Paper 2- Research Methods Flashcards
Define an ‘aim’:
An Aim is a general statement of what the researcher intends to investigate, the purpose of the study
Define ‘Hypothesis’:
A hypothesis is a clear, precise, testable statement that states the relationship between the variables to be investigated. It is stated at the outset of every study.
Define ‘Directional Hypothesis’:
Direction Hypothesis:
States the direction of the difference or relationship, (one-tailed)(more or less?)
Define ‘Non-directional Hypothesis’:
Non-directional Hypothesis:
Does not state the direction of the difference or relationship (two-tailed)
(there will be a difference in people who drink coke to people who drink vodka)
What is an IV (independent variable)?
An IV is some aspect of experimental situation that is manipulated by researcher (or changes naturally).
The change of IV allows DV to be measured.
What is a DV (dependent variable)?
The DV is a variable that is being measured by researcher. An effect on the DV should be caused by change in IV.
When do you use a directional or non-directional hypothesis?
A directional hypothesis is used when a theory or findings of previous research suggest a particular outcome.
A non-directional hypothesis is used when there is no previous research or findings from earlier studies are contradictory.
What is an extraneous variable (EV)?
An EV is any variable other than IV that may affect the DV if it is not controlled.
What are confounding variables?
Confounding Variables are a kind of EV that vary systematically with IV. Therefore, we cannot tell if any change in DV is due to IV or confounding variable.
E.g. varying emotions with IV.
What are demand characteristics?
Demand characteristics are any cue from researcher or research situation that may be interpreted by PPs as revealing intentions of the study.
This may lead to change in PP behaviour within research situation.
What are Investigator Effects?
Investigator Effects are any effect of investigator’s behaviour (conscious or unconscious) on research outcome (DV). May include the design of study to selection of (and interaction with) PP during research process.
E.g. when expecting a certain result, investigator may stimulate certain response to favour their hypothesis
What is ‘randomisation’ in experiments?
Randomisation is the use of chance methods to control for effects of bias when designing materials and deciding the order of experimental conditions.
What is ‘standardisation’?
Standardisation is using exactly the same formalised procedures and instructions for all PPs in research study.
What is independent groups design?
PPs are allocated to different groups where each group represents one experimental condition.
1 group does condition A, drinking coke,
1 group does condition B, drinking vodka
What is repeated measures design?
All PPs take part in all conditions of expt.
Each PP would do condition A, drink coke, AND condition B, drink vodka.
What is matched pairs design?
Pairs of PPs are matched on some variable(s) that may affect the DV. Then 1 member of pair is assigned to condition A and the other is assigned to condition B.
Attempts to control confounding variable of participant variables (individual differences that may affect results)
What is random allocation?
An attempt to control for PP variables in independent group design which ensures the each PP has same chance of being in one condition as any other.
What is counterbalancing?
An attempt to control effects of order in repeated measures design. Half of PPs experience condition A first then condition B, and half PPs experience condition B first then condition A.
Evaluate Independent groups design:
- Negative: PP variables: individual differences between two groups in the two conditions. If there is a mean difference between the groups on DV, may be to do with PP variables rather than effects of IV (= confounding variable). Reduces validity of findings. How to deal? Use Random Allocation.
- Negative: More expensive than repeated measures as twice as many PPs needed to contribute the same number of results. Increasing time/money spent recruiting PPs.
- Positive: order effects not a problem (which is a problem for repeated measure design).
- PPs also less likely to guess aim of expt (less demand char.)
Evaluate Repeated Measures Design:
- Negative: each PP has to do at least 2 tasks and order of these tasks may be significant (order effects).
- Use of counterbalancing may help reduce order effects.
- Order effects may also arise due to boredom/fatigue, might cause deterioration in performance on second task. OR PP performance may improve in terms of practise (in skill-based task). Therefore, order is a confounding variable.
- More likely in repeated measures study that PP work out aim of study = demand characteristics.
- Strength?? PP variables are controlled (therefore higher validity) and fewer PPs needed (time+cost effective!!)
Evaluate Matched Pairs design:
- PPs only take part in single condition so order effects and demand characteristics are less of a problem
- Although there is some attempt to reduce PP variables, PPs can never be matched exactly (even identical twins!)
