Number of citations because you really do not control this outcome so to me it is the arbiter of impact and external acceptance of one's work. If I have to choose a second then it would be recommendation

Chia-Lin - in terms of 'institutional' measurement, citation is the most important measure. It is often the more formal and universal 'acknowledgment'. One assumes, but it is not always the case, that a researcher has critically appraised your work and deemed it fit to include in their published work to support their findings. The number of 'reads' is still quite important. It demonstrates that readers are at least interested in your work. However, it is no indication that they then go on to use it to support their work. In my mind, recommendations is still a useful measure - but perhaps the lowest value of the three. It is more a subjective and personal acknowledgement that people 'like' your work. Again - it is no indication that it will be used for anything other than personal interest.

Dear Dean:

Thank you for your detailed, helpful and informative Answer.

It might be worth mentioning that the editors of some journals have been known to recommend journal self citations for accepted papers, which should count against Citations, but does not.

For me, it is the number of citations that reflects the quality of any academic research the most.

Not a problem Chia-Lin. That is true of some journals around 'forcing' citation from its sources - but it is not very common. Those journals soon get 'exposed' and it usually becomes a negative outcome for them.

Dear Amrit Kumar Bhandari :

Thank you for highlighting Citations as a measure of academic research quality.

Dear Dean:

Journal self-citations, and pressure to follow an editor's suggestions to follow suit, seems not to be uncommon in some disciplines in the social sciences.

Chia-Lin - Currently, one of the most useful measure of the quality of academic research and its impact is the H-index, which reflects both a researcher's number of publications and number of citations per publication (a measure of the publication's quality). Specifically, a scientist has a value h if h of their papers have at least h citations each, and their other papers are less frequently cited. The h-index is an author-level metric that measures the productivity and citation impact of the publications of a scholar. The index is based on the set of the scholar's most cited papers and the number of citations that they have received in other publication. An h index of 20 is usually considered as good, 40 as outstanding, and 60 as truly exceptional. The advantage of the h-index is that it combines productivity (i.e., number of papers produced) and impact (number of citations) in a single number. However, research gate too has many advantages as it exposes a scholar to a galaxy of other scholars, scholars learn and share ideas in their own field or related fields. Also, scholars from different fields can work together as collaborators, hence, it could be seen as a source or agency for furthering one's research goals and research contribution.

Dear Dr Manzoor Hussain :

Thank you for your detailed and informative Answer about the h-index, which is a matrix measure that depends on both Citations and Publications.

This is an interesting and challenging Question.

Citations generally refer to published articles and unpublished papers.

Read and Recommendations can refer to both Publications and Q&A.

Although Citations are generally a preferred measure of the perceived quality of academic research, alongside the h-index which balances Citations against Publications, Reads and Recommendations represent an indirect measure of the quality of academic research, as well as Q&A.

In the non-RG world, Citations is generally the most important. Citations reflects whether this article can bring more researchers inspiration or substantial help.

In the RG world, Reads is the most important. For recommendations in the RG world, there are usually only a few people who are willing to learn about your articles. I think RG's focus is on getting more people to read their articles, which is a very useful way to improve Citations.

So my answer is citations >>> reads > recommendations.

Best regards,

Yu-Da

Dear Michael John McAleer:

Thank you for your informative explanation about Reads and Recommendations representing an alternative and indirect measure of the quality of academic research and Q&A.

Dear Yu-Da Lin :

Thank you for your careful and helpful explanation of Citations in the non-RG world, and Reads and Recommendations in the RG world.

Sometimes I refer to articles, even if I disagree (or see limitations) with these ... Simple scientific decency requires mentions of predecessors. Even ignoring auto-citation, friendly citations, and citation to both "bosses" and "classics", it’s uncorrect to rate the quality of work by the number of citation. Sometimes the source of publication and journal name plays a role. In practice, very few "negative" citations, but these exists. How do you like this citation: "The authors of the work X completely misinterpreted own experimental data ..." But even such citation will added in citation system. The classic example of Évariste Galois (with H=1) is very descriptive.

As for the "reads", then this is not a criterion at all. It only means that the author chose the "fashionable" title. This is important, but does not determine the quality of the article. Who considers "re-reads", but these, in my opinion, better describe the level of the article.

There is nothing to say about recommendations, they reflect only "general opinion", but in science the truth is not established by a majority vote.

In my opinion, there is no objective criteria of scientific impact. Scientific activity can be measured. But who proved that squirrel is "better" than a gopher?

Dear Vadim S. Gorshkov :

Thank you for your highly informative and detailed Answer.

Your explanation of positive and negative Citations is exemplary.

Scientific research impact can be evaluated using alternative metrics, but there are substantial differences regarding their meaning and usefulness.

