- Foundation
- About 22 min
- 5 practice questions
Analysing the project data: averages, dispersion and correlation
Calculate the mean, median, mode and standard deviation of a grouped variable from the project and the correlation between income and spending, and interpret them for the report.
If something looks wrong, please tell us.
By the end of this lesson you can
- Calculate the mean and standard deviation of grouped data by the step deviation method.
- Find the median and mode of the same data and compare them.
- Calculate and interpret the correlation between two project variables.
Analysis turns tables into findings. The project uses measures of central tendency for the typical value, a measure of dispersion for how spread out the values are, and correlation for relationships.
Throughout this chapter we follow one made-up project: Navya, a young entrepreneur (made up), wants to start a small unit making bathing soap and asks you to study the market first. Every figure in the project is illustrative.
Standard deviation in brief
A measure of dispersion shows how far values are spread around the average. The project uses the standard deviation (σ): the square root of the average squared deviation from the mean. For grouped data by the step deviation method, with d' = (m − A) ÷ c:
Mean = A + (Σfd' ÷ N) × c. Standard deviation σ = c × √[Σfd'² ÷ N − (Σfd' ÷ N)²].
Worked example 1: mean and standard deviation of spending on soap
| Spending (₹) | Mid-value m | f | d' = (m − 125) ÷ 50 | fd' | fd'² |
|---|---|---|---|---|---|
| 0-50 | 25 | 10 | −2 | −20 | 40 |
| 50-100 | 75 | 30 | −1 | −30 | 30 |
| 100-150 | 125 | 50 | 0 | 0 | 0 |
| 150-200 | 175 | 22 | 1 | 22 | 22 |
| 200-250 | 225 | 8 | 2 | 16 | 32 |
| Total | 120 | −12 | 124 |
Mean: 125 + (−12 ÷ 120) × 50 = 125 − 0.1 × 50 = 125 − 5 = ₹120 a month. Standard deviation: Σfd'² ÷ N = 124 ÷ 120 = 1.0333; (Σfd' ÷ N)² = (−0.1)² = 0.01; difference = 1.0233; square root ≈ 1.0116; σ = 50 × 1.0116 ≈ ₹50.58. Interpretation: a typical household spends about ₹120 a month on soap, and spending typically varies by about ₹50 either side of that.
Worked example 2: median and mode of the same data
Median: N ÷ 2 = 60; cumulative frequencies 10, 40, 90 ... so the median class is 100-150 (c.f. before = 40, f = 50, h = 50). Median = 100 + ((60 − 40) ÷ 50) × 50 = 100 + 20 = ₹120. Mode: modal class 100-150; D1 = 50 − 30 = 20; D2 = 50 − 22 = 28. Mode = 100 + (20 ÷ (20 + 28)) × 50 = 100 + (20 ÷ 48) × 50 = 100 + 20.83 = ₹120.83.
Interpretation: mean (120), median (120) and mode (about 120.83) are almost equal, so spending is fairly evenly spread on both sides of ₹120. For Navya, a soap priced so that a household's monthly use costs around ₹120 would match the typical budget.
Worked example 3: income and spending on soap
Illustrative data for six of the surveyed households: monthly income X (₹ thousand) and monthly soap spending Y (₹).
| X (₹ thousand) | Y (₹) | x = X − 25 | y = Y − 110 | x² | y² | xy |
|---|---|---|---|---|---|---|
| 10 | 60 | −15 | −50 | 225 | 2,500 | 750 |
| 15 | 90 | −10 | −20 | 100 | 400 | 200 |
| 20 | 80 | −5 | −30 | 25 | 900 | 150 |
| 25 | 110 | 0 | 0 | 0 | 0 | 0 |
| 30 | 140 | 5 | 30 | 25 | 900 | 150 |
| 50 | 180 | 25 | 70 | 625 | 4,900 | 1,750 |
| ΣX = 150 | ΣY = 660 | 0 | 0 | 1,000 | 9,600 | 3,000 |
Means: X̄ = 150 ÷ 6 = 25; Ȳ = 660 ÷ 6 = 110. r = Σxy ÷ √(Σx² Σy²) = 3,000 ÷ √(1,000 × 9,600) = 3,000 ÷ √96,00,000 = 3,000 ÷ 3,098.4 = 0.97. Interpretation: a very high positive correlation: higher-income households spend more on soap. This suggests Navya could offer a low-priced soap for lower-income buyers and a premium one for higher-income buyers. Remember that correlation shows covariation; bigger families, which may also earn more, could be part of the reason.
Worked example 4: satisfaction by brand
| Brand (made-up names) | Users | Satisfied | Per cent satisfied |
|---|---|---|---|
| Brand P | 40 | 34 | 85 |
| Brand Q | 35 | 21 | 60 |
| Brand R | 25 | 20 | 80 |
| Brand S | 20 | 10 | 50 |
Working: 34 ÷ 40 × 100 = 85; 21 ÷ 35 × 100 = 60; 20 ÷ 25 × 100 = 80; 10 ÷ 20 × 100 = 50. Most-used brand (mode): P, with 40 users. Observation: users of brands Q and S are the least satisfied, so they are the buyers most likely to try a new soap.
Which tool answers which project question
-
Typical value
- Mean, median or mode
-
Spread
- Standard deviation, range
-
Relationship
- Scatter diagram, Karl Pearson's r, rank correlation
-
Most popular choice
- Mode of an attribute such as brand
Key takeaways: the mean and standard deviation of grouped data are found by step deviations; the median and mode give the middle and most common values, and when the three nearly coincide the distribution is roughly symmetric. Correlation shows whether two variables, such as income and spending, move together. Percentages and modes summarise attributes such as brand and satisfaction.
Check your understanding
-
1. Illustrative: with A = 300, c = 100, Σfd' = 20 and N = 50, the mean is:
- 340
- 320
- 302
- 400
Show answer
Answer: A. Mean = A + (Σfd' ÷ N) × c = 300 + (20 ÷ 50) × 100 = 300 + 0.4 × 100 = 340.
-
2. Illustrative: with c = 10, Σfd'² ÷ N = 2.25 and Σfd' ÷ N = 0, the standard deviation is:
- 22.5
- 2.25
- 15
- 1.5
Show answer
Answer: C. σ = c × √[Σfd'² ÷ N − (Σfd' ÷ N)²] = 10 × √(2.25 − 0) = 10 × 1.5 = 15.
-
3. In a project, which measure shows how spread out household spending is around the average?
- Mode
- Standard deviation
- Median
- Correlation coefficient
Show answer
Answer: B. The standard deviation is a measure of dispersion: it shows how far values are spread around the mean.
-
4. Illustrative: 18 of 24 users of a brand are satisfied. The percentage satisfied is:
- 18%
- 24%
- 80%
- 75%
Show answer
Answer: D. Percentage = (18 ÷ 24) × 100 = 0.75 × 100 = 75%.
-
5. Assertion (A): A correlation of 0.97 between income and soap spending shows that higher income alone causes higher spending. Reason (R): Correlation measures covariation, not causation.
- Both A and R are true, and R is the correct explanation of A
- Both A and R are true, but R is not the correct explanation of A
- A is true, but R is false
- A is false, but R is true
Show answer
Answer: D. A is false: a high r shows the two move together, not that one alone causes the other; other factors such as family size may play a part. R is true and is exactly why A cannot be concluded.
Sign in to keep track of the lessons you finish.