Research Methods Flashcards
What is the independent variable
The variable the researcher manipulates (experimental conditions)
What is the dependent variable
The variable being measured (must be operationalised)
What is an extraneous variable
Any variable that COULD affect the DV (not IV)
What is a confounding variable
A variable that HAS affected the DV (not IV)
What is a lab experiment
- Researcher directly manipulates IV to see effect on DV
- Controlled environment
- PPs randomly allocated to condition
What is a Field Experiment
- Researcher directly manipulates IV to see effect on DV
* Takes place in real world
What is a Natural Experiment
• Researcher makes use of naturally occurring IV to see effect on DV
• What is a quasi-experiment
- Researcher makes use of naturally occurring IV to see effect on DV
- IV is a difference between people that already exists (gender, age)
Strengths of Lab Experiment
• High level of control over variables.
- Prevents EV from becoming CV
• Cause and effect is easy to identify
• Easy to replicate by other researchers
- Check if findings are similar then the results are reliable
Weaknesses of Lab Experiments
- Strong chance of demand characteristics
- Setting is artificial
- Lacks mundane realism
- Lacks ecological validity
Strengths of Field Experiments
• Setting is real
- More mundane realism
- More ecological validity
• Cause and effect is easy to identify
• Less chance of demand characteristics
-PPs may not even be aware they are in a study
Weaknesses of field experiments
• Less control of EV
- DV effect wouldn’t be caused by IV
- Less valid
• Less control of sample
-Generalisability&Representativeness
• Difficult to replicate
Strengths of Natural Experiements
• High mundane realism
-High ecological validity
• Useful to use when it’s impossible/unethical to manipulate the IV/sample
Weaknesses of Natural Experiments
• Less control of EV
- DV effect wouldn’t be caused by IV
- Less valid
- Difficult to replicate
- Difficult to determine cause and effect
Strengths of Quasi-Experiments
• High mundane realism
-High ecological validity
• Useful to use when it’s impossible/unethical to manipulate the IV/sample
Weaknesses of Quasi-Experiments
• Less control of EV
- DV effect wouldn’t be caused by IV
- Less valid
- Difficult to replicate
- Difficult to determine cause and effect
What is an observation
Researcher watches/listens to PPs engaging in behaviour being studied
What is a Non-Participant Observation
Researcher is not directly involved with interactions of PPs
What is a Participant Observation
Researcher is directly involved with interactions of PPs
What is a Covert Observation
Psychologist goes undercover
Group is unaware they are being observed
What is an Overt Observation
Group knows about observation+psychologist researcher
What is a Natural Observation
- Researcher observes PPs in their own environment
* No deliberate manipulation of IV
What is a Controlled Observation
- Researcher observes PPs in controlled environment
* IV is directly manipulated
Strengths of Naturalistic Observation
• Pps usually unaware of being observed (covert)
- no observer effects
- no demand characteristics
- higher validity
• High mundane realism / ecologically valid
-results can be generalised
• Useful when manipulating IV is difficult or unethical
Weaknesses of Naturalistic Observations
• No control for EV
-results rendered invalid
• Cause & effect not always clear
-lack of control
• Risk of Observer Bias
- interpretations may be subjective, wrong
- results may be unreliable
Strengths of Controlled Observations
• Cause & Effect clear
-identify IV causes by change in DV
• EV controlled
-results more valid
• Yield qualitative data
-increases validity
Weaknesses of Controlled Observations
• Low Mundane Realism / ecological validity
-restricts results, can’t be generalised
• Observer effects may occur
- pps know (overt observation)
- social desirability bias
- data invalid
• Risk if Observer Bias
- researchers personal views, opinions influence data recording
- inaccurate
What is a Self Report Technique
Pps give info without researcher interference
What is an Interview
Researchers ask PPs questions face-to-face
What is a Questionnaire
PPs given written details of questions + instructions of how to record answers
What is a Structured Interview
• PPs all given same Qs in order
• Closed Qs
-therefore
• Quantitative data
What is an Unstructured Interview
- Informal in-depth conversational exchange
- Not planned, loose plan of themes to ask
• Open Qs
-therefore
• Qualitative Data
What is a Semi-Structured Interview
- Mixture of structured + unstructured interview techniques
