Paper 2Investigating

behaviour

Research Methods

Information Booklet

QUESTIONNAIRESA questionnaire is a self-report technique which allows participants to directly provide information about themselves. They consist of a set of pre-written questions which can be printed and given to Ps face to face, or can be posted, filled in by phone, completed over the internet, or simply left in public places.

Questionnaires are useful for surveying attitudes, opinions, beliefs and behaviours; they also offer some flexibility as questions can be quite broad and invite participants to answer in their own words (open questions) producing detailed qualitative data, or they can be quite narrow with forced choice responses such as tick boxes (closed questions) producing easy to analyse quantitative data.

Good questionnaires will avoid vagueness/ambiguity (e.g. ‘Do you drink coffee often?’ can be interpreted

differently by different people whereas ‘How many cups of coffee do you drink every day?’ is much clearer),

double-barrelled questions (where two questions are asked in one and P is left unsure how to answer if they are only permitted to give one response e.g. ‘Do you think crime is due to bad housing and poor education?)

leading questions (where people are encouraged to give a particular answer creating possible bias in responses e.g. ‘Many people think abortion is wrong: do you agree?’),

avoid the use of overly complex phrases or technical jargon. They will have the questions arranged in a logical order to avoid EV such

as demand characteristics.

AO2ADVANTAGES WEAKNESSESThey are highly replicable because it is easy to ensure procedures are the same for all participants – this allows researchers to check findings for reliability.

People may modify their answers to show themselves in the best light (dishonesty/social desirability bias); this reduces the validity of any subsequent findings.

They are time (and therefore cost) efficient, as a large sample of participants can be reached quickly and easily; a large amount of data can be gained.

Participant samples may be biased towards more literate people – this reduces population validity andmeans the sample is unrepresentative.

Investigator effects are reduced because researchers don’t need to be present to administer a questionnaire; this improves the validity of findings.

Because researchers are not always present, participants are unable to ask for help with unclear questions and may also miss sections/pages out.

INTERVIEWSAn interview is another form of self-report technique which allows participants to directly provide information about themselves. Here the researcher asks participants questions verbally about the topic being researched, usually face to face.

In a structured interview the interviewer has a pre-written set of questions which they do not deviate from; all participants are asked the same questions in the same order. In an unstructured interview the interviewer may have a few general questions in mind but there are no set questions; there is instead flexibility to pick up on issues in the participant’s comments and for them to expand on their responses.

When designing an interview, researchers must consider the categories of data required by their aim and must generate an appropriate set of questions, they must also decide whether to use a structured or unstructured interview. In addition they will consider issues such as social desirability, bias and ethical issues and will decide how responses are to be gathered during the interview (i.e. whether the interview will be recorded or whether the interviewer will make notes throughout).

AO2ADVANTAGES WEAKNESSESMay be more appropriate than other methods for dealing with complex/sensitive issues – the researcher can gauge if the participant is distressed or not.

Like with questionnaires, what people say they will do is not always what they would actually do in real life – ‘social desirability’ in answers reduces validity.

Because the researcher is present, interesting issues (as well as any misunderstandings) can be followed up immediately.

There can be low inter-rater reliability between interviews (of the same participant), due to interviewer effects of age, gender, ethnicity, personality etc.

Lots of rich data is gathered (especially in unstructured interviews) compared to e.g. a questionnaire, as there are far fewer constraints in place.

Interviews are extremely time consuming to prepare for and conduct; detailed data sets (particularly from unstructured interviews) take time to analyse.

CASE STUDIESA case study is an in-depth study of just one individual (or a particular group of people, an institution or an event). They are useful as they allow researchers to investigate unique cases in a lot of detail – often the research has a narrow focus on just one aspect of behaviour and is longitudinal (taking place over many years).

The case study method often incorporates the use of other research methods as techniques, for example researchers may give questionnaires to their participants, they may interview them, they may observe them, they may conduct experiments with them or they may conduct a content analysis on diaries/letters/school or employment records etc. The parents, other family members, teachers, managers, colleagues of the main participant may also take part in the research.

AO2ADVANTAGES WEAKNESSESCase studies allow researchers to investigate topics that it would be impractical and/or unethical to investigate experimentally.

Because only a limited number of people are investigated, it is difficult to confidently generalize the results to the population and to replicate the research.

Unique cases can challenge existing ideas and theories and also suggest new areas and/or hypotheses for future research.

There are often significant ethical issues with case studies due to the nature of the participants – e.g. if they are very young and/or have a significant disorder.

Complex interactions can be studied (using a wide range of techniques), rather than simple cause-and- effect relationships being found.

Data and subsequent findings may be unreliable due to subjective and/or biased recall and/or interpretation by the researcher, if they become too involved.

OBSERVATIONSAn observation is where behaviour is watched and recorded. In a naturalistic observation the researcher observes participants in their own environment, without manipulating the situation in any way. In a controlled observation the researcher actively manipulates variables and the observation would therefore usually take place in a specially set-up environment e.g. within a research laboratory.

Before any observational study takes place, the researcher has to decide on the particular behavioural categories or events to be investigated. In an unstructured observation (naturalistic or controlled) the researcher will not decide in advance exactly which behaviours they will record and will instead attempt to record a continuous stream of data, focusing on eye-catching or unusual events.

