|
|
 |
|
|
Qualitative Research
|
|
|
It is common for
researchers to make a clear distinction between quantitative research that
focuses on scientific sampling and the analysis of numerical data and
qualitative research that focuses on less rigorous (or convenience-based)
sampling and garnering textual information for non-statistical analysis.
Often, however, more mixed approaches are used in which both quantitative and
qualitative approaches are blended and used to inform the findings of each
other.
The advantages to
quantitative research are that well-selected, relatively smaller random samples
can be used to make generalizations about larger groups or populations. By
using strictly defined measurement approaches to assign numbers to observations,
it is also easier for other researchers to replicate findings and to confirm or
refine them in future research. The data we obtain from quantitative
research has the advantage of being amenable to highly sophisticated statistical
analyses and modeling procedures that can uncover interesting and important
relationships that are not visible to the naked eye. These are but a few
of the strengths of well-conducted, qualitative research.
Qualitative researchers
note, however, that quantitative approaches may be more prone to biases and
limitations in the knowledge base of the researcher. By restricting not
only the questions we ask but also the availability of response alternatives
(e.g., by using numerical rating scales), we are less likely to gain new
insights from the research participants that we work with. Qualitative
research, in contrast, seeks to gain more highly detailed responses in the
subjects' own words, rather than limiting them to responses (or even topics)
that we have predefined within our numerical measurement systems.
At GuideStar Research, we
agree with those who see the value in mixed models, where the focus of research
may shift or differ depending both on the nature of the research question that
we are asking and the nature of the people that we are asking it of. Some
business samples are too small to provide us with scientifically valid sample
sizes, yet too important to ignore (e.g., true strategic partners). And,
some topics are critical, but not yet sufficiently well understood to allow us
to reduce them to a small set of questions for which we can assign meaningful,
numbered response alternatives for the respondents to select from. These
are two instances in which we might suggest using qualitative approaches, such
as structured or semi-structured interviews and focus groups, to collect
information that may be vital to understanding your key business relationships.
We also suggest that some
of the distinctions between qualitative and quantitative approaches are somewhat
artificial constructions necessitated solely by the need to draw distinctions
between these two approaches. In practice, we usually obtain qualitative
responses to open-ended questions when we collect quantitative survey data and
even when we use qualitative methods in interviews or focus groups, we will
generally also seek to quantify responses in some areas or ask some questions
for which we solicit a quantitative response (e.g., "How highly would you rate
this company on...?"). Accordingly, although some of our research is more
clearly in one camp than the other, most of our research incorporates both
quantitative and qualitative aspects in collecting information.
We also believe and
recommend considering blending the two processes in an iterative, cumulative
learning approach whereby we use qualitative information to assist us in
selecting survey content, include both quantitative and qualitative items on
surveys to explore the views and intentions of respondents, and engage in
follow-on qualitative or quantitative research to further and refine our
understanding of the data that we obtain using predominantly quantitative
surveys.
Qualitative input is
especially useful in helping to select areas for surveying. Although the
combined experience of our clients and our staff can help to identify likely
dimensions and items for many different types of surveys, the input gained
through semi-structured interviews or focus groups conducted with key
participants in the business relationship often helps to surface important new
areas that warrant inclusion and investigation. In addition, the
qualitative responses on our surveys and the learning we gain from the
quantitative data we collect often reveals new issues that warrant further
understanding that can best be gained through additional qualitative research.
The learning that results from going back to key stakeholders to flesh out
complex issues and topics can then be used to refine subsequent surveys to
collect additional quantitative data and to generate more sophisticated
statistical models of how various critical factors interact in producing the
outcomes our clients seek to improve.
Here is a case in point.
Several years ago, we began a large survey program for a multi-national company
that is among the leaders in its field both in the breadth of its products and
its corporate customer base. We consulted with the client extensively and
conducted interviews with key people in the organization to draft our initial
survey of their customers. When we statistically modeled the findings of
our first survey, it was clear that the notion of "value" was a key intermediary
between many dimensions of satisfaction and customers' likelihoods of continuing
or expanding business with our client company. We then conducted focus
groups with senior staff in customer companies to better understand how they
defined and evaluated "value" within the context of relationships with providers
of products and services like those they obtain from our customer. We also
sought and obtained their impressions of how our client company compared to
other vendors that they deal with for these types of products and services.
This new learning was then taken back to senior leadership for further
discussions and we used the knowledge gained from all points of view to refine
the questions that we asked about value on the next survey (including adding an
open-ended question on the topic to gain further quantitative data). In
the following year, when analyzing the new survey, we replicated many of the
findings of the prior year and were able to advance our statistical modeling to
gain new insights. In that year, we learned that different aspects of our
client's relationship with their customers had different impacts on both value
ratings and customer loyalty. We noted, in particular, that customers were
distinguishing between what could be best described as the service
relationship and what could be characterized as the strategic
relationship that they have with our client. A series of semi-structured
interviews followed with senior representatives of customer companies and we
advanced our knowledge of this distinction further. Since that time, our
clients have focused on promoting strategic relationships with their large
customers and it has become a major focus of their sales training, customer
support, and our ensuing surveys. We are now continuing this process,
seeking to learn more about how partnering and value can be enhanced and we are
tracking progress in these areas with each new round of surveying.
While clear distinctions
can be made between quantitative and qualitative approaches to research, we
believe that a blending of the two is needed for a full understanding of
business relationships. This can help both to subvert the potentially
negative consequences of pre-existing biases that shape survey content and to
flesh out the most interesting and important findings obtained when collecting
quantitative survey data. When time and resources permit, partnering with
us to engage in an ongoing learning process is often your best path to
understanding the relationships that drive your business forward.
|
|
|
|
