What Is the Benefit of Publishing in a Peer-reviewed Scientific Journal?

Abstract

Many papers in economic science that are published in peer reviewed journals are initially released in widely circulated working paper serial. This raises the question about the benefit of publishing in a peer-reviewed periodical in terms of citations. Specifically, we address the question: to what extent does the stamp of approval obtained by publishing in a peer-reviewed journal lead to more subsequent citations for papers that are already available in working newspaper series? Our data ready comprises about 28,000 working papers from four major working paper series in economics. Using console information methods, we show that the publication in a peer reviewed journal results in around twice the number of yearly citations relative to working papers that never become published in a journal. Our results hold in several robustness checks.

Introduction

Ane of the most important metrics to assess the importance of a inquiry article, its impact or value is its citation count. The citation count across articles is in turn used in rankings of bookish journals, authors, departments, universities or as an input for tenure and promotion decisions (e.one thousand. Segalla (2008) or Seiler and Wohlrabe (2012)). Due to this, the expected citation count is 1 of the factors that determines researchers' decisions of how to disseminate their enquiry, be it as a published commodity or non. ^{Footnote 1}

Traditionally, publishing an article in a peer-reviewed journal has been an effective way to get one's work cited. For example, it increases the outreach of the article, makes it available online (for a fee or free equally open access), and it provides a postage stamp of blessing past peers. However, many of these benefits can readily be obtained by omitting the peer review procedure and publishing a paper in a working newspaper series or as a preprint. Indeed, in that location are many working papers that received a substantial number of citations even if they never successfully completed the peer review process to become journal articles.

In addition, the literature has identified several limitations of the peer review process that exercise non apply to working papers. Bornmann (2011) provides an excellent clarification. For example, Laband and Piette (1994), Hodgson and Rothman (1999) and Ductor and Visser (2020) propose that the editor and his or her relationship to the authors can have a big impact on the chances of a paper getting published in a high tier periodical. In addition, working papers are available without delay while the time from submission to publication in peer-reviewed journals can take up to multiple years and has increased over time (e.chiliad. Ellison (2002)). The well-nigh immediate availability of working papers allows a timelier discourse of the findings and potentially more citations. ^{Footnote 2}

While there are some shortcomings in the peer review process that is required to publish in a journal, this process too has some key advantages that can impact the number of citations a paper gets. As there is no formal peer review earlier preprints or working papers go available online, they can potentially include errors, central omissions, flaws or caveats. The peer review process thus acts every bit a quality control of enquiry papers and can foster trust in the science organisation. Additional potential benefits include reaching a new and potentially wider audience and an improved readability due to a more than standardized formatting.

Because there are both factors which suggest that working papers should become more citations (e.g. timeliness) and factors which propose that periodical articles should get more citations (e.thousand. potential flaws in working papers), it is not obvious whether publishing in a periodical increases the commendation count. If publishing an commodity in a peer-reviewed journal has footling or no impact on the citations a newspaper receives, this would put into question some of the motivations for publishing an commodity as well as the importance of working papers relative to published manufactures.

In gild to measure the affect of publishing in a journal, we first denote the drove of benefits that could lead to additional citations the stamp of blessing issue from publishing in a peer-reviewed journal. This proper name is chosen, equally at to the lowest degree the editor and the peer reviewers need to give their blessing for publication. We then choose to measure this benefit in terms of citations by comparing working newspaper articles that have been published either in a peer reviewed journal or not. Our information fix comprises about 28,000 papers published in four major economic science working paper series. The bibliometric information builds upon the RePEc website (Research Papers in Economic science, www.repec.org) and the citation information were retrieved from CitEc, which is closely related to RePEc. ^{Footnote 3} We estimate the postage of approval effect on the yearly citation count.

