Design Strategy Slides Temporal Dimensions
Navigating Change in Child Development Research
Lesson 1: Design Strategy
The Developmental Question
How do we measure growth over time?
Ontogenetic Change
Changes within an individual as they age (true development).
Cohort Effects
Differences due to historical period or generational experiences.
Are they developing, or is the world changing around them?
Cross-Sectional Approach
Snapshots across age groups at one time point.
Mechanics
Compare Age Group A (5yr) vs. Group B (10yr).
Efficient: Data collected in one session.
The "Fatal Flaw"
Susceptible to **Cohort Effects**. We cannot observe individual pathways or stability of traits.
Example: Comparing digital literacy in 10-year-olds vs 70-year-olds. Is it age, or the era they were born in?
Longitudinal Approach
Following the same individuals over time.
Individual Paths
Tracks intra-individual change and stability.
Attrition
Loss of participants over time may bias samples.
Practice Effects
Do participants improve because of development or the test?
Gold Standard for stability, but expensive and slow.
Case Study: The Flynn Effect
The Observation:
IQ scores have risen approximately 3 points per decade since the early 20th century.
The Critical Thinking Challenge:
If you compared 20-year-olds in 1950 vs 20-year-olds in 2020 (cross-sectional), you'd see a huge difference. Is it biological evolution, or a cohort effect?
Potential Confounds
Nutritional improvements
Formal schooling expansion
Test familiarity (environmental)
Environmental complexity
Cross-Sequential Designs
The Hybrid Solution.
Cohort A (born 2010)
Tested at 5, 10, 15
Cohort B (born 2015)
Tested at 5, 10, 15
Why do this?
By comparing Cohort A at age 10 to Cohort B at age 10, we can **quantify the cohort effect** while still tracking **individual growth** within each cohort.
Cohort Confounds Worksheet Cohort Confounds
Research Methodology: Psychology of Childhood
Name: ________________________
Date: _________________________
Objective
Analyze the **Flynn Effect** as a primary case study for distinguishing between ontogenetic development and cohort-driven environmental shifts. You will evaluate the internal validity of cross-sectional findings and propose a cross-sequential correction.
Part 1: Deconstructing the Flynn Effect
James Flynn observed that IQ scores increase globally by roughly 3 points per decade. In a hypothetical study, Researcher A compares the IQ scores of 20-year-olds in 1950 (Mean = 100) and 20-year-olds in 2020 (Mean = 121).
1. Identify the research design used by Researcher A. What is the primary threat to internal validity in this specific comparison?
2. List three environmental or historical factors (cohort effects) that could explain this IQ rise without necessitating a change in "innate" intelligence.
Part 2: The Longitudinal Alternative
Researcher B decides to follow a single cohort born in 1960, testing them every 10 years until 2020.
3. While this controls for cohort effects, it introduces "practice effects." Explain why practice effects are particularly problematic in longitudinal intelligence testing.
4. Define selective attrition and explain how it might result in an overestimation of cognitive stability as the cohort ages.
Part 3: Proposing a Sequential Design
You are tasked with designing a study that can definitively separate **age-related cognitive decline** from **cohort-related cognitive gain**.
Draft your high-level design strategy below. Include:
At least two separate birth cohorts.
Specific ages/years for testing (e.g., test Cohort A at age 10 and 20; test Cohort B at age 10 and 20).
How you will isolate the cohort effect from the developmental effect.
Design Strategy Teacher Guide Facilitation Guide
Research Designs Comparison: Longitudinal vs. Cross-Sectional
Learning Objectives
• Differentiate between individual change (development) and group differences (cohort effects).
• Evaluate the trade-offs of longitudinal, cross-sectional, and sequential designs.
• Apply design strategies to correct for historical confounds (e.g., Flynn Effect).
