6 Research Methods Flashcards
Falsifiability
The possibility that a statement or hypothesis can be proved wrong
Objectivity
Measurement of data is not affected by the expectations of the researcher
Replicability
Recording procedures carefully in order for another researcher to repeat them and verify the original results
Empirical Methods
Methods of gaining knowledge which rely on direct observation or testing (not hearsay or rational argument)
Laboratory Experiments
Taken place in a controlled environment, where variables can be carefully manipulated
Strengths of Lab Study
-high internal validity (controls CV’s and EV’s, cause-and-effect shown)
-more repliciable as there are controlled procedures
Weaknesses of a lab study
-low external validity (generalisability and mundane realism)
-low ecological validity
Field Experiments
Conducted in a natural setting, where the IV is still deliberately manipulated by the researcher
Strengths of Field Study
-higher external validity (realism)
-high ecological validity
-lack of demand characteristics
Weaknesses of a Field study
-lower internal validity (less controlled) - less easy to replicate
-ethical issues (consent not possible)
Natural Experiment
conducted when it is not possible, for ethical, or practical reasons, to deliberately manipulate the IV - it occurs naturally
Strengths of Natural Experiment
-high external validity (real-world application)
-only option for ethical reasons
-less bias
Weaknesses of a natural experiment
-no manipulation of IV means you cannot be sure of the relationship between IV and DV (causal relationship)
-no random allocation (CV)
Quasi-Experiments
IV is simply the difference between people that exist, gender, age , disorder , etc.
Strengths of Quasi-Experiment
-can be carried out in a lab
-allows us to have a comparison between 2 people
Weaknesses of a Quasi-Experiment
-no random allocation
-no manipulation of IV (cannot claim cause-and-effect)
Independent Variable
The thing that is changed
Dependent Variable
The thing that is measured
Extraneous Variables
Anything that might have an effect on the DV - can be controlled by experimenter
Confounding Variables
Are not controlled for in an experiment - and which do affect the results
Demand Characteristics
PPTs second guess the aims of the study, alter their behaviour
Investigator Effects
Influence of researcher on DV/design decisions
Randomisation
Chance methods to reduce researcher bias
Standardisation
ensuring all participants have the same experience
Independent Groups
PPTs in each condition of the experiment are different
IG Strenghts
-no order effect
-less demand charactersitics
IG Weaknesses
-extraneous or confounding variables decrease validity
-less economical
Repeated Measures
All PPTs take part in all conditions
RM Strengths
-PPT variables controlled
-more economical
RM Weaknesses
-order effect (use counterbalancing)
-demand characteristics
Matched Pairs
Similar PPTs paired on PPT variables, allocated condition A or B
MP Strengths
-less PPT variables
-no order effect
-no demand characteristics
MP Weaknesses
-less economical
-cannot match perfectly
Naturalistic Observation
Real-life setting, researcher does not interfere
Naturalistic Obsv +
-High ecological validity
-Less Demand Characteristics
Naturalistic Obsv -
-Low Internal Validity
-Difficult to replicate (lack of control)
Controlled Observation
Set up for the purposes of the observation, PPTs are aware they are being observed, lab setting
Controlled Obsv+
-High Internal Validity
-Easy to replicate
Controlled Obsv-
-Low ecological validity
-More Demand Characteristics
Overt Observation
PPTs are aware they are being observed
Overt +
-Ethically transparent (informed consent)
Overt -
-Demand Characteristics
-Limited to contexts
Covert Observation
PPTs are not aware of the observation
Covert +
-Avoids DC
Covert -
-Ethical Concerns
-Difficult to not give away
PPT Observation
Observer observes from within - joins the group being studied
PPT Obv +
-Greater Insight to behaviours
PPT Obv -
-Observer Bias risk
-Observer effect
Non-PPT Observation
Observer watched from a distance, does not interact with PPTs
NPPT +
-Objective Observation
-Reduce Bias
NPPT -
-Limited insight
Structured Observation
Organised observation with behavioural categories and sampling procedures
Structured Obv +
-Easier to analyse
-Replicability
Structured Obv -
-Restricts data
-Reduced depth
Unstructured Observation
Researcher records all relevant behaviour without a system in place
Unstructured Obv +
-Rich, detailed data
Unstructured Obv -
-Difficult to replicate
-Subjective analysis
Event Sampling
Recording the behaviour every time it happens
Time Sampling
Time intervals - record who is doing the behaviour
Questionnaire
A set of written questions (items) used to assess a person’s feelings and/or experiences
Questionnaire +
-Easy to replicate
-Easily distributed
