Component 2-research methods Flashcards
Identify the key features of an experiment [2]
The experimental method involves the manipulation of the IV to see if this has any effect on the DV, in order to establish cause-and-effect relationships. Additionally, any extraneous variables are controlled.
Explain the difference between the aims of a study and a hypothesis. [2]
Aims of a study is a statement of what a researcher(s) intends to find out in a research study, whereas a hypothesis is a precise, testable statement about the assumed relationship between variables.
Explain what is meant by ‘operationalisation’ [3]
To ‘operationalise’ is to ensure that variables are in a form that can be clearly tested. In order for a concept to be investigated, such as ‘educational attainment’, it needs to be specified more clearly, for example, this concept could be operationalised as ‘GCSE grade in maths’.
Explain why standardisation is important in research procedures. [2]
Standardisation is important in research procedures, as if the procedures are not standardised, the results may vary, due to changes in procedure rather than because of the IV. Additionally standardisation is important in research procedures, as if enables the study to be repeated.
what is meant by the AIM of the research?
A statement of what the researcher intends to investigate and the purpose of the study
Types of hypothesis: What is meant by the alternative/experimental hypothesis
A precise and testable statement which predicts what change(s) will occur to the DV when the IV is manipulated. Operationalisation is key to making the hypothesis testable.
Types of hypothesis: What is a Null Hypothesis?
A null hypothesis states that there will be no changes to the DV due to manipulating the IV. It states the results are due to chance and are not significant in supporting the idea being investigated.
Types of hypothesis: What is a non-directional hypothesis?
A non-directional hypothesis is
Deciding on a research question: alternative hypothesis
any hypothesis except the null hypothesis, that states that the IV will have an effect on the DV
Directional hypothesis
states the direction of the predicted effect that the IV will have on the DV
Non-Directional hypothesis
predicts simply that the IV will have an effect on the DV but does not state in what direction
null hypothesis
the assumption of no relationship between variables being studies
Independent variables (IV)
the variable that is manipulated in an experiment to test its effect on the DV
Dependent Variable (DV)
The variable being influenced by the IV, which can be measured
Co-Variables
2 variables that are examined to see whether a correlation exists between them
Operationalisation of variables
defining the variables clearly so that they can be objectively manipulated or measured
confounding variables
A varibale that is not the IV under study, but which varies systematically with the IV. Changes to the Dv may be due to confounding variables rather than the IV, so the outcome is meanngless. Coundfounding variables confound the results.
Extraneous variables
variables that do not vary systematically with the IV and therefore do not act as an alternative IV but may have an effect on the DV. They are nuisance variables that muddy the waters and make it more difficult to detect a significant effect.
Methodologies: experiments
- cause and effect measured by controlling and manipulating variables
- participants randomly allocated to experimental/control groups
Methodologies: laboratory experiments
- under controlled, artificial conditions
- IV manipulated, DV measured
- not necasarily a laboratory, just a controlled environment e.g. a classroom.
- experimental and controlled conditions
- researcher randomly allocates participants to experimental/controlled conditions
SRENGTHS:
- high level of control (can infer the IV caused the DV)
- easy to replicate (can check reliability)
WEAKNESSES:
- low ecological validity
- demand characteristics (if participants know they are being studied they may act in a certain way and affect validity)
Methodologies: experiments: Field experiments
- conducted in a natural environment
- IV manipulated so casual relationships can be formed between IV and measured DV
- participants unaware they are being studied.
STRENGTHS:
- higher ecological validity than laboratory experiments as conducted in a real setting without control of experimenter
- less demand characteristics, so increased validity
WEAKNESSES:
- unethical (participants unaware of study, less chance they can be debriefed)
- high chance of extraneous variables affecting results because less control than lab experiments.
Methodologies: experiments: quasi
quasi: researcher does not deliberately manipulate IV and participants not randomly allocated to experimental or control condition. E.G.
Natural: IV arises naturally and DV measured. Done when unethical to manipulate IV directly.
STRENGTHS:
- allows research where IV cannot be manipulated for practical or ethical issues, so different behaviours can be measures such as Schizophrenia
- real problems can be researched such as the effects of disaster on health which helps larger amount of people in more situations
WEAKNESSES:
- no casual relationships can be demonstrated as IV not directly manipulated so cannot be sure IV caused the DV.
- threat to internal validity due to less control of extraneous variables that may cause change to DV and not the IV.
