Exploratory research is the broadest and least theoreticaly-grounded of the three types of research. Essentially, you have no starting theory or hypothesis about a probem. In fact, you don't even know if there is a problem to begin with. In exploratory research, you simply start collecting information about something. It is similar to data mining, and it does not always produce results that can be useful. However, it generally gives you a better idea about something.

Descriptive research is, in a way, a higher level than exploratory research, because now we actually want to know something specific in nature and we then go about collecting that information (that is, we have a known set of parameters beforehand that we want to collect data about). Say we want to collect data about NASA astronauts' characteristics and we measure their height, weight, IQ, etc.. These measurements give us a clear idea or picture aboout the astronauts. We collect specific types of data that better describe an object, a person, or some phenomenon, but we know beforehand what we want to measure. We don't have a theory yet, however. For example, we don't theorize that NASA astronauts have higher IQ's than the rest of the population because they eat more cookies (a silly theory, of course).

Causal research uses data that we know we want to measure beforehand, so it is similar to descriptive research in some ways. But now we're not only collecting specific data, we want to prove (or disprove) a theory. Say we feel that people who sleep longer at night have better productivity at work. We expect to find that there will be a correlation between longer sleep at night and better work performance. We will collect data and model it in such a way that our theory becomes the most likely, the most plausible explanation. The biggest danger of causal research is to assume that if we find two things