Important Metrics of Quiz

by

Abstract

Several important metrics related to questions, answer variants, and quiz itself as well as their semantic meaning are described here. No implementation notes provided in this article as they can vary depends on used environment. Considered models are based on the Simple Quiz Model.

Quiz Metrics

Quiz Metrics are metrics, that are related to a quiz itself — not to the questions nor the answers. Some of these metrics are cumulative and derived from the question metrics.

Quiz Impressions

Quiz Impressions (QI) is an analogue of Viewable Impression metric, widely used in web marketing — the count of users who actually saw the quiz (minimum 50% of the quiz component area for at least one continuous second). 1

Quiz Starts Count

Quiz Starts Count (QS) is the number of users who starts a test.

Quiz CTR

Quiz CTR is the ratio of users who starts a test to the number of users who saw the quiz.

$CTR = QS I × 100 % ; CTR ∈ [ 0 % , 100 % ]$

Interpretation

The higher Quiz CTR, the higher user engagement, the better your quiz. This metric reflects quiz's attractiveness for user. Placing quiz into irrelevant content or making start screen look a little bit boring causes Quiz CTR downgrade.

Quiz Loads (QL) is the number of times the quiz was loaded on the page.

Usage

Quiz Impressions is the metric that sometimes can be hard to implement, so Quiz Loads can be used instead of it when there is no reasons to achieve a high accuracy.

Approximate CTR

Approximate CTR is the ratio of users who starts a test to the number of users who saw the quiz.

$CTR approx. = N s QLC$