Top

Day 2 Notes: Science of Change

June 20, 2013 by  

The science of measuring change

Science of Change

The challenge of Developmental Psychology is to apply the scientific method to people who are changing all the time. It makes the effort a unique science of change. 

Let’s try to make an outline:

  • Scientific Method
    • Observations = what senses tell
    • Inferences = conclusions drawn
    •  Theories: constructs
    • Models: variables
  •  Theories include:
    • Explanations
    • Descriptions
    • Predictions
  •  6 Characteristics of a theory
    • 1. Clear
      • Skinner is clear
      • Piaget is not
    • 2. Useful
      • Skinner & Piaget are useful
    • 3. Summarize facts
      • Organizing frameworks
      • Babylonian creation story
      • Jewish creation story
      • Theory of evolution
    • 4. Small number of assumptions
      • Morgan’s Cannon
    • 5. Internally consistent
      • Piaget is not consistent; written over many years
    • 6. Testable hypotheses
  • Verification: Did you see that?
  • Replication: Do it again?
    • Application:
      • What foods to eat
      • Acupuncture
      • Spankin
  • Theories determine
    • Which variables to study
    • What they look like (properties)
    • How they inter-relate
    • How to measure them
  •  Operational Definitions
    • Define in terms of operations
    • What is done
  • Chance & not chance
    • Normal curve
    • Symmetrical
    • Most people in the middle
  • 4 ways to use numbers
    • Nominal = use as names
    • Ordinal = place, rank or rating
    • Interval = count
    • Ratio = math numbers
  • Developmental Research
    • Hard to measure change
    • Too fast
    • Too slow
  • Principles
    • Qualitative vs quantitative
    • How ask question makes diff
    • Systematic underrepresentation
    • Lack of language
    • Some important behaviors are rare
  • Approaches to Research
    • Basic = improve theory
    • Applied = improve application
  • Data Collection Methods
    • 1. Documentation
      • Baby biographies
      • Diaries
    • 2. Observation
      • A. Naturalistic observation
        • Advantages
          • Non-lab environment
          • Less “artificial”
        • Disadvantages
          • Observer effect
          • Affected by presence of observers
      • B. Ethnography
        • in-depth descript of how live
        • emic perspective (insider’s)
        • etic perspective (outsider’s)
        • participant observation (long-term field study)
      • C. Structured observation
        • Advantages
          • Relatively naturalistic
          • Some lab-based controls
        • Disadvantages
          • Time consuming
          • Subjective results
          • Inter-rater reliability
      • D. Case history (case study)
        • Freud
      • E. Cohort effects
        • Baby Boomers
    • 3. Self-report
      • A. Questionnaires
        • Given to parents, friends, teacher…
        • Advantages
          • Convenient
          • Cheap
        • Disadvantages
          • Social desirability
          • Lack of objectivity
      • B. Interviews
        • Clinical interviews
        • Structured interviews
        • Targeted data (specific questions, open-ended)
        • Advantages
          • Can follow up kid’s answers
          • Probe topics of interest
          • Not stuck with ‘script’
        • Disadvantages
          • Diff. questions for each child
          • Difficult to generalize results
          • Experimenter demands
          • Leading questions
    • 4. Correlation
      • Relationship between variables
      • Paired observations
      • Simple correlations & used in Causal Modeling
      • Not prove cause-effect
        • Necessary, not sufficient
        • 4 possible interpretations
          • A can cause B
          • B can cause A
          • C can cause A & B
          • Chance
      • Significance
        • Relationship unlikely to be due to chance
        • Unlikely, not impossible
        • Shared variance: what have in common
      • Examples
        • Skill in baseball and shoe size
        • Shirt length and stock market prices
        • Ice cream sales and rates of drowning
        • Length of marriage and male baldness
        • Cause-effect inferences are at your own risk
      • Correlation Coefficient
        • Sign
          • Positive = When X goes up, Y goes up
          • Negative = When X goes up, Y goes down
          • Zero = When X goes up, Y wanders aimlessly
        • Magnitude
          • Between -1 and +1
          • Amount of relationship
    • 5. Experimentation
      • Observations (data)
      • Independent variable
        • Experimenter controls variable
        • Independent of the subject
        • Predictor
      • Dependent variable
        • Under control of subject
        • Outcome measure
        • Criterion
      • Random selection
        • Equal likelihood of being studied
      • Random assignment
        • Equal chance of getting wonder cure
      • Cause and effect inferences
        • More comfortable
        • Still inferring
    • 6. Meta-analysis
      • Comparing theories
  • Five Experimental Designs
    • A. Cross-sectional studies
      • Different groups, same time
      • 10, 20, 30 and 40 year olds
      • 1st, 3rd and 5th graders
      • Advantages
        • Snapshot
        • Least expensive
        • Quick estimate of developmental function
      • Disadvantages
        • Group data, not individual
        • Don’t know what will happen next
        • Cohort effects = not equivalent groups
        • Reading program
        • Year born
    • B. Longitudinal studies
      • Longitudinal = same group over time
      • Advantages
        • Avoid cohort effects
        • Act as own control
      • Disadvantages
        • Time consuming
        • Expensive
        • Subject loss; die, move away, quit
        • Practice effects
          • repeated exposure to a test situation
    • C. Split-Plot
      • Random assign to field (group)
      • Repeated measures (rows)
    • D. Sequential designs
      • Data collected as go
    • E. Delayed treatment
      • Group 1 right now
      • Group 2 in 6 months

 

 

Comments

Feel free to leave a comment...
and oh, if you want a pic to show with your comment, go get a gravatar!





Bottom