Reasons to NOT use mix-methods, why you should & protecting integrity of the data
Hello everyone, as an applied mix-methods researcher, I am often asked similar questions to the ones below:
- What are you more of qual or quant?
- In our project, we are doing x, y and z. Are mixed-methods researchers competent on taking on (x) qualitative need //OR// (x) quant needs? I am afraid if the person is “more of” than the other.
I fully listend to those needs and concerns. Yes, real practitioners of mixed methods research will be a great addition to your team. We, by default are trained and competent in both. Our trainings & careers are heavy witth both inductive and deductive thinking — know for sure we will never stop growing.
Word of caution, there are “researchers” with unstated reasons to utilize mix-methods that will compromise the integrity of the data (keep reading this post to learn more)
We are primarily hired because of our vast understanding of how to conduct research in different methodologies. Understanding how, it’s just one piece. We need agility to effectively switch and adapt research needs with very different ontological and epistemological points to address organizational and people’s needs. We need creativity to be able to visualize through patterns and think through long-term impact to deliver on-point innovative solutions. (See. An Ingredient to Innovation are Creative Mixed Methods Researchers)
Reasons you should NOT use mixed-methods
As you are probably aware, conducting high quality research takes time and not everyone understands its process. Here is a list of things you should NOT use to justify the use of mixed-methods research:
❌ Time does not permit in-depth qualitative/quantitative research
❌ Having access to participants/users is challenging
❌ Finding a large data set is challenging
❌ We need to use/apply/integrate/show numbers in some way
❌ We need to show we talked to our target audience
❌ We aren’t sure if we will have enough budget in [ insert timeline]
❌ Lets cover as much terrain as we can
Those are very real and valid practical concerns both researcher and non-researchers all-together will face as they collaborate through any project.
MYTH:
There is a huge misconception that quantitative research is a lot quicker than qual. Quantitative researchers may have already access to some data (eg., using Census data) but oftentimes they will find that a BIG data set that focuses on your exact research question does not exist.
Making good data takes time, whether it is qual or quant.
Reasons you SHOULD use mix-methods
The real reasons to justify our use mix-methods are beyond “practical” challenges. In his summary of mixed methods, Bryman (2006) outlines five methodological reasons for mixing methods:
✅ Complementarity: Deepen or enhance other data
✅ Expansion: expanding the inquiry to ask different questions
✅ Development: use one method to inform and improve the other
✅ Triangulation: corroboration of earlier data
✅ Initiation: resolving earlier contradictory findings
Protecting the Integrity of the Data
In the same study, Bryman notes, that many researchers who mix methods leave their reasons unstated. In his review, Bryman found a surprisingly large number of studies that did not mention why they chose to mix methods at all. Very few used “initiation,” or attempted to understand why their previous findings were confusing or inconclusive.
He cautions strongly that researchers who do not state a methodological reason for mixing methods risk mixing for mixing’s sake — and ultimately undermining the quality of the insights uncovered.
As a way to protect the integrity of the data I encourage you to take interest on the methods and ask why a particular method was chosen over the other one. This will not only protect the integrity of the data but will also bring along many other benefits like transparency of the data, generating high quality data that answers your target research questions, builds better organizational process, and so much more!
Ending Note
No matter the case, high quality applied researchers chooses mixed methods to enhance their overall research findings, rather than simply responding to a practical problem like lack of time.
If you are working with applied researchers, they will be transparent and should let you know the implications and challenges of the study, at what point of the study the data may be at a challenge, they will suggest ways to minimize bias or anything that will put your data at risk. This will yield to collective understanding of the limits and future iterations needed as you move forward with your project goals.
