How test batteries work

The catalog has 26 batteries, each built for a research area such as sleep and fatigue, ADHD, or decisions and reward. Each one says what it measures and how long it takes per participant, and its devices and languages come from its tests.

A battery runs on a phone or a tablet only when every test in it does. The app checks each participant's device before the first test.

A study started from a battery runs the battery's tests in the battery's order, with standard settings. The study records the battery's version, a number that changes whenever the tests, their order or their settings change.

Standard settings matter for percentiles. A test compared with a published norm has to run the way the norm's study ran it, and changing a setting keeps the scores but removes the percentile.

Adding repeat visits keeps the battery.

Changing a test, its order or its settings makes the study a custom study, and so does adding a questionnaire. The builder asks before it makes the change.

If you go ahead, the study no longer carries the battery and its participant battery reports go away. Results already collected keep their scores.

A battery study adds two reports: one for each participant, with a row per test on a 0 to 100 percentile scale, and one for the whole study. Each test keeps its own reference group, so the participant report has no total or average.

Every battery has a guide: what it measures, its tests in order, time and devices, languages, how to give it, how each test is scored, the session quality rules, the published norms it uses and how to cite it. Print it or save it as a PDF.

Each battery takes the grade most of its tests reach, with a minus when some tests are lower. The Data page of every battery's grade puts them in one table, grade A first, then A−, B and on down, with what each test below grade A has.