Dear Professor Chia-Lin Chang. Your question is indeed noteworthy. I might offer a different view about this issue. I currently work as a government bureaucrat, but I enjoy doing research. In my work, I make sure that every policy recommendation I made is research based, so that the government may produce an appropriate policy. Accordingly, reads, recommendations, or citations do not really matter to me. I measure the quality of my work by its impact on shaping the government’s policy. Still, reads, recommendations, or citations surely make me happy.

Dear Yudhi Dharma Nauly :

Thank you for your clear and helpful explanation about the important impact of research in shaping government policy.

Recommendation and h-index could be the most useful measure of the quality of academic research. The benefit of h-index is that it converts number of publications and number of citations in an identical benchmark.

Dear Mohammad Hassan Shakil :

Thank you for your helpful Answer regarding Recommendations and the h-index.

The h-index is a matrix representation that accommodates Citations and Publications.

Citations, Reads and Recommendations do not take account of Publications.

You are welcome Professor Chia-Lin Chang.

Dear Mohammad Hassan Shakil :

Thank you again for your interesting, informative and polite Answer.

My suggestion is the number of citation , recommendation and reads that is reflect the quality of paper.

Dear Fares F. Fares:

Thank you for your carefully ordered and thoughtful Answer.

H index and citation

Dear Amal Khudair Khalaf :

Thank you for your helpful Answer regarding Citations and the h-index, which depends on Citations.

The h-index is a widely-used measure of academic research quality as it balances Citations against Publications.

Comparable h-type indexes could be developed for Reads and Recommendations as they are both related to Publications Reads and Publication Recommendations, respectively.

Dear Michael John McAleer:

Thank you for the innovative suggestion of extending the concept of the h-index, which was developed to balance Citations against Publications, to balance each of Reads and Recommendations against Publications and Q&A.

It is clear that many academics do not understand the concept or definition of the h-index as a matrix representation of Citations against Publications.

On the other hand, TRI is a linear combination of four factors with given weights.

Anyone who understands elementary school arithmetic can calculate TRI and also understand its meaning.

TRI is indeed easy to understand as a measure of academic research quality as it is linearly dependent on Citations, rather than being a matrix function of Citations and Publications.

Dear Sajda Taha Mahmood :

Thank you for your interesting and helpful Answer in that all three metrics are useful in evaluating the quality of academic research.

It is necessary to define "academic excellence" to be able to measure it.

Academic research excellence metrics are always arbitrary functions of Citations, with no mathematical optimality properties.

The most widely-used function of Citations is the h-index, which is a matrix configuration that balances Citations against Publications to measure the quality, productivity and scientific impact of published academic research papers.

TRI is an alternative arbitrary function of Citations, and uses a linear equation with given weights to combine Citations, Publication Recommendations, and Full-text and Other Reads by RG members.

Q&A Recommendations are excluded from TRI, as are Publications, which differ from Publication Recommendations.

TRI will gradually become more widely accepted as an alternative measure of academic research excellence, primarily as it is heavily dependent on Citations.

A tricky question indeed.

On academic point of view ALL are important. But on usefulness point of view READ is probably the basic point to measure quality followed by its RECOMMENDATION and application. CITATION is more an academical measure than others. We cite a research article to enrich our article by incorporating the scientific logic in our s, without even applying it on personal level or directly promoting it in an wider field.

Dear Amit Baran Sharangi :

Thank you for your detailed, helpful and informative Answer regarding academic and scientific quality.

Unlike the RG score, Reads, Recommendations, Citations and TRI cannot decrease, so academic research effort is always rewarded.

I will go for citations.

Dear Michael John McAleer:

Thank you for your comparative and informative Answer regarding the RG score and TRI.

Dear Mohammad Hassan Shakil :

Thank you for a succinct and helpful Answer.

You are welcome professor.

Quotations first, recommendations II and finally read

Dear Sajda Taha Mahmood :

Thank you for your interesting and precise one-word Answer (I presume "Quotations" = "Citations").

Dear Djaafar Zemali :

Thank you for your ordered responses of Quotations ( = Citations), followed by Recommendations and Reads.

Neither Actions are more valuable than words. Did the reader take action and do something positive because of your publication??

Applications!

Kindly read this useful link:

What the Research Interest score includes When researchers read, recommend or cite a research item, its Research Interest goes up. Based on our data and feedback from scientists, we chose to focus on these interactions to reflect the life cycle of a scientist's increasing interest in a piece of research. First, a researcher accesses a research item. If it sounds of interest, they may read the full-text (if it is available). If they like what they read, they might recommend it. And if the work is really relevant, they might cite it in their own research.This is how we decided on a system for weighting the different forms of interaction:

• A read* has a weighting of 0.05.
• A full-text read* has a weighting of 0.15.
• A recommendation has a weighting of 0.25.
• A citation has a weighting of 0.5.
• https://www.researchgate.net/application.researchInterest.ResearchInterestHelp.html

Dear Rick:

Thank you for a helpful and interesting Answer.

