iMechanica - research - Comments

Forget deceptive numbers, think about bibliometrics makers!

Roozbeh Sanaei — Sun, 20 Jul 2008 14:04:51 -0400

I wish to point out that this strong citation influence actually has many negative side effects on scientific research and publication. Main problem is ,Authors who make large part of bibliometrics are not as careful as people who analyze them. ,Secondly, it has created an extreme situation where many good research papers are congested in the narrow pipelines of a few “top” journals. This congestion may cause delay in publication or, even worse, wasted time as a result of rejection which, as proudly claimed by some “top” journals, happened to over 90% of the submitted manuscripts. Thirdlly, it has helped to form an opinion that articles published in the “top” journals are, without doubt, the most important research and the most significant findings. Therefore, many readers have developed a habit of reading just, if not only, a few “top” journals. Fourthly, it actually encouraged authors to intentionally cite later follow-up articles published in journals with “high” impact factors instead of give due credit to the earlier pioneering works appeared in journals with “low” impact factors. This is because the reference to “top” journals may give a “strong” appearance of the manuscript and, when previous authors favorably cited are actually the board members of the “top” journals (for example NATURE!), the likelihood of acceptance may greatly increased. The drawbacks of citation-driven research may contain many other aspects which I simply have no time and space to write here. But a harsh reality may tell the actual consequence of this practice. Many people actually complain about seeing decrease in the quality of discovery and innovation. Why? I believe part of the reason may be many scientists are actually chasing for citation and not really looking for best methods.

Hirsch's original paper has already more than 250 citations... !

Mike Ciavarella — Sun, 20 Jul 2008 11:22:59 -0400

Despite a paper of 2005, the H-index original paper has already 250+ citations.... immense interest, as Nature's Editor said...

I select a few:

PDF] Dynamic h-index: the Hirsch index in function of time - all 6 versions »

Comparison of the Hirsch-index with standard bibliometric indicators and with peer judgment for 147 … - all 6 versions »

[PDF] On the Opportunities and Limitations of the H-index - all 2 versions »

What do we know about the h index - all 3 versions »
An extension of the Hirsch index: Indexing scientific topics and compounds - all 7 versions »

Generalized Hirsch h-index for disclosing latent facts in citation networks - all 7 versions »

[PDF] Simple models and the corresponding h-and g-index - all 3 versions »

Does the h-index for ranking of scientists really work? - all 5 versions »

On the h-index-A mathematical approach to a new measure of publication activity and citation impact -

from the citations of "measures for measures" more infos...

Mike Ciavarella — Sun, 20 Jul 2008 11:15:11 -0400

I attach below a list of papers which cite the 2006 "Measures for Measures" Nature commentary.

You will see that the discussion is very active, Hirsch proposed in 2007 to use his index to "predict" the future performances of researchers, some correct for self-citations, some are correlating this with proposal evalutation, some others suggest to use it in peer review. Some start to judge universities.

So the matter is certainly alive...

Self-citation corrections for the Hirsch index - M Schreiber - EPL, 2007 - iop.org
About one year ago the physicist Hirsch [1] proposed an easily computable index h as an estimate of the visibility, importance, significance, and broad impact of a scientist's cumulative research contribution. This index h is defined ... Cited by 10 - Related Articles - Web Search

Achievement index climbs the ranks - P Ball - Nature, 2007 - nature.com
The h-index was first proposed in 2005 by Hirsch 2 , and attracted immense
interest from scientists because of its claimed objectivity in ranking
scientific achievement. Although other bibliometric measures of ...
Cited by 9 - Related Articles - Web Search - BL Direct

Does the h index have predictive power? - JE Hirsch - Proceedings of the National Academy of Sciences, 2007 - National Acad Sciences
The h index of a researcher is the number of papers coauthored by the researcher with at least h citations each (1). We have recently proposed it as arepresentative measure of individual scientific achievement. Other commonly ...
Cited by 7 - Related Articles - Web Search - BL Direct

The strike rate index: a new index for journal quality based on journal size and the h-index of … - W Barendse - Biomedical Digital Libraries, 2007 - archive.biomedcentral.com
Quantifying the impact of scientific research is almost always controversial,
and there is a need for a uniform method that can be applied across all fields.Increasingly, however, the quantification has been summed up in the impact ...
Cited by 2 - Related Articles - Cached - Web Search