In a structured observation (naturalistic or controlled) the researcher decides in advance exactly which behaviours they will record; these are operationalized and a behaviour checklist is created. A behaviour checklist usually takes the form of a tally chart – observers count frequencies of the behaviours seen during the observation and record these in the relevant place on the tally chart, totals are then used to draw conclusions.

Recording of DataWith all observation studies an important decision the researcher has to make is how to classify and record the data. Usually this will involve a method of sampling. The three main sampling methods are:

1. Event sampling. The observer decides in advance what types of behaviour (events) she is interested in and records all occurrences. All other types of behavior are ignored.

2. Time sampling. The observer decides in advance that observation will take place only during specified time periods (e.g. 10 minutes every hour, 1 hour per day) and records the occurrence of the specified behavior during that period only.

3. Instantaneous (target time) sampling. The observer decides in advance the pre-selected moments when observation will take place and records what is happening at that instant. Everything happening before or after is ignored.

AO2

ADVANTAGES WEAKNESSESNaturalistic observations have extremely high ecological validity; if participants are unaware they are being observed their behaviour is completely natural.

If participants are aware they are observed their behaviour can quickly become unnatural, which massively reduces ecological validity.

Observations allow researchers to investigate topics that it would be impractical and/or unethical to investigate experimentally.

There are many ethical issues with observations if participants are unaware they are being observed (informed consent, confidentiality, right to withdraw).

Observations are a useful preliminary research tool; researchers investigating new areas can use them to produce hypotheses for future studies.

Low reliability – observed situations are often unique and therefore difficult to replicate; observers often disagree on judgments (low inter-rater reliability).

CONTENT ANALYSISA content analysis is where a researcher quite literally analyses the content of something, usually in order to transform complex qualitative data into quantitative data so that conclusions about patterns may be drawn more easily. This method can be used as a technique when applied to other research methods, for example a content analysis can be used to transform the qualitative data of an interview transcript into quantitative data.

The process of content analysis is extremely systematic. First the researcher decides what material to sample, they then decide what type of themes/categories might emerge from these materials and create a coding system based on these. A sample of material is then collected and analysed using coding units e.g. researchers may note each time they find a certain word/theme/character. By counting the frequencies of occurrence of each coding unit numerical (quantitative) data is obtained; statistical analysis can then be carried out.

AO2

ADVANTAGES WEAKNESSESStudies can easily be replicated by accessing archived materials and repeating the analysis; therefore findings can be tested for reliability.

Validity can be quite low; in identifying coding units researchers may be inconsistent or impose their own meaning on the data (subjective judgments).

As a method, content analysis has high ecological validity because it is based on real communications which have been gathered in natural settings.

Content analysis can be extremely time consuming both in terms of the preparation needed and the time spent on analysing large volumes of sample material.

Content analysis is an effective way of presenting qualitative data in a way that is easy to understand.

Ethical issues with content analysis as Ps may be unaware that their materials are to be used as part of the research.

CORRELATIONAL ANALYSISAO1Correlational analysis is a statistical technique used for investigating the strength of the relationship between two variables; it usually involves the researcher collecting two sets of secondary data i.e. no ‘hands on’ research takes place. The analysis will show either a positive correlation (as one variable increases, the other variable increases), a negative correlation (as one variable increases, the other variable decreases) or no correlation.

Correlations can be illustrated visually (using scatter graphs), or numerically (through correlation coefficients). A correlation coefficient is a number which ranges from -1 to +1; it shows the exact direction (sign) and strength (number) of the relationship.

When there is no relationship between two variables this is known as a zero correlation. For example there is no

relationship between the amount of tea drunk and level of intelligence.

When you draw a scattergram it doesn't matter which variable goes on the x-axis and which goes on the y-axis. Remember, in correlations we are always dealing with paired scores, so the values of the 2 variables taken together will be used to make the diagram.

Differences between Experiments and CorrelationsAn experiment isolates and manipulates the independent variable to observe its effect on the dependent variable, and controls the environment in order that extraneous variables may be eliminated. Experiments establish cause and effect.

A correlation identifies variables and looks for a relationship between them. An experiment tests the effect that an independent variable has upon a dependent variable but a correlation looks for a relationship between two variables.

This means that the experiment can predict cause and effect (causation) but a correlation can only predict a relationship, as another extraneous variable may be involved that it not known about.

AO2ADVANTAGES WEAKNESSESThey frequently utilise pre-existing or archival data; because participants/facilities are not required they are relatively time (and cost) efficient.

It is impossible to establish cause and effect between variables – we can say that they are related but we do not know in which direction the relationship functions.

Correlational analysis allows researchers to investigate topics that it would be impractical and/or

Inaccurate conclusions are commonplace – the media and/or policy makers may infer something different than is there

unethical to investigate experimentally.

(ethical implications re: misuse of data).

Correlational analysis is a precise method – it can tell researchers the exact strength and direction of the relationship between two variables.

Can only measure linear relationships (i.e. clear positives or negatives) – does not detect curvilinear relationships (e.g. where positive becomes negative).

