Image for post
Image for post

Triangulation: Beating the Bias

In its simplest form, Triangulation means using more than one research discipline/methodology in order to eliminate as much bias as possible.

In an ideal world every market research design would be comprised of multiple methodologies, layered upon each other to explore, test, refine and validate findings to ensure the original Business Objective is met. However, as with every ‘ideal’ in life, this would most likely take more time and resource than a business has available. Therefore it is essential to implement the theory of Triangulation from the very start of a research design to ensure the correct combination of methodologies is used to maximise validity whilst meeting your clients’ time and budgetary constraints.

In my experience you often get approached by a company with a pre-conceived idea of the methodology they wish to implement to get the results they desire. I’ve lost count of the number of times I’ve heard ‘We’d like to run a survey to tell us what we should include in our offering…’ Given the pace of today’s business climate, many companies require the validity and confidence that they believe a quantitative survey, with a robust sample size, would provide in terms of %’s. It is therefore our task, as researchers, to suggest alternatives, often a multi-discipline approach that is more likely to meet their business objectives — Triangulation!

Every methodology has its own merits, however combining them provides a result greater than its individual parts. Whilst a survey can give you strong statistical results that, for example, can test business hypotheses or gather information, who’s to say the hypotheses tested were the correct ones in the first place? Where is the context? Conversely, qualitative methodologies are fantastic at providing detail and reasoning, however the chances are slim, despite your very best efforts, that your group sample composition truly reflects the target market.

Even within different qualitative methodologies there are pros and cons. Focus groups provide a great platform for group insight, bouncing ideas around and gaining context, however they can be biased (especially in a face-to-face situation) by over powering or overtly shy personalities. Contrast this to a one-on-one depth interview where the potential for personality bias is removed but there is no facility to generate the effective idea generation that is achieved in a group situation.

There is also the consideration of the researchers themselves. Whilst bias can be created by the implementation of a singular research methodology, so too can using a singular researcher on a project. The assignment of multiple researchers to a suite of methodologies within a project greatly reduces bias. The process of collaborative analysis is a fantastic way to eliminate any sub-conscious bias a lone researcher may reflect in interpretation of the results and any recommendations made?

Throwing technology into the mix muddies the water further. When considering your research design you must now weigh up the merits of online qualitative methodologies vs more traditional face-to-face methods. An online focus group is much quicker (and often considerably cheaper) to implement and has huge benefits attached to it. No longer is geographical location a consideration, participants simply login, no more travel time for participants or clients, thus resulting in stronger levels of participation. Another selling point is the removal of bias from ‘strong personalities’; participants have a feeling on anonymity and are therefore more likely to be more honest.

But then some things are lost… including a researchers ability to analyse the un-said, for example participant body language and facial expressions. These don’t always marry-up with what participants are saying. Technophobes may self-exclude themselves from such methodologies and slow typers my struggle to get their point across at the speed required in online focus groups, thus creating bias.

I could go on and on, debating the bias created by each and every research methodology available but don’t worry I won’t, I think my point has been made — Triangulation of methodologies is key to reducing bias!

As is the rule with all research design, the sole aim is to ensure that the business objective is met, results are valid, and interpreted in a way that ensures they can be strategically implemented in context with confidence. The reduction of bias is key to ensuring you maximise the accuracy of your judgement in interpreting results.

Triangulation must be implemented from the very start of the research design. By using more than one researcher to create the research design you can limit bias from the outset, as the old saying goes ‘two heads are better than one!’

Obviously the nature of the research will determine the relevance of every methodology you consider and there are many iterations of approaches towards the implementation of Triangulation, however the following diagrams demonstrate simply the benefits of a multi-disciplined approach to research.

1. The parallel approach: an implementation of multiple complementary methodologies

Image for post
Image for post

2. Layered approach: a structured space to develop, refine and test ideas/concepts

Image for post
Image for post

3. Dimensions of triangulation

Image for post
Image for post

Written by

We empower brands to inform every decision at the speed of business by delivering on-demand insight and enterprise grade research technology.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store