Methodologies: participant observation
- researcher takes on role of a participant, whilst observing other participants’ behaviour
- researcher becomes part of group and doe not reveal themselves
STRENGTHS:
- less chance of demand characteristics as participants do not know they are being observed by researcher
- can research people who otherwise would be difficult to observe so researcher may find info they do not know existed
WEAKNESSES:
- observer bias (researcher expectations affect perception of events and they become subjective)
- unreliable findings as difficult to take notes during observation, so data relies on memory
Methodologies: non participant observation
- researcher observes and records participants behaviour from a distance without interfering
- participants unaware they are being observed
- pre-prepared categories decided and behaviour recorded categories as and when it happens
STRENGTHS:
- less chance of observer bias, as observer not taking part in the action
- no self report methods, researchers actually see participant behaviour
WEAKNESSES:
- observer bias (difficult to make judgements on thoughts and feelings of participants so observer may misinterpret behaviour based on own views and opinions)
- unethical
Methodologies:what is content analysis?
Evaluate content analysis
- exploration of behaviour to see what categories or themes emerge and tallying each time material fits a category/theme
- converts qualitative data into quantitative data, so more easily compared
- type of observational study
- written/verbal material such as magazines, television programmes analysed
- sample (artefacts being analysed)
STRENGTHS
- less chance of demand characteristics from participants and researcher in creating materials as artefacts already exist
- can replicate as long as artefacts are accessible to others (higher reliability)
WEAKNESSES:
- observer bias (affects validity as observers may interpret meaning of categories differently)
- cannot draw cause and effect relationships as authors of artefacts often unknown so cannot be easily questioned as to how or why the behaviour was explored
Methodologies: What are structured interviews?
Evaluate structured interviews
- standardised questions (like in a questionnaire), known as an interview schedule and usually asked face to face
- pre prepared questions asked in a fixed order
SRENGTHS:
- same questions asked each time so results easy to analyse
- replicable to more reliable because same questions can be asked in same way.
WEAKNESSES:
- can be restricted because there is no chance to ask further questions, may be frustrating for participants if frustrating or interesting issues arise
- does not allow for spontaneous questions, which may mean interviewer less responsive to participants.
Methodologies: What are questionnaires?
Evaluate questionnaires
- list of written questions which generate closed/open ended answers
- produce quantitative and/or qualitative data
STRENGTHS:
- can be used to assess psychological variables not obvious through observation
- data collected from larger group of participants faster than interviewing them.
WEAKNESSES:
- no guarantee participant is being honest
- different participants may interpret question in different ways.
Methodologies: What are Semi-structured interviews?
Evaluate semi structured interviews
- asking participants questions usually face to face. May be in the form of an interview schedule, but may include follow up questions to expand on answers to questions asked.
- begin with predetermined questions but further questions developed as a response to answers
STRENGTHS:
- more qualitative data can be collected, as questioned tailored to participant answers
- high validity as participants have opportunity to fully express true feelings/views
WEAKNESSES:
- same questions not used every time (difficult to analyse results and identify patterns and trends)
- not replicable due to different questions asked each time (unreliable)
Methodologies: What are self report methods?
- involves participant reporting on own thoughts and feelings through methods such as interviews and questionnaires
- whichever self report methods used, questions are often either open or fixed choice/closed
- open questions- allow free participant opinion, more likely to produce qualitative data which is in depth but hard to analyse
- fixed choice/closed- respondent has limited response and more likely to produce quantitative data which is easy to analyse but quite shallow.
Methodologies: What are Correlational studies?
Evaluate correlational studies
- comparing two co variables to see if there is an association/relationship between them
- scatter diagram/graph can be used to illustrate correlations
- pos correlation (high values of one variable associated with high values of the other)
- neg correlation (high values of one variable associated with low values of the other)
- strength of relationship measured with a correlation co-efficient, which is calculated with a statistical test by researchers.
- the closer the coefficient is to 0, the weaker it is, and the closer to 1 (1 or -1), the stronger it is
- tables of significance used to find out if correlation coefficient is significant, which show how big the correlation coefficient needs to be in order for correlation to count as significant
STRENGTHS:
- shows the direction and strength of relationship which can be used to make predictions on behaviour
- can be used when experiments not appropriate e.g. unethical to manipulate stress and illness
WEAKNESSES:
- correlations only show whether there is a relationship, not how or why co variables are related. There may be other external variables which explain the relationship
- difficult to establish cause and effect
Methodologies: What are case studies?
Evaluate case studies
- in depth investigation of a phenomenon which uses a descriptive analysis of a person, group or event. holistic study through one or more methodologies, usually longitudinal
- uses many research methods (interview, questionnaire, observation etc)
- most data qualitative, sometimes quantitative
STRENGTHS:
- produces rich qualitative data w/ high ecological validity as it is a study of a real life situation
- allows researchers to study cases not practical or ethical to manipulate in experiment
WEAKNESSES: researcher bias (researchers become too involved and lose objectivity, so may interpret data to fit in with own theories)
- difficult to generalise findings beyond individual /group studied as sample too small (low pop validity)
Methodologies: What are self reports?
