Business Context

InsightHub, a BI platform, is reviewing whether analysts choose appropriate visualizations for categorical versus numeric data. The analytics lead wants a simple statistical check on whether chart misuse is associated with data type.

Problem Statement

You are given counts from 200 recently published dashboards. Each chart was labeled by the primary data type it visualized and whether the chart choice was appropriate. Determine whether visualization appropriateness differs between qualitative data (categorical labels such as region, device type, plan tier) and quantitative data (numeric measures such as revenue, latency, or order value).

Also explain, in plain language, the conceptual difference between qualitative visualizations and quantitative visualizations, and identify which chart types are typically suitable for each.

Given Data

Examples used in the audit:

• Qualitative data: product category, country, subscription plan
• Quantitative data: daily sales, basket size, delivery time

Use a significance level of 0.05.

Requirements

1. Briefly define qualitative vs quantitative visualizations.
2. State the null and alternative hypotheses for a test of association.
3. Compute the expected counts under the null hypothesis.
4. Calculate the chi-square test statistic.
5. Compute the p-value and decision at $\alpha = 0.05$.
6. Quantify the effect size using the difference in appropriate-chart rates.
7. Give a business recommendation for analyst training and dashboard review.

Assumptions

• Each dashboard contributes one independent chart decision.
• The audit labeling is accurate and consistent.
• Expected cell counts are large enough for a chi-square approximation.