Intention-to-Migrate Prediction

Because intention is still one of the best predictors available: people who state plans migrate at much higher rates than those who do not, even if the relationship is far from certain. The point is not to treat intention as destiny but to use it as a calibrated signal whose error rate you measure. By quantifying how often intenders fail to move and adjusting for capability, the model extracts the genuine predictive content of intentions while being honest about their limits, which is far more useful than either ignoring intentions or naively believing them.

The intention-behavior gap is the discrepancy between what people say they will do and what they actually do. It has two sides: intenders who do not migrate (frustrated plans, central to Carling's involuntary immobility) and non-intenders who do (unplanned moves). It is measured with longitudinal data by computing the conditional probabilities of moving given prior intention, ideally as a full confusion matrix of predicted versus realized migration. Reporting both error directions characterizes the gap and reveals where intentions break down, which is itself a substantive finding about constraints and opportunities.

Because intention without the means to act often fails to produce migration. De Haas's framework holds that behavior results from the interaction of aspiration or intention with capability, so an intention backed by money, networks, and legal access is far more likely to be realized than the same intention without them. Including a capability measure and its interaction with intention lets the model distinguish realizable from frustrated plans, which typically improves predictive accuracy and yields interpretable insight into how constraints filter intentions into actual moves.