@Ewa: You mean there is something basically wrong in the system ... or rather recommend to try taking directions sparing failure reports.

I fully agree, what is understood as failures are not such. Sure, when taking into consideration what actually was aimed with a research (as it might be target-oriented), not reaching such aims is considered being a failure. But if research would be understood as such there would not be any failure at al - just research results that promote a hypothesis or don't. In the context of basic research this is nothing to discuss about. As far as I understand the situation, it is this context of target-oriented (project oriented) research that makes the situation so complicated. In basic research, any results are being considered valuable results ... but who in todays' times still has the funding to conduct basic research ...

Somehow it is strange: When one is in a new city and wants to reach a certain spot a comment from a peasant that one is taking the wrong direction is thankfully accepted. Going on with the wrong direction until finding this out by oneself may cost 10 minutes. Thus, the information led to 10 minutes saved time. If a research method does not lead to a suitable result according to a particular research problem, a hint regarding the possible wrong path may save a year of useless efforts or even more - but communicating is damaging? How do researchers get along with such problems? Simply accept them as maters of facts and repeat failures?

Thomas, you are right! In executing projects, we need more publications and in case of problems with concrete used methodology negative results is not posible to be published and we try to find any positive in our study! This give wrong information to the next scientists, following us in similar studies!

There are tons of examples, be it in the pharmaceutical industry (where usually research takes as long as is needed to find enough positive results) or in the field of humanist or nature science but also technological research. On the one hand, we propose not to newly invent the wheel but on the other hand, we force others to do exactly that, go the hard way and repeat failures that have earlier long been done.

Is a change of research and particularly publishing culture needed and how could such be initialized?

Thomas, I am shure that science system for publication and promotion of best ideas must be improved, if we wish to reac to community based of knowledge!

Ewa, in all scientific fields have pseudo-positive results, I suppose. This is cases when an amount of money are spend and for report we need publication, Then we publish such pseudo-positiwe result not sown that our direction of studies was wrong(editors of journals not published such papers)

My thoughts are also in the similar lines of what @Ewa Lipczynska-Kochany had written above. The amount of literature coming out now-a-days is huge in every field and getting updated with the same itself is consuming lot of time and energy. I suppose the idea of publishing "negative results" may further complicate the situation. Definitely, there will be some exceptions to this..depending on the field of study..

Sure Ewa - let us say that a scientist from the 2nd century proposes the 'reasonable' hypothesis that the Earth is the center of the Universe, with the Sun and other planets revolving around it. If that hypothesis is correct then each day the Sun will appear at one end of the sky and disappear at the other end. The prediction of that hypothesis (A) is indeed 'verified' each morning but, as we all know, this is not a decisive test since the same prediction can be made for an alternative hypothesis (B)- and a much better one, as your compatriot Copernicus was convinced in the 15th century - that it is the Earth that revolves around the Sun. But it took another century and the technological emergence of suitable telescopes, for Galileo to provide supporting evidence for B. So 'favorable' results are comforting but not always decisive.

On the other hand, disproving an hypothesis (by obtaining results clearly contrary to the expectations of that hypothesis) will be decisive in eliminating that hypothesis from the spectrum of possible explanations. Which is definitely worth publishing. For instance take the hypothesis that the 'homing behavior' of salmon (this fish find its native river after years spend in the open ocean hundreds of miles away) thanks to visual cues. Well, if that is correct, cutting the optical nerve will prevent the salmon to properly navigate to its home river. This experiment was carried out and resulted ('negative results') in the blinded fish returning home without problem. Great! You have eliminated the 'visual' hypothesis (which is highly publishable) and can propose another instead, based this time on 'olfactive' cues: i.e., you suggest that the salmon picks up the 'right' river based on olfactive memories. To test that, you experimentally block the olfactive sense of the fish. Bingo! Those fish cannot find their way back home. Your olfactory hypothesis is comforted. Further experimental tests (this time by manipulating with synthetic chemicals the scent of a salmon river) confirmed 50 years the validity of the olfactory hypothesis for the homing of coho salmon.

I have been in Shanghai for a week without access to this platform ... and in the meantime, the discussion starts getting vibrant. However, the fact itself makes me happy.

Thanks, Frederic for your supportive arguments which fully meet what I had in mind when I raised this discussion.

