Skip to content

Designed-pattern authoring: linking each teaching pattern to its generation rule, check, and reference analysis

Design · All designs
Evidence · no claims cited

Description

A designed pattern is "a trend, group difference, interaction, or exception intentionally built into a dataset for students to investigate." The workflow records each pattern in an executable case specification connecting it to its generation rule, comparison groups, measurement, and sample-size requirements, so that the generation rule, numerical check, reference chart, and rubric use the same definition of the intended finding. Verification examines the combined effect of interacting rules on the exported data, and the probability set in the generator is not itself evidence that the pattern appears in the data.

Design Implications

Context

Requirements

  • Each pattern requires one or more checks; the specification defines how overlapping rules interact, since an adjustment made for one pattern may affect another

Constraints

  • The order of generation rules affects what students observe; in WindowDash the ZIP 10011 cuisine adjustment overrides earlier cuisine rules

Target Learners

  • undergraduate business analytics students

Learning Goals

  • investigating unfamiliar data with AI
  • distinguishing counts from shares
  • justifying conclusions from evidence

Claims

Examples

-

Key Sources

  • Bang An, Maria Hamdani, Joseph Fox. (2026). DataCanvas-EDU: An Agentic Framework for Instructor-Guided Synthetic Data Generation in Business Analytics Education. https://arxiv.org/abs/2609.19617
Page metadata
Field Value
type design
id designed-pattern-case-specification
title Designed-pattern authoring: linking each teaching pattern to its generation rule, check, and reference analysis
description A designed pattern is "a trend, group difference, interaction, or exception intentionally built into a dataset for students to investigate." The workflow records each pattern in an executable case specification connec...
status review
generated by process:settle-candidates on 2026-10-10
trust unverified — no one has confirmed this page's content
sources 1 — bang-an-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.