The advantage of desk research is that it saves time and money. However, researchers must pay attention to the fact that many second-hand materials obtained in the market have serious defects.

Researchers need to pay special attention to the following aspects:

usability

In some countries, the statistical data are very complete, and enterprises can easily get the information they need, but in other countries (especially developing countries), the statistical means are backward, and it is difficult for researchers to get the information they need.

(2) timeliness

The data obtained from some information sources in some countries are often out of date for several years and cannot be used as the basis for enterprise decision-making.

(3) Reliability

The data provided by some countries are processed by scientific methods with high accuracy, while the data provided by other countries are only estimates with low accuracy.

(4) comparability

Sometimes the data obtained from different countries can't be compared with each other, which is because the situation in different countries is different and the data collection procedures and statistical methods are different. Sometimes, the same data may use different base periods in different countries, and the same indicator may have different meanings. For example, the consumption of TV sets is classified as entertainment expenditure in Germany and furniture expenditure in the United States. The comparability of national data in different countries will inevitably affect the usefulness of data, thus affecting enterprise decision-making.

Because of the above problems in second-hand materials, researchers should consider the following points when using these materials:

(1) What is the specific scope of statistical data, that is, which products are included?

(2) Who first collected the data? What's the purpose? Is there a motive to distort the data?

(3) Where did the data come from? What reason does the information provider have to distort the data?

(4) How is the data collected? Is the collection method reliable?

5] Is the data consistent with other local or international data? If there are serious differences, whether the reasons for the differences can be satisfactorily explained.