- Matching may be time-consuming and expensive (especially if pre-test required).
- Less economical than other designs.
What is a lab experiment?
A lab experiment takes place in a controlled environment where the researcher manipulates the IV and records the effect on DV, whilst maintaining strict control of EVs.
What is a field experiment?
A field experiment takes place in natural setting within which researcher manipulates IV and records effect on DV.
What is a natural experiment?
A natural experiment is where the change in IV is not brought about by researcher but would have happened even if researcher had not been there (e.g. PP was in car crash few months ago). Researcher records effect on DV. PPs may still be tested in a lab as only IV has to be natural, not necessarily the setting.
What is a quasi-experiment?
A quasi experiment is a study where the IV has not been determined by anyone, the variables simply exist (e.g. gender, race, height, weight, age). Strictly speaking it is not an experiment.
What is a strength and weakness of: Lab expt:
- Strength:
- high control over confounding variables and EVs.
- Researcher can be sure that any effect on DV is likely result of manipulation of IV.
- We can be more certain of cause and effect
- (high internal validity)
- Replication more possible than other expt types, due to high level of control. Replication is vital to check if findings are valid and not ‘one-off’.
- Weakness:
- lack generalisability- lab environment may be artificial and lack mundane realism
- In unfamiliar contexts, PPs may behave in unusual ways
- So their behaviour cannot be generalised beyond research setting (low external validity)
- Demand characteristics more prevalent: PPs aware that they are being tested in lab environment and may not behave naturally.
- The tasks may also lack mundane realism, not representative of everyday life - can results be generalised to normal life?
What is a strength and weakness of: Field expt:
- Strength:
- higher mundane realism than lab expt because environment is more natural.
- Field expt produce behaviour that is more valid and authentic.
- PPs may be unaware they are being studied (high external validity).
- Weakness
- loss of control of CVs and EVs; cause and effect between IV and DV more difficult to establish
- Precise replication of field expt is not possible
- Ethical issues: PPs unaware they are being studied, so cannot consent to being studied and so such research might constitute an invasion of privacy.
What is a strength and weakness of: Natural expt:
- Strength:
- Provide opportunities for research that are not undertaken for practical/ethical reasons, such as Romanian orphans by Rutter et al.
- They have high external validity because they involve study of real-world issues and problems as they happen.
- Weakness:
- Generalising findings difficult as there are reduced opportunities for research due to rare nature of naturally occurring event.
- PPs may not be randomly allocated to experimental conditions (only in independent group design).
- Therefore, cannot be certain that change in IV affects DV.
- If conducted in lab, then lack of mundane realism and demand char may be issue.
What is a strength and weakness of: quasi experiments:
- Strength: carried out in controlled conditions therefore share same strengths as lap expt (e.g replication, high internal validity)
- Weakness: cannot randomly allocate PPs to conditions and therefore may be confounding variables.
- Also, IV is not deliberately changed by researcher and therefore cannot claim that IV caused observed change.
What is random sampling?
All members of target population have equal chance of being selected. First obtain complete list of all members of target population. Assign a number to all names in list. Actual sample is selected through use of some lottery method (randomiser/picking from hat)
What is systematic sampling?
- A systematic sample is every nth member of target population is selected, e.g. every 3rd person.
- Sampling frame is produced, which is a list of people in target population into eg. alphabetical order.
- Sampling system nominated, e.g. every 3rd person or 6th or 5th
- May begin from randomly determined point to reduce bias
What is Stratified sampling?
- composition of sample reflects proportions of people in certain subgroups (strata) within target population or wider population.
- Researcher first identifies different strata that make up population.
- Then proportions needed for representative sample worker out
- Then PPs that make up each stratum selected using random sampling.
What is opportunity sampling?
- Researchers select anyone who happens to be willing and available.
- Researcher takes chance to ask whoever is around at time of study.
What is volunteer sampling?
- Involves PPs selecting themselves to be part of sample (self-selection)
- May place advert in newspaper or in online ad
- OR willing PPs may simply raise their hand when researcher asks.
Evaluate random sampling:
- potentially unbiased, so CV or EV should be equally divided between different groups, enhances internal validity.
