KIN330 Lecture Notes - Lecture 3: Royal Institute Of Technology, Sample Size Determination, Statistical Power
26 Dec 2017
School
Department
Course
Professor
Document Summary
Purpose of rq: to learn something new, to resolve debate (i. e. , conflicting info, to learn more about something incompletely understood. Identify the problem: clinical experience, clinical/research theory, professional literature, gaps/conflicts, replication (in different population/lab, answerable question has to be defined in terms of population, intervention/exposure, control/comparator, and outcome pico(t) In defining research problem, developing rq, and clarifying objectives, literature must be reviewed constantly and iteratively. Mediterranean diet, are older adults: manipulated intervention experiment, self-selected group looks at association. Intervention (i. e. , independent variable) may have multiple levels: presence vs. absence (control); different doses/intensities/frequencies. Defining the population: inclusion criteria, traits necessary to be considered a potential participant, exclusion criteria, traits that preclude potential participation. If source list is random, this approximates random selection: cluster sampling, sampling clusters of occurring groups of elements, often based on geographical location, non-probability sampling (sampling bias; only used when probability sampling is. Effect size: statistical term for quantifying strength of causal/correlational effect of iv on.
