Subjects Design Flashcards
What is Subjects Design?
The assignment of participants to experimental conditions (levels of the IV)
What are the 3 types of Subjects Design?
1) Between-subjects / Independent groups
2) Within-subjects / Repeated measures
3) Mixed-designs: Mixture of between and within
What is a Between-subjects design?
- There are 2 groups of different Ps
- Ps go through ONLY 1 level of the IV (different participants are used for each of the conditions)
What is a Within-subjects design?
- There is ONLY 1 group of Ps
- Ps go through 2 levels of the IV (repeated measures; same participants are used for both conditions)
What is the issue with the Within-subjects design?
Order Effects
- Participants get bored, tired or lose focus/attention when performing a task for the 2nd time
What is the issue with the Between-subjects design?
Participants in Group 1 and Group 2 may not have equal characteristics which may affect the result/induce bias
e.g. Group 1 majority may not like music but Group 2 majority do. So in a study investigating the effects of music on studying, Group 1 may perform worse
How do we ensure that any differences between the 2 groups in the Between-Subjects design are minimised?
(e.g. differences in age, tiredness, interest)
Random Allocation
What is Random Allocation?
Ensures that each participant is equally likely to be assigned to any IV level
Why do we randomly allocate participants to the conditions in a Between-Subjects design?
1) Randomising distributes the occurrence of potential moderating variables equally among experimental conditions (make it fairer)
2) Prevents experimenters (un)intentionally biasing their results
3) Enables the use of powerful statistical tests that can help determine causal relationships between variables
How do we minimise order effects following a Within-Subjects design?
Counterbalancing
What is Counterbalancing following a Within-Subjects design?
Split the group of participants in half (A and B)
- Group A can participate in Task 1 then Task 2
- Group B can participate in Task 2 then Task 1
Order effects will still influence Ps performance, but the effect of that influence will be evenly spread across each level of the IV
What are Factorial designs?
Experimental designs with 2 or more IVs
What design should we use if the experiment has 2 IVs or more IVs?
Factorial design
What do Factorial Designs allow us to do/ask?
1) What effect does IV 1 have on the DV?
2) What effect does IV 2 have on the DV?
3) What effect does the interaction of IV 1 and IV 2 have on the DV?
e.g. = Effects of alcohol consumption and work shift patterns on work productivity
DV: work output
IV 1: shift pattern
IV 2: alcohol consumption
What design is used when a study has 2 IVs and 4 groups, each group undergoing 1 condition?
Fully Independent Factorial Design (between subjects)
What design does this experiment follow?
Effects of alcohol consumption (IV1) and work shift patterns (IV2) on work productivity (DV)
DV = work output (productivity)
IV 1 = Time of day – between-subjects
(Randomly assign Ps to either the dayshift or nightshift group)
IV 2: Alcohol consumption – between-subjects
(Randomly assign Ps to either take the test with alcohol or without alcohol)
Fully Independent Factorial Design (between subjects)
How many groups does an experiment following a Fully Independent Factorial Design (between subjects) have?
4 groups (2 groups for IV 1 and 2 groups for IV 2)
What design is used when a study has 2 IVs and only 1 group, with all participants in the group undergoing 4 conditions?
Fully Repeated Measures Factorial Design (within subjects)
What design does this experiment follow?
Effects of alcohol consumption (IV1) and work shift patterns (IV2) on work productivity (DV)
DV = work output (productivity)
IV 1 = Time of day – within-subjects
(Split participants into 2 groups; each group takes the test during a day shift and during a night shift with counterbalance)
IV 2: Alcohol consumption – within-subjects
(Split participants into 2 groups; each group takes the test with alcohol and without alcohol with counterbalance)
Fully Repeated Measures Factorial Design (within subjects)
What design does this experiment follow?
Effects of alcohol consumption and work shift patterns on work productivity
DV = work output
IV 1 = Time of day – between-subjects
(Randomly assign Ps to either the dayshift or nightshift group)
IV 2 = Alcohol consumption – within-subjects
(Split participants into 2 groups; each group takes the test with alcohol and without alcohol with counterbalance)
Factorial Mixed Designs
What design is used when a study has 2 IVs and 3 groups? There are 2 groups that undergo 1 condition each and 1 group that undergoes 2 conditions.
