Skip to main content
Back

Type I & Type II Errors definitions

Control buttons has been changed to "navigation" mode.
1/14
  • Null Hypothesis
    Initial assumption in a test, often representing no effect or status quo, which is evaluated against sample data.
  • Alternative Hypothesis
    Statement opposing the initial assumption, suggesting a different effect or outcome in the population.
  • Type I Error
    Mistake made when evidence leads to rejecting a correct initial assumption, often linked to the significance threshold.
  • Type II Error
    Mistake made when evidence fails to challenge an incorrect initial assumption, potentially missing a real effect.
  • Alpha
    Threshold probability set before testing, representing the maximum risk accepted for wrongly rejecting the initial assumption.
  • Beta
    Probability of not detecting a real effect, representing the risk of failing to challenge an incorrect initial assumption.
  • P Value
    Calculated probability indicating how likely observed data would occur if the initial assumption were true.
  • Significance Level
    Pre-determined cutoff used to decide whether evidence is strong enough to challenge the initial assumption.
  • Hypothesis Test
    Structured procedure for using sample data to evaluate competing claims about a population.
  • Sample Data
    Observed values collected from a subset of the population, used to draw conclusions about broader trends.
  • Experimental Design
    Planned approach for collecting and analyzing data to answer specific research questions or test claims.
  • Ethical Considerations
    Moral factors influencing decisions about which risks or mistakes are more acceptable in research conclusions.
  • Probability
    Numerical measure of how likely an event or outcome is to occur, central to evaluating risks in testing.
  • Treatment Efficacy
    Effectiveness of an intervention, often the focus of claims tested in medical or scientific studies.