Question

The data given in tab one can be used to compare education spending by state. We cannot use the data as given, however. For example, to compare Texas and Vermont purely on total amount spent would give a false impression of how much each state spends on education because of varied population size. To make a more valid comparison we can calculate spending per capita by dividing total spent by state population. (although this is total state population and not just students it will give a much better basis for comparison)

Based on the regression line above, what amount would you predict for local spending if a state spends $40 billion? What might be some reasons that using that prediction would give you hesitation?

Answer 1

it is not given which 2 variables to tak e, hence we choose

spending per capita in billions and population as the 2 variables

we enter the data in excel and then goto data > data analysis tab and select regression

Recent Soces AIl Edit Links Advanced Co Tra Regressiorn 3 23 Input OK Input Y Range: Input X Range: Labels SB$1:5B$51 SAS1:SAS51 Constant is Zero Cancel IO Help .9 Confidence Level: | 95 | 96 .9 Output options | 0 output Range: O New Worksheet Ply ONew Workbook Residuals .4 .8 L.3 Residuals Residual Plots Standardized Residuals Lne Fit Plots Normal Probability Normal Probability Plots 1.9 .8

The regression equation is thus

population = -3.24 +1.236*spending

if spending = 40

then population
-3.24 +1.236*spending

-3.24 +1.236*40 = 46.2

as the model has not learnt for 40b and the value falls outside the learnt range of spending , hence the prediction might not be correct