Protected Area Effectiveness in the Scientific Literature: A Decade-Long Bibliometric Analysis

Round 1

Reviewer 1 Report

The authors described the methodology in detail. And the results based on the methodology are good. What I suggest is that there were limitations to the methodology. While the results of the analysis focused on those items the academics concerned, such as citation, journal …. It didn’t reveal those concerns by the study of PA MEE itself. I would suggest that it would mention the limitations. For example, there is a demand to know the popularity, efficiency, and effectiveness of different methods and methodologies. The results were not able to reveal much about it.  About the methodology, SLR, I would suggest that it would add its readability while adding some more literature on other cases by SLR.

Author Response

We would like to acknowledge and thank the three anonymous reviewers whose comments helped us improve the quality of our study. Please find below our responses to their remarks in the same order as mentioned.

Reviewer 1

 

• The reviewer is right that the two stated ‘thematic’ (i.e. PAME-related) research questions were not responded fully in the text. Actually, they were not the focus of this article, which is why just a small number of thematic variables (n=3) was kept in this study for greater meaning and richer bibliometric context. Thus, we deleted both confusing research questions and highlighted that the focus of the article was on bibliometric (not thematic) analysis, using those few key thematic variables just for context and more meaningful explanation of results.

 

• Regarding the remark on adding some more literature on SLRs, we think that we had included a large enough amount of studies on SLRs from different research fields, including PAs assessment, both in the Introduction and also in the Discussion. Our intention in this study was not to undertake a review of SLRs but just to provide an adequate context for the study, which we think we did in the submitted version without extending the article’s length overly. E.g.: “SLRs began in the mid-1990s as a novel way of conducting literature reviews. Their initial development was linked to the Evidence-based Medicine Working Group [2,7]. In the last decades, several SLRs have been addressed, especially focused on the medical field: epilepsy [8], chronic fatigue syndrome [9], Parkinson's disease treatment [10], sleep apnea [11], dementia [12], or infections [13]. More recently, there have been some interesting contributions in other disciplines as a result of the growing number of scientific publications. These relate to the social sciences [14], Information Systems [5], agriculture [15], climate change [16], forest fires [17,18], soil erosion [19], marine litter in transitional waters [20] or tropical marine science [21]. Bibliometric analyses are used for a variety of purposes, such as finding trends in journals, author or article performance, exploring collaboration patterns, or eliciting research interest on certain topics [22]. Other authors point out that bibliometric analyses are useful to understand patterns of research on a given topic, identify research gaps, and understand networks of collaborators on that topic [23,24]. Parallel to SLRs and bibliometric analyses, Protected Area (PA) effectiveness emerged as a research field in the early 2000s with the landmark publication “Evaluating Effectiveness. A framework for assessing the management of protected areas” [25]. Since those days, mostly qualitative and subjective PA assessment systems [26,27] have paved the way to increasingly objective, accurate and complex semi-experimental research designs [28-30]. Studies applying SLR on PAs are scarce. We only found few precedents. Caveen et al [31] conducted a SLR to determine the ecological effects of Marine Protected Areas (MPAs). Another study conducted a SLR that selected a sample of 407 national park studies to investigate the sociodemographic and psychographic characteristics of visitors, their behavioral patterns, and their integration into management activities [32]. On the other hand, bibliometric studies are also increasing in more specific research topics such as terrestrial PAs [33,34] and MPAs [35]. Despite this progress, bibliometric analyses from SLRs on the global effectiveness of PAs on land and at sea are still missing.”

 

Reviewer 2 Report

The authors analysed articles related to the ecological effectiveness of PAs. The topic fits well into the profile of the journal. The methodology used is correct and is suitable for answering the two research questions. The authors chose the 2010-2019 time period. A more detailed explanation for choosing this time interval is recommended. The authors took in consideration only the affiliation of the first authors. A more detailed explanation regarding this choice is also recommended. The first author is not always the main author. Taking in consideration only the first one, in some cases can be misleading.

There are 13 bibliometric and 3 thematic variables analysed. By the later, the authors enlist “the Human Development Index – HDI – of the country studied” (line 93-94) as one of the variables taken in consideration. Later, however, they use “the HDI of the country of first author’s” and there is no mentioning about the HDI of the analysed countries. It is recommended to clarify the used variable: the analysed country is not always the same as the country of the first author.

