A Bibliometric Analysis on the Topic of Social Policy

Abstract

Social policy relates to a variety of social phenomena. Over recent decades, it has attracted the interest of scholars from a variety of academic disciplines as well as that of national and international organizations. This study applied the bibliometric approach to identify and analyse articles present in the Web of Science database, published in the English language, containing the term social policy in the title, abstract, or author keywords. Bibliometric networks and analyses were conducted using bibliometrix and VOSviewer software. The results show that the publication trend of articles on social policy varied over time and in association with social phenomena; for instance, an increase in publications occurred in the wake of the 2008 financial crisis. COVID-19, environmental issues, migration, and austerity were the main social policy topics being addressed in more recent investigations. This paper advances our knowledge about the research trends on the topic of social policy. This analysis included all articles published until the end of 2023 and can thus be considered to be up to date at the time of publication. It highlights the latest topics social policy has become interested in and the range of research disciplines it is associated with.

1. Introduction

Research into social policy has long formed an integral part of both the political and social sciences, with the number of publications and topics being addressed duly increasing over time. Specific scientific journals on social policy have been launched, starting with the journal Social Policy & Administration—if we consider the UK context—at the end of the 1960s, and have since grown over time (Powell 2006). In addition, the range of social policies being addressed has broadened considerably over recent times, addressing new phenomena of present importance such as the environment (Cahill 2002), digital technology (Henman 2022), and the COVID-19 pandemic (Béland et al. 2021).

Whether considering the long history and wide breadth of research into social policy or its specific fields of application, applying the bibliometric analysis is both interesting and useful for gaining insight into how the corpus of knowledge on social policy has grown over time. For example, bibliometrics was applied to investigate publication trends on the following: social policy within the journal Social Policy & Administration (Powell 2006); welfare across the years 1974 to 1997 (Wormell 2000); the welfare state in the 21st century (Papachashvili et al. 2023); specific dimensions of social policy, such as poverty (Hassan et al. 2023); and the economics of subjective well-being (Dominko and Verbič 2019). However, no previous study has explicitly focused on social policy from an all-encompassing perspective.

The aim of this study was to assess the trends in academic research articles published until 2023 containing the term social policy in their title, abstract, or author keywords. Bibliometrics can be defined as “a quantitative analysis method that takes the external characteristics of scientific literature as research objects (Wang et al. 2020). It can be divided into literature statistical analysis, mathematical model analysis, system analysis, matrix analysis, network analysis, and so on, according to different display forms of empirical results” (Wang and Su 2020, p. 4). It is usually applied to very large data sets (i.e., hundreds to thousands of publications), as specified by Donthu et al. (2021). As underlined by Powell (2018), but see Donthu et al. (2021), bibliometrics can be used to provide a potentially valid, although imperfect, indicator of the importance of a scientific concept, organisation, country, or journal, and it is increasingly being valued as a tool for assessing scholarly quality and productivity. Thus, the results achieved through bibliometric analysis must be interpreted and described taking into account the technique’s limitations, which may be associated, for example, with the database from which the data were extracted, the search criteria set in the database, and those related to the specific methods applied (e.g., the exclusion of some records from the database due to incomplete data).

This paper is structured as follows: Section 2 introduces social policy research from a metric perspective; Section 3 presents the data and the methodology applied; Section 4 presents the results; Section 5 discusses the results and compares them with those of previous bibliometric analyses; and Section 6 draws some final conclusions.

2. Bibliometric Studies on Topics of Social Policy

Bibliometric studies on social policy in general or addressing specific areas of policy—such as social protection (Barr and Diamond 2010), inequality (McClelland et al. 2020), or social policy investment (Garritzmann et al. 2023)—are scarce and have yet to be consolidated in the literature. Considering the many studies being published on the wide range of topics connected to social policy (Powell 2006), it is important that this gap in the literature be addressed.

Powell (2006, 2016, 2018) is one scholar making an important contribution to this analysis through the use of the bibliometric and citation approach. Indeed, some of his empirical studies specifically analysed publications in journals dedicated to topics on social policy such as Social Policy & Administration. In one of his first studies on the topic, Powell (2006) analysed articles published in the first 40 years of the oldest British journal on social policy, namely Social Policy & Administration (launched in 1967 under the name Social and Economic Administration, S&EA, changing to Social Policy & Administration in 1979). As well as being one of the first journals on the subject area, it is also one of the most important. Powell (2006) also provides a useful chronological description of other British journals on social policy, namely Journal of Social Policy and Policy and Politics, launched in 1972; Critical Social Policy: a Journal of Socialist Theory and Practice in Social Welfare (CSP), launched in 1981; Journal of European Social Policy, launched in 1991; and Social Policy and Society (launched in 2002). In this study, Powell showed how the discipline of social policy has always exhibited great variety in its research areas, attracting leading scholars across the globe and leading to the production of important and influential research.

In a later study, Powell (2016) adopted the concept of “citation classics” (Garfield 1977)—he was the first author to explore the concept of citation classics in the area of social policy—to analyse 79 articles cited at least 50 times (computed by the Web of Science) and published in the five leading journals on social policy, namely Critical Social Policy, International Journal of Social Welfare, Journal of European Social Policy, Journal of Social Policy, and Social Policy & Administration. The concept of citation classics allows scholars to identify the most cited articles (which constitute important reference points within a field of research), emerging topics, and the main intellectual research groups, institutions, or countries. Powell (2016) identified that over half of the articles published on social policy were written by authors based in the UK. In addition, he noticed that about two-thirds of articles were published between 2000 and 2008. A small number of articles were of a qualitative nature or review articles about two thirds were classified as “conceptual”, and about a quarter were “quantitative” in nature. Nearly two-thirds were comparative in setting or focus, while about a quarter were UK-based. Welfare regimes were the leading issue. Around a third of the articles focused on a particular service area, with the leading areas being employment, health, social/community care, or long-term care. Of the top ten most cited articles, five articles were conceptual, two were reviews, and three were quantitative. The leading topic was welfare regimes. Only two papers were based on sectoral areas, namely health and employment; seven were comparative; and three were specific to the UK.