EXPERIMENTAL METHOD (LABORATORY)AO1These usually take place in a special facility (laboratory) within a university psychology department. The independent variable is directly manipulated by the researcher and its effect on the dependent variable is directly measured. All extraneous variables are controlled as much as possible.

AO2ADVANTAGES WEAKNESSESIsolation of the effect of the independent variable onthe dependent variable means that cause and effect can be inferred with reasonable confidence.

The artificiality of the research context usually means participants do not demonstrate real-life behaviour – this dramatically reduces ecological validity.

Strict controls and well-documented procedures mean other researchers can quite easily replicate laboratory experiments to check findings for reliability.

Demand characteristics are likely – participants will look to both the researcher and the research situation for clues about how they are predicted to behave.

As the research takes place in a designated research facility, specialist equipment can be used to deepen our understanding of behaviour e.g. MRI scans.

Laboratory experiments are impossible to use in situations where it would be inappropriate to manipulate the IV for practical or ethical reasons.

EXPERIMENTAL METHOD (FIELD)AO1Field experiments are the same as laboratory experiments in terms of the treatment of the IV/DV/EV’s, except the laboratory environment is swapped for a real-life setting such as a school, town centre or hospital.

AO2ADVANTAGES WEAKNESSESEcological validity is higher than in a laboratory experiment, due to the real world setting. This means findings relate better to real life for generalisation.

The researcher’s control over the environment reduces in the real world; more extraneous (later confounding) variables greatly reduce the validity of results.

Demand characteristics are greatly reduced compared to a laboratory experiment – if participants are unaware they are taking part they act more naturally.

If participants are unaware they are taking part they may become distressed by manipulations of the independent variable - it is impossible to debrief them.

As the effect of the independent variable on the dependent variable is still isolated we can still determine cause and effect in most situations.

Because the researcher has no real control over the participants who take part samples may be biased (e.g. on age/gender) so population validity is reduced.

EXPERIMENTAL METHOD (NATURAL)AO1In a natural experiment the researcher does not manipulate the independent variable at all – it is naturally occurring. They act only to measure the effect of the independent variable on the dependent variable (so this method is not truly experimental, more quasi-experimental). Examples include studies of the effects of child abuse on adult relationships, or the effects of anorexia nervosa on cognitive development.

AO2ADVANTAGES WEAKNESSESNatural experiments allow researchers to investigate topics that it would be impractical and/or unethical to investigate using other experimental methods.

The researcher has no control over the environment; extraneous (later confounding) variables greatly reduce the validity of results – they cannot be eliminated.

Ecological validity is extremely high compared to e.g. a laboratory experiment. The researcher is able to study completely ‘real’ problems and situations.

Ethical guidelines of informed consent, confidentialityand right to withdraw are breached if participants are unaware they are taking part in the research.

Demand characteristics are greatly reduced comparedto a laboratory experiment – if participants are unaware they are taking part they act more naturally.

As the natural events psychologists wish to study are rare (even one-offs) it is often impossible to replicate the research to test findings for reliability.

Quasi Experiments

Definition of Quasi-ExperimentA quasi-experiment is designed a lot like a true experiment except that in the quasi-experimental design, the participants are not randomly assigned to experimental groups.

In a true experiment, research participants have an equal chance of being assigned to any condition of the independent variable (the one being manipulated by the researcher) that is involved in the study. So, for instance, if a researcher was examining the effects of caffeine on reading comprehension, she might randomly assign participants to one of three independent variable conditions: those who drink one cup of soda, two cups of soda, or no soda. She might then assess each person's reading comprehension abilities following exposure to the independent variable. In a true experiment, each participant who volunteered would have an equal chance of being assigned to any of the three groups.

Quasi-experiments are employed when the researcher is interested in independent variables that cannot be randomly assigned. Usually this happens when the independent variable in question is something that is an innate characteristic of the participants involved.

Advantages Useful when it's unethical to manipulate the IV Studies the 'real effects' so there is increased realism and

ecological validity

DisadvantagesConfounding environmental variables are more likely= less reliableMust wait for the IV to occurCan only be used where conditions vary naturallyAware they're studied= less internal validaty

EXPERIMENTAL DESIGNSOnce a researcher has decided that they will use an experimental method and they have written a hypothesis, they have to decide which experimental design to use.

An experimental design is a decision about how to allocate participants to different experimental conditions.

Every experiment has two conditions: an experimental condition (where participants are exposed to the independent variable in order to see its effect on their behaviour) and a control condition (where participants are not exposed to the independent variable but otherwise have exactly the same experience as the participants in the experimental condition, in order to gain a ‘baseline’ measurement of behaviour so the researcher can assess if anything has in fact happened in the experimental condition).

There are 3 types of experimental design: Independent groups

Repeated measuresMatched pairs.

DESIGN A: INDEPENDENT GROUPS

AO1Different participants are placed in each condition i.e. there are two separate and different groups:

EXPERIMENTAL CONDITIONVitamin drink consumed while revising

CONTROL CONDITIONNothing drunk while revising

P1P2P3

P4P5P6

AO2ADVANTAGES WEAKNESSESBecause each participant only takes part once, researchers only need to produce one set of stimulus materials e.g. word lists. This makes for a fairer test.

