How to use Matrix Questions

A matrix question is a group of multiple-choice questions displayed in a grid of rows and columns. The rows present the questions to the respondents, and the columns offer a set of predefined answer choices that apply to each question in the row. Very often the answer choices are offered in a scale.

A blank template of a matrix question.

When to use Matrix Questions

Matrix questions are best used as a way to ask several questions about a similar idea when there is a scale involved.  They can be used for a variety of reasons, either as a mini-survey on their own, or as a single question type within a larger questionnaire. The closed-ended, predefined answers that apply to a series of questions make Matrix questions great for:

  • Customer experience/satisfaction surveys.
  • Questions about a subtopic in a larger questionnaire.
  • Consolidating many rating-scale questions in a more digestible format.

Customer experience surveys

Matrix questions are commonly used for customer experience surveys. For example, to ask a respondent about their experience on a flight, the rows might ask the respondent about the service, food, or entertainment while the columns ask them to choose a rating response.

Example of a customer experience survey using a matrix question.

Questions about a subtopic

Oftentimes a questionnaire includes many ideas, but some of them are specific to a subtopic within that survey. Matrix questions are an effective way to cluster these ideas into an easy format for the respondent to understand what is being asked of them.

For example, in a brand awareness survey, a customer might use a matrix to get more information on brand perception. 

Example of a brand perception question using a matrix.

Benefits  of Matrix Questions

The format and structure of matrix question types provide some unique benefits. Because it is actually a series of questions presented as a single table, it appears as a single question on the survey. This has the benefit of saving space (both on paper and in a digital survey) as well as reducing drop-offs from respondents who do not feel like answering 5 nearly-identical questions back-to-back. 

The grid is easy and intuitive for respondents to follow with closed-ended, predefined answer sets, which means quick responses and a clear, easy-to-analyze dataset as the outcome. 

Drawbacks of Matrix Questions

While there are many benefits, there are a few things to keep in mind when using Matrix questions. 

The table formatting, while easy for respondents to answer, can also result in activities such asstraightliningor other pattern-making within the table.

Another issue can be the addition of too many rows or columns, which may negatively impact the data quality. If there are too many choices, respondents might lose interest (and be more likely to provide quick, insincere answers to move quickly through it). In some cases, this can impact formatting as well, particularly in a digital survey environment such as mobile. If the matrix question is not designed for an optimal mobile user experience, it can be confusing or frustrating for respondents. 

Example of a matrix question that is optimized for mobile surveys.

Some survey companies will also charge each row in the matrix as an individual question, which may impact the cost of the overall survey. Keep this in mind when building your questionnaire. (Pollfish views matrix questions as a single question type, so pricing does not vary based on the number of rows and columns included).

Types of Matrix Questions (Single- selection vs Multiple-selection)

Matrix questions, like regular multiple-choice questions, can be either single-selection or multiple-selection. This means that a respondent can choose either a single answer choice per row, or they could choose multiple answer choices per row. 

These might be used in competitive analysis surveys to better understand how a product or brand is measuring up against competitive offerings. 

 

Example of a multiple-selection matrix question in a competitive analysis.

How Matrix questions differ from a Likert Scale

Many people believe that a matrix question is just a Likert scale, when in fact, it’s the other way around. 

A Likert Scale is a specific type of matrix question that is specifically designed to measure opinions in a linear fashion. Using a 5 or 7 point scale to collect user sentiments, a Likert Scale can be used to determine scaling attitudes such as:

  • Agreement (Strongly Agree- Strongly Disagree)
  • Likelihood (Very Likely- Not very likely)
  • Importance (Very Important- Unimportant)
  • Frequency (Always- Never)
  • Quality (Excellent- Poor)

A matrix question is a format for the question, meaning it is presented in a grid (or matrix). While matrix questions often happen to be Likert Scales, matrix questions can also be used across a variety of use cases outside of attitudinal measurement, as demonstrated above.

Best practices for writing a good matrix question

Writing a good matrix question follows many of the same best practices for writing good survey questions in general. However, due to the grid formatting, there are a few additional things to be aware of.

  • Limit the number of rows or columns. Keep it around 5 different options for questions and answers so as not to bore or overwhelm respondents.
  • Give respondents a way to opt-out of things they are not familiar with, such as a “no opinion” or “neutral” answer option.
  • Don’t make the questions too long. In the table format, long questions provide a poor respondent experience.
  • Try to group like-concepts together. For example, if you want to know about brand perception, keep the questions related to that subtopic. 
  • As in any closed-ended scaling question type, keep scaling answer choices in order so as not to confuse the respondent.

Matrix question types (both single and multiple-selection) are available in the “questionnaire” section of the Pollfish survey builder. Sign in or create an account to get started on your next survey.