Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

NEED BEFORE WEDNESDAY 8 pm EST I have no Idea how to do this I need help. This E

ID: 3133579 • Letter: N

Question

NEED BEFORE WEDNESDAY 8 pm EST

I have no Idea how to do this I need help.

This Excel file SAT Math shows the national average SAT Math scores for the years 1967-2005 (The College Board made significant changes to the SAT between the 2005 and 2006 exams; the class of 2005 was the last class to take the former version of the exam when it featured math and verbal sections).

In data sets that have large numbers such as "years", it is usually better to shift the data. Therefore, subtract 1966 from each Year value in the Excel file so that the shifted Year values range from 1 to 39.

Create a scatterplot with the shifted years as the explanatory variable x and the SAT Math scores as the response variable y.

Question 1. The scatterplot indicates that a linear model is not appropriate. The scatterplot also indicates that a ladder of powers transformation will not work for this data. A curvilinear (polynomial) regression model will work best. From the 2 polynomial models shown below, select the polynomial model that works best for this data. Base your decision on the results of the parameter hypothesis tests, the values of the standard errors and coefficients of determination, and analysis of the residuals. (one submission only)

y = 0 + 1x + 2x2 + y = 0 + 1x + 2x2 + 3x3 + (CORRECT)


Question 2. For the model selected in Question 1, consider the coefficient of the term with the highest power of x. What is the value of the test statistic for the hypothesis test that tests whether this coefficient is nonzero?

Question 3. What is the residual for 2004? Use 2 decimal places.

SAT Math Scores

Year

SATMath

1967

516

1968

516

1969

517

1970

512

1971

513

1972

509

1973

506

1974

505

1975

498

1976

497

1977

496

1978

494

1979

493

1980

492

1981

492

1982

493

1983

494

1984

497

1985

500

1986

500

1987

501

1988

501

1989

502

1990

501

1991

500

1992

501

1993

503

1994

504

1995

506

1996

508

1997

511

1998

512

1999

511

2000

514

2001

514

2002

516

2003

519

2004

518

2005

520

SAT Math Scores

Year

SATMath

1967

516

1968

516

1969

517

1970

512

1971

513

1972

509

1973

506

1974

505

1975

498

1976

497

1977

496

1978

494

1979

493

1980

492

1981

492

1982

493

1983

494

1984

497

1985

500

1986

500

1987

501

1988

501

1989

502

1990

501

1991

500

1992

501

1993

503

1994

504

1995

506

1996

508

1997

511

1998

512

1999

511

2000

514

2001

514

2002

516

2003

519

2004

518

2005

520

Explanation / Answer

Question 2. For the model selected in Question 1, consider the coefficient of the term with the highest power of x. What is the value of the test statistic for the hypothesis test that tests whether this coefficient is nonzero?
-5.536        or -5.54 ( two decimals)

Question 3. What is the residual for 2004? Use 2 decimal places.

-0.67

Regression Analysis

0.922

Adjusted R²

0.915

n

39

R

0.960

k

3

Std. Error

2.497

Dep. Var.

SATMath

ANOVA table

Source

SS

df

MS

F

p-value

Regression

2,579.4570

3  

859.8190

137.86

1.91E-19

Residual

218.2866

35  

6.2368

Total

2,797.7436

38  

Regression output

confidence interval

variables

coefficients

std. error

   t (df=35)

p-value

95% lower

95% upper

Intercept

525.4852

1.7666

297.454

3.76E-61

521.8988

529.0716

x

-4.3279

0.3777

-11.460

2.14E-13

-5.0946

-3.5612

xx

0.1846

0.0218

8.469

5.42E-10

0.1403

0.2288

xxx

-0.0020

0.00035837

-5.536

3.16E-06

-0.0027

-0.0013

Regression Analysis

0.922

Adjusted R²

0.915

n

39

R

0.960

k

3

Std. Error

2.497

Dep. Var.

SATMath

ANOVA table

Source

SS

df

MS

F

p-value

Regression

2,579.4570

3  

859.8190

137.86

1.91E-19

Residual

218.2866

35  

6.2368

Total

2,797.7436

38  

Regression output

confidence interval

variables

coefficients

std. error

   t (df=35)

p-value

95% lower

95% upper

Intercept

525.4852

1.7666

297.454

3.76E-61

521.8988

529.0716

x

-4.3279

0.3777

-11.460

2.14E-13

-5.0946

-3.5612

xx

0.1846

0.0218

8.469

5.42E-10

0.1403

0.2288

xxx

-0.0020

0.00035837

-5.536

3.16E-06

-0.0027

-0.0013