The conclusions of this research are correct and can help mapping the scientific behaviour of researchers working on the given topic. It was very interesting to read about the somewhat ambiguous relationship between the OA type of publications and the number of citations received.

At the end of the article, the authors mention the possibility of analysing keywords with softwares such as Vosviewer. It would have been interesting to see what keywords were the most common ones in the used database. This would have given a deeper understanding of the topics analysed.

The article is recommended for publication with minor revision.

Author Response

We would like to acknowledge and thank the three anonymous reviewers whose comments helped us improve the quality of our study. Please find below our responses to their remarks in the same order as mentioned.

Reviewer 2

 

• The reviewer is right that we did not specify why we had chosen the SLR time range. Thus, we added the following clarifying paragraph to section 2.1: “We selected that time range because relevant international decisions on PA effectiveness spanning until 2020 that are likely to have had a reflection in the scientific literature were made at the beginning of that decade (CBD, 2010) as well as on literature manageability grounds.”
• The reviewer also correctly points out that the first author is not always the main author of a study. Certainly, there is a variety of author ordering ‘styles’ in different research fields. Nevertheless, in biology, geography and environmental sciences, the fields most directly related to PA effectiveness studies, the first authors are usually the main contributors to the studies, with minor contributors taking other places and more senior ‘supervisors’ usually taking either the second place or the last place. Thus, we consider the first authors’ affiliations a good proxy for this study. Moreover, some studies included several authors. Therefore, analyzing the large amount of authors’ affiliations would have required very substantial time and effort, most likely requiring a specific study.
• As the reviewer says, the second of the three thematic variables analyzed is the “Human Development Index - HDI - of the country studied”, which we mentioned in lines 93-94. It is described in the search protocol (Appendix B, p.2) and its results are expressed in the "Stats" sheet in Appendix C (column B) and in rows 6-8 (Appendix D). Additionally, we analyzed the "Country of first author's institution", the twelfth of the bibliometric variables (see Search protocol, Appendix B, p.5). Its results, expressed as "HDI of the country of first author's", are shown in the "Stats" sheet in Appendix C (column I) and rows 27-29 (Appendix D). As the variable HDI of the country studied is a contextual variable we only mention its results in the 3rd point of Spearman's correlation analysis. On the contrary, we comment in more detail on the bibliometric results of the HDI of the country of first author's in relation to OA publications, the availability of supplementary data in the articles and the standardized number of citations. These issues often give rise to some discussion on this and other scientific topics.
• We agree that some other tools might have helped to clarify some aspects of the research. However, given the diversity of such tools, we just provided a brief overview of additional potential tools to complement the results we got from our study such as VOSviewer, CiteSpace or BRT for future developments in this field.

Reviewer 3 Report

I found the MS and its analysis really interesting and well-written. It can be really helpful and highly-cited in research of protected areas. The patterns oberved are really informative although scarce in bibliography, as a result I would suggest the study to be accepted and revised only with minor comments (see attached).

The only drawback that I found is the statistical significance of the increase in years (Fig 3- see also attached).

Comments for author File: Comments.pdf

Author Response

We would like to acknowledge and thank the three anonymous reviewers whose comments helped us improve the quality of our study. Please find below our responses to their remarks in the same order as mentioned.

Reviewer 3

• In figures 3a to 3d we decided to include a trend line. It is based on a linear regression analysis by the least squares method (Seber & Lee, 2003). We took into account its advantages: simplicity, ease of calculation and ease of understanding of the temporal evolution of the variables represented and their trend. However, it is true that it has some disadvantages: it does not consider other factors that could decrease predictability, some subjectivity. For these reasons, and to avoid methodological discussions that are outside the focus of the article, we decided to eliminate the trend lines in Figure 3. We think that this modification does not affect the readers' understanding of the results. In any case, comparing the group of articles from the first half of the decade (2010-2014) with those from the second half (2015-2019), the data evidence an increase in the average number of authors (+22%), pages (+24%) and references (+11%).

Seber, GAF., Lee, AJ. Linear Regression Analysis. 2003. Wiley.

• The last paragraph of the "methodological remarks" (lines 386-390) might seem to be unrelated to the content of the article. However, we believe that it does have a methodological relationship. As we have explained to reviewer 2, we provide a list of complementary methods and tools to develop this research in the future and to obtain new results (e.g., identification of the main authors/knowledge producers in this scientific topic, their networks, the main keywords, etc.). They are only pointed out as future lines of research.