Page 12 - The lraternational Journal of the Royal Society of Thailand.indd

The International Journal of the Royal Society of Thailand
Volume XV-2023

Materials and Methods

This article started by describing the co-authorship network of Thai researchers
who had been confirmed to buy authorships from paper mill services, which we

referred to as Index Researchers (IR_01 and IR_02). Then, we expanded our analysis
to include the co-authorship networks of additional researchers, affiliated with
an organization in Thailand, using a snowball sampling method. We subsequently

described if false authorships of these Index Researchers exhibited any distinct
characteristics compared to those of other researchers. All the analyses in this study

were performed using R programming language version 4.3.0 (R Core Team, 2022).

Publication records

We obtained full publication records for individual researchers from the

Dimensions Publication database (https://app.dimensions.ai/discover/publication),
which were subsequently utilized in the network analysis of each researcher’s
co-authorship. The dataset comprises the following details: article titles, corresponding

digital object identifiers (DOIs), publication dates, author lists with their respective
affiliations and countries, journal names, and the domains of research covered by each

journal, classified according to the Australian and New Zealand Standard Research
Classification (ANZSRC) 2020 (Australian Beureau of Statistics, 2020).

We restricted our subsequent analyses to articles published before January 31,
2023, to mitigate the potential impact of altered publication patterns following the

increased awareness of research integrity after January. Additionally, our analyses
focused solely on original articles published in academic journals published in English
language, excluding theses, preprints, corrigenda, and retraction notes.

Snowball sampling

As this study aimed to explore the potential distinctions in publication record

profiles and co-authorship network characteristics between Index Researchers and
other Thai-affiliated researchers, we extended our investigation of publication records
to the latter group through a snowball sampling method.

Briefly, we started our investigation by examining the publication records of

Index Researcher IR_01, who demonstrated a higher maximum annual publication
count than Index Researcher IR_02. Initially, we identified the frequent co-authors of

Index Researcher IR_01, comprising the top 15% of frequently co-published authors.

A Potential Use of Ego-centric Co-authorship Network Analysis for
4 Inspecting Research Misconduct: A Case Study of Thai Researchers