Evaluating the value of a paper is nothing more than evaluating it from both quantitative and qualitative aspects, or evaluating it from the perspective of journals and papers. The indicators mentioned below are quantitative, qualitative, periodic and articles.
Impact factor (IF)
The impact factor is specific to journals, not directly to papers.
The impact factor should be a quantitative and semi-qualitative indicator. Why? This is because journals with high impact factors are not without low-level articles, but also have high-level articles with low impact factors.
Of course, if the author's husband finds a good job, his status will naturally be high. Just like the former imperial concubine, the overall quality may be better than that of the general public, and there are also many outstanding figures among the civilians.
Therefore, the current influencing factor is also suffering from scale disease. Some people call SCI a "stupid China idea", so I won't go into details.
There are many ways to check the impact factors. Of course, the most classic method is the science network, but most people can't afford it. A simple and convenient way is to use medical literature assistant: use medical literature assistant to screen the quality of PubMed literature.
Citation times
I don't need to introduce this. Many databases and online query platforms can check the citation times, such as Google and Microsoft academic search, and use medical literature assistants to screen the quality of PubMed literature.
H index (H index)
H index was put forward by george hill Shi, a physicist at the University of California, San Diego, in 2005.
The calculation of H index is based on the number of papers by its researchers and the number of times their papers are cited. Hirsch believes that when a person's N papers in all academic articles are cited at least N times, the person's H index is N.
You can determine someone's H index in the following ways:
Rank all SCI papers published by them according to the number of citations from high to low;
Find the sorted list from beginning to end until the serial number of a paper is greater than the number of citations of the paper. The H exponent is obtained by subtracting 1 from the obtained serial number.
The above information about the H index comes from Wikipedia.
The easiest way to check the H index is to use Google Academic. Note that it is in English, not Chinese. In addition, FireFox and Chrome also have corresponding plug-ins to choose from.
I 10 index (I 10 index)
I 10-index was put forward by Google, which refers to the number of articles published by authors that have been cited more than 10 times.
For example, I have published 100 articles (hehe, it's a bit big), and 90 of them have been cited by others more than 10 times, so my I 10-index is 90.
If the impact factor is for journals, then the H index and I 10 index are for individuals. The paper has a high impact factor, so it can only be said that the paper has found a good husband's family, and the specific citation is not necessarily. H index and I 10 index are quantitative standards that accurately reflect the citation of papers.
G index
G index is a little unknown relative to the above indicators. G-Index was put forward by Leo Egghe in 2006, which is an index to evaluate the number of authors' papers.
The calculation method of G index is as follows
Arrange the articles published by all authors in descending order of citation times, and the serial number is G.
Square the serial numbers of all articles published by the author to get g2.
Add up the citation times of all the articles of the author to get ∑TC.
The last serial number where ∑TC is still greater than g2 is the G exponent.
It's a little detour It's okay. For example, it is more clear. For example, I published the following articles, sorted by the number of citations.
Citation times (TC) and serial number (G)
Sum of citation times (∑ TC) g 2
47 1 47 1
42 2 89 4
37 3 126 9
36 4 162 16
2 1 5 183 25
18 6 20 1 36
17 7 2 18 49
16 8 234 64
16 9 250 8 1
16 10 266 100
15 1 1 28 1 12 1
13 12 294 144
13 13 307 169
13 14 320 196
13 15 333 225
12 16 345 256
12 17 357 289
12 18 369 324
12 19 38 1 36 1
1 1 20 392 400
… … … …
As can be seen from the above table, my H index is 13, and my G index is 19, because the 20th document g2 is greater than the sum of all the previous quotations. The value of this example comes from an improvement of egghe.an index: G index.
Compared with H index and I 10 index, G index can better reflect the overall citation of the paper. For example, the articles I published are generally not high, and the H index may be relatively high, but once the G index is calculated, the original shape is revealed, and it turned out to be parallel imports.
H5 index (H5 index) and median H5 (median H5)
H index, I 10 index and G index are the statistics of the cited times of personal papers, while H5 index and H5- median are the evaluation systems of the cited times of magazines.
H5 index
H5 index is the minimum value of the ratio of the number of papers published in magazines to the number of citations in the past five years, such as
In recent five years, Nature has published 1000 articles (of course, the actual value is larger than this), which are arranged in descending order according to the citation times of each paper.
The citation times of 38 1 article are 38 1, and the citation times of 382 articles are 300, so the H5 index of Nature is 38 1.
