Feb 26, 2007

What do Art and Qualitative Research have in common?

Understanding Qualitative Research through Metaphors...
This paper, on using Quilting as a Metaphor to clarify the research process, caught my fancy since I had written a brief piece on the differences between qualitative and quantitative research inspired by another research paper that used the 'quilting' metaphor to explore the differences between these two disciplines. These papers have not only made for interesting reading but also helped me clarify some of my own thoughts about research through these metaphors.

In my early days as a researcher, I struggled to understand the rules of the game.

  • How does one define what an adequate sample size is – whether 4 groups or 8 interviews are enough to represent the reality we are trying to capture?
  • What exactly is meant by analyzing the data – other than placing data in a structure and looking for patterns – what does one actually do?
  • How do you define a good interview – was it with the chatty woman who couldn't stop talking and kept us amused OR the guy who was curt but spoke enough to help me answer all the questions I had?

The subconscious forces at work

Market Research textbooks for some reason do not carry more than a chapter on qual research. In the absence of formally documented knowledge (and here i am talking only about commercial qual, not academic) we gain it primarily on the job - by observing senior researchers or occasionally through informal conversations about the dos and donts. I suppose this happens because as a researcher gains proficiency in her craft - her skills in dealing with the complexity of this discipline improve subconsciously.

For instance, before starting to sketch, an artist if asked, can tell you what he is planning to sketch (let's say a leaf), what kind it would be, which colours he would use, his vision of it (a maple leaf symbolising the onset of autumn) etc. However, he would find it difficult to articulate exactly how much pressure he has to apply on his brush for the required amount of paint to stay on the canvas, at what angle he holds & tilts the brush head to get the accuracy of his stroke or how much of the leaf surface area should have the colour green/amber to reflect the changing seasons - since these aspects are part of his subconscious framework.

Likewise a researcher can be told about things like - possible information areas she needs to unearth in a discussion, the tools she has at her disposal... where, how and when she needs to execute those to be able to answer the burning question the marketer has. However for many other aspects it is difficult to lay down rules covering all probable outcomes. For instance what is the right level of comfort for the researcher to ask the respondent about that intensely emotional experience? How should one differentiate between a respondent who is stubbornly silent and one who is quiet, shy & will open up with prodding; between a person who is silent because she simply does not understand the question or or someone who has understood it & is uncomfortable answering it? The decision about whether a 20 minute discussion has covered enough ground on the respondent's habit or will she reveal her all important quirk in the next 5 mins? Much of what a researcher does or how she reacts in such situations is subconscious - often referred to as a 'matter of judgement'

Debunking the myth of being unscientific
Often qualitative research is perceived to be a discipline that is unscientific, lacking a rigid set of rules/structure or sometimes even esoteric. I suppose this perception emanates from the fact that as a discipline it is relatively more flexible, open and encourages deviations and creativity.

I remember how I learnt painting - I would always start of with a blank paper divided into smaller squares. Then in each square i would carefully replicate the required pattern and fill in the colours. Overtime I gained proficiency, I could replicate something without having to sketch it first or follow a grid. I could tweak the colours a bit. People who have mastered this skill create original works of art. If one has to observe such a person, especially in case of abstract work, the artist does not even have a reference for his painting. He starts with a canvas, his paints, brushes and a vision of what he wants to achieve and goes about creating that by putting colour to the canvas in a seemingly free form. Does that mean he does not follow any rules or worse still the discipline of painting because it allows creativity is unscientific and not governed by any principles at all? If that were true anyone / everyone would be able to paint effortlessly! These rules exist - its just that for regular artists the rules of how to balance colours, or create a symmetry between elements of the painting & the overall composition have become subconscious over time.

Likewise for researchers who have achieved a certain level of proficiency, the discipline offers an environment open/flexible enough for them to exercise their creativity in designing/ executing research, analysing & interpreting findings and experimenting with newer ways of representing these findings. Creative liberties are permissible, though grounds rules are followed. If i need to unearth an emotional trigger around a product i may chose to talk to a respondent at a time or a family or a bunch of strangers together depending on my need - though just for the sake of creativity or following a methodology in vogue i will NOT randomly decide to do ethnographic observations instead, since what I can achieve by conversation I cannot in this case understand by mere observation.

Whether it is art or research - the rules may not be always be obvious or apparent to an onlooker - but that does not mean the rules don't exist!

More to come - using the art metaphor to understand 'reliability in research' and what to consider while planning a research project.

Technorati Tags : ,

Feb 20, 2007

Adapting Projective Techniques in Research - Part 2

Taking off from where we left

Going beyond personification and sentence completions....
We were asked to design a study that involved understanding the brand equity of an auto-mobile brand. The 1st challenge was that - the brand in question was not one of the leaders in the category and therefore its salience and involvement with consumers was relatively low. The 2nd challenge was that part of the target audience comprised teen age girls - whose involvement with the brand was even more tenuous.

Using Logos as catalysts for generating brand associations
Instead of using the standard set of projective techniques we decided to experiment with some new ones. While we were designing that research - there was a logo quiz doing the rounds in our office. Most of us found that quiz addictive since irrespective of one's involvement or actual usage of a brand - logos are ubiquitous symbols that we are exposed to day in and day out. One may not spend too much time trying to read meaning into a logo design and how that is reflective of a brand but one does form associations around these logos - associations that are sometimes subliminal. So we decided to take that online logo quiz into the real world of our research.