In another, more recent exploratory study, Powell (2018) extended his research to reveal the most cited articles in social policy journals. From the outset, the author was faced with two problems: the first regarded the definition of social policy research; the second was connected to the list of “influential” social policy journals, which according to Powell (2018), revealed a further two issues, namely (i) how citations are linked with influence and (ii) difficulty in considering some journals, such as Policy and Politics, as strictly social policy journals. The empirical basis of this research regarded 24 works with over 2000 Google Scholar citations (Powell 2018). The key findings included the following: Esping-Andersen (1990) is the author with the highest number of total citations, while the most cited authors on social policy (or social administration) were Townsend (1979), Titmuss (1970), Lister (1997), and Lewis (1992). Of these 24 works, only 9 were by social policy researchers; instead, sociologists and political scientists dominated the list. Powell (2018) concluded his analysis by underlining that high citation numbers are indicative of impact and influence but not necessarily quality. Moreover, the author observed that the choice of the subject matter and its classification is subjective, and that the focus should lie more on the impact of individual works rather than the authors’ contribution to the field.

In a recent study, Papachashvili et al. (2023) analysed social policy publications concerning the welfare state in the 21st century and demonstrated how the number of scientific publications on this theme was influenced by the economic–social conditions at the time of publication. For example, they noticed an increase in the number of publications after the start of the 2007–2009 global financial crisis (Michie 2022). According to Papachashvili et al. (2023), a further stimulus for the increase in publications into the welfare state occurred during the deflection and recession period, identifiable with the time following the year 2015, and more recently the events associated with the COVID-19 pandemic. These are just some examples of how publications addressing specific social policy topics—in this case, the welfare state—correlate with social events. Thus, trends in scientific publications may reflect concurrent trends in social phenomena.

Although the use of metrics and citation statistics for the study of scientific production is characterised by certain methodological limitations (Bornmann and Hans-Dieter 2008; Holden et al. 2005b; Leydesdorff 2008; Meho 2007; Powell 2018), this approach has undeniably helped further scientific research and contributed towards helping higher education institutes and funders assess the research being conducted and evaluate how well investments in the area are managed (Powell 2018). Bibliometrics has also been beneficial for evaluating the impact of individual scientists, universities, journals, and countries in specific fields of research over extended periods of time (Holden et al. 2005a, 2010; Martinez et al. 2015), and the information collected, such as citation rates, authors, and author keywords, organized in a database can be analysed using a range of bibliometric techniques.

The specific aim of this study was to contribute towards filling the gap in the research (Boon 2017; Donthu et al. 2021; Sooryamoorthy 2020) by conducting an exploratory, quantitative bibliometric investigation and systematic analysis of the use of the term social policy (or social policies) in scientific journal publications.

The specific questions driving the analysis were the following:

RQ1. What are the trends in the output rate of the scientific literature for articles with the term social policy (or social policies) in the title, abstract, or author keywords until the end of 2023?

RQ2. Which of these scientific articles were the most cited, and which journals have had the greatest impact?

RQ3. What author keywords are most used in articles on social policy, and how is the network of co-occurrences configured?

RQ4. How is the network of co-citation resources, authors, and references configured?

3. Methods and Data

The bibliometric analysis followed the specific steps summarised in

Table 1. Principal steps to follow for a bibliometric analysis.

Order	Steps	Description	Expected Outcome
1	Define Research Objectives	Clearly outline the objectives of the bibliometric analysis	Clear research questions and objectives
2	Literature Search and Data Collection	Collect relevant literature from Web of Science, Scopus, and Google Scholar databases or collect raw data (e.g., from no database) and build your own database	A comprehensive dataset of relevant publications
3	Data Cleaning and Preprocessing	Clean and preprocess the data to ensure accuracy (e.g., removing duplicates and correcting author names)	A refined and accurate dataset ready for analysis
4	Selection of Bibliometric Techniques	Choose appropriate bibliometric techniques based on research objectives (e.g., co-citation analysis, co-word analysis, and bibliographic coupling)	Identification of suitable analysis techniques
5	Data Analysis	Conduct the analysis using chosen techniques	Insights and patterns in the literature
6	Visualization	Visualize the results to aid interpretation and presentation	Graphs, maps, and other visual representations of the data
7	Interpretation and Reporting	Interpret the results and prepare a report detailing the findings and their implications	A comprehensive report with insights and recommendations

Adapted from Passas (2024).

. The first step corresponds to the definition of the research objectives, as outlined in the previous section. The present section presents steps two to five.

The second step involves data source(s) selection and downloading the dataset. Since different bibliographic sources cover different scientific areas, the most appropriate bibliographic sources should be selected according to the research objective. Furthermore, not all bibliographic sources provide the information necessary to conduct a bibliometric analysis (Moral-Muñoz et al. 2020). The main bibliographic databases used in bibliometric analyses of human science disciplines are Web of Science (WoS), Scopus, and Google Scholar (GS); other databases focus more specifically on scientific disciplines, such as the Chemical Abstracts Service (CAS) databases (Neuhaus and Daniel 2008). The choice of database will determine the outcome of the search (Caputo and Kargina 2021).

Table 2. Characteristics of the databases: Web of Science (WoS), Scopus, and Google Scholar (GS).

Database	Description	Producer	Coverage	Subscription	Records	API	Formats
WoS	Provides access to multiple databases and citation data for 256 disciplines, i.e., science, social science, arts, and humanities.	Institute for Scientific Information (ISI) was the original producer, after the intellectual property passed to Thomson Reuters, and now the maintenance is in charge of Clarivate Analytics	Full-text articles, reviews, editorials, chronologies, abstracts, proceedings (journals and book-based), and technical papers. Total number of records is beyond 90 million. Temporal coverage from 1900 to the present.	Yes	500 records with a total of 100,000 per query	Yes	Plain text and tab-delimited
Scopus	Offers access to databases and citation data in life sciences, social sciences, physical sciences, and health sciences.	Elsevier	Book series, journals, trade journals, and patent databases. Number of records is around 69 million. Temporal coverage is from 2004 to present.	Yes	First 2000 records	Yes	RIS and CSV
GS	Index the full text or metadata of the scientific literature from most peer-reviewed online academic publications. It was launched in 2004, and it has been criticized for not banning predatory journals.	Google	The literature from most peer-reviewed online academic journals, books, conference papers, theses, preprints, abstracts, technical reports, court opinions, and patents. In 2018, it was estimated to hold 389 million documents.	Yes	n/a	No	n/a

Adapted from Moral-Muñoz et al. (2020).

summarises the main characteristics of WoS, Scopus, and GS.