We focus on the potential increase in citations due to publishing in a journal commodity but there are additional benefits like prestige or tenure and promotion which tin be important reasons to go through the publishing process every bit well. However, as citations tend to play a office for these benefits besides, our findings can have some relevance for those benefits as well. Specifically, our results suggest that the stamp of approval issue leads to a doubling in the number of yearly citations for journal manufactures and hence it makes sense to aspect a college value to journal publications than working papers.

This paper also complements the literature on the value of preprints and working papers. Specifically, information technology has been shown extensively that journal articles which are available as preprints and working papers (and open up access) besides are cited more than often (e.one thousand. Sarabipour et al. (2019), Fraser et al. (2020), Fu and Hughey (2019), or Wohlrabe and Bürgi (forthcoming)). Nosotros show here that the reverse is true also and hence the stamp of blessing is of import as well.

The remainder of the newspaper is structured as follows: the next section describes the data used followed by the panel regressions. The following section offers robustness checks, and the terminal section concludes.

Information

Nosotros build upon the data set of Baumann and Wohlrabe (2020a) who also provide more background data. The four working paper series considered stand out with respect to prestige and influence in the field of economics. They belong to the about cited and downloaded working paper series on RePEc. ^{Footnote 4} The four series are published by networks of economists. Submitting papers to the series is only allowed to members of the corresponding networks and joining the network is simply possible past invitation. One time an author is a member of the specific network he or she is gratuitous to submit any working paper. With this procedure, the networks want to clinch a specific level of quality of the submitted papers equally invitations are simply issued to established or promising researchers.

The original information prepare comprises 28,877 working papers from between 2000 and 2012. For our analysis we exclude those working papers that accept been published as a affiliate in a book (1120). This leaves united states of america with 27,757 papers. The majority of papers has been published in the NBER working paper series (10,364), closely followed by CEPR (6699) and IZA (6904). The least number of papers were issued in the CESifo series (3790). In Tabular array 1, we testify how many papers of each series have been published in a refereed journal. In full and for each series this share is approximately 50%. These numbers differ a little flake from the estimates provided by Baumann and Wohlrabe (2020a) who report a share of about 66% based on a random sample. In the robustness section nosotros accost this issue in more than item.

Our citation data comes from CitEc (http://citec.repec.org/). This website provides the citation data for the RePEc network. ^{Footnote 5} Each working paper and journal article has a unique identifier. We used this identifier to extract citations on a yearly basis for working papers and the periodical articles separately. Still, equally the working newspaper citations incorporate the periodical citations and vice versa, it is not possible to obtain accurate citation counts for the 2 versions separately. Due to this brake, nosotros apply the maximum citations per year from either version every bit the citation count for our dependent variable. ^{Footnote 6} Every bit Baumann and Wohlrabe (2020a) have shown, many papers accept been published in several working paper series simultaneously. CitEc also consolidates citations across versions, i.e. information technology assigns citations received by one working paper version also to the other versions. Due to time delays in the consolidation procedure at CitEc, the commendation numbers are non ever identical across serial. We therefore take always the maximum citation count across working papers.

In Table 1 we provide the descriptive statistics for the citation information. Besides the full sample, nosotros as well distinguish whether an article has been published in a periodical or not. The average citation count across all working papers is 43. This number is higher for papers in the NBER serial (61) and smaller for the IZA (25) and CESifo series (21). The citation distribution is quite dispersed as the standard deviation exceeds the hateful considerably. This is too confirmed in the left graph of Fig. i. The most cited paper published in a journal is the 1 past Melitz (2003) which appeared both in the NBER and the CEPR working paper series. The commodity past Pesaran (2004) is the almost cited newspaper (904) that never appeared in a periodical. Information technology was published both in the IZA and the CESifo working paper series. When comparing manufactures in our two subgroups, we run into a articulate difference betwixt the citation counts. Articles published in a journal received 70 citations on average whereas the average citation counts for unpublished papers merely amounts to xv. In the lower panel of Table 1, we report the p-value of a ii-sided non-parametric Wilcoxon-Isle of man-Whitney examination. The cypher hypothesis of an equal median of citations is conspicuously rejected both for the full sample and the iv series. ^{Footnote 7} The results reported in Table 1 are supported by the kernel estimates of the citation distribution depicted in the left graph of Fig. 1. ^{Footnote 8} They show that about of the papers with 10 or fewer citations accept non been published in a journal. Additionally, periodical manufactures in our sample exhibit more mass compared to the control group across almost the entire citation distribution.