- Difficult and time consuming to conduct. Complete list of target population hard to obtain.
- Sample may still be unrepresentative; could be that unrepresentative population is still selected by accident.
- Selected PPs may refuse to take part (becomes more of a volunteer sample)
Evaluate Systematic sampling:
- Sample method is objective.
- Once system for selection is established researcher has no influence over who is chosen
- method is time consuming, and PPs may refuse to take part.
Evaluate stratified sampling:
- produces representative population because it is designed to accurately reflect the composition of population
- Means that generalisation of findings becomes possible
- The identified strata cannot reflect all the ways people are different, so complete representation of whole population not possible.
- VERY time consuming and may be costly
Evaluate opportunity sampling:
- Convenient and cost effective and time effective.
- no need for list of members in target population or to divide population into different strata.
- BUT!! there are 2 forms of bias:
- sample is unrepresentative of target population as it is drawn from specific area, so findings cannot be generalised to target population.
- Also, researcher has complete control over selection of PPs, so may avoid people they do not like (researcher bias)
Evaluate Volunteer sampling:
- easy and requires minimal effort/input from researcher, less time consuming
- Researcher has PPs who are more engaged.
- Volunteer bias is a problem. Asking for volunteers may attract certain ‘profile’ of person, may want to please researcher, so can findings be generalised?
- e.g. college students may opt for research studies as it is easy way to make money. Sample = unrepresentative.
Why is deception an issue and how can it be dealt with?
- Deliberately misleading or withholding info from PPs at any stage of investigation.
- PPs have not received enough info when they gave consent to take part (or have been lied to) cannot be justified as informed consent.
- Can be justified if it doesn’t cause PP distress.
- Give a debrief, make PP aware of aims and details of investigation.
What is the BPS code of ethics?
- A quasi-legal document produced by British Psychological Society (BPS) that instructs psychologists in UK about what behaviour is and is not acceptable when dealing with PPs. The code is built around 4 major principles:
- Respect
- Competence
- Responsibility
- Integrity
What is Informed Consent and how does it prevent unethical practise?
- Prospective PPs should know what they are getting into
- Informed Consent = making PPs aware of the aims of research, procedures, rights (right to withdraw) and how their data will be used.
- PPs should make informed judgement whether or not to take part without being coerced or feeling obliged.
- From researcher’s p.o.v. asking for informed consent may make study meaningless because PP behaviour may not be as ‘natural’ as PPs are now aware of aims of study.
What is Protection from Harm and how does it prevent unethical practise?
- As a result of their involvement, PPs should not be placed at any more risk than they would be in normal life
- PPs should be protected from physical and psychological harm.
- Includes being made to feel embarrassed, inadequate, stress, pressure, mental torment, pain.
- PPs should also be reminded that they have the right to withdraw from investigation.
What is Privacy and Confidentiality and how does it prevent unethical practise?
- PPs have the right to control info about themselves
- This is the right to privacy
- If this is invaded then confidentiality should be protected
- Confidentiality is a part of the Data Protection Act
How would a researcher deal with informed consent?
- Consent letter
- details all relevant info that might affect their decision to participate.
- It must be signed for their consent to be given.
- for PPs under 16, a parental consent required.
- 3 types of consent:
- Presumptive consent - PPs themselves not asked, but a similar group. If they agree, then PP consent is presumed.
- Prior general consent - PP give permission to take part in multiple studies, including ones that involve deception. By consenting, they are consenting to deception.
- Retrospective consent - PP ask for consent (during debriefing), having already taken part in study.
How do researcher deal with deception and protection from harm?
- Given full debrief
- Made aware of true aims and details of expt
- Told how their data will be used and be allowed the right to withhold data.
- PPs should be reassured that their behaviour was typical/normal.
- If required, due to distress, give PPs counselling.
How do researchers deal with confidentiality?
- If personal details are held, they must be protected.
- Maintain anonymity
What is a pilot study and what are its aims?
- small scale run of actual investigation
- only handful of PPs
- tests procedure and checks if investigation runs smoothly
- includes questionnaires and interviews
- helps remove errors and introduce better measures, e.g. removing bad/ambiguous question form questionnaire.