Compared with IF, H5 is the article citation of Reaction Magazine in the past five years, and IF is the average citation of Reaction Magazine. Compared with the H index, H5 is aimed at the overall situation of the magazine, while the H index is aimed at the citation of individual papers.
Sometimes magazines have different influencing factors, and H5 may be the same. For example, the H5 of PLoS One and Nature Reviews Immunology are both 130, but the difference of influence factors between them can be more than one grade.
H5 median
H5- Median refers to the median number of citations of articles used. Why not use the average? Because the data is not normally distributed. The number of citations in each magazine will definitely not be evenly distributed, just like the income of China residents. Some must be quoted a lot, but some documents may be very watery and the number of citations is pitifully small. If the average can't reflect the real citation, the median is the best choice.
F 1000
F1000 (Facility of1000) is a secondary literature database, which provides evaluation services for biological and medical researchers. It is a new online research assistant tool released by BioMed Central, including biology and medicine. At present, it is a new evaluation system for life science researchers, not only to see whether it is included in SCI.
Medical F 1000: collected and evaluated by more than 2,400 world-class clinical experts and scholars, providing the most important information and development trend of medical papers in the world at present. Including 18 fields: anesthesia and analgesia, cardiovascular diseases, intensive care and emergency medicine, dermatology, diabetes and endocrinology, evidence-based medicine, gastroenterology and hepatology, hematology, infectious diseases, kidney diseases, neurology, oncology, psychology, public health and epidemiology, respiratory diseases, rheumatism and clinical immunity, urology and women's health. Website literature links with PubMed and PubMed Central.
Biology F 1000: Assessed by more than 2,300 experts and scholars, it provides the most important international biological paper information and research trends. Covering subject areas: biology, biochemistry, bioinformatics, biotechnology, tumor biology, cardiovascular biology, cell biology, chemical biology, developmental biology, ecology, evolutionary biology, gastrointestinal biology, genomics and genetics, immunology, metabolism and endocrinology, microbiology, molecular biology, molecular medicine, neuroscience, pharmacology and drug discovery, physiology, plant biology, kidney biology, respiratory biology.
main feature
This paper mainly makes an objective evaluation of the important papers included in PubMed, which is based on academic achievements rather than whether the journal is included in SCI.
The members participating in the review are composed of famous experts from internationally renowned institutions such as the United States and Europe. According to the contribution degree and scientific value of the papers to the current world biological and medical research, the index (F 1000 factor) which objectively reflects the academic level is used for grading, and several excellent papers in the last month are recommended to readers every day, and links to Pubmed are provided.
The three grades of F 1000 are 9 (excellent), 6 (required reading) and 3 (recommended).
The above content about F 1000 comes from Baidu Encyclopedia.
Therefore, compared with the influencing factors, the manual intervention of F 1000 is more, and its research significance is more significant. F 1000 should be a qualitative index, and the simplest way to check F 1000 in real time can also use the medical literature assistant.
After talking about the traditional evaluation indicators, let me take a look at the emerging paper evaluation indicators.
Substitution measure
Altmetric is a new indicator. Although it literally means a substitute indicator, I think "social influence" or "network influence factor" or "sharing factor" can better reflect its essence.
The background of replacing metric system.
You may encounter such a situation. After some papers were published, they were widely reprinted, reported by online news, and commented and shared on Twitter or G+. At this time, the impact factor and F 1000 can't reflect these.
Altmetric appears in this situation, which embodies the sharing, downloading and reading of a paper.
But now Altmetric is also controversial, and the name is also controversial. Personally, I think InterMetric is better, called IM for short, which is a bit mixed with real-time communication software. SocialMetirc, SM for short, is a little abnormal.
For more details about Altmetric, please refer to this article: Understanding the attention sharing of papers by using Altmetric evaluation system.
Similarly, Plum Metrics (using Plum Metrics evaluation system to understand the sharing of paper attention) and Impactstory (this may be an agent).
RG score
RG factor is an index to evaluate authors introduced by ResearchGate. The purpose of RG grading is to help you evaluate your level in the scientific circle. The calculation method is not how many articles I have published, but the degree to which my scientific research work is recognized by my peers.
RG score is different from the traditional evaluation index. It can count more information, such as downloading, browsing and sharing. The difference between RG score and Altmetric is that RG score is more important than sharing. If you share your ideas with your peers and get their approval and discussion, RG scores will increase rapidly.
More details about RG Score can be found in this article: ResearchGate researchers' own FaceBook.
Teachers arrive at school first every morning and finish school last every night; I'm going to share with you some topics of teachers'