This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Areas of rectangles=relative frequencies, the height=density (% / horiz.unit); Area= base x height (all must add to 100%) To Draw: make frequency/relative frequency table, use appropriate intervals on the x axis, and record the heights at the top of each drawn rectangle68, 95, 99.7 rule; Use RootMeanSquare to find SD ; Effects on Avg and SD of adding a constant to every number, multiplying by a constant Standard Units (z) = (ValueAvg)/SD; Percentiles “to the left”, use symmetry! Plotting Points (x,y); X goes over, Y goes up and down; Slope = rise over run (positive or negative); y = m( x )+b; Find b (y intercept) Correlation Coefficient and SD LineAverages of X and Y make point of averages; The coefficient, r , is between 1 and 1SD line goes through the point of averages, slope is (SD of y )/ (SD of x )To calculate r , convert each variable to standard units, find products, take average.Equation of the line for y on x estimates the average value of y for each xFor each increase of 1 SD on x , there is an incr. of r SDs on y , on averageRegression Method for individuals vs. groups“Regression Effect” as scatter spreads out, an increase in one and decrease in the other are often offsetting and misleading“Regression Fallacy” thinking that there is more to be interpretedEquation for Reg. Line goes thru the Pt of Avg; Slope=rise over run if our run is 1 SD x , our rise is r (SD y ) m= y = m x +b, write out the final equationOpposite rule P(event) = 1 P(opposite event) Key words “at least one”Mult Rule The chance that 2 things will both happen=chance the 1 st will happen x the chance the 2 nd will happen given the first will happen. will happen given the first will happen....
View
Full Document
 Spring '07
 MaryRusso
 Statistics, Normal Distribution, SNC, chance error

Click to edit the document details