The page below is a sample from the LabCE course Linear Regression Analysis. Access the complete course and earn ASCLS P.A.C.E.-approved continuing education credits by subscribing online.

Learn more about Linear Regression Analysis (online CE course) »
How to Subscribe
 MLS & MLT Comprehensive CE PackageIncludes 94 CE courses, most popular \$95 Add to cart Pick Your CoursesUp to 8 CE hours \$50 Add to cart Individual course \$20 Add to cart

Calculating Confidence Intervals

We cannot say for certain what a and β are without taking an infinite number of measurements. However, we can construct a range of values, called a confidence interval, and say that the parameter is in that range with a certain probability. The formulas on the next page show how to construct such confidence intervals.

Before constructing a confidence interval, you must choose the confidence level. The confidence level is the probability that that population parameter will be in the interval calculated. If you choose a high confidence level, you have a greater chance that the true parameter is in the interval; however, the confidence interval will be larger.

The most common choice for confidence level is the 95% level. That means, there is a 95% chance that the calculated interval contains the true population parameter, and a 5% chance the parameter is outside the interval (2.5% chance too high and 2.5% chance too low). t.05 refers to the t-value for the 95% confidence interval. These t-values must be looked up on a t-test table.