Please enable JavaScript.

Coggle requires JavaScript to display documents.

Method 😱 - Coggle Diagram

- - - - Reactivity
- - - - Start theories
      - Apply theories
  - - - Median Split
      - Extreme Group
      - Pre-Screen
    - - Nonequivalent group design:
        
        Differential research
        
        Posttest-Only Nonequivalent Control Group
        
        Pretest-posttest nonequivalent control group
      - Pre-post design studies
  - - - BG design
      - Advantage
        
        Time efficient
        
        Less attrition
      - Disadvantage
        
        Maturation
        
        Cohort effect
    - - WS design
      - Advantage
        
        No Cohort effect
        
        Temporal Sequence
      - Disadvantage
        
        Attrition
- - - - Descriptive
      - Correlational
      - Experiment
  - - - Nuremberg Code (Nazi)
      - Syphilis study at Tuskegee
      - Stanford Prison study
      - Belmont Report
        
        Respect for persons
        
        Beneficence
        
        Justice
      - APA ethics code
        
        informed consent
        
        Debriefing
        
        Confidentiality
        
        Institutional Review Board
      - Jim Coan Beth Loftus lost in the mall
        
        repressed memory debate
    - - Minimal Risk
      - risk benefit assessment ratio
    - - Active
      - Passive
    - - Types of Misconduct
        
        data fabrication (making up data)
        
        data falsification (manipulating data)
        
        plagiarism (using source without credit)
        
        patchwork plagiarism (multiple source without credit)
        
        self plagiarism
        
        P-hacking
      - Hauser scandal (Behaviour neurologist who faked data)
      - Consequences
        
        Violate (researcher and personal) ethics
        
        Impedes the progress of science and waste resources
        
        Damages the reputation of all science
      - How Conduct happens
        
        publication bias (outcome of a result biases whether one submits or getting it published
        
        experimenter expectancy effects
        
        dropping participants (dropping participant for unfit data)
        
        file drawer problem
        
        Reasons To
        
        Reason not to
      - Detecting Conduct
        
        P-curve
        
        Simulation
        
        Frequency of the mode / Simonsohn's randomness
    - - False positive
      - Two cultrips
        
        Ambiguity
        
        Desire
      - Simmons's 4 research DF
        
        choosing among multiple dependent variables
        
        choosing overly small or large sample size, and continuous collecting data
        
        Using covariates and abusing statistics
        
        reporting subsets of experi-mental condition
      - Other 4 misbheaviour
        
        Experimenter expectancy effects
        
        Misrecording of data
        
        Using an inappropriate statistical test
        
        Dropping participants
      - Flake Fried’s Questionable measurement practice
    - - Personal solution
      - Being transparent
        
        What is the construct
        
        What is he measure?
        
        What is the operationalization
      - Reporting
        
        Reporting all experimental condition
        
        Reporting covariate
        
        Reporting what the result is if all observation are kept
      - Pre-registration
        
        Terminal Rule
        
        Pre-registrated data analysis
      - Designing the experimental well
        
        Design it appropriately.
        
        Measurement is everything.
        
        Avoid experimenter biases.
        
        Avoid inappropriate analyses.
        
        Report with transparency.
        
        Replicate.
      - Replication
- - - - self report measures
      - behavioral measures
      - physiological measures
    - - random error
      - systematic error
    - - Categorical
      - Quantitative
        
        nominal
        
        ordinal
        
        interval
        
        ratio
    - - biases
      - ceiling effect
      - floor effect
  - - - cohen’s kappa
    - - Cronbach's ALpha
  - - - construct validity (5 component)
        
        face validity
        
        content validity
        
        predictive validity
        
        convergent validity
        
        divergent/discriminant validity
    - - conceptual validity (The degree to which a specific hypothesis maps onto the broader theory being test)
      - internal validity
      - external validity
      - ecological validiy
        
        generalizability vs econological
        
        Mook
        
        Stroop
    - - covariation
      - temporal sequence
      - removing confound
- - - - Type 1 error
      - Type 2 error
      - false alarm
      - false positive
      - false negative
      - alpha error
      - beta error
    - - a priori power analysis
      - post hoc power analysis
    - - direct replication
      - conceptual replication
      - replications plus extension
    - - Differences in sample and setting
      - The actual underlying theory is wrong or needs to be modified
      - Original study study is flawed, even though the effect exist
      - Methodology wasn’t accurate to the original study
      - Effect is actually smaller than the original study reported
  - - - maturation
      - history
      - practice effects
      - instrumentation
      - regression to the mean
      - attrition
      - assignment bias
  - - - participant expectancies/ participant reactance
        
        revealed through demand characteristics
      - evaluation apprehension/social desirability
      - acquiescence
    - - experimenter expectancies
      - experimenter attributes
      - Examples
        
