BIOL 001 Lecture Notes - Lecture 8: Framingham Heart Study, Selection Bias, Temporality
30 Jul 2020
School
Department
Course
Professor
Document Summary
Disease status is identified at the start of the study, then we look back to record their exposure status. Exposure and disease are determined at the same point in time (survey) Cannot determine temporal relationships or disease incidence (rate) Uses a representative sample from specified population. Random variation in results due to elements in sample being selected at random. Loaded, ambiguous, inaccurate or poorly posed questions. Processing error mistakes in data coding. Non-response failure to obtain data from sample. One variable shows patterns related to another variable (the two variables vary together somehow) Causation possibilities: a causes b, b causes a, third factor c (confounding factor) causes both a and b, a and b reinforce each other, a and b associated by chance. If you have a, you are more likely to have b, if you have b, you are more likely to have more a.
