Elicitation and data collection

One of Gordon’s most significant pieces of research, with Neil Maiden from City University, resulted in the ACRE framework.  Their research question was: Why is it so difficult to get complete and correct information and knowledge out of people, even when they are actively trying to help? The reason turned out to be that there are several main categories of knowledge, memory and skill, plus numerous sub-categories, each of which has different implications for choice of elicitation method. For example, working memory has a duration of just a few seconds, so the only way to access it is via real-time elicitation using methods such as think-aloud technique. There’s more about the ACRE framework here.

The short version of the framework can be summarised as do, don’t, can’t and won’t, as shown in the table below.

Things that people do tell you about can be accessed via any technique.

Things that people don’t tell you about, but would be perfectly willing to talk about, are semi-tacit. Examples include working memory; things people don’t think are worth mentioning; and things people don’t mention because they assume you know about the topic already. To elicit semi-tacit knowledge, you need methods such as laddering, think-aloud and card sorts.

Things that people can’t tell you about involve tacit knowledge in the strict sense; for example, the compiled physical skills often known as muscle memory, and expert recognition of familiar patterns – such as a doctor recognising a medical condition. To elicit tacit knowledge, you need nonverbal methods such as observation and experimentation.

Things that people won’t tell you about directly are a special case of semi-tacit knowledge. You can get some insights into this by using projective techniques, and indirect observation.


The ACRE framework reference: Maiden, N.A.M. & Rugg, G. (1996). ACRE: a framework for acquisition of requirements. Software Engineering Journal, 11(3).

Methods tutorials

Card sorts for finding which features of an item or concept are important to people

Content analysis for making sense of what is (and isn’t) in a text

Co-occurrence matrices for measuring similarities and differences between items and concepts, via card sorts

Hexagonal cards for generating ideas swiftly and informally

Idea commenting for developing ideas generated via other methods

Idea generation methods for larger-scale, more systematic generation of ideas, in ways that give everyone an equal voice

Laddering: an overview of the method as a whole

Laddering downward for explanations of subjective or technical terms

Laddering upward for goals and values

Likert-style scales and visual analogue scales for quantifying subjective opinions

Liking and disliking scales for getting better insights into people’s opinions

Multi-Criteria Decision-Making for choosing options systematically

Observation for systematically watching what actually happens, as opposed to what people say happens

Projective approaches for investigating topics that people are reluctant to talk about directly

Think-aloud technique for getting explanations, and for access to what’s in a person’s working memory

Visual analogue scales for finer-grained quantification of subjective opinions


Image: A card sorting session in progress

Material on this website is copyright of Hyde & Rugg unless otherwise stated.