GLBH0028 Econometrics for Health 2021-22
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
GLBH0028: Econometrics for Health
Econometrics for Health 2021-22
Final Assessment
Background
The smoking rates in the US are persistently high and the Federal Government has encouraged the different states to impose smoking bans in public places. While most US States have introduced a state-wide smoking ban in public places, 12 states have not adopted the smoking bans because their senior advisors were unsure whether smoking bans were the most effective way of reducing smoking rates and improving citizens’ health. The aim of this assignment is to assess whether the introduction of state-wide smoking bans have improved population health.
Data
To perform this task, you are given a subset of the Behavioral Risk Factor Surveillance System survey data for the year 2020 (the survey runs every year).1 The data set is called brfss2020.dta
and is available on Moodle. The variables included in the data are as follows:
• poorhlth: During the past 30 days, for about how many days did poor physical or mental health keep you from doing your usual activities, such as self-care, work, or recreation?
• smoking: Smoking status of the individual (1 if currently smoking, 0 otherwise).
• edu: Education level (1 none; 2 grades 1-8; 3 grades 9-11; 4 grade 12; 5 college 1-3 years; 6 college 4 or more)
• income: Income level (1 less than $10k; 2 $10k-$15k; 3 $15k-$20k; 4 $20k-$25k; 5 $25k-$35k; 6 $35k-50k; 7 $50k-75k$; 8 $75k or more)
• sex: sex (1 male; 2 female)
• age: age (in years)
• state: US State where the individual lives in.
• ban: Whether the state has introduced a state-wide smoking ban in public places (1 yes, 0 no)
Reporting style
You are asked to present your report by strictly following the question numbering. Present re- gression tables and estimated equations for all analyses. Use academic writing similar to journal articles while writing your response. We strongly encourage students to avoid giving simple ’yes/no’ answers or formulating these as bullet points. Presentation, language, clarity in the answers, and the do-file (to be submitted as an Appendix) are taken into consideration in the final mark. The word count is 2,000 words, and excludes any tables, figures, appendix, and references.
Questions
Using the brfss2020.dta data set, answer the following questions.
Q1 Provide a brief description of the data. Is there a difference in poor health days between
states which have introduced the smoking ban and those which have not? [10%]
Q2 Assess how comparable (according to individual characteristics) states with and without
smoking bans are. Is there a difference in the smoking rate between these two groups? [10%]
Q3 Use a linear regression model to estimate the effect of state-wide smoking ban on poor health
days, adjusting for relevant confounders. Interpret the results [10%]
Q4 Re-estimate the effect of state-wide smoking ban using linear regression, but accounting for
the fact that individual observations might be correlated within states. Discuss the results, contrasting them to those in Q3. [10%]
Q5 Estimate the average treatment effect of state-wide smoking ban on poor health days using
propensity score matching. Interpret the results, discussing the potential advantages of this approach compared to the linear regression approach. [15%]
Q6 Using the linear regression model, assess whether the effect of state-wide smoking ban differs
according to individual’s smoking status. Discuss the results. [10%]
Q7 Discuss the overall findings of your study in light of potential limitations of both the methods
and the data considered in the analyses [20%]
Q8 What are the implications of your findings for policy making? What suggestions would you
make for further research? [10%]
Presentation and do-file
Presentation, language, clarity in the answers, and the do-file [5%]
2021-12-31