Chapter 12 analyzed strength data for 57 female high school athletes. Upper body strength was summarized by the maximum number of pounds the athlete could bench press (denoted BP below, 1RM Bench in file). This was predicted well by the number of times she could do a 60-pound bench press (denoted BP_60 in output, BRTF (60) in file). Can we predict BP even better if we also know how many times an athlete can perform a 200-pound leg press? The table shows results after adding this second predictor (denoted LP_200 in output, LP RTF(200) in file) to the model. Multiple regression analysis of strength data

a. Does LP_200 have a significant effect on BP, if BP_60 is also in the model? Show all steps of a significance test to answer this.
b. Show that the 95% confidence interval for the slope for LP_200 equals 0.21 { 0.30, roughly (-0.1, 0.5). Based on this interval, does LP_200 seem to have a strong impact, or a weak impact, on predicting BP, if BP_60 is also in the model?
c. Given that LP_200 is in the model, provide evidence from a significance test that shows why it does help to add BP_60 to the model.

Predictor Coef SE Coef T Constant 60.596 2.872 21.10 0.000 BP 60 LP 200 1.3318 0.1876 7.10 0.000 0.2110 0.1519 1.39 0.171 Analysis of Variance Source Regression 6473.3 3236.6 51.39 0.000 Res. Error 54 3401.3 63.0 Total DF MS 56 9874.6 7.93640 R-Sq :-65.6%