BIOCHEM 301 Lecture 6: Midterm Review
Document Summary
Describes behavior, health outcomes of a particular group. Does not try to understand causes or relationships between variables. Investigates relationship between 2+ variables, does not provide causality. Experimental: tests whether a variable causes changes in behavior or health outcome; involve manipulation of one independent variable, to see the effects of a dependent variable. Quasi-experimental research: examines effects of naturally occurring events, used when researcher is unable to manipulate the iv or all other factors; ex. Measurement scales: nominal: names or identifies specific characteristic; ie categorical, numbers assigned as labels, ordinal: indicates when there is more or less of a variable; ie rank ordering, as in dislike, c. slightly dislike, neutral, slightly like, like. Reliability: how consistent or dependable a measurement is; inversely related to measurement error, reliability=true score variance/ total variance c. Inter-rater reliability: rating by one or more judges correlative: kappa, range = 0-1, . 8 is excellent e.