Choosing the right diagram or map

Detail

1. The basic rule: the data decides the method

• The kind of data decides the diagram or map, not the other way round. Before drawing, ask three questions:
• What is the data? Is it a count (number of people), a ratio (people per sq km, % share), or a reading that changes smoothly over space (temperature, rainfall)?
• What is the question? Is it change over time, a comparison of items, parts of a whole, or how a value spreads over space?

• How many variables are there? One variable, several variables, or three parts that add up to 100%?

• Diagram vs map:

• A diagram (graph, bar, pie) shows how much and when.

• A thematic map (a map that shows one theme, such as population or rainfall, on top of a base map) shows where.

• Wrong choice → misleading picture. For example, a choropleth of raw totals makes large districts look "high" only because they are big.

2. Diagrams: a key to match data with method

Data / question	Best method	Why it fits
Time series, one variable (e.g. decadal population growth)	Line graph	A continuous line shows the trend and its rises and falls
Time series, several variables (e.g. growth rates of 3 states)	Polygraph	Several lines on one set of axes, each in a different colour or pattern
Comparison of items (e.g. literacy rate of states)	Simple bar diagram	Bars of equal width and equal gaps; only the height (length) varies
Several variables per item (e.g. rice area in 2 years for each state)	Multiple bar diagram	Bars for each variable are placed side by side for every item
Parts of a total (e.g. rural vs urban population of states)	Compound (divided) bar	One bar per item, divided into its parts
Shares of one total (e.g. land use shares in one year)	Pie diagram	Circle = 100% = 360°; each part's angle = (value ÷ total) × 360
Frequency distribution	Histogram or frequency polygon	Grouped data in class intervals
Cumulative distribution	Ogive	Running total of frequencies
Climate of a station	Line and bar graph, climograph	Temperature as a line, rainfall as bars, both for 12 months
Wind directions	Wind rose	Bars from a centre point in each compass direction
Age-sex structure	Age-sex pyramid	Horizontal bars for each age group: males on one side, females on the other
Inequality	Lorenz curve	Distance from the line of equality shows how unequal it is
Three-part mix	Ternary diagram	Triangle with three axes that add up to 100%

3. Key diagram terms explained

• Line graph: points for each time period, joined by a straight line.
• The x-axis carries time (years). The y-axis carries the value.

• Best for continuous change over time, such as population growth across census decades.

• Polygraph: a line graph with two or more variables on the same axes.

• Each line has its own colour or pattern so that trends can be compared directly.

• Bar diagrams (three kinds):

• Simple bar: one bar per item.
• Multiple bar: several bars side by side for each item. Used when two or more variables are compared, such as rice area by state for two years.

• Compound bar (divided bar): one bar split into parts, showing the components of a total.

• Pie diagram: a circle divided into sectors.

• It suits one total and its shares.

• To compare two years, such as land use shares in 1950-51 and 1998-2001, draw two separate pies.

• Histogram: bars standing next to each other with no gaps, because the class intervals are continuous (e.g. 0-10, 10-20).

• Frequency polygon: made by joining the mid-points of the tops of the histogram bars.
• Ogive: a cumulative frequency curve.
• The "less than" ogive rises. The "more than" ogive falls.

• Where the two curves cross is the median (the middle value).

• Line and bar graph / climograph:

• The line and bar graph puts monthly temperature (line) and monthly rainfall (bars) on one chart for one station, such as Delhi.

• A climograph plots temperature against humidity or rainfall for each month. This gives the station's climate "shape" at a glance.

• Wind rose: lines or bars drawn out from a centre point in 8 or 16 compass directions.

• Each bar's length = how often the wind blows from that direction.

• The centre circle shows calm days.

• Age-sex pyramid: horizontal bars for 5-year age groups. Males are usually on the left, females on the right.

• Broad base → many births → growing population (e.g. Bihar, Uttar Pradesh).

• Narrow base → few births → ageing population (e.g. Kerala).

