Math Assignment
a. Regression output screenshot
Model summary - days
Model R R
2
Adjusted RMSE
2
R
M0 0.000 0.000 0.000 16.253
M1 0.299 0.089 0.070 15.673
Note: M 1 include eth, sex, lrn
ANOVA
Model Sum of Squares df Mean F p
Squares
Regression 3421.776 3 1140.592 4.643 0.00
4
Residual 34882.477 142 245.651
M1
Total 38304.253 145
Note: M 1include eth, sex, lrn
Note: The intercept model is omitted, as no meaningful information can be shown
Coefficients
Model Unstandardi Standardized Standardize t p
zed Error d
M 0 (Intercept) 16.459 1.345 12.23 <0.00
6 1
M 1 (Intercept) 18.932 2.570 7.365 <0.00
1
eth -9.112 2.599 -0.281 - <0.00
3.506 1
sex 3.104 2.637 0.095 1.177 0.241
lrn 2.154 2.651 0.066 0.813 0.418
Model summary
Adjusted R² = 0.070, meaning the model explains 7% of the variance in absenteeism.
F(3, 142) = 4.643, p = 0.004, indicating that the overall model is statistically significant.
ANOVA Results
The F-test result (F = 4.643, p = 0.004) confirms that at least one independent variable
significantly affects absenteeism at the 5% significance level (α=0.05).
The estimated regression coefficients are as follows:
Predictor Coefficient (B) Standard t- p-value Statistical
Error value Significance
(Intercept) 18.932 2.570 7.365 <0.001 Significant
Ethnicity (eth) -9.112 2.599 -3.506 <0.001 Significant
Sex 3.104 2.637 1.177 0.241 Not significant
Learning Ability (lrn) 2.154 2.651 0.813 0.418 Not significant
b. Regression Model Equation
Based on the JASP output, the estimated regression equation is:
Y^ =18.932−9.112 ( eth ) +3.104 ( sex ) +2.154 (lrn)
Where:
Y^ = Predicted number of days absent
eth = 0 (Aboriginal), 1 (Not Aboriginal)
sex = 0 (Female), 1 (Male)
lrn = 0 (Average Learner), 1 (Slow Learner)
This equation allows us to estimate absenteeism based on individual characteristics.
c. Interpretation of Slopes and Statistical Significance
Each coefficient represents the expected change in absenteeism for a one-unit increase in the
respective predictor while holding others constant.
1. Ethnicity ( β 1=−9.112 , p< 0.001)
The regression analysis reveals that ethnicity is a significant predictor of absenteeism, with
non-Aboriginal students missing, on average, 9.112 fewer days compared to Aboriginal
students. This difference is statistically significant (p < 0.001), meaning it is unlikely to have
occurred by chance.
2. Sex ( β 2=3.104 , p=0.241)
The regression model indicates that, on average, males are absent 3.104 more days than
females. However, the effect is not statistically significant, as the p-value of 0.241 exceeds
the 0.05 threshold. This suggests that sex does not significantly predict absenteeism. While
the coefficient shows a difference in absenteeism between males and females, the lack of
statistical significance implies that this difference is not meaningful.
3. Learning ability ( β 3=2.154 , p=0.418)
Interpretation: The model suggests that, on average, slow learners are absent 2.154 more
days than average learners. However, the statistical test shows that this difference is not
statistically significant, with a p-value of 0.418, well above the 0.05 threshold.
Statistical significance: With a p-value of 0.418, which is significantly higher than the 0.05
significance level, the effect of learning ability on absenteeism is not statistically significant.
Despite the regression coefficient suggesting that slow learners have, on average, 2.154 more
days of absence compared to average learners, this difference does not reach statistical
significance.
Conclusion:
Ethnicity is the only factor that significantly influences absenteeism.
Sex and learning ability do not significantly predict absenteeism in this dataset.
These results suggest that other, unexamined factors may contribute to absenteeism.
d. Calculating the residual for the second observation
Residual Formula
Residual=Actual Value−Predicted Value
For the second observation in the dataset:
eth = 0 (Aboriginal)
sex = 1 (Male)
lrn = 1 (Slow Learner)
Actual days absent = 11
Using the regression equation:
Y^ =18.932−9.112 ( eth ) +3.104 ( sex ) +2.154 ( lrn )
¿ 18.932−9.112 ( 0 ) +3.104 ( 1 ) +2.154 (1 )
¿ 18.932+3.104+ 2.154
¿ 24.19
Residual Calculation:
Residual=11−24.19
¿−13.19
e. Model Adjusted R-Squared Interpretation
Adjusted R² Value
Adjusted R² = 0.070, meaning only 7% of the variance in absenteeism is explained by
ethnicity, sex, and learning ability.
The adjusted R² value of 0.070 indicates that only 7% of the variance in absenteeism is
explained by the model, which includes ethnicity, sex, and learning ability. This low adjusted
R² suggests that absenteeism is influenced by other factors not included in the model.
According to Field (2018), a low R² may indicate that important variables, such as family
background, health issues, or school policies, are missing from the analysis. Despite the
model’s limited explanatory power, the overall model is statistically significant (p = 0.004),
suggesting that at least one predictor, such as ethnicity, is meaningfully associated with
absenteeism.
Reference:
Field, A. (2018). Discovering statistics using IBM SPSS Statistics (5th ed.). SAGE
Publications Ltd. http://repo.darmajaya.ac.id/5678/1/Discovering%20Statistics%20Using
%20IBM%20SPSS%20Statistics%20%28%20PDFDrive%20%29.pdf
