Comparison of bibliographic data sources: Implications for the robustness of university rankings

Bibliometric statistics are commonly used by university leadership, governments, funders, and related industries to quantify academic performance. This in turn may define academic promotion, tenure, funding, and other functional facets of academia. This obsession with excellence is highly correlated to various negative impacts on both academic behavior and research bias (Anderson, Ronning, et al., 2007; Fanelli, 2010; van Wessel, 2016; Moore, Neylon, et al., 2017). Furthermore, these metrics (such as citation counts and impact factors) are often derived from one of the large bibliographic sources, such as Web of Science (WoS), Scopus or Google Scholar (GS). Given the potential differences between their coverages of the scholarly literature, quantitative evaluations of research based on a single database present a risky basis on which to make policy decisions.

In a related manner, these bibliographic sources and metrics are also used in various university rankings. For example, Scopus is utilized by QS University Rankings and THE World University Rankings for citation counts, while Academic Ranking of World Universities makes use of WoS for a similar purpose¹. These rankings, and others, have been driving systematic transformations to higher education, including increased focus on student satisfaction and changes in consumer behavior. A focus on performance according to the narrow set of measures reflected in university rankings comes with a number of side effects, such as institutional homogenization, distorting disciplinary balance, and altering institutional focus (Shin & Toutkoushian, 2011; Hazelkorn, 2007). As a result of heavy criticism by the scientific community, university rankings (together with impact factors) have recently been boycotted by some academic stakeholders (Stergiou & Lessenich, 2014). This also includes domestic rankings². Nevertheless, they are still widely marketed and used, without necessarily being carefully comprehended by decision makers (e.g., policymakers and students).

Bibliographic data sources evidently make a significant impact on the academic landscape. This makes the selection and use of such databases essential to various stakeholders. As such, a number of important research questions arise:

• 1. 

  Are there differences across bibliographic databases?

• 2. 

  If there are differences, can we characterize them?

• 3. 

  Do these differences matter? How do they matter?

• 4. 

  And to whom do these differences matter?

Citation indexing of academic publications began in the 1960s, with the introduction of the Science Citation Index (SCI) by Eugene Garfield. This was followed by the annual release, starting from 1975, of impact factors through journal citation reports. This was initially developed to select additional journals for inclusion in the SCI. At that stage, much of the citation extraction was done manually (e.g., using punched cards as input to primitive computers) and the results were restricted to a niche selection of articles and journals. However, with the explosion of the Internet in the 1990s, citation indexing became automated and led to the creation of CiteSeer (Giles, Bollacker, & Lawrence, 1998), the first automatic public citation indexing system.

The rapid up-scaling of citation records created opportunities for new research explorations and bibliographic services. The former are often driven by citation analysis in the fields of bibliometrics and scientometrics, where quantitative evaluations of the academic literature play major roles. The latter is evidenced by the rise of large bibliographic and citation databases. Some of the most popular databases include WoS, Scopus, GS, and, more recently, Microsoft Academic (MSA).

WoS was the only systematic source for citation counts until 2004, when Scopus and GS were introduced. One of the earliest comparisons of these three sources was done by Jacsó (2005). The article reported on search results for citations to an article, citations to a journal, and citations to top 30 most cited papers in a particular journal. At that time, WoS had the highest number of records simply because of its longer time span, Scopus had the widest coverage for more recent years, and GS had the lowest number of records, with limited search functions and incoherent metadata records.

Other early studies showed that Scopus offered 20% more coverage (than WoS) in citations, while GS (although with good coverage) had issues with having a shorter reference list, less frequent updates, and duplicate references in its results (Falagas, Pitsouni, et al., 2008). A number of studies have shown that the average citation counts across disciplines varied by source (Bakkalbasi et al., 2006; Kulkarni et al., 2009; Yang & Meho, 2006). It was also shown that, for a small list of researchers, the h-index calculated from these three sources gave very different results (Bar-Ilan, 2008). The latest large-scale comparison showed that GS had significantly more coverage of citations than WoS and Scopus, although the rank correlations were high (Martín-Martín, Orduna-Malea, et al., 2018). Interestingly, Archambault, Campbell, et al. (2009) also showed that the rankings of countries by number of papers and citations were highly correlated between results extracted separately from WoS and Scopus.

Mongeon and Paul-Hus (2016) found that the journal coverages of both WoS and Scopus were biased toward Natural Sciences, Engineering and Biomedical Research. More importantly, their overall coverages differed significantly. Similar findings were obtained by Harzing and Alakangas (2016) when GS was added to the comparison, although for a much smaller sample of objects. Franceschini et al. (2016) also studied database errors in both Scopus and WoS, and found that the distributions of errors were very different between these two sources.

MSA was relaunched (in beta version) in 2016 as the newly improved incarnation of the outdated Microsoft Academic Services. MSA obtains bibliographic data through web pages crawled by Bing. MSA’s emergence and fast growth (at a rate of 1.3 million records per month, according to Hug and Brändle [2017]) has spurred its use in several bibliometrics studies (De Domenico, Omodei, & Arenas, 2016; Effendy & Yap, 2017; Portenoy & West, 2017; Portenoy, Hullman, & West, 2016; Sandulescu & Chiru, 2016; Wesley-Smith, Bergstrom, & West, 2016; Vaccario et al., 2017). At the same time, various papers have tracked changes in the MSA database and compared it to other bibliographic sources (Harzing, 2016; Harzing & Alakangas, 2017a, 2017b; Hug & Brändle, 2017; Paszcza, 2016). Its rapid development, especially in correcting some important errors, over the past two years and strength in coverage have been very encouraging. However, there remain concerns regarding the accuracy of MSA’s affiliation metadata. Ranjbar-Sahraei, van Eck, and de Jong (2018) found that a considerable number of publications in MSA have missing or incorrect affiliation information for a sample of output for a single university.