More participants are required than in other designs, as the sample size is halved when the participants are split. This makes the design expensive to use.

Order effects such as boredom, tiredness and/orlearning are reduced because participants only experience one condition – this increases validity.

Because results from different participants are compared, individual differences may negatively affect the results and therefore any subsequent conclusions.

DESIGN B: REPEATED MEASURESAO1The same participants are used in both conditions, i.e. each person takes part twice:

EXPERIMENTAL CONDITIONVitamin drink consumed while revising

CONTROL CONDITIONNothing drunk while revising

P1P2P3

P1P2P3

AO2ADVANTAGES WEAKNESSESFar fewer participants are required than in other designs, as the same sample is used twice. The design is cost-effective; only one set of participant expenses.

Participants experience both conditions, so order effects such as boredom, tiredness and/or learning negatively impact on results and conclusion validity.

Because results from the same participants are compared, individual differences do not affect the results or any subsequent conclusions.

At least two sets of stimulus materials are needed – this can create extraneous (even confounding variables) e.g. if word lists differ in difficultly.

NOTE – order effects are a significant issue with the use of repeated measures design. One technique used to overcome order effects is

‘counterbalancing’ – this is where the order of the conditions is mixed up, so that 50% of participants experience the experimental condition followed by the control condition, whereas the other 50% of participants experience the control condition followed by the experimental condition.

Although this does not eliminate order effects (as all participants are still experiencing both conditions), it does mean that any order effects are now equal across both conditions, so their negative effect is greatly reduced.

DESIGN C: MATCHED PAIRSAO1Different participants are used in each condition (like independent groups), but they are matched on key variables to form ‘pairs’ (to imitate repeated measures):

EXPERIMENTAL CONDITIONVitamin drink consumed while revising

CONTROL CONDITIONNothing drunk while revising

P1aP2bP3c

P4aP5bP6c

AO2ADVANTAGES WEAKNESSESBecause each participant only takes part once, researchers only need to produce one set of stimulus materials e.g. word lists. This makes for a fairer test.

The process of matching participants is difficult, time consuming and may be inaccurate/incomplete; participant variables are never fully eliminated.

Order effects such as boredom, tiredness and/or learning are reduced because participants only experience one condition – this increases validity.

If this design is used, attrition by just one participant will mean the loss of valuable data from the whole pair from the research; this is not cost or time effective.

FURTHER INVESTIGATION DESIGN: AIMSAO1All studies have an aim; an aim is the purpose of the study. Having a written aim makes research more focused – it clarifies what it is that the researcher is trying to discover.

TO INVESTIGATE A DIFFERENCETO INVESTIGATE A RELATIONSHIP

FURTHER INVESTIGATION DESIGN: HYPOTHESES

AO1A hypothesis is a precise, testable statement about the expected outcome of a piece of research i.e. a prediction. There are three types of hypothesis that are possible.

DIRECTIONAL HYPOTHESES when a researcher has a good idea what is going to happen in a study they will predict a specific outcome i.e. they will be specific about the direction of any differences in the way people behave

NON-DIRECTIONAL HYPOTHESES when a researcher is less sure what is going to happen in a study (i.e. findings could go either way) they will predict a more general outcome i.e. that there will be a difference in the way people behave, but not which direction this will be in

NULL HYPOTHESIS when a researcher is confident that the independent variable will have no effect at all on the dependent variable they select a null hypothesis

THERE IS A SIGNIFICANT DIFFERENCETHERE IS A SIGNIFICANT RELATIONSHIP

ONE TAILED- THERE IS A SIGNIFICANT DIFFERENCE WITH STUDENTS RECALL, WITH STUDENTS BEING ABLE

TO RECALL MORE INFORMATION ON A MONDAY MORNING THAN A FRIDAY AFTERNOON.

FURTHER INVESTIGATION DESIGN: VARIABLES

A variable is something that can change (or ‘vary’) within a study e.g. a score on a memory test.

An INDEPENDENT VARIABLE (IV) is the variable that is changed/manipulated by the researcher (cause)

A DEPENDENT VARIABLE (DV) is the variable which is measured by the researcher (effect)

‘Operationalising’ variables helps other researchers to replicate your research, as you detail as clearly and precisely as you can how each variable is measured.

Both the IV and DV must always be stated in a written hypothesis for it to be a complete statement.

An EXTRANEOUS VARIABLE is anything other than the IV that may have an influence on the DV – it is an uncontrolled variable that should really be eliminated. If the researcher fails to control for an extraneous variable and it negatively affects the research findings it is then known as a ‘confounding’ variable.

FURTHER INVESTIGATION DESIGN: PILOT STUDIES

No research is perfect. To help foresee any costly problems a small scale ‘pilot study’ may be carried out as a trial run before the researchers commit to conducting their full scale main study.

Pilot studies allow researchers to identify any potential problems in the method/design chosen, instructions given to participants, procedures, materials and measurements. These problems can then be rectified (or the decision can be made to scrap the study entirely), without an entire participant sample and a set of stimulus materials being wasted. This can save a lot of time and money.

Pilot studies are possible with most research methods, apart from natural experiments and case studies. This is because the events/participants are so rare that it would be too wasteful to sacrifice a sample for a pilot study.

