Displaying and Describing Quantitative Data

Introduction

Quantitative data analysis is essential in business statistics, providing methods to summarize, visualize, and interpret numerical information. This section introduces the primary tools and concepts used to describe and standardize quantitative variables, enabling effective data-driven decision-making.

Visualizing Quantitative Variables

Histograms

Histograms are graphical representations of the distribution of a quantitative variable. They divide the data range into intervals (bins) and display the frequency of observations in each bin.

• No categories: Data are grouped into bins, not categories.

• Bar height: Represents the number of observations in each bin.

• Example: A histogram of average stock prices over several years can reveal trends and patterns not easily seen in raw tables.

Relative Frequency Histograms

Relative frequency histograms display the percentage of observations in each bin, facilitating comparison between datasets of different sizes.

• Shape: The shape remains unchanged; only the y-axis changes from counts to percentages.

• Example: Comparing sales data from two stores with different total sales volumes using relative frequency histograms.

Describing Distributions

Key Aspects

When describing a distribution, focus on three main aspects:

• Shape

• Center

• Spread

Describing Shape

• Modes: Peaks in a histogram are called modes.

  • Unimodal: One main peak.

  • Bimodal: Two main peaks.

  • Multimodal: Three or more peaks.

  • Uniform: All bars are approximately the same height; no clear mode.

• Symmetry: A distribution is symmetric if the left and right sides are mirror images. If one tail is longer, the distribution is skewed (right or left).

• Outliers: Observations that stand apart from the rest of the data. Outliers should be reported and considered, as they can affect statistical measures.

• Example: A histogram of employee salaries may show right skewness due to a few very high salaries (outliers).

Describing Center

• Mean: The arithmetic average of all data values.

• Median: The middle value when data are ordered. If the number of observations is even, the median is the average of the two middle values.

• Resistant Measures: The median is resistant to outliers and skewed data, while the mean is not.

• Example: In a dataset of home prices, the median may better represent the typical price if a few luxury homes skew the mean.

Describing Spread

• Range: The difference between the maximum and minimum values.

• Interquartile Range (IQR): The range of the middle 50% of the data.

  • Lower quartile (Q1): 25th percentile

  • Upper quartile (Q3): 75th percentile

• Standard Deviation: Measures the average distance of data values from the mean. Appropriate for symmetric distributions without outliers.

• Example: The standard deviation of monthly sales figures shows how much sales typically vary from the average.

Choosing Measures of Center and Spread

• If the distribution is skewed or contains outliers, use the median and IQR.

• If the distribution is roughly symmetric and has no outliers, use the mean and standard deviation.

• Example: For income data (often skewed), median and IQR are preferred.

Standardizing Variables

Z-Scores

Standardizing allows comparison of values from different distributions by expressing them in terms of standard deviations from the mean. The standardized value is called a z-score.

• Z-score formula:

• A z-score tells how many standard deviations a value is above or below the mean.

• Comparing z-scores for different variables (e.g., price and square footage) helps determine which value is more unusual relative to its distribution.

• Example: If a house price has a z-score of 2.5, it is 2.5 standard deviations above the mean price, indicating it is unusually high.

Special Cases and Summary Statistics

• Identify modes to determine if data can be split into groups.

• Report outliers and consider their effect on measures of center and spread.

• Pair the median with the IQR, and the mean with the standard deviation for summary statistics.

Summary Table: Measures of Center and Spread

Additional info: Standardization is especially useful in business analytics when comparing performance metrics across departments or industries with different scales.