EERQI Final Conference, Brussels, 15-16 March 2011 This project is funded by the Socioeconomic Sciences and Humanities Section. Interrelations Of Indicators.

Slides:



Advertisements
Similar presentations
Writing up results This tutorial focuses on writing your results section. Click the next button in the bottom right hand corner to begin. Next QUIT.
Advertisements

Original Figures for "Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression Monitoring"
CHAPTER 1 Exploring Data
Chapter 13 Statistics © 2008 Pearson Addison-Wesley. All rights reserved.
Beginning the Visualization of Data
1 1 Slide © 2012 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
WISER: Bibliometrics I Who’s citing you? Angela Carritt & Juliet Ralph November 2011.
1 1 Slide © 2003 South-Western/Thomson Learning TM Slides Prepared by JOHN S. LOUCKS St. Edward’s University.
Five-Number Summary 1 Smallest Value 2 First Quartile 3 Median 4
Bibliometrics overview slides. Contents of this slide set Slides 2-5 Various definitions Slide 6 The context, bibliometrics as 1 tools to assess Slides.
Measures of Relative Standing and Boxplots
Not all Journals are Created Equal! Using Impact Factors to Assess the Impact of a Journal.
Warm-Up Exercises 1.Write the numbers in order from least to greatest. 82, 45, 98, 87, 82, The heights in inches of the basketball players in order.
Box and Whisker Plots A Modern View of the Data. History Lesson In 1977, John Tukey published an efficient method for displaying a five-number data summary.
Quartiles & Extremes (displayed in a Box-and-Whisker Plot) Lower Extreme Lower Quartile Median Upper Quartile Upper Extreme Back.
Descriptive Statistics Used to describe the basic features of the data in any quantitative study. Both graphical displays and descriptive summary statistics.
Chapter 2 Describing Data with Numerical Measurements General Objectives: Graphs are extremely useful for the visual description of a data set. However,
Chapter 3 - Part B Descriptive Statistics: Numerical Methods
Exploratory Data Analysis. Computing Science, University of Aberdeen2 Introduction Applying data mining (InfoVis as well) techniques requires gaining.
Class Meeting #11 Data Analysis. Types of Statistics Descriptive Statistics used to describe things, frequently groups of people.  Central Tendency 
Standards in science indicators Vincent Larivière EBSI, Université de Montréal OST, Université du Québec à Montréal Standards in science workshop SLIS-Indiana.
CHAPTER NINE Correlational Research Designs. Copyright © Houghton Mifflin Company. All rights reserved.Chapter 9 | 2 Study Questions What are correlational.
1 1 Slide © 2009 Thomson South-Western. All Rights Reserved Slides by JOHN LOUCKS St. Edward’s University.
Chapter 3: Central Tendency. Central Tendency In general terms, central tendency is a statistical measure that determines a single value that accurately.
© Copyright McGraw-Hill CHAPTER 3 Data Description.
Publication Bias in Medical Informatics evaluation research: Is it an issue or not? Mag. (FH) Christof Machan, M.Sc. Univ-Prof. Elske Ammenwerth Dr. Thomas.
Citation Searching with Web of Knowledge Roger Mills Catherine Dockerty OULS Bio- and Environmental.
Correlation Association between 2 variables 1 2 Suppose we wished to graph the relationship between foot length Height
Google Scholar as a cybermetric tool Alastair G Smith Victoria University of Wellington New Zealand
Introduction to Descriptive Statistics Objectives: 1.Explain the general role of statistics in assessment & evaluation 2.Explain three methods for describing.