THE BRITISH PSYCHOLOGICAL SOCIETY (BPS) CODE OF ETHICS

Psychology Is Dead Cool Really = Protection from Harm / Informed Consent / Deception / Confidentiality / Right to Withdraw

PROTECTION FROM HARMPsychologists have a responsibility to protect their participants from physical and emotional harm (e.g. embarrassment / humiliation, stress or loss of self-esteem/dignity). Ps should not be exposed to more risk than they would expect to experience in everyday life.

INFORMED CONSENTGiving consent means agreeing to something – a participant should always agree to take part in a study. When a participant is told the aims of research, as well as the nature of the procedure and the purpose of their role and agrees to it, this is known as informed consent i.e. the participant is fully informed before consenting.

DECEPTIONResearchers should not withhold any information from participants or actively mislead them about the true nature of the study they are to be involved in, either to encourage them to give consent or to get more valid results. Exceptions include when the deception is minor, deemed scientifically justified by an ethics committee and/or if participants are unlikely to object/show unease when the deception is revealed.

CONFIDENTIALITYConfidentiality means keeping information private. Participants should feel confident that the study’s report won’t reveal information or data which makes it possible for individual participants to be identified, or for their data to be linked to them i.e. they should remain anonymous.

RIGHT TO WITHDRAWParticipants should be allowed to leave at any point during the study if they decide they no longer want to take part, including retrospectively after the study has finished (their data would be

removed from the research and destroyed).HOW PSYCHOLOGISTS DEAL WITH ETHICAL ISSUES

PROTECTION FROM HARM Psychologists can ask their colleagues as well as ethics committees

to check their research proposals, to help spot any potential problems. At the start of their study they can ask their Ps about any pre-existing physical/mental conditions. During the study they can stop the research at the first sign of any harm occurring. After the study they can debrief all participants in full and offer aftercare.

INFORMED CONSENT Psychologists will ask adult participants to read and sign a consent

form, and will ask the parents of children under 16 years of age to give consent on their behalf. The carers / specialists of an adult with communication/ understanding difficulties will be consulted if it is felt they are unable to make an informed decision on their own.

DECEPTION Debriefing should be used after the study to explain the real aim and

rationale for the deception, as well as to reassure the participant and allow them to ask any questions they might have. The right to withdraw should be emphasised throughout, and the retrospective right to withdraw via destruction of data should also be offered.

CONFIDENTIALITY Psychologists should allocate numbers, letters or codes to participant

data in place of names to ensure it is anonymous, as well as keeping the location of the research as general as possible. Consent must be gained from Ps for their data to be used in situations where it is impossible to offer confidentiality

RIGHT TO WITHDRAW Participants should be informed of their right at the beginning of the

research, and should be reminded of this right at suitable points during and after the research. If they choose to exercise their right, pressure should not be put on them to stay and payment should not be used to coerce participants. Task avoidance should be taken as a wish to withdraw in child participants. If participants withdraw retrospectively their data must be destroyed.

Many institutions have ethics committees who approve or reject research proposals using a cost-benefit analysis (weighing up potential costs to Ps with potential benefits to society/science).

If psychologists ignore ethical guidelines or ethics committees they cannot be banned from research, but they may be expelled from their university/professional society, they may have their research licences revoked and they may face legal action from Ps.

PARTICIPANT SAMPLING TECHNIQUES

A ‘target population’ is a group of people (who share a given set of characteristics) about who the researcher wishes to draw a conclusion. As the target population is normally too large to allow all of the people in it to be tested, due to reasons of cost and practicality; the researcher obtains just a sample instead.

Because the researcher intends to generalise any conclusions generated using their sample to the entire target population, the sample should be representative of the target population i.e. they should share the same characteristics. If the sample is not representative it is said to have low population validity.

No method of sampling participants can guarantee a representative sample, but some are better than others.

TECHNIQUE A: RANDOM SAMPLINGWith this technique, every person in the target population has an equal chance of being selected.

The researcher first obtains a list of everyone in the target population, and then uses a computerised random generator or the ‘names out of a hat’ technique to select the required amount of participants.

AO2ADVANTAGES WEAKNESSESThe chances of selecting a biased sample are relatively slim, because everyone has a chance of being selected. This improves population validity greatly.

It can be difficult to obtain a list of the entire target population; even if you can, not everyone you select will be available and/or willing to participate.Even though this is a random method, a representative sample is not guaranteed – there is a chance some subgroups may be overrepresented or not selected.

TECHNIQUE B: OPPORTUNITY SAMPLINGWith this technique, the researcher selects anyone who is readily available and willing to take part. The researcher simply asks the people who it is most convenient for them to ask e.g. a researcher who also works as a university lecturer may ask students in their seminar group to participate.

AO2ADVANTAGES WEAKNESSESThis is a time (and therefore cost) efficient technique,as participants are readily available – sample sizes can be larger as the expenses per individual are smaller.

Samples are likely to be skewed in terms of participant backgrounds; an unrepresentative sample lacks population validity and findings cannot be generalised.There may be ethical issues with this technique regarding consent and right to withdraw if participants feel obliged to take part (e.g. students of a lecturer).