Tsay, Wu, and Tseng (2017) indicated that MSA had similar coverage to GS and the Astrophysics Data System for publications of a sample of physics Nobel laureates from 2001 to 2013, with MSA having a much lower internal overlap percentage than that of GS. MSA has also recently been used to predict Article Influence scores for open access (OA) journals (Norlander, Li, & West, 2018). Hug, Ochsner, and Brändle (2017) and Thelwall (2018), using samples of publications, showed there was uniformity between citation analyses done via MSA and Scopus. Harzing & Alakangas (2017a) also showed, for individual researchers, that the citation counts by MSA were similar to or higher than Scopus and WoS, varying across disciplines.

As discussed by Neylon and Wu (2009), using a singular article-level or journal-level metric as a filter for scientific literature is deeply flawed and incorporating diverse effective measurement tools is a necessary practice. In a similar vein, using a single bibliographic data source for evaluating specific aspects of academic work can be misleading. Given the immense social and academic impacts of the results of such evaluations, and the unlikeliness of them (as either part of research quantification or rankings) being completely discarded any time soon, one ought to be cautious in both interpreting and constructing such evaluation frameworks. With this in mind, we aim to provide a deep exploration in comparing the coverage of research objects with DOIs (digital object identifiers) in WoS, Scopus, and MSA³, in terms of both volume and various bibliographic variables, at the institutional level. In particular, a sample list of 15 universities is selected (ranging in geography, prestige, and size) and data affiliated with each university are drawn from all three sources (from 2000 to 2018). Less detailed data are also collected for another 140 universities to be used as a supplementary set where applicable. An automated process is used to compare the coverage of the sources and the discrepancies in the publication year are recorded. In addition, manual online searches are used to validate affiliation correctness and plausibility for samples of DOIs. The focus on DOIs also provides broader opportunities for cross-validation of bibliographic variables, such as OA status and document types from Unpaywall⁴, and citations data from OpenCitations⁵. These will assist in further understanding of the differences between data sources and the kind of biases that they may lead to.

Previous studies that compared WoS, Scopus, and MSA were limited to publications linked to an individual researcher, a small group of researchers, or one university. These comparisons were also mostly drawn in relation to citation counts. This article extends the literature by expanding the study set to include several universities and drawing institutional comparisons across a larger selection of characteristics and measures. We further sought to distinguish between the effects of completeness of research outputs in different data sources from the effects of that coverage on any specific performance metric (such as citation counts). Most previous studies have focused on the question of applying a specific data source to the specific problem of creating a citation-based ranking. An important difference for us was to ask the more general question of how bibliographic sources might affect a range of different performance evaluations, including, but not limited to, citations as the performance metric.

The study therefore includes analyses of potential effects in the exclusive selection of one source for evaluating a set of bibliographic metrics (i.e., potential effects on the ranking of universities). We use secondary data sources (Unpaywall and OpenCitations) to construct metrics for OA and citations. This gives standardized contrasting sets of records for comparisons across bibliographic sources. In turn, this allows us to disentangle the questions of research output completeness, and the separate effects this may have on quantitative measures. The results lead up to the main message that it is essential to integrate diverse data sources in any institutional evaluation framework.

The remainder of this article is structured as follows: Section 2 gives an overview of some global characteristics across the various bibliographic databases. Section 3 provides detailed descriptions of our data collection and manual cross-validation processes. All analyses and results are presented in section 4. Sections 5 and 6 contain discussions on limitations and conclusions, respectively.

WoS and Scopus are both online subscription-based academic indexing services. WoS was originally produced by the Institute for Scientific Information (ISI), but was later acquired by Thomson Reuters and then Clarivate Analytics (formerly a part of Thomson Reuters). It contains a list of several databases, where the level of access to each depends on the selection of subscription models. The search functionalities can also vary according to which databases are selected (for example, the “Organization-Enhanced” search option is not available when all WoS databases are included). On the other hand, Elsevier’s Scopus database seems to offer one unified database of all document types (the only exception is data on patents, which appear as a separate list in search results). A manual online search reveals a wider variety of document types in WoS. For example, it contains items listed as “poetry,” which does not seem to fit into any of the types in Scopus.

MSA is open to the public through the Academic Knowledge API, though both a rate limit and a monthly usage cap apply to this free version⁶. The subscription version is documented as relatively cheap at $0.32 per 1,000 transactions⁷. Its semantic search functionality and ability to cater for natural language queries are among the main differences from the other two bibliographic sources. Its coverage in patents has greatly increased through the recent inclusion of Lens.org metadata⁸. As a preliminary examination, we take a look at some global characteristics and features across the three sources. Table 1 provides an overview of coverage and comparative strengths in each source. WoS has several databases from which it extracts data. The most commonly used collection of databases is WoS Core, which allows for more functionality. On the other hand, WoS All Databases includes all databases listed by WoS (with increased coverage for Social Sciences and local languages, for example), but due to varying levels of availability of information its functionalities are limited (e.g., fewer search query options). Scopus does not seem to index Arts & Humanities, while MSA appears to have significantly more coverage in Social Sciences and Arts & Humanities than WoS Core and Scopus. With higher coverage for journals and conferences, MSA tracks a significantly larger set of records. It is also interesting to note that MSA had approximately 127 million documents only a couple of years ago (Herrmannova & Knoth, 2016).

The annual total numbers⁹ of objects for the various sources from 1970 to 2017 are displayed in Figure 1. In comparison to Jacsó (2005), and other studies mentioned earlier, there seem to be significant increases in both Scopus and WoS in terms of both growth over time and backfilling. However, both sources still have significantly less total counts than that of MSA. The figure also shows a high degree of correlation between Scopus, WoS Core, and WoS All Databases. However, this figure does not provide any information on internal or external overlaps across the sources (which we shall explore).

To get a better overview of research disciplines covered by each source, the percentage spread of objects across disciplines, for each source, is displayed in Figure 2. Evidently, all sources are dominated by the sciences, as commonly noted in the literature. However, MSA does seem to have relatively higher proportions for both Social Sciences and Arts & Humanities.