Table 1 Descriptive statistics for citation count

Full size tabular array

Fig. i

Kernel estimates for citations distributions

Full size paradigm

The comparisons and then far might be biased as the publication engagement of working papers ranges from 2000 to 2012. So, more contempo papers had less fourth dimension to accumulate citations. Therefore, we repeat the assay using a constant commendation window of nine years starting in the publication year of the working paper. As Table 1 and Fig. one show that the results stay qualitatively the same, papers that are eventually published in a periodical gather more citations. Every bit expected, the citation counts are smaller due to the smaller commendation window.

These simple hateful comparisons leave out many of the factors which might drive citation counts of papers and practise not necessarily reverberate a precise estimate. See Bornmann and Daniel (2008) and Tahamtan and Bornmann (2019) for a literature overview. Nosotros address these problems in the adjacent section.

Methodology and main results

Methodology

In social club to test the hypothesis that a periodical publication leads to more citations, we first create a dummy variable that takes a value of 0 if the paper was only available as a working paper in a specific twelvemonth and 1 otherwise (i.e. if information technology was published in a periodical). Considering this dummy takes value 1 in all years for papers that become journal manufactures in the start year, the stamp of approval effect cannot be separated from paper fixed furnishings and we exclude them from our analysis. This holds for 1878 manufactures. Baumann and Wohlrabe (2020b) report more on the differences between the publication dates of working papers and the corresponding journal articles.

Due to the panel nature of our data gear up, nosotros omit the controls used in the literature every bit they are either captured by the paper stock-still effects (e.g. the number of authors, the working paper series or the number of working papers) or captured by time stock-still effects (due east.thou. age of the papers). ^{Footnote 9} In improver, the panel construction with stock-still effects likewise controls for many unobservable characteristics specific to a paper that exercise not alter over time and can influence the outcomes. For example, only the better papers might somewhen exist published in a periodical and this quality difference is captured past newspaper fixed effects (Iron). ^{Footnote x} In a starting time step nosotros run a pooled regression and a random effects (RE) specification. The likelihood ratio test conspicuously rejects the pooled model in favor of the RE model (p-value = 0.000). Similarly, the Hausman examination rejects the RE model in favor of the FE model (p-value = 0.000). We then gauge the following equation as our main specification:

$$\brainstorm{aligned} citpy_{information technology}=\beta \mathrm{periodical}_\mathrm{it}+\nu _i +\mu _t+\varepsilon _\mathrm{it} \end{aligned}$$

(one)

where citpy are the citations per year for each article and journal is a dummy that takes value one for the years in which the journal article is bachelor. \(\nu _i\) are the newspaper fixed effects and \(\mu _t\) the fourth dimension fixed effects. These include also dummy variables representing each yr in our data sets. This captures general citations trends. Negative binomial panel regressions with fixed furnishings would exist natural pick for count data. However, these are non feasible (e.g. meet Wooldridge (1999)). Due to this limitation, we use Poisson regressions with both time and paper fixed effects instead. We run the regression both beyond all papers and for each working newspaper series individually and cluster the standard errors at the paper level. Nosotros report incident charge per unit ratios (the exponential of the coefficients) and the effect is multiplicative. This means for example that an incident rate ratio of 2 implies a doubling of the number of yearly citations and ane means no change in citations.