As seen in Table 2, the WoS, Scopus, and GS databases are the most useful databases for bibliometric research into the social sciences as they provide access to many different sources. Their coverages span different years and, as underlined by Öztürk et al. (2024), WoS has long been the world’s most comprehensive source of citation data, while Scopus is considered the second largest comprehensive citation database. The data source selected for use in this research was the WoS Core Collection database (https://clarivate.com/ accessed on on 29 July 2024), an internationally recognized database indexing the most prestigious journals (Merigó and Yang 2017) and comprising other “sub-databases”. In agreement with Öztürk et al. (2024), the main criteria guiding the choice of this database were the number of journals covered in the research field in question, access to the database through the institutional membership, the possibility to download the data in a format compatible with the analysis software, and the number of records that can be exported from the database.

To identify the articles of interest, I formulated a detailed query string to identify the studies that could be addressing matters related to social policy since the term “social policy” is also used in its plural form, the truncated form “social polic*” was always used so that either form occurring in the fields title, abstract, or author keywords would be detected. Since the full text is not always present in the indexing database, only these three fields were searched. However, it must be noted that whilst these fields are important for providing readers with a general overview of an article’s content, in particular, the author keywords, which summarise the most important concepts covered in the article, searching these alone and not the full text may lead to some articles relevant to social policy being missed. This aspect represents a limitation of this type of analysis, which must be considered when interpreting the results. The search was carried out on WoS and the “sub-databases” SSCI, ESCI, and SCI-EXPANDED. Only articles in the English language were considered. Proceedings papers, early access articles, book chapters, and data papers were excluded. The time span considered was set until 31 December 2023 without specifying the start date. All documents identified were extracted on 29 July 2024.

The query string identified a total of 9541 articles which were then subjected to a cleaning process—the third step of the bibliometric analysis (Passas 2024). The cleaning process controls for the presence of DOI, author names, titles, abstracts, and author keywords. Articles missing any of these were removed from the sample. A subsequent check identified and removed any duplicated records (same DOI and title). The final database comprised 6287 articles with complete DOI, author names, titles, abstracts, and author keywords. The choice to exclude articles missing a DOI from the database was dictated by the exploratory nature of this study in order to focus on clearly identifiable articles only, at both the national and international level; although, this again brought about a limitation to this study. The cleaning process also involved identifying the articles published in journals that had changed their name over the course of time (e.g., Social Policy & Administration was originally launched as Social and Economic Administration, and Journal of International and Comparative Social Welfare was formerly called Journal of Comparative Social Welfare and before that New Global Development) and grouping them together as belonging to the same journal independent of the journal name at the time of publication for the analysis.

The fourth step of the bibliometric analysis addressed publication- and citation-related metrics (Merigó and Yang 2017; Merigó et al. 2015) and involved the application of specific bibliometric techniques (Passas 2024). This was followed by a co-word analysis and by citation analysis (Cancino et al. 2017; Blanco-Mesa et al. 2017). Co-words are often derived from “author keywords”, and the analysis assumes that words that frequently appear together have a thematic relationship with one another (Donthu et al. 2021). Citation analysis is similar to co-word analysis but assumes that publications that are cited together are likely to be on similar themes (Hjørland 2013). These analyses can be used to reveal the intellectual structure of a field of research (Donthu et al. 2021).

Here, I used the R package bibliometrix (Aria and Cuccurullo 2017) and VOSviewer software (version 1.6.20 (0)) (van Eck and Waltman 2010, 2014b), commonly used in bibliometric research, to generate bibliometric network information and produce the correlated maps. As in previous research (e.g., Aria et al. 2022; Baraibar-Diez et al. 2020; Hassan et al. 2023), these open-source tools were used to analyse the data related to publications and citations metrics. All other analyses, including the cleaning process, were carried out in Microsoft Excel (version 16.81).

4. Results

The Results Section can be considered as corresponding to the six steps of the bibliometric process presented in Table 1, whereas the Discussion Section represents the final step of the process. Each sub-section presents the results of the different elements of the bibliometric analysis applied.

4.1. The Bibliometric Analysis of Social Policy Publications

This section presents the results of the analysis of the number of articles published, the citation rate, the most influential journals according to the database investigated, and the most cited articles containing the term “social policy”.

4.1.1. Trends in the Number of Annual Publications Containing the Term Social Policy

Figure 1 reports the number of scientific articles with a DOI code published per year containing the term “social polic*” (from here on referred to as simply “social policy”) in the title, abstract, or author keywords.

In 1991, just five English language articles were published containing the term “social policy”. From here on, the number of articles published in each subsequent year generally increased, exceeding 100 publications in 2006. This upward trend continued over the following decade, reaching its peak in the year 2020, when 486 articles matched the research criteria. In 2021, the number of articles began to decline, with a sharp decrease in 2023, when only 388 articles published. These figures are underestimated, however, as 3229 articles were removed from the database produced due to their absence of a DOI or author keywords.1 A review of the uncleaned database reveals the presence of articles with the term social policy dating back to 1991. A closer examination of the uncleaned database revealed that there were 38 articles containing the term social policy in 1985, 23 in 1986, and 25 in 1987. Nonetheless, the trend still depicts the overall increasing trend in the number of publications addressing the topic of social policy.

Table 3. General citation structure of published articles containing the term social policy (in the article title, abstract, or author keywords).

Number of Citations	Number of Articles	Accumulated Number of Articles	% Articles	% Accumulated Articles
n ≥ 500	12	12	0.19	0.19
250 ≤ n < 500	20	32	0.32	0.51
100 ≤ n < 250	183	215	2.91	3.42
50 ≤ n < 100	398	613	6.33	9.75
25 ≤ n < 50	807	1420	12.84	22.59
10 ≤ n < 25	1607	3027	25.56	48.15
1 ≤ n < 10	2674	5701	42.53	90.68
n = 0	586	6287	9.32	100.00
Total	6287		100.00

reports the following descriptive statistics on the general citation structure of the articles meeting the inclusion criteria in the cleaned database: the number of articles and accumulated number of articles per citation rate range; the percentage of articles and the accumulated percentage of articles per citation rate range. A total of 12 papers were cited more than 500 times (0.19%), 20 were cited between 250 and 500 times (0.32%), 183 were cited between 100 and 250 times (2.91%), and 398 were cited between 50 and 100 times (6.33%). Each of these citation rate ranges comprised less than 10% of all published articles. The most common citation rate range was by far 1 to 10 (n = 2674 articles; 42.53%). A total of 807 articles were cited 25 to 50 times (12.84%), and 1607 were cited between 10 and 25 times (25.56%). A total of 586 scientific articles (9.32%) had never been cited up until the end of 2023.

Of the 12 articles in the highest citation rate (>500 times), the most prolific was that by Marmot and Bell (2012), “Fair Society, Healthy Lives”, published in Public Health, which received more than 1230 citations.