Evaluate self reports
- participant reports own thoughts and feelings through interviews, questionnaires, inventory and diaries etc
STRENGTHS:
- offer insight as to why people behave in a certain way, so less reason for researchers to guess reason for behaviour
- qualitative data can be gathered
WEAKNESSES:
- risk of social desirability bias
- people may not be able to accurately recall, especially if asked for details over an extended period of time
Methodologies: What is quantitative and qualitative data?
Evaluate quantitative data
Evaluate qualitative data
Quantitative: can be measured numerically so statistical analysis can be completed e.g. scores on IQ test
STRENGTHS:
- data easily analysed using statistics
- easier to collect from large groups of participants
WEAKNESSES:
- tends to lose ‘human’ level of behaviour
- tends to offer shallow view of behaviour
Qualitative: data that can be observed but not measured numerically. Usually words, thoughts, feelings and is difficult to analyse e.g. feelings about a school
STRENGTHS:
- can offer more individual, ‘human’ view of behaviour
- provides in depth detailed data
WEAKNESSES:
- can be difficult to analyse collected data
- data tends to come from a limited number and range of people
Methodologies: What are primary and secondary sources?
Evaluate primary sources
Evaluate secondary sources
Primary: information/data directly collected by the researcher first hand for their research e.g. through a questionnaire
STRENGTHS:
- researcher can control format in exactly how data collected and it will specifically relate to research aims
WEAKNESSES:
- data may lack validity due to social desirability and demand characteristics
Secondary: info sources/data not directly collected/created by researcher and has already been obtained by other researchers e.g. literature reviews
STRENGTHS:
- data produced without ‘participant’ knowing artefact would be used in research may be more valid
WEAKNESS:
- researcher cannot control format of how data is produced/collected. May not be a specific match to research aim.
Methodologies: brain scans
What are CT (CAT) scans?
Evaluate CT (CAT) scans
What are PET scans?
Evaluate PET scans
CT (CAT) Scans: set of x rays combine to form 2d of 3d images of brain area being scanned
radioactive dye is injected into patient before x rays and then patient is placed in CAT scan machine
CT scanners use series of x ray beams passed through the head, creating cross sectional images of the brain showing structure but not function
STRENGTHS:
- high quality images better than those produced by x ray alone
- reveal abnormal brain structures e.g. tumours
WEAKNESSES:
- only show structure, not electrical activity in brain
- radiation exposure. More detailed scan uses more radiation
PET scans:
patient given a radioactive glucose
more active areas of brain will need more glucose
detectors in scanner highlight the most active areas
STRENGTHS:
- only PET scans allow brain chemical activity to be seen, so can distinguish between benign and malignant tumours
- useful for psychological research as they look for more active brain areas
WEAKNESSES:
- costly to run and maintain- limited availability for research
- not as precise scans as MRI
- radioactive dye needed which can only be administered a few times
Methodologies: What are cross sectional studies?
comparing one group of participants, representing a cross section of society against another at same point in time.
one group of participants representing one section of society are compared with participants from another group
Methodologies: What are Longitudinal studies?
Evaluate longitudinal studies
- conducting research over a long time period to observe long term effects of something on a specific behaviour. May use other methodologies such as case studies, interviews etc
- participants commonly assessed on two or more occasions as they get older. This allows researcher to log long term effects
STRENGTHS:
- same person tested so participant variables controlled (no. of siblings, age etc)
- developmental trends can be spotted easily as tests are repeated at regular intervals over many years and findings compared
WEAKNESSES:
- high attrition (drop out) rate as research takes so long, so sample left is small so is biased
- participants are more likely to be aware of aims of study so may show demand characteristics
Location of research: Laboratory environment
- most scientific location
- maximum control over variables (IV, DV, extraneous, confounding
- environment set up by researcher for purpose of research
- experiments and observations carried out here
Conducting research in a field
- outside laboratory in a more natural setting, such as schools, shopping centres, hospitals, nurseries
- environment may or may be new to participant BUT environment is natural
- experiment, observation, self report methods and nearly all other research methods can take place in field.
Conducting research online
- psychologists can observe new behaviour online and traditional psychological topics more efficiently on a scale and scope not possible 50 years later
- most common used methodologies online are surveys and experiments
- researchers utilise social networking sites and specific research sites to find samples and conduct research
conducting research in a laboratory: evaluation
Strengths:
- controlled environment. Increases internal validity, ensuring data collected is accurate and true
Weaknesses:
- demand characteristics- participants know they are part of study, so act unnaturally
- lacks ecological validity as unrealistic and don’t reflect real life and results cannot be generalised to real life situations
- research limited in scope. Ethical issues and logistical issues mean many studies cannot be researched
Conducting research in a field: Evaluation
Strengths:
- high ecological validity due to natural environment. Behaviour more authentic as often participants unaware they are part of a study, so results can be generalised to real life.
