Human Development – Research Paradigms & Time-Spanning Designs

Research Approaches

• Three core paradigms for studying human development:
  • Quantitative
  • Qualitative
  • Mixed-Methods
• Choice of paradigm must always align with:
  • Specific research aim
  • Specific research question(s)

Quantitative Research

• Definition: Collection of data that can be converted into numbers (scores, frequencies, amounts).
• Primary purposes:
  • Test theories or hypotheses
  • Examine relationships between variables
  • Offer generalisable, statistically verifiable findings
• Typical data sources & instruments:
  • Surveys with fixed response scales (e.g., $1\text{–}5$ Likert-type items)
  • Standardised tests & inventories (e.g., IQ tests, depression scales)
  • Observation schedules that yield frequency counts
• Example 1 – Prevalence of Bullying
  • Research question: “What is the prevalence of bullying among adolescents, and how does it compare to emerging adults?”
  • Hypothesis: Bullying rates differ across age groups.
  • Instrument: Survey asking “How often have you experienced bullying over the past week/month/year?” (scaled $1=\text{Never} \rightarrow 5=\text{Very often}$)
  • Expected finding: Both stages report similar prevalence of about $20\%\text{–}25\%$.
  • Classification: Non-experimental quantitative (observational, no manipulation).
• Example 2 – Experimental Distraction Study
  • Aim: Examine effect of distraction type on children’s reading comprehension.
  • Design: True experiment with random assignment.
  • Group 1: Read while up-beat music plays (auditory distraction).
  • Group 2: Read while visual movement occurs (visual distraction).
  • Outcome: Number of correct answers on passage-related comprehension test (quantitative measure of attention).
• Non-experimental quantitative scenarios:
  • Ethical/practical constraints prevent manipulation (e.g., cannot induce bullying).
  • Observational counts, e.g., bilingual word production – record number of words produced in a $5\,\text{min}$ play session.

Qualitative Research

• Definition: Collection and analysis of non-numerical data (words, images, observations, artefacts) to capture richness and complexity of human experience.
• Core goals:
  • Address how and why questions.
  • Illuminate participants’ subjective meanings, contexts, and perspectives.
• Common data-collection methods:
  • In-depth semi-structured interviews (individual)
  • Focus groups (shared experience + facilitator)
  • Participant observation, field notes, document or image analysis
• Example – Lived Experience of Bullying
  • Replace fixed-response survey with open-ended interviews among adolescents.
  • Capture emotional & psychological nuances: shame, fear, coping strategies.
• Hallmarks:
  • Flexible, less structured procedures; can adapt to participant responses.
  • Naturalistic settings (schools, homes, community spaces) rather than labs.
  • Particularly valuable for marginalised or under-represented groups—ensures their voices shape theory.
• Contribution to theory:
  • Through coding, thematic analysis, grounded theory, researchers generate context-embedded conceptual models.

Mixed-Methods Research

• Combines quantitative and qualitative strands in one project or programme.
• Rationale:
  • Capture breadth (numerical trends) and depth (lived meaning).
  • Enable triangulation—confirm, expand, or explain findings across data types.
• Core integration designs:
  1. Concurrent (Parallel) Design
  • Collect quantitative (e.g., survey) and qualitative (e.g., interviews) simultaneously.
  • Compare/merge results during interpretation.
  1. Explanatory Sequential Design
  • Phase 1: Quantitative survey → identify patterns (e.g., high vs. low depression scorers).
  • Phase 2: Qualitative follow-up interviews with selected sub-groups to explain scores.
  1. Exploratory Sequential Design
  • Phase 1: Qualitative exploration (e.g., focus group on what “bullying” means to young adults).
  • Phase 2: Develop/refine quantitative instruments based on Phase 1 insights; administer survey to larger sample.
• Strengths:
  • Produces comprehensive, valid understanding.
  • Allows instrument development rooted in participant language and context.

Time-Spanning Research Designs

• Goal: Determine when to measure development and how to infer change.
• Three archetypes:
  1. Cross-Sectional
  2. Longitudinal
  3. Cross-Sequential (Accelerated Longitudinal)

Cross-Sectional Design

• Definition: Collect data once from participants of different ages.
• Provides a snapshot of age-related differences.
• Example: Sample $10\text{-yr}, 15\text{-yr}, 20\text{-yr}$ olds simultaneously.
• Advantages:
  • Efficient (time & cost).
  • No attrition threats (single contact).
• Limitations & cautions:
  • Cohort effects: Differences may stem from historical/cultural contexts rather than age per se.
  • Influenced by contemporaneous events at time of data collection.
  • Cannot infer individual developmental trajectories or stability.
• Ethical/practical implication: Ideal for preliminary age comparisons but avoid over-attributing findings to developmental change.

Longitudinal Design

• Definition: Follow same individuals across multiple waves (weeks → decades).
• Enables:
  • Direct observation of developmental change.
  • Testing whether early factors predict later outcomes.
• Example research question: “Is IQ stable over the lifespan?”
  • Measure IQ at ages $4, 14, 24, 34, 44$; evaluate intra-individual stability/change.
• Advantages:
  • Insight into within-person trajectories & causal ordering.
• Challenges:
  • Time & resource intensive.
  • Attrition (drop-out) → potential selective attrition bias.
  • Practice effects: Repeated exposure may inflate scores independent of true development.
• Mitigation strategies:
  • Maintain engagement (incentives, regular contact).
  • Use alternate test forms to reduce practice effects.
  • Statistical techniques (weighting, multiple imputation) for attrition.

Cross-Sequential (Sequential, Mixed, Accelerated Longitudinal) Design

• Combines longitudinal tracking within cohorts and cross-sectional comparisons between cohorts.
• Structure:
  • Recruit multiple age cohorts at baseline.
  • Follow each cohort over several waves, creating overlapping age coverage.
  • Allows disaggregation of age effects, cohort effects, and time-of-measurement effects.
• Real-world exemplar: Longitudinal Study of Australian Children (LSAC)
  • Two cohorts:
  • B group (Baby): Started $2004$ at ages $0\text{–}1$.
  • K group (Kindergarten): Started $2004$ at ages $4\text{–}5$.
  • Follow-ups every $2$ years; Wave 11 launched in $2025$.
  • Provides up to $20$ years of within-cohort data and overlapping ages for cross-cohort comparison (e.g., compare both groups at $4\text{–}5$ years during Wave 3).
• Strengths:
  • Captures change more quickly than single-cohort longitudinal.
  • Identifies contextual era differences while still modelling trajectories.
• Drawbacks:
  • Still faces attrition and practice effects similar to longitudinal studies.
  • More complex sampling, data management, and analysis.

Practical & Ethical Considerations in Selecting a Design

• Align design with primary research question & theoretical aim.
• Evaluate feasibility: time frame, budget, participant accessibility.
• Ensure ethical compliance:
  • Avoid harmful manipulations (e.g., cannot experimentally induce bullying).
  • Provide informed consent, especially with minors.
• Plan for:
  • Long-term data management and participant retention (for longitudinal or sequential).
  • Appropriate analytical techniques to address potential biases (cohort effects, attrition).
• Aim: Generate valid, meaningful insights that accurately reflect developmental processes.