Citations require the following actions:

(1) academic A writes a paper;

(2) academic A cites a paper by academic B;

(3) academic A publishes the paper in the Web of Science / Clarivate Analytics, Elsevier Scopus, RG, or an alternative international rankings website, so that the Citation of the paper by academic B is recorded.

Dear Carine Temegne Nono :

Thank you for a clear and succinct Answer.

Dear Asaad Sh. M. Alhesnawi :

Thank you for the reference to the well-known RG website on TRI.

The citations is better

Dear Fatemeh Taghizadeh :

Thank you for your helpful and useful Answer.

I think Citations are Dominant on calculate research level. However Reads and Recommendations also favorite because if your research is high scientific quality more people read and recommend it.

Citations

Dear Elyor Berdimurodov :

Thank you for your interesting ordered Answer regarding Citations, Reads and Recommendations.

Dear Djaafar Zemali :

Thank you for your unambiguous Answer regarding the dominance of Citations.

In my own opinion, I will also go with this order: Citation; reads; recommendations.

Citations.

Interesting contributions

Dear Olutosin Ademola Otekunrin :

Thank you for your two Answers regarding the clear ordering of Citations, Reads and Recommendations.

Dear Samir M. Ismail :

Thank you for your clear and unambiguous Answer with respect to the dominance of Citations.

It is citation, but sometimes a paper may be read and not cited, that does not mean it is not of quality or less impactful. Also, if a paper is sincerely recommended based on its perceived quality, that is an applause.

The most useful measure of the paper quality is citation.

Citation

Citation

Citation answer is dominant than other on above discussion.

Dear Kamoru A. Adedokun :

Thank you for your detailed Answer regarding Citations, quality, impact, and applause.

Dear Latifa ZHOURI :

Thank you for your useful and helpful Answer regarding Citations.

Dear Elyor Berdimurodov :

Thank you for your succinct and informative Answer regarding Citations.

Dear Chuck:

Thank you for your succinct and unwavering Answer regarding Citations.

All of the above is important

Dear Sajda Taha Mahmood:

Thank you for the encompassing Answer.

Dear Afraa Ibrahim :

Thank you for your all inclusive Answer.

Citation by other authors is important in publishing published study results.

In my opinion, citations and h-index are the only credible quality measures for academic research.

Acceptance and implementation by industry.

Value to industry / Practical solution to a problem

I agree with Muhammad Saad Khan and do not agree with those to who talk by industry. They talk in a narrow sense because industry constitute a small percentage of the society.. Further, if one has to wait until industry accepts and implements, the march of science will die. Their views are unrealistic and narrow.Most inventions or discoveries are first discussed in papers and would take years and other peoples inputs before they really come to light. I guess Adam Smith theory of capitalism and Schumpeter analysis did not help us understand the economy and build industries. Or maybe industries came and approved their work!

Dear Djaafar Zemali :

Thank you for your helpful Answer regarding Citations.

Dear Muhammad Saad Khan :

Thank you for the informative Answer about Citations and the h-index.

Dear Rick:

Thank you for the instructive Answer about acceptance and implementation by industry.

Dear Kaukab Abid Azhar :

Thank you for the Answer concerning the value to industry and a practical solution to a problem.

Dear Chuck:

Thank you for your contrasting Answer with respect to the importance of scientific discoveries and their acceptance by industry.

Citations by other authors has been and is the standard all over the word

I think Citations are dominant on calculate research level.

Dear Chuck:

Thank you for your very clear Answer regarding Citations.

Dear Elsaid Elashry :

Thank you for your direct Answer regarding Citations as the best metric for evaluating academic research.

Reads and Citations

Dear Elena Tarkhanova :

Thank you for your helpful dual Answer of Reads and Citations.

Reads, recommendations and citations are all essential points of qualities of a good paper

I would rank citations higher than the other measures

I agree with Dr. Augustine S. Kukah!

Recommendations and reads are other ways of capturing the quality of paper because they can re;late to published. unpublished papers

Dear Augustine Senanu Kukah :

Thank you for your informative and helpful ranking of Citations, Reads and Recommendations.

Dear Chuck:

Thank you for the insightful discussion of Recommendations and Reads as capturing the quality of published and unpublished papers.

Both reading, recommendation, and citation are important indicators of the quality of scientific research and the characteristics of a scientist.

On the other hand, the citation of scientific works of a scientist by other authors has been and remains an important standard in assessing the scientific level of a scientist.

Dear Liliya Bazylyak :

Thank you for your detailed and helpful Answer regarding scientific research quality.

The h-index is a matrix balance of Citations against Publications.

An h-type index separately for Reads and Recommendations, h(Reads) and h(Recommendations), respectively, would be useful as they would be matrix balanced against Publications, and directly comparable with the h-index.

Number of citations and h-index

Dear Chuck:

Thank you for a helpful Answer regarding Citations and the associated h-index.

Dear Michael John McAleer:

Thank you for an innovative suggestion regarding h-type indexes associated with matrix balancing Reads and Recommendations, respectively, against Publications.