- Mixture if closed + open Qs
- Mixture if Quantitative + Qualitative data
Strengths of Interviews
A
Weaknesses of Interviews
S
What is a Closed Questionnaire
A
What is an Open Questionnaire
A
Strengths of Questionnaire
A
Weaknesses of Questionnaires
Ss
What is a Correlation
- Technique used to analyse relationship of 2 Quantitative variables (co-variables)
- Data usually obtained from non-experimental source (eg survey)
Advantages of Correlation
• Allow strength of relationship to be measured precisely
• Allow things to be researched that can’t be manipulated experimentally (practical / ethical reasons)
-eg stress + cardiovascular disease
• Predictions can be made about one of co-variables based on other co-variable
- due to previous correlations
- eg conclude as stress increases, cholesterol increases
Disadvantages of Correlation
• Correlational analysis cannot demonstrate Cause & Effect
• Correlation ≠ Causation
-third unrelated variable may be cause which influences both (usually population density)
• Correlations only measure linear relationships
- Not Curvilinear Relationships
- positive relationship up to certain point then negative (+ vice versa)
Differences between Correlations and Experiments
A
What is an Aim
- Precise statement about purpose of study and what it intends to find out
- Should include what is being studied (DV) and what it’s trying to achieve
- Eg
What is a hypothesis
• Precise, testable statement about expected outcome of study
Whatre the two types of hypotheses
- Null hypothesis - IV will have no effect on DV
- Alternative hypothesis - IV will have effect on DV
- Directional hypothesis - states direction of predicted difference between conditions
- Non-Directional hypothesis - does NOT state direction of predicted differences between conditions
Whats a pilot study
- Small-scale investigation conducted before research
- Identifies whether there needs to be modifications in study design
- Help determine if full-scale study would be feasible / worthwhile
What does target population mean
- Group that researchers are actively studying
* Results would ideally be generalised to apply to this group
What is random sampling
- Every member of target population has same chance of being selected
- Easiest method is drawing sample of names of target population out of hat
What is systematic sampling
• Every nth person from a list is selected
What is stratified sampling
• Classifies target population into categories
• Randomly choose sample consisting of PPs from each category
-same proportion as appeared in target population
Whats is opportunity sampling
Selects PPs who are readily available + willing to take part
What is volunteer sampling
- People self-select to participate in a study.
* Usually research participation is advertised for people to take part
What is experimental design
- How PPs are assigned to different conditions
* Independent groups, repeated measures, matched pairs
What is independent groups
• Different PPs are used in each conditions
-Therefore independent
• Usually randomly allocated to balance PP variables
Strengths of independent groups
• Order effects don’t occur
-different PPs each condition
- Order effects is when sequence PPs take part in conditions influence behaviour
- Eg PPs better w/ practice or bored/fatigued
- Demand characteristics reduced, only one condition (less chance guess purpose of study)
- Saves time because conditions can be done at same time
Weaknesses of indepedent groups
- More PPs needed
* Different results may be because PPs variables not IV
What is repeated measures
Each PP tested in all conditions
Strengths of repeated measures
- No PP variables (same people all conditions)
* Half number PPs needed compared to independent groups design
Weaknesses of repeated measures
• Order effects may affect results
-Avoided with counterbalancing
- Half PPs do 1 condition first / other do other condition. Then reversed
- doesn’t eliminate order effects, but controls impact (+allows order effects to be distributed evenly across both conditions)
• Demand characteristics more likely to occur
• Takes more time
-especially if time gap between conditions is required
What is matched pairs
• Different PPs used in all conditions
• PPs in groups are matched on characteristics important for study (age, gender, education)
-identical twins often used
Strengths of matched pairs
- Less risk of order effects
- Less risk of demand characteristics
- Participant variables unlikely, groups are closely matched
Weaknesses of matched pairs
• Twice PPs required compared to repeated measures
• Matching PPs process is v difficult.
