Research Methods

Aim

General statement of what the researcher intends to investigate, the purpose of the study

Hypothesis

A clear, precise, testable statement that states the relationship between the variables to be investigated. Stated at the outset of any study.

Use a directional hypothesis when

a theory or findings from previous research suggest a particular outcome

Use an nondirectional hypothesis when

there is no theory or previous research, or findings from earlier studies are contradictory

Extraneous variables

Any variable, other than the IV that may affect the DV if it is not controlled (e.g. age of ppts or lighting in the lab) => they do not vary systematically with the IV => they may or may not actually affect the results

Confounding variables

A specific type of extraneous variable => DO vary systematically with the DV => therefore we can’t tell if the change in the DV is due to the IV or the confounding variable => e.g. does excising improve health? - people who exercise more may also have a healthier diet - therefore we cant tell if the improvement comes from exercising itself or healthier food habits

Demand characteristics

Any cue from the researcher or from the research situation that may be interpreted by the ppts as revealing the purpose of the investigation => this may lead to them behaving in a way they think is expected of them (‘Please-U effect) or in a way to sabotage the results (‘Screw-U effect’)

Investigator effects

Any unwanted influence of the investigator on the research outcome

Randomisation

• Controls researcher effects

• = the use of chance methods to reduce the researcher’s unconscious biases when designing an investigation + deciding the order of experimental conditions

• e.g. order of word list (for a memory investigation) should be randomly generated so that the position of each word is not determined by the investigator

Standardisation

Using exactly the same formalised procedures and instructions for all ppts

Independent groups

When 2 separate groups of ppts experience 2 different conditions of the experiment

Strengths:

• Order effects are not a problem

• Ppts less likely to guess aim of study

Limitations:

• The ppts in the different groups are not the same - so differences in the DV may be due to ppt variables rather than the effects of the IV - these differences may act as a confounding variable, reducing the validity => deal with this using random allocation => ppts randomly allocated to the different experimental conditions - e.g. pieces of paper with A or B written on them are put in a hat and researcher selects them one at a time to assign ppts to groups

• More money/ time spent on recruiting ppts

Repeated measures

all ppts experience both conditions of the experiment

Strengths:

• Ppt variables are controlled

• Less time/money spent on recruiting ppts

Limitations:

• Order effects => may cause boredom or fatigue - ppt’s performance may improve or deteriorate on the second task => deal with this using counterbalancing => half the ppts take part in condition A then B and the other half take part in condition B then A (ABBA technique)

• Demand characteristics

Matched pairs

ppts are paired together on a variable relevant to the experiment => e.g. in a memory study ppts may be matched on their IQs => this is in an attempt to control for the confounding variable of ppt variables

Strengths:

• No order effects

• No demand characteristics

• Participant variables controlled to a degree

Limitations:

• However, ppts can never be matched exactly (even twins have important differences between them that may affect the DV)

• Matching can be time-consuming + expensive

Lab experiment

• Takes place in controlled environment within which the researcher manipulates the IV and records the effects on the DV, whilst maintaining strict control of extraneous variables

Strengths:

• High control over extraneous + confounding variables => means that researcher can be sure that any effect on the DV is due to manipulation of the IV => high internal validity

• Can be easily replicated => vital in checking the validity of the result (so we can be sure it is not just a one-off)

Limitations:

• Lacks generalisability =>artificial, not like every-day life => in unfamiliar context ppts may behave in unusual ways so their behaviour cannot be generalised beyond the research setting => low external validity

• Demand characteristics => know they are being tested in a lab experiment

• Low mundane realism => tasks do not reflect everyday experience

Field experiment

• IV is manipulated in a natural everyday setting

Strengths:

• Higher mundane realism => thus produces behaviour that is more valid + authentic => may be case that ppts don’t know they’re being studied (high external validity)

Limitations:

• Loss of control of extraneous/confounding variables => means that cause and effect between the IV and the DV is harder to establish

• Precise replication impossible

• Ethical issues => if ppts don’t know they’re being studied they cannot give consent

Natural experiment

• Researcher has no control over the IV and can’t change it - someone or something else causes the IV to change (e.g. before and after a natural disaster)

Strengths:

• Provide opportunities for research that may not otherwise be undertaken for practical or ethical reasons (e.g. study of institutionalised orphans - Rutter)

