SOCI 2P11 Lecture Notes - Lecture 3: Nonprobability Sampling, Statistical Inference, Convenience Sampling
12 Apr 2018
School
Department
Course
Professor
Document Summary
Sampling error: difference between sample statistic and population parameter, because sampling error can be computed, probability sampling enables use of inferential. Inverse relationship between sample size (n) and sampling error statistical tests: we can calculate sampling error with probability, whereas we can"t with non-probability. Incomplete sampling frame: non-probability sampling, non-response (and non-response bias, findings from a biased sample cannot be generalized to the parent population. Factors that affect sampling size: time and cost, non-response, diversity of parent population. Purposive sampling: selecting participants appropriate to the study topic and research objectives. Snowball: expanding the sample through referrals from participants. Involves identifying categories of people in the parent population and deciding on the number of people from each category to be selected for the sample. Convenience sampling: selection of people who happen to be available, produces highly unrepresentative samples; cannot generalize the results of a convenience sample.