• Lorenz curve: shows cumulative % of population against cumulative % of income (or land).

• The straight diagonal is the line of equality.
• The more the curve bends away from it, the greater the inequality.

• This gap is measured by the Gini coefficient (0 = perfect equality, 1 = total inequality).

• Ternary diagram (triangular graph): an equilateral triangle. Each side is a 0-100% scale for one of three parts that add up to 100%.

• Examples: sand-silt-clay in soil texture; primary-secondary-tertiary shares of the workforce.

4. Maps: a key to match data with method

Data type	Best map	Example
Absolute counts spread over an area	Dot map	Population distribution of India
Absolute values at points	Proportional symbol map	Populations of million-plus cities; capacity of power plants
Ratios per unit area or population	Choropleth	Population density, literacy rate, share of rice output by state
Ratios refined by land use	Dasymetric map	Density shown only on lived-in land, leaving out forest and desert
Continuous surface (a "field")	Isopleth	Temperature, pressure, rainfall, relief
Movement between places	Flow map	Trains, goods traffic, migration
Size-based comparison	Cartogram	Countries resized by population

5. Key map terms explained

• Dot map: each dot stands for a fixed number (e.g. 1 dot = 10,000 persons). Dots are placed where the people or things actually are.
• Needs: an administrative base map, the data, and a choice of dot value and dot size.
• Result: dense clusters (Ganga plain) and empty patches (Thar Desert, high Himalaya) show up clearly.

• Best for absolute numbers. Dot density maps show counts at a stated dot-to-count ratio. [5]

• Proportional symbol map: circles, squares or spheres are drawn at points. Each symbol's size is in proportion to the value at that point.

• These maps show raw totals (absolute size). Choropleths show rates. [5]

• Choropleth map: each unit (state, district) is shaded by its value. Values are grouped into classes, shown in shades that go from light (low) to dark (high).

• NCERT steps: arrange the data in ascending order → divide it into about 5 classes → give each class a shade → shade each unit.
• Rule: use ratios or rates (per sq km, per 1,000, %), not raw totals. Choropleths are meant for rates or densities, and normalising the data stops the map from misleading. [5]

• Weakness: the whole unit gets one shade, even if most of its people live in one corner.

• Dasymetric map: an improved choropleth.

• Cause: administrative boundaries do not match where people actually live.
• Process: other data, such as land use or land cover, is used to remove empty land (forests, deserts, water bodies). The value is then shown only on the land where it really occurs.
• Result: the boundaries follow the real spread of the data, not administrative lines. [5]

• Where it helps in India: districts with large forest or desert areas, such as in Rajasthan, Arunachal Pradesh and Madhya Pradesh.

• Isopleth map: an isopleth is a line on a map or graph that joins points with the same value of a variable. [4]

• Method: values are known only at stations, so the values in between are estimated. This is called interpolation.
• Common isopleths:

Isopleth	Joins places of equal...
Isotherm	temperature
Isobar	air pressure
Isohyet	rainfall
Isohaline	salinity of sea water
Contour (isohypse)	height above sea level
Isobath	depth of water

• Flow map: lines or arrows show movement from one place to another. The thickness of each line is in proportion to the volume that moves.
• NCERT calls it a "dynamic map", because it shows movement rather than a fixed pattern.
• NCERT example: the number of trains running on routes out of Delhi.

• Other uses: the flow of goods, passengers or migrants.

• Cartogram (value-by-area map): the map is stretched so that each region's size is in proportion to its data value, not its land area. [5]

• Example: in a population cartogram, densely populated countries such as India and Bangladesh swell, while large, thinly populated countries such as Australia and Canada shrink.