- research can be very diverse, so researchers can study a variety of topics
Weaknesses:
- Lower experimental validity: not as much control over confounding variables compared to laboratory. Cannot be sure results are accurate and not due to confounding variables.
Conducting research online: Evaluation
Strengths:
- samples less gender biased: 43% male
- samples more culturally diverse
- much larger samples: results can be generalised to more people
- easier to analyse electronic data
Weaknesses:
- cannot confirm participants identity- may pose ethical issues as to whether vulnerable people should take part. Robotic responses may be part of study. Researcher cannot be sure participants has required characteristics needed for study.
- samples generally westernised and from urban cultures
- monetary rewards can skew who participates
Participants: target populations
The group of individuals that a researcher is interested in studying e.g. people in the UK.
Participants: sampling frames
a group or population that is identified when it is unrealistic to study the whole target population. e.g. people in London.
Participants: What is Random sampling?
Evaluate random sampling
- participants selected from sampling frame and everyone has an equal chance of being selected e.g. names pulled out of a hat/random number generator
STRENGTH: - fair method. no researcher bias in the way they selected the sample as equal chance.
WEAKNESS: - may have a biased sample, not due to researcher bias but because the selection was due to chance there may be certain subgroups in target pop that are over or under represented in sample.
Participants: What is opportunity sampling?
Evaluate opportunity sampling
- participants selected at researcher’s convenience, without knowing any details about the sample in advance. e.g. picking people who were there at the time, in your specific location
STRENGTH: - easier for the researcher to administer compared with other sampling techniques (these may be more costly in time and resources), because researcher just uses the first suitable participants they can find, so less time consuming
WEAKNESS: - may have a biased sample, as the sample is whoever was available at the time and people tend to live in groups with people they share things in common with. if sample selected for example, from people walking around the town centre on a monday morning, then would unlikely include proffessional people as they are at work and those who live in rural areas
- may not be ethical to ask participants to be a part of research e.g. some students may feel obliged to take part in research projects of University professor, meaning they are not giving valid consent.
Participants: Systematic sampling
- use a predetermined system, such as every nth person on a list is selected by the researcher e.g. every 3rd house on a street or every 5th person on a register
STRENGTH: - no researcher bias,as participants selected using an objective system. Not biased as long as the first participant is selected randomly.
WEAKNESS: - may still have a biased sample, not due to researcher selection bias, but because the selection has an element of chance so no guarantee subgroups are not over or under represented in sample
Participants: stratified sampling
- sample of participants is produced by identifying subgroups (strata) according to their frequency in the target population. Participants are then selected randomly from the subgroups. Examples include ‘boys and girls’.
STRENGTH: - guarantee all subgroups in target pop will be represented in sample
WEAKNESS: - more difficult for researcher to administer compared with other sampling techniques such as opportunity sampling which may be less costly in time and resources
- participants randomly selected from each subgroup may not be representative of that subgroup.
- very time consuming to identify subgroups and then select participants and contact them
Participants: Quota sampling
- target pop divided into subgroups e.g. by sex, and participants chosen from each subgroup at the convenience of researcher, using a non random technique. Participants are obtained from each strata in proportion to their occurrence in the target population.
STRENGTH: - guarantee all subgroups in target pop will be represented in sample
WEAKNESS: - more difficult for researcher to administer compared with other sampling techniques such as opportunity sampling which may be less costly in time and resources
- participants opportunistically selected from each subgroup may not be representative of that subgroup.
- time consuming to identify subgroups and then select participants and contact them.
Participants: self selected sampling
- a sample of participants that relies solely on volunteers to make up the sample, such as people volunteer to participate, in response to an advertisement in a newspaper or on the internet.
STRENGTH: - valid consent given, as participants volunteer themselves
WEAKNESS: - people who volunteer may not be from appropriate subgroups within target pop
- researcher may not have sufficient participants who are willing to take part.
Participants: snowball sampling
- participants initially recruited by psychologist and then those participants recruit further participants from people they know
STRENGTH: - a helpful sampling technique when trying to investigate a rare characteristic/behaviour
- enables researcher to locate groups of people who are difficult to access, such as drug addicts.
WEAKNESS: - the initial participant researcher identifies may not know sufficient participants who share their rare characteristic
- can take more time to complete than other techniques, such as opportunity, as researcher has to wait and rely on initial participants to find more participants.