To perform a more detailed comparison of sources, we gathered outputs for a selected set of 15 universities, ranging in geography, prestige, and size, from each bibliographic source: WoS Core (referred to as just WoS from here on), Scopus, and MSA. This is done through the use of APIs for each data source. We extract records for the years from 2000 to 2018 via affiliation IDs (in the case of Scopus and MSA) and organization-enhanced search terms (for WoS)¹⁹. The results form three sets of data (one from each source) for each university. Subsequently, DOIs of objects (for those that do have them) are extracted from each set. A further 140 universities are also included as a supplementary set to be used where necessary. Our strategy is to explore various bibliographic characteristics related to these DOIs at the overall level (for all years and all institutions) and then contrast that with the corresponding results for individual universities focusing on a single year (i.e., 2016). Where applicable, the analysis for 2016 is also extended to the full set of 155 universities. We are mainly interested in the following characteristics:

• 1. 

  Distributions (e.g., Venn diagrams) of DOIs across sources,

• 2. 

  Discrepancies in publication year recorded by each source,

• 3. 

  Document types across various parts of the Venn diagrams of DOIs,

• 4. 

  Citation counts (as per OpenCitations) calculated across sources,

• 5. 

  OA levels (as per Unpaywall) calculated across sources, and

• 6. 

  Plausibility of assigned affiliation for DOIs exclusively indexed by a single source.

Analyses of characteristics 1 to 5 are mostly automated, with data collected into a data management systems implemented on the Google Cloud Platform. Data are gathered via the APIs of WoS²⁰, Scopus, and MSA. Specifically, each data source is queried for a specified time period for articles with specified affiliation. Our collection process is therefore dependent on both the affiliation and publication year metadata for each source. For example, a DOI may be indexed in both source A and source B, where it is assigned to university X by source A but not by source B. This DOI will not be retrieved from source B as an output from university X, even though it is indexed in source B. Similarly a significant number of DOIs are assigned to different publication years by different sources. Hence, the resulting Venn diagrams do not show the overall coverage of all research output for a given institution, but rather the discoverability via DOI, publication year, and affiliation.

Conceptually, we use each source to prepare the output data for 18 editions of an annually constructed evaluation or ranking. To examine the overlap between sources for our initial search criteria (i.e., affiliation and publication year), Venn diagrams for characteristic 1 are constructed by matching DOIs and publication years recorded by each source.

To explore the potential reasons for discrepancies across sources, we also examine characteristics 2 and 3. Metadata of each DOI from each source are compared to determine the level of agreement in publication years across sources. The “genre” field from Unpaywall is used to determine the document type of each DOI (e.g., journal articles, book chapters, conference proceedings).

Unpaywall data and OpenCitations’ COCI data are used to determine OA status and citation counts associated with each DOI. For this article, we only require the general OA status and not the type of OA (e.g., gold OA, green OA). Hence, we only use the “is_oa” field in the Unpaywall metadata to determine the OA status of DOIs in our data. COCI records citation links between Crossref DOIs. By querying and merging all links to a DOI, it allows us to determine the number of citations this DOI receives. The use of COCI data allows us to separate the effects of metadata coverage (i.e., publication and affiliation metadata) from the effects of inclusion of the sources of citations in a specific database. While COCI includes only those citations that are made openly available through Crossref (and therefore excludes citations from a number of large publishers, including Elsevier and the American Chemical Society at the time of writing) it nonetheless provides us with a consistent source of data that can be applied to the evaluation of all DOIs. We focus here on the differences in coverage among bibliographic data sources, and how this affects the results of evaluations based on other data sources. This is different from seeing how inclusion in a citation database affects the result of a citation-based ranking. That is, our goal is to compare the use of these data sources as instruments for discovering outputs that might then be evaluated by a range of external measures, rather than as instruments for constructing self-contained citation-based evaluations. For comparison, we also provide an evaluation of the differences in rankings constructed using citation data from each source in Supplementary Material 9.

We gather this information for a set of DOIs of interest (e.g., DOIs from WoS affiliated to one university) and obtain total citation counts for this set. This total can then be divided by the number of (Crossref) DOIs affiliated to this university to produce an average citation count²¹. The overall data collection process is summarized in Figure 3 ²². The codes and data to produce the results and figures in this article can be accessed at Huang, Neylon, et al. (2019).

A manual process is followed for checking characteristic 6. The procedure for the manual validation is focused on the nonoverlapping parts of the three sources (i.e., shaded sections in Figure 4). The overlapping parts indicate agreement by at least two sources over both affiliation and publication year records (when filtered down to a particular year). Given the different ways in which the sources gather data, the reliability of information for these parts is much more convincing. In contrast, the nonoverlapping sections are not validated by other sources. This leads to the need for the manual validation process.

The publication year can be a reason for the discrepancy of coverage due to inconsistencies in how date is recorded. For example, in the case of a journal article, a source may choose to record the date of the journal issue, the publication date for the article, or the date on which the article first appeared online. Hence, our first step is to check whether DOIs from the nonoverlapping sections are indeed in another source but fall in a different year. After removing these DOIs, which were identified via comparison to adjacent years, we sampled the remaining DOIs from each nonoverlapping section for manual validation (Figure 4). This is processed for DOIs from 2016.

The process that leads to the manual validation is summarized in the flowchart given in Figure 5. Once DOIs are sampled from each nonoverlapping section, they are compared against the other two sources (via DOI and title searches on each source’s webpage) and also the original document (online versions)²³.

Assume we have three sources, A, B, and C, and the current set of DOIs are from source A. The following questions are asked as part of the manual checking process (with a likewise procedure used for DOIs from the other two sources):

• 1. 

  Is this DOI found in the metadata record in source B?

• 2. 

  Is the title associated with this DOI found in source B?

• 3. 

  Is the exact affiliation phrase found in the metadata record in source B?

• 4. 

  If not, is the affiliation plausible?

• 5. 

  Is this DOI found in the metadata record in source C?

• 6. 

  Is the title associated with this DOI found in source C?

• 7. 

  Is the exact affiliation phrase found in the metadata record in source C?

