experimental design Flashcards

1
Q

what is an independent variable?

A

variable which the experimenter systematically varies or manipulates

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

what is a dependent variable?

A

the outcome variable as it is the variable upon which the experimenter is interested in observing effects

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

what are the 3 types of independent variable?

A
situational 
task variable (different tasks)
instructional variable (given instructions or none)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what are extraneous variables?

A

nuisance variables which may influence behaviour under investigation

but if not controlled may confound the results

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

what is a confounding variable?

A

vary systematically with the IV and may effect the DV to provide alternative results

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

what to do when deciding how to measure the DV in an experiment?

A

refer to previous studies

run a pilot study to make sure the task (DV) is of moderate difficulty and that the ceiling effect (too easy) and floor effect (too hard) do not disguise the differences between participants

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

what should you do if the DV is too unethical to perform?

A

have a relevance-sensitivity trade off

e.g using a simulator instead of an actual car when measuring effects of alcohol on drivers

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

what is a quasi-experiment?

A

IVs are comparing existing groups where differences already occur rather than manipulating the IV to make the differences occur

so the ppts are self-selected

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

what conclusions can/can’t be drawn from quasi-experiments?

A

CAN
group performed differently on the DV

CAN’T
control extraneous variables
make causal inferences
say that the IV is a cause of a DV

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

what conclusions can be drawn from experimental studies?

A

extraneous variables can be cotrolled for
causal inference can be made
can conclude that changes in IV cause changes in DV

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

what is experimental design?

A

both the process of constructing experiments and the resulting structure of the experiments

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

what is external valdity?

A

the extent to which the research findings can be generalised to other situations

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

what is operationalisation?

A

the process of deciding how to manipulate and / or measure the independent variable and measure the dependent variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

what is between-subjects manipulation?

A

systematic change to an IV where different ppts are exposed to different levels of the variable by the experimenter

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

what is a manipulation check?

A

a dependent measure that checks the manipulation of the independent variable has been successful

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

what is within-subjects manipulation?

A

systematic change to an IV where the same ppts are exposed to different levels of that variable by an experimenter

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

what is relevance-sensitivity trade off?

A

the more relevant the DV is to the issue that the researcher is interested in, the less sensitive it may be to variation in the IV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

what are demand characteristics?

A

cues that convey an experimental hypothesis to ppts

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

what is statistical conlusion validity?

A

using appropriate statistical procedures

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

what is construct vakidity?

A

operational definitions of the IVs and DVs fit the purpose of the study

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

ecological validity

A

generalising to other environments

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

temporal validity

A

generalising to otheirtimes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

what are the 2 types of external validity?

A

temporal

ecological

24
Q

what is naive empiricism?

A

when experimental findings aren’t used to validate a theory and are instead dircetly applied to the real world

25
Q

threats to internal validity

A
history effects
maturation effects
regression to the mean
testing effect
instrumentation effect
selection effects
mortality
26
Q

what are history effects?

A

event occurs between pretest and posttest and affects the study outcome

27
Q

what are maturation effects?

A

ppts naturally change between pretest and posttest

28
Q

what is regression from the mean?

A

ppts may perform extremely poorly in the pretest and likely to improve for posttest independently

29
Q

what is the testing effect?

A

ppts perform better at posttest because they had practice

ppts perform worse at posttest because they got tired

30
Q

what is the instrumentation effect?

A

performance differs between pre and posttests because different measures are used

31
Q

how to control for most threats to internal validity?

A

use a control group

32
Q

what are selection effects?

A

happen when random assignment fails e.g ppts in different conditions aren’t equivalent

33
Q

what is mortality?

A

ppts with certain characteristics drop out of the study e.g angry or shy

34
Q

what is between-subject design

A

i.e independent measures

ppts selected to one condition only and the different conditions are compared

35
Q

what is within-subjects design?

A

ppts take part in two or more conditions and a comparision is made within the group

i.e repeated masures

36
Q

when is a between-subject design used?

A

when experience gained from one condition makes it impossible to participate in another

37
Q

advantage of a between-subject design?

A

each ppts is fresh and naive to hypothesis

38
Q

disadvantages of between-subject design?

x2

A

more ppts to recruit

may be unexpected differences between the ppts

39
Q

two ways to create groups in between-subjects design?

A

random assignment

matching procedure followed by random assignment

40
Q

what is random assignment?

what is the obejective of it?

A

every ppt has an equal change of being placed in any of the groups
objective to spread individual differences across the groups

41
Q

what is the matching procedure?

A

1) get a score for each person on the matching variable
2) arrange scores from lowest to highest
3) create pairs of scores of those closest to each other
4) for each pair randomly assign one to one group and the other to the other

42
Q

conditions for the matching procedure?

A

must have reasons to believe matching variable will have effect on DV
must be an accurate way of measuring variable

43
Q

when is within-subjects design used?

A

conditions require brief time to test but extensive preparation
population of interest is small

44
Q

advantages of within-subjects design?

x2

A

more data for each ppts

reduced error variance

45
Q

disadvantages of within-subjects design?

A

threats to internal validity (testing effects, maturation etc)
order effects

46
Q

which order effects are there?

A

practice effects - improve due to practice
fatigue effects - reduced due to fatigue
carryover effect - one sequence of condition produces different result than aother sequence

47
Q

how to control order effects?

A

counterbalancing

48
Q

what is counterbalancing?

A

using more than one sequence of conditions

49
Q

what are the two types of counterbalancing?

A

complete

partial

50
Q

what is complete counterbalancing?

A

every possible sequence is used at least once

51
Q

what is partial counterbalacing?

A

using a subset of the total number of sequences
- randomising order of condition for each ppt or sampling from list of possible sequences

latin square

52
Q

what is experimenter bias and how to overcome it?

A

due to knowing hypothesis of study, experimenter may subconsciously help ppts guess the hypothesis

double-blind procedure so don’t know which condition is being tested

53
Q

what are the two types of subject bias?

A

hawthorne effect

demand characteristics

54
Q

what is the hawthorne effect?

A

ppts change behaviour as they know they are being studied

55
Q

what are demand characteristics?

A

being a “good subject” by trying to help confirm the hypothesis if they think they know it

56
Q

how to control for subject bias?

A

deception - keeping ppts naive to purpose of study (reducing demand characetristics) reduces “good subject” effects

placebo - conditions where ppts believe they are receiving treatment but they aren’t