10.1 Descriptive Statistics: Mean, Median, Mode & Weighted Averages
Key Takeaways
For a nonempty group, the mean times the number of observations gives their sum.
When disjoint groups use the same measurement, weight each group mean by its count to obtain the overall mean.
In sorted data, use the middle value for odd , or average the two middle values for even .
Deviations from the arithmetic mean sum to zero. For a target mean, the missing value must balance the other deviations.
The range is maximum minus minimum. The median uses middle positions and is less sensitive to extreme magnitudes than the mean.
Descriptive Statistics: Mean, Median, Mode & Weighted Averages
The Arithmetic Mean and the Total
For values of the same measure, the arithmetic mean is their sum divided by the number of observations:
If the exact mean is known, recover the total by multiplying by the count:
Subtract the known values from that total to recover a single missing value. A rounded mean may only give an approximate total.
The Missing Value Problem: Target Score Strategies
If scores count equally, an average target sets a required sum. Check that the resulting missing score is within the scale.
The Standard Total Sum Method
In this exercise, a student earns 82%, 78%, 91% and 84% on four exams. All five exams count equally. What score on the fifth exam will bring the mean to 85%? Work with scores on a 0-to-100 scale.
- Determine the total sum required:
- Sum the 4 completed exam scores:
- Subtract the current sum from the required sum:
The Deviation from the Mean Shortcut (Mental Math Accelerator)
If the scores are close to the target mean, calculate their differences from the target:
The algebraic sum of all deviations from the mean must always equal zero. Instead of summing large raw integers, measure each known score against the target average of 85:
- Score 1:
- Score 2:
- Score 3:
- Score 4:
Now, sum the deviations: . The student currently has a net deficit of 5 points relative to the target. To balance the sum of deviations to zero, the final exam must offset this deficit with a deviation of :
Both methods use the same target total. Choose the one with smaller arithmetic.
Combine Group Means
For disjoint groups using the same measurement and giving each observation equal weight, multiply each group mean by its group size to recover its sum. Divide the sum of those totals by the total count. The simple mean of the group means works when their sizes are equal. For two groups with different means and different sizes, it gives the wrong result.
Worked Group Mean
In this exercise, 30 participants in group 1 average 80 points and 20 in group 2 average 90 points on the same test, with one score per person. What is the mean for all 50 participants?
- The Trap: Averaging the two averages: (Incorrect).
- The Reason: Group 1 has 30 soldiers, while Group 2 has only 20 soldiers. Group 1 exerts 50% more weight on the outcome than Group 2. The combined average must pull closer to 80 than to 90.
- The Rigorous Calculation:
Use the Distance Between the Means
You can solve weighted averages mentally using the Inverse Ratio of Distances (the seesaw principle):
- Find the ratio of the sample sizes: .
- The total spread between the averages is points.
- The balance point divides this 10-point spread in the inverse ratio of the weights (, meaning 2 parts out of parts total):
- Add this shift to the lower average: points. The solution requires no three-digit multiplication.
Median: Determining Central Tendency for Odd and Even Datasets
Sort the observations and find the middle position or positions. The median depends on the central values. Making an observation already at the high end even larger will leave the median unchanged if the central values are unaffected.
Step-by-Step Median Protocol
- Sort the dataset: Never attempt to pick a median from unsorted data. Write the numbers in ascending order.
- Identify sample size ():
- Odd : The median is the single value located at position . For , the median is the 5th value: .
- Even : The dataset divides evenly into two equal halves. The median is the arithmetic mean of the two central numbers at positions and .
Even Dataset Example: An artillery section records 6 response times (in seconds): .
- Ordered sequence: .
- Here . The two central positions are (which is 28) and (which is 30).
- Compute their average: .
Mean vs. Median Under Skewed Conditions
In , the mean and median both equal 4. Replacing 6 with 26 leaves the median at 4 and raises the mean to 8. Adding or removing observations can change the median, because it can change which positions are central.
Mode and Range: Rapid Dispersion Identification
The Mode
The mode is the most frequently occurring value in a dataset:
- Unimodal: Exactly one value occurs with highest frequency (e.g., in , the mode is ).
- Bimodal: Two distinct values tie for the highest frequency.
- Multimodal: More than one value ties for the highest frequency. Bimodal is the two-mode case.
- No distinct mode: When all values have the same frequency, no value stands out. In these exercises, report no mode when every value appears once.
The Range
The range is the simplest measure of statistical dispersion, quantifying the total spread of the dataset:
The range changes if a new observation sets a new minimum or maximum. A new observation between those limits leaves it unchanged.
Descriptive Statistics Quick-Reference Table
| Statistical Measure | Formal Mathematical Definition | Calculation | Sensitivity to Outliers |
|---|---|---|---|
| Arithmetic Mean | Convert to sum: | Affected by magnitude | |
| Weighted Mean | For grouped data, weight by group size | Depends on the extreme value and its weight | |
| Median | Center of sorted array | Sort array; pick (odd) or average and (even) | Depends on the central sorted values |
| Mode | Value with highest frequency | Count how often each value occurs | Depends on frequencies |
| Range | Identify extremes and compute single subtraction | Depends only on the extremes |
Checks to Use During Practice
- Recover totals before combining groups or inferring a missing score. Keep the same units and weighting.
- When combining two different means and both groups have members, the result lies between the two means. If the groups have different sizes, the result is closer to the larger group’s mean.
- Sort the values before selecting the median. Count repeats as separate observations.
A student has scores of 76%, 84%, 81% and 87% on four modules. All five modules have equal weight, and the exact mean must be at least 83%. What minimum score is needed on the fifth module?
85%
87%
89%
91%
Every member of two disjoint groups takes the same fitness test once. The 60 people in group Alpha average 78 points. The 40 people in group Bravo average 88 points. What is the combined mean for all 100 people?
81 points
83 points
82 points
84 points
An instructor records the target engagement reaction times (in seconds) of eight recruits: 14, 22, 11, 18, 25, 16, 19, and 15. What are the median and range of this dataset?
Median = 17 seconds; Range = 14 seconds
Median = 17.5 seconds; Range = 14 seconds
Median = 17 seconds; Range = 11 seconds
Median = 18 seconds; Range = 14 seconds
Sections you finish are checked off in the contents.