Measures of central tendency Revision Notes

Measures of central tendency summarise a dataset by identifying a central or typical value. The three main measures are mean, median, and mode. Each provides different insights about the data's distribution and is useful depending on the context and data type.

The mean gives the arithmetic average, the median indicates the middle value when data are ordered, and the mode shows the most common value. Understanding how to calculate these from small datasets or ungrouped frequency distributions is fundamental in statistics.

• Mean: average value, sensitive to all data points.
• Median: middle value, robust to extreme values.
• Mode: most frequent value, useful for categorical or discrete data.

The mean is calculated by adding all data values and dividing by the total number of values. This gives a measure of the 'central' value considering every data point.

For ungrouped frequency distributions, multiply each distinct data value by its frequency, sum these products, then divide by the total frequency.

Formula for raw data: \( \text{Mean} = \frac{\sum x_i}{n} \) where \( x_i \) are data values and \( n \) is the number of values. For frequency data: \( \text{Mean} = \frac{\sum f_i x_i}{\sum f_i} \) where \( f_i \) are frequencies.

• Sum all data values or sum of (value × frequency).
• Divide by total number of data points or total frequency.
• Result is the average value representing the dataset.

Consider the dataset: 3, 7, 5, 9, 6. To find the mean:

1. Sum the values: 3 + 7 + 5 + 9 + 6 = 30. 2. Count the values: 5. 3. Divide sum by count: 30 ÷ 5 = 6. The mean is 6.

• Sum = 30, Number of values = 5.
• Mean = 30 ÷ 5 = 6.

The median is the middle value when data are arranged in ascending order. It divides the dataset into two equal halves.

If the number of data points is odd, the median is the value at position \( \frac{n+1}{2} \). If even, it is the average of the two middle values at positions \( \frac{n}{2} \) and \( \frac{n}{2} + 1 \).

For ungrouped frequency distributions, expand the frequencies into a list or use cumulative frequencies to locate the median position.

• Order data from smallest to largest.
• Find middle position: \( (n+1)/2 \) if odd, average two middle values if even.
• Median represents the central value, less affected by outliers.

Given the dataset: 8, 3, 5, 7, 9, arrange in order: 3, 5, 7, 8, 9.

Number of values \( n = 5 \) (odd), median position is \( (5+1)/2 = 3 \).

The third value is 7, so the median is 7. For an even number of values, average the two middle values.

• Ordered data: 3, 5, 7, 8, 9.
• Median position: 3rd value.
• Median = 7.

The mode is the value that appears most frequently in the dataset. A dataset can have one mode (unimodal), more than one mode (bimodal or multimodal), or no mode if all values occur equally.

For ungrouped frequency distributions, the mode corresponds to the value with the highest frequency.

The mode is especially useful for categorical data or when the most common value is of interest.

• Find the value(s) with the highest frequency.
• Mode can be single, multiple, or none.
• Useful for identifying common or popular values.

A cumulative frequency distribution shows the total number of observations less than or equal to the upper boundary of each class interval. It is useful for summarising grouped data and helps identify the distribution shape and central values.

Unlike simple frequency tables, cumulative frequencies accumulate counts progressively, making it easier to estimate medians and other percentiles graphically or by calculation. However, exact values of individual data points are not available, so approximations are necessary.

• Cumulative frequency at a class boundary = sum of frequencies up to that boundary.
• Data is grouped into class intervals with upper boundaries.
• Useful for estimating median and other quantiles.

The median is the value that divides the data into two equal halves. To estimate the median from a cumulative frequency distribution, first identify the total number of observations (n) and find n/2.

Locate the class interval where the cumulative frequency first equals or exceeds n/2 — this is the median class.

Assuming data is evenly distributed within this class, use linear interpolation to estimate the exact median value with the formula:

• Median = L + \( \frac{\frac{n}{2} - F}{f} \) × w
• where: L = lower boundary of median class
• F = cumulative frequency before median class
• f = frequency of median class
• w = width of median class

Suppose a cumulative frequency table shows 50 total observations. The cumulative frequency reaches 25 at the upper boundary of the 20–30 class, and the previous cumulative frequency is 15 at 20.

Here, n = 50, so n/2 = 25. The median class is 20–30.

Using linear interpolation: Median = 20 + ((25 - 15) / (25 - 15)) × 10 = 20 + (10/10) × 10 = 20 + 10 = 30.

• Median estimated as 30 units.
• Interpolation assumes uniform distribution within the class.

To approximate the mean from grouped data, use the midpoints of each class interval as representative values.

Calculate the frequency for each class by subtracting cumulative frequencies.

Multiply each midpoint by its class frequency, sum these products, then divide by the total number of observations (n). This gives an approximate mean:

• Mean ≈ \( \frac{\sum (f_i × x_i)}{n} \)
• where f_i = frequency of class i
• x_i = midpoint of class i
• n = total frequency

Given cumulative frequencies: 10 at 10, 30 at 20, 50 at 30, frequencies for classes are: 10 (0–10), 20 (10–20), 20 (20–30). Midpoints are 5, 15, and 25 respectively.

Calculate: (10×5) + (20×15) + (20×25) = 50 + 300 + 500 = 850.

