POSC300 Lecture Notes - Lecture 11: Sampling Frame, Random Assignment, Sample Size Determination
24 Oct 2018
School
Department
Course
Professor
Document Summary
Making judgements about the characteristics of the population from a sample of cases. Assumption that the investigation of a sample of cases allows us to make claims about the total population. Non-experimental methods rely on a random sample to make inferences the way experimental methods rely on random assignment. Universe of cases you want to describe and understand. Need to clearly define unit of analysis. Defines what you will study and what generalizations you can make. Costs a lot of time and money. Samples are a subset of the population. Each element in total population has a known or equal probability of being included in the sample. Each element in total population has an unknown probability of being included in the sample. The larger the random sampling error - the less our confidence in our interpretations and results. Random sampling error is a function of:
