Type I and type II errors

• Type I error


  • Researcher concludes that the results are significant, when the results are really due to chance factors
  • Rejecting the null hypothesis when it should have been accepted
  • Known as a false positive
  • The chances of making a Type 1 error are always the same as your level of significance
  • As significance level increases the chance of Type 1 error decreases
• Type II error


  • When the researcher concludes the results are not significant, when the results are significant
  • A real difference/correlation/ association has been overlooked
  • Accepting the null hypothesis when it should have been rejected
  • Known as a false negative
  • It is extremely difficult to determine the likelihood of making a Type 2 error
• Reducing Type I errors

  1. Replication: To get one significant result by chance is a misfortune but to get more than one is highly unlikely

  2. Increase the significance level i.e. from p≤0.05 to p≤0.01
• Reducing Type II errors

  Increase Sample Size