• 8. 

  If not, is the affiliation plausible?

• 9. 

  Is the DOI correctly recorded in source A (as per original document or doi.org)?

• 10. 

  Is the exact affiliation phrase found on the original document?

• 11. 

  If not, is the affiliation plausible?

The numbers of DOIs to be sampled for each institution are 30, 30, and 40 from (exclusively) WoS, Scopus, and MSA, respectively, after removal of DOIs that are found in another source for a different year.

Table 2 presents the total number of unique DOI records we have obtained from each source, the combined number of unique DOIs, and how many of these DOIs are recorded in Unpaywall, for each institution for 2016. The coverage of DOIs by Unpaywall is very high, as expected. The only slight exception is DUT, where a significantly higher portion of Scopus DOIs were not recorded by Unpaywall. A quick exploration²⁴ finds most of these DOIs to be registered with China National Knowledge Infrastructure (CNKI) or the Institute of Scientific and Technical Information of China (ISTIC), whereas Unpaywall currently only indexes Crossref DOIs²⁵.

In this section, we proceed with the detailed bibliographic comparisons across sources. We will start with exploring the coverage of DOIs by each source. This is followed by examining the amount of agreement, or disagreement, of publication year recorded by each bibliographic data source. The document types, citation counts, and OA percentages, as per source, are the subsequent analyses. Lastly, a manual cross-validation procedure is employed for samples extracted from nonoverlapping sections of the Venn diagrams for each institution in our sample of 15 institutions.

Here we take an exploration of the spread of the DOIs across the sources. Figure 6 shows the Venn diagrams of DOI counts for our initial set of 15 universities combined from 2000 to 2018 and for just 2016²⁹, respectively. Evidently, the central regions (overlap of all three sources) have the highest count in each Venn diagram. These are DOIs that have been indexed by all three sources and, given the intended global coverage of major publication venues by these sources, the relatively higher counts here are not at all surprising. However, there are also significant portions of DOIs exclusively accessed via a single source in both Venn diagrams. This gives rise to the potential biases in any bibliometric measure to be calculated from a single source.

This pattern of difference in coverage is mirrored at the institutional level, albeit the level of discrepancies varies across institutions. Supplementary Material 3 contains two Venn diagrams for each institution, both for 2016. In each case, the Venn diagram on the left records all DOIs as per bibliographic source and the one on the right is a subset of these DOIs that are also indexed in the Unpaywall database. It is noted that the two Venn diagrams for each institution are quite similar due to the high coverage of these DOIs by Unpaywall. The only exception is the Scopus coverage of DOIs for DUT, for which the DOIs exclusively indexed by Scopus significantly decreased when moving from the left Venn diagram to the one on the right. This is consistent with what we observed earlier, with many of these DOIs provided by agencies other than Crossref. The overall pattern is that there appear to be significant portions of DOIs only indexed by a single source. Hence, pulling together these sources can greatly enhance coverage. Interestingly, for most institutions, MSA has the most number of exclusively indexed DOIs (and appears to be more extreme for non-English-speaking and non-European universities), the only exception being UCL. ITB also represent an extreme case where the proportion of DOIs indexed by all three sources is much lower relative to other universities.

To have a better overview of how coverage of these three sources varies across institutions, we perform several analyses as follows. First, we identify each institution with the seven different counts (instead of percentages) as per its own Venn diagram of all DOIs (Venn diagrams on the left in Supplementary Material 3). We also include another 140³¹ universities for comparison. We view each (GRID ID, DOI) pair as a distinct object. Hence, we obtain a 155 × 7 contingency table. Each column of this table represents the number of DOIs falling in the respective section of the Venn diagram; for example, column 1 is the number of DOIs in section WSM of the Venn diagram (refer to Figure 4). We can also convert these counts to proportions by dividing them by the total number of DOIs for each institution. Figure 7 shows the distribution of these proportions for each section of the Venn diagram across all 155 universities. The higher proportion in the central region (section WSM) of the Venn diagram is again observed. The general pattern that emerged is that, for all sections of the Venn diagram, there appears to be a concentrated central location with many extreme cases (excess kurtosis of 2.29, 9.72, 5.96, 1.82, 22.24, 11.49, and 6.88, from sections WSM, WS, WM, SM, W, S, and M, respectively) and substantial skewness. We can also concatenate the respective sections to get the proportion of DOIs covered by each bibliographic source. The spreads of these proportions are summarized in Figure 8 as histograms.

Again, the pattern of high central peak, skewness, and long tail is observed. The peakedness and long tails are confirmed by the excess kurtosis of 4.29, 3.34, and 8.60 for WoS, Scopus, and MSA respectively. The skewness to the left with number of extreme cases highlights the low degree of coverage for some universities. Meanwhile, a correlation analysis of the proportions for the three sources is quite intriguing (see Table 3). Both Spearman’s rank correlation and Pearson’s correlation matrices are presented here. There appears to be a negative correlation between coverage by WoS and coverage by MSA: When there is a high proportion of coverage by WoS, the coverage by MSA is relatively low. There is also a low correlation between WoS and Scopus. While much of this may be attributed to the different methodological structure and focus across WoS, Scopus, and MSA, the degree of nonalignments is still quite a surprise³⁴.

We further performed tests of homogeneity across institutions to check whether the spread of DOIs across individual Venn diagrams come from the same probability distribution. The results of these tests are provided in Table 4. It is evident that the chance of rejecting homogeneity is very high. Bootstrapped samples from sample sizes 10 to 155, in increments of 5, all gave similar results as well.

Now that we have confirmed the differences in DOI distributions across institutions and negative to low correlations between the nonsymmetrical coverages by the three bibliographic sources, a follow-up question may be whether there are groupings among these universities. We proceed with a hierarchical cluster analysis for both the sample of 15 universities and for all 155 universities, using dissimilarities between the proportions of the Venn diagrams as clustering criteria³⁶. At the same time, we also color-code the universities by their regions and rank positions on the 2019 THE World University Rankings. Some of these are presented in Supplementary Material 6. While no striking patterns emerge, there do appear to be some interesting groupings. For example, there seems to be a block of European and American universities toward the left of the dendrogram colored by region. Perhaps unsurprisingly, around the same area for the dendrogram colored by THE ranking, there is also a rough cluster of the most highly ranked universities.