-Closed, fixed questions are easy to analyse
Questionnaire -
-Social desirability bias
-Anonymity can make me lie
Open Questions and Evaluation
Respondent provides their answers in words
+Not restricted
-Difficult to analyse
Closed Questions and Evaluation
Respondent has limited answers
+Easier to analyse
-Limited Response
Writing Good Questions
-avoid jargon
-avoid leading questions
-use appropriate language
-use of filter questions
Interviews
Face to Face interaction between interviewer and interviewee
Interviews +
-Better awareness of truthfulness
-Mostly flexible
Interviews -
-Risk of interviewer bias
Structured Interview and Evaluation
Pre-determined set of questions
+Easy to replicate
-Cannot deviate from the topic
Unstructured Interview and Evaluation
Free-flowing conversation, no pre-set questions
+Flexible
-Interviewer Bias
Semi-structured Interview
Set questions with follow up questions depending on the answer
Good Interviews
-Quiet room
-Rapport
-Ethics
Social Desirability Bias
Behaviours that present PPT in a positive light, giving socially favourable answers due to the presence of an interviewee
Interviewer Effect
PPT have demand characteristics due to the presence of an interviewer/investigator
Correlation
A method of data analysis used to find an association (relationship) between two co-variables
Where is a correlation shown?
Scattergraph
Correlation co-efficient
Represents the strength and direction of the relationship between the co-variables as a number between -1 and 1
How is the correlation co-efficient calculated?
Statistical Testing, Spearmans rho or Pearsons
Correlation analysis +
-Easily accessible
-Helpful in describing direction and strength of relationships
Correlation analysis -
-Does not show a causation, possibke third variable
Aim
Stated intentions of what questions are planned to be answered
Directional Hypothesis
One-tailed test
-outcome is greater or less
Non-directional Hypothesis
Two-tailed test
-there will be a difference
Null Hypothesis
Prediction of no difference
Pilot Study
Small-scale version of an investigation that takes place before the real investigation is conducted
Pilot Study aims
To check that procedures, materials, measuring scales, etc., work
-allow researcher to change anything if needed
Single blind design
PPT is unaware of the research aims of an investigation, researcher is aware
Double blind design
Neither PPT or researcher are aware of the research aims of an investigation
Control Group
Groups of PPTs who do not undergo a change in the IV condition, baseline behaviour measure
Confederate
Individual in a study who is not a real PPT but has been instructed by the researcher on how to behave
Random allocation
Technique used to reduce PPT variables, so each PPT has the same chance of being in any condition
Randomisation
The use of chance methods to control for the effects of bias when deciding materials and the order of conditions
Standardisation
Using the same standardised procedures for all PPTs in a study (avoids investigator effects)
Content Analysis
A research tool used to determine the presence of certain words, themes, or concepts within some given qualitative data
Steps in Content Analysis
-state aim and hypothesis
-decide sample
-read qualitative data and identify any recurring/emerging themes
-decide on units of analysis, develop a coding system
-analyse the findings and interpret them quantitatively in terms of the hypothesis
Thematic Analysis
Any emerging themes that are recurrent in the communication are studied in more depth, more descriptive than coding units
Opportunity sampling
Anyone in the vicinity who is willing and available
Opportunity sampling +
High ecological validity
Opportunity sampling -
PPT variables
Random sampling
All members of the target population have an equal chance of being selected
Random sampling +
Reduced Bias
Random sampling -
-May be an unrepresentative sample
-time-consuming
Stratified sampling
Reflects the proportions of people in subgroups of the target population
Stratified sampling +
Most representative
Stratified sampling -
-Time-consuming
-PPT variables
Systematic sampling
Every nth person is selected in target population
Systematic sampling +
Potentially unbiased
Systematic sampling -
-PPT variables
-may be unrepresentative
-time-consuming
Volunteer sampling
Self-selected sample, often replying to an advert
Volunteer sampling +
WIlling sample
Volunteer sampling -
Volunteer sample - certain personality (DC)
Informed Consent
-permission from PPT to use them and their data in the study
-parental consent for under 16
-given before the study
-informed on anything that may affect their willingness to participate
Dealing with informed consent
-PPTs should be issued with a consent letter or form detailing all relevant info and their right to withdraw at any time, if they agree it is signed
