is peer review any good? a quantitative analysis of peer review

20
Is peer review any good? A quan4ta4ve analysis of peer review Fabio Casa), Maurizio Marchese, Azzurra Ragone, Ma6eo Turrini University of Trento h6p://eprints.biblio.unitn.it/archive/00001654/01/techRep045.pdf

Upload: aliaksandr-birukou

Post on 28-Nov-2014

2.038 views

Category:

Education


0 download

DESCRIPTION

This is a presentation of the paper in which we focus on the analysis of peer reviews and reviewers behavior in conference review processes. We report on the development, definition and rationale of a theoretical model for peer review processes to support the identification of appropriate metrics to assess the processes main properties. We then apply the proposed model and analysis framework to data sets about reviews of conference papers. We discuss in details results, implications and their eventual use toward improving the analyzed peer review processes.

TRANSCRIPT

Page 1: Is peer review any good? A quantitative analysis of peer review

Ispeerreviewanygood?Aquan4ta4ve

analysisofpeerreview

FabioCasa),MaurizioMarchese,AzzurraRagone,Ma6eoTurrini

UniversityofTrento

h6p://eprints.biblio.unitn.it/archive/00001654/01/techRep045.pdf

Page 2: Is peer review any good? A quantitative analysis of peer review

Ini)alGoals

•  Understandhowwellpeerreviewworks

•  Understandhowtoimprovetheprocess

•  Metrics+Analysis–  (refertoliquiddoc)

•  Focusonlyongatekeepingaspect “Not everything that can be counted counts,

and not everything that counts can be counted.” -- Albert Einstein

Page 3: Is peer review any good? A quantitative analysis of peer review

MetricDimensions

Quality

Fairness Efficiency

Kendall Distance

Divergence

Disagreement

Biases

Robustness

Unbiasing

Effort vs quality. Effort-invariant alternatives

Page 4: Is peer review any good? A quantitative analysis of peer review

DataSets

•  Around7000reviewsfromvariousconferencesintheCSfield(moreontheway)– Large,medium,small– Somewith“youngreviewers”

Page 5: Is peer review any good? A quantitative analysis of peer review

Ispeerrevieweffec)ve?Doesitwork?

•  Andwhatdoesitmeantobeeffec)ve?HOWdowemeasureit?

•  Easiertomeasure/detect“problems”

•  Peerreviewrankingvs.idealranking

Page 6: Is peer review any good? A quantitative analysis of peer review

Comparingrankings

28 17 2

45 67 .. ..

89 33 ..

33 89 2

17 67 ..

28 .. ..

45

Page 7: Is peer review any good? A quantitative analysis of peer review

Idealranking(?)

•  Successinasubsequentphase•  Cita)ons

Suggested reading: Positional effect on citation and readership in arXiv, by Haque and Ginsparg

Page 8: Is peer review any good? A quantitative analysis of peer review

Comparingrankings

T=3

N=10

Divergence:Div(t,N)Kendallτ

Page 9: Is peer review any good? A quantitative analysis of peer review

9

Page 10: Is peer review any good? A quantitative analysis of peer review

Results:peerreviewrankingvs.cita)oncount

10

Divergence

Div

Normalizedt

Page 11: Is peer review any good? A quantitative analysis of peer review

Randomnessandreliability

•  Quality‐relatedbutindependentofthecriteriaforthe“ideal”ranking

•  Basicstats•  Disagreement

•  Robustness•  Biases

11

Page 12: Is peer review any good? A quantitative analysis of peer review

Quality‐relatedMetrics:Sta)s)cs

12

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0 1 2 3 4 5 6 7 8 9 10

Prob

ability

Marks

Distribu4onofmarks(integermarks)

Page 13: Is peer review any good? A quantitative analysis of peer review

Disagreement

•  Measurethedifferencebetweenthemarksgivenbythereviewersonthesamecontribu4on.

•  Thera)onalebehindthismetricisthatinareviewprocessweexpectsomekindofagreementbetweenreviewers.

Page 14: Is peer review any good? A quantitative analysis of peer review

NormalizedDisagreement(aferdiscussion)

14

C1 C2 C3

Computed 0,27 0,32 (high variance) 0,26 (high variance)

Reshuffled 0,34 0,40 0,32

Page 15: Is peer review any good? A quantitative analysis of peer review

Robustness

•  Sensi)vitytosmallvaria)oninthemarks– Triestoassesstheimpactofsmallindecisionsingivingthemark(e.g.,6vs7…..)

•  Measuresdivergenceaferapplyinganε‐varia)ontothemark

•  Results:reasonablyrobustexceptfortheconferencemanagedbyyoungresearchers

15

Page 16: Is peer review any good? A quantitative analysis of peer review

Metricdimensions

Quality

Fairness Efficiency

Statistics

Kendall Distance

Divergence

Disagreement

Biases

Robustness Unbiasing

Effort

Page 17: Is peer review any good? A quantitative analysis of peer review

Fairness

•  Defini)on:Areviewprocessisfairifandonlyoftheacceptanceofacontribu)ondoesnotdependonthepar)cularsetofPCmembersthatreviewsit

•  Thekeyisintheassignmentofapapertoreviewers:Apaperassignmentisunfairifthespecificassignmentinfluences(makesmorepredictable)thefateofthepaper.

Page 18: Is peer review any good? A quantitative analysis of peer review

Poten)albiases

•  Ra4ngbias:Reviewersarebiasediftheyconsistentlygivehigher/lowermarksthantheircolleagueswhoarereviewingthesamepaper

•  Affilia4onbias

•  Topicbias•  Countrybias•  Genderbias•  …

Page 19: Is peer review any good? A quantitative analysis of peer review

ComputedNormalizedRa)ngBiases

C2 C3 C4

top accepting 3,44 1,52 1,17

top rejecting -2,78 -2,06 -1,17

> + |min bias| 5% 9% 7%

< - |min bias| 4% 8% 7%

C2 C3 C4

Unbiasing effect (divergence) 9% 11% 14%

Unbiasing effect (reviewers affected) 16 5 4

Page 20: Is peer review any good? A quantitative analysis of peer review

20