Design Experiments

Design experiments refers to the process of systematically planning how to combine different product or service attributes to create choice sets for respondents to evaluate as tasks. This step is crucial because it influences the quality of the insights you can derive from the analysis. Here's a breakdown of what designing experiments in conjoint analysis entails:

  1. Attribute and Level Selection: First, identify the attributes (features, services, etc.) of the product or service to be analyzed and the different levels (variations) for each attribute. For example, for a smartphone, attributes might include screen size, battery life, and price, with each attribute having its own set of levels (e.g., screen size: 5 inches, 6 inches, 6.5 inches).

  2. Construction of Choice Sets: Based on the attributes and levels, create a series of choice sets. Each choice set is a combination of products or services with different attributes, and respondents are asked to choose their preferred option from each set refered as task. The design of these sets can follow different methodologies, such as balanced (full factorial), fractional factorial, or more sophisticated designs like orthogonal or efficient designs.

📝 Creating the design experiments

ngSurvey allows you to create your choice set either using

Once you have generated your designs you can review them when you edit your conjoint question. At that point its not possible anymore to edit any of your attributes or levels.

Last updated

(c) 2024 Data Illusion Zumbrunn. All rights reserved. Reproduction strictly forbidden.