Suggested Pacing
Hook: Flynn Effect Intro 15 min
Instruction: Design Showdown 30 min
Group Activity: Worksheet Synthesis 30 min
Debrief: Proposal Prep 15 min
Instructional Key Points
1. The Cohort Problem (Cross-Sectional)
Students often mistake age differences for age changes. Emphasize that in a cross-sectional study, age is confounded with the year of birth. Discussion Prompt: "If we compare the social media habits of 15-year-olds and 45-year-olds today, are we measuring development or a digital divide?"
2. The Attrition and Practice Problem (Longitudinal)
Longitudinal designs solve the cohort problem but introduce internal validity threats. Selective Attrition: Not all dropouts are random. High-achieving or high-income families often stay in studies longer, leading to an "optimism bias" in longitudinal data.
3. The Sequential Solution
Introduce this as the "gold standard." It allows researchers to check if a 10-year-old in 2010 behaves the same as a 10-year-old in 2020. If they don't, you've found a cohort effect.
Answer Key & Discussion Guide
Q1: Researcher A's Design
Cross-sectional. Primary threat: **Cohort Effects**. The difference is likely due to environmental changes (nutrition, education, tech) rather than individual biological development.
Q3: Practice Effects in IQ
Intelligence tests rely on novel problem-solving. If a participant sees the same pattern-matching puzzle every 10 years, they may develop a specific strategy for that test, masking any true cognitive decline or growth.
Q4: Selective Attrition
If individuals with lower cognitive functioning are more likely to drop out (due to health, mortality, or socioeconomic stress), the remaining group at age 80 will appear more "stable" or "advanced" than the original group at age 20. This creates a "survival of the fittest" bias.
Common Graduate Misconceptions
"Longitudinal is always better": Not true. If the research question is about a snapshot of current developmental stages (e.g., standardizing a new developmental milestone chart), cross-sectional is more appropriate and practical. Ethics Consent Slides The Ethics of Agency
Consent, Assent, and Vulnerable Populations
Lesson 2: Ethical Compliance
Why Children?
IRB Classification: Vulnerable Populations
Cognitive Limitations
Reduced capacity to understand long-term risks or abstract methodology.
Power Dynamics
Susceptibility to coercion (adult authority figures/researchers).
"The duty of the researcher is not just to do no harm, but to ensure the participant's agency is preserved."
The Two-Tiered Guardrail
Informed Consent
Legal Requirement
Given by parent/legal guardian.
Full disclosure of risks/benefits.
Legally binding.
Assent
Ethical Requirement
Given by the child (typically ages 7+).
Affirmative agreement to participate.
Developmentally appropriate.
Crucial: Failure to assent = Participant Withdrawal, even if parents gave consent.
Designing for Assent
Translating the "Why" and "What" for a 7-year-old.
Language
No jargon. Use "game" or "activity" instead of "protocol" or "stimulus."
Visuals
Use pictures or emojis to represent mood, activities, or the right to stop.
The "Stop" Button
Children must know they can quit at any time without getting in trouble.
Workshop Challenge:
"How do you explain a 20-minute eye-tracking study to a child who doesn't know what eye-tracking is?"
The Simulation
In small groups, one student will act as the **Researcher**, one as the **Guardian**, and one as the **Child (Age 7)**.
Researcher Goal
Obtain assent without coercion while being honest about risks.
Guardian Goal
Balance child's autonomy with the benefit of the study.
Child Goal
Decide if you feel comfortable. Ask questions about the 'game'.
Assent Architect Activity Assent Architect
Ethics in Child Research: Design Workshop
Name: ________________________
Date: _________________________
The Challenge
You are conducting a study on inhibitory control in 6-year-olds. The study involves a 15-minute computer task where children press a button for "rabbits" but not "turtles." They will wear a lightweight EEG cap (a "special hat with stickers") to measure brain activity.
Your Task: Translate this protocol into a visual assent form that a 6-year-old can understand and feel empowered to decline.
Part 1: The Script
How would you explain the "EEG cap" and "Inhibitory Control Task" using zero academic jargon?