-Retrospective consent
-Presumptive consent
-Prior general consent
Retrospective consent
Consent given after the event
-full debrief
-right to withdraw
Presumptive consent
Find a similar group without consent and then the OG with consent
-consent is ‘presumed’
Prior general consent
PPTs consent to potential studies but they don’t know which one they will participate in
Deception
Deliberately misleading or witholding info from PPT at any time in the investigation
-cost-benefit analysis by ethics committee should be used
Dealing with deception
Full debrief should be given at the end of the study and should be given the right to withdraw
-must leave feeling the same way as they arrived
Protection from harm
PPTs should not be placed at any risk any more than their everyday lives, should be protected from psychological and physical harm
Dealing with protection from harm
-full debrief
-right to withdraw
-counselling
-cost-benefit analysis
Privacy/Confidentiality
Data should be confidential
Dealing with privacy and confidentiality
Anonymity
-numbers instead of names
-never broadcast footage
-photos not publicised
BPS code of ethics
A quasi-legal document to protect PPTs based on four principles
-respect
-competenece
-responsibility
-integrity
Ethics Committee
Weigh up costs and benefits before deciding whether a study should go ahead (cost-benefit analysis)
Peer Review
Assesment of scientific work by other specialists who are in the same field, to ensure that any research intended for publication is high quality
Main aims of peer review
-to allocate research funding
-to validate the quality and relevance of research
-to suggest amendments and improvements
Evaluation of peer review
-Anonymity: usual practice that the peer reviewing remains anonymous, more likely to produce a more honest appraisal. Although some may use it to criticise their rival researchers.
-Publication Bias: can create a false impression of the current state of psychology if journal editors are being selective in what they publish.
-Burying groundbreaking research: may have an effect of slowing down the rate of change within a particular scientific discipline.
Reliability
How consistent a measuring device is, if it is replicated, results should be the same
Internal reliability
Each PPT in a study is treated the same way
External reliability
Same/similar results found after repeated test
Assessment of reliability
-Test-retest reliability
-Inter-observer reliability
-Measured using a correlation (should exceed +0.80) for reliability
Improving reliability
Repetition of study (check results correlate)
Internal validity
The extent to which the observed results represent the truth in the population we are studying
External validity
Generalisable beyind research setting
-ecological
-population
-temporal
Ecological validity
Realisitic setting
Population validity
Applicable sample
Temporal validity
Is it valid now?
Assessment of validity
-face validity
-concurrent validity
Face validity
Whether it looks like it measures what it should
Concurrent validity
Whether findings are similar to those on a well-established test
Improving validity
Larger sample size, more realistic setting
Paradigm
A shared set of assumptions about a subject matter of a discipline and the methods appropriate to its study
Paradigm shift
A major change in how people think and get things done that upends and replaces a prior paradigm
Reporting psychological investigations
-title
-abstract
-introduction
-method
-results
-discussion
-references
-appendices
Qualitative data
worded data
Qualitative data evaluation
+rich detail
+greater external valdidity
-difficult to analyse
-subjective interpretations of conclusions
Quantitative data
Numerical data
Quantitative data evaluation
+easy to analyse
+objective - less open to bias
-less detail
-lower external validity
Primary data
First-hand data
Evaluation of primary data
+authentic
-time & effort
Secondary data
Second-hand data, books, websites etc
Evaluation of secondary data
+Cheaper
+Easily accessible
-more likely to be inaccurate
-could be outdated or incomplete
-less validity
-may not match researcher’s objectives
Nominal data
Qualitative data, not able to be ranked
Ordinal data
Scaled or ranked data, subjective ratings, score
Interval data
Ranked data with equal measurement intervals, pre-existing measurement scales
Ratio data
Same as interval but includes an absolute zero
Type I error
Researcher accepts the alternative hypothesis (rejects the null)
How to reduce chance of a type I error
Use significance level of P=<0.01
Type II error
Rejects the alternative hypothesis (accepts the null)
How to reduce the chance of a type II error
Use significance level P=<0.05
Assessing Internal Reliability
Split half method - split test into two parts, PPT complete both parts, test the strength of the correlation between the two parts of the measure
