Publish Yet Perish?

Increasingly, the notion that seems to backfire is “publish or perish.” Publishing research papers in all areas of generations of knowledge (i.e., disciplines) seems to have updated its course. Lately, however, what seems to dominate the business of the generation of knowledge is the attribute of expansion. Expansion in volume and numbers; still doubtful of the quality, though. What seems to change its course is yet an obvious question that research organizations, universities, and other educational institutions increasingly face is the question of “Why fund research?”. Increasingly, private players' intervention seems to impact the existing research landscape worldwide. Looking at just the discipline of economics, the below graph from Aigner, Greenspon, & Rodrik (2021) demonstrates the irrational yet exponential growth in several journals and published research papers over the last four decades. 

While growth in numbers is evident, a question of perspective remains. As follows: [1] Are journals seen as the places of peer certification of quality and an intersection of supply and demand of reviewed scientific evidence, OR are more presented as means of individual quality signaling for demonstrating competence, and nobody cares about the audience and impact? [2] In the debate of quality vs quantity, is quality demonstrated through the quality of work or the quality of journals endogenous to several exogenous factors? [3] With more evidence on the role of networking in publication, frauds, etc., how much trust is left in the actual quality of work against the apprehensions of the general academic audience? [4] Who should fund research? Especially in developing countries and disciplines that don’t visibly contribute towards the betterment of the society with a huge scope for individualistic interpretations like social sciences and humanities. Is it the government? Or Private? [5] Who are the researchers writing papers for? For the academia that has invented innovative (least time consuming) ways to read a paper enough to cite it and promote the concept of “base paper”? Or for the government, which is confused with so many papers on the same topic, a new genre of papers, i.e., meta-analysis, has emerged to scale down similar findings to one or few? Or is it for the general public, who has shifted its limited attention towards short reels and long podcasts, meaning no time to go through papers and their supplementary materials? [6] Last but not yet least, are we at the peak of the “publish or perish” regime? I will argue in the affirmative of the last question while answering the others, with a contextual focus on the discipline of Economics.

            Focus on the two interesting, obvious graphs from Card & DellaVigna (2013) below. It might sound like a repetition of the expansion notion we discussed earlier, but it is and is not. In the figures, the authors demonstrate that since the 1970s, the top journals in economics have experienced sporadic submission rates (top figure) along with a steadily declining number of accepted articles in the top journals (bottom figure). What is interesting is the contrast between the previous figure and these two. It presents the inequality in publication prospects. However, we are here to talk about something different. It is a fact that, though the number of published articles is falling in the top journals, the total absolute production of articles in Economics is increasing. It is mainly because, with increasing decentralization, it is not difficult to get and run a journal these days, and most importantly, with the increase in volume, the desire for quality is also increasing. It is difficult to realize, but it is happening. If it does not, it will happen rapidly in the coming years. 

Why? I am not assuming that only top journals produce quality research, but it is safer to trust human intention that a good researcher with a good paper will try to publish in a journal that ensures a better reach. Increasingly, it is perceived that researchers succumb to petty regressions and variable permutations to get statistically sound findings (under very strong assumptions) without much sense of novelty. In disciplines like energy, pollution, and environment, where variables are new and can relate (impact in some other fashion) to many variables, papers are written and published with little research rigor and evident contribution to the literature. At times, this has taken publications to higher ranking journals through not research but rather academic means of authorship and other institutionally advantageous prospects. Undeniably, every coefficient is a piece of research, and every research is publishable, but for whom?  

For whom? If the question is,” For whom are researchers writing those papers?” answers could be many. Firstly, for fellow researchers in the domain. It is evident from several research findings that researchers increasingly don’t read the papers they cite. Using the term misprint, Simkin & Roychowdhury (2003) write that. As a result, misread articles or because of not following the original source misprints are rampant in research. The authors explain the problem using the figures below. Figure 3 (left) shows a distribution of misprints in citations to one such paper [10] in the rank–frequency representation, and Figure 3 (right) presents the same data in a number–frequency format. The authors conclude that misprints in scientific citations should not be discarded as mere happenstance. Greenberg (2009) uses the term citation distortion to highlight similar facts.

While it isn’t obvious, it increasingly doesn’t serve the purpose of writing long research papers, following which there is an evident homerun about shortening the paper size. Incredibly, several leading journals are offering small paper publication prospects to authors. What is very striking, though, is that, without strong theoretical contribution, intelligent methodological usage, or curation of new or novel datasets, the exercise of assimilating econometric methods and generating tables and interpreting succumbs to mere data science exercises and not thorough economics research. Maybe that is why petty publications are about to lose their credibility of numbers soon with the advent of GPT and LLM tools in research. Over the years, it might happen that such technical invention is left to the task of data analysis that doesn’t require much innovation, thus taking away the scope for petty regressions and, thus, frivolous publication prospects. It may very well happen that, policymakers don’t look-up to research papers for curated coefficients, rather they try estimating them themselves.