The "What": Description of activities
Write your child-friendly explanation here...
The "Right to Stop": Empowerment Statement
How do you tell them they can quit?
Part 2: Visual Layout
Sketch a layout for a single-page visual assent form. Use symbols, simple icons (draw them in), and very large text. Ensure there is a clear "Yes" and "No" section for the child to mark.
I WANT TO PLAY
NO THANK YOU
Review Questions:
Does your form inadvertently use "bribery" (coercion)? (e.g., "If you play, you get a huge toy!")
Is the "No" option visually as attractive as the "Yes" option?
Ethics Facilitation Guide Facilitation Guide
Ethics in Child Research: Consent and Assent
Learning Objectives
• Apply federal IRB regulations (Subpart D) to developmental research.
• Differentiate legal consent from ethical assent.
• Facilitate an ethical negotiation between guardians and minor participants.
Suggested Pacing
Instruction: IRB Standards 20 min
Workshop: Assent Architect 30 min
Simulation: Role-Play 30 min
Debrief: Ethics & Autonomy 10 min
Simulation Scenarios
Divide students into triads: Researcher, Guardian, and Child (Age 7). Hand out the following scenario cards.
Scenario A
The Hesitant Helper
The Study: A social-emotional learning task using a puppet show and stickers.The Conflict: The parent is desperate for the child to participate (maybe they are a fellow academic). The child is shy and says, "I don't like puppets today."Facilitator Note: Observe if the "Researcher" pressures the child or respects the refusal immediately.
Scenario B
The MRI Explorer
The Study: A non-invasive fMRI scan (lying still for 30 minutes in a loud machine).The Conflict: The child initially says "Yes" but becomes visibly anxious as the researcher describes the "loud noises" and the "dark tunnel."Facilitator Note: Watch for how the Researcher handles ongoing assent—is assent a one-time signature or a continuous process?
Debriefing Questions
"How does the researcher balance the need for data with the ethical requirement to stop a session when a child shows subtle signs of distress?"
"What happens when a child assents but the parent revokes consent midway? Who has the ultimate legal and ethical authority?"
Common Pitfalls in Graduate IRBs
Assuming Assent is Voluntary: Children may assent just because they want to please the adult. Researchers must look for non-verbal withdrawal (turning away, silence, crying) as a revocation of assent.Coercive Compensation: Offering high-value rewards (like an iPad) can coerce low-income families into pressuring their children to participate. IRBs require compensation to be "nominal" so it doesn't cloud judgment. Coding Calibration Slides The Coder's Lens
Systematizing Observation in Child Research
Lesson 3: Observational Methods
Ecological Validity
Observing behavior where it actually happens.
The Power of Observation
Captures spontaneous interactions.
Does not rely on self-report (crucial for children).
Reveals context-dependency of behavior.
The Challenge
"Human behavior is messy. Without a rigorous system, observation is just anecdote. How do we turn a playground interaction into quantifiable data?"
The Coding System
Operationalizing messy behaviors.
Exhaustive
Every behavior of interest must fit into a category. No "leftovers."
Mutually Exclusive
A single behavior should only fit into ONE category at a time.
Low Inference
Describe *what* the child does, not *why* they do it (e.g., "Child hits" vs "Child is angry").
Inter-Rater Reliability (IRR)
The cornerstone of observational science.
Percent Agreement
Simple but flawed: doesn't account for chance.
Cohen's Kappa (\(\kappa\))
The gold standard. Statistically adjusts for agreement that happens by pure luck.
Reliability Benchmarks
< .40 Poor
.40 - .60 Fair
.61 - .75 Good
> .75 Excellent
Ready to Code?
We will watch a 5-minute video of a playground interaction. You must use the provided coding sheet to tally every instance of **Prosocial Behavior**.
Caution: Definition matters. Is 'sharing a toy' the same as 'inviting a peer to play'?
