HSS 2381 Lecture Notes - Lecture 12: Confidence Interval, Statistical Power, Binomial Distribution
17 Dec 2017
School
Department
Course
Professor
Document Summary
Uncovers factors underlying relationships (are x and y correlated, or just linked by z) Examines relationship b/w x and y, while holding z constant. Pearson product moment correlation coefficient only works with ratio and interval data. Spearman, biserial, etc. (don"t have to know) Quantifies relationship of independent and dependent variables. Used to calculate/predict values of dependent variable. So for any value of x, you can predict the associated value of y (interpolation) Regression: statistical technique for finding best fitting line for data. Straight line that most closely fits data = regression line / line of best fit. The distance between the regression line, and a data point. Total squared error = sum of squared residuals. Line of best fit minimizes the sum of squared deviations around the line. Measures standard distance between line and data points. Considers spread of scores around regression line. Is the std deviation of distribution of data points around predicted values shown on regression line.