The WoS database also classifies each article according to “research area”. This information offers the possibility to gain insight into the general trends involving social policy topics. However, since the database download was subsequently cleaned, the information provided by WoS is not identical to the database analysed. Nevertheless, this information can still provide a general sense of the research areas involved in social policy. The most common research area cited—according to the WoS—in the uncleaned sample of articles into social policy was social work, involving 17.39% of all articles, followed by social issues with 14.43% of publications, public administration with 12.58%, political science with 11.05% of articles, sociology with 8.51%, and economics with 8.00%. Social policy itself did not appear as specific research area.

Data on “author productivity” were computed using bibliometrix, resulting in an overall total of 13,489 authors, of which 2140 were authors of single-authored documents. The most prolific author (as a single or co-author) was Béland D., with 29 papers (0.46% of the 6287 analysed), followed by Gal J. and Shang X., with 12 (0.19%) publications each. Daly M. has published 10 articles (0.15%) as have Fisher K. R., Roose R., Stephens C., and Weiss-gal I. Humpage L., Marston G., Mooney G., Parolin Z., and Powell M. have all published nine articles (0.14%).

The academic institution associated with the highest percentage of articles in the cleaned database was the University of Toronto (2.48% of the articles), followed by the University of Oxford (2.09%), Columbia University (1.78%), the University of Sidney (1.74%), the University of Queensland (1.59%), and King’s College London (1.52%). Articles from authors based in the USA comprised approximately 20% of the sample (n = 1213), whereas 17.51% (n = 1100) involved authors working in the UK, and 6.85% (n = 431) had authors based in Australia. Authors in Canada contributed to 5.64% (n = 355) of all publications, while those in Germany concerned 4.24% (n = 267). As expected, the most productive universities and researchers were all collocated in the United States, United Kingdom, and Australia.

4.1.2. The Prominent Journals Publishing Articles Containing the Term Social Policy

Table 4. The top ten journals publishing articles containing the term social policy (in the title, abstract, or author keywords).

Journal	Publisher	Articles	%
Social Policy & Administration	Wiley	298	4.74
Journal of European Social Policy	Sage	138	2.20
Critical Social Policy	Sage	115	1.83
International Journal of Social Welfare	Wiley	103	1.64
Social Policy and Society	Cambridge University Press	102	1.62
Social Science & Medicine	Elsevier	93	1.48
British Journal of Social Work	Oxford University Press	74	1.18
Social Indicators Research	Springer	61	0.97
International Journal of Sociology and Social Policy	Emerald	58	0.92
Journal of European Public Policy	Taylor & Francis	53	0.84

presents the top ten leading journals publishing articles containing the term social policy in their title, abstract, or author keywords. The journal with the highest number of articles published (until 2023) is Social Policy & Administration, with 298 documents (4.74% of the 6287). The Journal of European Social Policy was the second most prolific journal on social policy related research with 138 (2.20%) articles. The Critical Social Policy is next with 115 (1.83%), followed by International Journal of Social Welfare with 103 documents (1.64%). The fifth position is held by Social Policy and Society with 102 papers (1.62%), followed by Social Science & Medicine with 93 articles (1.48%), British Journal of Social Work with 74 (1.18%), and Social Indicators Research and International Journal of Sociology and Social Policy with 61 (9.97%) and 58 articles (0.92%), respectively. Journal of European Public Policy is ranked tenth with 53 (0.84%) articles.

The most relevant publishers are Wiley, Sage, Cambridge University Press, Elsevier, Oxford University Press, Springer, Emerald, and Taylor & Francis.

Finally, the names of the journals publishing the most cited papers on topics involving social policy can be observed to cover a variety of research areas, namely social policy, welfare, social work, sociology, public policy, and social aspects of medicine.

4.1.3. The Most Cited Articles Dealing with Aspects of Social Policy

The number of times an article is cited provides an indication of its academic attractiveness and influence in the research area (Garrigos-Simon et al. 2018).

Table 5. Most cited articles containing the term social policy.

R	Title	Authors	Journal	Year	TC	C/Y	Normalized TC
1	“Fair Society, Healthy Lives”	Marmot, M., Bell, R.	Public Health	2012	1233	94.85	33.07
2	“The relative influence of individual, social and physical environment determinants of physical activity”	Giles-Corti, B., Donovan, R.J.	Social Science & Medicine	2002	821	35.70	14.16
3	“New Frontiers in the Future of Aging: From Successful Aging of the Young Old to the Dilemmas of the Fourth Age”	Baltes, P. B., Smith, J.	Gerontology	2003	807	36.68	17.28
4	“Environmental and policy—Interventions to promote physical activity”	Sallis, J. F., Bauman, A., Pratt, M.	American Journal of Preventive Medicine	1998	801	29.67	13.23
5	“Three Worlds of Welfare Capitalism or More? A State-of-the-art Report”	Arts, W., Gelissen, J.	Journal of European Social Policy	2002	728	31.65	12.55
6	“Exposure to Opposing Views on Social Media Can Increase Political Polarization”	Bail, C. A., Argyle, L. P., Brown, T. W., Bumpus, J. P., Chen, H., Hunzaker, M. B. F., Lee, J., Mann, M., Merhout, F., Volfovsky, A.	Proceedings of the National Academy of Sciences (PNAS)	2018	680	97.14	44.38
7	“The Uses of Neoliberalism”	Ferguson J.	Antipode: A Radical Journal of Geography	2010	672	44.80	19.38
8	“The nature of social dominance orientation: Theorizing and measuring preferences for intergroup inequality using the new SDO7 scale”	Ho, A. K., Sidanius, J., Kteily, N., Sheehy-Skeffington, J., Pratto, F., Henkel, K. E., Foels, R., Stewart, A. L.	Journal of Personality and Social Psychology	2015	616	61.60	29.26

R: ranking; TC: total citations that the article has received from documents indexed on the WoS; C/Y: citation per year, the yearly average number of times each manuscript has been cited; Normalized TC: calculated by dividing the actual count of citing items by the expected citation rate for documents with the same year of publication.

lists the articles which have been cited more than 500 times. As the search does not focus specifically on social policy journals, it is expected that articles will also be published in journals focused on different research fields.

At the time of the present analysis, the most cited research article was “Fair Society, Healthy Lives” by Marmot and Bell (2012), published in Public Health in 2012. It has been cited 1233 times to date, with an average citation rate of 94.85 citations per year.

