Although more and more women are becoming physicians each year, it is well known
ID: 3217869 • Letter: A
Question
Although more and more women are becoming physicians each year, it is well known that men outnumber women in many specialties. Randomly selected specialties are listed below with the numbers of female and male physicians in each.
Specialty
Female
Male
Dermatology
Emergency medicine
Neurology
Pediatric Cardiology
Radiology
Forensic Pathology
Radiation Oncology
3482
5098
2895
459
1218
181
968
7500
20440
10088
1241
7674
399
3215
Suppose we would like to predict the number of male specialists from number of female specialists with the data provided in the table. Answer each of the following questions:
4)
a) Based on the type/form of the relationship between the two variables written above, you have made an assumption for your choice of the form of the predictive model in (1c). Write the assumption for the predictive model in a complete sentence.
b) In (2), it shows there are limitations of applying/using a predictive/regression model for prediction. Write complete sentences to explain the limitations of applying a predictive model.
Specialty
Female
Male
Dermatology
Emergency medicine
Neurology
Pediatric Cardiology
Radiology
Forensic Pathology
Radiation Oncology
3482
5098
2895
459
1218
181
968
7500
20440
10088
1241
7674
399
3215
Answer 1 Answer a The explanatory variable is number of female specialists and the response variable is number of male specialists. Answer b The scatterplot is given below. Scatter Plot y 3.4769x 119.18 25000 R2 0.8561 20000 15000 10000 5000 2000 5000 6000 3000 4000 Female Answer c The estimated regression coefficients are given below Regression Statistics Multiple R 0.9253 R Square 0.8561 Adjusted R square 0.8273 Standard Error 2846.7844 Observations Page I 1Explanation / Answer
4) a) The assumption is that the relationship between the response variable and indipendent variable is linear and they follow normal distribution. The other assumptions are
b) Regression has limitations with large samples; all p-values are statistically significant with an effect size of virtually zero.
Sometime if the model is not build properly, it might over estimate or less estimate the response variable.
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.