Scope of this review
This is a black-box review of the public interface. It does not describe source code, a private scoring algorithm, or a verified psychometric model. The visible page presents 48 statements on a five-point scale, grouped by axis, and describes a 12-axis result. It also says that answers are calculated in the browser. These are public interface claims, not independently audited implementation facts.
That distinction matters for political self-assessment. A polished chart can look more authoritative than the questions behind it. The interface should make its model and limits easy to inspect.
Keep three states visible
A useful questionnaire UI needs to communicate at least three different things:
Answer state: which items the person has answered, skipped, or changed.
Interpretation state: how answers relate to each displayed axis.
Context state: which parts of the result depend on a country-specific reference.
If the interface compresses all three into one label, users cannot tell whether a result reflects their answers, the test’s categories, or a localized comparison. A map with separate axes is easier to question than a single unexplained badge.
Treat political labels as interpretations
The page includes comparisons with PSOE, PP, VOX, and Sumar, described as editorial references for Spain. A test result should distinguish this kind of comparison from a poll, endorsement, or recommendation. The UI should also make it possible to read the underlying positions before displaying a party match.
For a design review, ask whether users can answer these questions without leaving the result screen: What does each axis mean? Which end of the scale is represented? Is a percentage a similarity score or a vote prediction? Which country and reference date does a party comparison assume?
A compact acceptance matrix
Area
Example check
Expected behavior
Completion
One answer is skipped
The result does not silently present an incomplete profile as complete
Revision
An answer changes
The displayed axes update or explain when recalculation occurs
Interpretation
A party match is visible
The page labels it as an editorial comparison, not advice
Context
A reader is outside Spain
The geographic scope remains clear
Privacy
The page claims local calculation
The actual network behavior should be checked separately
The last row is intentionally a test question. A visible privacy statement is useful, but an interface inspection alone cannot prove network behavior. If privacy is a release requirement, verify it with a network-level test and document the result.
A useful public example
12ejes organizes its explanation around 12 axes, including economy, rights, borders, ecology, taxation, Europe, work, digital policy, housing, and security. That breadth can help readers notice that two positions may point in different directions.
The Test Ideologia Politica also illustrates a key UX boundary: its party comparison is explicitly framed as indicative, and the page tells readers to consult official proposals before deciding. That boundary belongs close to the result, where people are most likely to overread a score.
Questions to carry into implementation
Before building a similar product, define how incomplete answers work, how uncertainty is communicated, how country references are updated, and how a user can inspect the scoring model. Then test the language with people who hold mixed views, not only with people who already identify with a familiar label.
The interface is not neutral merely because it uses numbers. Its question wording, axes, examples, and defaults all shape what users see. Make those choices inspectable, and treat every result as a conversation starter rather than an identity assignment.
Top comments (0)