Homework Help is Here – Start Your Trial Now!
Math
Statistics
Suppose you are asked to evaluate a new staff training program to reduce wait times at the Department of Motor Vehicles (DMV) locations. The program is instituted at a randomly chosen 250 of the 500 DMV offices in the state at the beginning of the previous year. You have data for all visits to all the offices for the whole year. This data includes these variables: y: Appointment duration in minutes (including wait time) T: 1 if the office was assigned to the new program (treatment), 0 otherwise urban: 1 if the office is in an urban area, 0 otherwise a) You start by estimating this equation, using data on the entire sample of DMV offices assigned to Treatment and Control: Yi = 30+ 3₁T; +i Assuming the program had no effect on the duration of the non-waiting portions of the appointment, is ₁ an unbiased estimate of the effect of the program on wait times? Why or why not? b) Which parameter does ₁ identify if DMV offices in the Treatment group can choose whether to implement the new program, and only 70% of them decided to implement it (but you still use data on the entire sample of offices assigned to Treatment and Control to estimate this equation)? What about in the case in which there is full compliance with the Treatment, so that all DMV offices assigned to treatment actually implement it? c) Suppose you believe the treatment effect is different in urban and non-urban locations. You estimate the following model: Yi = ẞo + B₁T; + 3₂urban; + ßzurban; × T; +€į. What is the difference in expected wait time between a treated and un-treated location? 1 d) Suppose you estimate the above model and get the following estimates: What is the effect of the program in (i) an urban location? (ii) a non-urban location? Bo=15, 3₁ = -2, 3₂ = 10, 33 = -2, e) Suppose that instead of rolling out this program randomly, the central DMV management had asked local offices to come forward and apply for this program, and the program had been rolled out on a first come, first serve basis. Would B₁ in the above regression still identify the causal effect of the program on wait times? Why or why not?

Suppose you are asked to evaluate a new staff training program to reduce wait times at the Department of Motor Vehicles (DMV) locations. The program is instituted at a randomly chosen 250 of the 500 DMV offices in the state at the beginning of the previous year. You have data for all visits to all the offices for the whole year. This data includes these variables: y: Appointment duration in minutes (including wait time) T: 1 if the office was assigned to the new program (treatment), 0 otherwise urban: 1 if the office is in an urban area, 0 otherwise a) You start by estimating this equation, using data on the entire sample of DMV offices assigned to Treatment and Control: Yi = 30+ 3₁T; +i Assuming the program had no effect on the duration of the non-waiting portions of the appointment, is ₁ an unbiased estimate of the effect of the program on wait times? Why or why not? b) Which parameter does ₁ identify if DMV offices in the Treatment group can choose whether to implement the new program, and only 70% of them decided to implement it (but you still use data on the entire sample of offices assigned to Treatment and Control to estimate this equation)? What about in the case in which there is full compliance with the Treatment, so that all DMV offices assigned to treatment actually implement it? c) Suppose you believe the treatment effect is different in urban and non-urban locations. You estimate the following model: Yi = ẞo + B₁T; + 3₂urban; + ßzurban; × T; +€į. What is the difference in expected wait time between a treated and un-treated location? 1 d) Suppose you estimate the above model and get the following estimates: What is the effect of the program in (i) an urban location? (ii) a non-urban location? Bo=15, 3₁ = -2, 3₂ = 10, 33 = -2, e) Suppose that instead of rolling out this program randomly, the central DMV management had asked local offices to come forward and apply for this program, and the program had been rolled out on a first come, first serve basis. Would B₁ in the above regression still identify the causal effect of the program on wait times? Why or why not?

