Research methodology

[luke77]

Hi guys,

I've been debating whether to go forward with a PhD in business for the past few months, and I'm wondering about research methodology in fields like accounting, management, and finance. To clarify, I'll describe my background in research. I have a science background, and research in the sciences pretty much follows the scientific method: a testable hypothesis is formed, and then you experiment in the lab to determine whether the hypothesis is supported by evidence. Thus, "research" is done by performing repeatable experiments.

So my question is, how is research performed differently in a field like management or accounting? In reading over publications in these fields it's pretty clear that financial research is performed in a different way than scientific research. In other words, what do business professors "do" to create new research? In a field like marketing I can see how you could perform surveys, get people in a lab, etc. to gather data and generate conclusions based on this data, but in the more technical fields I guess I have no idea how research is actually performed. If this is something I'm thinking about doing for the rest of my life, it;d be nice to know what business Phds actually do. Thanks for any responses.

Luke

[desimba]

There are 2 kinds of research which people do in business and economics. One is a more theoretical research which is a claim regarding how they believe the world works, starting from first principles. To take an example from game theory (in microeconomics), a theorist could develop a model which tries to explain how the game would evolve if two players were interacting under a set of circumstances, contingent on some assumptions he makes in his model . That may or may not be exactly how it works in the real world but as long as what he is claiming is reasonable, i.e. he makes reasonable assumptions, he would be seen as pushing the frontiers of the theory. Another example might be a macroeconomist speculating on how economic growth varies with inflation and whether level of capital investment in the economy influences that or not. He would be asserting something based on prior theory and on additional insight that he brings to the problem, his value add being that last bit of additional insight that he is contributing.

The second strand is empirical research. Here the researcher is not concerned a whole lot about pushing new theory but more interested in testing out what his already been theorized in the literature. He would (similar to science research) set up a hypothesis and then (almost always) using secondary data which is available from different statistical organizations such as the Census Bureau, US Deptt. of Commerce, try to draw a relationship between the two variables that he is examining. In the previous example where he is interested in economic growth and inflation, he might take the last 50 years of data for all US states and see if he finds a positive correlation or not among these two variables. This strand of research uses econometric techniques to make sure that the relationships which are being derived are robust to different specifications, say a different time frame, a different but related definition of how inflation is measured, etc.

Within empirical research, there is a sub strand which uses what is called natural experiments. Here researchers use events which have happened to test their hypotheses. e.g. The collapse of the Berlin Wall allowed researchers to test hypotheses regarding how two geographically proximate regions might develop once they are part of the same sovereign nation. Note that this (like almost everything that is non trivial in econ) cannot be tested in the same way as I would in science by going to a lab as we do not have that luxury of putting two nations together / breaking them into two so that we can study what happens when we do that! The way we get to it is by the use of natural expts. as I just mentioned.

Finally within econ, there is another approach which actually uses the laboratory setting and that is known as experimental economics. Various hypotheses of game theory regarding how individuals interact when they are playing a game with uncertain payoffs can be tested in that setting.

While all of what I have said is somewhat specific to econ, I would also say that it holds true for business in general, the differences being that there is a) less pure theoretical research
b) more empirical research using secondary data &
c) some empirical research that uses primary data generated through surveys and questionnaires.

[bauble]

Originally Posted by luke77

Hi guys,

I've been debating whether to go forward with a PhD in business for the past few months, and I'm wondering about research methodology in fields like accounting, management, and finance. To clarify, I'll describe my background in research. I have a science background, and research in the sciences pretty much follows the scientific method: a testable hypothesis is formed, and then you experiment in the lab to determine whether the hypothesis is supported by evidence. Thus, "research" is done by performing repeatable experiments.

So my question is, how is research performed differently in a field like management or accounting? In reading over publications in these fields it's pretty clear that financial research is performed in a different way than scientific research. In other words, what do business professors "do" to create new research? In a field like marketing I can see how you could perform surveys, get people in a lab, etc. to gather data and generate conclusions based on this data, but in the more technical fields I guess I have no idea how research is actually performed. If this is something I'm thinking about doing for the rest of my life, it;d be nice to know what business Phds actually do. Thanks for any responses.

Luke

I don't know that there is any notion of a "repeatable" experiment as you would see in the physical sciences, i.e., if you drop a ball in a vacuum, it will take the same amount of time every time you do it. Think of all business research as fundamentally an offshoot of the social sciences in which you are studying the most fickle of all things - people. Once you really consider this carefully and think about its implications for the types of questions that business fields are trying to answer, then the methodologies make sense, in my opinion. So I'd focus more on the questions that researchers in the various fields are trying to answer in deciding if this line of work is a good path to take.

[luke77]

Thanks guys, for the insightful responses.