Main results

Tabular array 2 reports the estimates both for the full sample and for each working paper series separately. The coefficient of the journal dummy is large and highly meaning beyond all regressions. The estimates imply that the postage of approving effect of publishing an article in a peer-reviewed journal results in more than a doubling of the yearly citation count relative to only making the article available as a working paper. This effect is non only statistically significant but also substantial in its size. Looking at the working newspaper series, the journal effect is larger for the CESifo and IZA working papers. For these, the publication in a periodical leads to an increase in citations past a factor of around ii.5 while it is a bear upon smaller for NBER and CEPR working papers with an increase in citations by a factor of ii.three.

Table ii Publication impact using panel regression

Full size table

Initial robustness checks

A working paper might appear in multiple working newspaper series. This might cause the paper to attain a wider audience before condign a journal article and hence the benefit from publishing it becomes smaller. To test this, we restrict our sample to papers that have been released just in one working paper series. The event in column one of Table 3 show that papers that are just in one working paper series indeed benefit more from being published. We can look at this argument also from a unlike perspective. Existence released in only i working paper series and then either published in a journal or not might exist an unfair comparing. The commodity finally published in a journal has an additional outlet and therefore a college visibility. In order to account for that issue, nosotros restrict our sample to manufactures published in a journal with but one working newspaper and compare them to articles bachelor in two working paper series but non in a journal. Column 2 of Tabular array iii shows that the result becomes a bear on smaller only remains highly significant. In column 3 we report the outcome of a comparing between articles that appeared in two working paper serial and in a periodical with those that appeared in three working paper serial just not in a periodical. Once more, publishing in a periodical leads to a significantly college citation count.

An additional concern is the issue of the time lag betwixt working paper and journal publication. Information technology could be the case that papers that take longer to get published in a periodical have a large change in quality. For case, they might go through more than review rounds and have more substantial revisions during the submission process. As a upshot, they might receive a larger benefit from publishing. Every bit already noted, papers with identical publication years in both outlets are excluded by construction hence we took simply those papers where the fourth dimension lag is one yr. This limits the revisions possible and hence the potential quality alter of the paper. In the fourth column of Table 3 we show that the journal effect is larger for the papers which only had limited time for quality changes. This suggests that the (unobserved) quality change is non the master commuter of our results.

Tabular array 3 Initial robustness checks

Total size table

Fig. 2

Evolution for cumulative citation count for a matched sample. This effigy shows the cumulative citations relative to the first year of the paper on the vertical axis for 151 matched papers that have at least 20 cumulative citations in the first 3 years. The bluish line are papers that have been published in a periodical in the third year and the red line are the matched papers that were not published. The years are on the horizontal axis

Full size image

Our next robustness check concerns a potential endogeneity issue. Specifically, one could imagine that only better papers become published and worse papers remain working papers. An editor could see that a working paper has already accumulated a substantial number of citations and might accept a positive bias towards these papers. This could cause our coefficients to capture the paper quality rather than the stamp of approving consequence. In society to test this, nosotros reduce our sample to papers that cumulatively received at to the lowest degree xx citations in the kickoff iii years and in improver only keep published papers that were published in the tertiary year of becoming bachelor as working paper. These papers are then matched to like working papers that take never appeared in a journal. Therefore, they have been issued in the same year and in the same working paper series. Furthermore, the cumulative citation ratio between two papers must have been betwixt 0.ix and 1.one. Based on these conditions we were able to match 151 published papers. For each of the commencement seven years, we calculate the cumulative citations relative to the first year and boilerplate this number across the two groups. Effigy 2 shows the resulting line for the published articles in blue and the line for the working papers in red plus the 90% conviction intervals. While both the published and unpublished papers have similar citations in the start iii years, the published articles receive more citations in one case they get published. This suggests that in that location is indeed a stamp of approval effect, as the papers in both groups should have a similar quality based on the citations received in the first three years, simply published papers receive more citations in subsequently years.

