Using test data on 20 types of laundry detergent, an analyst fitted a regression to predict CostPerLoad(average cost per load in cents per load) using binary predictors TopLoad(1 if washer is a toploading model, 0 otherwise) and Powder(if detergent was in powder form, 0 otherwise). Interpret the results. (Data are from Consumer Reports 68, no. 8 [November 2003], p. 42) R2 0.117
Adjusted R2 0.006 n19
R 0.341 k2
Std. Error 5.915 Dep. Var. Cost Per Load
ANOVA table
Source SS df MS F p-value
Regression 73.8699 2 36.9350 1.06 .3710
Residual 559.8143 16 34.9884
Total 633.6842 18
Regression output confidence interval
variables coefficients std. error t(df= 16) p-value 95% lower 95% upper
Intercept 26.0000 4.1826 6.216 1.23E-05 17.1333 34.8667
Top-Load −6.3000 4.5818 −1.375 .1881 −16.0130 3.4130
Powder −0.2714 2.9150 −0.093 .9270 −6.4509 5.9081