CITATION METRICS WORKSHOP (WEB of SCIENCE) BASIC LEVEL Prepared by Cited Reference Team, NUS Libraries. March 2013 Section Description Page I # Citation Searching of Indexed Works using the Author Search tab 1-5 Appendix A: Tips for Author with name shared by many other researchers 6-10 II Smith J (Geography) who publishes from 1985 onwards in many journals, Steps 2-5 Brown, J (Botany) who publishes from 1980 onwards only in a few journals Steps 6-11 III Appendix B: Rationale on choosing All Research Domains rather than just domains of Arts Humanities and Social Sciences 11-12 # Web of Science indexes only journals, etc which are of high impact. When we use the Author Search tab, we are limiting the results to documents indexed by Web of Science. If Web of Science indexes 10,000 journals, etc, then the search will be done ONLY on these 10,000 sources. Section I Citation Searching of Indexed Works using Author Search Tab The author we are working on is Professor Brenda Yeoh S.A (Saw Ai), Dean, FASS, Professor of Geography. There are up to three initials in her name. If the author you are working with has only one initial, or has a name shared by many other researchers, please read Appendix A as well. 1. Under the Search tab, click either of the Author Search links. Page 1
2. Enter yeoh for the Last Name and bsa for the Initials. Click the Exact Matches Only button. 3. Click the Author Name Variant link. It is necessary to do this step because: People who cite the author might not be consistent in the way they cite her initials Family names might be mistaken for personal names. In Prof Yeoh s case no one mistook Ai as her surname. To save time and paper, we did not document the searches under Ai, BYS, etc. a. Enter yeoh for the Last Name and sa for the Initials. Click the Exact Matches Only button. Repeat for yeoh b. b. Click Select Research Domain. Page 2
4. Choose All Research Domains. For rationale on choosing All Research Domains rather than just domains of Arts Humanities and Social Sciences, please read Appendix B. 5. Click Select Organization. 6. Choose the correct organization/s, click to choose Include records that do not contain organization information, and then click Finish Search. 7. By default, the list is sorted by publication date, newest to oldest. If you have a ResearcherID, select the latest and click I Wrote These Publications to add them to your ResearcherID My Publications list. ResearcherID is an online registry where you can build a publication list of your published research that can be made public or kept private. When you register you will be assigned a unique number that differentiates you from other researchers who may share your name and initials. Citation metrics will be generated automatically for certain publications in your profile, including Total Times Cited, Average Times Cited and h-index. Page 3
8. Glance through the list. If the list on the screen is long, scroll to the bottom of the page, choose up to 50 records to be shown per page. Most of the hits should be yours after you have done steps (2) to (6). 9. Scroll back to the top of the screen and click Create Citation Report. Page 4
10. You see a summary of the citation metrics, including the total citation count (Sum of Times Cited) and the h- index. 11. For Prof Yeoh, 90 of the 92 records are hers. The list of 92 records appear at the bottom. At this point, get rid of the false hits. a. If you are checking for someone else and you have a list of the author s work, for convenience, you can sort the list by other aspects like Source Title or First Author depending on how the list is organized. b. Click the checkboxes to select the false hits c. Click Go to generate the new report based on the 90 correct hits. b) Click the checkboxes to select the false hits a) Change sorting to Source Title or First Author if necessary c) Click Go to generate the new report based on the 90 correct hits 12. The new Citation Report shows that the 90 items authored by Prof Brenda Yeoh have been cited a total of 1104 times (965 without Self-Citations). Her h-index is 22. Page 5
Section II Appendix A: Tips for Author with Name Shared By Many Other Researchers 1. If you are not the author, it is best to get a list of publications from the author. If you cannot, try to obtain a list of: a. authors he/she normally work with b. journal titles he/she normally publish in c. information like when he/she starts publishing First Example: Smith, J, researcher in Geography, publishes from 1985 onwards in many journals 2. At Section I, step (4), if you choose All Research Domain (43,911 hits) you might end up having to look through many false hits just to get a few more real hits. Use the screen retrieved to decide. The following screen is for Smith, J. Since he researches in geography, it is better to choose only domains of ARTS HUMANITIES and SOCIAL SCIENCES (1987 + 3910 = 5,897 hits). a. If there are still too many hits, you can choose narrower fields. Click on the + next to the research domains to choose. b. You can narrow down significantly if you omit obvious, irrelevant fields of research like LITERATURE(462 hits), PSYCHOLOGY (1,248 hits) and EDUCATION EDUCATIONAL RESEARCH(544 hits). It is not advisable to choose just GEOGRAPHY alone as you would miss out relevant fields like AREA STUDIES and URBAN STUDIES. 3. Click Select Organization. Page 6
4. After narrowing down by Organizations, if necessary, narrow down by aspects like co-authors (represented by the Author tab), Publication Years, Source Titles by clicking on that heading. The following example is for Publication Years. a. Some of the years appear b. Click on more options / values... Page 7
c. Change the default sorting from Record Count to Alphabetical. Change to Alphabetical d. Select years, then click Refine. 5. Do step (7) to (12) of Section I. Page 8
Second Example: Brown, J, researcher in Botany, publishes from 1980 onwards in only a few journals 6. If you have the full publication list of the author and it is not long, and the published items are all from a few journals, do not use the Author Search at Section I step (1). In this example, Brown, J. publishes mainly in Plant Disease and Plant Physiology. Do as follows: a. The * after j represents truncation. Any name starting with J (John, James, etc) will be retrieved. Years of publication Change to Year Published Type in brown, J* to retrieve any name starting with J Title of journal Lastly, click Search 7. Choose the correct titles by clicking at the box on the left of each title. When done, click at the + at the top of the screen. For example, 25 titles were chosen. After you click +, the number 25 should appear. 8. Click Search to get back to the search screen. 9. At the search screen, change the title of the publication to Plant Physiology, then click Search. Page 9
10. Choose the correct titles, and click at the + at the top of the screen. For example, five titles were chosen. After you click +, the number 30 should appear. This is total of what you chose in steps (7) and (10). Next, click on the number 30. 11. Click on Create Citation Report. Page 10
Section III Appendix B Rationale on choosing All Research Domains rather than just domains of ARTS HUMANITIES and SOCIAL SCIENCES If only domains of ARTS HUMANITIES and SOCIAL SCIENCES were chosen, six records would be missed. They would be missed out because of the way Web of Science assigns the research categories and research domains. One of the six is a well cited item (12 citations). It is item 44 of the list of 92 sorted by Publication Date: newest to oldest. This list was obtained in step (6) of section I. Title: Values, change and inter-generational ties between two generations of women in Singapore Author(s): Teo, P; Graham, E; Yeoh, BSA; et al. Source: AGEING AND SOCIETY Volume: 23 Pages: 327-347 DOI: 10.1017/S0144686X0300120X Part: Part 3 Published: MAY 2003 Times Cited: 12 (from Web of Science) The research category assigned is Gerontology Page 11
Gerontology is assigned the research domain of LIFE SCIENCES BIOMEDICAL which is not what we would expect. Most of Brenda Yeoh s works are in the disciplines of ARTS HUMANITIES and SOCIAL SCIENCES. Does it mean we always choose All Research Domain? In this case, the difference in number of records retrieved is only seven. If we have chosen just domains of Arts Humanities and Social Sciences, we would have obtained 85 hits compared to the 92 obtained when we opted for All Research Domain. Of the seven, six are correct hits. If the author you are working on has only one initial or is a name shared by many others, and you choose All Research Domain you might end up having to look through many false hits just to get a few more real hits. By all means try All Research Domain first. Look at the hits retrieved and decide. Please read Appendix A. Page 12