I. Types, characteristics and functions of literature review

Literature review is "an information research report that concisely describes the most important part and indexes the source on the basis of analyzing, comparing, sorting out and summarizing all or most of the information on a specific topic in a certain time and space". The definition of literature review includes three basic elements: first, literature review reflects that the original literature has a certain time and space scope, and it reflects the content of the original literature in a certain period or space scope. Secondly, the literature review focuses on the contents of some related documents. Other secondary documents, such as titles, indexes, abstracts, abstracts, etc. , can only reveal the appearance information or content information of an original document, and there is no connection between items. However, a batch of related documents can be condensed and revealed as an organic whole, which is much more informative than secondary documents. Thirdly, literature review is an advanced product of information analysis. Bibliography, index, etc. Objectively describe the appearance characteristics of the original file, without involving the content of the file. The writer does not need to know the content of the original document, nor does he need to have basic knowledge of related disciplines. Abstract: It is a brief introduction and evaluation of the contents of original documents, and editors need to have some basic knowledge of related disciplines in order to identify and evaluate original documents. Literature review requires the author to have a deep understanding of the theme of the review and comprehensively, systematically, accurately and objectively summarize the content of a certain theme. Using the methods of analysis, comparison, sorting and induction, it is a creative research activity to deeply process a certain range of documents, which has a profound guiding role for readers.

1. The types of literature review can be divided from different angles. The most commonly used method is to divide the information content according to the depth of literature review.

According to the different information content of literature review, literature review can be divided into three categories: narrative review, critical review and special research report.

Narrative review is an information analysis product that focuses on a certain issue or theme, collects relevant documents extensively, analyzes, sorts out and synthesizes its contents, and gives a comprehensive and objective description of relevant theories, viewpoints, data, methods and development in refined and summarized language. The most important feature of narrative criticism is objectivity, that is, it is necessary to objectively introduce and describe various viewpoints and methods in the original literature. Generally, the author's comments and opinions are not put forward, but listed systematically. The characteristics of narrative enable readers to understand various viewpoints, methods, theories and data of this discipline, major or topic in a short time, grasp the overall situation and obtain information.

A critical review is an information analysis report that compares, analyzes and comments vertically or horizontally on the basis of a comprehensive description of a problem or topic, puts forward the author's own views and opinions, and makes clear the choice. The main feature of critical review is analysis and evaluation, so some people call it analytical review. While summarizing various viewpoints, theories or methods, critical review should also analyze and evaluate every viewpoint, every data and every technology, show the author's own views and put forward the final review results. It can inspire ideas and guide readers to find new research directions.

The special research report reflects and evaluates a special topic, which is generally a major topic related to the development direction of national economy and scientific research, and puts forward development countermeasures and trend prediction. It is a realistic, policy-oriented and targeted research result of information analysis. Its most striking feature is predictability. After introducing and describing all kinds of facts or data and theories, it demonstrates and predicts, and finally puts forward the prediction and planning of future development goals and directions. The special research report has reference and basis for scientific research departments to determine the research focus and discipline development direction, lead departments to make various decisions and effectively implement management. This review mainly includes forecast report, feasibility study report, special investigation report, suggestions, countermeasures and ideas report. According to the time range of literature review, it can be divided into dynamic review, retrospective review and predictive review. The dynamic review mainly describes various recent realistic trends, such as political trends, scientific research trends, sports trends, economic trends and so on. News summary is a typical dynamic summary. This kind of review has strong timeliness and reflects the latest development trend. The retrospective review describes the achievements and development process of a certain period in the past, which has a strong summary and can be used as a reference for the present. Predictive review is to predict the development direction and goal in a certain period in the future on the basis of review.

