HLTH201 Lecture Notes - Lecture 5: Main Source, Stratified Sampling, Cohort Study
Document Summary
27th of august 2018: research planning and design part one. Reliability: reproducibility of results of a procedure or tool. Internal consistency (how different items being examined correlate) Validity: accuracy of test/measure amount of measurement error. Confounding (similar to bias, but error in interpretation of a measurement, not measurement: research design and validity: Ability to generate findings to other sample (population) and settings (ecological) Any change/effect in dependent variable can be directly attributed to manipulation of independent variable: accuracy of sample estimate: Misclassification or information bias (measuring device or observer) Each member of the population have an equal chance of being selected. Computers are often used to generate randomization. A sample of (n) subjects is selected in such a way that every possible sample of the same size (n) has the same chance of being chosen. For example select every 50th person in the population. Use results that are easy to get.