1 An example
2 AirlinePercentage on time Complaints Southwest Continental Northwest US Airways United American Delta America West TWA Here is some data on airlines. We have the percentage of flights on time and the number of complaints per 100,000 passengers. What do you think happens to the number of complaints the greater the percentage of flights on time? I would think the complaints would fall.
3 The scatterplot suggests higher the % on time the lower the number of complaints per 100,000 passengers.
4 SUMMARY OUTPUT Regression Statistics Multiple R R Square Adjusted R Square Standard Error Observations9 ANOVA dfSSMSFSignificance F Regression Residual Total Coefficients Standard Errort StatP-valueLower 95%Upper 95%Lower 95.0%Upper 95.0% Intercept % on time
5 ON the previous screen I put a circle around the coefficients on the bottom left of the screen. We use them to write the equation as y hat = 6.02 – 0.07x, where y = complaints per 100,000 passengers and x is % of flights on time. Just to the right of the circle I have a rectangle around the t stat and p-value for the slope. Since the p-value is less than.05 we can reject the null hypothesis of a zero slope and we conclude that the percentage of flights on time has an influence on the number of complaints per 100,000 passengers. Since the slope is negative we would say the more flights on time means less complaints. IN the rounded rectangle on the previous page toward the top left you see the R-square. You really only what to look here after you see the slope is not zero. When the slope is not zero the R-square tells us the percent of the variation in y explained by the x variable. Here this means that 77.9% of the variation in complaints is explained by the % of flights on time. This is a good R-square. Can you think of what else would lead to complaints? Maybe bad snacks on the flight, or rude flight attendants.