SOCIOL 5 Chapter Notes - Chapter 16: Empirical Relationship, Farad, Linear Regression
Document Summary
Stat computations describing either the character of a sample or the relationship among variables in a sample. Descriptive stats merely summarize set of sample observations. Move beyond description of specific observations to make inferences about larger pop from sample pop observ were drawn. Proportionate reduction of error: logical model assessing strength of relationship asking how much knowing values on one variable would reduce our errors in guessing values on the other. For example, we know how much education people have, we can improve ability to estimate how much they can earn, thus indicating relationship between two variables. 60% yes 40% no make fewest errors in this process. Ex. know empirical relationship b/n first variable and some other; gender. If two variables related you should make fewer errors the second time. Possible to therefore, to compute pre by knowing relationship b/n 2 variables: greater relationship, greater reduction of error. Lambda: guess 100 people if male or female half right and wrong.
