4.2.3 Research Methods

Aim

Sets out what a researcher wants to meausre

Hypothesis

•A testable prediction, often implied by a theory
•measurable statement

difference between aim and hypothesis

aim has identified what the researcher wants to investigate
but the hypothesis is a testable prediction of what the results will be

Types of hypothesis

alternative and null

alternative hypothesis

'there will be a difference' between two conditions
testable statement which predicts the effect of one variable on another

null hypothesis

testable statement which predicts that one variable will not have an effect on another
'there will be no difference' between the conditions

Types of alternative hypothesis

directional and non-directional

directional hypothesis (one-tailed)

States the direction of the difference or relationship

non-directional hypothesis (two-tailed)

Simply predicts that there will be a difference or relationship between two conditions.
results could go either way

independent variable

variable that is manipulated

dependent variable

the variable that is measured in an experiment

Operationalisation definition

•putting something into operation
making it use able
•clearly defining variables in terms of how they can be measured
•make the variables measurable

how do you operationalise an IV

specifying how the IV will be manipulated

how do you operationalise a DV

specify how the DV will be measured

extraneous variable

•In an experiment, a variable other than the IV that might cause unwanted changes in the DV.
•not the IV but can affect the DV
'extra' variable

how do we 'get rid' of the extraneous variables?

control them

how do we control variables?

standardise or randomisation

standardisation

make sure all conditions are the same for each ppt
e.g. set of instructions given to all ppts

randomisation

leaving the differences in conditions up to chance
remove bias
e.g. random allocation of ppts, random allocation of materials (e.g. words in memory test)

confounding variable

variable whose influence on the dependent variable cannot be separated from the independent variable being examined
•occur at the same time in which IV manipulated or ppts are allocated. add a third variable to study

Extraneous variables can be either:

participant variables
or
situational variables

Participant variables

Individual differences in the personal characteristics of research participants that, if not controlled, can confound the results of the experiment.
•anything to do with the people used in the study, which could effect the DV other than the IV

examples of participant variables:

age, gender, mental illness, IQ score

Situational variables

features in the environment that the study was conducted in which could effect the DV other than the IV

examples of situational variables

Noise, temperature, smells, and lighting.

Reliability

consistent results

validity

accuracy - assume the test measures precisely what it aims to measure - meaning the data collected is accurate and represents some truth compared to others outside of the study

two main types of validity

internal and external

types of internal validity

face and concurrent

internal validity

Extent to which procedure/instrument is measuring what it intends to measure.

examples of internal validity

IQ tests, personality questionnaire

types of internal validity

Face validity and concurrent validity

What is face validity?

whether the instrument looks like it is measuring what it claims to measure

What is concurrent validity?

How well scores on one measure correlate with those on a related measure
•comparing new test with existing one to see whether they produce similar results...if new test does then concurrent validity

what is external validity?

the degree to which the investigator can extend or generalize a study's results to other subjects and situations.
• is it accurate representation of behaviour in real life?

what are the three types of external validity?

ecological, population, temporal

what is ecological validity

does the experiment relate to the real world?
can we generalise to the real world?

what is temporal validity?

can we generalise it to current time period? is it representative of modern day society?

Population validity

can it be generalised to target population
does it represent the whole population?

Two ways of assessing reliability

Test-retest
Inter-observer reliability

what is test-retest theory?

same method given to same ppts on two occasions, to see if same results obtained
(are same results obtained is process is repeated?)

What is inter-observer reliability?

The extent to which there is agreement between two or more observers involved in observations
must have 0.8+ correlation coefficient

Types of experiments

Laboratory, field, natural, quasi

Lab experiment

An experiment that takes place in a controlled environment where the researcher manipulates the IV and records the effect on the DV while maintaining strict control of extraneous variables.
•high control of variables
•manipulation of IV
•random allocation of ppts

strengths of lab experiment

•high control - manipulation of IV, standardisation, remove extraneous variables - increases internal validity
•Replicable - repeated easily due to high control - increases reliability - consistency

Weakness of a lab experiment

•low ecological validity (external validity)- doesn't reflect real-life - too artificial
•high control- more susceptible to demand characteristics- cues in environment lead to guessing aim - socially desirable - please experimenter

field experiment

•an experiment conducted in the participants' natural environment
•but IV is manipulated

Strength of field experiment

•High ecological validity- reflects real life-low control- own setting
•often without ppt awareness - act more naturally - less social desirability - increases internal validity

Weakness of field experiment

•not easily repeated -low control- not much standardisation - decreases reliability - can't find the same results again
•low internal validity - low control- no standardisation -decreases internal validity

natural experiment

no manipulation of IV
natural environment
naturall occurring

strength of natural experiment

•high ecological validity - reflects real life - low control
•less social desirability - ppt no awareness that they're being studied - increase internal validity
•can be used, in situations which would be unethical to manipulate the IV

weakness of naturak

•not easily repeated - low control- decreases reliability
•low internal validity - low control -no standardisation - decreases internal validity
•more time- consuming + expensive than lab

quasi-experiment

high in control/all variables controlled
IV- preexisting difference amongst/naturally occurring so can't naturally assign
IV's cant be randomly assigned

