both between and within- group measurements.
• Avoids most threats to internal validity if don’t properly
• Control groups are critically important
  o Having a repeat measure without a control group can result in internal validity errors, particularly the placebo effect.