Types of experiments

Question 1

Internal validity

Answer 1

Extent to what researcher is measuring what was intended, including control of variables to ensure changes in DV are solely due to IV

Question 2

External validity

Answer 2

Degree to which a research finding can be generalised to, eg, other settings (ecological validity), other groups of people (population validity) and over time (temporal validity)

Question 3

Replicated

Answer 3

Repeat investigation under same conditions to test validity and reliability of findings

Question 4

Laboratory experiment

Answer 4

. Controlled environment where EV + CV can be regulated
. Participants go to researcher
. IV manipulated and effect on DV is recorded
+ EVs + CVs can be controlled -> effect of EVs and CVs on the DV can be minimised -> Cause and effect between IV and DV can be demonstrated (high internal validity)
+ Can be more easily replicated -> greater control -> less chance new EVs introduced -> findings can be confirmed, supporting validity
- May lack generalisability -> controlled lab environment may be rather artificial and participants are aware they are being studied -> behaviour may not be ‘natural’ and can’t be generalised to everyday life (low external validity)
- Demand characteristics may be a problem -> these are cues in experimental situation that invite a particular response from participants - > findings may be explained by these cues rather than effect of IV ( internal validity)

Question 5

Field experiment

Answer 5

. Natural setting
. Researcher goes to participants
. IV manipulated and effect on DV recorded
+ More natural environment -> participants more comfortable and behaviour more authentic -> results may be more generalisable to everyday life
+ Participants unaware of being studied -> more likely to behave as normally so findings can be generalised -> study has greater external validity
- More diffiult to control CVs/EVs -> observed changes in DV may be due to IV but to CVs/EVs instead -> more difficult to establish cause and effect than in lab
- Ethical issues -> May not have given informed consent -> invasion of participants’ privacy

Question 6

Natural experiment

Answer 6

. Noone manipulated IV -> change but not by experimenter
. DV may be naturally occurring or may be devised by experimenter and measured in field or lab
+ May be only practical/ethical option -> may be unethical to manipulate IV eg study effects of instituitionalisation of children -> natural experiment may be only way casual research can be done for such topics.
+ Greater external validity -> Natural experiments involve real world issues -> findings more relevant to real experiences
- Natural event may only occur rarely -> reduces opportunity for research -> may limit scope for generalising findings to other similar situations
- Participants not randomly allocated -> experimenter no control over which parts in which condition as IV pre-existing -> may result in CVs that aren’t controlled eg Romanian orphans adopted earlier may also be friendlier ones

Question 7

Quasi-experiment

Answer 7

. IV based on pre-existing difference between people.
. Noone manipulated variable
. DV may be devised by experimenter and measured in field/lab
+ Often high control -> often carried out under controlled conditions -> shares some strengths of lab experiments -> replication possible
+ Comparisons made between people -> IV difference
- Parts not randomly allocated -> experimenter no control over which parts in which condition as IV pre-existing -> PV may cause change in DV acting as CV
- Casual relationships not demonstrated -> as with natural experiment, researcher doesn’t manipulate/control IV -> cannot say for certain that any change in DV from IV