Factorial Mixed Designs
Choice in experimental designs depends on… List 2 things
1) Participant availability
(WS = Smaller sample size; BS larger sample size)
2) Concerns in eliminating either order effects or individual differences effects
(WS = no individual differences but has order effects; BS = no order effects but has individual differences)
What is a Quasi-Experimental design?
Between-Subjects Designs without random allocation
Which experimental design follows the assignment of participants to experimental
conditions in a pre-determined manner?
e.g.: compare pre-existing alcohol consumption groups: alcoholics vs. nonalcoholic
or compare male vs. female
Quasi-Experimental design
How do we eliminate differences between the groups other than the variable of interest in Quasi-Experimental designs?
Matching groups
- Identify potentially moderating variables and match the groups on this basis
e.g. are 18-25-year-old men in prison more likely to have dyslexia compared to 18-25-year-old men who are not in prison?
- Any difference you measure in the incidence of dyslexia might be explained by other factors (e.g. intelligence, years in education, SES)
Solution = Match the groups based on IQ, education level & SES
Apart from matching groups, what else can we do to eliminate differences between the groups other than the variable of interest in Quasi-Experimental designs?
Matching Individuals
- Match (pair) individual participants on the basis of such variables
e.g. are 18-25-year-old men in prison more likely to have dyslexia compared to 18-25-year-old men who are not in prison?
- Compare individuals from the same background
How do we resolve the problem of an experiment which follows a Within-Subjects Design without counterbalancing?
Pre-test post-test control group design
How does the re-test post-test control group design help with resolving the problems in Within-Subjects Design without Counterbalancing?
1) Split Ps into 2 groups and manipulate the IV in one group only
e.g.
Treatment group: test → manipulation → test
Control group: test → no manipulation → test
- The inclusion of a control group allows us to account for any order effects that
might be present - We can then statistically control for the difference in the treatment group
accounted for by order effects
What are the specific terms used to differentiate between and within subjects designs in developmental research?
Between subjects = Cross-sectional (short-term)
Within subjects = Longitudinal (long-term)
What is a cross-sectional study?
Study participants in the short-term or between-subjects
What is a longitudinal study?
Study participants in the long-term or within-subjects
What is the problem with cross-sectional study?
Problem = It’s not possible to randomly assign participants
Solution = matching
What is the problem with longitudinal study?
Problem = it’s not possible to
counterbalance order
Solution = control group
Quasi-experimental designs are designs in which…
a. the same participants are used for each of the levels of the independent variable b. experimenters are able to control the assignment of participants to the different experimental conditions c. there are both between and within subjects components d. the assignment of participants to experimental conditions is pre-determined
D
Between or within?
Differences between males and females in attention span
a. Between b. Within
A
Between or within?
Following a group of children through development and measuring their reading ability at 5, 8 and 10 years.
a. Between b. Within
B
Between or within?
Giving a group of students a knowledge quiz before and after a lecture to see if their understanding has improved.
a. Between b. Within
B
Between or within?
Investigating the effect of employment status on motivation.
a. Between b. Within
A
Doron is designing an experiment. He doesn’t have many participants and is concerned that differences between participants might confound his results. What design is best for Doron?
a. Independent groups b. Repeated measures c. Mixed design d. Fully independent factorial design
B
In developmental designs, between subjects designs are often referred to as:
a. Longitudinal b. Critical Evaluation c. Case-Control d. Cross-Sectional
D
Factorial designs…
a. are designs which have more than 1 independent variable b. are designs which contain both within and between subjects factors c. are designs which have only 1 dependent variable d. all of the above
A
Which of the following is NOT an example of a quasi-experimental design?
a. Comparing the memory of girls and boys b. Comparing attention of addicts and non-addicts c. Comparing reaction times when listening to music and when not listening to music d. Comparing personality of psychology and engineering students
C
One way of avoiding bias in quasi-experimental designs is…
a. Random allocation of participants b. Counterbalancing conditions c. Matching groups or pairs of participants d. All of the above
C