The contrasts may be more apparent for the smaller sample of 15 universities. An example of this is presented in Figure 10. ITB is clearly an outlier from the rest of the group, as was the case for the Venn diagrams, and the two highest ranked universities are placed quite close to each other. Seven of the universities ranking from 201 and above are placed on the right of the dendrogram (perhaps in two clusters). One of these also consists mainly of universities from non-English-speaking regions (Loughborough being the exception). In general, there appear to be some patterns of prestige and regional clustering (for both the sample of 15 and the sample of 155 universities).

As mentioned earlier, discrepancies in publication year recorded by different bibliographic sources are possible, given that there is no universal standard to define publication year (or indeed publication date). This poses a problem when trying to combine sources to evaluate and track a bibliometric variable over time. A DOI can be double-counted (i.e., counted two or more times in different years via different sources). In the following, we explore the amount of agreement and disagreement on publication years by WoS, Scopus, and MSA. The overall numbers are presented in Table 5, covering all DOIs for 15 institutions and years from 2000 to 2018.

In this table, the number of DOIs jointly indexed by pairs of bibliographic sources and by all three bibliographic sources are recorded. It should be noted that percentages are calculated over different sets of DOIs (i.e., different denominators). For example, the number of DOIs common to all three sources (i.e., 404,710) is less than the number of DOIs common to only Scopus and MSA (i.e., 522,026).

It is clear that the overall levels of agreement are very high. However, two follow-up questions are (1) for DOIs that are present in a different source for a different year, what is the spread of these DOIs over years? And (2) while the overall agreement of publication years is high, does that carry over to individual institutions?

To answer these questions, we now focus our attention on the year 2016 and DOIs that are exclusively indexed by a single source for that year. Figure 11 displays the spread of such DOIs from a particular source when matched against the other sources for different years. These are again DOIs from our sample of 15 institutions combined. The majority of the discrepancies are within one year (i.e., falling in 2015 and 2017), while extending this window one further year in both directions covers almost all remaining cases. We also note some differences across the sources. The number of discrepancies between WoS and Scopus is relatively small compared to those involving MSA. This may be the likely result of MSA using the date when a document first appears online as their default publication date³⁸.

Next we explore how these discrepancies of the publication year are distributed for individual institutions. Table 6 records, for each source, the percentages of DOIs from 2016 that appear in the other two sources but differ by a year and two years, respectively. For WoS, the percentage of matches over one year is consistently small for all institutions, ranging from 0.8% to 2%. This also significantly decreases when moving to the two year gap. In contrast, Scopus and MSA seem to have more varied results for the one year gap across institutions and with generally higher percentages than those of WoS.

The one standout case is ITB, an Indonesian university situated in the City of Bandung. Its results for WoS are similar to other institutions, but one-year comparisons from Scopus and MSA yielded disagreements on publications year for 24.6% and 25.6% of all common DOIs, respectively. We believe that this may be due to two reasons. Firstly, WoS has a significantly lower coverage of ITB (see Venn diagrams for ITB in Supplementary Material 3) than Scopus and MSA. There is also a much lower number of DOIs exclusively indexed by WoS. Secondly, Indonesia has an extraordinarily large number of local journals owned by universities, and many of these are open access. This is largely driven by government policy, which requires academics and students to publish research results and theses in academic journals³⁹. Many of these journals are also linked to conference output. This may have resulted in a systematic difference on how publication years (or dates) are recorded (or defined). The other two cases that stand out, although less extreme, are Cairo and IISC.

In Supplementary Material 7, the directions of the comparisons are displayed in more detail for the three standout cases (i.e., Cairo, IISC, and ITB). The comparisons are also narrowed down to just Scopus and MSA. It is immediately clear that the differences between Scopus and MSA are the main contributors to these standout cases. Also, it appears that MSA tends to record the publication year one year earlier than Scopus. This is in line with our earlier comments regarding MSA recording the date of first online publication and the publishing venues in Indonesia.

Let us now focus on the outer parts of the Venn diagrams (i.e., DOIs that appear to be exclusively indexed by a single source). The results for these sets of DOIs are presented in Table 7. Columns 2, 5, and 8 lists the number of 2016 DOIs exclusively indexed by WoS, Scopus, and MSA (compare these again with Venn diagrams in Supplementary Material 3), without checking against DOIs listed in other years. The subsequent columns list the percentages of these DOIs that can be matched against DOIs in other sources for one-year and two-year gaps, respectively. Consistent with Table 6, significantly higher portions of DOI matches occur after incorporating the first one-year gap, as compared to including a further one year on both sides. ITB again sees the largest impact of these inconsistencies, which corresponds to the observation made in Table 6. In general, the effect on these exclusive sets of DOIs varies considerably across institutions and sources (more so than observed in Table 6, as expected).

Another important bibliographic variable is the document type (e.g., journal articles, proceedings, book chapters) that relates to each DOI. In particular, the coverage of different document types can lead to insights into potential disciplinary biases in data sources and differences in institutional focuses on output types.

For this study, we use the “genre” variable in Unpaywall metadata to determine the document type of each DOI. These are Crossref-reported types for all DOI objects in the Crossref database⁴³. Table 8 provides the counts of each document type within each part of the Venn diagram between WoS, Scopus, and MSA (for all 15 institutions from 2000 to 2018 combined)⁴⁴. An immediate observation is that journal articles make up (by far) the highest proportion of the DOIs. This is true overall and for individual parts of the Venn diagram, as would be expected. The scenario is again more interesting when we consider the outer parts of the Venn diagram (sections W, S, and M of the Venn diagram). The set of DOIs exclusive to MSA contains significantly more book chapters and proceeding papers relative to any other parts. It also provides almost all thesis entries in our data and is the only source to provide posted content (i.e., web pages and blogs). On the other hand, Scopus seems to provide many books and monographs not indexed by the other two sources.

