How does a Left Skewed vs. Right Skewed Distribution differ?

Skewness serves as a descriptive statistic that quantifies the degree and direction of asymmetry in a distribution. A value of zero indicates perfect symmetry, while positive or negative values confirm the presence of skew.

Visualizing the Direction of Skew

The most straightforward method for identifying the direction of skewness is by observing the location of the distribution’s “tail.” The tail is the elongated, thinner portion of the curve that extends away from the central cluster of data. The direction in which the tail points determines whether the distribution is left, right, or symmetrical.

A distribution is deemed left skewed (or negatively skewed) when the tail extends toward the smaller values on the horizontal axis. This happens because a small number of extremely low values are pulling the curve’s shape to the left, while the majority of observations are concentrated toward the right, forming the peak.

Conversely, a distribution is designated right skewed (or positively skewed) when the tail stretches out to the larger values on the right side of the axis. In this common scenario, the central mass of the data is situated on the left, and a few high outlying values pull the distribution towards the positive direction.

When a distribution exhibits no skew, it is perfectly symmetrical. This occurs when the left half of the curve is a mirror image of the right half, indicating that extreme values are equally balanced on both sides. This symmetrical shape is characteristic of a normal distribution.

It is important to remember that left-skewed distributions correlate with negative skewness coefficients, while right-skewed distributions yield positive coefficients.

The Relationship Between Measures of Central Tendency

A key statistical consequence of asymmetry is the altered relationship among the three main measures of central tendency: the mean, the median, and the mode. In a perfectly symmetrical distribution, these three values coincide at the peak. However, outliers present in skewed distributions pull the mean away from the median and the mode, creating a distinct statistical order.

In a Left Skewed Distribution, the sequence of central measures moves from right to left: the mode is the largest, followed by the median, and finally the mean, which is the smallest. This relationship is succinctly stated as: Mean < Median < Mode. The negative extreme values in the left tail drag the mean down significantly, making it less than the median, which remains closer to the center of the data set.

Conversely, a Right Skewed Distribution exhibits the opposite pattern. The presence of positive outliers pulls the mean up towards the long right tail, making it the highest value among the three measures. The order progresses from left to right: the mode is the smallest (located at the peak), followed by the median, and then the mean. The formal relationship is: Mode < Median < Mean.

Finally, in a symmetrical distribution (No Skew), all measures of center converge at the same point: Mean = Median = Mode. This equality is a defining feature of perfect symmetry, such as the standard normal distribution.

Interpreting Skewness Through Box Plots

While frequency histograms clearly show the tail, a box plot (or box-and-whisker plot) offers an alternative, highly effective visualization for diagnosing distribution asymmetry. The box plot graphically summarizes the five-number summary of a dataset, providing insight into its center, spread, and shape without needing to see the full data density.

The five critical values utilized in the construction of this statistical plot are:

• The Minimum Value (the smallest observation, excluding potential outliers).
• The First Quartile (Q1), which represents the 25th percentile of the data.
• The Median (Q2), which is the 50th percentile and bisects the data.
• The Third Quartile (Q3), which marks the 75th percentile of the observations.
• The Maximum Value (the largest observation, excluding potential outliers).

A box plot is constructed by drawing a box that spans from Q1 to Q3 (the Interquartile Range, or IQR). A vertical line is placed inside the box to mark the median. “Whiskers” then extend outward from the box to the minimum and maximum values (or non-outlier endpoints).

The asymmetry of the box and the length of the whiskers relative to each other determine the direction of the skew. In essence, the longer whisker or the larger segment of the box indicates the direction of the skew, mirroring the elongated tail seen in a frequency distribution.

When the data is right skewed, the median line is closer to Q1 (the bottom of the box), and the upper whisker (to the maximum value) is noticeably longer than the lower whisker. This indicates that the larger 25% of the data is more spread out than the smaller 25%.

When the data is left skewed, the median line shifts closer to Q3 (the top of the box), and the lower whisker (to the minimum value) is significantly longer than the upper whisker. This pattern confirms that the smaller values are responsible for the extended tail.

If the data is symmetrical, the median line rests near the center of the box, and the whiskers are roughly equal in length on each side.

Real-World Applications and Examples

Skewness is not merely a theoretical concept; it appears frequently across economics, biology, and demographics. Understanding the skew helps researchers correctly interpret population dynamics and resource allocation. Below are illustrative examples of datasets exhibiting left skew, right skew, and symmetry.

Example of Left-Skewed Distribution: Age at Death

The distribution of the age at which individuals pass away in industrialized populations is typically left-skewed. The vast majority of deaths occur at older ages, often concentrated between 70 and 90 years old. The long, thin tail extending to the left represents the relatively few individuals who die at younger ages due to accidents, disease, or other factors. This pattern causes the average age of death (the mean) to be slightly lower than the most common age of death (the mode).

Example of Right-Skewed Distribution: Household Income

The distribution of household incomes, particularly in countries like the U.S., provides a classic example of severe right skew. Most households cluster around a moderate income level (e.g., $40,000 to $80,000 per year), forming a tall peak on the left side. However, the presence of a few extremely wealthy individuals and families creates a prolonged, shallow tail stretching far to the right. These high outliers pull the mean income significantly above the median income, illustrating why the median is often preferred when discussing typical income levels.

Example of Symmetrical Distribution: Adult Male Height

The distribution of heights for adult males (or females) is widely regarded as a practical example of a symmetrical or normal distribution. Heights cluster around the central average, such as 69.1 inches for U.S. males, with deviations occurring equally in both the shorter and taller directions. Because there is no significant pull from outliers on either side, the mean, median, and mode are almost identical, resulting in a balanced, bell-shaped curve.