- In observational studies, pilot studies used to train observers.
- Helps identify problems and modify design of study to save money and time in long run.
What is a ‘single-blind procedure’?
- Aims of study and details kept secret from PPs, until after study.
- Any info that might create expectations is not revealed until the end of the study to control for confounding effects of demand characteristics.
What is a ‘double-blind procedure’?
- Neither PPs nor researcher aware of aims of investigation (third party conducting investigation)
- Often used in drug trials
- Prevents influencing PPs behaviour.
What is a control group?
Control groups are used for comparison, if change in behaviour of experimental group is greater than of control group, assuming all CVs are constant, then cause of effect was due to IV.
What is covert observation?
Undisclosed: PP not aware.
What is overt observation?
- PP is aware that they being observed.
What is strength/weakness of covert observation?
- Strength: investigator effects and demand characteristics less likely
- Weakness: less ethical, lack of informed consent.
What is strength/weakness of overt observation?
- Strength: more ethical, informed consent obtained
- Weakness: investigator effects and demand characteristics likely.
What is participant observation?
Researcher takes part (close proximity)
What is non-participant observation?
Researcher does not take part (not close proximity)
What is strength/weakness of participant observation?
- Strength: in depth data gathered due to proximity, e.g. body language, facial expressions, tone of voice.
- Weakness: Investigator effects and demand characteristics likely.
What is strength/weakness of non-participant observation?
- Strength: investigator effects and demand characteristics less likely.
- Weakness: researcher might miss behaviours of interest due to lack of proximity.
What is naturalistic observation?
Takes place in naturalistic (unaltered) setting
What is controlled observation?
Takes place in artificial (controlled) setting.
What is strength/weakness of naturalistic observation?
- Strength: higher level of ecological validity
- Weakness: issues with replication and ascertaining reliability.
What is strength/weakness of controlled observation?
- Strength: can be replicated to check reliability
- Weakness: lower levels of ecological reliability
What is structured observation?
Behaviour is coded in behavioural categories
What is unstructured observation?
Every instance of behaviour is recorded and described
What is strength/weakness of structured observation?
- Strength: higher levels of inter-observer reliability
- Weakness: less rich data can lack internal validity as researcher may miss important behaviours.
What is strength/weakness of unstructured observation?
- Strength: rich data leads to greater internal validity
- Weakness: lower levels of inter-observer reliability.
What is time sampling in observational studies?
Where an observer records behaviour at prescribed intervals, e.g. every 10 seconds.
What is event sampling in observational studies?
Where observer records the number of times the target behaviour occurs, e.g. using tally.
What is strength/weakness of time sampling?
Strength: time sampling allows for better use of time since fewer observations are made
Weakness: not every behaviour of relevance to investigation will be counted if it occurs in between time frames allocated.
What is strength/weakness of event sampling?
- Strength: every behaviour of interest to researcher will be counted from beginning through to the end of observation.
- There is possibility that some behaviours could be missed if there is too much happening at the same time, resulting in some not being coded.
What are questionnaires?
Set of written qs used to assess person’s thoughts and experiences. Used to assess the DV.
There are open and closed qs.
Open qs - no fixed range of answer (= qualitative data) difficult to analyse
Closed qs - fixed number of responses (quantitative data). Used for analysis but lacks depth and detail.
What are interviews?
A live encounter where one person, interviewer, asks qs to assess interviewee’s thoughts/experiences. The qs may be pre set or may develop as interview progresses.
- Structured interviews: pre-determined qs, asked in fixed order
- Unstructured interviews: like a conversation, no fixed qs. General aim that certain topic will be discussed. Interaction is free-flowing. Interviewee encouraged to expand and elaborate on their answers.
- Semi-structured interviews: between SI and UI. There is a list of qs but interviewer is free to ask follow-up qs.
Evaluate Questionnaires:
- Strength:
- cost effective
- large amounts of data quickly
- able to be distributed to large number of people
- large volume of data
- can be completed without researcher being present.
- Straightforward to analyse.
- Can use statistical analysis, graphs and charts.
- Weakness:
- Response may not be truthful
- respondents may be keen to present themselves in positive light, influence answers
- = social desirability bias.