Again we would like to examine how the situation plays out for individual institutions. After filtering the sets of DOIs to each institution and to the year 2016, we follow the same procedure as above to produce the spread of document types across each part of an institution’s Venn diagram. These are recorded in Supplementary Material 8. As we have observed for the combined data set, journal articles make up the highest portion of the DOIs for each institution. The next two most common document types are book chapters and proceeding papers. The only exception is ITB, where there are slightly more proceeding papers than journal articles. Interestingly, there are a few universities with more book chapters than proceeding papers (Curtin, UNAM, UCL, UCT, Giessen, and WSU).

There are high proportions of book chapters indexed exclusively by MSA for all institutions. MSA also has the highest proportion of exclusively indexed journal articles, except for MIT, UCL, and Giessen (WoS has the highest such proportion for these three institutions). It is also observed that MSA and Scopus seem to bring in more additional proceeding papers than WoS (the only exception being UNAM, where all three sources have similar exclusive coverage on proceeding papers). Scopus also seems to often add books and monographs not indexed by the other two sources. For all universities, journal articles make up the majority of DOIs exclusively indexed by WoS. In contrast, the document types of DOIs exclusively indexed by Scopus or MSA are more diverse. Overall, we observe that each source has a different exclusive coverage of document types, and this coverage also varies across institutions.

One set of commonly used bibliographic metrics in the evaluation of academic output are those that relate to citation counts. These include metrics such as h-index, impact factor, and eigenfactor. However, these citation metrics can also be calculated via different sources. WoS, Scopus, and MSA all record and maintain their own citation data. While some research has shown that the citation counts across these sources showed high correlations at the author level and journal level (Harzing, 2016; Harzing & Alakangas, 2017a, 2017b), the corresponding effects on a set of universities remain relatively unknown. We match each DOI against the list of DOI citation links in OpenCitations’ COCI data and obtain (if it exists) its total citation count. In Table 9, we present the results combining DOIs for our initial set of 15 universities for all years from 2000 to 2018.

To repeat, our goal is to identify the effects of coverage on the discoverability of sets of outputs that would then be evaluated using an external source of data. For this reason we use the COCI data from OpenCitations to provide an external and independent source of data that can be comprehensively applied to the evaluations of each set of DOIs that we discover for each institution and year. Similarly, when we use the same sets of outputs to compare OA status (section 4.5) we use an external data source (Unpaywall) that allows us to evaluate the performance of the sets of identified outputs. This is different from comparing the results of using each bibliographic data source as the source of both the sets of outputs and the source of performance data. For completeness, we also provide such an analysis in Supplementary Material 9. In the event, the conclusions of both analyses are very similar. This suggests that COCI is a viable source of comprehensive citation data for cross comparison at system level, even if it is not an appropriate source of data in its current form for analyzing the comparative performance of individual outputs due to its lack of coverage of some sources of citations.

The results show that the total number of citations to MSA DOIs is slightly lower than in WoS and Scopus. This is in addition to an already larger set of (Unpaywall/Crossref) DOIs. Hence, MSA resulted in a lower average citation number (from a smaller numerator of citation counts and a larger denominator of Crossref DOIs) from the OpenCitations citation links.

As a further analysis, we would like to investigate how the change of bibliographic source influences the perceived performance of an institution while holding the evaluative data source constant. Figure 12 presents two different charts for total citations and ranks by average citations for each of the sample of 15 universities. UCL and MIT experience the biggest changes in total citation counts: decreases of 34% and 38%, respectively (left-hand chart in Figure 12), when shifted from WoS to MSA. While the remaining universities’ total citation counts seem to have changed to a lesser degree across sources, the differing coverage of DOIs (i.e., different number of DOIs recorded) by each source can still significantly change the average citation counts. This is evidenced in the second chart of Figure 12. Only four universities’ rankings remain unchanged across sources (the top three and last place). All other universities’ positions have shifted at least once across the three sources, with the biggest changes affecting IISC, USP, and UNAM.

For a more general view, we now include the ranking results for the larger set of 155 universities in Figure 13. The results related to universities that have shifted by at least 20 positions (at least once) across the three sources are highlighted in color, with universities from English-speaking regions in red and non-English-speaking ones in orange. This includes 45 universities: 27 in red and 18 in orange. That means that almost one-third of the universities have shifted 20 or more positions. The most extreme cases include Charles Sturt University dropping 146 places when moved from WoS to Scopus, and Universitat Siegen and University of Marrakech Cadi Ayyad dropping 143 and 112 positions, respectively, when moved from WoS to MSA.

For further insight into the distribution of shifts across sources, we summarize the pairwise changes to average citations and rankings by average citations into box plots in Figure 14. The median change to average citations when moving from WoS to Scopus is just below zero, while the corresponding medians for WoS to MSA and Scopus to MSA are both just above zero. The corresponding mean values are −0.2, 1.2, and 1.3, respectively. As for the changes to rankings, the median and mean values are all close to zero. The distributions of these box plots are characterized by a concentrated center with long tails. This signifies the existence of two contrasting groups: those universities that were less affected by shifts in bibliographic sources, and those that can have their performance levels, in terms of average citations, greatly altered depending on the choice of source.

A recent topic of interest is the number of OA publications produced at different levels of the academic system. In particular, universities may wish to evaluate their OA standings for compliance with funder policies and OA initiatives. For objects with DOIs (and, in particular, Crossref DOIs), various information on accessibility can be queried through Unpaywall⁴⁹. We matched all DOIs from the sample of 15 universities to the Unpaywall metadata and calculated the percentage of OA output across each bibliographic source and for all (unique) DOIs combined. This is presented in Table 10.

There do not appear to be substantial changes to the overall OA percentage when shifting across sources for the combined sets of DOIs. However, we should keep in mind that there are significant differences in each source’s DOI coverage, as observed earlier.

