RESEARCH METHODS (1) 
Experimental Method
Aims & hypotheses
Independent variable (IV)
& dependent variable (DV)
Hypothesis = statement of what researcher believes to be true
Directional hypothesis states whether changes are greater or lesser, +ve or -ve etc.
Aims = general statement of investigation
Non-directional hypothesis doesn't state the direction
IV manipulated by researcher
The DV is measured
There are levels of the IV: control condition & experimental condition
The variables must be properly operationalised (measurable) before hypothesis can be tested e.g.. drinking 300ml of product compared to 300ml of water & recording how chatty they are
Control of Variables
Sources of bias
Controlling bias
Extraneous Variable (EV)
& Confounding Variable (CV)
EV = any variable other than the IV that may effect the result - nuisance that doesn't vary systematically with the IV e.g.. age of participants, lighting in the lab
CV = any variable, other than IV that has affected the results - varies systematically with the IV e.g.. if participants naturally loud & chatty can't be sure if change in DV due to this or energy drink
Demand characteristic = cue from researcher or situation that may reveal aim of study
Investigator effects = influence of researcher's behaviour on participants' behaviour e.g.. researcher may smile more at one group when recording how talkative they are, may encourage more chattiness
Randomisation = use of chance when designing investigations to control for effects of bias e.g.. order of list of words randomly generated
Standardisation = using same formalised procedure for all participants
Experimental Design
2) Repeated Measures
3) Matched Pairs
1) Independent Groups
Evaluation
One group to condition A and second to condition B
Random allocation is used to assign participants to groups
Strength - less of problem with participants guessing aims of study
Limitation - participant variables may act at EVs e.g.. condition A could have more naturally talkative people than B
Strength - order effects avoided e.g.. practice effect or boredom effect
Limitation - more participants needed than for repeated measures design
The order in which participants tested should be varied to reduce order effects e.g.. practice effect or boredom
Evaluation
Each participant does all conditions i.e. is tested again
Pairs of participants are matched & placed in separate groups eg. energy drink experiment - person who spoke most paired with person who spoke second most, one member of pair put in condition A & other in condition B.
Evaluation
Two groups of participants are used like in independent design - conditions A & B
Strength - fewer participants needed than independent design
Limitation is order effects can be a problem
Strength - controls for participant variables
Limitation - participants may guess aim of study
Strength - no problem with order effects
Limitation - matching takes time
Strength - participant variable are partly controlled
Limitation - more participants are needed than repeated measures design
Types of Experiment
Quasi-experiments
Natural
Field
Laboratory
Easier to control IV and EVs & CVs
Evaluation
Special environment where behaviour can be investigated under controlled conditions.
IV controlled but EVs less so
Evaluation
More natural environment, less controlled
DV may be assessed in natural environment or in lab e.g..Romanian Orphans the child's aggressive behaviour could be assessed in natural environment or lab
Evaluation
Experimenter doesn't manipulate the IV e.g.. Romanian Orphans - IV was whether children adopted early or late & would have occurred anyway without experimenter
As IV a difference between people, participants can't be randomly allocated to experimental conditions because participants in each group decided before experimenter comes along
Evaluation
IV is pre-existing condition, not a variable e.g.. age or gender
Sampling
4) Systematic Sample - system used, there is a pattern so every member of target pop does not have equal chance e.g.. every nth person chosen
Evaluation
Evaluation
5) Stratified Sample - method to obtain representative sample e.g.. opinion poll (contains selection of subgroups selected according to their frequency in target population)
3) Random Sample - equal chance of selection for every member of target population e.g.. lottery or random number table
Evaluation
Evaluation
1) Opportunity Sample - most easily available e.g.. people nearby
Populations - target population selected & sampling aims to obtain representative sample although often biased (certain groups like ages or gender over or under represented)
Strength - lab experiments easily replicated
Limitation - may lack generalisability as artificial
Limitation - demand characteristics may be a problem
Strength - EVs can be controlled
Strength - environment may feel more comfortable & natural for participants
Strength - participants not usually aware being studied
Limitation - more difficult to control CVs
Limitation - ethical issues re informed consent & not always debriefed at end
Strength - some cases the only ethical way causal research can be done
Strength - greater external validity e.g.. natural disaster & stress levels
Limitation - natural event may only occur very rarely
Limitation - participants may not be randomly allocated to experimental conditions
Strength - comparisons can be made regarding behaviours of different groups eg. people with & without autism
Limitation - random allocation to conditions is not possible
Strength - often high control as carried out under controlled conditions, effects of EVs on the DV minimised
Limitation - causal relationships not demonstrated as IV pre-existing
No control group
Limitation - inevitably biased sample as from specific area so can't be generalised
Strength - random sampling potentially unbiased
Limitation - representative sample not guaranteed because random
Strength - unbiased
Limitation - takes more time & effort that some other methods
Strength - likely to be more representative of target population than other methods
Limitation - not perfect as can't reflect all the ways people are different within target population
2) Volunteer Sample - ad on noticeboard or newspaper asking for participants so self-selecting unlike Opportunity
Evaluation
Strength - guarantees you get willing participants
Limitation - likely to be biased (volunteer bias - as volunteered likely to have certain characteristics not representative of target population)
Participants also related to each other by being paired
Counterbalancing - To limit order effects counterbalancing often used - half participants do condition A then B & other half do condition B then A (ABBA technique)
Strength - quickest method to use