Methodology

Myndora is built around the idea that personality is best understood through repeated measurement over time, rather than a single snapshot.

We treat personality as layered. Each layer describes a different part of how you tend to operate:

Behavior layer (Big Five) — your broad behavioral tendencies across situations.
Thinking style layer (16 Types) — how you process information and make decisions.
Motivation and stress layer (Enneagram) — what drives you and what creates pressure.

The app currently uses three established personality frameworks:

Big Five — to capture broad behavioral tendencies.
16 Types — to describe patterns in information processing and decision framing.
Enneagram — to explore motivational drivers and stress-related patterns.

For each framework, Myndora stores results across multiple test sessions. Over time, these repeated measurements are used to estimate a baseline profile and to calculate a stability indicator.

The baseline reflects the most frequently observed outcomes across sessions. Stability reflects how consistent those outcomes are over time. As more data points are collected, both estimates become more reliable.

Interpretations shown in the app are derived directly from stored results and predefined mapping rules. No manual interpretation or human review is involved.

Frameworks and sources

The following references inform our framework selection, question design, and mapping rules. This is a short, non-exhaustive list:

Big Five: Costa & McCrae (NEO-PI/NEO-PI-R), Goldberg (lexical Big Five), John & Srivastava (Big Five overview), and common trait/facet conventions.
16 Types: classic type-dynamics literature (including Gifts Differing) and related 16 Types dimension research.
Enneagram: Riso & Hudson, Palmer, and Naranjo (core type patterns and motivation/stress dynamics).

How tests are built and scored

Each test is built from structured question sets that map directly to the traits or dimensions of the framework:

Big Five: the current Big Five flow uses two lighter variants, Test A and Test B. Each variant covers the same 30 facets with one question per facet, and both variants contribute to the same longitudinal Big Five history. Facet scores are summed and then combined into trait scores, which are binned as Low, Mid, or High.
16 Types: each question contributes a signed score to one of four dimensions (E/I, S/N, T/F, J/P). Dimension totals are converted into percent bars and letter outputs.
Enneagram: answers contribute to nine type totals. The highest total is the provisional core type; small margins indicate uncertainty.

The goal is consistency and transparency. We avoid complex or opaque transformations so the results can be traced back to the questions you answered.

How results are stored

When you are logged in, your results are saved to your Myndora account so we can build a baseline and show stability over time. When you are not logged in, results are stored locally in your browser and can be claimed later after you create an account.

For details about data storage and retention, see the Privacy page.

Baseline and stability

Your baseline is the most frequently observed outcome for each trait or dimension across repeated sessions. Stability reflects how strong that baseline is relative to competing outcomes.

We compute a margin between the most common outcome and the next most common. Larger margins mean steadier results; smaller margins mean more variability or uncertainty. As you complete more sessions, these margins tend to stabilize.

How layers combine in Environment Fit

The Environment Fit page combines all three layers to compare how they point in the same direction (or not) for common situations. Each layer produces a simple signal (supports, neutral, or drains) and those signals are combined into an overall fit score.

This allows you to see where your behavior, thinking style, and motivation agree or conflict, which is often more useful than any single test in isolation.

Limitations

Myndora relies on self-reported answers. This means results can be influenced by temporary context, mood, or how a person interprets a question at a given moment.
The models used describe general tendencies rather than fixed traits, abilities, or exact predictions. They cannot capture the full complexity of a person or explain behavior in every situation.
Myndora does not validate results against external behavior, performance, or outcomes. It does not observe what people actually do outside the app.

To account for these limitations, Myndora focuses on repeated measurement over time rather than single results, applies consistent scoring rules, and explicitly surfaces uncertainty when results vary. This reduces overinterpretation of momentary states and makes instability visible instead of hidden.

The methodology is designed to support reflection and pattern recognition over time, not prediction, diagnosis, or behavioral validation.