To see whether such consistency in OA percentages carries over to the institutional level for 2016, we again filter the data down to each university. Figure 15 provides the percentages of OA output and the corresponding relative ranks for each institution, as per the set of 2016 DOIs indexed by each source and also recorded in Unpaywall. It is observed that, for quite a few universities, the OA percentages vary considerably depending on which source is used to obtain the sets of DOIs. The most extreme case is again ITB, which had about a 20% drop when moving from WoS to Scopus. Also, the direction of OA percentage changes differs across universities. For example, the OA percentage for MIT decreased when moving from Scopus to MSA, but the opposite occurred for USP. This is especially critical if one is to compare the relative OA status across universities, which can vary according to the source of DOIs used. As for OA ranks, it seems to indicate a group of universities not affected by changing source, while the other group have their ranks shifted significantly. The most affected cases seem to be USP, ITB, and UNAM.

The effects on OA levels and ranks are more difficult to express directly for the larger set of 155 universities. Again, instead of labelling the full set of universities, we highlight only those that have shifted by 20 positions or more at least once. This is displayed in Figure 16. There are 24 out of 155 universities that have shifted at least 20 positions in OA ranking when moved across sources. Seventeen of these are from non-English-speaking regions, including six Latin American universities (out of seven in the full set). This is an indication of the potential difference in coverage of the three sources due to language.

Analogous to the earlier analysis on citations, we calculate differences in OA percentages and OA ranks when shifting from one source to another and present these in a number of box plots in Figure 17.

Evidently, the median OA percentage changes when shifting from WoS to Scopus, WoS to MSA, and Scopus to MSA are all positive. The corresponding mean changes are also positive at 3.4%, 4.9%, and 1.5% respectively. The median and mean changes to rankings are all close to zero. However, in both OA percentage and OA rank changes, there are many recorded extreme points (including both negative ones and positive ones). These include an OA percentage change as large as 31.1% (moving from WoS to MSA) and an extreme drop in OA rank of 96 positions (MSA to WoS). The general distributions of both changes to OA percentage and changes to OA rankings are characterized by high central peaks and long tails. This implies that, while the changes are small for most of the universities, there are also a significant number of cases where universities are largely affected by shifts in data sources.

This section provides a summary of our manual cross-validation results of DOIs exclusively indexed by each source. For each of the 15 institutions, we randomly sampled 40, 30, and 30 DOIs from their sets of 2016 DOIs exclusively indexed by WoS, Scopus, and MSA, respectively (i.e., sections W, S, and M from the Venn diagram in Figure 4). This was done after the removal of DOIs that match up to other sources in a different year (this includes the neighboring two years: 2014, 2015, 2017, and 2018). Subsequently, these lists of DOIs go through a thorough manual cross-validation process. Various questions were asked against each DOI and compared across the three bibliographic sources. These are summarized in a table in Supplementary Material 10.

In the following, we shall highlight some of the main findings in a few simple charts, with further detailed analysis provided in Supplementary Material 10. Firstly, we focus on the plausibility of affiliation associated with each DOI.

In Figure 18, we present results related to affiliation of each DOI as per source. For each DOI, the target affiliation is checked against its online original document⁵². When the original document is not accessible (e.g., not OA), the affiliation is matched against the other two sources. The decision is made to indicate the affiliation as plausible when the target affiliation (i.e., affiliation as per our data collection process) appears exact (including obvious versions of the university name) on the document, a plausible affiliation name variant⁵³ appears on the document, or the affiliation is confirmed by at least one of the other two bibliographic sources. This should (roughly) inform us about whether each source has correctly assigned these DOIs to the target affiliations.

The result shows that all sources have only correctly assigned roughly 80% of their respective DOIs from our sample to the target affiliations, with very little difference in performance across the sources. When this is filtered down by university (see Figure 19), we see a more varied performance across universities.

Interestingly, not all percentages are very high across the universities. This is especially apparent for DUT and IISC, where MSA seems to have affiliated many DOIs to these two institutions without the target affiliations actually appearing on the original documents or confirmed by another source. Similarly, for DOIs that were assigned to MSU and UNAM by Scopus alone, only 46.7% (for both institutions) have a plausible affiliation match.

We have also checked each DOI against the DOI string actually recorded as per original document (where applicable) or via doi.org. These percentages (of correct DOIs) are 93.1%, 98.2% and 96.7 for WoS, Scopus, and MSA respectively (with all 15 institutions combined). While these numbers are relatively high, a significant number of errors suggests that DOIs are not being systematically checked against authoritative sources such as Crossref which we find surprising. In addition the nature of these errors which in some cases appear to be transcription or OCR errors is concerning (see examples in Supplementary Material 11).

We now take an overview of results from the DOI and title matching, given in Figure 20. As an initial analysis, no affiliation information is considered here and the results represent all DOIs for the 15 universities combined. Each bar represents the percentage of output corresponding to DOIs (that initially appear to be) exclusively indexed by one source that can be found in another source by DOI matching and title matching (via manual searches online). For example, the first bar corresponds to objects with DOIs sampled from Scopus. The height of the blue bar shows the percentage of these objects that can be found in WoS by DOI matching. The orange bar then indicates how much more can be found by title matching.

We found that in all cases where there is a DOI match, there is also a title match. However, the opposite is not necessarily true. Hence, title matching increases the coverage slightly in all scenarios. This does imply that all three sources have missing DOIs in their metadata, though there appear to be fewer cases for Scopus. Scopus also seems to have a good coverage of DOIs from WoS. More strikingly, a very high proportion of DOIs and titles from WoS and Scopus are found in MSA. In contrast, far fewer MSA DOIs and titles are covered by WoS and Scopus.

In Figure 21, we added affiliation matching to the mix; that is, we checked whether the target affiliation (i.e., affiliation as per our data collection process) appears in the metadata of the matching source after an object is found by DOI or title match. This decreased the coverage in all cases, indicating the potential disagreement of affiliation across sources. MSA is the most affected of the three sources.

