Editorial 28(4): Reflecting upon some unused IIER statistics
The term 'unused' seems an appropriate descriptor for the two kinds of statistics we report upon in this brief editorial. These are article processing times, and IIER website statistics. Both are important, though for different reasons, and both are to date, for similar reasons, 'unused', or more precisely, 'little used'.
We asked the participants whether they would like to change anything about the academic publishing system and what that might be. Approximately half (51.7%) indicated that they would like to change something about the publishing system and specified the change they wished to see in a subsequent section for comments. ... Of the suggested changes, time to publication, peer review process/quality, and fairness topped the list, which are indicative of the main concerns of authors worldwide (D'Souza, Kulkarni & Cerejo, 2018).Experimenting with ways to systematically monitor IIER's article processing times, we have used the 'DATEDIF' function in MS Excel to obtain some preliminary data. For IIER 28(4), with 16 articles, the average difference between submission and acceptance advice was 15.6 weeks (range 0.5 to 27 weeks), and between submission and publication was 29.3 weeks (range 10 to 43 weeks). The period of 29.3 weeks is longer than the author expectations reported by D'Souza, Kulkarni and Cerejo (2018), 3-6 months (13-26 weeks). However, on the positive side, IIER could be on an improving trend, as the comparable data for IIER 28(3) was 17.2 weeks (range 13 to 23 weeks) and 34.0 weeks (range 19 to 42 weeks), though of course data for two issues is insufficient to establish trends, if any. As time permits, we will extend the calculations to include previous issues, perhaps back to 23(1) in 2013.
Although the three key dates (submission, acceptance and publication) are defined exactly in IIER's records, in many individual cases interpretations have to be qualified by reference to IIER policies and to author-related events. IIER's policy with resubmissions of a rejected article is to record them as new submissions, for two main reasons. Firstly, IIER's submission records serve also as a recording of the editorial staff workload. As the amount of editorial staff time required to process a resubmission of a rejected article is usually about the same as the time required for a new article, all resubmissions are counted as new submissions. Also, editorial staff assessments of submissions tend to concentrate upon identifying problems, and in many cases this identifying is incomplete, because assessments may be ended when a sufficient number of significant problems have been found and noted. This is a time-saving approach, related to the colloquial expression, "Three strikes and you are out!" One consequence is that a resubmission may require more editorial staff time than the initial submission, because if authors have followed the advice given in the initial rejection advice, more time will be needed for more extensive examination of the resubmissions. Related to this are occasional cases where a resubmission has been deemed by editorial staff to have followed the reviewers' advice so well that further external review need not be sought (one such case occurred with a 28(4) article, leading to an exceptionally short period of 0.5 weeks between submission and acceptance).
Secondly, many resubmissions have been revised so extensively that characterising them as a 'new submission' is appropriate. Deeming this category to be more important than resubmissions which include only poor or minor revisions (and therefore are not deserving characterisation as 'new submissions'), we use 'new' as the only category. However, in IIER records we do take care to cross-reference to an earlier submission, if any, as that is important for initial assessment by editorial staff. The records also cross-reference any encouragements given by editorial staff and external reviewers, concerning a resubmission of a rejected article for a new review process. One difficult point is that the dividing line between recommendations to "reject, invite resubmission for a new review process", and "accept, subject to major revisions" is subjective, or 'fuzzy'.
Thirdly, author-related factors may have unpredictable impacts upon article processing times. Very occasionally, we encounter an unknown failure in timely delivery and reading of emails to authors. Wherever possible, we seek to email to all authors of an article, rather than only to the 'corresponding author', reasoning that multiple sending reduces the risk of an unknown failure in timely delivery and reading of advice from IIER. Occasionally, some authors ask for more time to undertake major revisions, such as adding new data and new references.
Finally, on article processing times, the absence of discussion about times for rejected articles does not mean that this topic is unimportant. It is next in the queue for stats procedures updating! We endeavour to provide rapid feedback on articles that editorial staff identify for rejection without referral to external reviewers. About 75% of submissions become 'editorial rejects'; for some more discussion, see Editorial 28(3) and Editorial 28(2). The key factor is minimising 'DATEDIF' between submission and rejection advice, so that authors can resume development of the research work and preparation for submission to another journal, with minimal delay.
The preceding discussion may have some implications for IIER's number of issues per year. We could reduce 'DATEDIF' between acceptance and publication by increasing the number of issues per year, with an appropriate reduction in the number of articles per issue. However, an increase in the number of issues per year would impact upon the scheduling of copy editing, formatting, website updating and alerts list activities, and the workloads required for these tasks.
Another factor that we have to consider as we approach the end of 2018 is the question, "Which will be better for authors and readers: publishing 28(5) in mid to late December 2018, or publishing the same articles in 29(1) in the first week of January 2019?" On current trends, it seems likely that the next issue will be filled with about 16 articles, 'just before Christmas'. However, as the latter is probably the more attractive option for the authors of the accepted articles involved, the next issue will be 'very soon after New Year', though that will add about 1.5 weeks to 'DATEDIF'. Without compromising confidentiality, we can reveal that the authors of one of the articles concerned very recently expressed their preference for a 2019 dating instead of 2018.
IIER's website provider, Netregistry, runs AWStats, "... a free powerful and featureful tool that generates advanced web, streaming, ftp or mail server statistics, graphically" (AWStats, undated). However, it will take some time to develop a routine for perusing the large amount of data that is available. In passing, we could note that three search engine 'bots' are responsible for large amounts of traffic from IIER's website (Figure 1), namely Microsoft's bingbot (Wikipedia, undated), Google's googlebot (Google, undated) and China's baiduspider (Chinese SEO Shifu, 2011).
Figure 1: Screen picture, IIER webmaster view, excerpt from AWStats output for Oct 2018
Roger Atkinson
Clare McBeath
Anne Power (on leave)
Co-editors 2018
AWStats (undated). What is AWStats. http://awstats.sourceforge.net/
Chinese SEO Shifu (2011). What is Baiduspider? Baidu's official web crawler. SEO Shifu Blog, 19 January. https://chineseseoshifu.com/blog/what-is-baiduspider.html
D'Souza, B., Kulkarni, S. & Cerejo, C. (2018). Authors' perspectives on academic publishing: Initial observations from a large-scale global survey. Science Editing, 5(1), 39-43. https://doi.org/10.6087/kcse.116
Google (undated). Googlebot. https://support.google.com/webmasters/answer/182072?hl=en
Heinrich, E., Henderson, M. & Redmond, P. (2018). How international is AJET? Australasian Journal of Educational Technology, 34(4). https://doi.org/10.14742/ajet.4830
Lenette, C. (2016). University students from refugee backgrounds: Why should we care? Higher Education Research & Development, 35(6), 1311-1315. https://doi.org/10.1080/07294360.2016.1190524 [note the data: 596 views; 4 CrossRef citations to date; and Altmetric 4 (or similar)]
Power, A., Atkinson, R. & McBeath, C. (2018). Editorial 28(2). Issues in Educational Research, 28(2), ii-v. http://www.iier.org.au/iier28/editorial28-2.html
Wikipedia (undated). Bingbot. https://en.wikipedia.org/wiki/Bingbot
Please cite as: Atkinson, R., McBeath, C. & Power, A. (2018). Editorial 28(4): Reflecting upon some unused IIER statistics. Issues in Educational Research, 28(4), ii-v. http://www.iier.org.au/iier28/editorial28-4.html |