Philosophy of Science

What science is, how it advances, and how different philosophical approaches understand it — each topic on its own page.

Philosophy of Science Approaches

  1. 01What Is Science?What distinguishes science from other kinds of knowledge?
  2. 02The Scientific MethodThe cycle of observation, hypothesis, test and revision
  3. 03The Problem of InductionHume's challenge to reasoning from past to future
  4. 04The Hypothetico-Deductive MethodDeriving testable consequences from hypotheses
  5. 05The Demarcation ProblemWhat separates science from pseudoscience?
  6. 06PositivismKnowledge only from observable facts
  7. 07Logical Positivism (Vienna Circle)Verifiability as the criterion of meaning
  8. 08Falsificationism (Popper)Science advances by attempts to refute, not to confirm
  9. 09Paradigms and Scientific Revolutions (Kuhn)Normal science, anomaly, crisis and paradigm shift
  10. 10Scientific Research Programmes (Lakatos)Hard core, protective belt, progressive vs degenerating programmes
  11. 11Epistemological Anarchism (Feyerabend)"Anything goes" — there is no single scientific method
  12. 12Scientific Realism and Anti-RealismDo theories describe the world, or merely work?
  13. 13Scientific Explanation (the DN Model)Explanation as deduction of the event from laws
  14. 14Bayesian Philosophy of ScienceDegrees of belief updated by evidence
  15. 15Research ParadigmsPositivism, interpretivism, pragmatism, critical realism
  16. 16Reproducibility and Open ScienceThe replication crisis, preregistration and transparency

Statistical & Methodological Foundations

  1. 01Hypothesis: Null and AlternativeThe logic of statistical hypothesis testing
  2. 02The Central Limit TheoremWhy the sample mean tends toward a normal distribution
  3. 03Sampling MethodsProbability and non-probability sampling
  4. 04Statistical Significance and the p-valueWhat the p-value does and does not tell you
  5. 05Type I and Type II ErrorsFalse positives, false negatives and power
  6. 06Statistical Errors and BiasSampling error, non-sampling error, bias vs variance
  7. 07Confidence IntervalsInterval estimation and its correct interpretation
  8. 08Effect SizeThe magnitude of an effect, independent of sample size
  9. 09Descriptive vs Inferential StatisticsSummarizing data vs generalizing to a population
  10. 10Variable Types and Levels of MeasurementNominal, ordinal, interval, ratio
  11. 11Validity and ReliabilityThe two core dimensions of measurement quality
  12. 12Correlation vs CausationAssociation does not imply causation
  13. 13Measures of Central TendencyMean, median and mode
  14. 14Measures of DispersionHow spread out the data are
  15. 15Skewness and KurtosisThe shape of a distribution
  16. 16Foundations of ProbabilityEvents, conditional probability and Bayes
  17. 17The Normal DistributionThe bell curve and z-scores
  18. 18Common Probability Distributionst, F, chi-square, binomial, Poisson
  19. 19The Sampling DistributionHow a statistic varies across samples
  20. 20Standard Error vs Standard DeviationTwo often-confused quantities
  21. 21Degrees of FreedomThe number of values free to vary
  22. 22Parametric vs Non-parametric TestsWhich family of test, and when
  23. 23Statistical AssumptionsNormality, homogeneity, independence
  24. 24One-tailed vs Two-tailed TestsDirectional vs non-directional hypotheses
  25. 25The Multiple Comparisons ProblemMany tests inflate false positives
  26. 26Bootstrapping and ResamplingInference without distributional assumptions
  27. 27Bayesian vs Frequentist InferenceTwo philosophies of statistics
  28. 28Statistical Power and Sample SizeThe chance of detecting a real effect
  29. 29Choosing the Right Statistical TestA practical test-selection guide
  30. 30Missing Data MechanismsMCAR, MAR, MNAR
  31. 31Outliers and Influential ObservationsDetecting and handling them sensibly
  32. 32Data Transformation and StandardizationLog, square-root, z, min-max

Statistical Literacy

  1. 01Sensitivity and SpecificityHow well a test catches positives and negatives
  2. 02Predictive Values (PPV and NPV)The real probability given a test result
  3. 03ROC Curves and AUCThreshold-independent classifier performance
  4. 04The Confusion MatrixThe basis of all classification metrics
  5. 05Classification Metrics: Accuracy, Precision, Recall, F1Evaluating a classifier correctly
  6. 06Likelihood RatiosHow a test result updates the odds
  7. 07Cohen's d and the Standardized Mean DifferenceThe magnitude of a difference between two means
  8. 08Variance Explained: Eta-squared and R-squaredHow much variability the model explains
  9. 09The Odds RatioA measure of association for categorical outcomes
  10. 10Relative Risk and Risk DifferenceComparing risk as a ratio and as a difference
  11. 11Number Needed to Treat (NNT)How many patients for one extra good outcome
  12. 12The Hazard RatioComparing instantaneous risk over time
  13. 13Measures of Association for Categorical DataEffect size after a chi-square test
  14. 14Overfitting and UnderfittingA model's ability to generalize
  15. 15The Bias–Variance TradeoffThe tension between two sources of error
  16. 16Training, Validation, and Test SetsSplitting data for honest evaluation
  17. 17RegularizationPreventing overfitting with a penalty
  18. 18Model Selection and Information CriteriaChoosing among competing models
  19. 19Goodness-of-Fit and Model ErrorHow well a model fits the data
  20. 20Regression DiagnosticsChecking the assumptions of a regression
  21. 21Interaction EffectsWhen one effect depends on another variable
  22. 22Random VariablesMapping outcomes to numbers
  23. 23Expected Value and VarianceA distribution's centre and spread
  24. 24The Law of Large NumbersWhy the sample mean converges to the truth
  25. 25Bayes' TheoremUpdating beliefs in light of evidence
  26. 26Joint, Marginal, and Conditional DistributionsHow several variables behave together
  27. 27How to Conduct a Hypothesis TestThe step-by-step testing procedure
  28. 28Checking Statistical Assumptions in PracticeWhat to do when assumptions fail
  29. 29Reporting Statistical ResultsReporting findings fully and honestly
  30. 30Interpreting p-values, Confidence Intervals, and Effect Sizes TogetherReading the whole picture, not one number