Uncertainty, standard errors and confidence intervals
- confidence intervals - help to quantify uncertainty around the estimate (of a population value)
- to construct a confidence interval, use the standard error (estimated by dividing the standard deviation by the square root of the no. of people in the sample
- lower and upper limits of the confidence intervals can be estimated as
- critical t value - the value of t which will give us the most accurate estimate of the confidence interval
- depends on degrees of freedom (N - 1) which in our case are related to the sample size
- How to interpret confidence interval:
- ‘assuming that our sample is one of the 95% producing confidence intervals that contain the population value, then the population value for … falls somewhere between … and …’