2. Common Terms and Equations
Confidence Intervals
Confidence intervals are frequently reported in scientific literature and indicate the consistency, or variability of the result. The confidence interval uses the sample to estimate the interval of probable values of the population. The confidence interval shows the range of values you expect the true estimate to fall between if you redo the study many times. Confidence intervals can provide important information to statistical significance of studies especially when a p-value is borderline (i.e., it is equal to the critical p-value). If the null value (the value that indicates no difference and is usually zero or one) is included in the confidence interval, then the result is not statistically significant.
If a study has 95% confidence interval calculated, then this means that if the study was repeated multiple times with samples from the whole population and the confidence intervals were calculated for each of those repeated studies, then the true value would lie within the calculated confidence intervals 95% of the time. For example, the true value for a whole population is 50. If a study is repeated multiple times with samples drawn from the whole population, the calculated confidence intervals of each repeated study may be slightly different such as 47-53, 48-54, 49-51, etc. However, the true value (50) would be contained within 95% of the calculated confidence intervals.
Reliability
Similarly, reliability refers to the consistency of the measurement, or the ability to repeat the measurement and obtain the same result. For example, if a scale is used to measure the weight of a cup at multiple times, the scale would be considered reliable if the weight is the same at each measurement, and the scale would be less reliable if it produced different results when measuring the same weight. Poor reliability of an instrument can impact study results.
(Image Source: iStock, Thiago Santos©)