Playground Prosocial Tally Coding Calibration
Observational Methodology: Prosocial Playground Behavior
Name: ________________________
Date: _________________________
Operational Definitions
Category A: Sharing
A child gives up a toy, material, or space to another child, or allows simultaneous use of a resource previously in their sole possession.
Category B: Assistance
A child provides physical or verbal help to another child who appears to be struggling with a task (e.g., helping someone up, pushing a swing).
Observation Log (5-Minute Clip)
Watch the video carefully. Mark a tally each time a behavior occurs within the 1-minute time intervals.
Interval Sharing (Tally) Assistance (Tally) Total for Interval 0:00 - 1:00 1:01 - 2:00 2:01 - 3:00 3:01 - 4:00 4:01 - 5:00
Part 3: Inter-Rater Reliability Workshop
Compare your results with a peer. Use the matrix below to identify points of agreement (both coded 'yes') and disagreement.
Calculations:
Percent Agreement:
The Kappa Challenge:
Critical Analysis:
Identify one specific moment where you and your partner disagreed. Was the disagreement due to:
Vague operational definition Inattention (observer fatigue) Subjective bias (interpretation of intent)
Describe the conflict and your consensus resolution...
Coding Facilitation Guide Facilitation Guide
Observational Coding Systems and Reliability
Learning Objectives
• Design low-inference coding categories for behavior.
• Conduct real-time behavioral tallies with precision.
• Calculate and interpret Inter-Rater Reliability metrics (IRR).
Material Prep
Select a 5-minute raw video of children on a playground. It should include at least 10-15 distinct interactions.
Pre-Lesson Tally: It is essential that you code the video yourself before the lesson to provide a "Gold Standard" comparison for students.
Teaching Strategy: The Drift
1. Countering "Observer Drift"
Explain that as observers tire, they tend to become more lenient or "drift" away from the strict operational definition. Activity: After minute 3 of the video, pause and ask: "Is your definition of 'Assistance' still the same as it was in minute 1, or have you started counting 'verbal encouragement' too?"
2. The Kappa Calculation Discussion
While students will calculate percent agreement, steer the discussion toward **Kappa**. Why? If sharing is very common (90% of intervals), two people could agree 80% of the time just by guessing "yes" for everything. Kappa "penalizes" this high-frequency agreement.
Resolution Masterclass
Common "Gray Area" behaviors students will struggle to code. Use these to guide the consensus discussion.
The "Accidental" Assist
A child holds a door open for themselves, and another child happens to walk through.
Verdict: Usually NOT coded as Assistance unless there is an intentional pause or look toward the peer. Emphasize *operational* markers of intent.
The "Reluctant" Share
A child gives a toy only after a teacher prompts them.
Verdict: Often coded as "Sharing (Prompted)" in complex schemes. For this workshop, decide as a class if teacher-intervention disqualifies the prosocial label.
Meta-Methodology Discussion
"If our reliability is only .50 (Fair), should we throw out the data, or should we refine our definitions and start again? At what point does the cost of retraining coders outweigh the benefit of the observational data?"
Infant Insight Slides Non-Verbal Windows
Experimental Paradigms in Infant Research
Lesson 4: Behavioral Proxies
Measuring the Unspoken
How do we know what a baby knows?
Infants cannot speak, point, or follow complex instructions. Researchers must use **Behavioral Proxies**—measurable actions that represent internal mental states.
"If a baby looks longer at A than B, does that mean they *prefer* A, or that A is *confusing*?"
Core Inference
Behavior \(\rightarrow\) Proxy \(\rightarrow\) Cognition
The Boredom Effect
Habituation Paradigms
1. Habituation
Repeatedly show stimulus X. Looking time decreases as the baby gets "bored."
2. Test Phase
Show stimulus Y (slightly different). If looking time increases (**Dishabituation**), the baby can perceive the difference.
