Purpose
In this homework, you will get practice with questions related to measurement and size. You will first practice with some classroom-type questions where you have a choice of answers, and then we will move to interpreting measurement and size given a table of actual paper. This then requires understanding information from a table, which is a skill I want yâall to practice constantly.
Finally, what you will practice is turning numbers from tables into actionable information that can help people make decisions.
The following questions are derived from real situations or questions about real policy evaluations. Unfortunately, this also means we will end up discussing sensitive and serious topics (for example, drug-related deaths in this homework). Please let the professor know if this is ever an issue for you.
Guidelines
- You can work by yourself or with groups of up to three.
- Submit your answers to Gradescopeâonly one per group.
- You will get points for correct answers. You will get points deducted if the answer contains more information thatâs not necessary or if the answer contains incorrect statements among correct statements. In short, we are trying to incentivize students to use the least amount of characters while maximizing the accuracy of responses.
- If a do-file is required, always work on a do-file that you can save.
- Your responses should be professionally formatted and written.
- The due date is Friday, February 7th, at 11:59 pm EDT.
- When you see things like âwrite it as non-technicalâ or âImagine you are talking to a policy-maker,â please refrain from using
any
technical jargon.
Short questions
- The National Institute of Health formed a review committee that reviewed a flu vaccine's effectiveness after flu season. The following measures were proposed, and the NIH had measurements of this before the vaccine was available, which can help make a before and after comparison; based on the insights from the lecture on measurement, pick the one that seems the most adequate of the four.
- Percentage of people vaccinated with the flu vaccine
- Number of people vaccinated with the flu vaccine
- Share of people with flu vaccine among the people who are hospitalized
- Share of people hospitalized because of the flu
- Now explain your answer. When explaining your answer, mention the issues you found with the ones you did not pick, one by one.
- A well-renowned public university in the South (đ) institutes a program to make its employees âhealthier.â Because of HIPPA (privacy around personal health information), the university doesnât have any measures for employee health. However, it does have health expenditures for each individual, and it can compare the changes in health expenditures before and after the program's introduction.
- If we see a reduction in personal health expenditures after the program, does this imply that their employees are less healthy? Explain. Provide a realistic counter-example. Use non-technical language. (Notice that the question will end in the future before the âexplainâ).
- If we see a reduction in personal health expenditures after the program, does this imply that their employees are healthier? Explain. Provide a realistic counter-example. Use non-technical language.
- After the program was implemented, they sent out a survey to a random set of employees so they could rate their health. They do their research and follow common questions used in many studies to assess health, which is rating health based on 5 âexcellent,â 4 âvery good,â 3 âgood,â 2 âfair,â and 1 âpoor.â They come for you for feedback on whether you think this measurement is âgood or not.â They acknowledge the limitations of the measure of subjectivity and ambiguity (i.e., Individuals may interpret the response categories differently, or one personâs âgoodâ might be anotherâs âfair.â) and lack of specificity (i.e., The measure provides a broad overview but does not identify specific health issues or dimensions). Aside from acknowledging those, What would your feedback be?
Table interpretation
Syringe exchange programs (SEPs), also known as needle exchange programs, are public health initiatives designed to reduce the harms associated with injection drug use. They do so by providing access to sterile syringes and facilitating the safe disposal of used needles for injection drug users (Centers for Disease Control and Prevention, 2016; Markell, 2016; Giroir, 2019). The idea is that by providing clean needles, the program prevents the spread of infectious diseases and that providing safe disposal of used needles reduces the risk of accidental needlestick injuries. Of course, there are other byproducts of SEPs, such as connecting individuals to healthcare and support services.
In light of the recent opioid crisis, many public health entities have called for an expansion of syringe exchange programs (SEPs). A paper by Vanderbilt Economics uses a newly constructed administrative dataset to estimate the effects of recent SEP openings on HIV diagnoses and drug-related deaths. Her main findings can be found in the following table.
The columns represent different models, while the rows represent different outcomes variables. They are using a difference-in-difference design for all of these models using a standard regression analysis, for the purposes of the homework, we can interpret the coefficients as coming from the following regression:
Where represents the different outcomes in county at time . represents the constant, represents a set of county characteristics for county at time , is the error term, and represents the effect of the policy (SEP) on the outcomes. The coefficients on table 1 represent the beta for each regression, with their standard error in parenthesis. Read the table notes for further information.
- How many regressions are embedded in this table? or how many regressions does table 1 have information from? (Just answer with a number).
- Asides from the HIV outcomes variable, there are a number of drug-related outcomes. Why doesnât the author just pick one instead of showing many? Explain what each of the measures captures that the other one doesnât.
- What is the mean number of HIV diagnoses in a county? Express your number with the appropriate units.
- The Albemarle county population is 115,676 in 2023 and in this year there were no SEP programs in the county. Using the information from this table what would be an expected number of drug-related deaths? Round to the first decimal. Show your process
- The point of the paper is to understand if SEP causes an increase in opioid-related deaths. Use the specification results from table 1 to provide a sentence that informs someone who doesnât know anything about statistics (say a legislator) on the effect of SEPs on opioid-related deaths (Notice how the question can and should end here, however weâll provide more guiderails for now). Tell them how many more deaths, and since number of deaths is hard to understand, express the effect as a percentage effect. Since there are many estimates, you can give them a range.
- In the county that this representative was elected from there were 17 opioid-related deaths (per 100,000) this past year. How many more deaths would we expect, if they open a syringe program in this county? Express it as a integer number and as a percentage. Show process.
- In the county that this representative was elected from there were 13.88 new cases of HIV diagnosis, the population is the one stated above. What would be the expected change in HIV Diagnosis if this program is implemented? You can provide a number if with up to two decimals. Show process. You can use results from column 1.
One step further
One more question that we could have added but would require more work and is a bit more sensitive is to then compare the expected âdecreaseâ in death because of less HIV diagnosis, relative to the âincreasedâ deaths due to drug-related issues for a given county. You could technically do this with the information that is online, but overall that would be furthering the exercises to being able to compare similar units and understand if the net-saving of lives is negative or positive. Either way, policies that save some people in some way, but then do the opposite are a bit hard to swallow, and hard to âevaluateâ, yet is important to know. Usually we can try risk-mitigation by asking the question, can we amend the policy such that it provides the benefits without the bad unintended consequences? But in some cases that may not be enough.
References
- Markell, J. (2016): âGovernor and Legislators Propose Expansion of Needle and Syringe Exchange Program,â Available at https://news.delaware.gov/2016/05/12/governor-and-legislators-proposeexpansion-of-needle-and-syringe-exchange-program/.
- Centers for Disease Control and Prevention (2015): âNCHHSTP Atlas,â Available at http://www.cdc.gov/nc (2016): âUse of syringe services programs increases, but access must improve for greater HIV prevention,â Accessed 31-July-2018 at https://www.cdc.gov/media/releases/2016/p1129-HIV-syringeservices.html.
PDF Version of HW
If you have issues downloading the PDF version, head to this site and try again: https://sebastiantello.notion.site/Homework-2-Needles-188a8428b27980a59abee6d07260cf50