Skip to content

What tagging is actually for

Tags describe. Banks assemble.

StudyDrome Exam Manager classifies items on two axes. Tags nest under named dimensions you define (Topic, Clinical Domain, Bloom's Level), and a question can carry tags from several at once. Clinical skills sit alongside them as a flat library. Three things read that classification: search, blueprint coverage, and must-pass domains.

Book a pilot

What is a tag type?

A tag type is a dimension. You name it, and the tags inside it belong to it.

Tag type

Tags inside it might be

Topic

Cardiology, Respiratory, Renal

Clinical Domain

Diagnosis, Management, Communication

Bloom's Level

Recall, Application, Analysis

The dimensions are yours to define. Each carries a color and a position in the list, so the picker reads the way your program already thinks.

A question can carry tags from several dimensions at once. That is the part that pays off later. Two dimensions of one item are what let a coverage matrix place Topic on one axis and Clinical Domain on the other.

Some tag types are owned by the system. Those cannot be renamed or deleted, so a dimension the platform depends on stays where it is.

How deep can a tag go?

Tags nest. A tag can sit under a parent, so "Arrhythmia" lives under "Cardiology" and takes its place in the tree rather than repeating it in its own name.

A tag with no parent is a root tag. There is no separate top-level object to maintain. The tree is just tags, some of which have parents.

One view returns the whole hierarchy at once, so you read the tree rather than click down through it.

Tags carry the same protection as their types. A system tag cannot be renamed or deleted either.

Where do clinical skills fit?

Clinical skills are a second axis, and they are deliberately flat.

The library belongs to your account. Each skill has a name and an optional code. Questions join to skills many-to-many, so one station can carry several skills, and one skill can run across many stations. You manage the library under "Clinical Skills".

Flat is the point. A skill says what a station asks the candidate to do. Depth belongs to the tag tree, where you already have it, and keeping the two apart stops one taxonomy from being asked to be both.

Medical schools tend to run both at once: tags for content, skills for performance.

What actually reads a tag?

Tagging is only worth the effort if something reads it. Three things do.

Finding items. The bank list filters on tags, so "every Cardiology item at Application level" is a filter rather than a hunt through folders.

Coverage. The blueprint matrix maps each axis to a tag dimension or your Clinical Skills library, then reports what is in each cell. It reports the coverage. The judgment about whether that coverage is good enough stays yours.

Must-pass domains. A scoring policy can require a minimum in a named domain, and the domain is addressed by tag. That is the point where classification stops describing and starts affecting who passes.

There is one more thing worth knowing, because it decides how you tag. A paper is assembled from banks. You choose the banks a test draws on, and the tags describe what came out of them. So tag for retrieval and for coverage, which is what tags actually drive, and let the bank tree carry the structure. How to structure a question bank makes that case in full, and question banks cover the tree itself.

A case vignette is a third kind of classification, shared by many questions rather than describing one. The eleven question types cover how that works.

What does the classification model guarantee?

A taxonomy you cannot trust is worse than none, because people tag around it. These are the properties you can build a scheme on.

  • The dimensions are yours to name. Topic, Clinical Domain, Bloom's Level — whatever your program already says out loud.

  • One question carries tags from several dimensions at once, which is what makes a two-axis coverage matrix possible.

  • Tags nest, and one view returns the whole tree rather than a single level.

  • Clinical skills stay flat and separate, so neither taxonomy is asked to be both.

  • Skills join many-to-many, so a station can assess several, and a skill can run across many stations.

  • System-owned types and tags cannot be renamed or deleted, so the scheme survives a tidy-up.

  • A test draws on banks. Tags describe the items, and coverage is reported back to you.

Who needs item classification?

  • Assessment leads who have to show the board what an exam actually covered.

  • OSCE leads mapping stations to the skills each one assesses.

  • Programs running one bank across several years, where retrieval matters more than folders.

  • Anyone who has inherited a bank in which every item is tagged “General”.

Frequently asked questions

What is a question tag type?

A named dimension that groups tags — Topic, Clinical Domain, Bloom's Level, or whatever your program uses. You define the dimensions yourself, and each carries a color and a position in the list. A question can hold tags from several dimensions at once, which is what lets a coverage matrix run two axes over the same bank.

Can question tags be nested?

Yes. A tag can sit under a parent tag, so “Arrhythmia” lives under “Cardiology” instead of repeating it in its own name. A tag with no parent is a root tag, and a single view returns the entire hierarchy at once. There is no separate top-level object to maintain alongside the tree.

How are tags different from question banks?

A bank is where an item lives and who can reach it. A tag is what the item is about. A paper is assembled from banks, so the bank tree carries your structure, while tags carry retrieval and coverage. Keeping the two jobs apart is what stops a bank tree from growing a branch for every attribute.

How do clinical skills differ from tags?

Skills are a separate, flat library held per account, joined to questions many-to-many. They describe what a station asks a candidate to do, rather than what the item is about. Tags carry the depth, because they nest. Medical schools usually run both: tags for content, skills for performance.

What are question tags actually used for?

Three things: read them. The bank list filters on tags, so finding every item on a topic at a given level is a filter. The blueprint matrix sets its axes to a tag dimension or to Clinical Skills and reports coverage per cell. And a scoring policy can require a minimum in a domain, with the domain addressed by tag.

Book a pilot

Bring the tagging scheme you use today, however loose. We will build the dimensions, load a slice of your bank, and show you the resulting coverage matrix.

Book a pilot

Related