research methods Flashcards
what is an independent variable
variable that you manipulatein a peice of research. so effect on DV can be measured
what is a dependant variable
The one that you measure, any effect on dv should be caused by change in IV
operalization
defining variables in terms of how they can be measure
hypothesis
clear,testeable and prescisestatement stating relationship between variables
directional
predicting the direction of the effect of IV on DV
non directional
predicting the effect but NOT the direction.
aim
**simply a statement* about aims of investigation.
extraneous variable
- variable other than IV(change) that may effect the Dv if not controlled
- eg age of pp
confounding variable
- when EV has not been properly controlled and interferes with DV(results)
situational variables
- EVs linked to the situation rather than pps
eg, temperature - eg intructions,lighting,temp,noise
participent variables
- EVs linked to participents rather than situation
- eg age,experience,gender,personality
investigator effects
- Any effect on DV(measure) as a result of the investigators behaviour
- eg smilimg or nodding (intentional or subconcously)
Demand characteristics
- any cues from research or situation to the pp which may lead to pp chaning behaviour
- eg knowledge of study,order of tasks
- screw you or please you effect
types of validity
face validity
- whether it looks subjectivley promising
- appears to measure what it is supposed to.
concurrent validity
- extent to which a psycological measure compares to a similar existing measure
- results obtained should either match or be closely similar to the results of a well established and recognised test.
temporal validilty
- generalisability to other historial times and eras.
ecological validity
- extent to which findings can be generalised to other situations and settings
ways of improving validity
- using a control group in experimental research to asses whether changes in DV were due to effect on IV
- standardised procedures
- single&double blind trials to achieve the same aim
- questionaires(assuring pps all data is anonymous)
- observations
- qualitative methods-higher ecological validity
social desirablity bias
- type of response bias tendancy to answer questions in ways that represtent themselves in socially acceptable terms or attempt to gain approval from others
randomisation
- use of chance to reduce effects of bias from investigator effects
- eg for design of matirials,deciding order of conditions,selection of pps.
random allocation
- random allocation of pps to experimental and control conditions in an extremely important process in research.
- decreases systematic error
- individual dfferences less likely to affect results
experimental designs
Independant group design
- pps allocated into different groups
- each group represent one experimental condition
experimental designs
repeated measures design
- all pps allocated to all groups
- take part in all conditions
experimental designs
matched pairs design
- pps take part in only one experimental condition
- eg intelligence,gender,age
experimental designs
How do we deal with issues in MPD&IGD
- random allocation eg put names in hat and pick
- counterbalancing attempt to control order of effects in RMD
- ABBA technique half take part in A then B half take part in B then A
experimental designs
order of effects
- when pps are tested more than once in repeated measures
- can lead to better performance due to practice
- or tiredness and bordem
experimental designs
how do we deal with issues in repeated measures design
-
counter balancing reduces order of effects
*
experimental designs
problem with independant variables
- variation betweenpps can effect DV
- eg age,race,socialclass
- problem for independant measures control this by random assignmement
types of experiments
Labortary
- takes place in a controlled environment
- manipulating Iv recording effect on DV
- :)easier to replicate, standardised procedure
- :(may produce unatural behaviour,low ecological validity
- demand characteristic,investigator effects
types of experiment
Field
- done in every day life environment of pps
- still manipulates IV but IRL setting
- cannot really control EV
- :)behaviour more likely to refelct real life,higher ecological validity
*:( difficult to replicateðical issues as no consent
types of experiments
Natural
- done in every day life of pps but NO CONTROL over IV
- takes advantage of pre existing Iv that wouldve happend anyway
- :) high ecological validity,can provide opportunities for reasearch that may have not been taken for ethical issues
- :( naturally occuring event may only happen rarely,difficult to replicate
types of experiment
Quasi
- contain a naturally occuring IV(change)
- eg difference in people that already exists
- :) carried under controlled variables,easier to replicate, control on EV&IV effect and cause established.
