English · Русский · Українська

In 90 seconds

These terms are easy to read in more than one way. Active pages should qualify the intended meaning at first use.

Model, function, aspect, and position

Model may mean a formal representation, an internal information model, Socionics Model A, or a statistical model. Name which one.

Function may mean a proposed psychic function, mathematical mapping, or software operation. Name the discipline.

Aspect is a system-specific dimension in a source typology. In the BWB Socionics reconstruction it is a proposed content-operation pair; aspects with similar names across systems are not automatically equivalent.

Position is a location in a type structure. In Model A it is kept distinct from the aspect and interpreted as a proposed use mode. Any cognitive interpretation assigned by BWB must be labelled a project hypothesis rather than a canonical alias.

Example of disambiguation

“Future predicts action” is ambiguous. It might refer to a Temporistics aspect, a future-time score, a scenario variable, or ordinary expectations. A research sentence must name the construct, instrument, population, and outcome.

Construct and measurement

Construct — a deliberately defined concept intended for description or measurement.

Operationalization — rules connecting a construct to tasks, items, indicators, coding, or outcomes.

Scale score — an instrument output calculated according to a declared scoring key; it is not automatically the construct itself.

Measurement error — variation due to the instrument or procedure rather than the intended construct.

Reliability and validity

Reliability — consistency under specified conditions; it does not establish that the intended construct is measured.

Content validity — whether items adequately represent the construct domain.

Convergent validity — expected association with relevant measures.

Discriminant validity — empirical distinction from rival constructs.

Predictive validity — performance on a prespecified future outcome.

Model alternatives and how to compare them

Before preferring a typological explanation, compare it with role, learned skill, language, culture, health, stress, response style, and a simpler established construct. The following terms describe tests that can distinguish competing models:

Measurement invariance — evidence that measurement behaves comparably across groups or occasions.

Calibration — agreement between predicted uncertainty or probability and observed frequency; discrimination alone is insufficient.

Incremental validity — improvement beyond a simpler or already established model.

Replication — a new test of a claim under declared conditions, not reuse of the development data.

Inference and uncertainty

Correlation — association; not by itself causation.

Causal mechanism — a claim about how change is produced; stronger than a latent-process story.

Uncertainty — limits arising from sampling, measurement, missingness, model choice, and unknown conditions.

Falsification condition — an observation or pattern declared in advance as challenging a hypothesis.

Source-and-norm vocabulary and practical questions

Source tradition — the original and later literature of a typological system; it may contain disagreement.

Raw source — preserved input material, not an endorsed BWB conclusion.

Source attribution — a claim about what a source says, distinct from project acceptance.

Normative source — an authority or framework used to determine what ought to govern an application.

  • What exactly does the source claim, and what evidence does it report?
  • Is the current sentence source attribution or BWB acceptance?
  • Which rival construct or contextual explanation has been tested?
  • Which explicit normative source governs what ought to be done in this application?

Research route for pair-level claims

Individual-level variable describes one person. Pair-level variable describes a relation, interaction, agreement, complementarity, or joint outcome. The former cannot be relabelled as the latter without a defined composition model.

Resource/friction profile is a qualitative map of possibilities to investigate. Decision support organizes evidence for human judgement; it is not automated decision replacement.

Use Compatibility Measurement Methods for measurement design, Validation Program for staged validation, and Epistemic Status and Inference Limits for claim classification.

Next reading

See Compatibility Measurement Methods, Latent Process, and Core Glossary.