TECHNIQUE C: VOLUNTEER SAMPLINGWith this technique, participants put themselves forwards for inclusion i.e. they self-select. The researcher places an initial advertisement in a magazine/newspaper, on the radio, on the internet/via email, or on a public noticeboard (e.g. a workplace/gym), asking for volunteers to take part in research. They may also place questionnaires somewhere public and ask people to return their answers.

AO2ADVANTAGES WEAKNESSESThis technique can sometimes be the only way of locating a particularly niche group of participants who are difficult to identify using the available information.

Only atypical members of the target population respond i.e. the most co-operative and motivated. This reduces population validity and generalizability.Only people who see the advertisement have a chance of being selected – this may reduce the overall size of the sample which reduces the significance of findings.

RELIABILITY

Reliability refers to the consistency of a measure. A measure is considered reliable if we get the same result repeatedly. A research method is considered reliable if we can repeat it and get the same results.

When assessing the reliability of a study, we need to ask 2 questions:

1) Can the study be replicated?2) If so, will the results be consistent?

A ruler for example would be reliable, as the results could be replicated time after time and the same results would be gained (consistency). If you measure the length of a book on Monday, and your ruler tells you it is 25 cm long, it will still tell you it is 25cm long on Friday.

An IQ test however may be unreliable, if a person sits the test on Monday and scores 140, and then sits the same test on Friday and scores 90. Even though it can be replicated, it shows low consistency and therefore is an unreliable test.

Internal reliability = a calculation of the extent to which a measure varies from another measure of the same thing over time.

How to Assess it Meaning How to improve it

Split Half Reliability

Splitting a test into two halves and comparing the

scores in both halves.

If the results in the two halves are similar, we can assume the test is reliable

Select the test items that

produce the greatest similarity.

External reliability = a measure of the extent to which something is consistent with itself.

How to Assess it

Meaning How to improve it

Inter-Rater reliability

If the measure depends upon interpretation of

behaviour we can compare the results from two or

more raters.

If there is high agreement between the raters, the

measure is reliable

Standardised procedures

Standardised instructions

Double blind techniques

Test-Retest reliability

The measure is administered to the same

group of people twice.

If the results on the two tests are similar we can

assume the test is reliable.

VALIDITYValidity is the extent to which a test measures what it claims to measure.

INTERNAL VALIDITY

INTERNAL VALIDITY The extent to which a procedure is measuring what it was intended to measure (rather than some other behavior).

In an experiment, internal validity concerns whether the outcome is the result of the independent variable, rather than of extraneous variables - eliminating extraneous variables helps to improve validity, as does improving the realism of research contexts so that participants show real-life behaviour.

EXTRANEOUS VARIABLE

HOW DOES IT AFFECT VALIDITY?

HOW CAN IT BE

OVERCOME?

Situational variables (anything to do with the environment of the experiment): time of day, temperature, noise levels etc.

Something about the situation could act as an EV if it has an effect on the DV.

E.g. poor lighting could affect Ps performance on a memory test

Standardized procedures

Participants variables (anything to do with differences in the Ps): age, gender, intelligence, skill, motivation, etc.

It may be that the differences between the Ps cause the change in the DV.

For example, one group may perform better because they are younger, or more motivated.

Repeated measures

design

Matched pairs

Investigator effects: the behaviour and language of the experimenter may influence the behaviour of the Ps: (experimenter bias)

Leading questions from the experimenter may consciously or unconsciously alter how the participant responds. Double blind

Participant Effects:Ps search for cues as to how to behave in an experiment.

The structure of the experiment could lead the participant to guess the aim of the study (demand characteristics).

They may be overly helpful and want to please the experimenter. Or, they may decide to deliberately act in a way which spoils the

It is important to try and create a situation

where the Ps will not be

able to guess what the aim of the study

is.

experiment. This is the “screw you” effect.

EXTERNAL VALIDITYThe degree to which a research effect can be generalized to other settings, other people and over time.

ECOLOGICAL VALIDITY: can findings be generalised to other settings.

POPULATION VALIDITY: can findings be generalised to other people.

TEMPORAL VALIDITY: can findings be generalised to other times.

Improving the realism of research contexts so that participants show real-life behaviour, ensuring your sample is representative of your target population and attempting to apply research only to the time it was conducted improves each type of validity respectively.

ASSESSING AND IMPROVING VALIDITY

TypeHow does this assess validity

Example OutcomeFa

ce V

alid

ity

On the ‘face’ of it, does it actually measure what it intends to measure?

Does Asch’s study appear to be measuring rates of conformity?

This only means that the test looks like it works. It does not mean that the test has been proven to work

Con

tent

Val

idity Does the

method actually seem to measure what you intended?

Does an IQ test actually measure levels of intelligence, or ability to solve puzzles?

To ensure content validity, a panel of experts (on IQ for example) may be asked to assess the measure for validity.

Con

curr

ent

valid

ity

How well does the measure agree with existing measures?

Does our IQ test agree with established tests of IQ?

Test Ps with both the new test and the old test. There should be high agreement between the scores on both measures.

Con

stru

ct v

alid

ity

Is the method actually measuring all parts of what we are aiming to test?

