Lotka's Law, Temporal Trends, Self-Citation Analysis and Inflibnet E-Shodhsindhu Coverage in Chemical Engineering Doctoral Dissertations: A Bibliometric Study of Universities in Andhra Pradesh, India

In the world of chemical engineering, the path to a doctorate is a journey of deep inquiry, where a student must stand on the shoulders of giants to build something new. This process relies heavily on how researchers find and use the work of others. For decades, scholars have observed that scientific output is not spread evenly; instead, a small number of experts produce the majority of the research, while many others contribute just a little. This pattern, known as Lotka's Law, helps map the landscape of who is doing the work. At the same time, the way researchers find information has changed dramatically. The shift from physical books in library stacks to digital files on screens has altered how quickly new ideas are discovered and how often researchers cite their own previous work or that of their mentors. Understanding these shifts is vital because it reveals how knowledge flows, how research groups are formed, and whether students have access to the most important tools they need to succeed.

A team of researchers set out to trace these exact patterns within the chemical engineering programs of five universities in Andhra Pradesh, India. They gathered the footnotes from 347 doctoral dissertations, a massive collection containing over 61,000 references. By examining these citations, they could see not just what was being studied, but who was guiding the students, how the research habits changed over time, and whether the digital libraries available to these students were actually providing the most critical journals. The study covers a span of more than seventy years, from 1950 to 2026, allowing the team to watch the evolution of the field in real time.

The first thing the researchers looked at was the productivity of the supervisors, the faculty members who guide these doctoral students. They found that the distribution of work followed a familiar pattern: a few supervisors guided many students, while most guided only one or two. The most prolific supervisor in the group, a professor at Andhra University, had guided sixteen doctoral theses. The next most active had guided thirteen, followed by another with twelve. While the overall pattern matched the expected mathematical rule for scientific output, there was a slight deviation in the middle range. A small group of supervisors who had guided between five and thirteen students was slightly larger than the standard rule predicted. The researchers suggest this is likely due to the specific culture of research groups in these universities, where several faculty members have built teams of intermediate size. This finding confirms that while the general rules of scientific productivity hold true, local institutional cultures can create unique shapes in the data.

As the researchers looked deeper into the timeline, they discovered a dramatic transformation in how these students gathered information. In the decades before 2010, the dissertations relied almost entirely on printed materials. The average number of references in a thesis was around one hundred, and the sources were often older, with a median age of nearly two decades. However, as the years progressed, the landscape shifted. By the period from 2020 to 2026, the average number of references in a dissertation had nearly doubled to over two hundred. More importantly, the nature of those references changed. Before 2010, zero percent of the citations came from online sources. By the most recent period, nearly half of all the sources cited were digital. This surge was not just a change in format; it reflected a fundamental shift in how researchers access knowledge. The availability of electronic journals allowed students to find more recent studies, with the average age of cited literature dropping significantly. The researchers also noted that the number of authors working together on the papers cited in these dissertations increased, moving from an average of fewer than two authors per paper to nearly three, mirroring a global trend toward more collaborative science.

Another critical question was whether these researchers were simply citing their own work or the work of their supervisors to pad their lists. The study estimated that self-citation, where a student cites their own past work or their supervisor's, made up less than five percent of all citations. This low number suggests that the research is driven by a broad engagement with the international scientific community rather than a closed loop of internal references. The dominance of major international journals in the citation lists further supports the idea that these students are looking outward to the best available science, rather than relying on a narrow, local pool of sources.

The final piece of the puzzle involved checking the library's digital holdings against the most important journals in the field. The researchers identified the top fifty-five journals that chemical engineers rely on most heavily. They then checked if the INFLIBNET e-ShodhSindhu consortium, a digital library network serving Indian universities, provided access to these specific titles. The results were overwhelmingly positive. The consortium provided full access to forty-nine of these top journals, covering more than ninety-three percent of the citations found in the dissertations. Five additional journals were available but with some limitations on how far back the archives went. Only one journal was completely missing, and even that appeared to be a recording error rather than a true gap in the collection. This high level of coverage means that the digital library is successfully delivering the core information these researchers need.

The study concludes that the digital library system has had a transformative impact on chemical engineering research in the region. It has enabled a shift from a print-based environment to one where nearly half of the sources are digital, allowing for faster access to current knowledge and more comprehensive literature reviews. While the system covers almost all of the most critical journals, the researchers identified a few specific gaps, particularly with journals published by certain American engineering societies, which could be addressed through targeted negotiations. The findings offer a clear picture of a research community that has successfully adapted to the digital age, supported by a library system that provides the vast majority of the tools required for high-level scientific discovery.