red thread of thinking

There should be a clear ‘red thread’ of thinking that arises from the client brief and connects the method, analysis and findings. Theories, models and conceptual frameworks are essential for research, because they describe, explain, and predict.  They enable the researcher to name and frame the problem in terms of concepts and identify the issues.

  • Rigorously analyse the brief, unpicking vague concepts, questioning and challenging.
  • Consider what frames of thinking, perspectives, models or theories might have some relevance. See below for some inspiration. Generate ideas about what you might want to explore.
  • Concretise the objectives – turn the concepts into things you can see, hear and feel – and ask about.  What are the relevant variables?
  • Choose methods that will surface the needed information – staying open to emergent ideas, to adapt and change your thinking. Ensure they will provide the evidence needed; and fit with the sample, timing and budget.
  • Develop a realistic and practical sampling strategy (demographics, psychographics, lifestages?)
  • Use the thinking as a framework for analysis – and even the storytelling.

Questioning the brief:

  1. Make sure you fully understand the brief and business problem. See the problem in its context, understand the challenges, and be aware of potential weaknesses in the organisation that could affect it.  Read between the lines. Ask questions of the business unit or client, look out for political/internal disagreements.
  2. Think about the end use of the data and what evidence will be required to solve the problem, and convince the stakeholders. Ask the problem owner what they will do when they have the research results.
  3. Define and analyse the concepts involved in the brief. For example – how will you know when you have come across ‘brand engagement’ in the research?


Theories for thinking

It is an overall framework that unites a range of disciplines and helps develop a theoretical viewpoint that can link context, emotions, motivations, biases and behaviour.  If you find behavioural economics frustratingly incomplete, this will provide you with the answers you need.

Evolutionary psychology has become a single seamless framework that can span disciplinary divides. (Dunbar 2008). The Uber Framework unites evolutionary psychology, emotions, cognitive biases, motivations and behaviour – because all of these are part of the adaptations we have made to succeed.

Download it here  The Uber Framework  and please help evolve it further – make it more comprehensive and useful.

There is a longer version that includes the main biases that researchers need to be aware of and how to manage them. Reference Guide to the evolutionary nonconscious.  Comments and suggestions to

What is The Unconscious? Reference Guide to the evolutionary nonconscious The Uber Framework

The classic, Freudian version is  defined as all the aspects of the psyche not available to awareness; and is inferred from observable phenomena. Psychoanalysts see it as the most influential part of the psyche, and the source of our motivations.  Unconscious ‘ideas’ have emotional charges and therefore strive for expression in the conscious. Because some of these ‘ideas’ can be perceived as threatening to the integrity of the personality, defence mechanisms are employed to moderate or subvert their effects.

The psyche is split into three main, conflicting parts,

  1. the Id -the source of desires that works according to the pleasure principle  “I want it and I want it now”
  2. the Ego, the mediator that works on the reality principle  “I will take care of my needs when and how its appropriate”
  3. and the Superego, which is the Conscience and the store of positive models created from rewards and punishments.

This is the foundation of the psychodynamic approach, which appears in other forms. In Transactional Analysis it is Child, Parent and Adult, for example.

It is the perfect description of many inner conflicts between needs and desires and ‘shoulds and oughts’. “I really want some chocolate, but I know I should have fruit instead.”  

Its still very useful for thinking about research issues where there is conflict within an individual.

The unconscious in the history of consumerism

The concept of the unconscious is embedded deeply in the history of qualitative research, in public relations and in the consumer society itself.