If we use a maths test to test intelligence, we miss out on other factors involved such as linguistic ability or spatial awareness.

To maintain construct validity, we need to define what it is we are aiming to measure, and ensure that all parts of that definition are being measured.

Pred

ictiv

e va

lidity

Is our measure associated with future behaviour?

If someone scores high on our IQ test, they would be expected to do well in GCSEs.

Follow up Ps to see if future performance is similar to performance on our measure.

DATA ANALYSIS & PRESENTATION: TYPES OF DATA

Depending which research method a psychologist decides to use, and the specific sort of tasks/activities participants complete, they will generate either quantitative or qualitative data.

QUANTITATIVE DATA is information that can be analysed numerically

(quantitative/quantity); it is behaviour measured in numbers/quantities and arises

from questions such as ‘How often…?’, ‘How much…?’ ‘How long…?’

AO2ADVANTAGES WEAKNESSESThis form of data is very easy to analyse using a range of descriptive and inferential statistics, tables and graphs. Patterns are easy to see.

Reducing complex human behaviour to numbersoversimplifies reality – lots of rich, useful data is simply lost.

QUALITATIVE DATA is information that is in narrative form

(qualitative/qualities); it is behaviour measured in words and arises from

questions about ideas and/or feelings e.g. ‘Tell me about…’ or ‘Describe…’

AO2ADVANTAGES WEAKNESSESCapturing the complexity of ideas/feelings in detailedresponses means that lots of rich, useful data is retained. We get a good overview of human behaviour.

This form of data is very difficult to analyse using any descriptive or inferential statistics, tables or graphs. Patterns are tricky to see.

ANALYSIS AND INTERPRETATION OF QUANTITATIVE DATE:MEASURES OF CENTRAL TENDENCY

Measures of central tendency analyse how close scores are to the average participant response.

THE MEANThis is the arithmetic average; it is calculated by first adding all of the data scores together and then dividing this total by the actual number of scores:

E.g. 2 + 5 + 6 + 7 + 10 = 30 30/5 = 6

ADVANTAGES WEAKNESSESUses all of the data i.e. takes all scores into account

Is not always an actual score (e.g. 2.4 children)

Is the numerical centre point of all actual values in the data

Is easily skewed by one anomaly

THE MEDIANThis is the middle score or value, found when the data is turned into an ordered list:

E.g.1, 2, 5, 6, 7, 9, 11= 6

ADVANTAGES WEAKNESSESUnaffected by extreme scores

May not be an actual score if there’s an even number of scores

Relatively quick and easy to calculate

Not good for small data sets where there are large differences e.g. 1, 2, 1000, 1001 = 501

THE MODEThis is the most commonly occurring/most frequent score:

E.g. 2, 2, 2, 5, 6, 9, 11 = 2

If there are two scores which are most common the data set is bi-modalIf there are three or more scores which are most common the data is multi-modal

ADVANTAGES WEAKNESSESIs always an actual score

Sometimes a data set doesn’t have a mode, or has many

Is not distorted by an extreme value

Doesn’t use all the data

MEASURES OF DISPERSIONanalyse how far away scores are from the average participant response i.e. their spread.

THE RANGEThis is difference between the highest and lowest score in a set of data; it is calculated by subtracting the lowest score from the highest score:

E.g. 6, 10, 35, 50 = 50 – 6 = 44

ADVANTAGES WEAKNESSESEasy to calculate Ignores most of the data – doesn’t reflect

the true distribution around the meanTakes into account extreme values

Easily distorted by extreme values

STANDARD DEVIATIONThis is the result of a calculation which measures (collectively) how much individual scores deviate from the mean, and presents this finding as a single number. It tells us how much data is spread (dispersed) around a central value (the mean).

A large SD score tells us there was lots of variation around the mean/scores were spread widely i.e. that participants in the sample were all responding very differently.

A small SD score tells us that scores were closely clustered around the mean i.e. that participants in the sample were all responding in very similar ways.

ADVANTAGES WEAKNESSESPrecise, as all values are taken into account

Not easy or quick to calculate

Detailed conclusions can be made

-

PRESENTATION AND INTERPRETATION OF QUANTITATIVE DATA

TABLES Tables can be used to present data in a clear and simple way.

Categories from the research are usually used as columns headings within a table, with the rows within the table containing individual participant scores, or frequency counts in summary.

Some tables may show results from descriptive statistics e.g. measures of central tendency/dispersion.

GRAPHSVisual displays summarising measures of central tendency are also useful ways of describing data.

BAR CHARTS are used when data falls into categories; the X axis is labelled with the categories, the Y axis with frequencies. The height of bars represents the number of times a category was recorded. Columns are equal widths, and do not touch each other.

HISTOGRAMS are use when results can be put

in a continuous order; the X axis shows all of the possible scores in order, the Y axis shows how many P’s got this score. The column heights show the frequencies. All columns are equal widths, and there are no spaces between columns.

LINE GRAPHS work in exactly the same way as a histogram, except lines are used to show where the top of each column would reach. They are particularly useful for comparing two distributions because they can be drawn on the same graph. SCATTER GRAPHS are

useful if you want to see how 2 variables compare. The X axis shows data from one variable, the Y axis shows data from the second variable. Each participant is represented by a single point, where their x/y values meet. When all participant scores have been plotted a line of best fit can be drawn, which highlights the overall trend of the results.