        Clever Hans
        
        Rosenthal & Fode (1963): right and dull rats
      - blind studies
      - double blind studies
    - - Generalizability of participants
        
        selection bias
        
        self selection effects
        
        WEIRD and homogeneous sample
      - Generalizability of research
        
        fatigue effect
        
        practice effect
        
        experimenter characteristic
        
        measurement
      - Nature of Internal Validity
- - - - probability
        
        simple random sampling
        
        stratified random sampling
        
        proportionate stratified Sampling
        
        cluster sampling
      - nonprobabiltiy
        
        conveince
        
        quota
  - - - add variance in the way you ask
      - avoid loaded question
      - multiple question for the same construct
      - positive and negative framing for reverse coding
      - be sensitive
      - avoid forced choice
      - avoid double barreled question
      - avoid negation
      - avoid jargon
      - keep it simple
    - - between 310
      - range and meaningfulness
    - - unipolar
      - bipolar
    - - Internal
      - Con/divergent
- - - - person confounds
      - environmental confounds
      - operational confounds
- - - - Holding variable constant
      - Matching the groups
      - Statistically control
      - Shortcomings
        
        Resource intensive
        
        Intrusive
        
        Impossible to control for everything
      - Randomization (assignment)
        
        Mechanism
        
        Advantage
        
        Disadvantage
      - Noise
    - - Measurement (IV + DV)
        
        Bad operationalization (Avoid!)
      - Manipulation (IV)
        
        Address directionality
        
        Address third variable
      - Comparison (DV)
        
        Experimental Group
        
        Control group
        
        No-treatment/passive
        
        Placebo/active
        
        Do: Run participant individually in multiple session
        
        Don’t: Run all participant all at once in a single session
      - Control (confound)
        
        Extraneous variables
        
        confounding variables
    - - covariation/correlation
      - Temporal sequence
        
        reverse causality issues
      - Eliminating (all relevant) confounds
        
        third variable problem
      - Internal Validity
    - - Participant characteristics
      - Orienting task
      - Materials
      - Assessment
    - - Ethics
      - logistics
  - - - Dittmar et al: Whether exposure to very thin body images causes young girls to experience negative feelings about their own bodies
      - Elliot & Colleague: Does red affect people’s performance on academic tasks?
      - Muller and Dweck: type of praise on persistence of student after hard questions
    - - Posttest
      - Pretest-Posttest (NOT MIXED FACTORIAL)
    - - More control
      - Indepdence
      - flexible
    - - Number of participant needed
      - Individual difference exist between grouos
      - Participant characteristics
    - - Randomization
        
        Restricted random assignment
      - Matching
      - Holding Constant
      - Composed of equivalent individuals
  - - - Concurrent-measure design
      - Repeated-measure design
    - - Minimal individual difference
      - Need less people/more power
      - More practical
    - - Sequence effect
        
        Attrition
        
        Fatigue effect
      - Carryover effect
        
        Order effect
        
        Pratice/Testing effect:
        
        Interference effect:
        
        Demand Charactersitics
    - - counterbalancing
        
        Helps with sequence and carryover
        
        Help with attrition
      - Using both WS and BS
  - - - Efficiency
      - Test boundary effect
      - Measure interaction
      - Virtually no disadvantage, with the only one being interpretation
    - - Interaction
        
        Disordinal
        
        Ordinal
      - Main drive by interaction
      - Simple effect
      - Main
    - - Bars vs Lines
      - Information to convey
        
        All IVs
        
        number Levels of IV
        
        Actual level descriptor
        
        Between or Within for each IV?
        
        Manipulated or measured for each IV?
        
        What the DVs are too