Type I and Type II errors and Level of Significance

• Type I errors are false positives
• Type II errors are false negatives
• Type I errors occur when the null hypothesis is rejected in error

  • the result is reported as significant when there is a possibility in occurred by chance
• Type II errors occur when the null hypothesis has been incorrectly retained - you conclude there is no effect when actually there is

  • the result is reported as not significant when there's a possibility it did show an effect
• Type II errors are more likely when using a smaller level of significance (1%) because there is a much smaller chance that the results occurred by chance
• Why do we use the p≤0.05 level of significance in psychology?

  It creates a balance between committing type I and II errors.

  If the result is significant at the p≤0.05 level, it means there is a 5% chance that it occurred by chance
• p = probability

  p ≤0.05 = 5% probability that the results occurred by chance