- questionnaires often produce response bias, where respondents reply in similar way, e.g. always ticking yes or same end of rating scale.
- May be because respondents did questionnaire too quickly or fail to read qd properly.
- Acquiescence bias = tendency to agree with items on questionnaire regardless of content of question.
Evaluate Interviews:
- Strength STRUCTURED:
- straightforward to replicate due to standardised format
- format reduces differences between interviewers
- Weakness STRUCTURED:
- if not deviation from topic or elaboration, limits richness of data collected an limits unexpected info.
- Strength UNSTRUCTURED:
- more flexibility
- can follow up points as they arise, gain better insight, elicits unexpected info.
- Weakness:
- But…. leads to more interviewer bias
- analysis of unstructured not straightforward
- irrelevant info?
- drawing firm conclusions difficult?
General weakness:
Interviewees may lie for social desirability reasons.
BUT a skilled and experienced interviewer should be able to establish sufficient rapport with PP to sensitive and personal topics are discussed, = truthful response.
What are the different types of questionnaire?
How should one design an interview
- interview schedule : list of qs that interviewer intends to cover
- standardised to reduce contaminating effect of interviewer bias.
- take notes during interview or record and analyse later
- quiet room, away from distractions
- begin with neutral qs, to help interviewee feel relaxed = rapport establishment
- Interviewees should be reminded that their answers will be dealt with strictest confidence.
How to write good questions for interviews or questionnaires?
- Avoid overuse of jargon - make wording simple and understandable
- Avoid emotive language
- avoid leading qs
- instead - use neutral qs
- avoid double barrelled qs ( 2 qs in 1)
- avoid double negative qs (i am not unhappy with job)
Define correlation?
Correlation is the strength and direction of association between two or more co variables.
There is positive correlation (CoV are directly proportional)
There is negative correlation (CoV are inversely proportional)
There is zero correlation (no relationship between CoV).
No cause and effect, only established idea of a relationship.
Evaluate correlation:
- Strength:
- useful preliminary tool for research
- precise and quantifiable measure of how 2 co variables are related.
- Show pattern, form premise for future research
- quick + economical to carry out
- no need for controlled environment and no manipulation of variables needed
- data collected by others (secondary data) can be used, less time consuming than expts.
- Weakness:
- due to lack of experimental manipulation, only tell how variables are related, not why,
- no cause and effect
- not clear change in which co variable affects other
- could be that another untested variable is causing relationship (intervening variable, also known as third variable problem).
- largely misused and misinterpreted
- often in media, correlations are presented as casual facts, = misleading (often a third unaccounted variable)
How would you write a hypothesis for correlation?
Directional:
there is positive correlation between price of chocolate and quality rating
Non directional:
there is correlation between price of chocolate and quality rating
What are the measures of central tendency?
Mean, Median, Mode
mean =
representative of data as whole
easily distorted by outliers (extreme values)
median =
easy to calculate
less sensitive than mean
ignores extreme values which may be important
mode =
crude measure of central tendency, not representative of sample
when there are several modes, not useful measure
sometimes, mode is the only thing you can calculate, e.g. favourite dessert.
What are the measures of dispersion?
Range (biggest - lowest + 1) and standard deviation.
Range:
influenced by outliers
does not indicate whether most numbers are closely grouped around mean or spread out, unlike standard deviation.
Standard deviation:
larger standard deviation = greater dispersion of data, so not all PPs affected same way, wider set of results, may suggest anomalous results.
Lower standard deviation = data is tightly clustered around mean, all PPs responded in similar way.
Can be distorted by extreme value
more precise measure of dispersion than range.
What is a Bar Chart and when is it suitable to use one?
- Data can be visually presented on a bar chart, so difference in mean values can be easily distinguished.
- Used when data is divided into categories, known as discrete data. Often useful for nominal data sets.
- Each condition is shown on a separate bar.
What is a Histogram and when is it suitable to use one?
- Bars touch each other, showing that data is continuous, NOT discrete.
- X axis has equal sized intervals of single category.
- Y axis represents frequency.
- If zero frequency, no bar.
What is a Scattergram and when is it suitable to use one?
- Scattergrams do not depict differences, but associations between co-variables.
