Research Methods

aqa a A2 psychology unit 4

HideShow resource information
  • Created by: lauren
  • Created on: 13-06-12 15:23

Features of the scientific method

Empiricism - information is gained through direct observation or experiment rather than by reasoned argument or unfounded beliefs

Objectivity - scientists strive to be objective in their observations and measurements - their expectations should not affect what they record

Replicability - to demonstrate the vailidity of an experiment is to repeat it - if the outcome is the same - affirms the truth of the original results - for replication to happen its important scientists record down the methods they used so that the same procedure can be used again

Control - seek to demonstrate a causal relationship to enable them to predict and control our world - experimental way is the only way to do this - we vary one factor (independant variable) and observe its affect on the dependant variable - all other conditions must be kept the same

Theory construction - use the facts to construct theories to help us understand and predict the natural phenomena around us - a theory is a collection of general principles that explain observations and facts

1 of 30

The scientific process

Induction - reasoning from the particular to the general - scientist may observe instances of a natural phenomenon and come up with a general law or theory - based on formulating theories from observation

Deduction - reasoning from the general to the particular - starting with a theory and looking for instances which confirm it - falsification is the only way to be certain - Popper 'no amount of observations of white swans can allow the inference that all swans are white, but the observation of a single black swan is sufficient to refute that conclusion

2 of 30

Application of the scientific method in psychology

Scientific research is desirable - enables them to produce verifiable knowledge as distinct from common sense or 'armchair psychology' - scientific research provides truth rather than just assumptions

Psychology is a science insofar as it shares the goals of all sciences and uses the scientific method - psychologists generate models which can be falsified and conduct well-controlled experiments to test these models - question to whether simply using the scientific method turns psychology into a science - Miller - psychologists who attempt to be scientists are doing no more than dressing up - take on the tools of sciences such as quantified measurements and statistical analysis but the essence of science has eluded them - psuedoscience - but psychologists can claim their discoveries are a fact

Kuhn's views - psychology couldnt be a science because unlike other sciences there is no single paradigm (shared set of assumptions) - biology or physics has a unified set of assumptions - psychology has a number of paradigms or approaches -psychology is a pre-science argues Kuhn

3 of 30

Application of the scientific method in psychology

lack of objectivity and control - some psychologists - human behaviour can be measured as objectively as physical objects but in psychology the object of study reacts to the researcher which leads to problems such as experimenter bias and demand characteristics which compromise validity  - but similar problems arise to the hard sciences - Heisenberg - not even possible to measure a subatomic particle without altering its behaviour in doing the measurement - experimenter effect - presence of an experimenter changes the behaviour of what is observed

4 of 30

Application of the scientific method in psychology

Science takes the nomothetic approach - looking to make generalisations about people and find similarities - the way to decide whether science is appropriate is to look at the results of psychological research - psychological approaches to treating mental illness have had a modest success which suggets that the goals of science are not always appropriate

Qualitative research - some psychologists advocate more subjective, qualitative methods of conducting research but these methods are still scientific insofar as they aim to be valid - data can be collected from interviews, discourse anaylsis, obersavtions etc and then triagulated - findings from these different methods are compared with each other as a means of verifying them and making them objective

5 of 30

Reductionism and determinism

Scientific approach is reductionist and deterministic - reductionist because complex phenomena are reduced to simple variables in order to study the causal relationship between them - reductionist in the development of theories - simpler theories are preferred - deterministic in its search for causal relationships - if we dont take a determinsitic view of behaviour this rules out scientific research as a means of understanding behaviour

Reductionism and determinism are mixed blessings - if we reduce complex behaviour to simple variables this may tell us little about real behaviour but without this reductionism it is difficult to pick out any patterns or reach any conclusions - dterminsims - also oversimplifies the relationship between casues and effects but it provides an insight into important factors such as the influences of nature and nurture

6 of 30

Validating new knowledge

Peer review - (refereeing) - the assesment of scientific work by others who are experts in the same field (peers) - intention of peer review is to ensure that any research conducted and published is of a high quality - peer reviewers generally unpaid - number of reviewers for each application/article/assesment - task is to report on the quality of the research and then their views are considered by a peer review panel

Parliamentary office of science and technology - peer review 3 main purposes - Allocation of research funding - Publication of research in scientific journals and books - Assessing the research rating of university departments

Peer review and the internet - sheer volume and pace of information available on the internet means that new solutions are needed in order to maintain the quality of information - scientific information is available in numerous online blogs, online journals and wikipedia - large extent the sources of information - readers decide whether it is valid or not and post comments or edit entries accordingly 

