📊 Statistics
SOCS: Shape · Outliers · Center · Spread
Describing Distributions (SOCS) — Never miss a component when describing a distribution
S
Shape
Describe the overall shape of the distribution: symmetric, skewed left, skewed right, uniform, or bimodal (having two peaks).
O
Outliers
Note whether any unusually high or low values stand apart from the rest of the data.
C
Center
Report a measure of center — typically the mean or median, depending on the shape (covered in a later lesson on choosing between them).
S
Spread
Report a measure of spread — such as the range, interquartile range (IQR), or standard deviation — describing how variable the data is.
1
A dataset of exam scores has a long tail toward lower scores, one score far below the rest, a median of 85, and an IQR of 10.
2
Shape: skewed left (the tail points toward lower values). Outliers: yes, one unusually low score exists.
3
Center: since the distribution is skewed, the median (85) is the appropriate measure of center, rather than the mean (which the low outlier would pull down).
4
Spread: the IQR of 10 describes the range of the middle 50% of the data, giving a resistant measure of spread that isn't distorted by the outlier.

Exams test whether you address all four SOCS components every time you're asked to describe a distribution, and specifically penalize incomplete descriptions that skip shape, outliers, center, or spread.

The most common trap is describing only shape and center while forgetting to explicitly mention outliers and spread — a complete answer must address all four components, even if the answer to one (like "no outliers") is brief.

1. What does the SOCS mnemonic stand for?
Shape, Outliers, Center, Spread.
Tap to reveal / hide
2. Name three possible shapes a distribution could have.
Any three of: symmetric, skewed left, skewed right, uniform, bimodal.
Tap to reveal / hide
3. Why is it important to check for outliers when describing a distribution?
Because outliers affect which measures of center and spread are most appropriate to use.
Tap to reveal / hide
4. If a distribution is skewed, which measure of center is more appropriate — mean or median?
Median.
Tap to reveal / hide
5. Why do exams penalize incomplete distribution descriptions?
Because a complete description requires addressing all four components: shape, outliers, center, and spread.
Tap to reveal / hide