Social Mobility Table
- Introduced by
- Pitirim Sorokin; refined by Hauser, Hout, Featherman
- Year
- 1927 (concept); 1970s–1980s (modern analysis)
- Type
- Cross-classification of social origins by destinations
- Family
- Social stratification & mobility
What does it do?
A social mobility table is a cross-classification of individuals by their social origin (typically a parent's class or occupation) and their own destination class, forming the empirical foundation of intergenerational mobility research.
Analyzing it separates how much people move between classes, distinguishes movement forced by changing class sizes from genuine exchange, and isolates the underlying origin–destination association that measures the openness of a society.
The intuition behind the method
Lay all parent–child pairs in a grid whose rows are the parent's class and columns are the child's class. The diagonal counts those who stayed put; the off-diagonal counts those who moved up or down. But a raw count of movers is misleading: if the class structure itself shifted — say manual jobs shrank and professional jobs grew — some movement was simply forced by the changing number of slots, not by any loosening of class barriers. The mobility table lets us peel that 'structural' movement away from the 'exchange' between classes, and to focus on odds ratios that capture how strongly origins still bind destinations regardless of marginal change.
History
Pitirim Sorokin's 1927 Social Mobility established the field, and David Glass's 1954 British study popularized the formal mobility table. The decisive methodological advances came in the 1970s and 1980s, when Robert Hauser, David Featherman, Michael Hout, and John Goldthorpe brought log-linear and association modeling to mobility tables, separating marginal change from the structural association and enabling rigorous cross-national comparison — most prominently in Erikson and Goldthorpe's comparative project on industrial societies.
When to use it
Use mobility-table analysis when you have categorical origin and destination data (class, occupational grouping, education) and want to characterize intergenerational movement, compare openness across countries or cohorts, or test specific hypotheses about which origin–destination associations are strong. It is the natural framework when both variables are categorical and the question is about relative rather than absolute positions. It is not appropriate when origin or destination is continuous (use intergenerational elasticity), when cells are too sparse for stable estimates, or when the fine structure of the association requires the formal log-linear and association models that build on the table rather than the descriptive indices alone.
Limitations and considerations
- Requires categorical class schemes whose boundaries are debated and affect every conclusion.
- Descriptive indices like total mobility confound structural and exchange components unless explicitly decomposed.
- Sparse cells in detailed class schemes destabilize estimates and odds ratios.
- A two-way origin–destination table omits mediators such as education unless extended to higher-dimensional tables.
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Provides the standard, internationally comparable framework for categorical intergenerational mobility.
- Distinguishes structural (marginal-driven) from exchange (circulation) mobility, clarifying what 'more mobility' means.
- Odds-ratio focus yields measures of relative mobility that are invariant to changing class sizes.
- Serves as the data structure on which powerful log-linear and association models are built.
Applications
- Measuring intergenerational class mobility within a country and tracking it across birth cohorts.
- Cross-national comparison of social fluidity and openness.
- Studying occupational inheritance and the strength of class reproduction.
- Analyzing educational mobility between parents' and children's attainment categories.
- Examining marriage and assortative-mating tables using the same origin–destination logic.
How it works
How it works
The table is built by counting individuals in each origin-by-destination cell n_ij. Rows index class of origin, columns index class of destination, and the cell entries are frequencies. The marginal totals give the origin and destination class distributions for the two generations.
Formula
n_{ij} = \#\{\text{individuals with origin } i,\ \text{destination } j\}Two conditional views interpret the table. Outflow percentages divide each cell by its row total to ask 'of those from origin i, what share reached destination j?'. Inflow percentages divide by the column total to ask 'of those now in destination j, what share came from origin i?'. They answer different mobility questions.
Formula
o_{ij} = \frac{n_{ij}}{n_{i\cdot}},\qquad f_{ij} = \frac{n_{ij}}{n_{\cdot j}}Simple summary indices include total mobility — the proportion of cases off the main diagonal — and its complement, the proportion immobile. These describe gross movement but conflate movement caused by structural change with movement reflecting genuine fluidity.
Formula
\text{total mobility} = 1 - \frac{\sum_i n_{ii}}{n_{\cdot\cdot}}Total observed mobility is decomposed into structural (or forced) mobility, which is the movement compelled by differences between the origin and destination marginal distributions, and exchange (or circulation) mobility, the reciprocal movement that would occur even if the class structure were unchanged. Sociologists generally treat exchange mobility as the better indicator of social openness.
Formula
\text{total} = \text{structural (forced)} + \text{exchange (circulation)}The substantively central quantities are the odds ratios formed from cell counts, which measure the origin–destination association net of the marginal distributions. Because odds ratios are invariant to the changing sizes of classes, they capture relative mobility — how tightly origins constrain destinations — and are the focus of formal log-linear and association models fitted to the table.
Formula
\text{analyze } \frac{n_{ij} n_{i'j'}}{n_{ij'} n_{i'j}}\ \text{(odds ratios = relative mobility)}Common pitfalls
- Comparing total (absolute) mobility rates across countries or eras without recognizing that differing marginal distributions drive much of the difference.
- Treating outflow and inflow percentages as interchangeable when they answer opposite questions.
- Reading the diagonal as 'no mobility' while ignoring that broad class categories hide within-category movement.
- Drawing conclusions about social openness from raw movement counts rather than from marginal-free odds ratios.
Frequently asked questions
What is the difference between absolute and relative mobility?
Absolute mobility is the observed amount of movement between classes, including movement forced by changes in the sizes of classes across generations (structural mobility). Relative mobility, captured by odds ratios, describes the competition between people of different origins for the same destinations independent of how class sizes changed. Two societies can have very different absolute mobility yet identical relative mobility, which is why sociologists emphasize the latter when assessing openness.
Why are odds ratios central to mobility analysis?
Odds ratios formed from the table's cells are invariant to the marginal (origin and destination) distributions. This means they isolate the genuine origin–destination association — how strongly where you start binds where you end up — from the mechanical effects of a changing class structure. Log-linear and association models parameterize exactly these odds ratios.
How is structural mobility distinguished from exchange mobility?
Structural (forced) mobility is the minimum movement required by the difference between the origin and destination marginal distributions: if the destination structure has more high-status slots than the origin structure, some upward movement is unavoidable. Exchange (circulation) mobility is the additional, reciprocal movement that would occur even if the marginals were identical, and it reflects fluidity rather than structural change.
Related methods
Sources
Hauser, R. M. (1978). A structural model of the mobility table. Social Forces, 56(3), 919–953.
Open DOI referenceCopied.Hout, M. (1983). Mobility Tables. Sage University Paper Series on Quantitative Applications in the Social Sciences, 07-031. Sage.
ISBN: 978-0-8039-2056-9Copied.