Unit 8: Group Experimental Research: Single-Factor Designs Flashcards

1
Q

true experiment

A

A true experiment in which the researcher has complete control over the independent and extraneous variables, so that a cause-effect relationship can be established.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

quasi experiment

A

A true experiment in which the researcher has less than complete control (i.e. the independent variable is a participant characteristic)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

factor

A

the independent variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

level

A

the particular value of the independent variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

condition

A

a broad term used to discuss the independent variable(s)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

treatment

A

another word for condition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

between-subjects design

A

each subject is only exposed to one level of the treatment condition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

within-subjects design

A

each subject is exposed to every treatment condition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

order effect

A

within subjects: changes in behaviour resulting from the order of the conditions - ex. participants getting tired, bored, more experienced

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

sequence effect

A

within subjects: effect that results from interaction between the conditions - they have different effects depending on what order they come in

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

block randomization

A

control procedure in which the order of the conditions are randomized but with each presented once before any are presented a second time. Useful if you want to present conditions multiple times, and are using multiple sessions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Counterbalancing

A

conditions are presented in one order the first time, and the opposite the second time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Define true experiment

A

A true experiment that shows a cause-effect relationship between the independent and dependent variables.

  • researcher has complete control over the independent variable
  • random assignment of P to control and experimental groups
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Describe the two basic elements of good experimental design.

A
  1. presence of a control group or condition

2. random assignment of P. In single-subject, multiple conditions should be random or counterbalanced

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are within-subject designs? What is the basic strategy for achieving control?

A

Subject acts as their own control (pretreatment baseline is the control condition).
Control of order effects and sequence effects within (ideally) or between subjects.
Better ability to detect true differences.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

How are order effects and sequence effects different?

A

Order effects are more general - the conditions can go in any order and the same influence will be exerted on them (things like fatigue, learning).
Sequence effects are interaction between the conditions themselves.

17
Q

How do you control for order and sequence effectswithin subjects?

A

Counterbalancing: ensuring that each condition appears in the ordinal position an equal number of times.
Randomization can be used when the sample is large enough that results of one subject’s sequence effects are unlikely to have an influence on the end outcome.

18
Q

When is reverse counterbalancing ineffective in controlling for order effect?

A
  • when relatively few P will be tested
  • when you have several conditions that can be presented only a few times
  • works well when you think the confounding variables will act in a linear manner over conditions
19
Q

How do you control for order and sequence effects within groups?

A
  1. counterbalance within the group by presenting each condition combination an equal number of times
    Problem: as the number of conditions increases, the number of combinations increases mathematically.
  2. control for order effects but not sequence effects if you give up the condition that every condition follow every other an equal number of times.
    Latin square
20
Q

Latin Square

A

Each P gets a different order of conditions. Order is controlled for, but not sequence.
Permits smaller number of Pp.

21
Q

Balanced Latin Square

A

in each set, each condition is immediately preceded by a different condition

22
Q

What is complete counterbalancing? What does it control for? What is its greatest disadvantage?

A

Complete counterbalancing is a technique in single-factor design that controls for order and sequence effects.
It’s biggest disadvantage is that the more conditions, the exponentially more combinations or sets there are.

23
Q

What are between-subject designs? Explain the basic strategy for achieving control.

A

In between-subjects designs, each P only experiences one condition.
Control is achieved by using a control group.

24
Q

When are between-subjects designs preferred over within-subjects designs?

A

Between subject designs are preferred when P can’t be their own controls because of carry-over effects between conditions (experiencing one will effect experience of all of the others).

25
Q

Describe three designs to avoid in research, and give reasons for avoiding them.

A
  1. One-group post-test design - measures one group, only after treatment. No control or baseline data=no comparison to demonstrate the effect of the treatment. Internally invalid, too many variable uncontrolled.
  2. Post-test only design - using nonequivalent control/experimental groups (not randomly assigned) - too many potential confounding variables - results could be treatment or could be inherent differences.
  3. One-group pre-test/post-test design - measures a single group before and after the treatment. too many variables uncontrolled for.