Science illuminates human development, improves health and education, and reveals hazards.
Examples: reduced infectious diseases in children, progress against illiteracy and sexism/racism, and fewer early deaths due to scientific advances.
But science can also mislead or be misused; caution is required.
Correlation vs Causation
Correlation: a relationship where two variables tend to occur together; defined as Corr(X, Y).
Positive correlation: both increase or decrease together; negative correlation: one increases while the other decreases; zero correlation: no evident connection.
Correlation strength (numerical): range from −1.0 to +1.0.
Noteworthy thresholds: +0.3 or −0.3; astonishing when +0.8 or −0.8.
Correlations are not evidence of causation: two variables can be connected due to direct causation, reverse causation, or a third variable.
Examples (potentially misleading): first-born with asthma, teenage girls’ mental health, dentists with obesity rates; the dentist-obesity link could be due to a third factor.
Mantra: "correlation is not causation".
Quantitative vs Qualitative Research
Quantitative (quantity): numerical data; easier to replicate, compare across cultures, and less biased in measurement.
Trade-off: quantitative data may overlook nuance; qualitative data enriches understanding but is less generalizable.
Practical note: many studies benefit from combining approaches (mixed methods).
Measuring Education and Health
Debate: should we rely on tests as objective measures of learning?
Pro-test perspective: standardized tests indicate mastery; often used to determine promotion or school quality.
Anti-test perspective: tests can narrow learning to facts, neglect attitudes, creativity, and values.
Health measurement issue: living quality is not fully captured by physiological data alone (e.g., ventilator status vs. meaningful life).
Mixed methods in health/education yield richer, more verifiable details.
Example from very old vs. younger elders: quantitative health scores showed poorer physical health but better psychological health; qualitative quotes illustrated resilience and meaning.
Ethics in Research
Core requirement: follow ethical standards across all research.
IRB (Institutional Review Board) reviews ensure consent and minimize harm; pre-IRB-era research often violated ethics (especially with children, minorities, prisoners, or animals).
Informed consent and minimizing risk are essential.
Vaccines and outbreak ethics:
Ebola vaccine development showed urgency vs. long-term efficacy data.
Decisions often made under time pressure; sometimes long-term data are unknown.
During outbreaks, rapid action (e.g., vaccination of frontline workers) can be justified to save lives.
Global health ethics: balancing safety, efficacy, and equity; international cooperation and transparency are crucial.
Collaboration, replication, and transparency are essential to avoid biased or culturally driven conclusions.
Big Questions and Cautious Exploration
Five contentious questions highlighted:
1) Prenatal drugs: safety for fetuses.
2) Education: what knowledge and conditions prepare children for the future?
3) Poverty: what conditions enable healthy development?
4) Transgender children, contraception, romance: what evidence supports wellbeing and rights?
5) Family structure: single parenthood, divorce, same-sex marriage and optimal family life.
The takeaway: answers are not fully known; research is influenced by culture, politics, and personal values.
Scientists should pursue answers that benefit everyone, while acknowledging limits and diversity of opinions.
Toward Responsible Science
Next generations will build on what is known, while exploring unanswered questions thoughtfully.
Remember the core goal: help everyone fulfill their potential.
Critical mindset: think critically about what is known, what remains uncertain, and what questions should be asked next.