Category: Blog
-
Bolt Performance Calibration — When Students Choose Different Questions, the Same Mark Can Come From Different Test Forms
Optional exam questions can create hidden form differences when students choose different routes through the same paper. Bolt calibrates what equal marks can and cannot justify before comparing performance.
-
Student/Studying Interface Learning Manual: Graphing-Tool Interface | A Correct Equation Can Look Wrong When the Window Is Wrong
A student-facing Learning Manual for operating digital graphing tools: preserve the entered expression, inspect axes and window, verify scale and points of interest, distinguish graph state from mathematical meaning, and return the displayed evidence to the problem.
-
Student/Studying Interface Learning Manual: Audio-Description Interface | Hearing the Dialogue Is Not the Same as Receiving the Visual Information
A student-facing Learning Manual for operating audio description with educational video: identify which visual information is missing from dialogue, choose an appropriate description route, preserve timing and source context, verify task relevance, and return the described information to the learning job.
-
Bolt Performance Calibration — A Mathematics or Science Test Can Become Partly a Language Test
Mathematics and science assessments inevitably use language, but unnecessary linguistic demand can obscure the subject knowledge a score is meant to represent. Bolt calibrates what the performance actually supports.
-
Student/Studying Interface Learning Manual: Spreadsheet Interface | A Correct Number Can Become Meaningless When Its Row, Column or Formula State Is Lost
A student-facing Learning Manual for operating spreadsheets as study tools: preserve row and column meaning, distinguish entered values from formulas, verify references and units, inspect copied formulas, and keep each result attached to the data state that produced it.
-
Bolt Performance Calibration — Two Students Can Answer Different Adaptive-Test Questions and Still Be on the Same Score Scale
Computer-adaptive tests can give different students different questions while estimating performance on a common scale. Bolt explains what makes that comparison defensible—and what can still go wrong.
