Why the middle of a scale matters
A rating-scale question often gives people a middle option, such as “Neither agree nor disagree.” It can be tempting to summarize the results as a contest between positive and negative answers. That shortcut may be useful for a specific decision, but it leaves out people who chose neither side. If neutral responses make up a substantial share, a chart of positive answers alone can make an opinion look more settled than it is.
Neutral does not always mean indifference. A respondent might have mixed views, lack enough experience to judge, or find that the statement does not quite fit. The chart cannot tell those explanations apart. It can, however, preserve the response they gave and prompt a useful follow-up question. When presenting a five-point scale, start by showing all five options rather than treating the middle as space to be discarded.
Choose a chart that preserves the distribution
For one question, a simple bar chart with a bar for each response option is often the clearest starting point. Keep the options in scale order, from the most negative to the most positive, and label each bar with a percentage or count. Readers can then see whether answers cluster at the extremes, near the middle, or across the scale.
To compare several questions or groups, a stacked bar can save space. A diverging stacked bar places negative responses on one side and positive responses on the other, with neutral responses visibly occupying the center. This makes the balance of opinions easier to scan while keeping the middle category in view. If you split the neutral segment around the center line, explain that choice in the chart note; otherwise, readers may mistake the two halves for different responses.
Use color consistently across questions and groups. A restrained sequence of shades can indicate the progression from negative through neutral to positive. Do not rely on color alone: include a legend or direct labels, and check that adjacent segments remain distinguishable. For a small number of questions, a chart with more room for labels is usually more useful than a dense display that requires readers to decode tiny segments.
Keep neutral separate from unanswered
A neutral choice is an answer. A skipped question is not. “Don’t know” or “Not applicable,” when offered as choices, may also mean something different from the midpoint. Combining any of these into one bucket can obscure both the distribution of opinions and how many people could answer the question.
State the base used for each percentage. For example, a chart might show percentages among people who selected one of the five rating options, while a note reports how many skipped the question or selected “Don’t know.” If those responses are important to the interpretation, display them alongside the scale rather than burying them in a note. Counts can be especially helpful when comparing groups of different sizes: the same percentage may represent very different numbers of respondents.
Use summaries without letting them replace the chart
Sometimes a team needs a compact measure, such as the share who selected either of the two positive options. That measure is commonly called top-two-box. It can help readers compare many items quickly, provided the label explains exactly which answers are included and the response base stays consistent. Pairing that figure with a full-distribution chart prevents a high positive share from concealing a similarly large neutral group.
An average rating can be convenient, but different patterns can produce similar averages. One question may divide respondents between opposite ends of the scale, while another draws mostly middle responses. Show the distribution when those patterns could lead to different decisions. Likewise, avoid calling a neutral response “negative” merely because it is not positive. If you calculate a net positive measure, define the calculation and show the neutral share nearby.
Make the takeaway match the evidence
Write captions that describe what the chart shows before suggesting what it means. “Positive responses exceed negative responses, while many respondents chose neutral” is more informative than “Customers approve” when the middle segment is prominent. If one group has a larger neutral share than another, describe that difference without assuming why it exists.
Before publishing, check the wording and direction of every scale, especially if some questions are phrased negatively. Confirm that percentages add up after rounding, labels match the options respondents saw, and each comparison uses an appropriate base. A reader should be able to tell what was asked, who is represented, and where every response went. When neutral answers remain visible, the chart supports a more careful decision: whether to act on a clear preference, investigate uncertainty, or ask a better question next time.