The next most cited documents were the following: “The Relative Influence of Individual, Social and Physical Environment Determinants of Physical Activity” by Giles-Corti and Donovan (2002) and “New Frontiers in the Future of Aging: From Successful Aging of the Young Old to the Dilemmas of the Fourth Age” by Baltes and Smith (2003), with 821 and 807 citations, respectively. The first had an average citation rate of 35.70 citations per year, while the latter was slightly higher at 36.68 citations per year. Another document with a considerable number of citations (29.67 citations per year) is that by Sallis et al. (1998), published with the title “Environmental and Policy—Interventions to Promote Physical Activity”. Another article above the 700 citations threshold was that by Arts and Gelissen (2002) “Three Worlds of Welfare Capitalism or More? A State-of-the-art Report”. Three articles had 600 to 700 citations: Bail et al. (2018), with 680 citations and a citation rate of 97.14 citation per year; Ferguson (2010), 672 citations and 44.80 citation per year; and Ho et al. (2015), 616 citations and 61.60 citation per year.

This ranking, presented in the seventh column of Table 5, considers the number of citations per year, but if we consider the normalized total citations (eighth column), the ranking changes. In this case, the first article in the classification is that by Bail et al. (2018), followed by the publication by Marmot and Bell (2012). In third and fourth places are the articles by Ho et al. (2015) and Ferguson (2010), respectively, followed by the papers by Sallis et al. (1998), also the oldest paper in the ranking, and Arts and Gelissen (2002).

4.2. Network of the Author Keyword Co-Occurrence

Author keywords were analysed to obtain information on their frequency of use. The magnitude of the frequencies is represented by the relative sizes of the circles indicating each node in Figure 2.

The curved lines connecting two nodes indicate that the two author keywords appeared together in the same paper, and the thickness of the line indicates the strength of the link between the two items, i.e., social policy and social work, and the shorter the distance between two nodes, the greater the frequency of the keyword co-occurrences (van Eck and Waltman 2014b). The analysis was performed using the co-occurrence tool provided via the software VOSviewer, which counts the number of articles in the sample in which two keywords appear together (binary counting method). No weight was applied to the analysis, and only the keywords reported by the authors were considered. Since the item “social policy” was considered in both the singular and plural form, these occurrences were merged, but they are presented here in the singular form for the sake of simplicity. The results presented from here on reflect this arrangement.

A total of 14,389 author keywords were identified in the 6287 articles. To create a readable map, I applied a threshold of “25 keyword occurrences” using the fractional counting method, as used for all analyses contained herein (Tsilika 2023). A total of 147 keywords met this criterion, meaning that these 147 keywords occurred at least 25 times in the sample of articles. Furthermore, for each of these 147 keywords, the strengths of the co-occurrence links with other keywords were computed as well as the links between two keywords.

Table 6. The top 20 author keywords co-occurring in the articles analysed.

Rank	Keyword	No. of Articles in Which the Keyword Occurred	Links	Total Link Strength of Co-Occurrences	Cluster
1	Social policy	2182	146	1620	2
2	Welfare state	498	122	433	3
3	Poverty	305	114	266	7
4	Gender	159	93	129	3
5	Inequality	148	80	123	3
6	Health	141	91	118	1
7	Social work	135	82	117	2
8	Education	120	81	98	8
9	Neoliberalism	115	74	99	8
10	China	113	64	85	1
11	Disability	99	66	86	6
12	Employment	95	68	83	4
13	Unemployment	94	64	78	1
14	COVID-19	90	65	73	8
15	Governance	84	54	67	4
16	Latin America	80	53	70	5
17	Family	74	56	56	1
18	Social exclusion	74	47	58	2
19	Children	72	47	58	2
20	Social protection	70	50	62	7

presents 20 of the 147 keywords that occurred in more than 50 of the 6287 articles related to social policy. The column “links” indicates the strength of the co-occurrence connecting the keyword with other keywords (the higher this value, the stronger the link); the column “total link strength of co-occurrences” indicates the number of publications in which two keywords occur together; and the cluster membership indicates in which cluster the keyword is included, provided by default via the VOSviewer software algorithm.

Most frequent keywords were social policy, welfare state, and poverty, then gender, inequality, and health (Table 6).

The lines (Figure 2) connecting the node “social policy” are thickest going towards the keywords welfare state (129.99), poverty (79.80), social work (45.78), gender (32.49), and inequality (27.45), being the keywords with the highest link strength value for total co-occurrences.

The VOSviewer program identified eight clusters for the 147 keywords, represented by eight different colours in Figure 2. Each colour clusters words according to a subfield of social policy. In addition, from the figure we can notice that more than one cluster is characterized by a geographical reference. The largest cluster, shown in red (cluster 1, 31 items), is led—according to occurrence frequency—by the item health, followed by China, unemployment, and family. This cluster also includes other sub-dimensions, including care, well-being, and aging. A dominant trait of this cluster appears to be health conditions. The geographical areas related to this cluster are mainly China, Japan, and the United States. The green cluster (cluster 2, 28 items) is led by the keyword social policy. Other items with a significant occurrence in this cluster are social work, social exclusion, and children. This cluster appears to depict aspects related to the social dimension of exclusion; thus, the field of social work. Russia is the nation most frequently mentioned in this cluster. In addition, the cluster includes the keyword qualitative research, indicating the most prevalent research methodology used in this research area. Welfare state is the most frequent keyword in the blue cluster (cluster 3, 23 items). It is also associated with sub-dimensions related to gender and inequality, and includes aspects related to migration. In the yellow cluster (cluster 4, 18 items), the most cited keywords are employment and governance, and it mainly concerns Australia, Germany, Italy, New Zealand, Spain, Sweden, and countries making up the UK. The purple cluster (cluster 5, 15 items) is led by the keyword social security. It includes sub-dimensions concerning health care and the European union as well as the topic of housing. The cluster in light blue (cluster 6, 13 items) is led by the keywords disability and COVID-19, and the countries most considered are Brazil, Mexico, and South Africa. The orange cluster (cluster 7, 11 items) focuses on the dimension of poverty, and it is associated with the dimensions of social protection and development. The geographic area associated with cluster 7 is Africa. The last cluster, shown in brown (cluster 8; 8 items), mainly focuses on education and neoliberalism, with Turkey and India being two associated geographical areas.

Figure 3 shows the co-occurrences of the author keywords according to the average number of articles published per year. This allows us to consider how topics have changed over the years. Trending topics from 2013 to 2017 are presented as dark purple for earlier years and brighter lighter colours for later years. The curves connecting the nodes denote their appearance in the same publication. The shorter the distance between two nodes, the greater the frequency of co-occurrences of the two keywords.

