Data Visualization Issues
I once had trouble understanding data while working on a survey analysis for employee satisfaction across different departments of our company. The dataset included responses to multiple questions on job satisfaction, work-life balance, and opportunities for growth. With hundreds of responses and a variety of rating scales, the raw data was presented as a long list of responses, which made it challenging to extract meaningful insights. I found it difficult to see how different departments compared, where specific issues were more pronounced, how different factors influenced overall satisfaction, and what relationships could be found.
One way that would have made this data easier to understand would have been to have a pie chart. That would have helped with understanding categorical data, like the proportion of employees satisfied with their jobs in each department. Pie charts are good for getting a quick and clear visual breakdown.
A stacked bar chart also could have helped. It could have been used to display multiple variables in one visual, like comparing overall satisfaction and work-life balance ratings across the different departments. That would have made it easier to see how different factors contributed to the total satisfaction scores.
Another idea would have been a word cloud. Since part of the survey included open-ended responses, using a word cloud to visualize the most frequently mentioned keywords (like "management," or "workload," or "flexibility") could have given us a clearer picture of the main concerns without having to sift through hundreds of text responses.
A correlation matrix might have helped too as this would show relationships between different factors like work-life balance, growth opportunities, and overall satisfaction. This would be a way to identify which factors were most strongly linked to satisfaction, so that we could have seen where to focus on areas needing improvement.
