Feb 25

Is Grounded Theory an Option with Big Data?

This week’s readings really harp on the importance of theory and the scientific method. While I agree with the authors for the most part, I’m wondering if theory is always necessary. Perhaps there are some phenomena that are new and/or don’t fit nicely into an established theory. In these cases, a grounded theory approach may be useful. According to the Grounded Theory Institute (http://www.groundedtheory.com),

“Grounded Theory is an inductive methodology.  Although many call Grounded Theory a qualitative method, it is not.  It is a general method. It is the systematic generation of theory from systematic research.  It is a set of rigorous research procedures leading to the emergence of conceptual categories.  These concepts/categories are related to each other as a theoretical explanation of the action(s) that continually resolves the main concern of the participants in a substantive area.  Grounded Theory can be used with either qualitative or quantitative data.”

Grounded Theory is a research methodology which directs the researcher on issues concerning data collection and describes procedures for data analysis. With Grounded Theory, the researcher doesn’t have any a-priori hypotheses about the phenomenon. Rather, the researcher develops a theory from the ground up via extensive data analysis.

I think Grounded theory is useful for new phenomena or in situations in which using an existing theory would be like putting a square peg into a round hole. I remember reading an article that used Grounded Theory during the dissertation phase of my Ph.D.  This article investigated the political and rational roles of a systems development methodology. I found the article relevant and interesting and actually cited it in my own work.

Overall, while I strongly believe in using theory, I also think the importance of theory needs to be balanced. There are some phenomena that are new or don’t neatly fit into established theory; for these phenomena, a Grounded Theory approach may be useful.

Feb 24

The pragmatic importance of a good theory

In “The End of Theory: The Data Deluge Makes the Scientific Method Obsolete”, Chris Anderson argues that theory may be irrelevant in a world of big data:

“This is a world where massive amounts of data and applied mathematics replace every other tool that might be >brought to bear. Out with every theory of human behavior, from linguistics to sociology. Forget taxonomy, >ontology, and psychology. Who knows why people do what they do? The point is they do it, and we can track and >measure it with unprecedented fidelity. With enough data, the numbers speak for themselves.”

When I first started by Ph.D. program ten years ago at Virginia Tech, I would have argued that theory is irrelevant; that, if the data gives the result, then it’s not important to know why. This mentality reflects a very practice-based or industry-based approach to research. I used to believe that organizations don’t really need to understand why a best practice works, just that it works. Leave the why to the ivory tower academics. After doing research for ten years, I look back at myself from 10 years and think my line of thinking is folly. Theory is important. Theory tells us the why. Theory provides boundaries for understanding a phenomenon. Without theory, we don’t know why a statistically significant result holds true. If we don’t know why, we can’t tell whether or not this result will hold true in another scenario with different conditions or even a future scenario with the exact same conditions. Without knowing why, research can’t tell whether the same result will happen again in the future.

Feb 18

The Importance of Critical Mass

During yesterday’s domain of one’s own class, I started thinking about network technologies. Network technologies increase in utility as more people and more people use them; that is, the benefits of using network technologies increase as they gain critical mass. A perfect example of a network technology is the telephone. If I’m the only person using the telephone, I can’t talk to anyone else using my phone, so my utility from using the telephone is fairly low. As more and more people use the telephone, I can use my phone to talk to others, thereby increasing my own utility from using the telephone.

So, how is this relevant to social media instructional technologies (e.g., twitter)? I think social media technologies can be perceived as network technologies. If I’m the only person blogging or using Twitter, the benefits of using these technologies aren’t that great. On the other hand, as more and more of my colleagues and students use these technologies, the benefits of using these technologies increase for me as well as for my students and colleagues. As more of my colleagues and students blog and tweet, I can learn from their blogs and tweets and the more others who read my blog can learn as well. Essentially, the benefits of using these technologies will increase as they are more widely adopted. So, please continue to blog and tweet.

Feb 10

Good learners make good teachers

When I was working in industry before I decided to pursue a Ph.D., a good friend of mine sarcastically told me that I would make a good teacher because I like to talk about things that no one else cares about. While that may be true, I more strongly agree with Alec Couros that the best learners make the best teachers. Whether I learned about a new software package at a conference, found a new case to use in the classroom, or read about an emerging issue in Accounting or Information Technology, I find that the concepts I am most passionate about teaching are the ones I have recently learned about. For example, this summer I read a book titled “George Washington’s Expense Accounts”, which provided an overview of how George Washington kept his expense accounts during the Revolutionary War. The book also discussed the historical significance of each expense account entry. I used this book as the basis for a group presentation for my Honors BUAD 131 course. The presentation required the students to read the book, comment on George Washington’s ability to keep accurate expense reports, and then review UMW’s policy for expense reports. When I was reviewing this group project at the beginning of the semester and listening to the group’s presentation at the end of the semester, I sensed that I really taught this well. Probably because it was something I recently learned more about myself.

Feb 03

The benefits of sharing information

It seems like the authors believe that privacy is paramount and that sharing personal information is a bad thing. The authors comment that we sell our personal information for discounted products and services. We receive personal benefits, in the form of free goods in services, such as free Facebook accounts, in exchange for revealing personal information.

While the authors, seem to believe that sharing personal information is always bad, could there be times when sharing personal information could be valuable to society? John Wilbanks, a healthcare IT expert, gave a TED talk describing how the intense desire to protect our privacy is slowing medical research. When we engage in medical clinical research, we need to give our informed consent indicating that we understand the objectives and purposes of the research study. This informed consent is designed to protect us from joining a clinical research study we don’t understand. Wilbanks further argues that the way we gain informed consent creates silos around valuable medical data. For example, the data collected for a colon cancer clinical research study could only be used to study colon cancer. This data can’t be networked, or used for other purposes or by other people. According to Wilabnks, while informed consent protects us from joining clinical studies we may not full understand, it can have hinder innovations in the medical field. More specifically, Wilbanks comments that informed consent as it exists today is preventing medical researchers from aggregating medical data from various clinical trials. This is problematic because many clinical trials have small sample sizes. By pooling our medical data, we can create a “commons’ which can be used to create larger sample sizes, thereby increasing statistical power. Using big data techniques, correlations and associations can be extracted and analyzed from these large medical data sets. These correlations, in turn, could be used to help find new treatments for various illnesses, ranging from cancer to Alzheimer’s disease.

I think Wilbanks has a point. Taking his point into the realm of commercial business, revealing our personal information has certainly cut the cost of purchasing goods and services on the web. I wonder how many people would use Facebook to stay in touch with friends and family if it wasn’t free? I wonder how many people would use Google for searching if they were charged? So, yes, while we are selling our personal information for lower priced goods and services, I’m not sure this is always bad. Perhaps I’m willing to accept less privacy in exchange for free goods and services. Or, perhaps I’m just cheap.

The full Ted talk can be found here: http://www.ted.com/talks/john_wilbanks_let_s_pool_our_medical_data.html