6. NCERT exercises show the pattern

Exercise item	Correct choice	Logic
Population distribution	Dot map	Absolute count spread over space
Decadal growth of population	Line graph	One variable over time
Change in several variables over time	Polygraph	Several time series
"Dynamic map"	Flow map	Shows movement
Land use shares, 1950-51 vs 1998-2001	Two pie diagrams	Shares of one total, one pie for each year
Rice area by state (several years)	Multiple bar	Several variables per item
Share of rice output by state	Choropleth	A ratio (% share) shaded by unit

7. Present-day practice: GIS and spreadsheets

• Today, diagrams are made in spreadsheet software and maps in GIS (Geographic Information System: software that stores, analyses and displays data linked to locations). See the remote-sensing-gis note.
• The logic of choice is the same. Software only draws faster. A choropleth of raw totals is still wrong, even when GIS draws it.
• India's official source of thematic map layers is Bhuvan, the geoportal of NRSC (National Remote Sensing Centre, ISRO, Hyderabad):
• Land Use Land Cover (LULC) maps are available at 1:250K, 1:50K and 1:10K scales. [2]
• Other layers include Carto DEM, geomorphology, lineament, flood hazard zonation, flood annual layers, erosion, salt-affected and waterlogged areas, the Water Bodies Information System and urban sprawl. [2]
• Bhuvan also offers wasteland, urban land use, and glacial lakes and water bodies of the Indian Himalayan region. [3]

• Users can add these maps into their own systems as OGC Web Services (standard web formats that GIS software can read). [2]

• How the choice works on Bhuvan data:

• LULC classes are categories (forest, crop, built-up). They are shown as qualitative colour maps, not light-to-dark choropleths.
• LULC can serve as the ancillary layer (the extra data layer) for a dasymetric population map.
• Carto DEM (Digital Elevation Model: a grid of heights) gives contours, which are isopleths of height.
• Flood annual layers in two different years can be compared side by side, the same way as two pies for two years.

Prelims Hooks

• "Dynamic map" (NCERT) = flow map. Line thickness is in proportion to the volume moving.
• Population distribution → dot map. Population density → choropleth. This pair is a classic trap.
• Decadal growth of population (one variable) → line graph. Several variables over time → polygraph.
• Pie diagram: angle of a part = (value ÷ total) × 360°. To compare two years, use two pies.
• Isohyet = equal rainfall. Isohaline = equal salinity. Isobath = equal depth. Isobar = equal pressure. Isotherm = equal temperature.
• Choropleth shading goes from light (low) to dark (high) and needs ratios, not totals. A dasymetric map refines it using land-use data. [5]
• Ogive: "less than" and "more than" curves cross at the median. Histogram bars have no gaps.
• Lorenz curve → inequality (Gini). Ternary diagram → three parts adding up to 100% (e.g. soil texture). Wind rose → frequency of wind direction.
• Cartogram = value-by-area map, so region size is in proportion to the data value. [5]
• Bhuvan (NRSC, ISRO) gives LULC at 1:250K, 1:50K and 1:10K scales, as OGC web services. [2]

Mains Points

• Why the choice of method matters for policy (GS-I):
• A choropleth of raw totals hides real need: large districts look "high" just because they are big.
• Density, sex ratio and literacy must be mapped as ratios.

• Dasymetric refinement shows where people actually live (Ganga plain vs Thar, forested Northeast). This leads to better targeting of services. [5]

• Matching the tool to the theme helps in planning:

• Isopleths (isohyets, isotherms) → agro-climatic zoning.
• Flow maps → transport and migration corridors.
• Age-sex pyramids → demographic dividend vs ageing (Bihar vs Kerala).

• Lorenz curves → land and income inequality.

• GIS has made thematic mapping routine but not automatic (GS-III):

• NRSC's Bhuvan layers (LULC, flood hazard zonation, erosion, waterlogging, glacial lakes) feed disaster management and land-resource planning. [2][3]

• The planner must still choose the right method, or the result can mislead.

• For answer writing:

• A neat dot map, choropleth or flow diagram sketched in a GS-I answer (population, migration, crops) shows spatial thinking.
• The rule to follow: counts → dots or symbols; ratios → choropleth; fields → isopleths; movement → flows.