- sample not likely to be a good cross section from the population because it is the friends of people who were initially recruited, so are likely to be very similar
Participants: observational sampling (event and time sampling)
techniques can be used in both non participant and participant observations
- EVENT: participants observed and specific behaviour (event) recorded each time it occurs to create a total score, e.g. to see how happy 3 children are during a lesson, you may create a tally chart of how many times each child smiles, laughs and jokes.
STRENGTH:
- records all events during observation unlike some time sampling methods (e.g. time point sampling)
WEAKNESS:
- limited as it puts all events together, so cannot say whether a behaviour is more common at the start, middle or end of the observation
- TIME: psychologist observes and records behaviour (such as a score) at specific time intervals e.g. every 15 mins and then creates an average score for each participant. researcher may pick specific interval points to make their recording (time point) or they may create a series of intervals within the observation (time interval) and record the number of events in each interval.
if I wanted to observe the anxiety people felt during a horror film, I might use a time sample. In terms of measuring ‘anxiety’ I could record the participants pulse rate every 15 minutes, throughout the film and then create an average score for each participant and for each time interval.
STRENGTH:
- researchers do not get overwhelmed with data unlike an issue that may exist in event sampling
WEAKNESS:
- can be more complex to organise than event sampling. The researcher needs to keep track of the different time intervals when they are recording events, which can be quite confusing if they are observing multiple behaviours or participants.
Experimental design: Independent groups
- participants take part in only one experimental condition
- researcher may randomly allocate participants to either a control or experimental condition but not both
STRENGTH: - no order effects as participnats only take part in one condition
WEAKNESS: - no control over participant variables
- need twice as many participants as repeated measures design to produce same amount of data.
Experimental design: Repeated measures
- participants take part in both the control and experimental condition
- The performance of the participants on the control is compared to their performance on the experimental. If there is a significant difference in the performance of the participants in the two conditions, we can say that the IV has caused an effect on the DV.
STRENGTHS: - need half as many participants as independent groups design to produce same amount of data
WEAKNESS: - demand characteristics
Experimental design: matched pairs
- form of independent groups design where the experimental and control participants are deliberately similar e.g. there is a balance between gender and IQ levels in each condition.
- no random allocation of participants. researcher uses pairs who have been matched for any participant variable that may influence performance on DV.
- matching process can be costly of time and money but eliminates participant variables present in independent groups design.
STRENGTH - eliminates problem of participant variables that exist within independent groups design
WEAKNESS: - demand characteristics and order effects
Levels Of measurement: Nominal Data
- shows categories of data represented by frequencies. Data sets have no relative numerical value, e.g. boys and girls
- least sophisticated form of data. NOIR (nominal least, ordinal, interval, ratio most)
- Sophistication refers to how much detail. nominal gives basic detail e.g. seven calssmates own a dog and 7 dont. No info about why this is the case or types of dog etc.
- Sophistication matters as the level impacts what statistical test is used and indicates how complex the data is, which impacts the genralisability of the research results.
Levels of measurement: Ordinal data
- ## data can be placed into ascending or descending order, but intervals between data not necessarily equal e.g. the times for 1st, 2nd and 3rd in a race.
Levels of measurement: Interval Data
- equal numerical intervals between scores e.g. temperature. The interval between 1 and 2 degrees is the same as between 22 and 23 degrees.
Levels of Measurement: Ratio data
- equal intervals between scores and has an absolute or true zero point e.g. speed (mph).
- Most sophisticated data
Reliability: Internal reliability
The extent to which a test or measure is consistent within itself. e.g. the use of standardised procedures and instructions for all participants. e.g. all the questions on an IQ test (form of questionnaire) should be measuring the same thing. This may not be relevant to all questionnaires because sometimes internal consistency is not important, such as a questionnaire about fear experiences may look at many different aspects of being fearful.
- issues can arise when standardised procedures are not used. Cannot establish cause and effect relationship if everyone is not given same experiment.
- estimations and self reports not reliable as difficult to compare.
- mean and standard deviation reliable descriptive statistics. More reliable than mode and range.
- operationalised variables increases internal reliability.
Reliability: External reliability
- the extent to which a test produces consistent results over several occasions. e.g. if an interviewer conducted an interview and then conducted the same interview with the same interviewee a week later, the outcome should be the same, or the interview is not reliable.
- high internal, higher external
- more controlled experiment/study has higher external reliability. e.g. field experiments have higher amounts of extraneous variables so less externally reliable
- lab experiments are highly controlled, so high external reliability, but lowered external validity.