Who should fund research? This question is apocalyptic, as it could be to any individual researcher, as it deals with the stipends of research students, salaries of faculties, logistics, and whatnot. In my view, free research is something you do while living on your retirement deposits before dying a theoretical death. Increasingly, two specific trends are spotted in the genre of research funding, and they both have their hits and misses at the research objective. First, if you expect the government to spend on R&D, the government asks, “Why?”. More specifically, there is an increasing inertia among the governments with regard to funding specific types of research that doesn’t suggest something worthy implementation, rather ends up question the very funding agency i.e. government. Such is the case of hesitance towards funding social sciences in several countries around the world. The governments want to focus more on physical sciences and engineering that gives tangible growth in innovation around the country rather that different verbal ways to questioning the being of the governments.

Second, if you expect the private entities to support your research it should have some meaning to the profit or altered profit motives. As seen in disciplines like Economics, private thinktanks have emerged in large numbers, but with their defined agendas. This results at times in them validating and publishing research that is biased and unscientific.

With these, it is true that, the question of who should fund research remains. Moreover yet, what is well answered is that the condition ex-ante and ex-post funding acquisition is providing results worthy replication, validity and implementation. This is where, the mushrooming publication business might see a downgrade. We are at a high time to realise that, at the advent of this AI age, people are trying to avoid a research paper with, Introduction, Literature Review, Gap, Objectives and Findings of 30 pages. Avoidance is a must when they just need the magnitude and sign of a few (or a single) coefficients. Thus, unworthy papers, with unworthy length just adding another coefficient of average relationship is definitely not going to mean much funding in future.

The questions [1],[2],[3] regarding the changed role of publications in in the research landscape (specifically the case of petty publications), the debate of quality vs quantity in research output and the ethics of publication orientation, draws directly upon the signaling thesis of Spence (1978), but I am not sure publications as a signal of research output has bettered or worsened the situation. In his seminar, Spence argues to get rid of asymmetric information in a job market, workers will be inclined to signal their competence in form signals. Coming to the academic job market in India, someone’s research aptitude suffers a usual limitation of being intrinsic, unobservable thus always in search of better signals. But with publications the asymmetric game used to get solved easily. But with mushrooming of publication houses, widespread lobbies among writers, reviewers and editors, psychological and systemic biases and lot many things, number of publications doesn’t stand much count today. If this statement is not true now about countries like India, in five years it will be.

Though the prospects of publication initially started with a goal of dissemination scientific knowledge to a greater audience and to record the contributions of thinkers in the current landscape it seems to have changed. This ethos of publication is only maintained in the higher stratas of elite researchers who with their hard work have secured a living and can now dedicate to the knowledge creation of the greater good. But they are handful, and the dirty reality is that it has become a rat race. Being at the top comes at an ease of publication that always keeps you there however securing a place there from the bottom of the pyramid is no ballgame.

Last words of this essay could be just that, doing good research of quality takes time but the rat race takes away the patience. In countries like India, clearly it is an exclusive choice that, you must choose between, waiting for quality results or publish petty results. However, this choice, what I suggest will make the market for quality research more difficult in coming years. It might get late to reach the downtrodden regions on earth, but it eventually will. As the number of petty publications skyrocket, lesser people will read them and with a question arising on funding, eventually research will emerge as an elitist profession rather than something that everyone with keen interest in knowledge creation could pursue.

References:

AIGNER, E., GREENSPON, J., & RODRIK, D. (2021). The Global Distribution of Authorship in Economics Journals. https://ssrn.com/abstract=3960280

CARD, D., & DELLAVIGNA, S. (2013). Nine facts about top journals in economics. Journal of Economic Literature, 51(1), 144-161. DOI: 10.1257/jel.51.1.144

GREENBERG, S. A. (2009). How citation distortions create unfounded authority: analysis of a citation network. Bmj, 339. https://doi.org/10.1136/bmj.b2680

SIMKIN, M. V., & ROYCHOWDHURY, V. P. (2003). Read Before You Cite!. Complex Systems, 14, 269-274. https://doi.org/10.25088/ComplexSystems.14.3.269

SPENCE, M. (1978). Job market signaling. In Uncertainty in economics (pp. 281-306). Academic Press. https://doi.org/10.1016/B978-0-12-214850-7.50025-5