-may still have different motivation levels + fatigue
• Matching PPs process is v time consuming
What are pp variables
• Characteristics of PPs which may affect DV
-Eg IQ, age, gender, personality
• Random allocation of PPs to conditions w/ independent groups ensure no bias
What are environmental variables
• Characteristics of environment that may affect DV
-Eg temperature, time of day, lighting, noise
• Standardisation used to counteract this
-making sure conditions, materials, instructions same for all PPs
What are investigator effects
• Person collecting data has knowledge of research aim which affects data
- Eg observer bias
• Double blind technique used to overcome
-neither PPs nor investigator know hypothesis or condition PP is in
• Standardised scripts written to ensure investigator acts in similar way all PPs
What are demand characteristics
• PPs guess study’s purpose + expected behaviour
-Eg observer effects + interviewer effects
- PPs try to give researcher desired results
- Or PPs try to give researcher opposite of desired results (the screw-you-effect)
- Or PPs acting unnaturally out of nervousness or social desirability bias
What are operationalised behavioural categories
- Specific observable behaviours recorded during observation
- Behavioural categories represent more general construct under investigation
- Behavioural categories;
- allow observers to tally observations into pre-arranged groupings
- provide clear focus for researcher
- enable proposal of testable hypothesis
- allow more objective/scientific data recording
- provide data easier to quantify/analyse
- should result in greater reliability
What is observer bias
- Occurs when observer knows aims of study or hypotheses then influences observations
- Overcome with inter-observer reliability
- each observer agrees on interpretation of behavioural categories beforehand
- observes PPs independently
- recordings correlated
• Overcome with intra-observer reliability
-observation is video recorded so watched several times
What is event + time sampling
• Event sampling - recording every time certain behaviour (/event) occurs in target individual(s)
-Eg every time someone smiles
• Time sampling - recording behaviours within given time frames
-Eg every 30 seconds
How do you write good questions
• Clarity - no ambiguity for PPs. Avoid double negatives, double-barrelled Qs
• Bias - no leading Qs. No social desirability bias.
-Respondents may prefer to give answers that make them seem better (especially if Qs are socially sensitive)
• Analysis - Qs easy to analyse. Closed Qs easy to analyse compared w/ open.
-Though PPs may be forced to select answers that aren’t detailed enough
How do you write good questionnaires
- Filler questions - Adding irrelevant Qs distracts PP from Mai purpose, reduces demand characteristics risk
- Sequence of questions - start w/ easy Qs and save so anxiety / defensive inducing Qs for later
- Sampling technique - often use stratified sampling so no bias
- Pilot study - Qs tested on small group, refined later on
How do you record interviews
• Could take written notes throughout interview
-likely to interfere w/ listening skills
- If researcher doesn’t write smth PP may feel they said smth useless
- Could also be recorded audio or video
What are the effects of the interviewer
• Strength of interview compared to questionnaire is presence of interviewer who is interested in PP answer may increase amount of info provided
• Interviewers need to be aware of non-verbal communication (don’t sit with arms crossed, don’t frown)
+ listening skills (don’t interrupt)
What is informed consent
• PPs sign consent form
-explains general purpose of study
• No pressure to consent and withdraw at any time
• Investigators should inform PPs of objectives of investigation
-in some cases they can’t because demand characteristics
What is protection from harm
- Investigators have responsibility to protect PPs from physics & psychological harm
- Risk of harm should be same as normal life and study should be stopped if harm is suspected
- PP should leave research in same condition they entered it
What is right to withdraw
• At start of research all PPs must be aware they can leave at any time
-regardless of payment or inducement is offered
- Difficult to implement during covert observations
- If PPs appear distressed during study they should be reminded of right to withdraw
What is confidentiality
- PPs data not disclosed to anyone unless agreed in advance
- Numbers or letters used instead of names if research is published
- Confidentiality means data can be traced back to a name
- Anonymity means it cannot since PPs never provided their name
What is avoiding deception
- Withholding of info/misleading of PP is unacceptable if PP likely to show unease once found out about deception
- Intentional deception of PPs over purpose and general nature of investigation should be avoided wherever possible (medical/scientific justification)
- If unavoidable then ethics committee must approve. Cost-benefit analysis determine if research should be carried out
- Id deception used then real purpose of study/what occurred in other conditions must be explained after study (also told why deception was necessary)
What is debriefing