1 1 Slide Slides Prepared by JOHN S. LOUCKS St. Edward’s University © 2002 South-Western/Thomson Learning.
Descriptive Statistics becoming familiar with the data.
ERT 207-ANALYTICAL CHEMISTRY
A Bibliometric Comparison of the Research of Three UK Business Schools John Mingers, Kent Business School March 2014.
ICOLIS 2007 AN ATTEMPT TO MAP INFORMATION LITERACY SKILLS VIA CITATION ANALYSIS OF FINAL YEAR PROJECT REPORTS N.N. Edzan Library and Information Science.
Measures of central tendency are statistics that express the most typical or average scores in a distribution These measures are: The Mode The Median.
Citation Searching with Web of Knowledge Roger Mills Catherine Dockerty OULS Bio- and Environmental.
Categorical vs. Quantitative…
Chapter 3 Descriptive Statistics II: Additional Descriptive Measures and Data Displays.
To be given to you next time: Short Project, What do students drive? AP Problems.
1 Descriptive Statistics 2-1 Overview 2-2 Summarizing Data with Frequency Tables 2-3 Pictures of Data 2-4 Measures of Center 2-5 Measures of Variation.
Chapter 3, Part B Descriptive Statistics: Numerical Measures n Measures of Distribution Shape, Relative Location, and Detecting Outliers n Exploratory.
Chapter 3: Central Tendency. Central Tendency In general terms, central tendency is a statistical measure that determines a single value that accurately.
IMPORTANCE OF STATISTICS MR.CHITHRAVEL.V ASST.PROFESSOR ACN.
Citation Searching To trace influence of publications Tracking authors Tracking titles.
DfE Statistical First Release – 23 Oct 2014 The DfE published the ‘Statistical First Release’ of the 2014 results at the end of last month. You can follow.
CiteSearch: Multi-faceted Fusion Approach to Citation Analysis Kiduk Yang and Lokman Meho Web Information Discovery Integrated Tool Laboratory School of.
Statistics topics from both Math 1 and Math 2, both featured on the GHSGT.
1 CS 430: Information Discovery Lecture 5 Ranking.
1 Accountability Systems.  Do RFEPs count in the EL subgroup for API?  How many “points” is a proficient score worth?  Does a passing score on the.
2-6 Box-and-Whisker Plots Indicator  D1 Read, create, and interpret box-and whisker plots Page
Bell Ringer Find the median of the following set of numbers.
Copyright © 2016 Brooks/Cole Cengage Learning Intro to Statistics Part II Descriptive Statistics Intro to Statistics Part II Descriptive Statistics Ernesto.
Date of download: 6/27/2016 Copyright © 2016 SPIE. All rights reserved. Flow chart of the imaging processing. See Sec. 2 for details. Figure Legend: From:
EERQI Final Conference, Brussels, March 2011 This project is funded by the Socioeconomic Sciences and Humanities Section. EERQI Innovative Indicators.
EERQI Final Conference, Brussels, March 2011 This project is funded by the Socioeconomic Sciences and Humanities Section. EERQI Basic Features Part.
Where Should I Publish? Journal Ranking Tools
WIS/COLLNET’2016 Nancy, France
Sign critical appraisal course: exercise 1
Bryan G. Cook, University of Hawaii
CHAPTER 3 Data Description 9/17/2018 Kasturiarachi.
Journal Citation Reports® – “the JCR” February 2009 Enhancements
Citation Searching with Web of Knowledge
Chapter 1 Warm Up .
EERQI Basic Features Part 1:
Citation Searching with Web of Knowledge
Fig. 1 Patterns of productivity during a scientific career.
Understanding How the Ranking is Calculated
EERQI Innovative Indicators and Test Results
Presentation transcript:

