Why mixed method is preferred in social as well as health science research nowadays is because, it can give more clear and real answers of the research question (s). It might have other more benefits as well.
In much of the research on the human condition, whether it focuses on health or other aspects, such as economic processes or religious beliefs, it is advisable to use both qualitative and quantitative methods, because each provides diffferent kinds of information about the topic at hand. In qualitative research, the perspective tends to be up-close, holistic, and capable of characterizing process. It also teaches us how to ask the most relevant questions about the topic, ar as Agar says, "the hip questions." Qualitative investigation cannot, however, tell us how prevalent a behavior is, or how generalizable a pattern of behavior is. That is the province of the quantitative inquiry, which usually relies on short answers administered to large samples of respondents. The most productive way of joining the two kinds of inquiry involves "front-loading" the qualitative study so that it can inform the formulation of the quantitative interview instrument. Both kinds of research have to be as efficient as possible, and both have strategies for assuring efficiency - quantitative by means of power analysis and qualiative by means of saturation of categories (grounded theory) or redundancy (ethnography). This is just a thumbnail sketch of a series of lectures on the question of qual and quant.
You can only use the method that is appropriate to the question! Thus a quantitative approach is sometimes entirely inappropriate where a qualitative response is required. For example when one is asking 'how many times do you clean your teeth in a day?' a quantitative reply is required. When one asks 'why do you clean your teeth'? only a qualitative response will embrace ambiguity and meaning!
In reply to John Page - i would question the whole knotty problem of 'generalizability'. Does quantitative work really produce an answer that is 'generalizable'? The whole world? Every corner of society? Does it ever address the concept of CONTEXT? Qualitative work can be perceived as generalizable to certain contexts, although this statement is bound to be refuted by quantitative methodologists! Again, I would reiterate, each answers its' own question, only to be refuted (or not) by future work . But make no mistake, qualitative work can stand on its own. (From someone who uses both methods of research!)
The canons of sampling help to decide how generalizeable a quantitative result is. If you have a good grasp of the universe (good enough to give all units an equal chance of being in the sample) and have a well framed probability strategy for sampling, then your results will be generalizeable to
(I ran into an uncooperative prompt) that particular universe that you have sampled. Qualitative inquiry has implications of generalizeability, but one can't be sure until one has engaged in quantitative follow-up. Qualitative inquiry can provide a quantitative study with some aspects of context and some hypotheses about process, I absolutely agree that both qualitative and quantitative inquiry can stand alone.
It was a nice discussion. I appociate everyone's opinion. I feel that I got some useful information from this discussion. As other things have, mixed method has also pros and cons. However I think mostly quantitative data can ba well interpreted with the help of qualitative information and qualitative information will be objectified by the help of qualitative data. Therefore, these two methods are found cooperating each other rather than contrasting. But, we should be careful during selecting methods. I think common sanse and logic should necessarily be used where we are selecting mixed or a single (qualitative or quantitative) method. If someone use mixed method unnecessarily could be no more than waste of resources.