BPSC Auditor, Panchayati Raj Dept · Module 52 · standard depth · 15 min
Statistical analysis, graphs and diagrams for Mains
What BPSC Auditor, Panchayati Raj Dept asks from this chapter
Syllabus lines this chapter covers:
- Prelims: General Studies · General mental ability
- Mains: General Studies I · Statistical analysis, graphs and diagrams
BPSC Auditor, Panchayati Raj Dept questions from this module are being sorted into our bank.
Cut to standard depth for BPSC Auditor, Panchayati Raj Dept. 7 deeper paragraphs are in the full chapter.
Start from zero: from a heap of numbers to a picture
A labour officer collects the daily wages of 50 workers at a road site. Fifty separate numbers on a sheet tell nothing at a glance. The officer needs to group them, find a typical wage, and draw a picture that a senior can read in seconds.
That is what this chapter teaches. We group raw data into a frequency table, find the mean, median and mode of grouped data, and draw histograms and ogives. We then learn how to pick the right diagram and how to write a short, clear paragraph about any data set.
Why this matters for Mains. Several state Mains papers carry a section on statistical analysis, graphs and diagrams. Those questions are descriptive: you must show the working, draw neatly and explain the result. Marks go to method and presentation as much as to the final number.
How this builds on the last chapter. The previous chapter worked with a short list of single values. Real surveys produce hundreds of values, which must be grouped before any measure can be found. The formulas here are the grouped versions of the same three averages.
Frequency tables and the grouped mean
Making classes. Find the smallest and largest values, choose a convenient width and make five to ten classes of equal width. A value on a boundary, such as ₹300, goes into the class that starts with it, so 300 belongs to 300–350. Count how many values fall in each class with tally marks; that count is the class's frequency.
The class mark. Each class is represented by its midpoint, called the class mark. For 300–350 it is (300 + 350) ÷ 2 = 325. We assume, as an approximation, that every value in the class sits at this point.
| Wage class (₹) | Workers (f) | Class mark (x) | f × x | Cumulative f |
|---|---|---|---|---|
| 200–250 | 5 | 225 | 1,125 | 5 |
| 250–300 | 10 | 275 | 2,750 | 15 |
| 300–350 | 18 | 325 | 5,850 | 33 |
| 350–400 | 12 | 375 | 4,500 | 45 |
| 400–450 | 5 | 425 | 2,125 | 50 |
The direct mean. Multiply each frequency by its class mark, add, and divide by the total frequency. Here Σfx = 16,350 and Σf = 50, so the mean wage is 16,350 ÷ 50 = ₹327. The Greek letter Σ simply means "add up".
The step method. Big products invite errors, so take an assumed mean A = 325, the class width h = 50, and u = (x − A) ÷ h, which gives −2, −1, 0, 1 and 2. Then Σfu = −10 − 10 + 0 + 12 + 10 = 2. The mean is A + (Σfu ÷ Σf) × h = 325 + (2 ÷ 50) × 50 = ₹327, the same answer with tiny numbers.
Why it matters. Examiners give marks for a neat table with all columns shown. The step method is the standard way to show working in a descriptive paper and cuts arithmetic slips.
Grouped median and mode
The median class. With n = 50 workers, the median is the value at position n⁄2 = 25. The cumulative frequencies are 5, 15, 33, 45 and 50, so the 25th worker falls in the class 300–350. That class is the median class.
The median formula. Inside the median class we assume values are spread evenly. The median = L + ((n⁄2 − cf) ÷ f) × h, where L = 300 is the lower limit, cf = 15 is the cumulative frequency before the class, f = 18 and h = 50. So the median = 300 + (10 ÷ 18) × 50 ≈ ₹327.8.
The modal class and mode. The class with the highest frequency, 300–350 with 18 workers, is the modal class. The mode = L + ((f₁ − f₀) ÷ (2f₁ − f₀ − f₂)) × h, where f₁ = 18, f₀ = 10 is the class before and f₂ = 12 the class after. So the mode = 300 + (8 ÷ 14) × 50 ≈ ₹328.6.
Reading the three together. Mean ₹327, median ₹327.8 and mode ₹328.6 are very close. That tells us the wage distribution is nearly symmetric, with no long tail of very high or very low wages. When the three measures sit far apart, the data is skewed, and the median is the safer summary.
Why it matters. The two formulas look heavy, but each is just "start at the class's lower limit and move part of the way across it". Understanding that idea makes them easy to remember and hard to misuse.
Histograms, frequency polygons and ogives
The histogram. Draw wage classes on the horizontal axis and frequency on the vertical axis, and raise a bar over each class. The bars touch, because the classes run on without gaps. This is the key difference from a bar graph, whose bars stand apart because its categories are separate things.
Unequal widths. When classes differ in width, the bar's area, not its height, must match the frequency. If two 50-rupee classes were merged into one 100-rupee class with 30 workers, its bar would be 15 high, not 30, in units of workers per 50 rupees. Forgetting this makes wide classes look far more crowded than they are.
The frequency polygon. Join the midpoints of the bar tops with straight lines, and close the shape on the axis at the class marks just beyond each end. The polygon shows the shape of the distribution and is handy for comparing two distributions on one chart.
The ogive. Plot cumulative frequency against the upper limit of each class: (250, 5), (300, 15), (350, 33), (400, 45) and (450, 50). Join the points with a smooth curve, which is the less-than ogive. Draw a line across from 25, half of 50, to the curve and down to the axis; it lands near ₹328, the median.