For example, the items social capital, social security, and unemployment were the keywords most associated with social policy around the year 2015, while the themes COVID-19, migration, social investment, and austerity appear in more recent papers.

4.3. Analysis of Co-Citations

The following three sections present the results of the co-citation analysis between journals, authors, and references. Co-citation analysis observes the simultaneous citation of two items (such as journals, authors, or articles) by a third item (i.e., a journal, author, or article). According to this perspective, the more two items are cited together, the more likely they are to be related, and the closer they will be located to each other in the network figures reported in the following sections. The size of the node reflects the number of citations that an item (i.e., journal, author, or article) has received (van Eck and Waltman 2014b; Small 1973).

4.3.1. Results of the Journal Co-Citation Analysis

Figure 4 depicts the results of the journal co-citation analysis. A total of 117 journals (of the 132,118 sources indicated by the VOSviewer software) were cited at least 200 times. Journal of European Social Policy is the only journal to be cited around 2000 times, precisely 1910 citations (links: 179; total link strength: 1613.13). Social Science & Medicine was cited 1763 times (links: 112; total link strength: 1315.82), and Social Policy & Administration was cited 1546 times (links: 179; total link strength: 1522.38). The Journal of Social Policy was cited 1428 times (links: 112; total link strength: 1248.04), while American Sociological Review was cited 1271 times (links: 112; total link strength: 1152.44). The American Political Science Review was cited 1197 times (links: 112; total link strength: 1074.79).

The software VOSviewer recognized five clusters of journals. The biggest cluster (cluster 1, 33 journals), shown in red, is led—according to the citations rank—by Social Science & Medicine,2 the Lancet (citations: 974; links: 112; total link strength: 636.48), and the American Journal of Public Health (citations: 859; links: 111; total link strength: 715.05). It also includes the Journal of Personality and Social Psychology with 753 citations (links: 109; total link strength: 537.69). This cluster mainly groups journals from social, psycho-social, and social medical areas.

The next biggest cluster, shown in green (cluster 2, 24 journals), is led by the journals Social Policy & Administration, Journal of Social Policy, and the Critical Social Policy (citations: 755; links: 111; total link strength: 584.63). The British Journal of Social Work (citations: 874; links: 106; total link strength: 577.72) is also a member of this cluster. This group clusters journals which mainly publish in the area of social policy.

Cluster 3, comprising 20 journals and represented in blue, is led by the Journal of European Social Policy, American Political Science Review, American Journal of Political Science, Journal of European Public Policy (citations: 854; links: 105; total link strength: 695.36), and Comparative Political Studies (citations 746; links: 109; total link strength: 689.83). This cluster contains the highest number of journals which mainly focus on political aspects of social policy research.

The next biggest cluster (cluster 4, 19 journals) is shown in yellow. Its leading journals are American Economic Review (citations: 777; links: 111; total link strength: 689.87) and Social Indicators Research (citations: 734; links: 112; total link strength: 579.84). This cluster is the most heterogeneous and includes journals dealing with economics and the measurement and the development of studies on the sociological dimension of social policies.

Finally, the fifth cluster, composed of 17 journals and shown in purple, is led by American Sociological Review, American Journal of Sociology (citations: 855; links 112; total link strength: 781.24), Journal of Marriage and Family (citations: 805; links: 106; total link strength: 645.14), and European Sociological Review (citations: 721; links: 111; total link strength: 647.55), together representing the sociological journals.

4.3.2. Results of the Author Co-Citation Analysis

Figure 5 depicts the co-citation of referenced authors in the field of social policy (since only the first author of a cited document was included in the analysis of WoS data). Furthermore, organisations such as Eurostat, OECD, and World Bank as authors were excluded to keep the focus on individual researchers. The analysis included the 215 most cited authors from the total of 137,692 (threshold applied: 50 citations).

Figure 5 indicates G. Esping-Andersen (citations: 1199; links: 209; total link strength: 1072.18) to be the most cited author, followed by P. Pierson (citations: 552; links: 189; total link strength: 505.18) and W. Korpi (citations: 452; links: 192; total link strength: 421,48). Other cited authors are M. Foucault (citations: 395; links: 145; total link strength: 211.95), G. Bonoli (citations: 327; links: 180; total link strength: 303.65), and E. Huber (citations: 321; links: 160; total link strength: 290.47).

The cited authors are clustered into five different groups. Cluster 1, shown in red, contains a total of 64 authors. The most cited authors in this cluster are M. Foucault, A. Giddens (citations: 264; links: 180; total link strength: 222.17) and P. Bourdieu (citations: 263; links: 165; total link strength: 184.47). Cluster 2 (green, 50 authors) is led by P. Pierson followed by W. Korpi and G. Bonoli. The blue cluster (cluster 3, 49 authors) is most influenced by A. Sen (citations: 264; links: 180; total link strength: 222.17), A. Alesina (citations: 164; links: 141; total link strength: 131.96), and E Diener (citations: 139; links: 80; total link strength: 82.68). The yellow cluster (cluster 4, 29 authors) is led G. Esping-Andersen and E. Huber. The purple cluster (cluster 5, 23 authors) is led by M. Ferrera (citations: 283; links: 189; total link strength: 260.57), W. Streeck (citations: 176; links: 156; total link strength: 159.13), and A. Hemerijck (citations: 165; links: 157; total link strength: 157.40). The authors with the highest values of total link strength are G. Esping-Andersen, P. Pierson, W. Korpi, G. Bonoli, and E. Huber, and this ranking is different compared with the citation ranking, which placed M. Foucault before G. Bonoli and E. Huber.

4.3.3. Results of the Reference Co-Citation

The analysis of co-citated references reveals pairs of papers that are both cited by the same document. The references comprise all those listed in the bibliographies of the 6287 documents analysed, and this aspect must be considered in the interpretation of the results. I applied a minimal citation threshold of 50 citations in the analysis, and only 37 of the 261,043 cited references met this threshold and are represented in Figure 6.

The most cited reference was “The Three Worlds of Welfare Capitalism”, Princeton University Press (citations: 641; links: 31; total link strength: 430) by Esping-Andersen (1990), presenting the highest value of interconnection. The second is another publication by Esping-Andersen (1999), namely “Social Foundations of Postindustrial Economies”, Oxford University Press (citations: 133; links: 29; total link strength: 107). This is followed by “Dismantling the Welfare State? Reagan, Thatcher and the Politics of Retrenchment” by Pierson (1994), Cambridge University Press, which received 116 citations (links: 29; total link strength: 101), and the article by Korpi and Palme (1998) entitled “The paradox of redistribution and strategies of equality: Welfare state institutions, inequality, and poverty in the Western countries”, American Sociological Review, 63(5) (citations: 115; links: 29; total link strength: 101). These are followed by a work by Pierson (1996) entitled “The New Politics of the Welfare State”, Cambridge University Press (citations: 94; links: 30; total link strength: 85).