Newness = Interest
Violation of Expectation
The "Magic Show" Methodology
The Logic:
Infants will look longer at events that are physically impossible or mathematically incorrect.
Looking Time = Surprise.
Example: Mickey Mouse Math
Show 1 doll, hide with screen.
Hand adds another doll behind screen.
Screen drops. **Impossible Event:** only 1 doll remains.
The Validity Question
Does "looking longer" mean a baby has a concept of addition? Or are they just reacting to a visual discrepancy?
Rich Interpretation
Infants have innate core knowledge of number, physics, and agency.
Lean Interpretation
Infants are reacting to simple perceptual features (e.g., total surface area of dolls).
Proxy Paradox Analysis The Proxy Paradox
Infant Research: Evaluating Behavioral Inferences
Name: ________________________
Date: _________________________
Case Study: Wynn (1992)
In a classic Violation of Expectation (VoE) study, 5-month-olds saw one Mickey Mouse doll placed behind a screen, followed by a second. When the screen was lowered to reveal only one doll (the "Impossible" event), infants looked significantly longer than when they saw two dolls (the "Possible" event). Wynn concluded that infants have an innate "accumulator" mechanism for number.
Part 1: Rich vs. Lean Interpretations
Review the following two interpretations of Wynn's data. Identify the strengths and weaknesses of each.
Interpretation A (Rich)
"Infants possess an innate mental representation of number and can perform basic arithmetic (1+1=2)."
Identify one piece of evidence or logic that supports this interpretation...
Interpretation B (Lean)
"Infants are simply reacting to a change in the 'total amount of doll stuff' (perceptual area) or a violation of object permanence."
Identify one way a researcher could test Interpretation B specifically...
Part 2: Critiquing the Proxy
Consider the following "confounds" in VoE looking-time studies. How might they undermine the claim of "infant math"?
1. Low-Level Perceptual Salience: How does the visual complexity of two dolls vs. one doll affect looking time regardless of mathematical "correctness"?
2. Familiarity Bias: In some studies, infants look longer at what is *familiar* rather than *novel*. How would you know which bias is driving your infant's behavior?
3. The "State" Variable: Factors like hunger, fussiness, or parent positioning. How should a researcher control for these without biasing the sample through high attrition?
Part 3: Methodology Redesign
Propose a control condition that would isolate "numerical concept" from "visual area."
Think about varying the size of the objects.
Think about varying the color or shape while keeping number constant.
Infant Paradigm Facilitation Guide Facilitation Guide
Experimental Infant Paradigms: The Validity of Proxies
Learning Objectives
• Critically evaluate the link between behavior and cognitive inference.
• Differentiate between rich and lean interpretations of infant data.
• Identify confounding variables in habituation and VoE designs.
Core Concept
"Just because a baby looks longer doesn't mean they are thinking 'arithmetically.' In graduate research, the burden of proof is on the researcher to rule out perceptual low-level explanations before claiming cognitive high-level ones."
The Interpretation Debate
Interpretation A: The "Rich" View (Wynn, Spelke)
Argues for **Core Knowledge**. Infants have innate biological modules for number, space, and agency.Key Logic: If an infant reacts to 1+1=1 as "impossible," they must have a mental symbol for "2."
Interpretation B: The "Lean" View (Clearfield, Mix)
Argues for **Perceptual Salience**. Infants react to continuous variables (brightness, area, volume) rather than discrete number.Key Logic: Two Mickey dolls have twice the surface area of one. The baby is just reacting to the "amount of Mickey" they see.
Troubleshooting VoE Designs
Potential Flaw The Impact The Fix Familiarity Preference Baby looks longer at what they've seen before, not what is "impossible." Strict counterbalancing of conditions across subjects. Parental Influence Parents squeeze the baby or gasp when they see the "magic trick," cueing the infant. Parents must wear darkened glasses or listen to music on headphones. Coder Blindness Coders might look for "longer looks" in the condition they think *should* be longer. Coders should only see the baby's face, not the stimuli the baby is watching.
