# Alpha level

## Alpha level is the odds a researcher is willing to take to commit a Type I error

The alpha level, also known as the level of significance, is the chance that researchers are willing to take to commit a Type I error. The alpha level is often set at .05. The alpha level defines the boundaries for determining the critical region that leads to "very unlikely" or significant results that differ from the null hypothesis. Researchers that use an alpha value or level of significance of .05 know that there is a 1 in 20 chance of committing a Type I error.

In instances where Type I or false positive findings are deleterious to patients, a small alpha level such as .01 or .001 can be chosen. This makes significant treatment effects much harder to detect. With these alpha levels, researchers take on the risks of 1 in 100 and 1 in 1,000, respectively, of committing a Type I error.

When multiple hypotheses are tested concurrently, such as when multiple statistical tests are run at the same time using the same statistical test, the chances of committing a Type I error increase DRASTICALLY. If you were to test five hypotheses using an alpha value of .05, then the odds of committing a Type I error increase to 5 in 20 or 25%. This phenomenon is often accounted for using a Bonferonni corrected alpha value to be used for significance. Simply divide the alpha level by the number of hypotheses being tested to yield the new alpha level that needs to be achieved to denote statistical significance. For example, .05 / 5 = .01. So, when testing my five hypotheses, unless the test statistic falls in the critical area outside of .01, it is not statistically significant.

In instances where Type I or false positive findings are deleterious to patients, a small alpha level such as .01 or .001 can be chosen. This makes significant treatment effects much harder to detect. With these alpha levels, researchers take on the risks of 1 in 100 and 1 in 1,000, respectively, of committing a Type I error.

When multiple hypotheses are tested concurrently, such as when multiple statistical tests are run at the same time using the same statistical test, the chances of committing a Type I error increase DRASTICALLY. If you were to test five hypotheses using an alpha value of .05, then the odds of committing a Type I error increase to 5 in 20 or 25%. This phenomenon is often accounted for using a Bonferonni corrected alpha value to be used for significance. Simply divide the alpha level by the number of hypotheses being tested to yield the new alpha level that needs to be achieved to denote statistical significance. For example, .05 / 5 = .01. So, when testing my five hypotheses, unless the test statistic falls in the critical area outside of .01, it is not statistically significant.

Click on the

**Beta Level**button to continue.## Hire A Statistician

**DO YOU NEED TO HIRE A STATISTICIAN?**

Eric Heidel, Ph.D., PStat** **will provide the following statistical consulting services for undergraduate and graduate students at $100/hour. Secure checkout is available with Stripe, Venmo, Zelle, or PayPal.

- Statistical Analysis
- Research Design
- Sample Size Calculations
- Diagnostic Testing and Epidemiological Calculations
- Survey Design and Psychometrics