The concept existed before Freud, but he popularised it. His nephew, Edward Bernays (the ‘Father of Spin’) used Freudian concepts to reinvent pre-war propaganda into Public Relations. American politicians and planners believed that powerful primeval unconscious urges had manifested in the brutality of Nazi Germany. Anxious that the melting pot of the USA could contain similar dangers, they planned to divert these urges into the acquisition of consumer goods. (In Freudian terms – sublimation). They believed this was the best way of maintaining democracy – and it had the happy by-product of driving the expanding economy. Bernays worked closely with politicians and corporations to develop what he called “the engineering of consent” – controlling the minds of the public without their awareness.  Watch the extraordinary documentary “The Century of the Self” to really grasp the extent to which everything we take for granted in consumer society is suffused with Freudian ideas.   Adam Curtis Documentary

A number of psychoanalysts came to the US to work in advertising and research, most famously Ernest Dichter, founder of Motivational Research. His most famous book, the Strategy of Desire, is still an amazing read. However he was defeated by hubris and his work discredited, while the general manipulation of the public was exposed in books like The Hidden Persuaders.

Focus Groups versus Group Discussion

Mike Imms Roots and Theoretical Basis of Qual puts forward the hypothesis that the introduction of focus groups in the US was  a reaction against the now discredited psychoanalytical approach. Whereas the UK had developed more slowly and retained its psychological approach, resulting in a deeper and more analytic approach. The term ‘focus group’ now covers almost everything, but in the 1990’s it characterised what Mary Goodyear called the ‘Cognitive approach’.

Characteristics Cognitive (USA) Conative (UK)
Purpose Demonstration Exploration
Interviewing Logical, closed, hand shows Opportunistic, open, probing, facilitation
Response Give answers Debate issues
Interviewer Moderator Researcher
Observer’s role To get proof To understand
Transcripts Rarely Usually
Analysis On the spot In depth
Output Information Understanding


“They are, indeed, groups ‘focused’ on specific issues with all the disciplines that follow from convergent thinking and control.  The analysis or articulation of the problem has been worked on before, and so the interview is largely a question of confirming or expanding known issues.” (Goodyear)

Goodyear, M, (1996) Divided by a Common Language: Diversity and Deception in the world of Global Marketing, Market Research Society Conference

Further reading:  Download Confronting the Unconscious  (AQR InDepth article) or read online

If you want to know more about the tripartite psyche, defence mechanisms, archetypes and myths, Jung and the Myers Briggs, download Essentials of Psychoanalysis for Researchers

But be aware that thinking has moved on. The new definition of unconscious is ‘brain activity occurring out of conscious awareness’ and this has given rise to the nonconscious and methods of neuroscience.

Imagine a person who is very quick thinking. He or she can sum up a situation quickly, knowing what key cues to look for, and can take immediate action. (Its called ‘thin slicing’ – the ability to find patterns in events from very narrow windows of experience). It would be good to know a person like this, who is very adaptable, can keep out of danger, and can stay on top of fast moving situations where a lot of decisions have to be made. The good news is – that person is you. In Daniel Kahneman’s terms, that is what System 1 thinking would be like if it were a person.

Other great characteristics are:

  • It generates impressions, feelings and inclinations
  • Operates automatically and quickly
  • It mobilises attention, especially when there is a change from the normal
  • Executes skilled responses and generates skilled intuitions, (after adequate training.)

Fast thinking is going to have some drawbacks too. It

  • Jumps to conclusions
  • Operates automatically (subconsciously) so you don’t know whether its short cuts are necessarily the best ones
  • makes decisions on the basis of impressions, feelings and stereotypes, even when a more rational approach would be better
  • Infers and invents causes and intentions
  • It is subject to a range of biases as a result of these limitations

You can see how such a way of thinking could have evolved, but you will also be aware that you can think quite rationally. And yes, there is System 2 thinking. The System 2 thinker is logical, calculating and conscious. It can do compare and contrast, compare objects on attributes, make chains of inferences – everything we think of as the higher thought processes. (It has its own biases too).

But sadly it’s also relatively slow – and gets tired easily. Its called cognitive depletion – that desire to reach for a chocolate bar after intellectual effort. Or you just switch to System 1. So having narrowed down your choice of fridge freezers on all the rational criteria – you go for the one with the cool blue light on the outside.

Your systems work together. So all those impressions, feeling and inclinations from System 1, when they are endorsed by System 2, become beliefs, attitudes and intentions. System 2 can train System 1 on what it should look out for, and it can program the memory to override habitual responses.