7 of 30

Peer review commentary

Clear why peer review is essential - without it we wouldnt know what is opinion and what is fact - we need to have a means of establishing the validity of scientific research - certain features of peer review is criticised - unachievable ideal - isnt always possible to find an appropriate expert to review a research proposal or report - poor research may be passed because the reviewer didnt really understand it - anonymity - usually practised so that reviewers may be honest and objective - but it may have the oppostie effect  - if reviewers use the veil of anonymity to settle old scores or bury rival research - research conducted in a social world where people complete for research grants and jobs and make friends and enemies - social relationships affect objectivity - some journals now favour open reviewing - publication bias - peer review tends to favour the publication of positive results because editors want research that has important implications in order to increase the standing of their journal - preserving the status quo - peer review results in preference for research that goes with exisitng theory rather than dissenting or unconvetional work - science generally resistant to large shifts in opinion and change takes a long time and requires a revolution in the way people think - peer review may slow down change in scientific theories as it favors research that backs up old theories

8 of 30

Conventions for reporting psychological investigat

Scientific journals contain research reports which tend to be organised into the following sections - Abstract - summary of the study covering the aims/hypothesis, method/procedures, results and conclusions - Introduction/aim - what the researchers intend to investigate - includes review of previous research, may state their hypothesis and research predictions - Method - detailed description of what the researchers did providing enough information for replication of the study - information about pp's (sample) the testing environment, procedures used to collect data, any instructions given to pp's before and afterwards (brief and debreif) - Results - what the researchers found, statistical data which includes descrptive statistics and inferential statistics - Discussion - offer explanations of the behaviours they observed and might also consider the implications of results and make suggestions for future research - References - full details of any journal articles or books that are mentioned

9 of 30


All experiments involve an IV and a DV - the IV is varied (manipulated) in order to see its affects on the DV - demonstrating a causal relationship - as far as possible all other variables are controlled so any changes in the DV are due to the IV rather than extrenuous variables

Lab experiments - controlled environment - high internal validity as many EV's can be controlled - but some (investigator/experimenter effects and demand characteristics) may reduce internal validity - control increases replicability but reduces external validity as its less likely to represent real life where EV's are in play

Field experiments - experiment conducted in a more natural environment - may be possible to control EV's although its more difficult to control them than in a lab exp - experimenter effects are reduced as pp's are usually not aware of being in a study - demand characteristics may still be problematic as the way the IV is operationalised may convey the experimenteral hypothesis to pp's

10 of 30


Natural experiment - makes use of existing IV's - such as treatment used for people with schizophrenia - experiment involves the deliberate manipulation of the UV by the experimenter so causal relationships cannot be drawn from a natural experiment - pp's are not randomly allocated which may reduce validity but its often the only way to study certain behaviour or experiences such as the effects of privation

Experimental design - several levelsof the IV - experimenters have a choice - either each pp is tested on all of the IV's (repeated measures) or there are separate groups for each IV (independant groups) or the pp's in each independant group can be matched with pp's in the other grou on key variables such as age and IQ (matched pairs design)

11 of 30

Self-report methods

Questionnaires and interviews - find out what people think and feel - interviews are ressentially real-time, face to face questionnaires though theres an option to conduct a fairly ustructured interview where the questions are developed by the interviewer as the answers are given by the interviewee - structured interviews/questionnaires are more easily repeated in exactly the same way

Main problem for self-report methods is honesty - social desireability bias means that respondents may provide answers to put themselves in good light which aren't exactly true

Questionnaires and interviews mat involve open questions which permit a respondent to provide their own answer - can produce unexpected answers providing rich insights but they are more difficult to analyse than closed questions

12 of 30

Observational studies

simply watch what people do - observational studies - not that simple to observe behaviour as there is so much information to collect - so they use behavioural categories to record particular instances of behaviour and also sampling methods such as recording behaviour every 30 seconds - time sampling - every time a certain behaviour occurs - event sampling - even in naturalistic observations structured techniques are used to study behaviour - observational studies provide a rich picture of what people actually do rather than what they say they do - observers may be biased (observer bias) - where their observations can be effected by their expectations

13 of 30

Correlational analysis

Some studies concerned with the relationship between two variables such as IQ and Alevel results or reaction time and age - some studies use correlational analysis which doesnt demonstrate a causal relationship but is useful in identifying where relationships between co-variables exist - can be done with large sets of data and can be easily replicated - but other unknown variables (intervening) that can explain why the co-variables being studied are linked - may lack internal/external validity 

14 of 30

Case studies and other research methods

detailed study of an individual, institution or event - uses information from a range of sources such as from the person concerned and from their family and friends - many techniques may be used such as interviews psychological tests observations and experiments - usually longitudinal - complex interaction of many factors can be studied - difficult to generalise from individual cases as each one has unique characteristics - often necessary to use a recollection of past events as part of the case study and this info may be unreliable

numerous other methods - content anaylsis (kind of observational study) - cross-cultural research (comparing effects of different cultural practices on behaviour) - meta-analysis (combines results of many studies on the same topic to reach overall conclusions