Two ogives. A more-than ogive plots the number of workers earning at least each lower limit, and it falls as wages rise. The two ogives cross at the median. This gives a second check on the formula and is a favourite Mains question.
Choosing the right diagram and writing about data
Start from the question. Before drawing, ask what the reader must see: a share, a comparison, a trend, a spread or a link. Each job has a diagram that does it best. A wrong choice, such as a pie for change over ten years, hides the very thing you want to show.
| Purpose | Best diagram |
|---|---|
| Parts of one whole | Pie chart |
| Compare separate categories | Bar graph |
| Change over time | Line graph |
| Spread of continuous data | Histogram or frequency polygon |
| Running totals, median, quartiles | Ogive |
| Link between two variables | Scatter diagram |
Good drawing habits. Give every diagram a title, label both axes with units, state the scale, and add a legend when there is more than one series. Start bar and histogram axes at zero, and use a ruler. These small things carry marks in a descriptive answer.
Writing about data. A good paragraph has four parts: an overview, the key figures, one comparison and one caution. For the wages, a model paragraph follows.
"Most of the 50 workers earn between ₹300 and ₹350 a day, and the mean wage is ₹327. The mean, median and mode are almost equal, so wages are spread fairly evenly around the centre. Only 5 workers earn ₹400 or more. Because grouped figures assume values sit at class midpoints, these averages are close estimates, not exact values."
Why it matters. Officers write notes on data every week. A paragraph that names the main pattern, backs it with two numbers and states its limits is exactly what a senior wants to read.
Where and why this is asked
Why examiners test it. Planning, budgets and welfare schemes run on survey data. An officer must summarise a distribution, draw it honestly and explain what it shows. A descriptive statistics question tests all three in one answer.
Where it shows up. Several state Mains General Studies papers include a section on statistical analysis, graphs and diagrams, with questions on grouped mean, median, mode, histograms and ogives. Prelims-level aptitude papers ask shorter versions, such as finding a class mark or the median class. Interviews sometimes ask candidates to explain an average in plain words.
When it goes wrong. The usual slips are using class limits instead of class marks in the mean, taking cf of the median class itself instead of the class before, drawing gaps between histogram bars and plotting ogive points at class marks instead of upper limits. Another is ignoring unequal class widths.
The habit to build. Lay out a full table with f, x, fx or fu and cumulative f before any formula. Write each formula, then the values, then the answer, so the examiner can follow every step.
Practice set: twelve solved problems
Problems 1 to 9 use this table of marks scored by 40 students: 0–10 has 3 students, 10–20 has 7, 20–30 has 12, 30–40 has 10 and 40–50 has 8. Try each problem first, then read the steps.
1. What is the class mark of the class 20–30?
Take the midpoint of the limits. It is (20 + 30) ÷ 2 = 25.
2. What are the cumulative frequencies?
Keep adding the frequencies: 3, 3 + 7 = 10, 10 + 12 = 22, 22 + 10 = 32 and 32 + 8 = 40. They are 3, 10, 22, 32 and 40.
3. Which is the median class?
With n = 40, n⁄2 = 20, and the cumulative frequency first reaches 20 in the third class. The median class is 20–30.
4. Find the mean mark.
The class marks are 5, 15, 25, 35 and 45, and Σfx = 15 + 105 + 300 + 350 + 360 = 1,130. The mean is 1,130 ÷ 40 = 28.25.
5. Find the mean by the step method with A = 25 and h = 10.
The u values are −2, −1, 0, 1 and 2, so Σfu = −6 − 7 + 0 + 10 + 16 = 13. The mean is 25 + (13 ÷ 40) × 10 = 28.25, as before.
6. Find the median mark.
Use L = 20, cf = 10, f = 12 and h = 10. The median is 20 + ((20 − 10) ÷ 12) × 10 ≈ 28.3.
7. Find the mode.
The modal class is 20–30 with f₁ = 12, f₀ = 7 and f₂ = 10. The mode is 20 + (5 ÷ 7) × 10 ≈ 27.1.
8. What percentage of students scored 40 or more?
That class has 8 of the 40 students. The share is 8 ÷ 40 × 100 = 20%.
9. If the classes 30–40 and 40–50 were merged into 30–50, how high should its histogram bar be, in students per 10 marks?
The merged class has 18 students over a width of 20. Its height is 18 ÷ 2 = 9, not 18.
10. Which diagram best shows the share of each crop in a district's sown area?
The data is parts of one whole at one time. The answer is a pie chart.
11. Which diagram best shows monthly rainfall through a year?
The data shows change over time. The answer is a line graph, though a bar graph is also acceptable.
12. Where do the less-than and more-than ogives of a distribution meet?
At the crossing point, as many values lie below as above. Its horizontal coordinate is the median.
Quick revision
- Class mark = (lower limit + upper limit) ÷ 2; boundary values go to the class that starts with them.
- Grouped mean = Σfx ÷ Σf, or A + (Σfu ÷ Σf) × h by the step method.
- Median = L + ((n⁄2 − cf) ÷ f) × h, with cf from the class before the median class.
- Mode = L + ((f₁ − f₀) ÷ (2f₁ − f₀ − f₂)) × h in the modal class.
- Histogram bars touch, and with unequal widths the area, not the height, shows frequency.
- Ogive points go at upper limits; read the median at n⁄2, or where two ogives cross.
- Pick the diagram by its job, and write about data with an overview, key figures, a comparison and a caution.