Graphical Representation of Data
Reflective Piece (Charts and Histograms Worksheet)

What?

This assignment built directly on the previous summary measures worksheet but shifted the focus from numerical description to visual representation. Using Excel, I constructed percentage frequency bar charts, clustered column charts, and relative frequency histograms to explore patterns in categorical and continuous datasets, including brand preferences, ecological survey data, and weight loss outcomes under two diets. My responses and interpretations were embedded directly into the worksheet alongside the charts.

Rather than relying on automated chart tools, the exercises required manual decisions about chart type, class boundaries, scaling, and labelling. This made the process slower, but also more deliberate, and forced me to think about why a particular visual was appropriate for a given type of data rather than treating charting as a cosmetic final step.

So what?

The key learning for me was how much analytical judgement is baked into visualisation choices. In the brand preference exercises, the same underlying data told a much clearer story once converted into percentage frequency bar charts. Seeing Area 1 and Area 2 side by side made it immediately obvious that the overall preference ordering was the same, but that the strength of those preferences differed. That distinction is far less intuitive when looking at raw tables, yet it is exactly the kind of nuance decision makers care about.

This resonates strongly with my work in the aviation exams department. I frequently produce charts comparing cohorts, attempts, or semesters, and this worksheet reminded me that visual structure can either clarify or distort interpretation. A clustered bar chart makes comparative differences explicit, whereas a single aggregated chart can flatten meaningful variation. The exercises showed that the “right” chart is not the one that looks best, but the one that best matches the analytical question being asked.

The histogram exercises were particularly valuable. Constructing class intervals manually, choosing bin widths, and plotting relative frequencies made the relationship between data distribution and interpretation very tangible. In the diet examples, the histograms revealed differences in shape, spread, and central tendency that were not obvious from summary statistics alone. Diet A appeared more symmetrical and slightly shifted toward higher weight loss, while Diet B was more concentrated at lower values. That visual evidence made the comparison feel more grounded than simply stating that one mean was higher than the other.

This has clear parallels to exam analytics. Two exams or cohorts can share a similar average score while having very different distributions, one tightly clustered around the pass mark, another spread across a wide range. A histogram makes risk visible in a way a single number never can. The worksheet highlighted that graphs are not decorative, they surface structure that directly affects how outcomes should be interpreted.

Now what?

Going forward, I want to be more intentional about when and why I use specific chart types. For categorical comparisons, I will default to percentage based visuals rather than raw counts, especially when cohort sizes differ. For performance data, I want to use distribution plots more often to complement averages, particularly when results feed into academic decisions or quality assurance discussions.

This assignment also reinforced that good visualisation is a form of ethical reporting. Poorly chosen charts can exaggerate differences, hide variability, or imply certainty that is not really there. Being explicit about scales, class boundaries, and comparisons is part of being honest with data. The practical discipline of building these charts step by step has made me more confident that when I present a visual, I understand exactly what assumptions sit behind it.

Overall, this worksheet helped close the gap between statistical concepts and real world communication. It showed that understanding data is not just about calculating the right values, but about choosing visuals that make the underlying patterns visible without oversimplifying them.

References

• University of Essex Online (Unit 8). Charts and Graphs Worksheet (completed assignment). :contentReference[oaicite:1]{index=1}