Q2. In their paper titled “Superstar effect in sport: Evidence from Italian socc
ID: 3367216 • Letter: Q
Question
Q2. In their paper titled “Superstar effect in sport: Evidence from Italian soccer”, Lucifora and Simmons look at whether there exists a superstar effect in wage determination among professional soccer players appearing in the Italian league. A regression table (presenting OLS regressions with team fixed effects) from this paper is presented below:
SUPERSTAR 1 denotes a dummy variable equal to one for serie A goal-scoring rates of greater than 0.25 and less than or equal to 0.4, whereas SUPERSTAR 2 denotes the exceptional achievement of more than 0.4 goals per serie A game (comprising the top decile of serie A goal-scoring forwards). Consider the third set of coefficients (i.e. the last two columns-coefficients and associated p values-that are under the heading “3”)
1.What is the dependent variable?
2.What does it mean for this regression to have “team fixed effects”?
3.Interpret the coefficient on the variable SUPERSTAR 1.
4.Is this coefficient statistically significant at a 5% level? Why or why not?
5.Based on the variables SUPERSTAR 1 and SUPERSTAR 2, do you see evidence of existence of a superstar effect?
Explanation / Answer
Sol:
What is the dependent variable?
The dependent variable is natural log of player's salary (ln(SALARY)).
What does it mean for this regression to have “team fixed effects”?
This means that the data has been gathered from all the levels of the team that are of interest.
Interpret the coefficient on the variable SUPERSTAR 1.
ln(SALARY) are higher for players with goal-scoring rates of greater than 0.25 and less than or equal to 0.4 than players with goal-scoring rates of less than 0.25 by 0.337.
Is this coefficient statistically significant at a 5% level? Why or why not?
The p-value for the coefficient of SUPERSTAR1 is 0.017. As, the p-value is less than the 5% significance level, we conclude that there is significant evidence that the coefficient of SUPERSTAR1 is statistically significant at a 5% level.
Based on the variables SUPERSTAR 1 and SUPERSTAR 2, do you see evidence of existence of a superstar effect?
Similarly, p-value for the coefficient of SUPERSTAR2 is 0.024. As, the p-value is less than the 5% significance level, we conclude that there is significant evidence that the coefficient of SUPERSTAR2 is statistically significant at a 5% level. As, both the variables SUPERSTAR 1 and SUPERSTAR 2 are statistically significant, there is significant evidence of existence of a superstar effect.
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.