Changing coefficient when an independent variable is added?

In summary, the regression results in parts b and c show that education is an important factor in explaining the income difference between men and women, and that women may have a higher average education than men in this data set.
  • #1
Tim 1234
11
0

Homework Statement


(b) (10 points) Estimate the equation income = β0 + β1female. Interpret theβ1 coefficient. How does β1 compare to your difference in means in part a?
(c) (10 points) Estimate income as a linear function of highest grade and female. Report results. How does the coefficient on female change?
(d) (10 points) How could we use the regression results in (b) and (c) to tell if women have higher average education than men?

2. Homework Equations

b) income = 39331.2-13506.8female
c)income = -14676.4+3896.5highest_grade-15487.9female
The coefficient on female decreased by 1981.10 with the inclusion of highest grade completed.

The Attempt at a Solution



I know from the data set that women do have a higher average education than men.
Obviously adding the new variable increased the amount by which income is decreased if the individual is a female, but I don't really understand why that indicates a higher average education for women than for men.
 
Physics news on Phys.org
  • #2

The coefficient on female in part b represents the difference in average income between men and women, assuming all other variables are held constant. This means that on average, women in this data set earn $13,506.80 less than men. This difference in means is a simple comparison of the average income for men and women, while the coefficient in the regression equation takes into account other factors that may influence income, such as education and experience.

In part c, the coefficient on female changes to -15,487.90, which means that on average, women earn $15,487.90 less than men with the same highest grade completed. This suggests that education is an important factor in explaining the income difference between men and women. By including education in the regression equation, we can better understand the relationship between education and income, and how it may differ for men and women.

To determine if women have a higher average education than men, we can compare the coefficients on female in part b and part c. If the coefficient is larger in part c, it suggests that women have a higher average education than men, as education is a significant factor in determining income. In this case, the coefficient on female did increase, indicating that women may have a higher average education than men in this data set. However, further analysis would be needed to confirm this conclusion.
 

1. How does adding an independent variable affect the coefficient in a regression model?

Adding an independent variable can potentially change the coefficient in a regression model. This is because the new variable may have a correlation with the dependent variable, causing the relationship between the original independent variable and the dependent variable to change.

2. Is it necessary to change the coefficient when adding an independent variable?

No, it is not always necessary to change the coefficient when adding an independent variable. The coefficient will only change if the new variable has a significant impact on the relationship between the original independent variable and the dependent variable.

3. How do I know if the coefficient has changed due to the addition of a new independent variable?

You can determine if the coefficient has changed by comparing the regression model before and after adding the new independent variable. If the coefficient for the original independent variable has changed, it indicates that the new variable has a significant impact on the relationship.

4. Can adding an independent variable improve the accuracy of a regression model?

Yes, adding an independent variable can potentially improve the accuracy of a regression model. This is because the new variable may provide additional information that can better explain the variation in the dependent variable.

5. Are there any potential drawbacks to changing the coefficient when adding an independent variable?

One potential drawback is multicollinearity, which occurs when the new independent variable is highly correlated with one or more existing independent variables. This can lead to unstable coefficients and make it difficult to interpret the impact of each individual variable on the dependent variable.

Similar threads

  • Set Theory, Logic, Probability, Statistics
Replies
5
Views
981
  • General Discussion
Replies
8
Views
1K
  • Set Theory, Logic, Probability, Statistics
Replies
13
Views
2K
Replies
19
Views
1K
Replies
1
Views
823
  • Set Theory, Logic, Probability, Statistics
Replies
7
Views
469
Replies
80
Views
4K
  • Classical Physics
Replies
3
Views
1K
  • Quantum Interpretations and Foundations
2
Replies
45
Views
3K
  • Set Theory, Logic, Probability, Statistics
Replies
7
Views
340
Back
Top