How do you view the proliferation of applied micro-empirical research in reduced form in domestic economic research?

Reply to questions about Bai and Jia (2016)

Baiying Jia Ruixue

September 24, 2022

First of all, thank you for the discussion on Zhihu. We treat our essays like our own children: on the one hand, we want it to grow up with criticism, and on the other we get angry at the accusations that come when we don’t read it carefully, or when the reader misunderstands it. In order to avoid more misunderstandings, we will reply to the questions one by one. This is not to say that this article is perfect, and we are aware of its limitations (which have little to do with the question posed by this skeptic). There are many related issues that we hope to better understand in future research.

After the explanation, please consider that we need to focus on our work, not quarrel on Zhihu. We have turned off the reply feature for this article. If you want to continue serious discussions, you can email us. Our contact details are easily found in the paper.

Reply about the relevant data (Original question 1, 2, 3):

(1) Our data is based on six major revolutionary groups (the article is summarized in the figure below), not just the Tongmenghui. If the people who participate in the alliance have participated in other organizations before, they will be classified as members of other organizations first. The picture shows that there were indeed many members before 1905.

(2) Regarding the year, we are based on the earliest time each revolutionist appears in the data. Instead of simply using the time it joined the Alliance. For example, some people mentioned in the original question added the following years to our data: Song Jiaoren 1904, Huang Xing 1903, Zhang Binglin 1904, Chen Tianhua 1903. As with (1), we do not fail to take into account the previous organization as the skeptics assume.

(3) This is not to say that there are no measurement errors in the data we use. The challenge certainly exists. We did at least the following four sets of related analyses:

(3.1) Our double-difference estimate is biased only if revolutionaries from prefectures with more places before the abolition of the imperial examination are mistakenly believed to have joined after the abolition of the imperial examination. Our article is not a simple comparison of before and after 1905.

(3.2) We did not shy away from talking about ” a large number of revolutionaries joined the Tongmenghui in June-August 1905 ” (this is the main question of our article). We mentioned in the article that after removing those who participated in other organizations in the early years, 35% of the remaining members of the Alliance are still recorded as joining before September:

Also, to confirm that the effect of 1905 occurred after September (abolition of the imperial examination) rather than before September (abolition of the imperial examination), we conducted a monthly analysis using the time of joining the Tongmenghui.

(3.3) We use information from the 1911 uprising to demonstrate the robustness of the results. We think that although the uprising information also has the problem of measurement error, it has little to do with the measurement error in the revolutionary information.

(3.4) Our instrumental variables can also help with the problem of measurement error.

Reply to the Foreign Imperial Examination (Original Question 4):

There are different ways out after the imperial examination, and we have done some background discussions in our article. Some empirical analyses, such as studying abroad, are also provided. We also discussed the foreign imperial examinations, and our main finding is that compared with the imperial examinations, the relationship between the number of students and foreign scholars in the foreign imperial examinations is weaker (see columns 7-8 in the table below).

Of course, this is not to say that we analyze all possible outcomes – it is not the purpose of our article to analyze all possible outcomes.

Reply to Jinshi data (original question 5):

In our original data, jinshi are divided into flag registration and national registration. Our jinshi data is based only on the latter and does not include flag status.

Summarize:

Our articles are not perfect and can be criticized, but only after reading the article carefully and giving reasonable criticism is meaningful.

Source: Zhihu www.zhihu.com

Author: Ruixue Jia

[Zhihu Daily] The choice of tens of millions of users, to be a big cow to share new things in the circle of friends.
click to download

There are 31 more answers to this question, see all.