Assessing reliability: Inter-rater reliability
- when 2 or more psychologists produce consistent results by using a standardised procedure, to prevent variation of data collection by multiple psychologists/researchers.
- In observational research, the issue is that observations should be consistent. We would expect two observers watching the same behaviour to produce exactly the same data
- the extent to which two or more observers agree is inter-rater reliability
- measured by correlating the observations of two or more observers.
Assessing reliability: test-retest reliability
- testing and retesting same participants over time with the same test (when participants have had time to forget it) and comparing their scores. If the scores are the same, the test has external reliability.
- if the test produces scores, these can be compared by calculating a correlation coefficient
Assessing reliability: Split-half reliability
- splitting the test answers from a participant in half and seeing whether the individual got the same or similar scores on the 2 halves. If so, internal reliability is high.
- the scores can be compared by calculating a correlation coefficient
Validity: Internal Validity
- findings accurate and effects of DV caused by IV, so study measures what it intends to measure (confounding variables have been controlled and will not affect results)
- The degree to which an observed effect was due to the experimental manipulation rather than other factors such as confounding/extraneous variables
Validity: External
- whether the study depicts real life e.g. if task has mundane realism, and whether findings would apply to different places, different times or different people (population validity).
Validity: mundane realism
- whether the tasks are everyday usual occurences e.g. exams are normal but having an MRI scan is not so much.
Specific validity issues: researcher bias
- where the researcher either directly/indirectly influences results of a study through either the design of the study or through the way the research is conducted/analysed
Specific validity issues: demand characteristics
- type of confounding variable where participants unconsciously work out the aim and act differently (either through social desirability or the screw you effect- act in a way way that will deliberately ruin results).
Specific validity issues: social desirability bias
- participants give the response they think will show them in the best possible light. May mean responses are not a true reflection of real thoughts/feelings.
Ways of dealing with issues of validity: double blind procedure
- Neither researcher or participant knows true aims of study.
- overcomes researcher bias and reduces demand characteristics
Ways of dealing with issues of validity: single blind procedures
- participants completely unaware of research hypothesis until after their role is complete. The researcher knows the aims, so does not overcome researcher bias.
- overcomes social desirability bias and demand characteristics.
Ways of dealing with issues of validity: Changes to location of research and nature of task
- conduct research in a field rather than a lab. Have participant complete realistic everyday tasks.
Ways of dealing with issues of validity: sampling technique and target population
- changing sampling technique and target pop will improve external validity
Assessing validity: Face validity
- least sophistaicated measure
- whether the test appears (at face value) to measure what it claims, and so is subjective
- high face validity if purpose of test is clear even to naive respondents ( participants who do not know aims of study)
Assessing validity: predictive validity
- the degree to which a test can predict a future outcome of a more broad topic such as behaviour, performance or disease.
- The degree to which the findings can be applied to different more varied situations.
Assessing validity: Content Validity
- objectively measures whether the method of measuring behaviout is accurate and decides if it is a fair test which achieves the aims of the study (internal validity)
- can be carried out with an expert in specific area of behaviour to check if the test is valid.
Assessing validity: concurrent validity
- comparing a measurement with an established one that has known validity
- if both tests have similar results, then new test has concurrent validity, if not then new test needs to be redesigned and carried out.
Assessing validity: Construct validity
- most sophisticated test of validity as it looks at whether the overall results reflect the phenomena (external validity)
- achieved by checking the existing definitions of behaviour being studied and redesigning the test if it measures a different construct.
Ethics: Deception
- deliberately misleading of falsely informing participants about the nature of research (e.g. Milgram (1963) deceived participants as they believed they would harm participants by giving them the ‘shocks’ which were not actually real).
- overcome by use of roleplay, where participants play different parts in the study but still know
- if participants are deceived they cannot give valid consent
Ethics: valid consent
giving participants enough information (in a form they understand) so they can make an informed choice about whether the choose to participate
- deception means valid consent cannot be given as participant has not received enough information to make an informed choice.
Ethics: confidentiality
ensuring third parties are not able to trace information back to participants
- usually achieved by providing anonymity e.g. using participant numbers not names
Ethics: risk of stress, anxiety, humiliation or pain
- occurs in instances where research could induce more than minimal pain through repetitive or prolonged testing
- invasive testing such as the administration of drugs or vigorous physical exercise would not be encountered in everyday life and so is unethical (physical and psychological harm)
- research which induces unnessary stress or humiliation of participants
Ethics: risk to participants values, beliefs, relationships, status or privacy
- research that focusses on socially seneitive topics likley to have these risks e.g. sexuality, gender etc. and includes potentially sensitive data (e.g. confidential documents such as medical records)
- includes breaches of confidentiality such as showing recordings of participants without valid consent to do so
Ethics: working with vulnerable individuals
- vulnerable individuals face greater risks- researcher holds a position of power over these individuals
- granted additional protections through ethical guidelines and use in psychological research is restricted
- children under age of 16
- those lacking in mental capacity
- those in care
- people in custody (prison) or probation
- people engaged in illegal activities such as drug use.