Generally System 2 thinks its in charge, and its true that most formal knowledge is in System 2 format.

Implications of BE for research

  • Do research as much in context as possible (memory is strongly influenced by context)
  • Use immersive research and ethnography as people are poor witnesses to their own behaviour
  • Be careful of the order of talking about stuff/presenting stimuli (Anchoring bias)
  • If you are researching decision-making processes don’t recruit people who have already made the decision – they will likely justify it
  • If you are following a decision-making process expect a mixture of quick short cuts and rational evaluations – watch out for when people get overwhelmed by information
  • Many biases relate to a need to maintain a sense of consistency and self-esteem – make it alright for people to change their mind during the research
  • Aim to work out what short cuts people generally use to make decisions in that segment
  • Be suspicious when people say they don’t respond to free offers and numerous other ‘marketing gimmicks’
  • If you are assessing a number of product options (combinations of features and prices) try to use the experimental method where possible, because the parameters set by one will frame the other.
  • Be very careful how you write stimulus material: especially in relation to implied gains or losses and be aware that ‘cognitive ease’ (easy to read and understand) tends to make material more believable.

Kahneman’s theory sums up a great deal of work on cognitive psychology, but this approach is not totally new. Petty & Cacioppo had developed the Elaboration Likelihood Model in the 1980s, which contrasted Central (slow, System 2) and Peripheral (Fast System 1) routes to decisions.  It looked at the implications for advertising effectiveness and persuasion, but I suspect it never took off because it didn’t have a catchy System 1 kind of name. “Let’s use the Elaboration Likelihood Model – maybe not”

More detail on specific cognitive biases – download: Overview of BE and research implications

Slides and notes from the AQR webinar: The New Unconscious is the Nonconscious

There has been an evolution in the conception of the unconscious. In the 1990’s, the ‘decade of the brain’, neuroscience reconceptualised the unconscious as brain processes that occur outside of conscious awareness – 80-90% of them, apparently.  Habits, language, programmed and learned social and cultural behaviour; there is a large amount of automaticity in what we do.  But since we are only aware of conscious thinking, it seems to us like most choices are consciously made.

Self-delusion continues as a theme of behavioural economics.  There is a cognitive bias for everything, including believing you are not biased.

‘Strangers to Ourselves’ by Timothy Wilson is a powerful review of mental processes that “that are inaccessible to consciousness but that influence judgement, feelings, or behavior”.  This is now called the Adaptive Unconscious – because it helps us to solve problems and adapt to circumstances, and it plays a role in many important life decisions.

Kenrick and Griskevicious in The Rational Animal  take an evolutionary perspective to explain the apparent ‘irrationalities’ so keenly discovered by behavioural economists.

Evolution is about surviving to pass on your genes, and those who are best at the goals of avoiding harm and disease, getting friends and status, attracting and keeping a mate and caring for their family, are those who have survived. And they have done so by having largely unconscious systems and strategies for triggering appropriate behaviour for each evolutionary challenge.  This is the idea of adaptive unconscious behaviour guidance systems.

So the unconscious is finally seen as being largely helpful and goal–directed. Usefully, this thinking helps link the proximate causes of behaviour – the ones people can talk about, with stimuli, emotions, and the ultimate causes. The Uber Framework is a way of putting this thinking into practice. Please try it and send feedback to help make it more useful:

The nonconscious is providing a field day for the purveyors of neuromarketing, biometrics, facial coding etc., all ways that avoid asking people directly what they are thinking and feeling.  At the 2014 IIeX, which brings together top notch clients such as Unilever, General Mills and Google, with research suppliers, the clientside most requested meetings were with suppliers in the nonconscious measurement category.

Notice the ‘measurement’ part – they are drawn to the safety of numbers.  Hopefully they are used as part of a suite of techniques, since they too require context and interpretation. But there is never just one answer to these marketing questions.  There is not one theory that could explain how all advertising works. There is no one theory that can explain everything about how consumer decision-making works. The skill of the researcher is to identify what kind of a model is needed to explain the scenario at the nub of the research question.