DATA ANALYSIS & PRESENTATION: QUALITATIVE DATA

Qualitative data is sometimes seen to be of limited use because it’s difficult to analyse and open to bias during interpretation (e.g. data from an observation, open questions from a questionnaire or an interview). For this reason it’s often converted into quantitative data via content analysis.As a reminder: First the researcher decides what material to sample, they then

decide what type of themes/categories might emerge from these materials and create a coding system based on these. A sample of material is then collected and analysed using coding units e.g. researchers may note each time they find a certain word/theme/character. By counting the frequencies of occurrence of each coding unit numerical (quantitative) data is obtained; statistical analysis can then be carried out.

If a psychologist is attempting to analyse the data gathered from an observation, or to present the data in quantitative form there must be a good amount of data to begin with, definitions of behaviours must be operationalized and observer bias must have been avoided.

With questionnaires, the selection of data items must be unbiased, and interpretation of ambiguous answers must be attempted with caution.

With interviews, the interview context must be taken into account, and bias in the interpretation of participant responses must be avoided.

AO2ADVANTAGES of transforming data

WEAKNESSES of transforming data

Patterns are easier to see; statistical analysis can be carried out

Detail is lost when data is converted into numbers

Data is easier to summarise and present

Care is needed to avoid bias in defining coding units,or deciding which behaviours fit particular units

Longitudinal vs cross-sectional studies

Longitudinal studies differ from one-off, or cross-sectional, studies. The main difference is that cross-sectional studies interview a fresh sample of people each time they are carried out, whereas longitudinal studies follow the same sample of people over time.Features of longitudinal vs cross-sectional studies

Types of Brain Imaging Techniques

PET

Positron Emission Tomography (PET) uses trace amounts of short-lived radioactive material to map functional processes in the brain. When the material undergoes radioactive decay a positron is emitted, which can be picked up be the detector. Areas of high radioactivity are associated with brain activity.

EEG

Electroencephalography (EEG) is the measurement of the electrical activity of the brain by recording from electrodes placed on the scalp. The resulting traces are known as an electroencephalogram (EEG) and represent an electrical signal from a large number of neurons.

EEGs are frequently used in experimentation because the process is non-invasive to the research subject. The EEG is capable of detecting changes in electrical activity in the brain on a millisecond-level. It is one of the few techniques available that has such high temporal resolution.

CAT scans

Computed tomography (CT) scanning builds up a picture of the brain based on the differential absorption of X-rays. During a CT scan the subject lies on a table that slides in and out of a hollow, cylindrical apparatus. An x-ray source rides on a ring around the inside of the tube, with its beam aimed at the subjects head. After passing through the head, the beam is sampled by one of the many detectors that line the machine’s circumference. Images made using x-rays depend on the absorption of the beam by the tissue it passes through. Bone and hard tissue absorb x-rays well, air and water absorb very little and soft tissue is somewhere in between. Thus, CT scans reveal the gross features of the brain but do not resolve its structure well.

Evaluation:

different people's brains operate in different ways—not everyone's brain responds to a given task in exactly the same way. Even for a single person, their brain responds differently to the same task on each occurrence of the task. So researchers have everyone do the task tens or hundreds of times, and then they take a statistical average across all of the tasks. And after that, they average across everyone's brain. So those colorful pictures you are seeing represent an average—and it doesn't mean that every person, every time, displays exactly that activation pattern.

And finally, the colorful pictures hide a very important fact: The entire brain is active pretty much all of the time. Neurons are always firing, at least a few times every second. The brain is a complex system, and every cognitive activity is widely distributed across the neocortex.

Peer review in Psychology

Peer review is a process that takes place before a study is published to ensure that the research is of a high quality, contributes to the field of research and is accurately presented.  The process is carried out by experts in the related field of research

Peer review has an important function, as it ensures that only high quality research is disseminated and available as a body of scientific evidence.  Such evidence frequently becomes part of mainstream thinking and practice, so it is vital that conclusions that these are based on are the subject of valid methods and accurate presentation.  

If research was published without this process of review and checking, poor research might be disseminated which would damage the integrity of that field of research, or that of the discipline as a whole.  

In addition, research often has clear practical applications for society or people’s day to day lives; if research was not reviewed to ensure quality, then any recommendations or guidelines could not be founded and may have negative consequences for affected individuals

Strengths of Peer Review

Peer review promotes and maintains high standards in research, which has implications for society and funding allocation so that it is assigned to high quality research.

Helps to prevent scientific fraud, as submitted work is scrutinised.

It promotes the scientific process through the development and dissemination of accurate of knowledge and contributes new knowledge to the field.

Limitations of Peer ReviewIf anonymity is not maintained experts with a conflict of interest might not approve research to further their own reputation or career.

Contributes to the “file drawer effect” – as only statistically significant findings are published.  This means that findings that challenge existing understanding might be overlooked as they are not published

Research experiment

Milgram

  https://www.simplypsychology.org/milgram.html

Kohlberg

https://www.simplypsychology.org/kohlberg.html