Ethics: working with animals
- UK ethical guidelines introduced in 1986 as part of the Animals Act, since updated to the Animals (scientific procedures) Act (1986). BPS also provide guidelines.
- Psychologists must:
- follow relevant legislation
- replace the use of animals (only use them if nothing else can be)
- consider choice of species and strain
- reduce the number of animals used
- refine their procedures (no unnessary harm caused to animal)
- consider type of animals (only use captive bred animals when possible)
- provide care for animals
How can psychologists manage the risk posed by ethical issues in their research using ethical guidelines and debriefing?
- ethical guidelines in relation to working with human participants governed by British Psychological society (BPS)
- ethical guidelines provide moral principles that guide research
- one guideline is that participants must be DEBRIEFED to overcome ethical issues that may have risen from research. Aims to ensure participants aware of true aims of study (overcomes deception), to return participant back to original state.
Ethics: managing the risk posed by ethical issues (ethics committees)
- provide moral principles that guide research. They approve research before it goes ahead.
- administer a cost benefit analysis of research (considers the potential outcomes against potential costs)
- if benefits outweigh costs, they consider VALID CONSENT (whether participants know what they are signing up to in advance)
- if valid consent possible, likely study will be approved as participants can choose to not take part if they wish not to (right to withdraw)- can leave and have data removed at any time
- at end of study participants must be fully debriefed.
- committee may continue to monitor research after approval if e.g. participant risks are high, and they may stop study at any time if they deem it no longer ethical.
What is peer review?
- allows scientific community to ensure new information is valid. Stops low quality research being published and misleading scientific community.
- a peer would be another scientist who is in the same field as the researcher
- 2 types: pre-publication peer review (most common, most accepted in scientific community). post-publication peer review (newer, less accepted)
what is ‘format for reporting psychological investigations ‘?
- most scientists use a similar format to report on investigations.
- This means scientists know how to structure reports
- this format is also helpful as other scientists reading investigation can recognise features of report easily.
FORMAT:
- TITLE
- ABSTRACT (short summary with key details of investigation including reseacrh question, method, results, conclusion)
- INTRODUCTION
- METHOD (key details e.g. materials, participants, procedures etc)
- RESULTS (summary of key findings, may include result tables, graphs, statistics)
- DISCUSSION (main conclusions and possible implications (both theoretical and practical), limitations of research and suggestions for future/further research)
- REFERENCES (for all sources cited in article)
- APPENDIX (copies of any items used in research, such as scales used to measure variables).
Describe the process of peer review
- researcher submits article to journal
- journal assessed by editor of journal
- if accepted by editor of journal, article sent to reviewers who are experts in the field the article is written about
- reviewers normally kept anonymous from article’s editor
- reviewers submit comments to editor
- editor may reject article or return it to author to make revisions
- revised article re submitted to editor for publication.
What are descriptive statistics?
mathematical summaries of the data set. They either focus on various forms of average or on the spread of data. By looking at either one of these summaries, a brief insight into the overall results of the study can be gained and how likely it is that the data collected was reliable.
* two types:
* measures of central tendency: provide an average score from within the data set. It includes mean, median and mode, each one appropriate for a different situation
* measures of dispersion: provide an insight into how spread out the data was. The closer the scores are, the smaller the dispersion will be. Lower/smaller dispersions imply more reliable data with fewer anomalies. Include range and standard deviation
What is the Mean and how is it calculated?
- the arithmetic average of a data set. Takes the exact values of all the data into account.
- the mean is calculated by adding up all the data items and dividing by the number of data items
- can only be used with ratio and interval data
What is the Median and how is it calculated?
The Median is the middle value in an ordered list. All data items must be arranged in order and the central value is then the median. If there are an even number of data items there will be two central values. To calculate the median add the two data items and divide by two.
* can be used with ratio, interval and ordinal data
What is the Mode?
The Mode is the value that is the most common data item. With nominal data, it is the category that has the highest frequency count. With inteval and ordinal data, it is the data item that occurs most frequently. To identify this, the data items need to be arranged in order. The modal group is the group with the greatest frequency. If two categories have the same frequency the data have two modes, so are bi-modal
Evaluate the Mean (AO3)
- influenced by anamalous results, which may mean it is misrepresentative of the data as a whole
- cannot be used with nominal or ordinal data
Evaluate the Median (AO3)
- not influenced by anamalous results so can be useful under such circumstances
- appropriate for ordinal, ratio and interval data and can be easier to calculate than the mean
- BUT not as sensitive as the mean, as the exact values are not reflected in the media, where as the mean is the most sensitive measure of central tendency because it takes account of the exact distance between all the values of all the data.