Outdated or not, sometimes the idea of an inner conflict between temptation and conscience is just what is required to understand a process.  There are tools for surfacing and clarifying it, from simple bubble drawings to role play. Jung had moved on Freud’s ideas to include symbols and archetypes – and yes, there are ways of identifying and working with archetypes in brands.

Habits, technically nonconscious, have been analysed (see Duhigg, The Power of Habit) and we understand how they are made and changed. And there are simple techniques for observing and researching habitual behaviour, made all the more powerful by the mobile phone.

Social influence is something people are little aware of; it’s not very empowering to say you did something simply because thousands of others have.  Ironically one of the best ways of observing social influence at work is the focus group – as long as the moderator is not busy trying to make people NOT influence each other.  But we also see it in looking at user imagery, in process analysis and customer journeys.

Most consumers can’t talk about the cultural influences that frame their thinking because these are too pervasive. But techniques from anthropology and semiotics put cultural analysis in the grasp of qualitative researchers, even if they need to hire an expert to do it.

System 1 thinking is fast and automatic, but it’s not random. There are sets of associations, recurring choices of heuristics, impressions and prejudices, and it can be tapped into through word association, collaging, enactment and the Implicit Association test, amongst others.  There are many ways of eliciting how people connect and classify products, brands and how they link product benefits to higher level emotional and social needs.

Behavioural economics has brought to the fore the influence of context in priming and nudging behaviour while mobile technology has made it easier than ever to see directly into the environments where decisions are taken and products are consumed.

So qualitative research has a rich set of tools for working with the nonconscious that need the application of some informed eclecticism – and both Systems 1 and 2.  See The Source – the Qualitative Technique Sourcebook to understand how  projective and enabling techniques can help us work with the nonconscious. Brain scanners not required.

In consumer psychology, they are the short cuts consumers use in making decisions. Kahneman and Frederick think there may be two stages of processing:

1. The completely unconscious ‘noticing’ stage. This selects a small part of all the available information about the product/brand (including associations, identity, emotions, objective characteristics, advertising, pack cues etc) to  provide simple substitutions for the complexity that the brand entails. Or the consumer may use the ‘availability’ heuristic – judging by what is more available in memory. Memory is often biased towards vivid, unusual or emotionally charged examples.

2. The second stage may involve comparison between a number of products and is also shaped by the needs, preferences and desires of the consumer at the time. There are cognitive influencers such as framing, hierarchies and connections of ideas that affect the terms in which we think about something. For example it is hard to think of tax as a positive construct – even the idea of tax relief contains within it the idea that tax is an affliction.

In addition, there will be strategies consumers have learnt to use successfully in the past. For example, not wasting time getting the optimum possible combination of features for the money, but satisfycing – find goods that exceed a certain threshold of satisfaction.  On other occasions they may seek to minimise risk, or buy the most expensive to impress somebody.

Other heuristics include

  • Price as a surrogate for value
  • More or less than a previously anchored price
  • Familiarity is linked to trust – the mere exposure effect
  • Trusting the experts, authority
  • Symbols of functionality – does a face cream with nanospheres work better than one without?
  • Alternatively feeling uneasy about something e.g. about Genetically modified foods
  • Negativity bias – a negative piece of information has more power than a positive one
  • Feeling valued because you are offered a concession (bargain)
  • Contrast effect – compared with a really expensive item it looks good value
  • Sunk costs – you have already invested time/money so you might as well buy it
  • Social proof – in uncertain situations we would do what most people to
  • Admission to a reference group – the product will make you part of an admired group (or one you would rather avoid –so you don’t buy)
  • Liking, attractiveness, similarity to ourselves
  • Associations with other items or people that have power or value
  • Optimism bias – tendency to be over-optimistic about the outcome of something you have planned
  • Ambiguity aversion versus cognitive ease. We prefer things that are easy and clear to buy and aim to avoid those where there is conflicting or incomplete information.
  • Framing how we think about something – it’s not a cost, it’s an investment

Personal Construct Psychology (PCP) was founded by George Kelly, who wanted to understand how individuals ‘construe’ (make meaning of) the world.  It is an individual-based psychology. One of the main premises is that we all see the world through our own goggles, the lenses of which are fashioned from previous experience. However because of shared experience, language, social and cultural factors, we have many shared constructs that enable us to communicate and make predictions about other people as well as ourselves.

The theory is a form of Constructivism – according to some academics, the underlying philosophy of qualitative methodology. Qualitative research is about understanding the meaning people ascribe to things and events and PCP gives us some powerful tools for this.

Kelly’s theory has been described as “every man his own scientist”. Each person notices repeated themes in his life, names and categorises them, and then applies these labels to the people and situations he meets.

For example, a man might notice that the people with big noses he meets all are friendly and have a good sense of humour.  He is using the constructs ‘big nose – small nose’ and ‘good sense of humour – no sense of humour’, and when he meets a new person he will (subconsciously) notice their nose and anticipate what kind of sense of humour they will have.  We make no value judgements about the validity or logic of these constructs in PCP. We have to accept that it may be odd, but it is his way of looking at the world.

PCP is based on working with the individual’s construct system, a very extensive and complex hierarchical network of inter-relating constructs. Construct hierarchies have a pyramidal structure, with lots of detail at the bottom, and as you go up the levels they become more general and inclusive – and generally more important to the person’s sense of identity.

Construct systems are motivational – they provide reasons for doing things, and they respond to the environment.

You may not have heard of George Kelly, but you may well have heard of the Kelly Grid, technically the Repertory Grid, and of Laddering, or Means End Chain analysis. Nowadays its not used by many people, but those who do use it, find it revelatory.


Rep grid used for a website comparison
Rep grid used for a website comparison

In PCP, the things constructs are applied to, are called elements. They can be people or roles, situations or products, types of packaging etc.  The Repertory Grid or Kelly Grid (which comes in different varieties) uses comparisons between elements to elicit the constructs, and then requires some form of rating of each element.  The grid then shows the patterns of interaction between them, rather like a cluster analysis.

A repertory grid can be used before laddering to identify the key constructs, the basic distinctions. The process called Laddering is essentially going up the pyramid, from very basic distinctions about the world, to very meaningful distinctions. This explains its value as a research method, and why it needs great sensitivity – because you are dealing with a person’s core constructs.

Laddering exposes the structures of another person’s thinking, emotions and values. It shows the links between the basic attributes of a product (or organisation), their higher benefits, and the core needs and values they meet.

The process of Laddering is used in creating the Hierarchical Value Maps that come out of Means End Chain analysis. although its an individual process there are means of coding and aggregating ladders, although often two or three used as a pilot study are revealing enough. More information about Laddering, and associated techniques

Here is one of the most thorough reviews of models relating to behaviour change: Behaviour Change Knowledge Review from Government Social Research and the accompanying Practical Guide to Behaviour Change Models
Download NLP Overview for Researchers  for an explanation of:

The basic principles, the communication model, logical levels, setting outcomes, working with states and modalities, sensory acuity, rapport  and perceptual positions.

The Uber Framework document contains an outline of Plutchiks’s theory of emotions and an updated version of Maslow – that includes both deficiency and abundance needs. The framework connects these to environment, thinking and behaviour.

In qualitative research we are more likely to be working with less intense emotions than terror or joy. Emotion and Social Judgement, a short review of the literature on mood by Bower is very relevant. Putting  respondents in a good mood is a common moderating style, but it may be introducing a bias:

  • Mood dependent retrieval shows that the mood people are in at the time will affect what they remember about their life. People in a good mood will recall more happy events and vice versa.
  • Mood congruent processing shows that people pay more attention to material that matches their current mood.
  • People who are primed to be in a good mood will evaluate things more positively, and make more positive judgements of future events. Those in a bad mood are more critical.

There is a case for priming respondents to be in the mood that matches the one in the situation being researched.

There is some evidence that people vary in their tendency to think affectively (emotionally) or cognitively and that they are more motivated to action when a message is framed in the same way.