Additional robustness checks

Bookkeeping for not-tracked articles

So far, nosotros causeless that we were able to find all published articles respective to working papers. However, Baumann and Wohlrabe (2020a) showed that the actual share of working papers in our sample which are finally published in a periodical is nearly 66% using a random sub-sample (instead of the 50% nosotros constitute). In this department, we desire to make sure that our results are non driven by our potential inability to track all published papers. In social club to test this, nosotros repeat the regression in Table 2 just care for a random 30%, twenty%, x%, v% and 1% of the published papers equally if they had zero citations in every year. Specifically, we run a simulation with yard repetitions for each threshold where we randomly assign aught citations to the specified share of published manufactures. This simulation is equivalent to adding a like percentage of the least cited unpublished working papers randomly to the published journals. Replacing 10% of the citations with zeros is equivalent to not having plant effectually 1400 periodical publications. Replacing the citations with zeros are likewise a clear worst-case scenario, every bit it is likely that at to the lowest degree some of the published articles we could non discover accept more than than zippo citations.

Tabular array 4 Robustness: accounting for not-tracked articles using simulations

Total size table

As Table four shows, all simulations have a positive and significant coefficient and the coefficient becomes smaller the more periodical articles are inverse to have 0 citations. The mean coefficient of the 1000 simulations is also reasonably close to the highest and lowest coefficient in the simulation, fifty-fifty if the difference is several standard errors. With 30% missed articles, the stamp of approval result is smaller but even so corresponds to a 75% increase in citations for articles that are published in a journal relative to the ones that remain working papers.

Skewness of the citation distribution

It is well-known that the citation distribution is skewed (due east.g. encounter Seiler and Wohlrabe (2014)). Our sample includes several papers that take very large numbers of citations which could potentially drive our results. Also, there is a number of papers that have never been cited, which could also impact our results. In guild to check whether the papers with many citations bulldoze our results, nosotros repeat the regressions in Table 2 column 1 but exclude papers that have at least 1000, 500 and 100 citations from our sample. The results are presented in the first three columns of Table 5.

Tabular array v Robustness: bookkeeping for skewed citation distribution

Full size table

Compared with the results where we did non exclude these papers, the impact of publishing a paper are broadly unchanged or at that place is a larger postage of approval event. This result is as well robust to omitting the papers with zero citations as shown in column (4) and to omitting papers with both 0 citations in addition to the ones with at to the lowest degree 100 citations shown in the last column in Tabular array 5.

Further robustness checks

Finally, nosotros consider two boosted issues. Outset, our data has more observations for older papers than for newer ones as our sample stops in 2020 for all papers. This could potentially influence our results every bit this causes older papers to have a higher weight in the panel. In order to appraise the bear upon of this, we repeat the regressions but only include the first nine years for each paper (a 9-year citation window). The results for this regression are shown in the first cavalcade of Tabular array 6. While the coefficient declines to a 90% increase, it is notwithstanding highly pregnant.

Tabular array 6 Farther robustness checks

Full size tabular array

Second and equally mentioned in the methodology department, we take multiple sources for citations. CitEc reports citations for the working paper and for the periodical commodity separately, even if both commendation numbers comprise citations for either. So far, we used the maximum of the two numbers as the periodical citations in the analysis. In column 2 of Table 6 we report the regression result for the smaller of the two numbers. The coefficient and the significance are about unchanged when compared to the ones obtained in Table 2.

Conclusion

This paper showed that the yearly citation count of working papers which are subsequently published in a periodical more than doubles relative to unpublished working papers. This increase in citations is substantial, controlling for a number of observable and unobservable variables that are fourth dimension or paper specific in a panel regression. In that location is thus a substantial postage of approval effect when publishing in journals.

While we were able to identify what nosotros call the postage stamp of approving outcome, nosotros were not able to identify what component of publishing an article causes this issue. In that location are many possible channels, which could crusade this increase in citations. For example, the peer review process that approves and improves the paper or the additional outreach of a journal could both be potential causes that increase the citation count and we constitute evidence for both of these channels. This paper also focused on citations only. There are other potential benefits beyond citations to both the researcher and the field as a whole, which we did not address hither.

