Within our group we divided the work to among the 4 group members. All of the members worked on the slides that we created, while we all focused on certain pages within the slides. The manager was able to oversee everyone and ensure that everyone in the strip was on pace. Our analyst was in charge of leading the research methods that we used and how we went about finding our conclusion. All group members assisted in the research, but our analyst ensured that everyone was on the same page. Our presentation specialist was in charge of making sure our slides and our NetPhi were all on theme, and ensured everything is cohesive with each other. Our problem solver was essential in ensuring that everything flowed together.
Methods
The methods that we used is the connotative and deconstructive methods of reading datasets from Poirier and the “How to Look at it Method” from Koopman. The connotative method allowed our group to understand the reasoning behind the dataset and the historical context that would lead to reasoning why certain variables are included or not. We used the deconstructive method to find the missing variables in which we think would be important to the dataset, with weight being missing for the majority of the data entries, and a lot of the variables only providing binary data options, with everything else being unknown/othered. Our group used the “How to Look at it Method” by asking questions about the missing variable of weight. We asked if a patients weight affected the readmittance of patients due to the other health problems that arise with the heavier a person is.
Findings
This dataset is not representative of the patient or provider experience with diabetes and patients readmission into the hospital or what causes it, but instead the data set is a limited insurance/administrative dataset from the time during a push for more profitable medicine in America. Diabetes is limited to medicated, and shorter inpatients stays, which reflects an institutional press for efficiency in hospitals. By not including weight, and making other variable limited in their categories or nearly, researchers generalize and clean the data and ignore the complex reality of the disease.
Our group analyzed this dataset based on the patients admittance into the hospital, but lacked sufficient variables to determine whether the patient was readmitted due to diabetes. This dataset did not include the weight variable which is essential in determining whether the medicine that was used on the patients worked effectively.