Skip to main content

User Studies (Explore, Test)

We conduct different kinds of user-centered studies at different intervals:

  • Developmental feature tests are conducted in preparation for and following software releases in order to determine the technical adequacy of implemented features.
  • Internal user challenge workshops are conducted to assess the disciplinary suitability of LAUDATIO features and to brainstorm new feature ideas. These occur every 2-3 months.
  • External user workshops are conducted both as a means of literacy training and to gather feature requests from a multitude of disciplinary researchers.

Test workflow​

Most of the tests come in the form of guided challenges that take inspiration from heuristic walkthroughs (Sears, 1997).1 In them, we provide detailed steps to do one thing in LAUDATIO and have the participants assess the operability of LAUDATIO and optionally record their screens while doing it.

Task structure​

We provide test users with Markdwon test sheets that include general information on:

  • what is being tested
  • which minimal version of LAUDATIO can be tested with
  • and whether or not a user needs to be logged in to complete a task.

Each task is then defined as a set of steps to follow, acceptance criteria for the main outcomes of a task, and a freeform text field for test user feedback.

We optionally ask for so-called negative tests, where the intended absence of something is tested as well.

A minimal example is shown here (from our first test phase, April 20-30, 2026, task 6: 6_PLAINTEXT_SEARCH_DISTRIBUTIONS_IN_CORPUS.md):

## Task Description

This test serves to determine the adequacy of the document distribution visualization in combination with the plaintext search functionality (within a single corpus).

## LAUDATIO Version

1.1.0

## Login Status

You need to be logged in to do this test.

## Steps

1. Select an unpublished corpus of yours for which you extracted the plaintext (cf. `2_PLAINTEXT_EXTRACTION.md`).

2. Click on "Preview"

3. Go to "Documents"

4. On the right above the documents table, click on "Show distributions"

5. Try to roughly memorize (or take a screenshot) of how the distribution looks

6. Close the distribution window

7. On the right above the documents table, click on "Filter by plaintext"

8. Type a full word you know or expect to appear in only a subset of the documents and press enter to activate the search

9. If there are now less documents than before, click on "Show distributions" again

10. Are the distributions different than before (e.g. fewer columns)?

## Acceptance Criteria

* [ ] For a corpus, there is a button on the right above the documents table called "Show distributions"

- [ ] The distributions are shown

* [ ] The plaintext search changes the distributions

## Notes/Problems

Evaluation​

TBD

Conducted studies (sorted by date in descending order)​

May 6, 2026: Internal user challenge workshop​

  • 3 tasks
  • 10 participants

April 20-30, 2026: Developmental feature test​

  • 7 tasks
  • 6 participants
    • 5 who recorded screens while doing the tasks

March 4: Internal user challenge workshop​

  • 5 tasks
  • 6 participants (+ 1 who went through the tasks prior to the workshop)

Footnotes​

  1. Andrew Sears (1997). "Heuristic Walkthroughs: Finding the Problems Without the Noise". In: International Journal of Human–Computer Interaction 9, 3, 213-234. ↩