These documents are clustered into six groups. The red cluster includes eight documents and is led by the paper by Pierson (1994) “Dismantling the Welfare State? Reagan, Thatcher and the Politics of Retrenchment”. The green cluster contains seven items and is led by the two publications of Esping-Andersen (1990, 1999). The blue cluster is composed of five publications and is led by the article by Pierson (1996) “The New Politics of the Welfare State”. The yellow cluster (five items) is led by Marshall (1970) “Social Policy in the Twentieth Century”, published by Hutchinson. The purple cluster, composed of four items, is led by the book by Wilensky (1975), “The Welfare State and Equality: Structural and Ideological Roots of Public Expenditures”, Oxford University Press (citations: 69, links: 29, total link strength: 222). The last cluster (four items), shown in light blue, is led by the publication by Korpi and Palme (1998) “The paradox of redistribution and strategies of equality: Welfare state institutions, inequality, and poverty in the Western countries”. To identify the aspects common to these citations is not easy, since co-citations not only reflect the content characteristics of the publication, but, as underlined by Powell (2016, 2018), they may reflect “ceremonial citations” or be governed by the phenomenon of “cronyism”.

5. Discussion

This analysis shows that the publication of articles dealing with the term of social policy is characterised by an upward trend, indicating a growth in the academic interest in the field (Papachashvili et al. 2023). In addition, comparing the results of the present study with those of previous works, such as those by Powell (2006, 2016, 2018) and Papachashvili et al. (2023), is not always possible. Whilst the present study considered all articles addressing the subject of social policy (in the title, abstract, or author keywords) independent of the publication source, Powell (2006, 2018) focused on social policy journals only and the cited articles therein, and Papachashvili et al. (2023) specifically addressed the social welfare dimension. If, on the one hand, the open nature of the present study is a limitation, on the other hand, it offers the advantage of demonstrating how the term social policy was used and cited in other research fields and how it has spread across to these other disciplines. After the 2008 financial crisis, and following the 2015 European refugee crisis, there was a noticeable increase in the number of the papers published containing the term social policy, demonstrating the possible presence of a relationship between contemporary social phenomena and the research activities of those interested in social policy, sustaining social policy to be a vigorous field of research. The interest towards phenomena such as COVID-19, environmental issues, migration, and austerity also provides tangible evidence of the relationship between social policy studies and the actual phenomena characterising post-modern societies.

The results revealed the topic of social policy to be present in the WoS database from 1991 onward; however, as already underlined, earlier publications may have been removed due to the application of the article selection criteria. Nevertheless, the finding still indicates social policy to be a consolidated area of research, confirmed by the rise in the number of articles published in subsequent years as well as by the increasing trend in citation rate. Indeed, the WoS database contains articles using the term social policy in the title, abstract, or author keywords dating back to the year of its establishment in 1985.

From 2021, the number of publications related to social policy started to decrease, and the data do not allow us to identify the causes of this trend. However, one possible hypothesis to investigate in future studies is that increased specialisms in the discipline may have led to a decline it the term’s use. For instance, since social policy has come to encompass a number of policy fields—education, health, housing, poverty, employment, and, most recently, environmental social policy (Hirvilammi et al. 2023)—it may have become less useful for researchers to speak of “social policy” in general terms, but rather to engage in conversations that target the specific or subfields of social policy they are interested in. This might be evident for journals specifically focusing on social policy as well as for all other journals.

The majority of papers using the term social policy have been cited 1–10 times, whereas very few (twelve) had a high citation score of over 500. The most cited article was published by Public Health and provides an overview and reflections on the findings from the World Health Organization Commission on the Social Determinants of Health and the Strategic Review of Health Inequalities in England—i.e., it focuses on a very specific dimension of social policy. However, considering the broad range of journals publishing the other most cited articles shows how the topic of social policy has grown to encompass a wide range of disciplines, from medicine and psychology to geography.

Nonetheless, as expected and already demonstrated by Powell (2006, 2016, 2018), this study confirms that the top journals publishing articles containing the term social policy specifically relate to the research area of social policy, namely Social Policy & Administration; Journal of European Social Policy; Critical Social Policy; International Journal of Social Welfare; and Social Policy and Society.

Béland D. is the most prolific author, followed by Gal J. and Shang X. However, it should be noted that the most productive authors are not necessary the most expert in the field of social policy since the research performed by these authors was not necessary published in social policy journals but included journals covering other disciplines. The University of Toronto, the University of Oxford, Columbia University, the University of Sidney, the University of Queensland, and King’s College London were the most productive organisations. The results on productivity according to nation confirmed, as expected, the USA and the UK to be those contributing the most, in line with previous studies (Garrigos-Simon et al. 2018; Patrick and Hee 2020). This finding was also reflected in the outcome of the analysis on the co-authorship networks between countries (the results for which are not graphically presented in the Results Section), which indicated the home institutions of the most collaborative authors’ to be situated in the USA and UK.

Focusing on the author keywords, the most frequently used keyword was that of the search criterion, namely social policy. Other frequently used keywords were welfare state, poverty, gender, inequality, health, and social work, thus denoting the specific topics addressed in social policy research, as also emerged in the research by Papachashvili et al. (2023), as well as the topics being discussed in relation to the concept of social policy. This aspect highlights how the theme of social policies is linked to many other dimensions of daily life. Furthermore, these keywords led some of the other clusters linked with social policy, but which were composed of other keywords. One of the clusters concerned health and its related aspects (depression and aging), but also unemployment. Another cluster was led by the keyword social policy, which linked directly with the keywords social work and social exclusion, and the scientific articles in this cluster mainly dealt with these topics. This cluster also included the keywords children, migration, participation, and community; the latter two could emphasise the importance of the communitarian function of social policy. The keyword social capital was also present in this cluster, which one might consider as further evidence of the communitarian character of the articles associated with this cluster. The keyword welfare state appears to indicate a specific pilar of research into the area of social policy. The articles containing this author keyword are mainly related to gender-related inequality, as well as to class and race and organisations. Another sub-dimension concerns publications on unemployment and thus topics concerning the labour market and investment into policies on employment incentives. Other sub-groups concern a traditional dimension of social policy, namely social security, while others still focus on new phenomena such as the aftermath of the COVID-19 pandemic. This result demonstrates how the term social policy has been linked and considered in relation to many other aspects of daily life, underlining the importance of this discipline for the development of research in the social sciences. Furthermore, some of the sub-topics treated were associated with specific countries, often treated as single-country cases—thus, their presence as keywords—whilst other countries were associated with social policy in broader ways and were not mentioned in keywords. The articles analysed in the present work revealed how some nations paid particular attention to certain topics; for example, Germany, New Zealand, Sweden, and the UK were all present in the sub-cluster employment, indicating that the articles focusing on this topic used evidence form these countries. This may mean that reforms were occurring in those countries related to social policy, or that the listed countries were treated as comparative case studies.