- Either co-variable occupies x/y axis.
What is a normal distribution?
- bell-shaped curve = normal distribution. It is symmetrical
- Within normal distribution, most items are located in the middle area of the curve with ver few at extreme ends.
- The mean, median, mode all occupy the same midpoint of curve.
- The tails of curve never cross x axis
- This is because extreme data is always theoretically possible.
- E.G. look at IQ distribution graphs.
Include positive and negative skews.
What are skewed distributions?
- A skewed distribution is a spread of frequency data that is not symmetrical, where data clusters to one end.
- Positive skew = type of frequency distribution where long tail is on + right side of peak, and most of distribution is concentrated on the left.
- Negative skew = type of frequency distribution in which the long tail is on – left side of peak and most of distribution is concentrated on the right.
What conditions meet the requirements for a Sign Test to be carried out
- Need to be looking for a difference rather than association
- We need to be using repeated measures
- We need data that is nominal data.
What is probability?
Probability is the likelihood that certain events will occur.
What is the null hypothesis?
Null hypothesis = states that there is no difference/association in population. Any difference/association found is likely due to chance.
What is the accepted level of probability?
0.05 (95%). This level means that findings are significant and will reject the null hypothesis and accept the alternative hypothesis.
Sometimes 0.01 is used (99%).
How is the critical value calculated in Sign Test and what does it show?
- When statistical test has been calculated, ( +s and –s are added up and less frequent one is chosen).
- Less frequent sign is chosen, = ‘S’
- Ensure that if there is the same value in both conditions, that that value should be ruled out and not included in sign test. Total Number, N, would be adjusted.
- Calculated value S needs to be compared to critical value on on a critical value table.
- Level of significance is generally 0.05
- N = number of PPs (also called degrees of freedom) (ensure PPs who had same values in both conditions are not included in this number)
- Recognise if hypothesis is directional (one-tailed) or non-directional (two-tailed)
- Compare the critical value to S.
- The calculated value of S must be equal to other less than (≤) the critical value at 0.05 level of sig.
What are case studies?
A detailed + in depth analysis of individual, group, institution or event. These often involve unusual individuals/events, or ‘typical’ cases as well on an individual basis.
Conducting case study involves production of qualitative data (sometimes also quantitative)
Researchers may construct ‘case history’, using interviews, observations, questionnaires, or a combination. The individual may be subject to experimental/psychological testing, which may produce quantitative data.
These are often longitudinal studies, take place over long period of time.
Data may be gathered from family or people around them.
Evaluate Case Studies:
- Strength: offer rich, detailed insights may provide info on unusual forms of behaviour. May be preferred over superficial data, like questionnaires. May also contribute to knowledge of typical functioning, like HM - MSM. Case studies can. beused to help make hypothesis for future studies, or creation of theories.
- Limitation: generalisation of findings issue when dealing with small sample sizes. Information gathered is based on subjective selection + interpretation of researcher. Personal accounts from PPs and their family may be prone to inaccuracy and memory decay. Evidence from case studies tend to be low on internal validity.
What is content analysis?
Content Analysis - type of observational research where people are studied indirectly via communications they have produced. May include spoken interaction (conversation, speech/presentation), written form (books, magazine, films). Aim is to summarise and describe this communication in systematic way.
Coding is the initial stage of content analysis, some data sets may be extremely large so this info needs to be categorised into meaningful chunks. Could be counting how many times a word is said to produce quantitative data. e.g. how many times women are portrayed as inferior in TV shows etc.
Thematic analysis is a form of content analysis but outcome is qualitative. The main process involves identification of themes (idea that is recurrent). These are likely to be more descriptive than coding units. Once researcher is satisfied that the themes cover most aspects of data they are analysing, they may collect new set of data to test validity of themes and categories. Then researcher will write up final report, using direct quotes from data.