Deja vu A study of duplicate citations in Medline - M Errami, JM Hicks, W Fisher, D Trusty, JD Wren, … - Bioinformatics, 2008 - Oxford Univ Press
Scientific misconduct comes in many forms. The Office of Science and TechnologyPolicy defines research misconduct as ‘fabrication, falsification or plagiarism in proposing, performing or reviewing research, or in reporting
Cited by 4 - Related Articles - Web Search - BL Direct

Predicting impact factor one year in advance - CM Ketcham - Laboratory Investigation, 2007 - nature.com The first impact factor (IF) to reflect the sole efforts of a new editorial team occurs 4 years into what is usually a 5-year editorship, owing to the lag times of: paper accrual and publication, accumulation of citations in derivative ...
Cited by 2 - Related Articles - Web Search - BL Direct

How do ecological journals stack-up? Ranking of scientific quality according to the h index
JD OLDEN - Ecoscience, 2007 - bioone.org The competitive nature of today's scientific environment requires theavailability of ranking indicators that are both fair and easy to compute. An ecologist's publication record is the paper-trail that defines his/her ...
Cited by 1 - Related Articles - Web Search - BL Direct

[PDF] Making science count: Open Access and its impact on the visibility of science - all 9 versions » D Law - 2007 - eprints.cdlr.strath.ac.uk
Strathprints is designed to allow users to access the research output of the University of Strathclyde. Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. ...
Related Articles - View as HTML - Web Search - BL Direct

[CITATION] Academic Institutions in the United States and Canada Ranked According to Research Productivity in … - all 6 versions » F Password - Conservation Biology, 2007 - Blackwell Synergy Skip to main content. ...

Related Articles - Web Search

Scientific indicators of productivity: time for action - all 4 versions » SM Dainesi, R Pietrobon - Revista Brasileira de Psiquiatria, 2007 - SciELO Brasil
OBJECTIVE: This article focuses on the differential activation of the
hypothalamic-pituitary-adrenal axis in generalized anxiety disorder and panicdisorder. METHOD: The results of recently reported reviews of the ...
Related Articles - Cached - Web Search

[PDF] Academic Institutions in the United States and Canada Ranked According to Research Productivity in …C Education - sillimanlab.com
Conservation biology is a young, but quickly maturing, scientific discip- line
currently under much scrutiny (Meine et al. 2006). Recent studies highlight what conservation biolo- gists publish (Fazey et al. 2005a), who publishes in ...
View as HTML - Web Search

[PDF] NHMRC grant applications: a comparison of “track record” scores allocated by grant assessors with … - all 3 versions » B Nicol, K Henadeera, L Butler - The Medical Journal of Australia, 2007 - mja.com.au
The primary aim of our study was to examine the track record score given toapplicants for project grants and to compare this with bibliometric analysis of the publications on which that assessment was based. A secondary aim was ...
Related Articles - View as HTML - Web Search - BL Direct

Assessment of Research Performance in Biology: How Well Do Peer Review and Bibliometry Correlate - all 2 versions »
BG Lovegrove, SD Johnson - BioScience, 2008 - bioone.org
Bibliometric indices based on publishing output, and citation records used to
measure scientific quality, are increasingly being employed to supplement and even replace traditional alternatives, such as the peer-review system. In ...
Related Articles - Web Search - BL Direct

The h-index and self-citations - all 2 versions »
L Engqvist, JG Frommen - Trends in Ecology & Evolution, 2008 - Elsevier
The h-index [1] has been claimed to provide a simple way to compare objectively the scientific achievement of researchers and has rapidly become one of the most favoured measures of scientific output [2]. The h-index is an author's ...
Related Articles - Web Search

Escape from the impact factor P Campbell - Nature, 2005 - int-res.com
As a competitive editor, I focus on (among many things) the added value that my colleagues provide to authors in enhancing their papers, and to readers in the assessment of a paper’s importance. Citations are one measure of our ...
Cited by 3 - Related Articles - Web Search

[PDF] The w-index: A significant improvement of the h-index
Q Wu - Arxiv preprint arXiv:0805.4650, 2008 - arxiv.org
I propose a new measure, the w-index, as a particularly simple and useful way to assess the integrated impact of a researcher’s work, especially his or her excellent papers. The w-index can be defined as follows: A researcher has ...
View as HTML - Web Search

Measure for Measure - play by Shakespeare or Nature commentary?

Mike Ciavarella — Sun, 20 Jul 2008 11:04:46 -0400

Measure for Measure is a play by William Shakespeare,
which deals with the issues of mercy, justice, truth and their relationship
to pride and humility: "Some rise by sin, and some by virtue fall".

