Describing One Quantitative Variable

Overview

When analyzing a single quantitative variable, it is essential to summarize its distribution by describing its shape, center, variability, and identifying any outliers. These summaries help us understand the main features of the data and make meaningful comparisons.

Visual Displays for Quantitative Data

Types of Visual Displays

• Dot Plot: Displays each data value as a dot above its value on a number line. Useful for small datasets and for identifying individual values.

• Histogram: Groups data into intervals (bins) and displays the frequency or proportion of values in each bin as bars. Best for visualizing the overall shape of the distribution.

• Box Plot: Summarizes data using the five-number summary (minimum, Q1, median, Q3, maximum) and identifies outliers as asterisks. Useful for detecting skewness and outliers.

Note: Box plots show the median but not the mean. Dot plots and histograms do not directly show the mean or median, but these can be estimated visually.

Describing Distributions

Shape

The shape of a distribution describes the general appearance of the data when plotted. Common shapes include:

• Skewed to the Right: Most values are on the left with a long tail to the right. Example: Birth rates in countries, where most countries have low birth rates but a few have high rates.

• Skewed to the Left: Most values are on the right with a long tail to the left. Example: Life expectancy, where most countries have high life expectancy but a few have low values.

• Bell-Shaped: Data are symmetrically distributed around a central value, forming a bell-like curve. Most values are near the center, with fewer values in the tails.

• Symmetric but Not Bell-Shaped: The left and right sides are mirror images, but the distribution may be bimodal (two peaks) or uniform (rectangular).

Example: In a bell-shaped distribution, such as heights of adult humans, most values cluster around the mean, with fewer extremely short or tall individuals.

Outliers

An outlier is a value that is notably distinct from the rest of the data. Outliers can be high (unusually large) or low (unusually small). They are identified in box plots as asterisks and should always be interpreted in context. Outliers may result from data entry errors, natural variation, or indicate subgroups within the data.

• Example: An elephant with a lifespan of 40 years is an outlier among mammals with much shorter lifespans.

Numerical Summaries

Measures of Center

• Mean (\( \bar{x} \) for samples, \( \mu \) for populations): The arithmetic average of all values. Not resistant to outliers.

• Median: The middle value when data are ordered. More resistant to outliers and skewed data. (Covered in more detail in Section 2.3.)

Formula for the Mean:

Measures of Variability

• Range: Difference between the maximum and minimum values. Not resistant to outliers.

• Standard Deviation (\( s \) for samples, \( \sigma \) for populations): Measures the typical distance of values from the mean. Not resistant to outliers.

• Interquartile Range (IQR): The range of the middle 50% of values. More resistant to outliers. (Covered in more detail in Section 2.3.)

Formula for Sample Standard Deviation:

The 95% Rule (Empirical Rule)

For bell-shaped distributions, approximately 95% of values lie within two standard deviations of the mean:

• Interval: [mean - 2 × standard deviation, mean + 2 × standard deviation]

Example: If the mean pulse rate is 69.6 bpm and the standard deviation is 12.2 bpm, then about 95% of pulse rates are between 45.2 bpm and 94.0 bpm.

Z-Scores

The Z-score of a value indicates how many standard deviations it is from the mean. Z-scores help identify unusual values.

• Formula for Sample Z-score:

• Formula for Population Z-score:

• Z-scores greater than 2 or less than -2 are unusual in bell-shaped distributions.

• For non-bell-shaped distributions, Z-scores with absolute value 3 or more are considered unusual.

Example: A systolic blood pressure of 204 mmHg with a mean of 120 mmHg and a standard deviation of 38.5 mmHg has a Z-score of 2.18, indicating it is 2.18 standard deviations above the mean.

Summary Table: Visual Displays and Their Uses

Additional info: The notes reference StatKey, an online tool for statistical visualizations, and mention that the interquartile range and median are covered in more detail in Section 2.3. The image included is a bell, which visually reinforces the concept of a bell-shaped distribution.