I would like to hear your views on the literature survey in a paper:
1. Do you find it fun to write, or not?
2. Do you find that it is very important in every paper?
3. Do you find it hard to write it, or it is easy to pick the right papers?
4. What is your strategy/goal when searching for papers to cite and summarise?
5. How long back in time should one go?
6. Feel free to add your questions, if you can also answer them! :-)
/Michael
@Michael Patriksson For me, literature survey is no a fun. I consider it important for every paper. Nowadays, it's easy to pick right papers. While hunting papers, my emphasis lies on recently (3-5 years) published articles in the concerned field. However, I don't ignore to cite the original basic work (references) as per necessity either in methodology or discussion even if such work/ references are more than half century old.
Michael Patriksson i find that literature survey is important and at the same time weird.
I assume that we are talking about a fairly new PhD student, who has not read much, and who have been asked to gather many good papers from the last decade or so from good journals - but not open access-only papers.
If you are a beginner, you must discuss with your main supervisor about how to gather papers - new as well as far back in time. Categorize them, and keep notes on the papers, so you can check which ones "belong together". You should of course check with your supervisor regularly, to make sure you're doing it right.
Let me know how it goes - perhaps I can browse a survey from you and give you hints!
Do you find it fun to write, or not? It is okay for me to write as it is my passion.
Do you find that it is very important in every paper? Yes, it is.
Do you find it hard to write it, or it is easy to pick the right papers? It depends on the topic you want to write in, and usually depending on the mood you are taking.
What is your strategy/goal when searching for papers to cite and summarise? Again, that will depend on the topic you want to write, basically, composition.
How long back in time should one go? Depending on the mood you are taking.
1. Do you find it fun to write, or not - yes, it follows from already done job on getting knowledge on the topic.
2. Do you find that it is very important in every paper? NO. Many introductions are generall and not well adressed. However it is important to address well the background of your research.
3. Do you find it hard to write it, or it is easy to pick the right papers? it follows from already done job on getting knowledge on the topic.
4. What is your strategy/goal when searching for papers to cite and summarise?
Check newest papers in the field, go to the milestone papers/source papers, include some summary based on review papers
5. How long back in time should one go? source papers (even from 1905) should be cited. If somebody reported some new phanomena, he should be cited instead of newest review paper on this topic.
regards,
GB
Yes, I love surveys and find that I personally learn alot from them... for others they serve as a sort of baseline for an area that they want to explore, like How many EAPs are there in the US; how many locations have support services for Sexual Assault/Hassment victims etc,,.... Yes, please keep doing surveys - very helpful in many ways...
Literature survey is very very important in my opinion and if it is nicely written , it is always fun reading it. I think most recent findings are the most important ones.
Surveys should also be a walk past older publications, so that the readers - whom may be slightly less aware about the development of the topic - can follow more easily what YOU are doing.
Dear Michael Patriksson
Thank you very much for opening this discussion. When I have read your papers or ones written by Professor Adam Letchford, a question raised in my mind: How can one write these nice literature reviews? Is there any systematic way to do that?
Ehsan Monabbati: Sure there is!
A few pointers: try to categorise the papers in a natural way: not all papers that you have gathered and read will be of the same style; older papers will of course be marked by time, if you are dealing with a rapidly advancing field. Some old stuff (or new) might even be wrong, and those you should deal with especially.
Make sure that you have found the best recent survey papers, so that you have an idea on what is the state-of-the-art in the field. But also mention in the same detail what happened before. How long back in time you want, or need, to go to find papers depends on how large and old the field is. The survey should be written such that each paper you talk about is excellent, and also so that the sum of the papers' results and conclusions build a complete synthesis of the field that you are reviewing. This is not easy!
Also make sure that you have listed previous surveys (and a short text decscribing each survey's focus and results), so that you can focus on the latest and most interesting parts of your sub-field.
Make sure that you don't write much more about your stuff than about others' - it's impolite to concentrate on YOU. :-)
This is my five minutes of thought.
It would be nice to hear what you think, and also to see how your review is structured. I can tell you that 95% of the surveys I have read are way too short, have very vague structure, and do not really tell what is the most important bit in each paper. You can make a difference, if you work hard, and make sure that the better papers also get more room than the ones that are less informative. You should even skip those surveys or papers that are badly done.
Ok?
Let me know how it goes!
The passion for writing could be developed by practising on a daily basis. You mat start bey committing your self too write on a daily basis for 15 minutes. Gradually you will discover through practice that your passion for writing has increased and it is fun that you enjoy doing.
There are several items to keep track of while you write or make notes, and you need to have the connections between the sub-fields clear, so I think you need to spend more time than that. I think I can recall that I made notes on paper with key words that, for a given paper, I ticked the box if some aspects were there in the paper, or not. That would also inform me on the quality of the paper. Nowadays, after several hundred manuscripts reviewed, it's automatic. :-)
Citations about a project study varies with project.