Evaluate Content Analysis:
- Strength: useful as it can circumnavigate many ethical issues normally associated with psychological research. Much of the material that an analyst might want to study may already exist within public domain. This means no issues obtaining permission. Therefore, these communications have benefit of high external validity, may access data of sensitive nature provided the authors consent to use it. Content Analysis is flexible in that it may produce both qualitative and quantitative data…
- Limitation: People tend to be studied indirectly, so communications studied are usually analysed outside of the context within which it occured. There is. adanger that researcher may attribute opinions and motivations to the speaker or writer that were not intended originally. However, most modern analysts are clear about their preconceptions (‘reflexivity’). BUT, content analysis may suffer from lack of objectivity.
What is the role of peer review in scientific process?
Purpose of peer review = to assess validity of another study before it is published in a journal.
Peer review process on image.
Aims of Peer review:
1. Validate quality + relevance of research (all elements of report assessed)
2. Suggest amendments + improvements before publication
3. To allocate research funding
Evaluate Peer review:
Anonymity
It is usual practice that the ‘peer’ doing the reviewing remains anonymous throughout the process as this is likely to produce a more honest appraisal. However, a minority of reviewers may use their anonymity as a way of criticising rival researchers who they perceive as having crossed them in the past! This is made all the more likely by the fact that many researchers are in direct competition for limited research funding. For this reason, some journals favour a system of open reviewing whereby the names of the reviewer(s) are made public.
Publication bias
It is a natural tendency for editors of journals to want to publish significant ‘headline grabbing’ findings to increase the credibility and circulation of their publication. They also prefer to publish positive results (file drawer problem).
This could mean that research which does not meet these criteria is ignored or disregarded. Ultimately, this creates a false impression of the current state of psychology if journal editors are being selective in what they publish.
Burying groundbreaking research
The peer review process may suppress opposition to mainstream theories, wishing to maintain the status quo within particular scientific fields. Reviewers tend to be especially critical of research that contradicts their own view and much more favourable to that which matches it.
Established scientists are the ones more likely to be chosen as reviewers, particularly by prestigious journals and publishers. As a result, findings that chime with current opinion are more likely to be passed than new and innovative research that challenges the established order.
Thus, peer review may have the effect of slowing down the rate of change within a particular scientific discipline.
What are some implications for the economy of psychological research:
Attachment research into the role of the father
Attachment research has come a considerable way since John Bowlby first asserted that a child can only ever form a secure and lasting monotropic bond with its mother. Thus, at the time Bowlby was writing, childcare was seen as a mother’s responsibility, and hers alone, whilst the father was free to carry out his ‘natural’ role as provider for the family.
Morerecentresearchsuggeststhatthefathermayfulfilaqualitativelydifferentrole from the mother. Crucially, this role is no less valuable in the child’s upbringing.
Psychological research has shown that both parents are equally capable of providing the emotional support necessary for healthy psychological development, and this understanding may promote more flexible working arrangements within the family. It is now the norm in lots of households that the mother is the higher earner and so works longer hours, whilst many couples share childcare responsibilities across the working week. This means that modern parents are better equipped to maximise their income and contribute more effectively to the economy.
The development of treatments for mental disorders
Absence from work costs the economy an estimated £15 billion a year. A recent government report revealed that a third of all absences are caused by mild to moderate mental health disorders such as depression, anxiety and stress (The Telegraph 2014). Psychological research into the causes and treatments of mental disorders therefore has an important role to play in supporting a healthy workforce.
In Chapters we looked at treatment of mental disorders. Patients are able to have their condition diagnosed quickly and gain swift access to treatment. Many conditions are treated through the use of psychotherapeutic drugs such as SSRls for depression and OCD. Antianxiety drugs are used for stress conditions.
Referrals can also be made by GPs for psychotherapies such as systematic desensitisation or CBT. Individuals can also engage in self- treatment using similar methods.
This means that, in many cases, people with mental disorders are able to manage their condition effectively and return to work. Thus, the economic benefit of psychological research into disorders such as depression is considerable.
What is reliability?
measure of consistency
What are the different ways. of assessing reliability?
- Test-retest : administering the same test to same people on different occasions. If test is reliable, then results obtained should be same/similar. Must be sufficient time in between to ensure that PP cannot recall answer from previous occasion. The set of scores from questionnaire could be correlated to make sure they are similar. If correlation is a strong positive correlation, then reliability of test is assumed as good.
- Interobserver reliability - prevents subjectivity bias, need to watch same event but record data independently. Also called inter-rater, and inter-interviewer.