According to the space-time scope of commentary reports, it can be divided into vertical commentary and horizontal commentary. Longitudinal review is described in the order of time development, which can reveal the development speed of review topics; The horizontal review is described according to the theme or region, country, product, etc., regardless of time sequence. , conducive to peer comparison. According to the clients, it can be divided into decision-making review, research review and promotion review. The main purpose of decision review is to provide decision-making reference for leading departments at all levels. On the basis of summing up, it is important to put forward suggestions as the basis for decision-making. Scientific research summary is to serve the scientific research work of researchers, which is highly professional and does not need to introduce basic concepts and principles. Universal review is aimed at readers, aiming to let readers know the general situation of a certain subject, topic and thing, and have a basic understanding, which generally includes a simple and comprehensive introduction to basic concepts, conditions, status quo and trends. According to the reported objects of literature review, it can be divided into subject review, literature review, conference review and special review. Summary of disciplines, summarizing the development trend of a certain discipline; Literature review, summarizing the research results of a number of documents and analyzing the internal laws of these documents; The minutes of the meeting, the comprehensive report of various viewpoints and theories put forward at the meeting, are generally not evaluated by the author; Thematic review, summarizing the history, present situation and development trend of a special field.

Characteristics of literature review

The content and characteristics of literature review have been mentioned before. Literature review is an advanced information research product, which recreates the original literature. Its content features are as follows:

Content synthesis: This is the most basic feature of literature review, which contains two meanings. On the one hand, literature review first shows a comprehensive description of a large number of documents. All kinds of comments are mainly based on comprehensive narration. "Every original document that needs to be summarized must be put in context and analyzed and synthesized with certain knowledge". On the other hand, it summarizes the development and situation in a wide range of time and space, with both vertical descriptions and horizontal reports.

Language generalization: Literature summarization is not simply copying or abstracting various theories, viewpoints and methods in the original literature, but summarizing with concise language on the basis of understanding the original text. Therefore, literature review is different from abstract, it does not extract the central content of the original literature; It is also different from extracting, and it is not necessary to extract exactly according to the original text. Instead, the useful theories, viewpoints and methods in the literature are summarized and described in the most refined language, and the data are extracted, while the details of the argumentation, calculation and derivation process in the original literature are abandoned.

Information concentration: Literature review focuses on the contents of a batch of documents in a certain period, and concentrates a lot of information. A review can reflect dozens to hundreds of original documents with high information density. Scholars at home and abroad have done a lot of research on how many references are needed for a review, and there are also some different views. To evaluate the compression degree of review literature, we can use the average number of references cited per page of review literature or the ratio of the number of pages of original literature to the number of pages of review literature. The concentration of review in different disciplines is different, depending on whether enough original documents are concentrated and whether the theme of review is fully reflected.

Objectivity of review: There are two aspects to the objectivity of comprehensive literature. On the one hand, we should objectively describe and enumerate various theories, viewpoints, methods, technologies and data, and truthfully reflect the contents of the original documents, and we are not allowed to distort them at will, or take them out of context, regardless of the context, and at the same time avoid misunderstanding due to different understandings; On the other hand, when analyzing, comparing and commenting on various theories, viewpoints and methods, we should have an objective attitude, analyze and evaluate objectively, and we should not comment out of personal preferences and tendencies, let alone deliberately take sides or attack out of personal feelings. In addition, when making predictions, we should try to be objective based on facts and data and scientific deduction methods, instead of imagining them out of thin air and blindly proposing them out of subjective will.

Formal characteristics of literature review

Literature review is different from other information products and documents in content, and also has some remarkable characteristics in form. Its formal characteristics often become the main basis for judging whether a document is a comprehensive article. There are many references. As mentioned above, the literature review condensed the contents of the original literature. Therefore, its most obvious formal feature is a large number of quotations. Mikhainov proposed to summarize the articles that quoted 40- 100; Price believes that any paper containing more than 25 quotations from other works is a comprehensive paper; Woodward believes that any scientific literature with at least 40 references can be regarded as a review paper. The title is very eye-catching The titles of general review articles directly reflect the types of reviews, such as those containing words such as "review", "review", "review", "progress" and "dynamic", or titles such as "current situation, trends and countermeasures" and "analysis and thinking". This document is usually a review article. In addition, the titles of abstract articles often reflect the time range of the abstract, such as "in recent years". A summary of China's summative research in recent ten years is a typical summative article title.