15 of 30

Few other concepts

Aims and hypotheses - researchers start by identifying what they intend to study (the aims) and then make a formal statement of their expectations using a hypothesis. Hypothesis may be directional and non-directional. Good hypothesis should be operationalised so that the variables are in a form that can usually be tested

Investigator effects and other problems - investigators may communicate their expectations unwittingly to pp's so leading pp's to fulfil the investigators expectations - demand characteristics and social desireability bias are other problems

Pilot study - small-scale trial run of a  research study to test any aspects of the design, with a view to making improvements

16 of 30

Issue of reliability

Experimental research - reliability refers to the ability to repeat a study and obtain the same result - replication - essential that all conditions are the same otherwise any change in the result may be due to changed conditions

Observational techniques - observations shuld be consistent - ideally two or more observers should produce the same record - extent to which the observers agree i called inter-rater or inter-rater reliability - calculated by dividing total agreements by the total number of observations - a result of 0.80 or more suggests good inter-rater reliability - can be improved by training observers in the use of a coding system/behaviour checklist

Self-report techniques - 2 different types of reliability ivolved with self-report techniques - internal reliability - measure of the extent to which something is consistent within itself - external reliability - measure of the consistency over several different occasions 

Reliability also concerns whether two interviewers produce the same outcome - inter-interviewer reliability - split-half method (compare a persons performance on two halves of a questionnaire or test - close correlation - meaures internal reliability - test-retest method - person given a questionnaire etc on one occasion then its repeated again after a reasonable interval - measure is reliable if the outcome is the same

17 of 30

Issue of validity

Internal validity - whether the researcher did test what he intended to test

External validity - extent to which the results of the study can be generalised to other situations and people - ecological validity

Experimental research - internal validity often affected by EV's which may act as an alternative IV - changes in the DV are due to EV's rather than the IV and so conclusions about the effect of the IV on the DV are erroneous - important to consider issues to whether the pp's knew they were being studied and whether the task itself was aritifical rather than the setting and so low in mundane realism which reduces the generalisability of the results

Observational techniques - internal validity - observations will not be valid if the coding system/behaviour checklist is flawed - some behaviour may belong in more than one category which reduces internal validity of the data collected - internal validity also affected by observer bias - what someone observes is often influenced by what they expect - reduces objectivity of results - likely to have high ecological validity as they involve more natural behaviours but naturalistic research is not necessarily higher in ecological validity

18 of 30

Sampling techniques

Use sampling techniques to minimise cost while maximising generalisability 

Opportunity sample - pp's selected by using those people who are most easily available - easiest method to use but is biased as the sample is drawn from a small part of the target population

Volunteer sample - pp's selected by asking for volunteers - placing adveritsements in newspapers - method can access a variety of p's if the advertisement is in a national paper - sample would be representative - but samples are biased because pp's are likely to be highly motivated and/or with extra time on their hands

Random sample - pp's selected using random number technique - all members of target population identified then individuals are selected either by the lottery method or by using a random number generator - potentially unbiased as all members of target population have an equal chance og being selected  - though may be biased as some people may refuse to take part

19 of 30

Sampling techniques

Stratified and quota samples - sub-groups (strata) within a population are identified - then a predetermined number of pp's is taken from each sub-group in proportion to their representation in the target population - done by using random techniques, in quota sampling its done by opportunity sampling - method more representative than other methods as here is a proportional respresentation of sub-groups - but selection in each sub-group may be biased because of opportunity sampling

Snowball sampling - difficult to identify suitable pp's - study of eating disorders, start with 1 or 2 people with eating disorders then ask them to direct you to some other people with these problems and so on - useful when conducting research with pp's who are not easily indentified but prone to bias as researchers may only contact people within a limited section of the population

20 of 30

Ethical issues with human participants

Informed consent and deception - issue arises as full information may compromise the integrity of a study as knowing the full amount of information may affect the pp's behaviour making the results meaningless - harm and what constitues too much harm 

Code of conduct - BPS code of ethics and conduct identifies four ethical principles - Respect for the dignity and worth of all persons - standards or privacy, confidentiality and informed consent - intentional deception is only acceptable when it is necessary to protect the integrity of research and when the nature of the decpetion is disclosed to pp's at the earliest opportunity - one way to judge deception is to consider whether pp's are likely to object or show unease when debriefed in which deception may be judged unacceptable - should be aware of the right to withdraw from the research at any time 

Competence - psychologists should maintain high standards in their professional work