The general picture that has emerged is that MSA seems to have good coverage of DOIs that initially appeared to be exclusively from WoS or Scopus. However, it falls short on correctly assigning affiliations and recording DOIs corresponding to each output. MSA also seems to have substantially broader coverage, including many objects that genuinely appear exclusively in MSA. The correctness of affiliation metadata for these is high overall, but tends to vary across institutions.

One obvious limitation is our focus on DOIs and our dependence on the uniqueness of DOIs. We do note that there may be research objects with multiple DOIs and related objects may also be assigned a common DOI (e.g., books can fit both cases). A related matter is the correctness of DOIs; that is, whether they were recorded correctly (as per doi.org) in each source’s metadata⁵⁴. DOIs that did not generate Unpaywall returns could include such cases. While our manual cross-validation process did check our samples against doi.org, it is not clear what the scale of this issue is for the overall data.

Our manual cross-validation process was carried out over a number of months after the initial data collection process. This means that there may be potential discrepancies between metadata content at the time of collection and time of manual search. However, we expect such cases to be few, given that we are focused on 2016 data, and a number of manual spot checks did not reveal any obvious such cases.

Both API and manual searches for WoS and Scopus may be limited to the subscription model of the authors’ home institution at the time of access. On the other hand, matching identifiers have also proved to be challenging. For example, a few institutions have multiple Scopus IDs (e.g., multiple campuses), without an overarching ID. For the three cases we have encountered, out of the 155 universities we have selected what appeared to be the main campus IDs. A recent study (Donner, Rimmert, & Van Eck, 2020) have also highlighted issues regarding the affiliation disambiguation systems in WoS and Scopus. Other challenges include Unpaywall and OpenCitations coverage is limited to Crossref DOIs, manual cross-validation is limited to DOI and title searches only, and there is inherent subjectivity in linking plausible affiliation names.

This article has taken on the task of comparing and cross-validating various bibliographic characteristics (including coverage, publication date, OA status, document type, citations, and affiliation) across three major research output indexing databases: WoS, Scopus, and MSA. This is done mainly with a focus on identifying institutional-level differences and the corresponding effects of using different data sources in comparing institutions. Our data consist of all objects with DOIs extracted from the three bibliographic sources for an initial sample of 15 universities and a further supplementary 140 universities (used only where applicable).

Firstly, we found that the coverage of DOIs not only differs across the three sources, but their relative coverages are also nonsymmetrical, and the distribution of DOIs across the sources varied from institution to institution. This means that the sole use of one bibliographic source can potentially seriously disadvantage some institutions and advantage others in terms of total output. While the general level of agreement on publication year is high across sources, there were individual universities with large differences in coverage per year. The comparison of document types showed that different sources can systematically add coverage of selected research output types. This may be of importance when considering the coverage of different research discipline areas.

Our subsequent analyses further showed that while the aggregate levels (i.e., for 15 universities combined) in citation counts and OA levels varied little across sources, there are significant impacts at the institutional level. There were clear examples of universities shifting dramatically in both of these metrics when moving across sources, some in opposite directions. This makes any rank comparison of citations or OA levels strongly dependent on the selection of bibliographic source.

Finally, we implemented a manual cross-validation process to check metadata records for samples of DOIs that initially appeared to be exclusive from each source, for each of the 15 universities. The records were compared across the three bibliographic sources and against (where accessible) the corresponding online research documents. The process revealed cases of missing links between metadata and search functionalities within each database (for both affiliation and DOI). This means that the real coverage of each source is unnecessarily truncated. Overall, it appears that MSA has the highest coverage of objects that initially appeared exclusive to other sources. However, it often has missing DOIs and affiliations that do not match with WoS, Scopus, or online documents.

There is also strong evidence that the effects of shifting sources may be more prominent for non-English-speaking and non-European universities. Similar signs were observable for universities that are low ranked or medium ranked in both citations and OA levels, while those that achieve high rankings in these measures show much smaller shifts in position when the data source is changed. Universities that are highly ranked on these measures also tend to be highly ranked in general rankings like the THES, suggesting a bias in reliability and therefore curation effort toward prestigious universities.

Our concluding message is: Any institutional evaluation framework that is serious about coverage should consider incorporating multiple bibliographic sources. The challenge is in concatenating unstandardized data infrastructures that do not necessarily agree with each other. For example, one primary task would be to standardize the publication dates, especially for longitudinal study. This may be possible, to a certain degree, by using Crossref or Unpaywall metadata as an external reference set. The development of the Research Organization Registry⁵⁵ may also provide further opportunities to improve on disambiguation of institution names via a community-managed data source. Tackling these problems is by no means trivial. However, it has the potential to greatly enhance the delivery of fairer and more robust evaluation.

Chun-Kai (Karl) Huang: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing. Cameron Neylon: Conceptualization, Data curation, Investigation, Methodology, Software, Project administration, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing. Chloe Brookes-Kenworthy: Conceptualization, Data curation, Investigation, Writing—review & editing. Richard Hosking: Conceptualization, Data curation, Investigation, Software, Writing—review & editing. Lucy Montgomery: Conceptualization, Project administration, Supervision, Writing—review & editing. Katie Wilson: Conceptualization, Writing—review & editing. Alkim Ozaygen: Conceptualization, Writing—review & editing.

The authors declare there to be no competing interests. The funder and internal university sponsor had no part in designing the study or describing the results.

This work was funded by the Research Office of Curtin University through a strategic grant, the Curtin University Faculty of Humanities, and the School of Media, Creative Arts and Social Enquiry.

The raw data collected from WoS and Scopus cannot be made available publicly due to the respective licensing terms. However, the curated and derived secondary data are made available, alongside the code used for the analyses, at Huang et al. (2019), as referenced in the main text. The data collected for the manual cross-validation (section 4.6) is fully available at Brookes-Kenworthy, Huang, et al. (2019), as referenced in the main text.

The authors would like to thank Alberto Martín-Martín and an anonymous reviewer for their valuable feedback, which has helped to improve this article.