The role of literature review

Concentrate on information and report a number of related documents, so that researchers can keep abreast of the development trends of disciplines, majors and special topics. With the increasing role of information in modern society, the trend of global informatization is becoming more and more obvious. The amount of literature in various disciplines has surged; A large number of interdisciplinary and marginal disciplines have emerged; There are more and more documents in other languages, and the degree of document dispersion is increasing. The percentage that users can read is getting smaller and smaller. When the number of documents reaches a certain level, users will lose interest in reading documents. Literature review condenses all aspects of information, so that users can learn a lot of information in a very short time without language barriers. A comment covers dozens to hundreds of original documents, so users don't need to read a lot of original documents, which reduces repeated reading. System report is convenient for users to fully and systematically understand the progress or overview of a certain aspect. Literature review analyzes and synthesizes the contents of the report to make it orderly. When reporting, you can extend vertically, compare horizontally, or crisscross. Organize and systematize various theories, viewpoints and methods from different angles to reflect the whole picture of the problem. Evaluate and predict the original document. Literature review should reflect some related documents, so in the process of writing the review, we should evaluate the selection, generalization and introduction of the original documents, and provide the essence through evaluation. On the basis of summing up, this paper puts forward reasonable reasoning, scientific prediction, and future development direction and prospect, so that the leading departments can make decisions according to the evidence. It can be used as a reference for researchers to determine new research directions and avoid repeated research and development. Provide retrospective retrieval methods to expand the number of documents controlled by users. Literature review articles are accompanied by references and related information, because reviews are usually based on one or several topics and problems, so references usually belong to the same category or related. Readers can start with the references after the article to search back, directly find and read the original documents they are interested in, and concentrate on mastering some related documents. Literature review has been paid more and more attention. In particular, users engaged in scientific research and decision-making prefer to read relevant literature reviews. Some users have regarded literature review as the main source of information, and put forward many requirements for the quality of literature review. How to write high-quality abstracts is a concern of information workers and readers at present.

Second, the compilation of literature review

1. Writing steps of literature review

Literature review is an important achievement of information analysis. Intelligence analysis is also called intelligence research, intelligence investigation, intelligence analysis and synthesis. It refers to "a scientific thinking activity and re-creation activity in which information workers collect and occupy information purposefully around a certain theme according to the needs of society and users, and systematically summarize relevant information by means of sorting, evaluating and screening, and by means of analysis and synthesis". In the process of information analysis, a series of information products such as titles, indexes, abstracts and abstracts will be produced, and literature review is the highest product of information analysis. Different from previous products, it simply deals with the form and content of the original document, but it is a re-creation activity as well as a research activity. Mikhailov believes that "writing a summary is usually between scientific information work and research work itself, and sometimes it is outside the dividing line between them". The writing process of literature review is the process of information analysis, and it is the research work of writing literature review on the basis of comprehensive analysis of all or most related literature contents.

The compilation process of literature review is as follows: (1) Determine the topic and research focus; Collection, tracking and accumulation of documents; Analyze and evaluate the value of data and screen it; Organize the data and make it systematic; Write a summary.

Determining the topic and research focus is the first step of information analysis and literature review. Generally speaking, there are two ways to choose the topic of literature review: active and passive There are the following situations to choose a topic actively: first, intelligence personnel or intelligence research institutions find problems and clues through daily intelligence research and service activities to choose a topic; First, in the contact and communication with intelligence users, intelligence personnel or institutions know what users need, so as to determine the topic; First, professional researchers accumulate a large number of systematic data in their daily research work, and combine their own research results and experiences to actively write summaries. Passive topic selection is divided into topics assigned by superior leadership departments and topics entrusted by other units. Whether it is passive topic selection or active topic selection, it is necessary to consider whether summarizing the topic selection has research value; Whether you have the ability to complete the topic according to your own manpower and material resources. Whether the topic is suitable is related to whether the writing task can be successfully completed and whether certain social and economic benefits can be obtained.

