Topic 1.6 Notes – Descriptions for One Quantitative Variable Distributions

What a One-Variable Quantitative Distribution Shows

A distribution tells you two things about one quantitative variable: the values it takes and how often those values occur.

What you’re describing is the data, not the graph type. Saying “this is a histogram” does not answer the question. You want to say what the histogram shows about, for example, quiz scores, wait times, or heights.

A complete description includes:

• shape
• center
• spread (variability)
• unusual features such as outliers, gaps, and clusters

Always write in context:

• name the variable
• include units if there are units
• use approximate values or intervals from the graph

Different graphs show different detail:

• Dotplots and stemplots show exact values more clearly
• Histograms group values into intervals, so you usually describe ranges, not exact data values

This comparison is helpful because it shows the same data displayed three ways. Focus on how the dotplot and stemplot make the individual values easy to see, and how the histogram groups those values into bars.

Same data shown as a dotplot, stemplot, and histogram

SOCS is a good memory tool for Shape, Outliers, Center, Spread. It helps you organize, but your answer still needs to sound like an actual description.

Shape and Unusual Features

Symmetry and skewness

Shape has two parts. One is whether the graph is symmetric or skewed.

• Approximately symmetric means the left and right sides are roughly mirror images
• Skewed right means the tail extends farther toward larger values
• Skewed left means the tail extends farther toward smaller values

The AP trap here is huge. Skew is named for the tail, not where most of the data are. If most values are on the left and the tail stretches right, it is skewed right.

Some graphs are messy and irregular. Don’t force a label if none fits well.

Peaks and modality

The other part of shape is peaks.

• Unimodal means one main peak
• Bimodal means two prominent peaks
• Approximately uniform means about the same frequency across values, with no clear peak

You can combine these, like skewed right and unimodal or approximately symmetric and bimodal.

Minor bumps do not create extra modes. Look for the overall pattern. Also, the peak is not always the center.

Outliers, gaps, and clusters

These are the unusual features you’re expected to notice.

• Outlier means an unusually high or low value relative to the rest
• Gap means an interval with no observations
• Cluster means a concentration of values, often separated by a gap

Give locations in context, like “an apparent outlier around 51 minutes,” not just “there is an outlier.” In the example below, most response times form one main cluster, there’s a clear gap before the far right, and a single outlier at about 51 minutes.

Dotplot showing an outlier, gap, and main cluster

Center and Spread

The center is the typical or middle value of the distribution. On a graph, you estimate it.

Center is not automatically:

• the midpoint of the axis
• the midpoint of min and max
• the tallest bar
• the most common exact value

The spread tells how much the values vary. You can describe it with:

• approximate minimum and maximum
• overall span
• width of the main cluster
• whether values are tightly packed or spread out
• long tails or isolated extremes

If there’s an outlier, separate the main spread from the total spread. For example, “most response times are from about 8 to 28 minutes, but overall they range from 8 to 51 minutes.”

Writing a Complete Description

A strong AP-style description usually sounds like this:

“The distribution of response times is skewed right and unimodal, centered at about 15 minutes, with most values from about 8 to 28 minutes and an overall range from about 8 to 51 minutes. There is an apparent high outlier at about 51 minutes.”

That works because it includes, in order:

1. variable and units
2. shape
3. center
4. spread
5. unusual features

If you compare two distributions of the same variable, use comparative language such as “Group A has a higher center and less variability than Group B.”

Using the Graph to Justify Claims and Common Mistakes

When you justify a claim, point to evidence from the graph itself:

• counts or proportions in an interval
• center above or below some value
• tail direction
• peaks, gaps, clusters, or outliers

Example: “Most requests were answered in under 19 minutes because about 23 of the 30 observations are between 8 and 18 minutes.”

That claim is about the observed data shown. It does not automatically describe a whole population.

Common mistakes:

• saying “most data are on the left, so skewed left”
• naming the graph instead of the distribution
• leaving out variable or units
• calling empty axis space a gap
• claiming exact values from a histogram bin
• saying “the graph looks like it” without evidence