- :( maybe confounding variables(when EV not properly controlled), cannot randomly allocate pps.
sampling techniques
population
general population in whole world
sampling technoques
target population
large group pf individuals researcher may be interested in
sampling techniques
sample
group of people who take part in reasearch drawn from target population
sampling techniques
opportunity sampling
- taking sample from people available at the time
- &fit for criteria
- :) easy and inexpensive to carry out
- :( may not be representive,could be subjective to bias
sampling techniques
volunteer sampling
- pps select themselves to be apart of sample
- :) easy and less time consuming
- :( asking may attract certain people leading to bias and less valid results
sampling techniques
random sampling
- everyone has equal change being selected
- eg ranomdly selected on computer
- :) no bias or researcher influence
- :( not representive of whole population
sampling techniques
stratified sampling
- people in certain subgroups within target pop
- randomly selected in proportional amounts
- :) increases chances of attaining representive sample
- :( takes more rescourses×
sampling techniques
systematic sampling
- listing all then selecting the nth person till you reach desired population size
- eg every 4th
- :) unbiased if randomised representitive data
- :( if not randomised can be biased
Bias
- if certain groups maybe over or underestimated within sample selected
- eg too many young people
generalisation
- extent to which the findings and conclusions from a study can be applied to a population
- only possible if sample of of pps is **representitave ** of population.
standardisation
- using the same procedures for all pps
- eg same environment,instructions and experience
ethical issues
BPS(british phsycological society) ethical issues
- depection
- protection from harm
- informed consent
- debreifing
- right to withdraw
- confitientiality
ethical issues
informed consent
- making pps aware of aims of research,procedure and their rights
- they can choose whether to take part or not
how to deal
- ensure pps have been given all info that may affect their descision
- under 16=parental consent signature
ethical issues
deception
-
deliberetly misleading or witholding information from pps
*informed consent not given
how to deal
- need for deception should be approved by ethics comitee
- debriefed at end of study
ethical issues
protection from harm
- pps not placed in any more risk than in daily lives
- protected from physical&physcological harm
- eg embaressed,stress,pressure
- they have the right to withdraw
how to deal
- at the end of study full debreif of the aims
- given right to withdraw data
- reassured behaviour is typical&normal
- may require councilling researcher may provide
ethical issues
privacy&confidentiality
- pps have right to control information about themselves
- right of privacy extends to where study took place
- geological location not named.
how to deal
- maintain anonymity eg numbers or initials of pps
- during debrefing pps reminded data is protected throughout process
ethical issues
debreifing
- post research interview inform pps true nature of study
ethical issues
right to withdraw
- pps have the right to withdraw themselves&data at any time.
pilot studies
what is a pilot study
- a small scale version of an investigation which is done before the real investigation is undertaken
- allow potential problems to be identified
- allows money&time saved in the long run
- self reports orinterviews important to reword or remove wods that may be ambigious orconfusing.
pilot studies
what is important to check?
- timings
- understanding of instructions
- ethical issues
- fatigue
- damand characteristics and investigator effects
- equipment functioning
*
single blind procedure
- researcher does not tell pps if they are being given a test treatment or a control treatment
- avoid demand characteristics
*
double blind procedure
- **neither pp or experimentor ** knows who is recieving particular treatment
- prevents bias due to demand characteristics or placebo effect
- reduces investigator effects
control groups&conditions
- set as a baseline resulys from experimental condition are compared to this one
- is change is siginificantly greater easier to conclude it was the effect of IV.
observational techniques
Naturalistic observation
- watching and recording behaviour in a natural setting wouldve taken place normally
- people expected to be observed
strengths&weaknesses
- :) high ecological validity as its naturally occuring behaviour
- :( cannot be replicated tocheck reliability as researcher is not in control of variables
observational techniques
controlled observation
- watching&recording behaviour from a structured envioronment
- less natural as ps may be aware they are in a study.
strengths&weaknesses
- :) more control over extraneous&confounding variabl,can be replicated to check reliability
- :( low ecological validity behaviour in artificial environment leading to demand characteristics
observational techniques
Covert observation
covert(undercover)
- pps are unaware they are being watched or observed
- without consent
- observer usually tucked away from view
strengths&weaknesses
- :) natural behaviour decreases investigator effects & removes demand characteristics
- :( ethical issues could be invading pps privacy as they didnt give consent
observational techniques
Overt observation
- behaviour is observed and pps are aware they are being watched
- they have given full consent
strengths&weaknesses
- :) ethically acceptable since conent is given
- :( can lead to demand characteristics, and more likely to have unatural behaviour through investigator effects through social desirability bias
observational techniques
participent observation
- observer becomes part of group that is being observed
*
strengths&weaknesses
- :) more insightful increasing validity of findings,unlikelyto overlook or miss any behavious
- :( could lead to researcher bias as they may lose nobjectivity their behaviour moight influence pps.
