Research Methods Flashcards
What is operationalisation?
- process which contructs that can’t be directly measured and redefined in a measurable way
What is the difference between psychological and physiological questionnaires?
Psychological: more subjective
Physiological: more objective
Outline the operationalisation process
- construct: concept being studied
- conceptual definition: theoretical definition of construct
- operational definition: how the construct is measured in study
What is the difference between a strong and weak operationalisation?
Strong:
- discriminative
- supported by previous research
- use multiple questions to measure complex constructs
- explicit for replication
Weak:
- vague
- doesn’t cvapture complexity of concept
List three common types of measurement
- self report
- observational
- physiological
What is the self-report measurement?
- P provide info about themselves -> interview or questionnaire
- feelings/ actions reported with accuracy
What is a negative of self reports?
- risk of response bias: tendency participants provide false or inaccurate response
What are observational measurements?
- observer records participants behaviours
- multiple observers used to reduce bias
- different types: covert and overt
- more objective
What are physiological measurements?
- records biological data
- uses specific hardware
List some criticisms of physiological measurements
- expensive and less accessible
- Includes level of subjectivity
- reductionist
What is nominal data?
- categorical data assigned numbers with no numeric meaning
- described using mode
What is ordinal data?
- categorical data with some order
- no equal intervals between values
- described using median, mode or range
What is interval data?
- numerical data
- values represent equal intervals between levels
- no true 0
- described using mean, median, range or SD
What is ratio data?
- numerical data
- values have equal distance
- true 0 -> reflects absence of variable
- described using mean, median, mode or SD
What are the 3 types of validity?
- content validity
- face validity
- construct validity
What is construct validity?
- testing if construct truly captures construct of interest
- testing how correct what is measured
What is face validity?
- does the construct appear to measure what it says to be measuring
- based on researcher’s subjectivity
What is content validity?
- does the measure capture all elements of construct
- based on researcher’s subjectivity; considered after reviewing previous research on construct
What is measurement error?
- instances measured value differs from true value
- measured value = true value + error
What is random error?
- occurs due to chance
- natural variability in measurement process
- unpredictable & occurs equally in both areas
How is random error reduced?
- taking lots of measurements
- using precise instruments
- control variables affecting constructs of interest
What is systematic error?
- occurs consistently in same direction across all observations
- due to issues in data collection process
- identified & avoided with proper design strategies
How do you limit the effects of systematic error?
- use multiple measures to collect data
- callibrate instruments consistently
- establish protocol & train experiments properly
Tell me the research process
Theory -> research question -> hypothesis -> study design -> data collection -> analysis & inference
What should a good theory have?
- provide reasonable explanation for construct
- provide benefit beyond current theories
What are the characteristics of a good research question?
- fills gaps in literature
- answers in doubt
- answers with important practical implications
When does the research question stage occur?
- before the identification of a theory or after forming a hypothesis based on theory
What is a hypothesis?
Educated guess or prediction on natural phenomenon
What are the 3 characteristics of a strong hypothesis?
- testable
- falsifiable
- makes a prediction
What is a null hypothesis?
- no difference or effect any experimentally observed difference due to chance
What is an alternative hypothesis?
- there’s a relationship between variables
What is the difference between a directional and non-directional hypothesis?
Directional: predicts expected direction of effect
Non-directional: effect predicted but direction not specified
- two-tailed
What is an independent variable?
- variable manipulated by researcher; predictor
What is a dependent variable?
- variable measured by researcher; outcome
What is an extraneous variable?
- variable not intentionally being studied
What is a confound variable?
- extraneous variable could influence outcome of experiment
What are social variables?
Qualities of relationship between participants and experimenters
What is experimenter bias?
Experimenter’s behaviour giving cues for desired response or behaving differently towards participants in different conditions
What are the 5 ways to control confounding variables?
- elimination
- constancy
- balancing
- double-blinding
- statistically
What is the control condition?
- group where IV not manipulated by researcher
What is the experimental condition?
- groups where IV manipulated by experimenter
What is the placebo control condition?
- participant receive something similar to treatment but lack effectiveness
What is the waitlist control condition?
- participants placed on waitlist and told will receive treatment once experimental group receive it
What is the comparison with best alternative?
P receive all treatment but experimental group receive treatment being tested, control group receive best alternative available.
What is a type I error?
- reject H1 when fail to reject it
- conclude IV had an effect on DV when it didn’t
- data doesn’t accurately reflect what it is in the world
What is a population?
- entire group of interest
What is a sample?
- subset of the group investigated to form inferences about population
Why might a sample not generalise to population?
- biased sample