The analysis of the author keywords demonstrates how the topics have changed over time. One example is the link between social policy and the keyword COVID-19. This might be an indicator of researchers’ propensity to investigate concepts related to recent social phenomena or it might reflect the introduction of new concepts into social policy studies, showing that the discipline does not purely focus on the “traditional” concepts treated in the past, such as welfare reform, class, and ethics. Since the journals considered here were not specific to the area of social policy, it could be interesting to conduct the same analysis again but with a focus on social policy review and to compare the output with the findings reported here.

As far as the co-citation network of journals is concerned, the leading journals were Journal of European Social Policy, Social Policy & Administration, and Journal of Social Policy, in line with previous research (Powell 2006, 2016, 2018). However, since this analysis did not limit the analysis to journals specifically related to social policy, other leading journals were revealed, including journals related to sociology, such as Social Science & Medicine, and political studies, namely American Political Science Review. The clusters identified a number of precise characteristics (Figure 4): one included journals specifically concerned with the field of social policy (green cluster), another included journals on social medicine and psychology (red cluster), and one included journals on social and political policy related to specific geographic areas, such as Europe and America. One cluster represented journals in the field of sociology, once again, concerning either America or Europe. The final cluster was more heterogeneous, including journals dealing with fields of economics and methodology. The different clusters indicate how the concept of social policy also attracts journals that do not specifically deal with the discipline of social policy and how topics in the sociological, economic, and political areas could be related through the dimension of social policy. The connections between journals belonging to different clusters indicates that articles dealing with topics on social policy are not isolated.

Esping-Andersen G. is the social policy researcher who had the greatest impact on subsequent research activities in the field. Pierson P. and Korpi W. were the next two most cited authors, as well as the authors of some of the most cited publications. This study confirms Esping-Andersen G. to be the most cited author—as identified by Powell (2018)—but the findings from this analysis also add that Esping-Andersen G. is the author most cited by researchers from other disciplines, and not only by social policy researchers. The other most cited social policy researchers revealed in this study differed from those reported by Powell (2018), reflecting the differences between the two studies.

6. Conclusions

The objective of this study was to provide a detailed description of the scientific publications present in the Web of Science database addressing the topic of social policy. Previous bibliometric studies on social policy focused on social policy journals only or on specific dimensions of social policy, such as the welfare state, leaving a gap in the literature, which the present study has now filled.

Although the data obtained using the bibliometric approach applied are mainly descriptive and must be interpreted with caution due to the limitations inherent to bibliometric analysis, they provide a panorama of the main authors, journals, and articles addressing the topic of social policy in the literature, which is of great practical value for social policy scientists.

The results demonstrate a positive trend in the output rate of publications containing the term social policy (or social policies) in the title, abstract, or author keywords up until 2021, when the number of articles began to decline. However, the recent drop in output rate should not be interpreted as a decline in the interest in social policy issues since the period coincides with the entrance of other new concepts, described by other keywords (e.g., migration and social investment), which gives clear evidence of the relationship between social policy studies and the actual phenomena. Thus, young researchers approaching the discipline of social policy should consider these new topics to help them navigate the field and identify the lines of research that are actual and fertile and with the potential to kick-start their career.

The results provide social policy researchers and researchers from other related fields, such as sociology and public policy, with some suggestions relating to the best publication channels to use in terms of journals and publishers. This is important considering the increasing pressure placed on academics and universities to publish, especially in journals with a high impact factor. By indicating which journals have the greatest impact in the field of social policy, this work can assist researchers in maximising the visibility of their research through a sage choice in journal. Furthermore, the sub-topics revealed in this work may help researchers identify new, fruitful research directions to embark upon in future analyses.

The data presented here can also be used by scholars to identify the most important authors publishing in the area of social policy, in terms of co-citated authors and co-citation of cited references. Through being informed about these authors and their work (the contexts, findings, and research approaches used), scholars can draw relevant stimulus to develop their work in the best way.

Future research should also compare the results generated here with those from similar analyses focused on social policy journals only.

This study presents some limitations. One relates to how exhaustive the dataset produced was, since studies were sourced from the WoS database only. To deal this problem, further research could consider other databases such as Scopus or Google Scholar or use more than one database and merge the data together. Even in this latter case, caution in interpretating the data would be necessary since the database merging strategy is not always the preferable one (Donthu et al. 2021). A second limitation is that the analysis only included papers written in the English language, associated with a DOI code, and supplied with author keywords. This selection criteria led to the elimination of 3229 studies from the sample. The inclusion of these articles would have increased the volume of data and may have produced different results, although this would have complicated the data cleaning process. Moreover, from a methodological perspective, it is essential to take the journals into account that have undergone a name change over time and address this aspect in the analysis.

The analysis focused on the term social policy (or social policies) contained within the article title, abstract, and author keywords but did not consider the text of the paper or the “KeyWords Plus”—the index terms automatically generated from the article titles within the database. Further research should consider these limitations and endeavour to include the text of the article and the KeyWords Plus in the analysis. Thus, from the methodological perspective, future analyses could be developed using different bibliometric software such as CitNetExplorer version 1.0.0 (van Eck and Waltman 2014a), which would also allow the researcher to compare the results achieved through different tools. Different kinds of publications, not just peer reviewed articles, could also be considered to obtain a more comprehensive analysis of the literature. Finally, the fact that bibliometrics promote the “publish-or-perish culture”, where citation counts and egos are prioritised over quality (Breznau 2021), should be taken into consideration. That said, alternative ways of getting research published are also being developed, for example, thanks to the Open Science Movement which strives to make scientific research (including, for example, access to data, detailed methods, and peer reviews reports) accessible to all levels of society, focusing more on the qualitative aspect of the publications and not their quantitative characteristics.