Journal of Personality and Social Psychology Bulletin (Mayer & Tormala, 2010) found that people who thought about the world in cognitive terms were more persuaded to give blood when the message was framed in terms of ‘thinking’. On the other hand the group that used emotional words was more persuaded when ‘feel’ was used.  There were the expected gender differences with women giving more weight to emotionally-framed arguments.

This is potentially a problem for mass communication, but emotional messages can be couched in emotional language and in cognitive language. I can say “I feel happy,” or “I’m thinking happy thoughts.” Both messages contain emotional content but people respond to them in different ways. Read the original post on Psyblog

A quick version of the Marlowe-Crowne Social Desirability Scale

Social desirability is a well-known bias in both qualitative and quantitative research.  There are ways of constructing questions to try to avoid it. But it’s connected to many other biases that are designed to help maintain self-esteem – egocentric bias, attribution errors, illusory superiority, self-serving bias, etc.

So it may not be that easy to discount. Plus, researchers are affected by it too. (Except we suffer from a cognitive bias that says we don’t have any biases). So it’s interesting to do this quick test to get a deeper understanding of what is involved. Just add up the scores and read the comments below.

Statement True False
1 I never hesitate to go out of my way to help someone in trouble 1 0
2 I have never intensely disliked anyone 1 0
3 There have been times when I was quite jealous of the good fortune of others 0 1
4 I would never think of letting someone else be punished for my wrongdoings 1 0
5 I sometimes feel resentful when I don’t get my way 0 1
6 There have been times when I felt like rebelling against people in authority, even though I knew they were right 0 1
7 I am always courteous, even to people who are disagreeable 1 0
8 When I don’t know something, I don’t mind admitting it 1 0
9 I can remember playing sick to get out of something 0 1
10 I am sometimes irritated by people who ask favours of me 0 1


Some self-deception is thought to be healthy; believing we are more talented or intelligent than we really are can help us influence and win over others, according to Robert Trivers, anthropology professor at Rutgers University and author of The Folly of Fools.

A score of 7-10 is considered high. Crowne says “it means you are extremely concerned with what others think of you, and you want to believe that you are as good a person as you say you are. You may have a strong desire to conform when you are in a group.”  You may be using this as a protective mechanism, but finding it stressful to keep up.

A score of 4-6 is average, and indicates that you are fairly honest with yourself about your strengths and weaknesses. You are not overly concerned with the views of others, nor oblivious to them.

A score of 0-3 Crowne says” These people are either extremely frank, with high levels of self-disclosure, or actually quite distressed about themselves”.  In other words you may be too self-deprecating and be quite hard on yourself.

The Marlowe-Crowne Social Desirability Scale was devised by Douglas Crowne, Distinguished Professor Emeritus at the University of Waterloo in Canada, and colleague David Marlowe.  It works by asking questions that, in most cases, cannot be true of you all the time.  The extent to which you believe them to be true of you indicates how strongly you feel a need to look good in the eyes of others, and how honest you are about yourself.

The Default Qualitative Philosophical Framework: Constructivism –  reality is socially constructed

Simplifying enormously, this is a view that reality is fundamentally mind-dependent. People create their own reality by the way they interpret the world. Interpretations are mediated by culture, language, social norms and so on, but the same thing can have different meanings in different contexts.  HSBC constructivismThe old HSBC advertising makes the point perfectly.

This is in contrast to Positivism, which underpins quantitative research, assuming that an objective world exists ‘out there’, and it is subject to regularities and laws that the scientific method can uncover.

Implications of Constructivism

  • Validity lies in getting as close as possible to the world-view of the participants
  • Research needs to involve an analysis of what things mean to people and why
  • Since researchers also construct the world, it is important they aim for ‘empathic neutrality’ – making their own assumptions and values transparent while being open and non-judgemental.
  • Context: physical, social, emotional, cultural – is important because it affects meaning

(A good introduction to the theoretical foundations of qualitative research can be found here:  Qualitative Research Practice – A Guide for Social Science Students and Researchers )