Reading: Repertory Grid as a simple Qualitative Data Collection Technique

There are many ways to collect data, some of which you will explore in more advanced courses. Previously, we discussed the need to develop a clear value proposition and to understand customer choice criteria. In this chapter, we look at one simple example of a way to collect such information

Repertory Grid Analysis

Repertory Grid analysis was developed by a clinical psychologist called George Kelly.^[1][2] Kelly also developed the Personal Construct Theory (PCT). PCT suggests that people develop their own mental models of the world, usually based on their own personal experiences. These models are composed of ‘constructs’, which are the rules that determine the model of the world that people have in their heads. In some ways, constructs can be seen as similar to beliefs. We all carry these models in our heads, determining what we like, what we don’t like, and what our preferences are for any given situation. The model changes over time.

As a means to measure PCT, Kelly developed the Repertory Grid interview technique. This has been widely used in research and it is particularly useful in marketing research because of its simplicity combined with the ability to elicit a clear model around what is important to respondents.

Repertory Grid Interviews

Repertory Grids are developed through interviews or focus groups. For marketing research, a company may gather together a small group of people who are representative of the customers in a particular market. The group should represent the demographics and other characteristics of customers in the market. Often, a study will feature several differing focus groups with slightly differing sets of customers. For example, one group could be people in their teens, while another could feature just people aged 30 or older. Others might be limited by gender or by where they live and shop.

The focus group will usually be conducted by a moderator who is trained both in the repertory grid method and also understands the problem the client company needs to solve.

The focus group will begin with the moderator explaining the purpose of the discussion. They would then introduce so-called Elements which respondents will evaluate to elicit Constructs. For a marketing repertory grid, the Elements might be a list of brands, or more general products or product categories, stores, new product ideas, package designs, shopping channels, or something else relating to the client’s business.

For example, imagine you represent a carbonated soft drink brand and need to develop a list of choice criteria for how customers evaluate and choose the different brands of carbonated soft drinks available in a particular market. You could do this through a focus group that uses Repertory Grid.

There is no one way to conduct a Repertory Grid focus group, but a simple example would be to write each of the Elements on a single piece of card (which the moderator would prepare in advance). The moderator would then present these cards in sets of three to the focus group participants.

• The Elements would be the various brands of carbonated drinks in the market in question.
• The Constructs would be the rules or perceptions that focus group respondents elicit during the focus group discussion.

The first three cards might show Coca-Cola, L&P, and Vibe Sparkling Water.

After presenting three cards to the focus group, have respondents discuss the characteristics of each one and split the three into a pair (two cards) and a separate card. Respondents should be free to separate cards however they wish, with as little input as possible from the focus group moderator:

Once respondents have made their choice, the moderator should have respondents explain the difference. Sometimes, respondents are able to identify multiple different factors for the same split:

• Coca-Cola is American – L&P and Vibe Sparkling Water are New Zealand brands
• Coca-Cola has many different package sizes – L&P and Vibe Sparkling Water have only a few package sizes
• Coca-Cola is expensive – L&P and Vibe Sparkling Water are relatively cheap

And so on. Each point made in this way is a Construct in the Repertory Grid. Remember: it is the respondents’ perceptions that are key to this exercise, so the moderator should not make any judgement. Once respondents have run out of ideas for a particular split of cards, the moderator can then ask respondents for other ideas. For example:

Once again, the moderator should ask respondents to explain why they separated the cards in this way. For example:

• Coca-Cola and L&P are high sugar content – Vibe Sparkling Water is low sugar content
• Coca-Cola and L&P come in large, plastic bottles – Vibe Sparkling Water comes in cans that are easier to recycle

And so on again. Constructs do not need to be matching pairs, and they do not need to be explained. These are genuine perceptions of genuine customers, and so point directly to the factors they consider when making a choice between multiple options.

Once respondents can no longer separate the three cards into two and one, the moderator should then introduce a new set of three, keeping in mind that they should re-use items that respondents have already evaluated. For example:

The focus group discussion would continue until all the cards had been evaluated by respondents, mixing and matching them into different pairs. At the end of the focus group, moderators should have written down a great many paired constructs. By analysing these constructs in more detail, it is then possible to consolidate the list into the commonly elicited choice criteria used by customers.

These choice criteria can then be used in competitive analyses or for other activities. For example, as focus group research requires respondents to be present to answer questions, it is relatively time-consuming and, consequently, high cost. However, once a list of constructs is developed, that list can then be used in a much larger questionnaire survey across a broader sample of respondents to gain further insight. As the constructs are developed in pairs, they are easily adapted to a Likert-type scale, with questionnaire respondents indicating their relative agreement between the two constructs presented.

There are many marketing research techniques that can be used to elicit clearer data on customer choice criteria. The Repertory Grid technique is a relatively simple one. This same technique can be used in many other ways when choice criteria are required, such as evaluating possible job options or choosing which paper to take.