PSY1022 Lecture Notes - Lecture 13: Statistical Inference, Null Hypothesis, Statistical Hypothesis Testing
Document Summary
You will remember from previous weeks that it is not possible to test every person in the population (imagine having to test every man and woman around the world in the above example!). Instead, we typically recruit a sample of individuals and draw conclusions about the population from this smaller group. To do this, we need to use inferential statistics. There are several possible methods, all with the aim of evaluating whether or not the data we"ve collected support a research hypothesis. If we want to know how likely it is that our specific sample has been drawn from a specific population, then we need to use probability. The basis of inferential statistics is to determine the probability (i. e. , the likelihood) of observing the experimental result. Take for example playing heads or tails, and getting heads eight times in a row. Conclusion: this is an improbable (but note: not impossible!) result.