From my perspective, there is too much irrelevant literature available and really relevant stories (such that bring research further) are missing ... the failure stories. This, on the one hand forces us (as researchers) to put much more energy into the selection of relevant papers (Ewa's mentioning of getting too complicated) , on the other hand, the outcome is more or less the same to before (the explosion of offered papers) but many relevant questions still keep unanswered. What means that more literature does not necessarily make the result more relevant or helpful. To be honest, I am sick of listening to little contextually limited success stories on conferences without hearing anything about the limitations of these studies.

In my own research, I have the aim to reduce cultural conflicts in international educational scenarios. I would say I did my quite painful (because of the massive number of publications from various fields) homework, e.g., by analyzing reports on (general) cultural conflicts in educational scenarios (which generally led me to aspects that at least are worth being further investigated). Basing on such aspects, I developed a learner-focused questionnaire, translated it into various languages (still ongoing) and conducted it in various national contexts (still ongoing and further support is very welcome). So far ... now I have results that show me that indeed, learners from different contexts show cultural differences regarding several aspects of behavior and expectations in educational processes. Those results are sufficient to help educators building up cultural competences in terms of achieving an understanding that such conflicts are not generally their fault but happen because of the differences. They also may be able to conclude which strategies are suitable in order to prevent or solve such conflicts. However, what still is absolutely unclear is how large such a difference needs to be that a conflict may occur (level of acceptance) and how the various aspects (102 items) are interrelated. At which point is the level of differences relevant for the implementation of changes (and does such a point exist at all)? In terms of developing a recommender system, this is a crucial question to be answered.

What I would need at this point are reports on very particular conflicts that (repeatedly) have been monitored; conflicts that occurred between educators/learners and/or learners/learners from different contexts, and if any kind of attempt has been made to solve the conflicts, what has been done and what was the result? My very optimist idea is that with a large enough number of such reports, a clustering should be possible that allows us to find patterns which enable us to forecast conflict potential and implement preventive steps (such as preparing students, encouraging students to deal with expectable differences instead of dropping out, adapting learning materials to provide less conflict potential, etc.)

However, the response rate on my various calls for contributions in different media was close to zero. As a possible explanation I was told that it simply is unlikely to report failures (In educational scenarios, not being able to solve cultural conflicts might be understood as a failure). I wanted to know, what I can do in order to to encourage educators to report such issues without fearing to loose anything ... what can I do to overcome this particular conflict?

Thank you all for your contributions in this discussion,

yours,

Thomas

@Ewa: You mean there is something basically wrong in the system ... or rather recommend to try taking directions sparing failure reports.

I fully agree, what is understood as failures are not such. Sure, when taking into consideration what actually was aimed with a research (as it might be target-oriented), not reaching such aims is considered being a failure. But if research would be understood as such there would not be any failure at al - just research results that promote a hypothesis or don't. In the context of basic research this is nothing to discuss about. As far as I understand the situation, it is this context of target-oriented (project oriented) research that makes the situation so complicated. In basic research, any results are being considered valuable results ... but who in todays' times still has the funding to conduct basic research ...

Somehow it is strange: When one is in a new city and wants to reach a certain spot a comment from a peasant that one is taking the wrong direction is thankfully accepted. Going on with the wrong direction until finding this out by oneself may cost 10 minutes. Thus, the information led to 10 minutes saved time. If a research method does not lead to a suitable result according to a particular research problem, a hint regarding the possible wrong path may save a year of useless efforts or even more - but communicating is damaging? How do researchers get along with such problems? Simply accept them as maters of facts and repeat failures?

Thomas, the new question you raised re: cultural context vs educational expectations is a serious one. My own post focuses on the perceived 'failure' (in certain circles) of a falsified research hypothesis, whereas such a finding should be considered as added value to science instead. It is important to report/ publish that a certain research hypothesis is no longer "on the menu" as soon as experimental results / or observations based on new technologies convincingly reject it. This is worthwhile science as you remark with your metaphor on saving precious navigation time.

Now Ewa remarks that a scientific system may not wilingly accept a paper (and her/ his author) that disproves a fashionable paradigm. True: the history of science provides repeated examples of bright researchers who were so ahead of their time that they suffered in their career as a result (think of Alfred Wegener pioneering theory on plate tectonics; of Charles Darwin's long hesitation to publish his revolutionary vision on the evolution of species ...) But this blind, 'immune response' of the academic system to accept radical novelty remains the exception, not the norm - fortunately. And in the end scientific truth will triumph anyway. Even if it takes time ... even posthumously.