Measures for measures is also a Commentary in Nature 444, 1003-1004 (21 December 2006). It proves the following interesting results:

1) impact factor as defined by ISI-Thompson (average number of citations acquired during the past two years for papers published in the same period) is "determined by a small fraction of highly cited papers, so the citation rate for individual papers is largely uncorrelated to the impact factor of the journal in which it was published".

2) The H-index does better than publication frequency, although (as in my comment to Roozbeh) Hirsch choice to define H-index is to equate h=C(h) where C(i) is number of citation of paper i, and so this is a special choice of more general h=A C(h)^K, where A=K=1 decided by Hirsch is not supported by any data and requires further study.

3) the probability for papers with large citations is indeed a power law (at least for the SPIRES database) a/k^gamma, where gamma=2.8 --- therefore, about 50% of all papers have 2 or less citations, top 4.3% of papers produces 50% of all citations

4) "Institutions have a misguided sense of the fairness of decisions reached by algorithm; unable to measure what they want to maximize (quality), they will maximize what they can measure".

5) compared to H-index, mean number of citations per paper is a much better indicator, and simple statistics shows this is especially reliable for large number of papers (like for esthablished researchers). So we should try that, perhaps on the isihighlycited.com
scientists.

Hence we seem to be able to restart discussion with Roozbeh, Bijwait, Chad and others based on this. Also, please consider that if we use Google Scholar, we are able to avoid in the long term the ISI-database, which means we will avoid the ISI journals, and hence we can open to open access. Therefore, there is a huge business and lobby to stop (US$ 9 bn) which presently is not doing its job with accuracy (see the poor representativity of impact factor which has for example Int J Plasticity very high whereas it has very low H index in fact), and which we will change for newer and more scientifically based (non-profit) systems.

Some useful background reading on this... from ScienceBlogs

Mike Ciavarella — Sat, 19 Jul 2008 17:26:18 -0400

Are biofuels a core climate solution?

Category:
Posted on: July 16, 2008 4:23 PM, by Joe Romm

In
order to know whether biofuels could be a major climate solution, the
scale of low-carbon energy deployment needed to avert climate
catastrophe must be understood.

According to the Latest Intergovernmental Panel on Climate Change report, if
carbon emissions average 11 billion tons a year (GtC/yr) this century,
then carbon cycle feedbacks will probably take us to atmospheric
concentrations of 1000 ppm of carbon dioxide and some 5.5°C warming from preindustrial levels. And I think we can probably all agree that is the end of life on this planet as we know it
with the loss of the inland glaciers that provide water to a billion
people, widespread desertification for up to one third of the planet,
loss of most species on land and sea, and ultimately an ice free planet
with sea levels up to 250 feet higher.

Unfortunately, unless we sharply reverse national and global energy
policy, we'll be over 11 GtC/yr in 2020. So what would it take to then
freeze at 11 GtC/yr for most of the rest of the century? It would
require 11 of Princeton's "stabilization wedges"
-- strategies and/or technologies that over a period of a few decades
each reduce global carbon emissions by one billion metric tons per year
from projected levels. Click to enlarge the figure.

SEE THE ORIGINAL BLOG FOR MORE DETAILS AT THE LINKS:

Are biofuels a core climate solution?

original August 2004 Science article on the wedges:

http://www.epa.gov/chp/documents/biomass_chp_catalog_part5.pdf

dear Peter I am afraid it is not a problem that I can fix

Mike Ciavarella — Sat, 19 Jul 2008 17:00:43 -0400

It looks fine on my computer, and indeed there were carriage returns. Since you are the first to complain it may be your setup? Sorry anyway we can ask the imechanica people when they have to check this maybe they can answer you. Regards, m

Carriage return

Peter J. Attar — Sat, 19 Jul 2008 12:26:47 -0400

Mike,

I am not sure if this is you or the imechanica page but your text scrolls

way to the right when I view it with FIrefox. If it is you could you please try and hit the carriage

return a couple more times when you type? It will make reading your informative posts

much easier. Thanks.

Geometry effects

Ajing Cao — Sat, 19 Jul 2008 11:14:51 -0400

Dear Wei,

Thanks for this interesting introduction. Acutually, in my opinion, the strength also becomes sensitive to factors other than the sample size per se, when the characteristic dimensions (such as the diameter of a pillar) of the metallic object are on the scale of a few nanometers to a few tens of nanometers (or larger). In other words, on this scale, the ‘‘size effects” can be derived not only directly from the sample dimension itself, but also from accompanying consequences that may not be immediately apparent. Our recent paper has demonstrated and explained such effects: unusually large influence is derived from the sample shape and temperature on the apparent strength of nanoscale objects. This work underscores the importance of acquiring detailed information about the sample geometry, including local curvature and atomic-scale features, before predicting and explaining the mechanical properties in measurements.

