A normal curve, a dot plot, a box-and-whisker, a multi-rater chart — each is built for a different job. Picking the one that fits the question makes a profile clearer; picking the wrong one just adds clutter. Here's when to reach for each.
The starting point isn't "which graph looks best" — it's "what is the reader supposed to take away?" A single score explained to a parent, a whole profile at a glance, the spread hidden inside a composite, agreement across raters: these are different questions, and each has a chart that answers it well. Before the visual, name the takeaway. The graph follows from that.
The bell curve places a score in the context of the whole population — where it sits relative to average, and how common or rare it is. It's the most intuitive way to show standing, because it turns a percentile into a position the eye reads instantly.
Explaining one score, or a few, to an audience that isn't steeped in statistics. A single dot on the curve, with the average band shaded, makes "your child is right around the middle" or "this falls well below the typical range" immediately clear. It's the natural choice for a parent conference or the summary section of a report.
Showing many scores at once. Pile a dozen subtests onto one curve and the markers collide, the labels overlap, and the clarity that made the curve useful is gone. Past a handful of points, a dot plot usually reads better.
A dot plot lines scores up along a shared scale, one marker each, so an entire profile becomes visible in a single view. Where the curve shows one score's standing, the dot plot shows a whole set's pattern.
Seeing the shape of a profile at a glance — which scores cluster, which one separates, whether the profile is flat or jagged. It's the workhorse for comparing many scores across domains, and it's especially good at making a single outlier obvious, since the eye is drawn straight to the marker that sits apart from the rest.
Conveying precision or spread. A dot is a single point; it says nothing about the confidence interval around a score or the scatter beneath a composite. When how certain or how variable a score is matters, a bare dot can imply more exactness than the data supports — add interval bars, or use a box plot.
A box-and-whisker plot shows a distribution rather than a point: the median, the middle range, and the reach out to the extremes. In assessment, its most valuable use is making spread visible — the variability that a single summary number conceals.
Showing that a composite hides real scatter. An index score of 100 can sit on top of subtests ranging from 70 to 130. Reported as a single number, that 100 looks like a tidy summary of "average" ability. Drawn as a box, the enormous spread underneath is impossible to miss — and that spread is often the clinically important part, because it can mean the composite isn't a meaningful summary of anything.
Explaining a single score to a lay audience. Box plots carry more statistical convention than a parent should have to decode — quartiles and whiskers invite questions that pull focus off the point. For a general audience, a curve or dot plot communicates faster; save the box plot for showing variability to a reader who needs to see it.
Behavior and adaptive rating scales are built to be answered by several people — a parent, a teacher, the student themselves. A multi-rater chart plots every reporter on one graph, so their scores can be read together instead of flipping between separate pages.
Comparing raters side by side on the same scales. Putting parent, teacher, and self scores on one chart — shape-coded per rater — turns three separate score sets into a single readable picture of where reporters agree and where they diverge.
Disagreement between raters is information, not noise. A behavior that shows up sharply at school but not at home, or a concern a student reports about themselves that adults haven't noticed, is a genuine finding — and it's exactly what a single-rater view erases. The multi-rater chart is valuable precisely because it preserves those differences instead of averaging them away. It doesn't tell you what the disagreement means; it makes the disagreement visible so you can consider it. (More on reading rater agreement →)
Beyond the type of graph is a second choice that matters just as much: how you organize what goes on it. The same scores can be arranged to tell different parts of the story, and moving between those arrangements is itself part of how a profile gets understood.
Index and composite scores are usually the headline — they're what summaries and eligibility discussions hang on. Subtests are the supporting detail. Keeping them on separate graphs lets the headline stand clear: an index-level chart shows the broad profile without the visual noise of twenty subtests, and a reader isn't asked to sort the important scores from the granular ones. When the goal is "what's the overall picture," a clean index chart delivers it.
Sometimes the relationship between the parts and the whole is the story. Showing subtests grouped under the index they feed makes it visible when a composite is built from wildly uneven pieces — the moment where "this index score isn't really interpretable" stops being a caveat in the text and becomes something a reader can see. When scatter within a domain is the finding, combining is what reveals it.
A middle path is often the clearest: keep the main graph as the headline, and add a small, focused plot for the one area that needs a closer look. A Working Memory index that's a relative weakness, shown beside a mini dot plot of just its subtests, lets the reader see both the summary and the evidence behind it — without burying the whole report in detail. The main chart carries the story; the mini plot carries the proof for the part that matters.
| If you want to show… | Reach for… |
|---|---|
| Where one score falls in the population | Normal curve |
| The pattern across many scores | Dot plot |
| The spread hidden inside a composite | Box & whisker |
| Agreement and disagreement across raters | Multi-rater chart |
| The broad picture, uncluttered | Index-only chart |
| How the parts build (or undercut) the whole | Index + subtest combined |
| A closer look at one key area | Main chart + mini section plot |
None of these choices is an interpretation — they're ways of presenting evidence so the interpretation is easy to follow. Which one fits depends entirely on the question you're answering, and that question is yours to define.
If you haven't yet, the companion guide makes the broader case for why a visual beats a table in the first place — and how to keep a visual honest.
Switch between a normal curve, dot plot, and multi-rater chart with a click — and see which one tells the clearest story. Free, nothing stored.
Open the Normal Curve Plotter Multidisciplinary Plotter