Before You Watch

This video describes the concept of Inferential Statistics - in particular hypothesis testing for a single population mean. This is the area of Statistics where we use a sample (some subset of the whole group we wish to describe) in order to infer (deduce) something about the whole group (known as the population).

For example, we may want to know the population mean response time to a medication, i.e., how long on average before a painkiller takes effect. It’s not possible to test the medication on an entire population of individuals requiring such assistance (time, costs, logistics prevent this). Instead we must assess a sample of individuals from that population. We use the results from the sample to make an inference about the entire population.

By taking a sample we introduce a source of uncertainty, as there are many possible samples that could be obtained. Hence we need to account for this uncertainty when making decisions about the population. It is recommended that you are familiar with the content of the Introduction to Statistics and Random Variables videos before viewing this video.


Now What?

This video introduced how we can make conclusions (infer) about a population by looking at a sample from that population.

The videos Inferential Statistics: Two Groups and Chi-squared Test extend hypothesis testing to situations where we want to compare two groups or look for associations between two categorical variables.

But When Am I Going to Use This?

Market research is one example where information obtained from a sample are used to test hypotheses about the population. The process of sampling and gathering data if done well will provide useful information to assist marketing strategies. Further, medications have been safely introduced into common use following clinical trials (tests on samples) which test whether significant improvements are achievable.

Chosen samples need to be representative of their population; having an understanding of the sampling and analysis process enables you to drive or contribute to such tests (or studies) as there is always a demand for quality research and reporting in medicine, business, government and all fields of study, e.g., experimentation in science (CSIRO). When viewing or hearing of statements made in the media, remember to ask on what sample these comments are based!

Other Links

You may like to consider the Statistics Learning videos Understanding Statistical Inference and Understanding the p-value.

Please rate this page