Cao A, Ma E, Sample shape and temperature strongly influence the yield strength of metallic nanopillars, Acta Mater (2008), doi:10.1016/j.actamat.2008.05.044

Ajing Cao

I will immediately sell my Google stocks now!

Mike Ciavarella — Sat, 19 Jul 2008 10:16:57 -0400

;)

Search Mechanical+Engineering+Forum in Google.

Roozbeh Sanaei — Sat, 19 Jul 2008 10:12:45 -0400

Search Mechanical+Engineering+Forum in Google.
This
one comes before Imechanica. (at least in my computer!).

I dont think any body tries to PUMP this blog into Google.

Laptop for College (Mechanical Engineering Student)

General Questions
1) What is your budget?
$1000-$1600
2) What size
notebook would you prefer?
a. Ultraportable; 12" screen or less
b. Thin
and Light; 13" - 14" screen
c. Mainstream; 15" - 16" screen
d. Desktop
Replacement; 17"+ screen
-B-
3) Please select your country's flag as a
post icon and tell us what country are you buying this in.
U.S.
4) Are
there any brands that you prefer or any you really don't like?
5) What tasks
will you be performing with the notebook?
Pretty much all the normal stuff
except for gaming. I dont really need my laptop to be a great gaming machine.
Also, I'm majoring in mechanical engineering so I'll need to run programs like
autoCAD.
6) Will you be taking the notebook with you to different places or
leaving it on your desk?
Mostly desk.
7) Will you be playing games on it;
if so, which games?
No
8) How many hours of battery life do you
need?
Over 5 preferably.
9) Do you mind buying online without seeing the
notebook in person?
No
10) What OS do you prefer? Windows (XP or Vista),
Mac OS, Linux, etc.
Not sure. Xp or Vistas fine.
Screen Specifics
11)
From the choices below, what screen resolutions would you prefer?
[B]d.
WXGA or WXGA+ - 1280x768/800 or 1440x900; Wider viewing version of XGA, good for
movie viewing or spreadsheets.[/B
e. WSXGA+ - 1680x1050; Wider viewing
version of SXGA, good for movie viewing or spreadsheets.
f. WUXGA -
1920x1200; Wider viewing version of UXGA, good for movie viewing or
spreadsheets.
-D-
12) Do you want a glossy/reflective screen or a
matte/non-glossy screen?
Glossy
Build Quality and Design
13) Are the
notebook's looks and stylishness important to you?
Somewhat
14) When are
you buying this laptop and how long do you want this laptop to last?
Early
August at the latest. Hopefully throughout college.
Notebook
Components
15) How much hard drive space do you want; 40GB to 500GB?
Not
really sure. 250GB sound good?
16) Do you need an optical drive? If yes, a
CDRW/DVD-ROM, DVD Burner or Blu-Ray drive?
DVD burner

but don't forget that "ranking" is a key to the Google success!

Mike Ciavarella — Sat, 19 Jul 2008 06:02:27 -0400

Bijwait

however, don't underestimate the importance of "ranking" --- it's the very key to the early success of Google. Google was the first company to heavily invest into ranking, and so documents you find on Google are high up ONLY if they have high access.

Of course, the very same success of Google, hides that advertissment and other forms of "pumping" the ranking by paying, is the other side of the medal. So very soon this will happen also in Science. People will pay, or at least form "citation clubs" to improve ranking and citations. I think I have posted already something about this when I discussed about the Research Excellence Framework in UK.

Regards

Mike

Some further links on Matlab and Abaqus, including the reverse!

Mike Ciavarella — Sat, 19 Jul 2008 05:03:12 -0400

Running Abaqus in Matlab environment « FEG: A journey of a mind

Following shows you a snip how to perform the analysis repeatedly in Matlab. for i = 1 : 10 istatus = dos(’abq661 job=filename interactive’); % Run Abaqus, ...
fegs.wordpress.com/2007/08/08/running-abaqus-in-matlab-environment/

ABAQUS: ABAQUS FEA Solver - Using Matlab in Abaqus

I am new to both Abaqus and Matlab and i would appreciate if some one can help me a bit in interfacing Matlab with Abaqus. I am doing a buckling analysis of ...
www.eng-tips.com/viewthread.cfm?qid=218317&page=3 - 24k

Essays.se: ABAQUS AND MATLAB

Biswajit, excellent reply -- are you sure we disagree? :)

Mike Ciavarella — Sat, 19 Jul 2008 04:06:28 -0400

I agree with most of what you say, except the comparison of H-index with Porn and Drug I think goes too far --- but is a good rethorical example.

