Inferential Statistics
Get introduced to sampling methods and terms.
Inferential statistics is a branch of statistics that makes inferences or predictions about a population based on a sample. In inferential statistics, we use models, formulas, and other varied methods to make estimates, predictions, and generalizations about a population. Using these methods, we can estimate population parameters, find confidence intervals, conduct hypothesis tests, and compare samples.
Population and sample
Population refers to all the individuals or elements we are interested in for a particular statistical study. All people living in a country are an example of a population.
Sample, on the other hand, is a part of the population that was selected by the researchers and fits particular criteria. For example, a company makes election predictions by surveying 2,000 people around the country. This group of 2,000 people is called the sample of the population.
Sampling error means that there is a divergence between the parameters of the population and the sample. For example, we take a group of 100 people who live in a neighborhood and find that the average age of the sample group is 38. However, the average age of all the residents (the population) is 33. This discrepancy is called sampling error. Sampling errors are unavoidable. They occur no matter what. The task of the researchers is to approximate the error to zero as much as possible.
Bootstrap sampling means that when samples are chosen from a population, they are not removed from the pool so they can be selected again. This method generates replicate values in sample groups. We can create larger samples than the original populations with this method.