Hi everyone! My name is Bruna, and I’m part of Cohort 26 from Sydney Data School. Dashboard week day 4 and we had a Power BI day focused on exploring a dataset about dengue cases. Each of us was requested to bring in additional datasets to create a unique analysis, and I was really excited to dive into this topic because dengue is such a significant health issue in Brazil (my home country).
Initial Goals and Challenges
My initial plan was to explore how internet and TV access might correlate with dengue cases, thinking regions with more information access might have better awareness or preventive measures. However, after spending almost an entire day searching, I couldn’t find any comprehensive datasets on internet cables or TV access specifically for Brazilian states. It was a bit frustrating because this was a perspective I was really eager to analyze!
Switching Gears: Education, and Population Data
With my original plan hitting a wall, I decided to pivot to other potentially relevant data sources. I gathered weather data, educational statistics, and population data to normalize the dengue cases and see if any trends emerged. I even found a dataset on student enrollment in Brazil, but I had to do a lot of cleaning and merging to make it work with my dengue data. Despite my best efforts, these variables didn’t reveal any strong relationships with the dengue case numbers.
I did manage to find some basic data on internet usage, though it wasn’t quite what I’d initially envisioned. I was hoping for a dataset that combined internet and TV access as indicators of information availability, but instead, I worked with the internet access data I had. Unfortunately, even with this, I didn’t get the insights I was aiming for.
The Unexpected Outcome
After hours of exploration, creating charts, and trying to connect the dots between dengue cases and factors like internet access, education, population, and states, I was surprised to find no clear patterns. None of the variables seemed to correlate strongly with the number of dengue cases, which made me feel like I might have missed the mark in my approach. It was a tough day, putting in so much effort for results that felt inconclusive, but it was also a learning experience.
Alteryx Exploration for Data Comparison
In addition to Power BI, I explored the dengue dataset in Alteryx since it contained multiple levels of temporal and spatial data. This gave me a chance to compare different types of data structures and see which would work best for my analysis. While this added complexity, it helped me understand the dataset better and choose the data that best suited my approach.
Wrapping Up
In the end, while my analysis didn’t yield the clear insights I had hoped for, I still found value in the process. Working with a real-world dataset related to a topic close to my heart was rewarding, and I enjoyed tackling the challenges, even if the results didn’t align with my expectations.
Thanks for reading, and feel free to check out the screenshot of my final dashboard below. It was a tough day but a worthwhile one in terms of learning and exploring new angles!
