Google Scholar launched in November 2004 with the goal of making it easier to access scientific knowledge. But did it live up to this aim or simply create new problems? Kyriakos Drivas shows that contrary to early concerns, the platform has made older papers more discoverable and helped knowledge spread more efficiently across academic disciplines.

The rise of the internet accompanied two transformational changes in research. The first was a massive increase in the volume of research papers. Based on OpenAlex data, the universe of scientific works grew from 1.4 million in 1984, to 5 million in 2004 and 8.9 million in 2024, marking a dramatic accumulation of knowledge.

This growth in research output posed a problem: how do we review this vast body of literature and navigate it without being overloaded with information? The answer was to rely on the second transformational change brought about by the internet: a massive reduction in search and access costs.

Online search platforms such as Google Scholar, PubMed, Web Of Science, Scopus and OpenAlex, among others, were created to reduce information asymmetry, provide completeness of information and supply researchers with tools for efficient navigation of science literature. As Google succinctly puts it in their mission statement, they seek: “to organise the world’s information and make it universally accessible and useful”.

However, some researchers wondered early on about potential dangers lurking in online availability. They argued that it could sweep older knowledge under the rug and habituate scientists into searching for works close to their scientific field, thereby hurting interdisciplinary science and possibly leading to a Balkanisation of science.

The first problem largely failed to materialise. Research has now shown that as older papers come online, their visibility and citations typically increase. It is nevertheless worth assessing the legacy of tools like Google Scholar to see precisely how they have affected research over the last two decades.

Measuring the impact of Google Scholar

In a new study, I examine how the rate of citations of older papers published in the 1950s and 1960s changed after 2004, when Google Scholar was launched. While researchers have identified problems with Google Scholar since its launch, online traffic data indicates there was rapid acceptance of the platform.

To provide convincing evidence that any change that occurred after 2004 was not due to other factors such as the increased volume of papers or events influencing citation behaviour, I carefully matched pre-citation patterns of these older papers with newer published ones. The citations of a paper usually follow an inverse U-shaped relationship, with few citations immediately after publication, followed by a spike and then a drop off. To examine whether results are sensitive to such citation lags, I separately matched papers published in the early and late 1980s.

I tested whether Google Scholar indeed favours new knowledge at the expense of old knowledge. On the contrary, in a simple numerical analysis, I found that older papers are represented on par with their overall frequency in the universe of science, which is in line with previous research findings.

Older papers experienced an uptick in their citations that cannot be explained by citation patterns up until then, pointing to Google Scholar as the cause.

Indeed, in the more rigorous part of the analysis, I found that older papers experienced an uptick in their citations that cannot be explained by citation patterns up until then, pointing to Google Scholar as the cause. More importantly, results showed that these older papers were cited beyond their own broad field, suggesting that fears online availability would lead to a “narrowing of science” did not materialise.

Research transformed

There is currently a broader debate taking place about whether researchers are now reading and producing a narrower and more homogenous body of work. For instance, in an influential study in Nature, researchers showed that papers and patents have become less disruptive over time, though these results have been challenged on the basis that the measure of disruptiveness and approach of the study may be flawed.

My research suggests online platforms are unlikely to be the cause of any observed homogeneity in citation behaviour.

Regardless of where one stands in this debate, my research suggests online platforms are unlikely to be the cause of any observed homogeneity in citation behaviour. This is not to say these platforms and the overall digitisation and indexation of science are free from pitfalls entirely. One potential issue with online platforms is a lack of transparency over the coverage and ranking of search results, which could create a loss of trust among researchers. Another is that while habitual use of these platforms can lead to efficiency, it can also backfire if online platforms are flawed.

With this said, if developed and used properly, these platforms can undoubtedly do a lot of good, even in ways that may be less visible to some users. For instance, some younger researchers may be unaware that in the past, the citation of a paper could only be attributed to the first author. In the words of one of the founder of Scientometrics, Eugene Garfield: “If you search the Citation Index for the cited work of a given scientist, you will find only those publications in which the scientist was listed as first author… Obviously, this characteristic can affect the accuracy of someone’s citation rate.”

The digitisation brought initially by the launch of Web Of Science has promoted fairness among scientific teams by reducing the importance of authorship order.

This naturally created huge incentives to be listed first. Informal discussions with senior academics revealed this was indeed an issue of contention due to citation recognition. There is evidence that in this context at least, the digitisation brought initially by the launch of Web Of Science has promoted fairness among scientific teams by reducing the importance of authorship order.

*

This blog draws on the author’s article, The role of online search platforms in scientific diffusion, published in the Journal of the Association for Information Science and Technology (JASIST).

The content generated on this blog is for information purposes only. This Article gives the views and opinions of the authors and does not reflect the views and opinions of the Impact of Social Science blog (the blog), nor of the London School of Economics and Political Science. Please review our comments policy if you have any concerns on posting a comment below.

Image Credit: Postmodern Studio on Shutterstock.