I think our discussion is very fruitful, and I have already said you and Roozbeh are extremely good partners in the discussion, including when you are irritating :) See my recent comment from Anne-Wil Harzing directly A very interesting contribution from Anne-Wil Harzing !

I have to insist that there seem to be more interest in H-index, than in our REAL problems like the energy crisis. I think, and I am not alone (Sir Richard Branson said the same), that the energy crisis is a GOOD think--- in fact, finally we will agree to REDUCE the energy consumption. Maybe our USA friends will start to buy LESS of the Porsche Cayenne to go to Shopping Malls.

Where do you really think we disagree? I particularly like that you agree a LIST of the TOP papers is useful as a start for the young students.

I conclude that we probably agree at about 99%, and we can start work together on the same direction! :) If you read my post about the VIRGIN Challenge more carefully, I also suggested it is too hard for an individual (indeed as the prize of 25millions suggests), but for imechanica as a whole???

A very interesting contribution from Anne-Wil Harzing !

Mike Ciavarella — Sat, 19 Jul 2008 03:46:13 -0400

I received these comments from Anne-Wil Harzing, who is an authority in the field, partly in favour of Roozbeh.

Mike,

I know SCIMAGO, but as you indicated their results
are based on Scopus, which has its own limitations, the most notable of
which being that for many journals it only covers the period from 1996
onwards. For those of you interested in citation analysis, you might
want to read up on the work published in information sciences and
library sciences. The most important journals here are Online
Information Review (which has several articles on Google Scholar and
Scopus in comparison with ISI), Scientometrics and JASIST (Journal of
the American Society for Information Science and Technology).

Of course there is lots of debate about the use of citation analysis,
many more papers similar to those that Roozbeh mentioned are referenced
and discussed in my various papers (white or published) on this topic.
See my research programme Quality and Impact of Academic Research for
more information (http://www.harzing.com/program6.htm). You and Roozbeh
might also be interested in a recent theme issue on "The use and misuse
of bibliometric indices in evaluating scholarly performance" (http://www.int-res.com/abstracts/esep/v8/n1/).

I thought it was useful to provide some background information.

Best wishes,
Anne-Wil

P.S. You are welcome to publish these comments on your blog and attribute them to me.

Re: H-index

Biswajit Banerjee — Sat, 19 Jul 2008 02:38:24 -0400

Dear Mike,

Your answers were not entirely satisfying. Let me reply to some of your points.

1) " That H-index is popular is a matter of fact, and does NOT depend on me." Indeed, drugs and porn are also quite popular and may even be useful to some people - does that mean that we should all start using them? Yours is not a scientific answer at all :)

2) The Virgin earth challenge is probably too difficult not any single person to address. Also, Eric's and Rozzbeh's (and also my) observations suggest that the number of reads is linearly related to the number of comments and may not reflect the number of readers. I have a suspicion that we may have only around 50 human readers per comment with the rest of the readers being in the form of bots and webcrawlers. So the number of hits does not mean that a topic is necessarily popular.

3) I don't think ranking is the prime motivation of scientists as a group or most people in general. However, you may be right that some academic scientists may be rank driven - academic departments do tend to select such individuals. Personally, I have found that many such rank-driven people are assholes (see Bob Sutton's No Asshole Rule [video ]).

4) Once again, I find the idea of ranking countries based on scientific output ridiculous. What generalization are uyou going to make about, say Cameroon? That all Cameroonians are idiots because their h-index is low? That reminds me of 19th century European racists who decided on the intelligence or otherwise of entire groups of people based on the shapes of their chin !

5) Surely you don't think that Dugdale's was the best paper ever to be published in JMPS! What do you mean by best?

6) It's good that you can afford to read the literature voraciously. I hope our younger readers get that message.

7) The Google scholar rankings do not separate out self citations and citations in arbitrary pages. I would like to see the actual number of citations of a paper by people who have not known the author personally. That would be a better measure of the actual impact of a paper. My opinion still remains that the exercise is pointless and a waste of our time though a list of highly cited papers (400 or more in our field, 2000 or more in physics) can be useful to a beginning graduate student.

Regards,

Biswajit