- PPs told aim of study and info about other conditions after study
- Researcher check on PP’s welfare and remind of right to withdraw and right to confidentiality
- Does not provide justification for unethical aspects of research
What is the process of peer review
• Research proposal submitted to panel of psychologists for peer review
-panel decides if research is worth funding
- Published in scientific journal for peer review again after research conducted + research report written
- Reviewers can
- accept manuscript
- accept with revisions
- suggest author makes revisions & re-submit
- reject without possibility of re-submission
What happens during peer review
- Psychologists conduct independent scrutiny of research report before deciding if should be published
- Psychologists work in similar field to research investigating
- Research considered in terms of
- validity
- significance and originality
- appropriateness of methodology
- experimental design
• Purpose is
- ensure quality + relevance of research
- ensure accuracy of findings
- evaluate proposed designs for research funding
• Prevents
- dissemination of irrelevant findings
- unwarranted claims
- unacceptable interpretations
- personal views
- deliberate fraud
Strengths of peer review
• Independent scrutiny increases probability of errors being identified
-authors are less objective about own work
- Double blind procedure used so researcher is kept anonymous + research who does peer review isn’t known
- Involves specialist psychologist in field judging work, have exceptional knowledge and expertise to make best judgement
- Not always possible to find appropriate expert
- Possible that poor research may be positively peer reviewed because reviewer didn’t understand it
Weaknesses of peer review
• Journals prefer positive results since editors want to increase standing of journal
- therefore published work biased
- therefore misperception of facts
- eg research finding gender differences more likely published than no differences in gender
- leads to misperception men and women are Very different
• Unfair process whereby some reviewers have connections with certain uni’s
-therefore favouritism or bias towards researchers dependent on institution
How can research on social influence improve economy
- Helps understand how behaviour and attitudes can be changed
- Used to encourage people to engage in more healthy behaviours (eating healthily)
- Results in healthier people therefore reduced pressure on NHS resources and improved productivity (less time off work sick)
How can research on memory improve economy
- Cognitive interview improved amount of accurate information collected from eyewitnesses
- Implication is amount of money spent on wrongful arrests/imprisonments & wasted police time is vastly reduced
How can research on attachment improve economy
- Research disputing Bowlby’s monotropic theory allows both parents to equally provide emotional support for children
- Households have more flexible working arrangements
- Examples include mothers earn more at work and fathers stay at home, couples share child care evenly
- Modern parents are better equipped to maximise their income and contribute more effectively to the economy
How can research on mental health improve economy
• Drug therapies research has been essential in reducing cost of mental illness to economy + returning people to work
• Findings lead to improvements in psych health/treatment programmes
-People manage health better, less time off work
• Findings lead to better ways of managing people prone to mental health issues whilst at work
-improving productivity
- May encourage investment from overseas companies
- Providing treatments but be financial burden.
- New therapies > old therapies
- New therapies £ > old therapies £
What is quantitative data
• Involves numbers and can be measured objectively.
-quantifiable
• Examples include
- dependent variable
- closed questions in questionnaires
- structured interviews
- tally of no. times behavioural category seen in observation
What is qualitative data
• Involves words
- data based on subjectve interpretation of language
- only quantifiable if data categorised + frequency counted
• Examples include
- Open questions in questionnaires
- Transcript from unstructured interview
- Researchers describing what they see in an observation
• Challenging to analyse because it relies on interpretation by the researcher
-prone to subjectiveness, bias, innacuracies
• Difficult to categorise into sensible number of answer types
What is primary data
Data collected directly by researcher for purpose of investigation
What is secondary data
• Information collected for a purpose other than current use
• Could be for different study. Could be different researcher
- Could use government statistics
• Substantial variation in quality + accuracy
What is a meta-analysis
- Combining results from different studies on specific topic for overall view
- Results generalised across larger populations, data viewed with more confidence
- Prone to publication bias, researcher leaves out studies with negative or non-significant results
What is a table
Table
What is a scattergram
Graph
What is a bar chart
Bars
What is a distribution curve
Curvy
What is desceiptive statistics
Mean median and mode
