Posts

  1. One simple step to improve statistical inferences
  2. Priors for Bayesian estimation of visual object processing speed in humans
  3. Simple steps for more informative ERP figures
  4. Robust effect sizes for 2 independent groups
  5. the percentile bootstrap
  6. how to fix erroneous error bars for percent correct data
  7. How to chase ERP monsters hiding behind bars
  8. The Harrell-Davis quantile estimator
  9. the shift function: a powerful tool to compare two entire distributions
  10. How to quantify typical differences between distributions
  11. A few simple steps to improve the description of neuroscience group results
  12. Matlab code for the shift function: a powerful tool to compare two entire marginal distributions
  13. Problems with small sample sizes
  14. A clearer explanation of the shift function
  15. How to illustrate a 2×2 mixed ERP design
  16. How to compare dependent correlations
  17. Trimmed means
  18. Your alpha is probably not 0.05
  19. Your power is lower than you think
  20. Bias & bootstrap bias correction
  21. What can we learn from 10,000 experiments?
  22. Can someone tell if a person is alive or dead based on a photograph?
  23. Reaction times and other skewed distributions: problems with the mean and the median [part 1]
  24. Reaction times and other skewed distributions: problems with the mean and the median [part 2]
  25. Reaction times and other skewed distributions: problems with the mean and the median [part 3]
  26. Reaction times and other skewed distributions: problems with the mean and the median [part 4]
  27. A quick review of Reader et al. EJN 2018
  28. Cohen’s d is biased
  29. Bayesian shift function
  30. A new shift function for dependent groups?
  31. Test-retest reliability assessment using graphical methods
  32. Small n correlations cannot be trusted
  33. Correlations in neuroscience: are small n, interaction fallacies, lack of illustrations and confidence intervals the norm?
  34. Small n correlations + p values = disaster
  35. Power estimation for correlation analyses
  36. Planning for measurement precision, an alternative to power analyses
  37. Illustration of continuous distributions using quantiles
  38. Cluster correction for multiple dependent comparisons
  39. Hierarchical shift function: a powerful alternative to the t-test
  40. R functions for the hierarchical shift function
  41. Comparing two independent Pearson’s correlations
  42. When is a 95% confidence interval not a 95% confidence interval?
  43. Comparing two independent Pearson’s correlations: confidence interval coverage
  44. The bootstrap-t technique
  45. Analog teaching activities about sampling and resampling
  46. Mean or median reaction time? An open review of Miller (2020)
  47. Why normality tests are great…
  48. How to test for variance homogeneity
  49. Can we use cluster-based permutation tests to estimate MEG/EEG onsets?
  50. You get the symptoms of a replication crisis even when there isn’t one: considering power
  51. Simulation-Based Power Analyses for the Smallest Effect Size of Interest: A Confidence-Interval Approach for Minimum-Effect and Equivalence Testing
  52. Using cluster-based permutation tests to estimate MEG/EEG onsets: How bad is it?
  53. The Declaration of Helsinki mandates pre-registration
  54. A warning about data-driven simulations
  55. Confidence interval definition: wrong again!
  56. Improving statistical reporting in psychology
  57. Tukey mean-difference plot
  58. Pre/Post design: the fallacy of comparing difference scores
  59. All your p-values are wrong

Bonus:

Zombie scientist

Forget about getting definitive results from a single experiment

Normality is a myth