First Week of Research Progress

My first week of research started out with some confusion. As I’m pretty sure anyone reading this will know, Monday, the 27th, was Memorial Day. Despite my knowledge of this, I still intended on meeting with my research professor on this date because she specifically stated it while we were emailing over the break. So, after preparing to meet with my instructor at 10 am and going to her office and another place I thought I might find her, I sent her an email only to find out that she decided to take the federal holiday off. In her response email she apologized and recommended that I do the same and, although, I did, I also decided to start on my portion of the research.

At the meeting the next day, we were able to narrow down the search to a compressed data set from the wonder search section on the Centers for Disease Control and Prevention website, also known as CDC Wonder. In this research lab, the information we are looking for is online and public access like that of the CDC and BJS databases. One problem with public access information is that it can be hard to find. So, at the first meeting we decided upon what data set on mortality would be best to use. It was decided that we would use the mortality data set that would let us differentiate between underlying causes of death, which would help us to find the information we wanted on maternal mortality.

I downloaded the data set within the timeline of 1999-2017 and with the parameters of year, race, and state – this included the race and state codes designated by the CDC. The measurements in the data set consisted of total number of deaths, population, and crude rate (also known as the mortality rate). The crude rate calculated by the CDC is the number of deaths divided by the total population of each state. The goal from this research is to find the maternal mortality rate across the country while delving into how race, class, and incarceration affect these numbers. The plan is to combine the CDC U.S. mortality data with information on social factors from the BJS that may influence the rate.

During the first week I downloaded two data sets from the CDC, one with the parameters that I mentioned above and one just by year in each state. I converted both data sets into excel spreadsheets. The first data set I had to reformat in notepad before would transfer into excel because excel truncated the information and put it into only one column the first time around; but for the second one I was able to use Rstudio to do the same thing in a lot less time. The meeting scheduled for Thursday went something like Memorial Day for me and I, once again, emailed my professor along with the work that I had completed.

I continued to work into the weekend, since my professor emailed feedback on my work, once she was feeling back to her normal self. Thus, overall my first week was productive, but things could have definitely gone better had there not been a breakdown in communication. Before the meeting on Monday, I investigated the ‘unreliable’ phrase that was reoccurring in the crude rate column that turned out to be a put in place of the small number that the CDC likely thought insignificant. During the second week, I plan on calculating the crude rate in place of the unreliable fill-in. Additionally, it will be important to download the BJS data and look into economic control variables for the mortality rate that can consider poverty and incarceration.