- We used it specifically with the older males
- The exercising involved guessing the brand name from the logo. This got people pretty involved in the game - it was challenging and fun though we used this just to warm them up to the idea
- Post the guessing game we zeroed in on the 3 - 4 brands that mattered to us and got them to reel out associations they had about those brands based on their logos.
- From the data we captured - it was evident that did not merely talk about the brand based on its logo - rather the logos acted as stimuli that triggered / unearthed thoughts that they had about these brands. Things they saw in the logo sparked off thoughts about their experiences around these brands that they would then narrate to us.

Taking cues from Popular culture

Movies and music from movies have for long been quite a vibrant part of the pop culture in India. Each week a dozen new movies or more are released and viewed. People discuss movie actors, their lives, their costumes, the songs...so much so that even some of the national news channels in India now have regular coverage on this. There is a film song for every situation or so it is said and it is not unusual for people to recall them and break into a spontaneous session of song and laughter. So we decided to include a game of songs in our research

- Designed for the teen age girls - this tool worked well - since such games are common in college The girls had to work in small groups and come up with the songs they thought characterised the brands
- The groups helped in ideating. The songs - since those were part of a larger context of a film carried some residual meaning and association with them - which helped the girls articulate what they felt about the brands which were being spoken of
- We rounded up the exercise by asking them their reasons for choosing specific songs and getting reactions from other groups on songs chosen by a group.

Were these experiments an efficient way of gathering information?
Did these exercises help us get information we were seeking - Yes and in hindsight I can say that it was in a manner that was intuitive keeping in mind the audience and therefore they seemed to talk about these brands with relative ease. To that extent we achieved 'validity' in research. When we compared the set of brand associations across groups that we derived from these two and some more exercises we used in that research - we found consistency amongst them and to that extent the goal of 'internal reliability' was achieved as well. Would this have been possible using standard techniques? I don't know - since we never had a matched sample where we used the standard projectives - though I can say that we did not lose anything as a result of trying to experiment with newer ones and it only turned out to be easy and fun for the respondents

The question of adaptation
Projective techniques have basically been adopted in research from another discipline. Research was never the home ground for these tools. Also in their adaptation and use, we have not over time formulated norms against which findings are benchmarked or analysed. Analysis and inferences mainly happens by way of triangulation (comparing findings from multiple techniques), juxtaposing the data from these tools against what is said in the general group discussion and by understanding from the consumers themselves the motives for their responses. If I assume my understanding of using projectives in research is not too far from the mark then - the question in my mind is - why do the few that we originally inherited from psychology continue to dominate research even today? Based on some ground rules would it not be possible to design newer techniques that are specifically suited to market research and / or the target audience being researched?




Projective Techniques,Projective Techniques and qualitative research,qualitative research,

Feb 15, 2007

Adapting projective techniques in research - Part 1


Projective Techniques - Why do we use those in Research

Originally used by psychologists to understand the subconscious aspects of one's personality - projective techniques involve getting people to respond to ambiguous stimuli with the assumption that it 'projecting' their inner thoughts and feelings on to the stimulus would be less threatening / intimidating than speaking about those directly

These techniques have found their way into market research a) As way of tapping into the consumer subconscious and b) an alternative and often easier way (v/s direct questioning) to get consumers to express themselves. In the 1st application these tools allow 'projection' of repressed subconscious thoughts and in the 2nd case they are meant to facilitate / enable the articulation of thought (may / may not be subconscious) otherwise difficult to express .

What to consider while designing research involving Projective Techniques

The use of projective / enabling techniques in research can be fun or NOT depending on which ones are used, how many are used in a given research session and how those are incorporated and executed. I have sat through some research sessions which only involved executing one technique after another - for 3 hours. Like anything else stretched beyond the limit, it left me with a bunch of tired and sometimes even angry respondents waiting to charge out of the research facility. There is a another particularly disastrous interview that comes to mind - with an oldish man who was quizzed about computer peripherals. I had to first ask him to personify the product and then the brands (yes not one but 3) in the category. The man could not conceive computer parts / brands as human beings to save his life. I could only empathise with him ! (In my defence I did not design those studies :)

Which One to use - These techniques becomes counter-productive if we do not keep into mind the audience we are talking to, while designing the research. An older audience may not respond well to the idea of imagining products / brands as human beings - a concept that a younger audience may relate very easily to. Similarly I have often found housewives who have been out-of-touch with 'writing' find it very difficult and sometimes even embarrassing to fill out their responses in writing. Talking comes more naturally to them as we know only too well :) The output of activities such as completing thought bubbles, sentence completion, or even simple self completion questionnaires end up as half completed sentences amidst a paragraph full of scratched out words.

How many to use - When used intermittently with discussions they sometimes act as energizers besides fulfilling their primary purpose. But when someone decides to do a marathon run with executing 10 techniques in a row - then its a case of greed and foolishness.

How to execute - The 3rd factor to watch out for is how is one actually going to execute the tool. In everyday life people do not go around filling out thought blurbs, playing word games or imagining their toothpaste as a human being. Although these techniques are meant to 'facilitate expression' we often forget that switching from our normal mode to doing what these techniques expect can pose a barrier to expression in itself. It helps therefore if people are not expected to respond to a cold question, if the researcher leads by example, if the activity is made to be a little bit of fun rather than an exercise that expects the person on the other end to perform.

I have learnt that part of the ease of expression happens because of the mood / frame of mind people are put in and if they are able to relate to the activity they are asked to do, intuitively.

As much as I've been part of disastrous research , there have also been some studies which have been sensitively designed with some interesting adaptations to the use of projectives - which was the reason why I started writing this post today. I'll save those anecdotes for part 2



Projective Techniques,Projective Techniques and qualitative research,qualitative research,