21 of 30

Ethical issues with human participants

Responsiblity - have a responsibility to their clients, to the general public and to the science of psychology - protecting participants from physical and psychological harm as well as debriefing and to identify any unseen harm or to give them assistance if needed

Integrity - should be honest and accurate - reporting the findings of any research accurately and ackowledging any potential limitations - bringing instances of misconduct by other psychologists to the attention of the BPS

Dealing with ethical issues - code of conduct offers ethical guidelines for psychologists to follow - psychologists deal with ethics by using ethical committees to assess research proposals by punishing psychologists who contravene with the code with disbarment from the soceity

22 of 30

Ethical issues with non-human animals

Animals may be chosen to study because: they are fascinating to study in their own right - animals offer the opportunity for greater control and objectivity in research procedures - use animals when we cant use humans - animals have been exposed to various procedures and events that would not be possible with humans such as Harlows monkey experiment - humans and non-human animals have enough of their psychology and evolutionary past in common to justify conclusions drawnf from experimenting one or the other

23 of 30

Moral justification

Sentient beings - in terms of pain there is evidence that animals respond to pain but it may not be the same conscious awareness as humans - some evidence that animals other than primates have self-awareness - brain-damaged humans lack sentience but wouldnt be used in research without informed consent

Speciesism - discrimination on the basis of species is no different from racial or gender discrimination - suggests use of animals is an example of speciesism - gray argues we have a special duty of care to humans and speciesism isnt equivalent to racism

Animal rights - whatever produces the greater good for the greater number is ethically acceptable - so if animal research can alleviate pain and siffering it is justifiable

24 of 30

Existing constraints for animal rights

BPS published guidelines for research with animals but there is a legislation too - in the UK an axct requires that animal research only takes place in licensed labs with licensed researchers on licesenced projects - licensces only granted if - potential results are important enough to justify the use of animals - the research cannt be done using non-animal human methods - minimum nuber of animals will be used - any discomfort or suffering is kept minimal by appropriate use of painkillers or anaesthetics


25 of 30

Inferential analysis, probability and significance

Probability - inferential statistics allow psychologists to draw conclusions based on the probability that a particular pattern of results could have arisen by chance - if it could have arisen by chance then it would not be correct to conclude - if it could not have arisen by chance or if its extremely unlikely to have arisen by chance then the pattern is described as significant

Significance - in order to use an inferential test we need a null hypothesis and an alternative hypothesis - what we are looking for is a significant different between samples to be sure that they are actually different - otherwise we assume the populations are the same and accept the null hypothesis

Chance - Psychologists use a probability of less than 0.05 which means that there is less than or equal to a 5% probability that the results are due to chance - when researchers want to be more certain they use a probability of less than 0.01 or even 0.001 - in other studies 0.10 may be used - chosen level of probability is called the significance level

26 of 30

Levels of measurement

when deciding which test to use you may need to identify the level of measurement that was used - 

Nominal - data are in separate categories such as grouping you class into people who are tall, medium or short

Ordinal - data are ordered in some way - lining up your class mates in order of height - the difference between each item is not the same

Interval - data are measured using units of equal intervals such as when counting correct answers or measuring your classmates heights

27 of 30

Using inferential statistical tests

Different tests - used for different research desgins - looking at correlation between 2 variables then you use spearmans rho test

Observed and critical values - taking the data collected and doing some arithmetical calculations which produce a single number called the test statistic - in spearmans rho - test statistic called rho - rho value is the observed value - to decide if observed value is significant the figure is compared to the critical value and this is the value that a test statistic much reach in order for the null hypothesis to be rejected - to find appropriate critical value you need to know

degrees of freedom (df) - looking at the number of pp's in the study (N)  - in chi sqaure you calculate the degrees of freedom based on the number of cells there are 

one-tailed or two-tailed hypothesis - if the hypothesis was directional you use the one-tailed test, if it was non-directional you use the two-tailed test

significance level selected - probability usually less than 0.05

28 of 30

Importance of R

some tests are significant when the observed value is equal to or exceeds the critical value - for others its the reverse - if there is a letter R in the test then the observed value should be greater than the critical value - if there is no R then the observed value should be less than the critical value

type 1 error - rejecting a null hypothesis that is true

type 2 error - accepting a null hypothesis that is in fact not true

29 of 30

Tests - acronym to remember which test to use

                                  no (nominal)                         * starts --> charlie (chi square)

                     **** (correlational)                                          sucks (spearmans rho) 

inside (interval)                                                    mens (mann-whitney U)

 ok (ordinal)                                                   willys (wilcoxon T)


30 of 30



thanks :) 

Similar Psychology resources:

See all Psychology resources »See all Research methods and techniques resources »