EERQI Final Conference, Brussels, March 2011 This project is funded by the Socioeconomic Sciences and Humanities Section. Interrelations Of Indicators Work in Progress Prof. Dr. Stefan Gradmann / Dr. Frank Havemann Humboldt-Universität zu Berlin / Berlin School of Library and Information Science (IBI)

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Overview Base Data Intrinsic Indicators: Interrelation Extrinsic paper data from search engines and social-network services Citations in Google Scholar Correlation of intrinsic total score with extrinsic scores

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Base Data Assessments of 179 papers based on intrinsic criteria two files of extrinsic data: –citation numbers of rated papers obtained with Google Scholar (on March 8, 2011) –data from search engines and social-network services. extrinsic author data suffer from homonymic authors → we only use paper attributes. Papers in English and in German distributed over three thematic groups: –Group 1 includes papers about "assessment, evaluation, testing & measurement" (35 / 35) –group 2 about "comparative and inter-/multicultural education" (33 / 17) –group 3 about "history and philosophy of education" (34 / 17)

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Intrinsic Indicators: Interrelation Rigour ratings –average of nine ratings of different aspects Originality ratings –average of three ratings of different aspects Significance ratings –average of four ratings of different aspects Combined rating score for each paper: the average ratings of all 16 aspects (total score on a scale from 0 to- 7). To do: weight the mean ratings of each paper with its number of ratings (→ we need all individual ratings by different persons that until now have not available). The scatterplots in the three figures of mean scores of rigour, originality, and significance show that the latter two correlate best, especially for English-language papers.

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Originality – Rigour Interrelation → Lowest Correlation Strength

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Rigour - Significance Interrelation → Medium-low Correlation Strength

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Originality – Significance Interrelation → Maximum Correlation Strength

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Intrinsic Indicators: Distributions of Total Scores Box-and-whisker plots of distributions of total scores per language and group can be compared in the figures We display only distributions of rated papers which also have data from search engines or social-network services. –The box in each plot contains 50 % of papers around the median (black horizontal line). –The range of ratings are visualised by the "whiskers". –Lonely points show outliers (which are more distant from the box than 1.5 times the box's height).

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Intrinsic Indicators: Distributions of Total Scores

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Extrinsic paper data from search engines and social-network services Sources: –CiteULike, LibraryThing, MendReader –Google, Metager Many papers have only hits in one service. To get useful data we apply the in-dubio-pro-reo rule and select maximum values. We assume that zero hits cannot be used as a valid value of an indicator and thus exclude papers without hits from the analysis. The hit distribution of papers with at least one hit is heavily skewed to the left: Many papers have only a few hits and only a few papers have many hits. We therefore use the logarithm of hit numbers as a more adequate representation.

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Extrinsic paper data from social-network services

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Extrinsic paper data from search engines: similar to social networks

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Extrinsic paper data from search engines and social-network services All papers with social-network hits also have search engine hits. Both hit numbers correlate –quite well in each of the three groups for papers in English –and less well for papers in German

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Extrinsic paper data from search engines and social networks: English

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Extrinsic paper data from search engines and social networks: German

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Citations in Google Scholar: citation distributions for samples of the three groups Not all papers are listed in Google Scholar. Only a few papers in German are in the sample. We omit them. Here we use the y-scale of dual logarithms of numbers of citation + 1. The addition of 1 is a usual bibliometric method to include papers without citations into the analysis of log-values. It can be justified with the argument that publishing a new result is its first citation.

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Citations in Google Scholar: The total scores (mean ratings) Note, that the first (red) group is rated best but cited worst (in contrast to the results for search engines and social- network services, where for papers in English ratings and hit numbers on the aggregated level of thematic groups seem to correlate). Extrinsic author data remain a to do: an effective method for disambiguating authors is needed first

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Correlation?

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Correlation of intrinsic total score with extrinsic scores In the case of hits in social networks and in search engines there is no correlation with intrinsic total score as the scatterplots show.

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Correlation of intrinsic total score with extrinsic scores: social networks

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Correlation of intrinsic total score with extrinsic scores: social networks

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Correlation of intrinsic total score with extrinsic scores: search engine data

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Correlation of intrinsic total score with extrinsic scores: search engine data

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Correlation of intrinsic total score with extrinsic scores The same is true for citations of papers in English drawn from Google Scholar

Interrelations of Indicators / Stefan Gradmann, Frank Havemann EERQI Final Conference, Bruxelles March Conclusion As a consequence, to do any correlation analysis (including rank correlation) of these intrinsic and extrinsic paper data does not make any sense... … as long as such an analysis is based on paper attributes exclusively! → Effective author name disambiguation and disciplinary allocation is key Preliminary results do not yet invalidate the correlation methodology … … but they are revealing in terms of source data quality! How to understand the variance among the sub-samples?