The core idea
Migration concerns a change of residence under a stated definition; urbanisation concerns the increasing urban share of a population. Stocks, flows, reasons and geographic classifications answer different questions. A sound explanation states its period, population, boundaries and evidence before assigning causes.
1. Define residence before counting movement
A journey is not automatically migration. A daily commuter can cross a district boundary while keeping the same residence; someone changing residence within a state may count as a migrant under an appropriate definition. Internal migration stays within a country's borders; international migration crosses them. Surveys also distinguish residence from birthplace and specify duration rules. In the PLFS migration material for July 2020–June 2021, usual residence involved staying, or intending to stay, continuously for at least six months; a migrant's last usual residence differed from the place of enumeration. This is a dated survey definition, not a claim that every institution uses the same threshold. The 2011 Census D-03 table organises evidence by last residence, duration and reason. Choose the measure that fits the question rather than treating all movement counts as interchangeable.
Sources: MoSPI/PIB: Migration in India, 2020–2021 (14 June 2022) ↗ · Census of India 2011: D-03 migration table metadata ↗
2. A stock and a flow tell different stories
A stock counts people with a specified characteristic at a time, such as residents whose last usual residence was elsewhere. A flow counts movements during a period, such as arrivals this year. A long-settled person can belong to the stock without belonging to this year's inflow. Someone may move more than once, so movements and unique persons can also differ. Reasons require similar care. Work, education, marriage, household movement and other circumstances may overlap in a person's life even when a questionnaire records one main reason. A person who reports marriage as the reason for moving may also work after arrival. The recorded reason does not establish every motive, later activity or degree of choice. Read the question wording and categories before making a broad social claim.
Sources: Census of India 2011: D-03 migration table metadata ↗ · MoSPI/PIB: Migration in India, 2020–2021 (14 June 2022) ↗
3. Account for urban growth before explaining it
Urban population can grow through natural increase, net migration and reclassification. Natural increase is births minus deaths; net migration is arrivals minus departures. Reclassification can add a settlement to the urban category without its residents moving. Boundary changes must also be handled consistently. For a fixed town boundary, a simple model starts with 1,000 residents, adds 30 births and 80 arrivals, and subtracts 10 deaths and 20 departures: the ending population is 1,080. This is bookkeeping under complete counts, not a theory of why the changes occurred. Urbanisation usually refers to an increasing urban proportion, while urban growth refers to an increase in urban numbers. A country's urban population can grow while its urban share stays constant if the total population grows proportionately. Always specify the denominator.
Sources: UN DESA: World Urbanization Prospects 2018, Highlights, Box 2 ↗
4. Worked case: a small net change can hide large movement
In a fictional town over one year, there are 80 arrivals and 20 departures across the same boundary. Net migration is 80 − 20 = 60. Gross movement in this simplified record is 80 + 20 = 100 recorded moves. The numbers are not competing answers: they answer different questions. Net change helps population accounting, while the separate flows may matter for housing turnover, school places and transport. We have not established that 100 distinct people moved, because a person could appear in both flows. Nor can we infer that 60 people obtained jobs or that every arrival came from a village. To investigate those questions, we need origin, timing and activity information. The arithmetic identifies what the count means; it cannot supply missing social characteristics.
Sources: Census of India 2011: D-03 migration table metadata ↗ · UN DESA: World Urbanization Prospects 2018, Highlights, Box 2 ↗
5. Worked case: a changing percentage without a move
An invented region has 10,000 residents, of whom 4,000 are classified as urban. Its urban share is 4,000 ÷ 10,000 = 40%. A settlement containing 500 residents is then reclassified from rural to urban, with no births, deaths or movements in this exercise. The urban count becomes 4,500 and the share becomes 45%. The change is 5 percentage points, not a 5% relative increase: relative to 40%, the increase is 12.5%. None of those 500 residents migrated in the model. Therefore a graph of urban share alone cannot tell us how many people moved to cities. When comparing years, inspect definitions and geographic units as carefully as the figures. A classification is useful for administration and research, but it is not identical to every resident's daily experience of a place.
A share can change without anyone moving
| Population | Before | After reclassification |
|---|---|---|
| Urban | 4,000 | 4,500 |
| Rural | 6,000 | 5,500 |
| Total | 10,000 | 10,000 |
| Urban share | 40% | 45% |
45% − 40% = 5 percentage points
Sources: UN DESA: World Urbanization Prospects 2018, Highlights, Box 2 ↗
6. Connect movement with opportunities and constraints
People can move toward education, employment or family, away from insecurity or livelihood pressures, or through combinations of these conditions. A simple push-and-pull description is a starting point, not a complete causal explanation. Travel costs, contacts, housing, care responsibilities and information affect whether an opportunity is reachable. Migration can connect origin and destination through money, visits, language and care; it need not sever one place completely from the other. Urban outcomes also depend on transport, land use, services and decisions about investment. A rise in rents alongside arrivals does not by itself prove that migrants caused the rise: supply, redevelopment and other demand may change simultaneously. Compare plausible mechanisms and evidence. Avoid describing migrants as a single homogeneous problem or assuming that every move produces the same gain or loss.
Sources: OpenStax, Introduction to Sociology 3e: Urbanization ↗ · Census of India 2011: D-03 migration table metadata ↗
7. Read surveys as designed observations
A census table, a sample survey and an interview have different strengths. The 2011 Census table cited here offers a dated classification of migration evidence. The 2020–2021 PLFS material added migration questions to a labour-force survey; the official release explicitly notes that its sample design was not specifically focused on migration and temporary visitors. These sources should not be presented as current counts or interchangeable measurements. Examine coverage, question wording, weighting, fieldwork conditions and the population a percentage describes. Interviews can reveal how a move was experienced but cannot alone estimate its national frequency. Quantitative patterns can show a distribution without explaining every individual's decision. Combining methods is useful when each addresses a clearly stated question, not when several sources merely repeat the same attractive claim.
Sources: MoSPI/PIB: Migration in India, 2020–2021 (14 June 2022) ↗ · Census of India 2011: D-03 migration table metadata ↗
PUT IT INTO PRACTICE
Practice: separate three measures of change
- For 80 arrivals and 20 departures, calculate net and gross movement. Answer: 60 net and 100 recorded moves; unique persons are not established.
- Compute the fixed-boundary population model. Answer: 1,000 + 30 − 10 + 80 − 20 = 1,080; births and deaths contribute 20, migration contributes 60.
- Calculate the reclassification change. Answer: 40% to 45%, or 5 percentage points; no migration occurs in the stated model.
- Repair “the survey proves all migrants move for jobs.” Answer: the sources record several reasons, and a recorded main reason does not exhaust a person’s circumstances.
Check your understanding
Why can a daily commuter fall outside a residence-based migration count?
The journey may leave the person’s usual residence unchanged; the definition counts a different event.
Can a long-settled resident belong to a migrant stock?
Yes, if the definition refers to a previous residence, even though the person did not arrive in the current period.
What does net migration omit?
It does not separately show the scale of arrivals and departures, or identify each person’s characteristics.
Why is 40% to 45% not a 5% relative rise?
The difference is 5 percentage points; dividing 5 by the original 40 gives a 12.5% relative rise.
Can urban growth happen without migration?
Yes. Natural increase or reclassification can raise the urban count.
Why date a migration statistic?
The period, definitions and fieldwork conditions matter; an older measurement is not automatically a current description.