This is about multiple regresion

What determines the price for a used car? A random sample of 805 cars from one manufacturer that were advertised in a large metropolitan area was selected and the mileage, number of cylinders, number of doors, the size of the engine (liters), and if the car has a sound system and leather seats were measured. A multiple regression model was fit to predict the price. The value ofR²when all six variables are included in the analysis is 0.385. The value ofR²when only the variable mileage is included in the analysis is 0.019. The value ofF-statistic to determine if the other five variables are useful in explaining variation in selling prices is:

a) 0.366

b) 95.08

c) 0.59

d) 159.97.

Two different models for predicting prices of cars were fit. In the first model, the coefficient associated with the number of doors was positive. In the second model, the coefficient associated with the number of doors was negative. In both cases, thep-values for the associated hypotheses were small. Even though the same data were used, the results seem contradictory. Which is the best explanation for this finding?

a) An error must have been made by one of the researchers.

b) Both researchers could have had nonstatistically significant results.

c) A different set of explanatory variables could have been used in the two models.

d) There could have been an influential observation in the data making the analysis inappropriate.