Measurement models | Reflective factor (CFA) | Bifactor | Second-order factor | Exploratory SEM (full cross-loadings) | Formative measurement and MIMIC | Item response models | Hierarchy, growth, and longitudinal models | Mixtures | Networks | General SEM assembler | Causal license: the badge

Scaling marks for latent variables | The counting rule | Label switching | Conditional independence and d-separation | A single summary

The grammar in one paragraph | Building a one-factor model by hand | From a lavaan model | From an lme4 formula | TikZ export | Model-data duality | Inspecting a diagram as data | Round-tripping to lavaan

0. The data format | 1. Simulate a study with self-censoring | 2. Declare a dynamic missingness graph and check recoverability | 3. Ask the data whether silence was informative | 4. Estimate, profile, calibrate | 5. Report | 6. Full-information maximum likelihood under missing at random | Where to go next

Why estimation alone cannot repair self-censoring | The running example | One tilt: fit_tilt() | The profile: tilt_profile() | Break-even values and identified sets: break_even() | Calibrating the sensitivity value: calibrate_delta() | From a passive sensor | From randomized probes | From the post-skip contrast | Burden together with self-censoring | Bootstrap intervals | Reporting: missingness_declaration() | Practical recommendations | References

The generating model | Every motif and its parameters | Design features | Reproducibility | Simulating from a fitted model | Planning a design: three small studies | How much does self-censoring bias the default analysis? | Which sensor is good enough to calibrate? | How many probes? | Comparing estimators across mechanisms | References

The question | The silence test | Reading the sign | Inference | Curvature | The M1 + M3 collider | The sensor-gap test | The fatigue check | How often do the tests reject? A small Monte Carlo | What to do with a rejection | References

What a dm-graph is | The seven motifs | Asking the graph questions: d-separation | The recoverability report | A table over the taxonomy | Observed contexts, sensors and probes | Choosing a declaration in practice | References

Measurement models | Reflective factor (CFA) | Bifactor | Second-order factor | Exploratory SEM (full cross-loadings) | Formative measurement and MIMIC | Item response models | Hierarchy, growth, and longitudinal models | Mixtures | Networks | General SEM assembler | Causal license: the badge

Scaling marks for latent variables | The counting rule | Label switching | Conditional independence and d-separation | A single summary

The grammar in one paragraph | Building a one-factor model by hand | From a lavaan model | From an lme4 formula | TikZ export | Model-data duality | Inspecting a diagram as data | Round-tripping to lavaan