Issue 184

xusichuangUID: 2653998

184"It is a grave mistake to theorize before one has data."



In scientist field, whether it is a mistake to theorize before one has data has been a controversial question. I do not agree with the author’s opinion that data is prior because sometimes it is more efficient to theorize firstly and then use data to check the theory.
In scientific studies, especially when it is a long time ago, there are little scientific theory and the data is relatively simple to analysis. It is easy for scientist to see through the data and develop theory from it.
However, as the scientific study develops, the data becomes complicated, it is hard sometimes even impossible to directly do analysis with the data. A better way is to generize a logical theory and then uses data to check whether this theory holds.
This method, when comparing with “data first” has advantages. The most significant one is that theory first enables a researcher a goal-oriented way to analysis on the data. Almost every economic research is such an example. Facing with the macroeconomic data, one may feel helpless because it is so intertwined and so complicated that one even can not find where to start s/he’s analysis. On the opposite, however, when a logical theory is developed, a researcher will have its goal, that is, to check whether the data goes well with s/he’s theory, if so, the theory is right, if not, the researcher will focus on the shortcomings of s/he’s theory. From the above analysis, we can see that it is obviously more efficient for one to theorize before data.
On the other hand, in some scientific field such as theoretical math, quantum theory that is as abstract as an abysmal, there are not correspondent data, at least in a certain time interval. The convincing example is Albert Einstein’s relative theory. When Einstein put forward his theory, he just practiced an imaginary experiment because in reality, there is no such an object whose speed is as high as light. At that time, because of lacking data, many of his colleagues do not believe him. However, other people’s distrust doses not prevent the theory from being true. Many years past, when technology developed to the extent that researchers can experiment and collect data in the micro world on the relative theory, he was finally proved right. Thus, the most distinguished researchers’ theory is so advanced that data does not exist in his time. However, their research, even without correspondent data, might be both true and monumental.
To sum up, I concede that it is a more direct way to develop theory form the exist data in the relatively simple research. However, to theorize first will be more efficient. Moreover, many prominent scientists’ study, although important and conductive, can not find data in reality.

qisaimanUID: 2703629

似乎没理会grave mistake 这个提法。
加点让步更好。比如刻意用数据验证理论。