Evaluate the Mode (AO3)
- only method that can be used with nominal data (data in categories)
- not influenced by anomalous results and is much more useful for discrete data and is the only method that can be used with nominal data (data in categories)
- may be more than one result (more than one mode, bi-modal), or no results at all if the data set is quite varied, so is not a useful way of describing this data.
What are measures of dispersion?
- a set of data can be described in terms of how dispersed or spread out the data items are. These descriptions are known as measures of dispersion
- include Range and Standard Deviation
What is the Range and how is it calculated?
- The Range is a value that shows the spread of data, representing the difference between the lowest and highest scores.
- in psychology, it is customary to add 1 (‘difference +1’) e.g. calculate the range taking the lowest score from in the set from the highest and add 1. This is known as Inclusive Range, which accounts for every score in the data set. The addition of 1 is because the bottom number, e.g. 3, could represent a value as low as 2.5, and the op number, e.g. 15, could represent a number as big as 15.5. This is done as a correction factor, so both the top and bottom scores are included.
- two data sets may have the same mean but a different range, so the range is helpful as a further method of describing the data. If the mean is the only method used to decribe data, the data would appear to be the same.
What is Standard Deviation and how is it calculated?
- Standard Deviation is a more precise method of expressing dispersion. It is a value which represents the amount of variation of results from the mean score. It is a measure of the average distance between each data item above and below the mean
- lower Standard deviation indicates a narrower distribution of results
- higher Standard deviation indicates a wider distribution of results
Evaluate Range (AO3)
- easy to calculate (strength)
- affected by anomalous results (weakness)
- fails to take account of the distribution of the number e.g. it does not indicate whether most numbers are closely grouped around the mean or spread evenly
Evaluate Standard Deviation (AO3)
- a precise measure of dispersion because all the exact values are taken into account
- not difficult to calculate if you have a calculator
- it may hide some of the characteristics of the data set e.g. extreme values
What is the Chi-squared Test and when should it be used?
- used with nominal data (data that are in categories), so the data are in frequencies. Frequencies must not be percentages
- used when there is an independent groups design
- a test of difference (enables us to consider whether or not two samples of data are different from each other e.g. may want to know whether people produce more accurate work in a noisy or quiet environment, so would be looking at the difference in participant’s performance in the two conditions) and a test of association (tests of association look at whether two variables both increase at the same time (positive association/correlation) or as one increases, the other decreases (negative association/correlation).
- Tests of difference generally used for experiments e.g. an experiment to see if noisy conditions reduce the effectiveness of revision
What is the Mann-Whitney U Test and when is it used?
- appropriate for tests of difference, where there is an independent groups design
- used when the hypothesis predicts a difference between two sets of data
- used when the data are at least ordinal (i.e. not nominal).
What is the Sign Test and when is it used?
- A Test of difference which enables us to consider whether or not two variables of data are different from each other.
- used with nominal data
- used when there is a repeated measures design (may be from a matched pairs experiment)
- used when the hypothesis predicts a difference between two sets of data
What are the key characteristics of a research aim?
- clear and concise statement of intent
- should be relevant to the field of psychology and contribute to existing knowledge
- identifies variables or behaviours being studied
- the aim must be clear, specific, testable through empiracal investigations and relevant to psychology.
- aims should be clearly defined to make it easier to replicate the study
What is meant by Standardised procedures?
a set of procedures that are the same for all participants in order to be able to repeat the study. This includes standardised instructions, which are the instructions given to participants to tell them how to perform the task.
What is an Experiment?
a research method where causal conclusions can be drawn because an independent variable has been deliberately manipulated to observe the causal effect on the dependent variable.
What is the difference between extraneous variables and confounding variables?
extraneous variables are any variables that are not the IV that may cause a change to the DV, if not controlled. Counfounding variables are a type of extraneous variable. Extraneous variables become confounding variables if they are not controlled and then do affect the results/have an effect on the DV.
Give examples of extraneous variables
- participant variables (e.g. age, intelligence)
- situational variables (e.g. experimental setting, surrounding environment)
- experimenter variables (personality, appearance, conduct of researcher)
What is Spearman’s rank order correlation coefficient test?
- a statistical test that is used when the research hypothesis predicts a relationship (Correlation)
- it is used when the data is related data (repeated measures or matched pairs design)
- it is used when the level of data is at least ordinal