There are several factors effecting any specific article to be cited in your work such as.
1- During your study you were a reviewer for taylor and francis online hence, your citation will generally come from that group.
2- During study time you had a limited time frame hence any relavent study will be cited.
3- During study you had ample time so you start with the begining of the variable hence, older studies will also be cited.
4- During study phase you were following a specific researcher who generally cites from a specific group.
5- You intend to submit article to target journal , citations will be mostly from that journal.
6- Your funding agancy has made certain requirements for you.
There are plenty more hance, it is a dependent variable with handful of independent !
Shafiq: You should not focus on one publisher because you would like to publish with them - the citations you choose must only be decided based on relevance!
No-one shall base the review on only one or two publishers' or journal's output, but the whole spectrum of relevant journals from all publishers. You are not allowed to favour one publisher against others - it is in fact immoral.
I can remember from some years ago that a few journals actually requested that papers from the journal should, "if possible", be cited in the article. That is an insult, to begin with - I know better than the editor which precise articles are relevant to cite - from the whole spectrum of journals. That is how it should be.
If a journal would tell me to cite their own articles, I would definitely never publish with them, that's for sure.
Samy: I would ask the newcomer to also actually read a book on how to write scholarly papers. There are plenty, and if the status quo here at RG is what it is, I would say that it is below par.
For researchers and teachers there are pretty good books on mathematical writing. My favourite is this, now in its new 3rd edition:
- Paperback: 353 pages
- Publisher: SIAM - Society for Industrial and Applied Mathematics; Third edition (December 16, 2019)
- Language: English
- ISBN-10: 161197609X
- ISBN-13: 978-1611976090
Hermann Gruenwald I find it fun to write literature overviews - and I am pretty good at it, too, I think. But I would never throw in papers that do not belong in the paper, just because a scrupulous editor tells me so. So far I have been asked perhaps five times or so, and I do tell them what I think of that attitude.
Michael Patriksson
Thank you very much for your time, and sorry for my too-late response.
Let me briefly state what I get from your insightful answer:
Writing a literature review is not listing all relevant papers! but, It is to choose carefully the key papers (including recent surveys) that cover the field, which itself depends on "how large and old the field is".
In my opinion, a good literature review should give a map of the field and the location of the current paper on it!
So, this is what I usually try to do for writing my literature reviews:
I begin by gathering all the relevant papers. Then, I write down the main contribution of each paper, and (after reading your answer) select the important ones. Next, I seek to do the most difficult step, which is to find a logical path that could connect the selected papers and to put all parts together to declare the map of the filed and the location of the paper on it.
Thanks again.
The review should also pinpoint threads of research that either still not exist, or should be explored more, perhaps because of necessity - like global warming, famine, corona-induced economic hardship in industry, as well as in the academy, and the list goes on and on.
Michael Patriksson here are a few short answers from my side:
1. NO (it's boring)
2. YES (the list of references is the backbone of the paper)
3. NO (thanks to SciFinder)
4. I try to cite the most important recent references but also give fully credit to the pioneers in the field.
5. This can be very different. In the paper cited below we included a reference dating back to 1863. The situation is different if some review articles can be cited.
Article Synthesis and Structural Investigation of Brightly Colored O...
Dear Michael Patriksson
1. Do you find it fun to write, or not?
Yes, as I can learn about the recent developments in the field and find new connections. I think its the task of an author to cite at least some references, which allow an easy entry into the field.
2. Do you find that it is very important in every paper?
Yes. At least a recent review should be mentioned.
3. Do you find it hard to write it, or it is easy to pick the right papers?
Sometimes it is hard. But during the writing I some times exchange some references to get a better fit with the intention of the paper (see 1.)
4. What is your strategy/goal when searching for papers to cite and summarise?
Databases. Back and forward citations and keyword search.
5. How long back in time should one go?
It depends on the topic. Beginning with the pioneering work (as Frank Edelmann already mentioned) and ending with the newest articles in the field.
6. Feel free to add your questions, if you can also answer them! :-)
- Should someone do a more or less randomized Google search based on one keyword or a chemical element ? NO!
- Should someone cite lab/university mates and/or prominent persons of you university even if the topic doesn't fit? NO!
Should the first sentences recall text book knowledge? No!
@Michael Patriksson For me, literature survey is no a fun. I consider it important for every paper. Nowadays, it's easy to pick right papers. While hunting papers, my emphasis lies on recently (3-5 years) published articles in the concerned field. However, I don't ignore to cite the original basic work (references) as per necessity either in methodology or discussion even if such work/ references are more than half century old.
1. It's boring.
2. Yes. some recent references should be included in the document.
3. Sometimes it's hard. It depends on the theme
4. I'm looking for recent work using keywords on the famous
Data base.
5. Less than a week.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing