Beyond longitudinal data: Causes, consequences, changes, and continuity.

Abstract

Longitudinal data have very important advantages for both measurement and the testing of causal hypotheses on the causes or course of psychopathology, but cross-sectional studies should usually be used first. The investigation of causes needs to encompass the several different types of causal question. The study of within-individual change constitutes a most important research strategy to test causal hypotheses, but it is not the only approach. Their testing requires specification of possible mechanisms, together with attention to the differential impact of risk experiences, of the possible role of person-environment interactions and protective mechanisms. This article addresses several strategies needed in using longitudinal data to test cause-and-effect relationships, including natural experiments, testing of competing hypotheses on mechanisms, study of reversal effects, multiple replications in different circumstances, use of designs to dissociate possible mechanisms, testing for dose-response relationships, examination of effect-specificity, considering biological plausibility, and assessing the strength of effects.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 153 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.