Functions and Change: A Modeling Approach to College Algebra (MindTap Course List)
Functions and Change: A Modeling Approach to College Algebra (MindTap Course List)
6th Edition
ISBN:9781337111348
Author:Bruce Crauder, Benny Evans, Alan Noell
Publisher:Bruce Crauder, Benny Evans, Alan Noell
Chapter5: A Survey Of Other Common Functions
Section5.3: Modeling Data With Power Functions
Problem 6E: Urban Travel Times Population of cities and driving times are related, as shown in the accompanying...
See similar textbooks
icon
Related questions
Question
Suppose you are asked to evaluate a new staff training program to reduce wait times at the Department of Motor Vehicles (DMV) locations. The program is instituted at a randomly chosen 250 of the 500 DMV offices in the state at the beginning of the previous year. You have data for all visits to all the offices for the whole year. This data includes these variables: y: Appointment duration in minutes (including wait time) T: 1 if the office was assigned to the new program (treatment), 0 otherwise urban: 1 if the office is in an urban area, 0 otherwise a) You start by estimating this equation, using data on the entire sample of DMV offices assigned to Treatment and Control: Yi = 30+ 3₁T; +i Assuming the program had no effect on the duration of the non-waiting portions of the appointment, is ₁ an unbiased estimate of the effect of the program on wait times? Why or why not? b) Which parameter does ₁ identify if DMV offices in the Treatment group can choose whether to implement the new program, and only 70% of them decided to implement it (but you still use data on the entire sample of offices assigned to Treatment and Control to estimate this equation)? What about in the case in which there is full compliance with the Treatment, so that all DMV offices assigned to treatment actually implement it? c) Suppose you believe the treatment effect is different in urban and non-urban locations. You estimate the following model: Yi = ẞo + B₁T; + 3₂urban; + ßzurban; × T; +€į. What is the difference in expected wait time between a treated and un-treated location? 1 d) Suppose you estimate the above model and get the following estimates: What is the effect of the program in (i) an urban location? (ii) a non-urban location? Bo=15, 3₁ = -2, 3₂ = 10, 33 = -2, e) Suppose that instead of rolling out this program randomly, the central DMV management had asked local offices to come forward and apply for this program, and the program had been rolled out on a first come, first serve basis. Would B₁ in the above regression still identify the causal effect of the program on wait times? Why or why not?
Transcribed Image Text:Suppose you are asked to evaluate a new staff training program to reduce wait times at the Department of Motor Vehicles (DMV) locations. The program is instituted at a randomly chosen 250 of the 500 DMV offices in the state at the beginning of the previous year. You have data for all visits to all the offices for the whole year. This data includes these variables: y: Appointment duration in minutes (including wait time) T: 1 if the office was assigned to the new program (treatment), 0 otherwise urban: 1 if the office is in an urban area, 0 otherwise a) You start by estimating this equation, using data on the entire sample of DMV offices assigned to Treatment and Control: Yi = 30+ 3₁T; +i Assuming the program had no effect on the duration of the non-waiting portions of the appointment, is ₁ an unbiased estimate of the effect of the program on wait times? Why or why not? b) Which parameter does ₁ identify if DMV offices in the Treatment group can choose whether to implement the new program, and only 70% of them decided to implement it (but you still use data on the entire sample of offices assigned to Treatment and Control to estimate this equation)? What about in the case in which there is full compliance with the Treatment, so that all DMV offices assigned to treatment actually implement it? c) Suppose you believe the treatment effect is different in urban and non-urban locations. You estimate the following model: Yi = ẞo + B₁T; + 3₂urban; + ßzurban; × T; +€į. What is the difference in expected wait time between a treated and un-treated location? 1 d) Suppose you estimate the above model and get the following estimates: What is the effect of the program in (i) an urban location? (ii) a non-urban location? Bo=15, 3₁ = -2, 3₂ = 10, 33 = -2, e) Suppose that instead of rolling out this program randomly, the central DMV management had asked local offices to come forward and apply for this program, and the program had been rolled out on a first come, first serve basis. Would B₁ in the above regression still identify the causal effect of the program on wait times? Why or why not?
Expert Solution
trending now

Trending now

This is a popular solution!

steps

Step by step

Solved in 1 steps

SEE SOLUTIONCheck out a sample Q&A here
Blurred answer
Recommended textbooks for you
Functions and Change: A Modeling Approach to Coll…
Functions and Change: A Modeling Approach to Coll…
Algebra
ISBN:
9781337111348
Author:
Bruce Crauder, Benny Evans, Alan Noell
Publisher:
Cengage Learning
Linear Algebra: A Modern Introduction
Linear Algebra: A Modern Introduction
Algebra
ISBN:
9781285463247
Author:
David Poole
Publisher:
Cengage Learning
Glencoe Algebra 1, Student Edition, 9780079039897…
Glencoe Algebra 1, Student Edition, 9780079039897…
Algebra
ISBN:
9780079039897
Author:
Carter
Publisher:
McGraw Hill
SEE MORE TEXTBOOKS