Strength of quasi experiment

•high control - standardisation -increases internal validity
•replicable- repeated easily due to high control increases reliability

Weakness of Quasi experiment

•low ecological validity -doesn't reflect real life - too artificial
•high control - more demand characteristics - socially desirable as easier to guess aim

types of experimental design

Independent measures, repeated measures and matched pairs

independent group design

ppts only go through one condition

repeated measures design

ppts go through BOTH conditions

matched pairs design

ppts matched based on similar characteristics (e.g. gender or IQ) then allocated to either condition (one from each pair does conditions A, second does condition B)

strengths of independent group designs

•no order effects - only going through one condition - ppts won't become bored, tired or practiced
•reduce demand characteristics - harder to guess aim as only doing one conditions - harder to guess aim

weakness of independent group design

•no control over ppt variables - individual differences
•require LOTS/MORE ppts

Strength of repeated measures

•ppt variables controlled - ppts take part in both conditions...experimenter comparing like with like (removes ppt variables) increasing internal validity
•fewer ppts needed
(removes ppt variables and extraneous variables...high internal validity)

Weakness of repeated measures

• increase order effects - ppts complete both conditions - can become bored/tired/get better from practice (practice effect)

how to deal with the weakness of independent group design

random allocation of ppts - reduces experimental bias + theoretically distributes variables evenly

how to deal with the weakness of repeated measure designs

counterbalancing- ensures each condition= tested first/second in equal amounts - does not remove order effects but attempts to balance out the effects of order between the two conditions

strengths of matched pair design

only take part in one condition - order effects = removed
demand characteristics less of an issue
lower individual differences and the effect they have

weakness of matched pair design

ppts can never be matched exactly! - ppt variables are only reduced! not removed!
time consuming- ppts need to complete a task before experiment in order to be matched

Sample

a group of people who take part in the study

sampling

process of gaining the sample

target population

the larger group of people that the experimenter is aiming the study at - sample taken from this group

representative

whether the sample reflects the target population

generalisation

The extent to which findings and conclusions from a particular experiment can be broadly applied to the population. This is possible if the sample of people is representative of the population.

what are the 5 types of sampling?

random, opportunity, volunteer, stratified, systematic

random sampling

every member of the target population has an equal chance of being selected

method of random sampling

each member of target population given number....put all numbers in a generator- select one number out until the desired sample number has been reached - ignore any repeats

Strengths of random sampling

highly representative- every member = equal chance of being selected - good as removes experimenter bias - higher population validity

Weakness of random sampling

Cannot guarantee the sample is totally representative of the target population - by chance could still be bias
time consuming

systematic sampling

picking the nth number from a list of names or numbers

strength of systematic sampling

removes experimenter bias - higher population validity - more representative

weakness of systematic sampling

by chance - sample can still be biased - therefore less representative lower population validity

volunteer sample

consists of people who are willing to volunteer for a study - ppts choose to take part

strength of volunteer sample

Easy to collect - quick
Participants are more engaged since they want to be there.
​

weakness of volunteer sample

Volunteer bias is a problem asking for volunteers may attract a specific type of person who is keen to please the researcher. This may affect how findings can be generalised.

opportunity sample

A sampling technique where participants are chosen because they are easily available + willing to take part

strength of opportunity sample

quick and easy - data will be collected at a faster rate

weakness of opportunity sample

-experimenter bias - researcher could ask ppts who they think would prove the hypothesis - reduce population validity + make sample less representative
- ppts will be collected from same place which means ppts = likely to have similarities - which = not reflective of whole target population

order effects

A extraneous variable arising from the order which Ps take part in all conditions, so their performance improves/worsens across conditions through practice/fatigue.

way to reduce order effects

counter balancing - reduce fatigue/boredom

participant variables

Individual differences in the personal characteristics of research participants that, if not controlled, can confound the results of the experiment.

way to reduce participant variables

matched pairs

situational variables

features in the environment that if not controlled may alter results of the experiment

way to reduce situational variables

lab experiments - have minimal situational variables

investigator effects

Any effect of the investigator's behaviour (conscious or unconscious) on the research outcome (the DV).
This may include everything from the design of the study to the selection of, and interaction with, participants during the research process.

way to reduce investigator effects

standardised instructions
double-blind technique

pilot study

-> small scale trial run study
-> before the real thing
to check the investigation runs smoothly - solve methodological problems/gain feedback e.g. ppts guessing aim/not understanding task/remove extraneous variables

what are the 6 ethical guidelines according to the BPS (British Psychological Society)?

Deception
Informed Consent
Protection of ppts
Confidentiality
Right to withdraw
Debrief

protection of ppts

ppts cannot be subject to any:
psychological harm or physical harm

solution of protection of ppts

debriefing
opportunity of sampling

Confidentiality

keeping data safe/protected:
ppts results must remain anonymous
and must be kept secret/locked away at all times

solution of Confidentiality

encrypted data/password protected data
no names + use of numbers or initials instead