The Fatal Flaw Challenge
Ask students: "If we found that infants looked longer at a square turning into a circle than a square turning into a triangle, have we proven they understand geometry, or have we just proven they can see shapes? Where is the line between perception and cognition?"
Funding Frontiers Slides Funding the Future
Grant Writing and Proposal Mastery
Lesson 5: Project Synthesis
The Specific Aims
The most important page of your proposal.
Key Components:
**The Gap:** What is unknown in the field?
**The Innovation:** Why is your approach better?
**The Aims:** 2-3 concrete, testable goals.
"A reviewer should be able to read just this page and understand the entire logic, feasibility, and impact of your study."
The Methodology
Precision and Pitfalls.
Design Choice
Justify: Cross-sectional or Longitudinal? (Remember Lesson 1!)
Sample Power
How many children do you need? Account for attrition.
Ethics Plan
Describe your assent process for minors. (Lesson 2)
Why Grants Get Rejected
Avoiding the "Death Knell."
The "Fishing Expedition"
No clear hypothesis; just looking for "anything interesting."
Aim Interdependence
Aim 2 cannot be done if Aim 1 fails. This is a massive risk.
The Reviewer's Mindset:
"I have a limited budget. Why should I trust *your* design over the other 90 proposals on my desk?"
Your Master Proposal
You will draft a 2-page proposal for a study that addresses a developmental gap using one of the methodologies from this course.
Specific Aims
Methodology
Peer Review
Grant Proposal Builder Proposal Blueprint
Capstone Project: Methodological Design and Ethics
Principal Investigator: ________________
Date: _________________________
Section 1: Specific Aims
PAGE LIMIT: 1 PAGE
A. The Significance & Gap
Describe the current state of knowledge and the specific gap your study will fill. Why is this research needed now?
B. Aim 1
C. Aim 2
Section 2: Research Strategy
PAGE LIMIT: 1 PAGE
A. Design Selection
Justify your choice (e.g., Longitudinal, Cross-Sequential, Microgenetic). Explain how this design mitigates cohort effects or attrition.
B. Sampling & Power
Who are your participants? How will you recruit them? What is your plan for participant retention?
C. Ethical Protections (Assent)
Outline your specific plan for obtaining developmentally appropriate assent from minors and consent from guardians.
Grant Facilitation Guide Facilitation Guide
Grant Writing Mastery and Proposal Synthesis
Learning Objectives
• Synthesize methodological knowledge into a cohesive research proposal.
• Structure a high-impact Specific Aims page.
• Critically review peer proposals for methodological flaws and ethical gaps.
Final Session Pacing
Workshop: Drafting Aims 30 min
Instruction: The Reviewer's Eye 20 min
Peer Review Session 30 min
Final Debrief & Wrap-up 10 min
The Reviewer's Perspective
Prepare students for the peer review session by discussing these common NIH/NSF critique points.
Red Flag: Over-Ambitiousness
Proposing a 10-year longitudinal study with a 500-person sample on a $50,000 budget.
Advice: Guide students to scale their design to be feasible within a 2-3 year "Early Career" grant framework.
Red Flag: Descriptive Only
"We will observe children playing and see what happens."
Advice: Proposals must be hypothesis-driven. "We predict that X will correlate with Y because of Z."
Peer Review Protocol
Instructions for Students:
1
Read the Specific Aims page. Can you identify the "Gap" and the "Innovation" within the first paragraph?
2
Check the Methodology for a "Fatal Flaw" (e.g., confounding cohort effects, unethical assent process, or circular logic).
3
Provide one "Warm" feedback (strength) and one "Cool" feedback (critical design question).
Course Synthesis
"You have moved from understanding basic designs to critiquing the deepest proxies of the infant mind, and finally to architecting your own scientific inquiry. The ethics and methods you've practiced here are the foundations of trustworthy developmental science."
