Scenario:

Several information are provided including those of their grades,

about their parents, and results of some psychological exams. This

exercise will cover many other topics later. For now, just (1) find the

estimate of the regression line (or estimated regression) where the

student’s score on math achievement test is dependent on father’s

education, mother’s education, grades in high school, competence

score, motivation score, and pleasure score; (2) discuss the results as

we had in the examples above; and, (3) which independent variables

seem to be significant?

Problem 1: Run a regression using gender, father’s education,

mother’s education, grades in high school, competence, motivation,

and pleasure as explanatory variables for student’s scores in math

achievement.

Computation 1:

Model Summary

Adjusted R Std. Error of

1 .653a .426 .360 5.33111

a. Predictors: (Constant), Pleasure, Faed, Gender, Motivation,

Sum of
Model Squares df Mean Square F Sig.

1 Regression 1285.412 7 183.630 6.461 .000a

Residual 1733.665 61 28.421

Total 3019.077 68

a. Predictors: (Constant), Pleasure, Faed, Gender, Motivation, GradeHS, Competence,

b. Dependent Variable: Mathac

Coefficientsa

Standardize
Unstandardized d 95% Confidence
Coefficients Coefficients t Interval for B

Std. Lower Upper

1 (Constant) -6.859 4.849 -1.414 .162 -16.556 2.838

Gender -3.546 1.441 -.267 -2.461 .017 -6.428 -.665

Faed .301 .332 .125 .906 .369 -.363 .965

MoEd .327 .407 .107 .803 .425 -.487 1.141

GradeHS 1.901 .478 .449 3.974 .000 .945 2.858

Competence .125 1.388 .012 .090 .928 -2.650 2.901

Motivation 1.608 1.238 .151 1.299 .199 -.868 4.083

Pleasure .910 1.215 .083 .749 .457 -1.521 3.340

a. Dependent Variable:
Mathac

Discussion

Multiple linear regression analysis was conducted was

conducted to determine the influence of gender, father’s education,

mother’s education, grades in high school, competence, motivation

and pleasure, as determinants of student’s scores in Math

significant as indicated on the acquired f-value of 6.461

With p<0.05. It is therefore stated that the independent variables,

gender, father’s education, mother’s education, grades in high school,

competence, motivation and pleasure predict the student’s scores in

Math Achievement. Meanwhile, the computed adjusted R2 value of

0.426 indicates that gender, father’s education, mother’s education,

grades in high school, competence, motivation and pleasure has

contributed significantly in the variability of student’s scores in Math

Achievement by 42.60% from the total variability. Therefore, the

difference of 57.40% was credited to other factors not covered in the

topic.

In addition, table shows that gender and grades in high school

significantly influence the student’s scores in Math Achievement. It

could be seen in the coefficient models that at 0.05 significant level,

regression coefficients of (-3.546) for gender, and (1.901) for grade in

high school predicts the student’s scores in Math Achievement. This

means that a unit of change on gender, and grades in high school

correspond to (-3.546) and (1.901) units change on the student’s

scores in Math Achievement, respectively. Thus, this leads to the

rejection of null hypothesis that none of the independent variables

that significantly influence the student’s scores in Math Achievement.

motivation, and parent’s education as explanatory variables for

student’s scores in math achievement.

Computation 2:

Model Summary

Adjusted R Std. Error of

1 .647a .419 .383 5.23530

a. Predictors: (Constant), ParEduc, Motivation, GradeHS,

ANOVAb

Sum of
Model Squares df Mean Square F Sig.

1 Regression 1264.942 4 316.236 11.538 .000a

Residual 1754.135 64 27.408

Total 3019.077 68

a. Predictors: (Constant), ParEduc, Motivation, GradeHS, Gender

b. Dependent Variable: Mathac

Coefficientsa

Standardize
Unstandardized d 95% Confidence
Coefficients Coefficients Interval for B

Lower Upper
Model B Std. Error Beta t Sig. Bound Bound

1 (Constant) -4.552 3.890 -1.170 .246 -12.323 3.220

Gender -3.310 1.383 -.250 -2.393 .020 -6.073 -.547

GradeHS 1.850 .442 .437 4.188 .000 .968 2.733

Motivation 1.946 1.031 .183 1.887 .064 -.114 4.007

ParEduc .662 .316 .224 2.093 .040 .030 1.294

Standardize
Unstandardized d 95% Confidence
Coefficients Coefficients Interval for B

Lower Upper
Model B Std. Error Beta t Sig. Bound Bound

1 (Constant) -4.552 3.890 -1.170 .246 -12.323 3.220

Gender -3.310 1.383 -.250 -2.393 .020 -6.073 -.547

GradeHS 1.850 .442 .437 4.188 .000 .968 2.733

Motivation 1.946 1.031 .183 1.887 .064 -.114 4.007

ParEduc .662 .316 .224 2.093 .040 .030 1.294

a. Dependent Variable: Mathac

Discussion

Multiple linear regression analysis was conducted was

conducted to determine the influence of gender, grades in high school,

motivation and parent’s education, as determinants of student’s

scores in Math Achievement. As shown on the table, the regression

equation is significant as indicated on the acquired f-value of 11.538

with p<0.05. It is therefore stated that the independent variables,

gender, grades in high school, motivation and parent’s education,

predict the student’s scores in Math Achievement. Meanwhile, the

computed adjusted R2 value of 0.419 indicates that gender, grades in

high school, motivation and parent’s education, has contributed

significantly in the variability of student’s scores in Math Achievement

by 41.90% from the total variability. Therefore, the difference of 58.10%

was credited to other factors not covered in the topic.

high school significantly influence the student’s scores in Math

Achievement. It could be seen in the coefficient models that at 0.05

significant level, regression coefficients of (0.020) for gender, (0.000)

for grades in high school, and (0.040) for parent’s education predicts

the student’s scores in Math Achievement. This means that a unit of

change on gender, grades in high school, and parent’s education

correspond to (0.020), (0.000) and (0.040) units change on the

student’s scores in Math Achievement, respectively. Thus, this leads

to the rejection of null hypothesis that none of the independent

variables that significantly influence the student’s scores in Math

Achievement.

QUEENIE MARIE O. ALAS