A bar chart compares categories; a line chart is useful for ordered time observations; a pie chart shows parts of a meaningful whole; a histogram represents the distribution of a quantitative variable over intervals. These forms are not interchangeable. Choose the chart after deciding what relationship must be communicated.

Every graph needs a title, labels, units and an intelligible scale. Where relevant add the reference period, source and explanation of missing values. A polished chart without these features may be impossible to interpret correctly. Visual attractiveness cannot compensate for an unidentified denominator.

The numbers below are invented teaching datasets, not Bihar expenditure or performance data. They show how an aspirant can interpret and criticise administrative information. Do not attach them to a real district or department without replacing them with sourced observations.

Suppose a hypothetical ₹200 crore programme has four components: water ₹80 crore, roads ₹60 crore, health ₹40 crore and training ₹20 crore. Shares are 40%, 30%, 20% and 10%. A full circle is 360°, so sector angle = component/total × 360°. The angles are 144°, 108°, 72° and 36°, which sum to 360°.

A pie chart is meaningful here because the four mutually exclusive components sum to the stated programme total. It becomes misleading if categories overlap or the total is incomplete. If health spending includes training already counted separately, the chart would double-count part of the whole unless the classification is corrected.

A share is not a growth rate. If the overall programme budget doubles while one component’s share falls, the component’s absolute allocation may still rise. For example, 40% of ₹200 crore is ₹80 crore; 30% of ₹400 crore is ₹120 crore. Always inspect both the whole and the part.

Bars encode magnitude through length, so a zero baseline is normally important. If values are 95 and 100 and the vertical axis begins at 90, their visible bar lengths may appear to be 5 and 10, making a modest difference look like a doubling. The actual relative increase is (100−95)/95 × 100, about 5.26%.

A truncated axis can sometimes reveal small time-series fluctuations, but it must be clearly marked and interpreted carefully. The issue is not that every non-zero axis is forbidden; it is whether the visual encoding exaggerates or conceals the relationship relevant to the question. Bars and line plots use baseline cues differently.

Three-dimensional decoration can distort apparent areas and make front segments look larger. Use simple two-dimensional forms for exact comparison. When criticising a graph in an exam, identify the specific distortion and explain its effect instead of writing only that the chart is misleading.

A line chart with equally spaced horizontal positions implies equal time intervals unless labelled otherwise. If observations from 2010, 2011 and 2020 are placed as if they are consecutive equally spaced years, apparent trends and slopes can mislead. Show the real spacing or make the categorical treatment explicit.

An index with base 100 summarises relative change. A value of 125 means 25% above the base-period value, not an increase of 125%. Two series both starting at 100 can be compared for relative growth but not for their original absolute size. A small economy and a large economy can have the same index.

Nominal and real series should not be mixed without explanation. A rising current-price expenditure line may reflect inflation as well as more services. Before concluding that service volume increased by the same percentage, examine price changes, population and the actual output measure.

A histogram groups numerical observations into intervals. With equal-width intervals, frequencies can be shown through bar heights. With unequal-width intervals, area rather than raw height must represent frequency, so frequency density equals frequency divided by class width. Otherwise a wide interval can appear more important merely because of its width.

Hypothetical example: the 0–10 interval has 20 observations and the 10–30 interval has 30. Their widths are 10 and 20, so densities are 2 and 1.5 observations per unit. The second interval has more observations but lower density. A histogram using raw frequencies as heights would misrepresent the distribution when widths differ.

Averages can conceal spread and subgroup differences. Two districts can have the same mean rainfall but different month-to-month variability. Without a distribution or variation measure, the mean alone cannot establish equal reliability. This connects statistical interpretation with Bihar’s agriculture and water-management questions.

First check whether the title matches the data. Then inspect units, denominators, time intervals, category overlap, missing observations and the scale. Finally ask whether the written conclusion goes beyond what the graph shows. A chart can be technically well drawn yet used to support an unjustified causal claim.

Consider a hypothetical graph showing higher expenditure and better outcomes in the same years. It demonstrates co-movement, not necessarily that expenditure alone caused the change. Population, prices, other policies and reporting may also change. Explain what further evidence would help, such as comparable measures and a credible evaluation design.

For an exam response, state two or three concrete findings with calculations, followed by relevant limitations. Do not list every possible flaw regardless of the actual graph. Precision means matching criticism to observed features and explaining whether correcting them might change the conclusion.