HDFS 1060 Lecture Notes - Lecture 3: Random Assignment, Data Quality, Recall Bias
22 Aug 2016
School
Department
Course
Professor
Document Summary
Because research creates knowledge, and we want our knowledge of the world to be accurate. Because common sense isn"t always (a) common or (b) right, and philosophy or logic is not the same as science. Because its important to be able to critically evaluate facts and opinions (and to know the difference) Generalizable- study sample will generalize all the people like them. Apolitical- cant have a political agenda; cant have a stake in an outcome. Sample- the people who are actually in the study. Samples are often weird cultures- western, educated, industrialized nations, rich, and democratic. However, just because its not perfect sample, doesn"t mean the study doesn"t contain any truth. True experimental- key is random assignment- only kind of research that can definitively tell you real results. Quasi-experimental- control for variables you think might have led people to self-select into groups. Participant interpretation of questions and the situation does not always match your own.
