In my opinion all of them are possible source. It depends on the subject and the topic. A lot of theses have unpublished material inside. On the other hand, the really hot spots inside theses are published in journals or conference proceedings. As a result, there is no exact answer to take for granted.

If you find a SURVEY on the topic or a ROADMAP on the topic, and published in a good IF journal, it will be one of the best guide papers.

I agree that we can find good reviews in journals and conferences looking of high quality. But I believe we can find a lot of valuable information in theses and dissertations, because articles in the information is objective. However theses and dissertations usually describe the steps from start to the achievement of results.

During my dissertation, I found lots of important information in theses and dissertations (searched them based on published articles).

PS: Sorry for the mistakes in English

Dear John K. Marco Pima, Theses and dissertations in parallel work provides you a better and quick way of understanding. However, Theses and dissertation are usually not documented globally. Also, a researcher by the end of the submission of theses and dissertation would have published some papers.

So, in my opinion give wait to journal / conference papers in your review work.

Thank you for sharing your experiences. Until now,the two sources are giving me a lot of relevant data. However, for the most recent, I have found journal articles to be more useful than theses. Thanks.

For official work these are the things that have to be referenced. However, for bleeding edge, in software engineering at least, the blogosphere of the consultants who are interacting directly with the needs of customers provide a much more current view of ideas and results. This includes both individual blogs, communities of interest (Linked-in, Yahoo, Google, etc.) and archives of good blogs sometimes offered by industry or interest groups. Care must be taking in referencing them alone, but some interesting statistics can be developed on the use of certain terms or the discussions. Unfortunately, in fast moving disciplines, the formal literature can be months or years behind. Sometimes this is a plus, but it is often annoying.

Let me quote from an answer that I provided to a different question:

"Let me ... criticize many of the recent ("so-called" structured) literature reviews that focus on common data bases, works with Impact factors and alike. But they neglect grey literature or sources. If there is an interest on the topic and knowledge about ideas in grey literature, this must be inlcuded. So, if a good idea in a thesis, let it be a starting idea from a supervisor or a student or an extension or modification from the same persons, then get it in.

Food for thought, indeed. We should not primarily be interested in getting papers into the CV, we should strive to advance science."

Where can we find grey literature related to computer science?

In my opinion Theses, Dissertations and Journal publications will contribute a lot towards Literature Review. I do not relay much on Conference proceedings as they mostly wont carry the entire data what we need. Theses and Dissertations are best choices, but reviewing it may consume a lot of time at the same time these may carry entire information about a particular research and are more reliable. Another problem with Theses are that, they wont be documented/ available globally. Journal publications on the other had would carry sufficient information that may be needed for one's research but wont guarantee full information. So if i list my choice of selection of material for literature review, it will follow the following pattern 1. Theses and dissertation, 2. Book chapters, 3. Journal publications, 4. Conference Proceedings, 5. Reliable internet sources...

Dear John K. Marco Pima,

PhD and Masters theses and Dissertations are better sources for someone who is working in the same research group(area) since it will be easier to understand and clarify while doing the reserch literature review.

Only in selecded cases of high quality up-to-the-date PhD and MSc theses covering full topic.

Review journals are very important and you can solicit them.

Dear John K. Marco Pima,

In general reviewed journal papers are better source of references in review work because for readers it is easier to trace / acquire the documents and hence more transparent source of referencing.

PhD and Msc Theses are solicited, because the most of thesis includes a part related to literature review.

@ Pima: Certainly because these are considered to be the primary sources of information. So, it is always better to give priority for theses / dissertations while doing literature review.

PhD thesis has no stringent page limit as compared to research articles. Therefore, PhD thesis might have more information than associated published papers.

Literature Review in Dissertations, peer reviewed journals act as a spring board for reference only to use ones own ingenuity. However, status queued following of a literature review by another fellow would hamper independent thinking process that may mar the value of the subsequent research and independent findings.

Yes Maya Daneva, I agree with you on "a systematic approach" which is a requirement for all (Masters and PhD). What about rigorousness of the process?

Journal and IEEE transactions are best resources related to the research areas. But Masters and PhD Thesis can be important resource but it should be very close to research topic for which Literature review is being taken.

For students undertaking Masters Thesis, i think they need to use more journal papers given the level of depth their thesis require. For the PhD theses a combination of all sources is needed and to broaden the sources of the Literature and the depth of the study.

This was my understanding too that only journals and thesis are enough for literature review. But I have discovered that if you are really into good research work and innovation then you have to go through patent databases too. Mere Google-ing won't help at all. Most of the patent databases are free and should be searched before completing literature review phase. This will not only help you in giving you a near to perfect state of work done but also a great confidence to innovate too. All the best.

Peer-reviewed scholarly journal is the highest quality and accurate.

Please refer to the attachment.

Social Research Methods : Qualitative and Quantitative Approaches by W. Lawrence Neuman (2011, Hardcover, Seventh Edition, International Edition) page 128.

I would say dissertations only offer a direction for drilling down to the roots of the concept by the original gurus. That would require opening up their books / material for grasping their original thoughts before articulating in own literature review

Aatif Saif,

It is true that Dissertations offer a direction and if you make follow up on the references, then you get the most of it. I have tried this and it has worked for me to supplement other sources.

Well, in order to have a good literature review, one should look for latest review papers from journals. Review papers cover in-depth presentation of past, present and future works done in a particular research field. They also mention the challenges in doing the future works.

You are right Prashant, review papers provide up-to-date review of the applied methodology and literature trends.

Dear John K. Marco Pima

PhD, Masters: Theses and Dissertations provide us a quick access to methodologies and tools along with scope of future works. Since these documents are quickly available for research groups at same institute. Hence, referring these are essential if research idea/methodology/tool/comparison benchmark is used.    

Since Theses and Dissertations are not accessible or publicized immediately therefore, it is not commonly referred by other researchers of the same field. Because most of the the research papers are from the Theses' and Dissertations' work, so generally, referring of these are not prime issue. 

Conference papers are commonly discusses new ideas, tools, methodologies, technologies and comments but not to the extents of completeness. Some good conferences adhere a standard like IEEExplore. Standard conference papers and /or conference papers designed from same research groups / fields are justified for referencing.

Standard journal papers are the best source of breakthrough of your research and make a practice of it and therefore, this the best source of reference. 

Are PhD/Masters Theses and Dissertations or Journals/Conference proceedings the best sources for a Literature Review? Why?

In my personal view, all of them i.e. theses / dissertations & journal articles / conference proceedings are good sources for academic research & literature review.  However, each of them have their own unique characteristics & use cases as shared below:

Also want to add - besides theses / dissertations & journal articles / conference proceedings, sometimes references at the end of mainstream academic books also serve as good sources for researchers to conduct further literature review.

All above and cross references, in addition, prove to be helpful, while writing dissertations.

I always go for quality journal articles because they provide concise insights about the research topics I am pursuing. Amir

Journals/Conference proceedings are the best sources for a Literature Review

Journals will be good source especially Review Articles