Type III error
Type III errors occur as a result of miscommunication and lack of hypotheses
Type III errors were first postulated by a statistician at Oak Ridge National Laboratory (ORNL) back in 1957.*^ A Type III error is when a statistical test is used to answer the wrong question, or the error committed by giving the right answer to the wrong problem.* Miscommunication between a statistical consultant and a research often occurs due to a lack of competencies and understanding of each other's science and empirical frame of reference. The lexicon or language of statistics is very nebulous and induces cognitive dissonance in human beings. Promoting a clear understanding of the research question, research design, and effect size in statistical consultation is important to detract from committing these types of errors.
Type III errors also occur as a result of applying statistical tests without hypothesis-driven reasoning underlying the analysis. Another way of saying this is running statistics just to find a significant p-value! A Type III error may also occur when multiple hypotheses are tested concurrently and the only significant finding is focused on. Essentially, a Type III error can occur when researchers are only looking for significant results and not answering real research questions in a hypothesis-driven fashion...or POOR SCIENCE.
Type III errors also occur as a result of applying statistical tests without hypothesis-driven reasoning underlying the analysis. Another way of saying this is running statistics just to find a significant p-value! A Type III error may also occur when multiple hypotheses are tested concurrently and the only significant finding is focused on. Essentially, a Type III error can occur when researchers are only looking for significant results and not answering real research questions in a hypothesis-driven fashion...or POOR SCIENCE.
Click on the Alpha Level button below to continue.
Statistician For Hire
DO YOU NEED TO HIRE A STATISTICIAN?
Eric Heidel, Ph.D. will provide statistical consulting for your research study at $100/hour. Secure checkout is available with PayPal, Stripe, Venmo, and Zelle.
- Statistical Analysis
- Sample Size Calculations
- Diagnostic Testing and Epidemiological Calculations
- Psychometrics
*Kimball, A. W. (1957). Errors of the third kind in statistical consulting. Journal of the American Statistical Association, 52, 133-142.
^Oak Ridge National Laboratory is in my hometown of Oak Ridge, TN!
^Oak Ridge National Laboratory is in my hometown of Oak Ridge, TN!