• High external validity => involve study of real-world issues and problems as they happen (e.g. effects of natural disaster on stress levels)

Limitations:

• Naturally-occurring event may only happen rarely => limits opportunity for research

Quasi-experiment:

• IV based on an existing difference between people (e.g. gender)

Strengths:

• Often carried out under controlled conditions => replication

Limitations:

• Confounding variables => cannot randomly allocate ppts to conditions

• IV not deliberately changed by researcher => so can’t claim that IV has caused any observed change

Population

A group of people who are the focus of the researcher’s interest

Sample

A group of people who take part in an investigation. Sample is drawn from a target population and is presumed to be representative of that population

Random sample:

• All members from a target population have an equal chance of being selected

• Obtain complete list of all members of the target population

• All names on list are assigned a number

• The actual sample is selected through the use of some lottery method (e.g. picking names out of a hat)

Strengths:

• Unbiased => means that confounding or extraneous variables should be equally divided between the different groups => enhances internal validity

Limitations:

• Difficult + time-consuming => complete list of the target population difficult to obtain

• Whilst laws of probability suggest that it is likely that random sampling will choose a representative sample, it is possible that the random method may not actually be (e.g. may pick 20 female doctors from Twickenham called Sarah)

• Ppts may refuse to take part (means you end up with something more like a volunteer sample)

Systematic sample:

• When every nth member of the target population is selected (e.g. every 5th student from a school register)

• A sampling frame is produced which is a list of people in the target population organised, for example, into alphabetical order

• A sampling system is nominated (e.g. every 3rd person)

• Researcher works through sampling frame until the sample is complete

Strengths:

• Objective => once the system of selection has been established, the researcher has no influence over who has been chosen

Limitations:

• Time consuming

• Ppts may refuse to take part (end up with more of a volunteer sample)

Stratified sample:

• When composition of sample reflects the proportions of people in certain subgroups (strata) within the target population or the wider population

• Researcher identifies the different strata that make up the population

• Then, the proportions needed for the sample to be representative are worked out

• Finally, ppts that make up each stratum are selected using random sampling

Strengths:

• Produces representative sample as it is designed to accurately reflect the composition of the population => generalisation of findings becomes possible

Limitations:

• Stratification not perfect => the strata cannot reflect all the ways people are different => so complete representation of the target population is not possible

Opportunity sample:

• When researchers decide to select anyone who happens to be willing and available

• Researcher asks whoever is around at their time of study (e.g. in the street)

Strengths:

• Convenient + less costly (time/money)

Limitations:

• Sample unrepresentative of target population as it is drawn from very specific area (e.g. one town) => findings cannot be generalised to target population

• Researcher has full control over selection of ppts => may avoid people they don’t like the look of (researcher bias)

Volunteer sample:

• Ppts select themselves to be part of the sample

• E.g. researcher may place ad in newspaper or noticeboard

Strengths:

• Less time consuming => ppts come to researcher

• Ppts more engaged

Limitations

• Volunteer bias => asking for volunteers may attract certain types of people => e.g. people who are more curious or who are more likely to please the researcher => affects how far findings can be generalised

Ethical issues

• Arise when a conflict exists between the rights of ppts in research studies and the goals of research to produce authentic, valid and worthwhile data

• E.g. a researcher may not want to reveal the true purpose of a research study to ppts in order to study more ‘natural’ behaviour, but this raises the concern about whether it is acceptable to mislead ppts in this way

Informed consent

Making ppts aware of…

• aims of research

• the procedure

• their rights (including the right to withdraw)

• what the data will be used for

Deception

• Deliberately misleading or withholding information from the ppts

• Occasions where deception can be justified if it does not cause the ppt undue stress

Protection from harm

• Ppts should not be put at anymore risk than they would be in their everyday lives

• Should be protected from physical or psychological harm => includes being made to feel embarrassed or being put under a lot of stress

• Important feature => ppts reminded they have the right to withdraw

Privacy and confidentiality

• Ppts have right to control info about themselves => this is the right of privacy

• Right of privacy extends to the area the study took place => e.g. institutions or geographical locations are not named

BPS code of conduct

• Set of ethical guidelines

• Researchers have a professional duty to observe these guidelines when conducting research

• Guidelines closely match the ethical issues

• Attempt to ensure that all ppts are treated with respect and consideration during the research

• Guidelines are implemented by ethics committees in research institutions who often use a cost-benefit approach to determine whether particular research proposals are ethically acceptable