How is reliability measured?
Using correlational analysis. The correlation soefficient must exceed +0.80 for it to be reliable.
How can reliability of questionnaires be improved?
Using test-retest method; ensure correlation coefficient exceeds +0.80.
A questionnaire that doesn’t have strong positive correlation may need to be re-written.
How can reliability of interview be improved?
Use same interviewer everytime, interviewer must be trained to not ask leading Qs or ambiguous Qs, perhaps through use of structured interviews (fixed Qs).
How can reliability of observations be improved?
Ensure behavioural categories are operationalised, measureable, and self-evident (obvious). Categories must not overlap, and all possible behaviours should be covered on list.
How can reliability of experiments be improved?
Procedures must be standardised
What is validity?
The extent to which an observed effect is genuine/legitimate, is a measure of what was intended to. bemeasured (internal validity) and if the findings can be generalised beyond the research setting (external validity).
What is internal validity?
= whether effects observed in experiment are due to manipulation of IV and not some other factor. This other factor could be something like demand characteristics.
What is external validity?
= are findings generalisable to other settings
What is ecological validity?
= type of external validity, concerns generalising findings from study to ‘everyday life’.If the task that is used to measure DV in expt is not like ‘everyday life’ (low. mundane realism), then low ecological validity.
What is face validity?
Basic form of validity where measure is scrutinised to determine whether it appears to measure what it is supposed to measure. It is. away of assessing validity.
What is concurrent validity?
= extent to which psychological measure relates to existing similar measure. It is a way to assess validity.
What is temporal validity?
Temporal validity is the extent to which findings from research study can be generalised to other historical times and eras. (form of external validity).
How can validity be improved in experiments?
Using a control group, help distinguish if change in IV results in change in DV. Standardise procudres to minimise impact of PP reactivity and investigator effects on validity. Use single-blind (PP not made aware of aims of study reduces demand char) or double-blind procedures (third party conducts expt and PP unaware, reduced demand char and investigator effects).
How can validity be improved in questionnaires?
Most questionnaires + psychological tests incorporate ‘lie scale’ within questions to assess consistency of respondent’s response and control effects of social desirability bias. Validity may be enhanced by reassuring respondents that all data submitted will remain anonymous.
How can validity be improved in Observation?
Covert observations, so behaviour of those observed is natural + authentic.
How can validity be improved in qualitative research?
Demonstrate interpretive validity, which is extent to which researcher’s interpretation of events matches that of PPs. This can be seen through coherence of researcher’s narrative and inclusion of direct quotes, and also using ‘triangulation’, using number of diff sources as evidence.
What are all the inferential tests and their positions on the table?
What are features of science?
- Constructing Theories and testing hypotheses.
- falsifiability
- replicability
- objectivity
- having a paradigm
- peer review process
How does theory construction occur? Describe both inductive and deductive scientific methods.
What is a Type I error
When null hypothesis is rejected and alternative hypothesis is accepted, when it should have been the other way around, as in reality the null hypothesis was true. (often called false positive, researcher falsely claimed to find significant difference)
What is a Type II error?
Null hypothesis is accepted and alternative hypothesis rejected, when it should have been the other way around. Called false negative.
What is a paradigm?
A set of shared assumptions and agreed methods within scientific discipline.
What is a paradigm shift?
Result of scientific revolution when there is significant change in dominant unifying theory within scientific discipline.
What are the sections of a scientific report?
- **Abstract **- short summary includes aims, hypotheses, procudres, results and conclusions.
- **Introduction **- literary review of general area of research, detailing relevant theories and studies related to current study, should follow logical progression
- Method - includes subsections: design, sample, apparatus, procedure, ethics
- **Results **- summary of key findings, often feature descriptive statistics (tables, graphs and charts, measures of central tendency, measures of dispersion), inferential statistics could. beapplied and critical values calculated and level of significance, any qualitative data should be summarised also in themes/categories.
- **Discussion **- summaries results, discuss limitations of investigation and suggestions of improvements that may be addressed in future studies. Inclusion of the wider implications of the research considered, like real-world application.
- Referencing
What does a reference look like for:
* books
* journals
* websites
