"Are you a qually or a quanty?" is a question I've been asked a countless number of times in my years as a market researcher. But recently, we've seen the emergence of lots of new methods which have blurred the lines between qual and quant, for example the use of image based online discussions where you can achieve both in depth qualitative feedback, as well as instant stats and sentiment maps.
With this overlap, more and more agencies are looking to recruit 'all-rounders'. But in the process, have they lost sight of the skill sets that are essential to demonstrating expertise in both types of research? I would suggest it's better to play to individual strengths, to create a team of experts, ones who specialise in different areas with capability in others. In my experience it is this approach that provides the collectively exceptional, creative, agile team of insight professionals you need to succeed. Put simply, don’t dilute your agency expertise with generic research skills only.
| Tweet This | |
| Jack of all trades. master of none - Don't dilute your agency's market research expertise |
Here, I'd like to outline some of the specialist skills and traits that every research agency should be able to demonstrate in terms of qual and quant...
Fundamental to qualitative research success is the ability to uncover the values and attitudes hidden behind the feedback elicited from customers during the research process. Let’s take a look at a few of the skills necessary for achieving just that.
Design
With a quant study, the design and questions are agreed by the researcher prior to launch and every participant is asked the same thing - a bit of routing aside. Qual design on the other hand is much more fluid. Its great practice to prepare qual research materials in advance of fieldwork, but awareness and the ability to adjust based on how the participant/s respond to questioning is key. After all, it isn’t until the responses start coming that the researcher can establish which areas represent a problem or opportunity for the client, which are of most importance to the customer, which require probing and so on. Operating on the fly like this requires an inherent knowledge of qual techniques, group management, prompting and so on plus a solid understanding of the business in question.
Preferred researcher would be:
Moderation
In qualitative studies, the researcher will most often share the ‘research space’ with participants. They are more heavily involved in the direct data collection process than in quant. Therefore, every agency should have moderators who can speak the customer language, who can really engage with them to deliver the detail required for the research, who know when and how to ask the fundamental ‘why’ questions; and who can remain objective throughout all. The communication skills of the researcher are essential for depth without bias in qual. The researcher is the data collection tool. Their skills are imperative to the relevancy of the data captured and its quality.
| Tweet This | |
| In qual research the researcher is the data collection tool - Their skills are imperative to data quality |
Preferred researcher would be:
Analysis
Analysing qual can be a 'messy' process - whilst there may be best practice guidance, there is no hard and fast rule. Effective qualitative analysis lies in the researcher’s ability to take large volumes of unstructured data, which can be text, audio, visual or audio-visual, and transform it into meaningful insight. They must be able to sift though, identify significant themes and interpret. The end result must address the client’s objectives, whilst also highlighting additional relevant insights that may not have been considered at the beginning of the project. Ensuring outcomes are translated back to the client in a meaningful way requires a significant level of creativity and judgement from the researcher.
Preferred researcher would be:
In quantitative studies, the understanding of statistical techniques and the ability to validate and interrogate numerical data for actionable insight is a key focus. Let’s now look at some of the specific skills required for this.
Design
When designing quant, a more structured approach is required and getting your design right before you launch is paramount to the reliability of your results. There isn’t the same level of flexibility in quant design as in qual - you can’t adjust your questioning during fieldwork, nor do you want this. The whole point of quant is to validate or otherwise attitudes, opinions, behaviours, and other predefined variables which is impossible if questions are changed inter-study – your results will be skewed at best!
To get your quant design right the first time around, the researcher or researchers’ responsible need to be capable of both big picture and detail orientated thinking. They need to look at the client objectives and visualise their analysis then dip into the survey design and question detail accordingly. I have previously written about some of the key characteristics of good survey design - ensuring a good survey length and structure and choosing the right question types and wording. Knowledge of different questioning techniques is an absolute.
Preferred researcher would be:
Sampling
You’re live with your survey – what next? I sometimes feel there is a misconception that the researcher sits back at this point and waits until sampling is complete. Not at all! At this stage, the researcher’s project management skills come to the forefront. It’s their job to ensure the quality of the data coming in is to an acceptable standard, that quotas and completes are carefully monitored and that ‘sample’ is achieved within the specified time scale.
The researcher’s ability to spot poor quality data is paramount to ultimate insight quality. They should immerse themselves in the data and know their survey inside and out. Speeders (those whizzing through the survey much quicker than expected), flat liners (those picking the same point on every scale) and extreme response bias (those who only go for the extreme ends of the scale) must be identified, excluded and replaced with considered participant completes.
Preferred researcher would be:
Analysis
The analysis of quant is much more structured than that of qual but distinct skills are still needed in order to filter and interpret data for meaningful insight. This doesn’t mean running each and every question through an analysis program and providing the client with every single stat.
Rather, the researcher should have the ability to interrogate the data output, both with respect to the client objectives and independently, identify correlations and link findings together for meaningful insight. They should be able to tell the story, not present a sea of numbers. Technical ability in terms of survey and digital analysis software is also key here – knowing what types of analysis can be conducted and technically how to do it.
Preferred researcher would be:
As well as their research abilities, all agency team members should be self-motivated, have strong social skills and a desire to constantly improve. The team must remain supportive of each other and professional even under pressure. It is equally important that your research agency management have a full and thorough understanding of the team’s abilities and specialisms in order to allocate work effectively for quality insight at optimum speed.