observational techniques
non participent observation
- observes from a distance not apart of group being observed,remains seperate
- moreobjective way
strengths&weaknesses
- :) can be less bias,less likley to identify with pps
- :( due to lack of proximity may overlook or miss behaviour, or open to observr bias eg steryotypes that they are aware of.
observational designs
behavioural catogories
- when target behaviour being observed is broken up into more prescise components that are observable&measureable
strengths&weaknesses
- :) more structured and objective
- :( difficult to make clear,all forms of behaviour must be listed can be unambigious
observational design
unstructured observation
everything
- consists of continious recording where researcher writes everything they see
*
stregths&weaknesses
- :) more richness&depth of detail increasing validity
- :( risk of observer bias,only see the behaviour they’re interested in,producing qualititative data
observational design
structured
- necessary to simplify target behaviours
- reasearcher quantifies what they are observing with a predetermined list of behaviours&sampling mehthods
strengths&weaknesses
- :) quantitative data produced easy to analyse and compare with other data,less risk of observer bias
- :( difficult to recieve high interobserver reliability,as filling predetermined lists is subjective
observational designs
time sampling(structured)
- recording/tallying behaviour within a certain **timeframe ** that is predetermined before study
strengths&weaknesses
- :) less time consuming as it reduces numbe rof observations,more structured&systematic
- :( important details can go missed data unpresenntitative of whole observation
observational designs
event sampling(structured)
- counting/tallying when particular behaviour is carried out.
strengths&weaknesses
- :) useful for infrequent behaviour,researcher still picks up behaviour that doesnt occur in regualr intervals
- :( important details of behaviour may be overlooked as its complex,counting errors,effecting validity
self report techniques&design
self report technique
- method in which a person is asks to state or explainir own feelings,opinions/experiences related to a given topic
self report techniques&designs
questionnaire
- set of written questions used to asses a persons thoughts/experiences
strengths&weaknesses
- :) cost effective,saves time&money,easy to analyse
- :( social desirability bias (changing answers to present themselve better)
- :( acquiesnce bias(tendancy to agree with items regardless how they feel)
- no researcher present unable to clear up misunderstandings
self report techniques&designs
open questions
- allow pps to elaborate on their answers and give more detail
strengths&weaknesses
- :) qualitative data rich in detail
- :( difficult to analyse
self report techniques&designs
closed questions
- offers a fixed number of responses eg choosing an option “yes or no”
*
strengths&weaknesses
- :) quantative data easier to analyse,saves money
- :( lacks depth
self report techniques&designs
interview
- live encounter where interviewer is asks a set of qs to asses interviewees thoughts/experiences
- qs may be preset or may develop
self report techniques&designs
structured interview
- predetermined list of qs
- asked in a fixed order
strengths&weaknesses
- :) easy to replicate,high reliability,quick
- :( cant elaborate, limited richness of data
self report techniques&designs
unstructured interview
- free flowing questions
- general aim, interviewee encuraged to expand on answers.
* :) greater flexibility,greater insight on interviewee
* :( analysi
self report techniques&design
designing interviews
- standardised list of qs
- rapport (establishing reationship)
- ethics
- interview scedual-qs intent to cover
- treat interviewees with confidence
self report design
likert scales
- indicates agreement or otherwise
- statement using 5 points
self report design
rating scales
- respondents identify a value that represents their feeling about a certain topic
self report designs
fixed choice option
- includes a list of possible options
- respondents requires to pick what applies to them.
self report designs
writing good questions
- overuse of jargon (using technical terms where those within particular feild or area will know)
- emotive language and leading qs(attitude toawrds topic is clear)
- doublebarrelled questions/double negatives(two qs in one,or difficukt to decifer) e.g i am not unhappy in my job(agree or disagree)
role of peer review
- allocation of research funding
- assess the quality&relevance of research
- suggesting improvements
- assessing the research rating of university departments
evaluation of peer review
- not always possible to find appropriate expert
- publication bias research may publish significant positive results, result can create false impression of current state of pyscology if editors are selective/bias.
- burying ground breaking research-may surpress research that might contradict the views of reviewer.rearch that poses a challenge or fresh/new less likely to be passed.
psycology&the economy
*