Topic selection principle

The principle of pertinence: some people call it the principle of applicability to users, which means that the topic must meet the needs of certain users, that is, it has certain social and practical significance. If the topic is selected or entrusted by the superior leadership department, the department or unit that selects the topic should consider the topic from the needs of the unit and the department, combined with the practical problems in production and work; If the intelligence department chooses a topic spontaneously, it is necessary to understand not only the direct needs of users, but also the potential needs of users, and dig out those needs that users have not expressed in words or are unaware of. Predictability principle: the predictability of the selected topic requires that the evaluated topic should pay attention to the current hot issues and report them in time. However, if we simply pursue hot reports and work behind "hot spots", there will be information lag, which can not meet the needs of readers in time. Therefore, when choosing a topic, we should be good at finding those problems that should have attracted people's attention, but they have not been paid attention to, so as to fill the gap and attract people's attention.

Feasibility principle: the feasibility of selecting a topic refers to the human, material and time conditions for completing this topic. Material conditions include access to information, amount, source of funds, etc. Human resources refers to the number of people who participate in writing and collecting materials, whether the writer has all the conditions for writing a summary, his knowledge structure, academic level, skills, comprehensive analysis ability, writing organization ability, foreign language level, etc. In addition, we should also consider the time limit for completing the summary writing, whether it can be completed on time under the existing manpower and material resources, how likely it is to be completed, and whether it is capable of undertaking the writing work. Whether it is passive topic selection, active topic selection, or spontaneous topic selection by researchers, we should seriously demonstrate its feasibility to decide whether it is a negative topic selection. When considering the feasibility of selecting a topic, we should give full play to the advantages of the unit and the author.

Principle of originality: The purpose of literature review is to avoid a lot of repeated reading and save time and energy, so we should also avoid repeated topics. The topic selection should be something that no one else has done. Before deciding on the topic, it is necessary to extensively investigate whether predecessors have made similar reviews or are doing them. If you choose the topic blindly without any investigation, it will easily lead to repeated topics, waste a lot of manpower and material resources, and cause readers to repeat reading.

How to choose a topic

The topic selection of literature review should adhere to the principles of pertinence, foresight, originality and feasibility. Specifically, the topic can be considered from the following angles. Cooperate with the state to formulate scientific research development direction and scientific research planning topics. For example, the selection and determination of projects funded by the National Ninth Five-Year Natural Science Fund and Social Science Fund need comprehensive reports from various disciplines for the leading departments to choose and make decisions. Topic selection of major planning, policies, decisions and key construction projects in national economic construction and development. For example, during the start-up and construction of Gezhouba Water Control Project, relevant personnel submitted hundreds of special intelligence investigation reports. Select the related topics of administration. Such as technology introduction, making production plans, developing new products and other management problems, so that managers can understand the advanced management experience at home and abroad for reference and learning. Controversial, there are problems to be solved. By summing up their own opinions or putting forward their own opinions on the basis of summing up, it is helpful for readers to compare, identify and choose controversial and unresolved issues. Inspire readers' ideas. In view of the new disciplines, new problems, new ideas, new theories and new growth points of discipline development, as well as the major problems reflected by scientific development trends and trends, the topic is selected. We should start with some promising new problems as soon as possible and keep tracking and reflecting them with their development. Cooperate with the national economic and political situation to carry out basic theoretical research topics. Such as the current research on "market economy" and "Deng Xiaoping Theory".

