PSC 41 Lecture Notes - Lecture 7: Quota Sampling, Sampling Error, External Validity

• School: UC-Davis
• Department: Psychology
• Course: PSC 41
• Professor: Vazire

Document Summary

Populaion: usually quite large, you can"t measure everyone so you take a sample: representaiveness is needed for external validity. Example: literary digest (1936: ran polls before presidenial elecions, 2. 4 million surveys, had a biased sample; did not have external validity. Error in sampling: the extent to which the sample difers from the populaion: sampling error: no two samples will be the same, there will be variaion. The larger your sample size helps: sample bias: if your sample is selected systemaically. Simple random sampling: sample will be representaive of the populaion and will be able to draw conclusions about the enire populaion. Straiied sampling: gather informaion about all members of the populaion and idenify meaningful subgroup; divide populaion into subgroups and randomly select proporional samples from each subgroup. Select a non-representaive sample that will exaggerate the efects such as. Snowball sample: if you are interested in a hard-to-access populaion comparing characterisics of the top 10% and botom 10%