Secondary data

Although the collection of statistical data mainly refers to the collection of original data, and the statistical investigation methods to be introduced later also focus on the collection of original data, the collection of data actually includes not only the collection of original data but also the collection of secondary data (second-hand data). In many cases, statistical research is based on mastering secondary data.

Secondary data (second-hand data) refers to statistical data collected and collated by others. Under certain conditions, researchers may not be able to collect data in person, or they may know that some data have been investigated by others, so it is not necessary to do it again. At this time, it is necessary to collect second-hand information to meet the needs of research. This kind of secondary data based on other people’s investigation is also called the indirect source of data.

Common indirect sources of data mainly include:

(1) published data. Mainly from government departments, organizations, schools, scientific research institutions, etc., such as: China Statistical Yearbook, Compilation of Population Census Data, Beijing Statistical Yearbook, World Development Report, research data released by a university or scientific research institution, survey results data released by professional survey consulting institutions, statistical data released by various media, books and newspapers, etc.

② Unpublished data. For example, the business report data of various enterprises and the unpublished survey results data of professional survey and consulting institutions. It should be noted that if you cite unpublished data, you should pay attention to compliance, get the consent of the data owner, and be responsible for the consequences of using these data yourself.

③ Data crawled by the network. In the era of big data, the scale of data is also growing massively. There is a large amount of data in the Internet, which can be in the form of numbers, tables and other structured forms, or in the form of voice, pictures, text, video and other unstructured methods. People can automatically or manually obtain data by using technical means such as web crawler, and process and sort out these crawled data for analysis. These data are also second-hand data relative to the data crawlers, because the process of data from scratch was realized by others, not by the data crawlers, and the data crawlers only completed the work of data integration or sorting.

Proper use of indirect data can save manpower, material resources, financial resources and time in practice, and achieve better results and benefits. However, we should pay attention to its applicability and timeliness when using indirect data. Researchers should analyze whether the purpose of collecting original data is consistent with their own research purpose, find out whether the method of collecting original data is scientific, whether the provider of original data is fair and objective, and also pay attention to whether the meaning, calculation caliber and calculation method of data are comparable to avoid data misuse or abuse. In addition, try not to use outdated data, and when quoting second-hand data, be sure to indicate the source or source of the data, and respect the labor achievements of others.