Total n = 50, so mean ≈ 850 / 50 = 17.0 units. This is an approximation because actual data points within classes are unknown.

• Mean estimated as 17.0 units.
• Relies on midpoint assumption for each class.

A grouped frequency distribution organises data into class intervals (ranges) and records the frequency (number of data points) within each interval. This summarises large datasets by showing how data values are spread across intervals rather than listing individual values.

Each row in the distribution includes a class interval and its corresponding frequency. For example, a distribution of test scores might group scores into intervals like 0–9, 10–19, 20–29, and so on, with frequencies showing how many students scored within each range.

• Class intervals represent ranges of data values.
• Frequency shows how many data points fall into each interval.
• Grouped distributions simplify large datasets for easier analysis.

The modal class is the class interval that contains the highest frequency in a grouped frequency distribution. It represents the range where the most data points are concentrated, indicating the most common values in the dataset.

Unlike the mode in ungrouped data, which is a single value, the modal class is an interval because data are grouped. This means the mode is approximated to a range rather than an exact number.

• Modal class = class interval with the greatest frequency.
• Indicates the most common range of data values.
• Provides a summary measure of central tendency for grouped data.

To find the modal class, follow these steps: 1) Look at the frequency column in the grouped frequency distribution. 2) Identify the largest frequency value. 3) The class interval corresponding to this largest frequency is the modal class.

For example, consider the following grouped frequency distribution of daily temperatures (°C):

| Temperature (°C) | Frequency | |-----------------|-----------| | 10–14 | 3 | | 15–19 | 7 | | 20–24 | 12 | | 25–29 | 9 | | 30–34 | 5 | Here, the highest frequency is 12, which corresponds to the class interval 20–24. Therefore, the modal class is 20–24 °C.

• Scan the frequency column for the highest value.
• Match this frequency to its class interval.
• That interval is the modal class.

The modal class indicates where the bulk of data points lie within the grouped distribution. It helps identify the most typical range of values in the dataset, which can be useful for summarising and comparing data.

However, the modal class does not provide information about the exact mode value within the interval or the distribution shape. It is a descriptive measure that complements other statistics like the mean and median.

• Shows the most common data range.
• Useful for understanding data concentration.
• Does not specify an exact mode value.

A survey records the number of hours students study per week, grouped as follows:

| Hours Studied | Frequency | |---------------|-----------| | 0–4 | 5 | | 5–9 | 8 | | 10–14 | 15 | | 15–19 | 12 | | 20–24 | 10 | Step 1: Identify the highest frequency. Here, it is 15.

Step 2: Find the class interval with frequency 15, which is 10–14 hours. This is the modal class, indicating most students study between 10 and 14 hours per week.

• Highest frequency = 15
• Modal class = 10–14 hours
• Interpretation: Most students study 10 to 14 hours weekly.

Measures of central tendency summarise a dataset by identifying a central or typical value. The three main measures are the mean, median, and mode, each providing different insights depending on the data's nature. Choosing the correct measure is essential for accurate data interpretation.

The mean is calculated by summing all data values and dividing by the number of values. It uses every data point, making it sensitive to all information in the dataset.

For example, for data values 4, 7, 9, 10, 20, the mean is (4 + 7 + 9 + 10 + 20) ÷ 5 = 50 ÷ 5 = 10.

The mean is appropriate when data are roughly symmetric and free from extreme values (outliers). However, a single very high or low value can distort the mean, making it unrepresentative of the typical data point.

• Mean formula: \( \bar{x} = \frac{\sum x_i}{n} \) where \(x_i\) are data values and \(n\) is the number of values.
• Sensitive to outliers and skewed data.

The median is the middle value when data are arranged in order. If there is an even number of values, it is the average of the two middle values.

For example, for data 3, 5, 7, 9, 11, the median is 7. For data 3, 5, 7, 9, the median is (5 + 7) ÷ 2 = 6.

The median is not affected by extreme values, making it a better measure for skewed distributions or data with outliers. However, it does not use all data points, so it may ignore some information.

• Order data before finding median.
• Robust against outliers and skewed data.
• Does not reflect all data values.

The mode is the most frequently occurring value in a dataset. It is especially useful for categorical data where mean and median cannot be calculated.

For example, in the data set 2, 3, 3, 5, 7, the mode is 3 because it appears twice, more than any other value.

The mode can be used to identify the most common category or value. However, some datasets may have no mode (all values unique) or multiple modes (bimodal or multimodal), which can complicate interpretation.

• Useful for categorical or nominal data.
• May be no mode or multiple modes.
• Does not consider numerical magnitude.

Selecting the best measure depends on the data type, distribution shape, and presence of outliers.

Use the mean for symmetric, numerical data without outliers, as it reflects all values.

Use the median for skewed data or when outliers are present, as it better represents the central location without distortion from extremes. For example, income data often use the median because a few very high incomes skew the mean upwards, making the median a fairer central value representation. Use the mode for categorical data or when the most common value is of interest, such as the most popular product size sold.

• Consider data type: numerical (mean, median) vs categorical (mode).
• Assess distribution shape: symmetric (mean), skewed (median).
• Check for outliers: median preferred if outliers exist.
• Contextual relevance: what does the measure represent in the real-world scenario?