I have long argued that the difference between qualitative and quantitative research is more about what a researcher does with the method & data rather than the method itself. One key difference is the amount of structure in gathering data. Open-ended methods such as interviews and observation are unstructured ways of gathering data. On the other extreme, surveys etc. are close-ended. Respondents can choose from a few given options.
In the middle are semi-structured methods like card-sorting, freelisting etc. which can be used for either qualitative or quantitative research.
The data collection method needs to be distinguished from the data-analysis method. It is possible to gather unstructured data from methods like interviews / observation, then turn around and apply quantitative methods to the analysis.
One way to get the best of both types of methods is to use a hybrid of qualitative and quantitative research, using each to its strengths. This is becoming more common – in many conversations, I have heard user researchers, anthropologists talk about using a hybrid methodology, and the best way to marry qualitative and quantitative approaches. I personally believe that hybrid methodologies are superior because they keep in mind both the needs of the designer and the business stakeholder. The designers want rich textured insights best derived from qualitative approaches, business stakeholders do want to bet their products on a designers insights, they want some way of verifying that those insights meet a broad market demand.
We kept this in mind with MindCanvas and created a flexible methodology suitable for both qualitative or quantitative research. Qualitative for us means watching a few people do the research exercises, while quantitative means sending out the study url to lots of people for data which will be statistically analyzed. Early reactions to the system show that people perceive this flexibility of the system and appreciate its hybrid usage.
For qualitative insights – you can observe someone do a sorting or a divide-the-dollar exercise, while doing a think-aloud. This can be followed up with an interview at the end. For quantitative research, er simply send out a url and let respondents complete the exercise on their own. I find that its most effective to do some qualitative research before and after the quantitative exercise. Before the quantitative exercise, it helps almost as a pilot, to refine the research questions, make sure people interpret it in the way the researcher intended.
After the quantitative phase, it serves as a way to better understand the quantitative data. For example, you might have noticed that for a cell phone divide the dollar exercise, people who care about mp3 players also care about world ringtones. You might want to explore this relationship and understand it better in a qualitative manner.