Notes

1. Run across Spiewanowski and Talavera (2021) for a recent study of publication behaviour for UK-based economists.

2. Brown and Zimmermann (2017) provide a detailed discussion on this issue. Sarabipour et al. (2019) outline the value of preprints for early phase researchers.

3. An comprehensive overview provides Zimmermann (2013). RePEc data has been used in bibliometric analysis by Rath and Wohlrabe (2016), García-Suaza et al. (2020) or Wohlrabe and Gralka (2020), among others.

5. Nosotros thank Jose Manuel Barrueco for help with the citation data. The citations were retrieved in February 2020.

6. Instead of the maximum, nosotros also ran the estimations the minimum and our results remained qualitatively the same equally shown in Table 6.

7. We obtain the aforementioned results in case of two-sided t-test.

8. In social club to increase readability, we capped the citations at 200.

9. This setup also controls for more than sophisticated age structures as in Anauati et al. (2016) and is broadly in line with the structure in Fraser et al. (2020) or Fu and Hughey (2019).

10. Obviously, unobservable characteristics that change over time are not captured by the paper and time fixed effect like for example a quality change betwixt the working paper and the published paper. We address this issue in "Initial robustness checks" department.

References

• Anauati, V., Galiani, South., & Gálvez, R. H. (2016). Quantifying the life bicycle of scholarly manufactures beyond fields of economic research. Economical Inquiry, 54(2), 1339–1355.

  Article  Google Scholar

• Baumann, A., & Wohlrabe, G. (2020a). Where have all the working papers gone? Evidence from four major economics working paper series. Scientometrics, 124, 2433–2441.

  Commodity  Google Scholar

• Baumann, A., & Wohlrabe, K. (2020b). Where take all the working papers gone? Evidence from iv major economics working paper serial. CESifo Working Newspaper 8328.

• Bornmann, L. (2011). Scientific peer review. Annual review of information scientific discipline and engineering science, 45(ane), 197–245.

  Article  Google Scholar

• Bornmann, L., & Daniel, H.-D. (2008). What exercise citation counts measure? A review of studies on citing behavior. Journal of Documentation, 64(i), 45–80.

  Article  Google Scholar

• Brown, A. J., & Zimmermann, K. F. (2017). Three decades of publishing enquiry in population economics. Journal of Population Economics, 30(1), xi–27.

  Commodity  Google Scholar

• Ductor, Fifty., & Visser, B. (2020). Why are connections to editorial board members of economics journals valuable? Technical report, Department of Economic Theory and Economic History of the University of Granada.

• Ellison, G. (2002). The slowdown of the economics publishing process. Journal of Political Economy, 110(5), 947–993.

  Article  Google Scholar

• Fraser, North., Momeni, F., Mayr, P., & Peters, I. (2020). The effect of biorxiv preprints on citations and altmetrics. Quantative Scientific discipline Studies, 1(two), 618–638.

  Google Scholar

• Fu, D. Y., & Hughey, J. J. (2019). Releasing a preprint is associated with more attention and citations for the peer-reviewed article. Elife. https://doi.org/ten.7554/eLife.52646.

  Article  Google Scholar

• García-Suaza, A., Otero, J., & Winkelmann, R. (2020). Predicting early career productivity of PhD economists: Does counselor-match affair? Scientometrics, 122(1), 429–449.

  Article  Google Scholar

• Hodgson, Thou. Thou., & Rothman, H. (1999). The editors and authors of economic science journals: A example of institutional oligopoly? Economic Periodical, 109(453), 165–186.

  Article  Google Scholar

• Laband, D. N., & Piette, Yard. J. (1994). Favoritism versus search for practiced papers: Empirical testify regarding the behavior of periodical editors. Journal of Political Economy, 102(1), 194–203.

  Article  Google Scholar

• Melitz, M. J. (2003). The touch of merchandise on intra-industry reallocations and aggregate industry productivity. Econometrica, 71(6), 1695–1725.