Steps of data collection, tracking and accumulation After determining the topic, you can start collecting data. Collecting information in a planned and step-by-step way can get twice the result with half the effort.

Clarify the scope of retrieval: it is necessary to further clarify the contents and topics involved in this review topic, so as to clarify the general scope of the required materials, and then determine the scope and focus of information retrieval, which are the core data sources closely related to the review topic, which are the background data sources needed for writing review, and which are the reference data sources and tool data sources, and delineate the retrieval scope.

System search: after determining the scope and focus of the search, start the system search and collect all possible relevant information comprehensively and extensively. "Systematic retrieval is a method to obtain literature information comprehensively and systematically through retrieval tools according to the content characteristics (disciplines or themes) or external marks (such as authors, titles and publishing institutions) of documents". In manual retrieval, the commonly used retrieval tools are abstracts, indexes, bibliographies, directory announcements, directory cards and so on. In computer testing, organic reading indexes, abstracts, full-text databases, etc. Often used. As far as retrieval approaches are concerned, manual retrieval approaches include classification, subject, author, book title, references and various special terms and symbols. There are many retrieval methods in computer detection system. In addition to the above manual retrieval methods, in the full-text retrieval system, you can also use free words, keywords, time, place, abstract words, organization names and other retrieval methods to conduct fuzzy retrieval, implicit retrieval, combined retrieval, and comprehensively retrieve relevant information. There are many retrieval methods and tools. When carrying out systematic retrieval, we should determine the retrieval theme and category according to our own conditions and the concept and content of the review topic, and then determine reasonable and feasible retrieval methods and strategies, and select appropriate retrieval tools to ensure the recall and precision of retrieval. In the process of retrieval, the relevant information of the retrieved materials should be recorded in order to obtain the original documents.

Extensive browsing: the publication of general retrieval tools has a certain time lag, about one year to several months. Therefore, it is necessary to read and browse all kinds of documents in the past six months extensively to get the latest information. The types of documents and materials mainly include: books, periodicals, conference documents, patent materials, standard documents, newspapers, encyclopedias, yearbooks, handbooks, institutional guides, atlases, dictionaries, scientific and technological reports, dissertations, manuscripts, statistical data, product materials, government publications, internal materials, microform and machine-readable documents, etc. And pay attention to browsing the new book bibliography, new book announcements, etc. How to use these documents to obtain information when searching and browsing? You can use books, professional dictionaries and encyclopedias to understand the existing definitions and viewpoints; Use newspapers and government publications to understand the relevant national guidelines, policies and regulations, as well as the general situation of the subject, and obtain background materials; Use relevant yearbooks and handbooks to obtain facts and data and grasp the general situation of development; Grasp the latest news and research trends by using comprehensive literature and meeting minutes; Understand the performance index of technology or product by using patent specification; You can also use patent novelty search to find out whether a technology or product has been developed. It should be emphasized that attention should be paid to the collection and utilization of comprehensive information. Information exchange: including visits, meetings or individual conversations, field surveys, and distribution of questionnaires. , directly get first-hand materials and the latest information. The information obtained through communication should be accurately recorded, and its source and reliability should be carefully checked before it can be quoted or referenced in the abstract.

Principles of data collection

In the process of collecting data, we should pay attention to the following principles in order to ensure the quality of collected data.

Principle of universality: There are three meanings here. First, it refers to a wide range of disciplines, which not only collects relevant documents of this subject, but also collects relevant documents of some interdisciplinary and basic disciplines; Second, it refers to a wide range of literary types, including printed literature, micro-literature, machine-readable literature and other forms of literature; Third, the collection has a wide range of time and space. According to the content of the review, the collected data can be traced back to decades or hundreds of years in time, and can include domestic and foreign literature in space.

Representative principle: Pay attention to collecting representative documents. For example, the literature published in the core journals of this discipline, the articles written by academic leaders or well-known scientists, and the speeches of leaders of relevant state departments can all represent the current level of development and understanding.