Dealing with informed consent

• Ppts get a consent letter or form

• Details all the relevant info that might affect their decision to participate

• If agree => sign

• For investigations involving children under 16, a signature of parental consent is required

Alternative ways of getting consent:

Presumptive consent

• Rather than getting consent from the ppts themselves, another group of similar ppts is asked if the study is acceptable

• If group agrees, then consent of the og ppts is presumed

Prior general consent:

• Ppts give their permission to take part in a number of different studies (including one that involves deception)

• By consenting they are consenting to being deceived

Retrospective consent:

• Ppts asked for consent during debriefing (so have already taken part in study)

• May not be aware of their participation or that they may have been subject to deception

Dealing with deception and protection from harm

• At the end, ppts should be given a full debrief

• Ppts made aware of the true aims of the investigation + any details they weren’t given at the start (e.g. existence of other conditions)

• Told what data will be used for + right to withdraw from study + right to withdraw data

• In extreme cases, if ppts have been subject to stress or embarrassment, they may require counselling which the researcher should provide

Dealing with confidentiality

• Maintain anonymity => researchers usually refer to ppts using numbers of initials

• Case studies => often use initials when describing individual involved in study

• Standard practice during briefing and debriefing that ppts are reminded that their data will be protected throughout the process and that it will not be shared with other researchers

Cost-benefit analysis

Benefits => the value or the ground-breaking nature of the research

Costs => damaging effects on individual ppts or reputation of psychology as a whole

Pilot study

• A small-scale version of an investigation that takes place before the real investigation is conducted

• Involves only a handful of ppts rather than the total number

• Aim => to check that procedures + materials work => allows researcher to identify any potential issues + to modify the design or structure (saves time and money in the long run)

Single-blind procedure

= ppts unaware of the test being conducted

• Ppts not knowing the aim of the research

• Ppts not knowing which condition of the experiment they are in or whether there is another condition at all

• => controls for the confounding effects of demand characteristics

Double-blind procedure

= Ppts and researcher unaware of the test being conducted

• Neither the ppts nor the researcher know the aim of the study

• Often important features of drug trials => treatment administered by someone independent of the investigation + who doesn’t know which drugs are real or placebo

Naturalistic observation

• Take place in setting or context where target behaviour would usually occur

Strengths:

• High external validity => findings can be generalised to everyday life

Limitations:

• Lack of control over the research situation makes replication of the investigation hard

• May uncontrolled extraneous/confounding variables that make it more difficult to judge any pattern of behaviour

Controlled observations

• Some control over variables - including manipulating variables to observe effects + control confounding/extraneous variables

Strengths:

• Extraneous/confounding may be less of a factor so replication becomes easier

Limitations:

• Cannot generalise to everyday life

Covert observation

Ppts are unaware they are the focus of the study + that their behaviour is being observed in secret

Strengths:

• Removes demand characteristics => increases internal validity

Limitations:

• Ethics => people may not wish to have their behaviour noted down - some behaviour is private (e.g. how much money spent on shopping)

Overt observation

Ppts know their behaviour is being observed + they have given informed consent

Strengths:

• More ethically acceptable

Limitations:

• Demand characteristics => low internal validity

Ppt observations

• Observer becomes part of the group they are studying (e.g. becomes a cult member)

Strengths:

• Increased insight into ppts lives => high external validity

Limitations:

• Researcher may come to identity too strongly with those they are studying and lose objectivity => line between researcher and ppt becomes blurred

Non-ppt observations

• Researcher remains separate from those they are studying and records behaviour in a more objective manner

• May be impractical or even impossible to join particular groups (e.g. middle aged female researcher observing behaviour among Year 10 boys)

Strengths:

• Objective psychological distance from ppts => so less danger of them adopting a local lifestyle

Limitations:

• May lose the valuable insight to be gained in a ppt observation as they are too far removed from the people + behaviour they are studying

Evaluation of all observations

Strengths:

• Capture what people actually do (may include unexpected behaviour)

Limitations:

• Observer bias => observer’s interpretation of a situation may be affected by their expectations (however, this may be reduced by using more than one observer)

• Cannot determine causal relationships

Structured observation

• Researcher simplifies the target behaviours that will become the main focus of the investigation using behavioural categories

• There may be too much going on in an investigation for a researcher to record it all

Strengths:

• Make the recording of data easier and more systematic

• Data likely to be numerical (quantitative data) => means that analysing and comparing the behaviour observed between ppts is more straightforward

Limitations:

• No richness or depth of detail in the data collected

Unstructured observation

• Researcher writes down everything they see

• Produces accounts of behaviour that are rich in detail

• May be appropriate when investigations are small in scale and involve few ppts

Strengths:

• Richness + depth of detail in the data collected

Limitations:

• Qualitative data => more difficult to record and analyse

• Increased risk of observer bias (as there are no behaviour categories) => observer may only record behaviour that ‘catches their eye’

Behavioural categories

• = a checklist of the target behaviour

• Target behaviours to be studied should be precisely defined and made observable and measurable (e.g. target behaviour ‘affection’ can be broken down into categories such as kissing, hugging etc.) => clear and unambiguous as possible

• There should be no need for inferences to be made (e.g. being loving) as 2 observers may interpret this differently making it unreliable

• Behaviour categories should not overlap (e.g. difference hard to tell between ‘smiling’ and ‘grinning’)

Event sampling

= counting the number of times a particular behaviour occurs in a target individual or group (e.g. event sampling of dissent at a football match would mean counting the number of times a player disagrees with the ref)

Strengths:

• Useful when target behaviour happens infrequently and so could be missed if time sampling was used

Limitations:

• However, if the specified event is too complex, then the researcher might overlook important details if using event sampling

Time sampling

= recording behaviour within a pre-established time frame (e.g. recording behaviour every 30 seconds)

Strengths:

• Effective in reducing the number of observations that have to be made

Limitations:

• That said, those instances when behaviour is sampled might be unrepresentative of the observation as a whole

Inter-observer reliability

• Recommended that researchers do not carry out observational studies alone (single observers may miss important details or only notice events that confirm their opinions or hypothesis - bias)

• 2 observers => makes research more objective and unbias

• Data from different observers is compared to check for consistency

How to do inter-observer reliability:

• Observers should familiarise themselves with behaviour categories

• Then, observe behaviour at same time (perhaps as part of a small-scale pilot study)

• Compare data recorded and discuss any differences in interpretation

• Analyse data from the study - inter-observer reliability is calculated by correlating each pair of observations made and an overall figure is produced

Evaluation of questionnaires

Strengths:

• Cost-effective => can gather large amounts of data quickly because they can be distributed to large numbers of people

• Can be completed without researcher present => reduces effort involved

• Data that they produce is usually straightforward (particularly the case if questionnaires have fixed-choice closed questions) => the data lends itself to statistical analysis, and comparisons between groups of people can be made using graphs and charts

Limitations:

• Responses not always truthful => ppts may want to present themselves in a positive light which may influence their answers - social desirability bias

• Questionnaires can produce a response bias (when people respond in a similar way e.g. always ticking ‘yes’ => may be because they read it too quickly and don’t read the qs properly - a particular form of response bias is acquiescence bias (the tendency to say ‘yes’)

Open and closed questions

Open:

• Does not have a fixed range of answers

• Respondents free to answer in any way they wish

• Produce qualitative data (but may be difficult to analyse)

Closed:

• Offers a fixed number of responses

• We might restrict them to 2 options (e.g. ‘yes’ or ‘no’) or rate on a scale (quantitative data) => easier to analyse but lacks the depth and detail of open questions

• NOTE: Closed questions that produce qualitative data (e.g. ‘yes’ or ‘no’) can be turned into qualitative data (e.g. counting the number of ‘yes’ or ‘no’ responses)

Interviews

Structured:

• Made up of a pre-determined set of questions that are asked in a fixed order

Unstructured:

• Works like a conversation

• No set questions

• General aim that a certain topic will be discussed, and interaction tends to be free-flowing

• Interviewee encouraged to expand and elaborate

Semi-structured:

• List of questions that have been worked out in advance but interviewers are also free to ask follow-up questions based on previous answers

Evaluation structured interview

Strengths:

• Straightforward to replicate due to standardised format - this also reduces differences between interviewers

Limitations:

• Not possible for interviewers to deviate from the topic or explain their questions as this will limit the richness of the data as well as limit unexpected information

Evaluation unstructured interview

Strengths:

• More flexibility => interviewer can follow-up points as they arise => gain insight into the worldview of the interviewee, including eliciting unexpected information

Limitations:

• Increased risk of interviewer bias

• Analysis of data not straightforward => researcher may have to sift through much irrelevant information + drawing conclusions can be difficult

Designing interviews

• Most interviews involve an interview schedule (the list of questions the interviewer intends to cover)

• This should be standardised => to reduce interviewer bias

• Interviewer will take notes throughout interview or it will be recorded and analysed later

• Quiet room away from other people

• Begin with some neutral questions to build rapport + make interviewee feel more relaxed/comfortable

• Interviewees should be reminded that their answers will be treated with the strictest confidence

Writing good questions

Overuse of jargon:

• the best questions are simple and easily understood

Emotive language and leading questions:

• Emotive language should be replaced with neutral alternatives

• Leading questions guide the respondent towards a particular answer

Double-barrelled questions and double negatives:

• DB questions contain 2 questions in one - so the respondent might agree with one and disagree with the other

• Double negatives can be confusing

Correlation

A mathematical technique in which a researcher investigates an association between 2 variables, called co-variables

Positive correlation

As one co-variable increases so does the other

Negative correlation

As one co-variable increases, the other decreases

Zero correlation

When there is no relationship between the co-variables

Evaluation of correlations:

Strengths:

• Provide a precise and quantifiable measure of how 2 variables are related => this may suggest ideas for possible future research if variables are strongly related or demonstrate an interesting pattern

• Often used as a starting point to assess possible patterns between variables before researchers commit to an experimental study

• Quick and economical to carry out => no need for a controlled environment and no manipulation of variables is required => secondary data can be used which means correlations are less time-consuming than experiments

Limitations:

• Can only tell us that variables are related but not why => cannot demonstrate cause-effect between variables - we do not know which co-variable is causing the other one to change

• Also, it might be the case that another untested variable is causing the relationship between the 2 variables (an intervening variable)

Qualitative data

Expressed in words and is non-numerical e.g. a transcript from an interview

Strengths:

• More richness of detail than quantitative data => much broader in scope + allows the ppt to more fully report their opinions and feelings on the subject

• Greater external validity - provides researcher with more meaningful insight into the ppt’s worldview

Limitations:

• Hard to analyse => cannot really be studied statistically so patterns between data hard to identify

• Conclusions rely on the subjective interpretation of the researcher - can lead to bias

Quantitative data

• Data that can be counted - usually given as numbers

• Data can be analysed statistically + easily converted into graphs

Strengths:

• Simple to analyse - thus, comparisons between groups can be easily drawn

• Data in numerical form tends to be more objective and less bias

Limitations:

• Narrower in meaning and detail => fails to represent ‘everyday life’

Primary data

• Original data that has been collected specifically for the purpose of the investigation by the researcher

• Data that arrives first-hand from the ppts themselves

Strengths:

• It fits the job => authentic data obtained from the ppts themselves for the purpose of a particular investigation (e.g. questionnaires can be designed in a way that specifically targets the information the researcher requires)

Limitations:

• Requires time and effort => conducting experiment requires planning + resources

Secondary data

• Conducted by someone other than the person conducting the research => this data already exists before the psychologist begins their research or investigation

• Often the case that secondary data has been subject to statistical testing and therefore the significance is known

• May be located in journal articles, books or websites

Strengths:

• Inexpensive + easily accessed using minimal effort

Limitations:

• Variation in the quality + accuracy of secondary data => may be outdated or incomplete => challenges the validity of conclusions

Meta-analysis

• Uses secondary data

• When a number of studies are identified which have similar hypothesis => they can be pooled together and a joint conclusion produced

Strengths:

• Allows us to create a larger and more varied sample => so results can be generalised across much larger populations (increases validity)

Limitations:

• May be prone to publication bias => researcher may choose to leave out studies with negative results or non-significant results

Evaluation of measures of central tendency

Mean

Strengths:

• Representative of the data as a whole => it is the most sensitive as it includes all the values in a data set

Limitations:

• Easily distorted by extreme values => so in the presence of extreme anomalies, the mean may not actually be representative of the data set as a whole

Median

Strengths:

Limitations:

Sign Test

To use sign test:

• Difference not association

• Repeated measures design

• Nominal data

• 0.05 => findings are significant so reject the null hypothesis

• 0.01 => a stricter + more stringent level => to be even more confident that the findings were not due to chance (e.g. in cases where there is a human cost - drug trials