  MathSciNet  Commodity  Google Scholar

• Pesaran, H. (2004). General diagnostic tests for cross-sectional dependence in panels. CESifo Working Newspaper 1229.

• Rath, Thousand., & Wohlrabe, K. (2016). Recent trends in co-authorship in economics: Evidence from RePEc. Applied Economic science Messages, 23(12), 897–902.

  Commodity  Google Scholar

• Sarabipour, Due south., Debat, H. J., Emmott, E., Burgess, S. J., Schwessinger, B., & Hensel, Z. (2019). On the value of preprints: An early career researcher perspective. PLoS Biology, 17(2), e3000151.

  Article  Google Scholar

• Segalla, M. (2008). Publishing in the correct place or publishing the right matter: Journal targeting and citations' strategies for promotion and tenure committees. European Journal of International Management, two(2), 122–127.

  Google Scholar

• Seiler, C., & Wohlrabe, Chiliad. (2012). Ranking economists on the footing of many indicators: An alternative approach using RePEc data. Periodical of Informetrics, six(three), 389–402.

  Commodity  Google Scholar

• Seiler, C., & Wohlrabe, K. (2014). How robust are journal rankings based on the affect factor? Prove from the economic sciences. Journal of Informetrics, eight(four), 904–911.

  Article  Google Scholar

• Spiewanowski, P., & Talavera, O. (2021). Journal rankings and publication strategy. Scientometrics. https://doi.org/ten.1007/s11192-021-03891-5.

  Article  Google Scholar

• Tahamtan, I., & Bornmann, L. (2019). What do commendation counts measure? An updated review of studies on citations in scientific documents published betwixt 2006 and 2018. Scientometrics, 121(three), 1635–1684.

  Article  Google Scholar

• Wohlrabe, M., & Bürgi, C. (forthcoming). Do working papers increase journal citations? Evidence from the pinnacle five journals in economic science. Applied Economic science Messages.

• Wohlrabe, Thousand., & Gralka, South. (2020). Using archetypoid assay to allocate institutions and faculties of economics. Scientometrics, 123, 159–179.

  Article  Google Scholar

• Wooldridge, J. G. (1999). Distribution-gratuitous estimation of some nonlinear console data models. Journal of Econometrics, 90(i), 77–97.

  MathSciNet  Article  Google Scholar

• Zimmermann, C. (2013). Bookish rankings with ReREc. Econometrics, 1(3), 249–280.

  Google Scholar

Download references

Funding

Open Access funding enabled and organized by Projekt Deal.

Author data

Affiliations

1. ifo Insitute for Economic Reserarch, Poschingerstr. 5, 81679, Munich, Federal republic of germany

   Klaus Wohlrabe

2. St Mary'south College of Maryland, 47645 College Drive, St. Mary'south City, Medico, 20686, United states of america

   Constantin Bürgi

Respective author

Correspondence to Klaus Wohlrabe.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution four.0 International License, which permits use, sharing, accommodation, distribution and reproduction in whatsoever medium or format, as long every bit you lot give appropriate credit to the original author(southward) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party cloth in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the cloth. If textile is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted utilise, you will need to obtain permission straight from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/iv.0/.

Reprints and Permissions

About this article

Cite this commodity

Wohlrabe, Thousand., Bürgi, C. What is the do good from publishing a working paper in a journal in terms of citations? Evidence from economics. Scientometrics 126, 4701–4714 (2021). https://doi.org/10.1007/s11192-021-03942-x

Download citation

• Received: fifteen June 2020

• Accustomed: 04 March 2021

• Published: 26 March 2021

• Issue Appointment: June 2021

• DOI : https://doi.org/10.1007/s11192-021-03942-ten

Keywords

• Peer review
• Citations
• RePEc
• Working newspaper
• Preprint

JEL Classification

• A10
• A12

riveraentreirs.blogspot.com

Source: https://link.springer.com/article/10.1007/s11192-021-03942-x

Share this post