Statistics 401, Week 4
Statistics 401, Week 4 01:960:401
Popular in Basic Statistics for Research
verified elite notetaker
Popular in Statistics
This 3 page Class Notes was uploaded by Wendy Liu on Thursday September 29, 2016. The Class Notes belongs to 01:960:401 at Rutgers University taught by Hei-ki Dong in Fall 2016. Since its upload, it has received 57 views. For similar materials see Basic Statistics for Research in Statistics at Rutgers University.
Reviews for Statistics 401, Week 4
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/29/16
Week 4: Bivariate Data 27 September 2016 Basic Statistics for Research Professor HK Dong Wendy Liu Bivariate/multivariate data – observations on two or more variables Marginal totals – total frequency of any row or column of a data table, given in the right-hand margin or bottom margin Simpson’s paradox – reversal of conclusions from a data table after combining several data tables together due to appearance of unreported variables Experimentation Random assignment – subjects placed randomly into control/experimental groups Placebo effect – subject’s expectations of a treatment to work cause positive results, even though the treatment itself has no therapeutic value o Placebo – treatment that has no physiological effect; usually a sugar pill, for drug testing Double-blind procedure – experimenters don’t know which subjects are in which group, and subjects themselves don’t know what group they are in o eliminates placebo effect and experimenter bias Scatter diagram/plot – pairs of observations plotted as dots on a graph, with one observation as one variable (x,y) Positive correlation – x and y increase/decrease together Negative correlation – x and y increase/decrease in opposite directions Correlation coefficient r – measures strength and direction of linear relationship between x and y Ranges from -1 ≤ r ≤ 1 o Magnitude of r indicates strength |r| = 1, perfect linear relationship o Sign of r indicates direction r > 0, positive correlation r < 0, negative correlation o r = 1, perfect positive correlation o r = -1, perfect negative correlation o r = 0, no correlation r: sample correlation coefficient ρ: population correlation coefficient Calculating r: Definitional formulas: sum of squared deviations of x: sum of squared deviations of y: sum of cross products of x & y deviations: Alternative formulas: sum of squared deviations of x: sum of squared deviations of y: sum of cross products of x & y deviations: Example calculation of r w/ definitional formulas: n = 4 for (2,5) (1,3) (5,6) (0,2) 2 5 0 1 0 1 0 1 3 -1 -1 1 1 1 5 6 3 2 9 4 6 0 2 -2 -2 4 4 4 Example calculation of r w/ alternative formulas: n = 4 for (2,5) (1,3) (5,6) (0,2) 2 5 4 25 10 1 3 1 9 3 5 6 25 36 30 0 2 0 4 0 Spurious correlation – observed correlation btwn two variables that is false, due to influence by a third variable (the lurking variable) Predictor/input/independent variable – denoted by x Response/output/dependent variable – denoted by y Method of Least Squares – for the line of best fit – minimizes the average amount of residual residual (SSE) – vertical error btwn data point and line of best fit o sum of squared error: regression equation for line of best fit: o slope: o intercept: 2 Coefficient of determination r – amount (%) of variation in Y due to variation in X
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'