Scribd
Upload
4 views10 pages

Sec 4175

The document presents a regression analysis predicting Final_Score based on Attendance_Percent and multiple predictors including Sleep_Hours, Internet_Access, Hours_Studied, and Family_Income. The simple regression shows a very strong correlation with R² = 0.970, while the multiple regression explains 99% of the variance in Final_Score, indicating the effectiveness of the predictors. Key findings suggest that Attendance_Percent and Hours_Studied are the most significant predictors of Final_Score.

Uploaded by

sameed.fict23
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
4 views10 pages

Sec 4175

The document presents a regression analysis predicting Final_Score based on Attendance_Percent and multiple predictors including Sleep_Hours, Internet_Access, Hours_Studied, and Family_Income. The simple regression shows a very strong correlation with R² = 0.970, while the multiple regression explains 99% of the variance in Final_Score, indicating the effectiveness of the predictors. Key findings suggest that Attendance_Percent and Hours_Studied are the most significant predictors of Final_Score.

Uploaded by

sameed.fict23
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 10

Name - Sameed Farooqui

Roll No. - 23HCS4175


Course - Computer Science

Regression: Simple and Multiple

Dataset -
Simple

1. Variables Entered/Removed
2. Model Summary
Metric Value Meaning

R 0.985 Very strong correlation between


Attendance_Percent and Final_Score.

R Square (R²) 0.970 97% of the variation in Final_Score can be


explained by Attendance_Percent alone.

Adjusted R 0.968 Slightly adjusted for sample size; still very high,
Square showing a robust model.

Std. Error of 2.312 Small standard error → predictions are close to


Estimate actual values.

Interpretation:
This model is extremely strong.
An R² = 0.970 means 97% of a student's final score can be predicted just by
knowing their attendance percentage.

3. ANOVA Table

Interpretation:

● The F-value = 573.017 is very large.

● The p-value (Sig.) = < 0.001, meaning the model is highly significant.

● Conclusion: Attendance_Percent significantly predicts Final_Score.


4. Coefficients Table

Regression Equation (using B values):


Final_Score=−45.683+1.478×Attendance_Percent.

Interpretation of Coefficients:

● Constant (-45.683):

If a student’s attendance was 0% (hypothetically), their predicted


Final_Score would be -45.683.
(Of course, practically, no one gets 0% attendance.)

● Attendance_Percent (1.478):
For every 1% increase in attendance, the Final_Score increases by
approximately 1.478 points.

The t-values are very large, and p-values are very small (< 0.001), meaning the
coefficients are statistically significant.
Conclusions
● Attendance_Percent is an extremely strong predictor of Final_Score.

● The regression model is highly significant and very reliable.

● The model can be used for prediction purposes confidently.

● Students who improve their attendance will likely see a direct, positive
impact on their final scores.

Multiple
1. Objective
The goal of this analysis is to predict Final_Score based on multiple
predictors. The independent variables used are:

● Sleep_Hours (hours of sleep per night)

● Internet_Access (1 = Yes, 0 = No)

● Hours_Studied (hours of study per day/week)

● Attendance_Percent (percentage of classes attended)

● Family_Income (family’s monthly income in ₹ ’000)

2. Model Summary
Statistic Value Interpretation
R 0.995 Very high correlation between the observed
and predicted values.
R Square 0.990 99.0% of the variance in Final_Score is
explained by the predictors.
Adjusted R 0.988 Adjusted for degrees of freedom; still
Square extremely high.
Std. Error of 1.356 Low standard error indicating accurate
Estimate prediction of scores.

Interpretation:
The model explains nearly all the variation in students' Final_Score. The
high R Square and Adjusted R Square values suggest that the chosen
predictors are very effective in explaining the outcome.

3. ANOVA Table

Interpretation:

● The F-statistic (1200.00) is extremely high, and the associated p-


value (< 0.001) is far below the typical alpha level (0.05).

● This indicates that the overall regression model is statistically


significant.

● Thus, at least one of the predictors reliably predicts Final_Score.


4. Coefficients Table

Regression Equation

Based on the coefficients, the regression equation is:

Final_Score=−15.000+(1.200×Sleep_Hours) + (3.500×Internet_Access) +
(4.000×Hours_Studied) + (0.900×Attendance_Percent) +
(0.250×Family_Income)

Interpretation of Coefficients:

● Intercept (-15.000):
If all predictors are zero (which may be hypothetical), the predicted
Final_Score would be -15. While not realistic, the intercept serves to
adjust the baseline level.

● Sleep_Hours (1.200):
For each additional hour of sleep, Final_Score increases by about 1.2
points. However, with a p-value of 0.145, this effect is not statistically
significant at the 0.05 level.

● Internet_Access (3.500):
Students with Internet Access score, on average, 3.5 points higher than
those without it; this is statistically significant (p = 0.002).

● Hours_Studied (4.000):
Each extra hour of study contributes about 4 points to the Final_Score,
and this effect is highly significant (p < 0.001).
● Attendance_Percent (0.900):
For every 1% increase in attendance, the Final_Score increases by
about 0.9 points. This predictor shows a very strong effect (p < 0.001).

● Family_Income (0.250):
Each increase of ₹1,000 in family income is associated with an
increase of about 0.25 points in Final_Score (p = 0.047), which is
statistically significant.

5. Final Conclusions & Recommendations


● Overall Model Strength:
The model is extremely strong, explaining 99% of the variation in the
Final_Score. This implies that the combined effects of Sleep_Hours,
Internet_Access, Hours_Studied, Attendance_Percent, and
Family_Income are very powerful predictors.

● Key Predictors:
○ Attendance_Percent and Hours_Studied have the most robust
contributions (high t-values and very low p-values).

○ Internet_Access and Family_Income also significantly enhance


the model’s prediction.
○ Sleep_Hours did not reach statistical significance in this model.

You might also like