Principle of timeliness: when collecting data, we should determine the time range of collection and look back at what time period. Because the literature has its timeliness, the literature on some topics has a short half-life and a fast aging speed. Determining a reasonable search time can avoid getting some useless information and reduce the work in the data screening stage. Data collection is not limited to this period, but a long-term process. In the process of writing a summary, we should constantly pay attention to the development of this discipline, constantly consult and collect the latest information, and reflect the latest and recent data in the summary as much as possible. With the process of data collection, data accumulation is also synchronized. This process of tracking and accumulation will continue until the final summary article is finalized. The methods of data collection and accumulation mainly include abstraction (translation), clipping, copying and memory. Either way, the source (document type, name, annual volume, page number), author, publishing unit, document collection unit and request number of the data must be recorded. Therefore, in the process of data collection and accumulation, by-products such as "title record" or "catalogue" can be produced. When analyzing and evaluating the value of data, we only skim the data and judge whether it is available according to the title, abstract, abstract and directory arrangement of the data. At this stage, it is necessary to read through all the collected materials, make theme analysis and factor analysis, judge whether and how much they are available, and then filter the materials to narrow the scope of reference materials.

Principles of identifying data

The evaluation and appraisal of materials must be considered in combination with the content value of materials, and there are mainly the following principles.

Reliability principle: that is, the authenticity of data and whether the contents of documents are objective and true. With the development of modern science and technology, there have also been mixed problems in literature, and a large number of unscientific and pseudoscientific things have appeared. When selecting materials, we should pay attention to the authenticity of the materials and choose them carefully.

Principle of objectivity: refers to the appraisal and evaluation of documents and materials, and the attitude of deciding whether to accept or reject should be objective and fair, not subjective and not mixed with personal feelings.

Relevance principle: refers to how relevant the collected literature content is to the review topic and how much content is related to the review topic. Documents with more relevant information should be given priority, and documents with less relevant information or of little use value to review topics and materials with repeated contents can be eliminated.

Applicability principle: refers to the extent to which documents and materials can be used by users. That is to say, although the contents of documents and materials are closely related to the review topics, due to the background of documents and materials, such as geographical environment, climate, natural resources, level of scientific and technological development, economic capacity, and ethnic, cultural and national differences, the information provided by some documents and materials is not universally applicable and can be used for reference, but only applicable to situations similar to the document background. For example, the introduction and evaluation of products and technical information should consider whether it is suitable for the national conditions and the technical level of the country. The introduction, evaluation and selection of theoretical viewpoints and theories should consider the political system, ideology and humanistic factors of the country, and cannot be quoted at will. The principle of intelligence application is a dynamic principle, and there is no clear standard, which needs to be treated according to the writer's specific situation.

The principle of novelty: on the one hand, it depends on whether the ideas, principles and methods put forward in the literature have new contents and new developments compared with other materials, and the facts described have never been published or new conclusions have been put forward on the basis of the original facts; Including new historical materials, new data or new skills, new concept system, new theoretical model, etc. Whether new technologies, inventions and new management experience have been introduced in the research methods. Summarizing the new contents contained in the literature is the most enlightening for readers, so that readers can keep abreast of the development trends. The novelty of data content can be judged from the external characteristics of literature, the content of literature and the basis of scientific and technological development, as well as from the aspects of popularization and application, comparison with similar achievements and technical and economic effects. On the other hand, it depends on whether the viewpoints, theories and methods reflected in the literature are outdated and lose their use value. In particular, the information about technical methods is extremely timely. With the development and use of a new generation of products, the old technical methods will soon be eliminated, with little practical value. Therefore, it is necessary to understand its timeliness and advancement. This is particularly prominent in the intelligence investigation of introducing equipment and putting into production new products. It is necessary to avoid drawing wrong conclusions because of outdated information and ignorance of the current development situation, and blindly introducing outdated technologies and equipment.