| 0
SciVal
For Administrators
Alexander van Servellen
Consultant, Research Management (SEA)
Elsevier Singapore
UNIMAP Malaysia
February 23rd, 2016
| 1
Agenda
10:00-11:00 Open Discussion on Needs
11:00-12:00 Setting up SciVal part I
12:00-14:00 Lunch & Informal Discussions
14:00-15:00 Setting up SciVal part II
15:00-15:45 Analysis & Reporting
15:45-16:00 Break
16:00-17:00 Generating Reports
| 2
Open Discussion on Needs
| 3
Typical uses
Measure scientific
production and
benchmark research
performance at
multiple levels
Assess the
Monitor research
international impact
trends
of research
Map collaboration
Identify leading
networks and
organizations and
identify collaboration
competitors
opportunities
Assess the impact of
research funding on Identify who is doing
the scientific output of what and with whom
researchers and in a variety of fields
graduate students
Source: http://www.science-metrix.com/en/expertise/bibliometrics/know-how
| 4
Reporting at various levels
Research Office Top
Management
Faculties
Research Units
| 5
Progress in setting up SciVal
1. Who has never logged into SciVal?
2. Who needs to setup SciVal for a particular school, department or centre of excellence?
3. What will you use SciVal for (as far as you know)?
4. What is UNIMAPs process for internal reporting?
5. What kinds of reports are produced currently?
6. Have you been able to set the relevant groups up in SciVal?
7. Do you have the list of researcher names and the corresponding Scopus Author IDs ready?
| 6
Notes
| 7
Setting Up SciVal
| 8
My SciVal Researchers and Groups
Creating Researchers
Creating Groups of researchers (departments etc.)
| 9
Master List
| 10
First create a Master List of researcher names
| 11
Who has a master list?
Can those of you who have a master list for their
department, or school, show them to me.
Can you let me know if there are any questions
about the master list?
| 12
Research with more than one
Scopus Author Profile
| 13
We start with the authors who have more than one
Scopus Author ID
| 14
Go to My SciVal and Define a new Researcher
| 15
Do this only for the authors who have more than one
Scopus Author ID!!
add the Scopus
author ID field
twice
Fill in the
Author IDs
of one
author
| 16
SciVal will ask you to confirm both profiles before
merging them into one profile in SciVal & Scopus
You can save the
researcher, or
validate the
publications first
| 17
Add a tag to the researcher indicating which
department he/she belongs to
Here I added a tag, so that
once there are hundreds of
researchers in SciVal, each
one is clearly tagged.
| 18
Finding individual authors
For the authors in the master list who have more than one
Scopus Author Profile
Please use the discussed method to add those authors to
SciVal.
Ask for my help if you need it.
| 19
Bulk Import Researchers that
have one Scopus Author Profile
| 20
The researchers with one Scopus Author ID can be
bulk imported into SciVal
You can paste the
author IDs belonging
to one department in
the box. Past just the
ID, not the author
name etc.
| 21
SciVal will show you if any Author ID is not found
If an author ID is not
found, then you
should search for
this author manually
by name in SciVal.
| 22
Create the Department and give it a name. Remember
to tag the researchers.
Give the
group a
name
Check this box
to create a
group for these
researchers
Tag the
researchers
| 23
Bulk uploading authors
For the authors in the master list for whom you have found
one Scopus Author Profile
Please use the discussed method to bulk upload them into
SciVal.
Ask for my help if you need it.
| 24
Notes
| 25
Finalize the Department
| 26
Now you have created a department in SciVal.
Remember to add the researchers who previously had
two Author IDs and were not bulk imported
Use edit function
to add or remove
researchers from
this group
Open the
definition editor
| 27
Identify the author to be added to the Department
Here you can clearly see
one author is not yet in the
group. It is the one we
created manually, because
he had two Author IDS
Drag and drop his name into the group
| 28
Save the Group again
| 29
Now you have created a department in SciVal which
you can easily share with other SciVal users.
Use the
sharing
function to
give others
access to this
department
| 30
Groups can be combined into larger organizational units
Drag and drop units to
combine
| 31
Bulk uploading authors
Now its time to finalize the department.
Please make sure the department includes all the
authors, and it has been given correct name, tags
etc.
| 32
Notes
| 33
Lunch -- meet back here at
14:00
| 34
Setting Up SciVal (continued)
| 35
My SciVal Research Areas
Creating Researcher Areas using keywords
Creating Research Areas by grouping entities
| 36
Create Research Area
| 37
Use Keywords to define a very specific area of science
Useful
search
tips
| 38
Refine & Validate
Validate that
publications are
representative of
topic
| 39
Prepare your Research Areas in advance. Areas larger
than 5000 publications can take 48 hours
| 40
Create a Research Area
Pick a specific topic you are interested in.
Use keywords to create a small Research Area in SciVal.
Let me know if you need help.
| 41
Notes
| 42
Sharing and ownership
| 43
Share Departments & Research Areas with others
Use the share
icon
43
| 44
Invite to share using email of registered SciVal user
44
| 45
Grant editing rights or change ownership of
departments you created, but have shared with others
By default, you are the only one who can edit entities which you created, and
have shared with others. You can grant editing rights to others, or even transfer
ownership to someone else.
| 46
Sharing
Now that you have made Departments and Research Areas in
SciVal
Please share one of them with a neighbor.
You could experiment with transferring ownership to them and
back again.
| 47
Notes
| 48
Analysis and Reporting
| 49
Now that you have customized SciVal you can
explore and analyze using SciVal
Overview module Benchmarking module is Collaboration module
provides dashboard flexible, it allows you to looks at specifically at co-
style reports based on a select a time period publication trends. It
3 or 5 year period. You anywhere from 1996 to allows you to identify the
can analyze countries, one month ago. You can top collaboration partners
institutions, any choose from over 17 of an institution or
researcher, group of metrics, and compare country, as well as
researchers and across different types of identify potential
research areas you entities (i.e. you directly collaboration partners.
have created. It gives a can compare researchers Research areas can be
comprehensive with departments, applied here as a subject
overview of one entity at institutions, countries in filter.
a time. one analysis)
| 50
Main purposes of metrics
Evaluation of performance
Typically top-down performed by an authority
Essential to account for variables that can affect metrics values besides
differences in performance, so that the evaluation is fair
Demonstration of excellence
Typically bottom-up performed by entity requesting resources
Generally aiming to find a way to showcase strengths compared to peers;
non-performance variables may be used to advantage in this case
Scenario modeling
Non-performance variables may or may not be important depending on
scenario
e.g. if modelling recruitment in nanoscience, might not need to worry about
different citation rates between fields because only looking at one field!
| 51
20 sets of metrics at your disposal
Slice and dice your data from multiple angles to identify your core strengths and
weaknesses
Productivity metrics Collaboration metrics
Scholarly Output Authorship Count
h-indices (h, g, m) Number of Citing Countries
Collaboration (geographical)
Citation Impact metrics Academic-Corporate Collaboration
Citation Count
Citations per Publication
Cited Publications Usage metrics
h-indices (h, g, m) Views
Field-Weighted Citation Impact Views per publication
Publications in Top Percentiles Field-Weighed Views Impact
Publications in Top Journal Percentiles
Disciplinarity metrics
Collaboration Impact (geographical)
Journal count
Academic-Corporate Collaboration Impact
Journal category count
Snowball Metric; www.snowballmetrics.com/metrics
| 52
Notes
Non-Performance Variables
| 54
When making comparisons, we must take into
consideration Non-performance variables
Size
Discipline
Publication-type
Database coverage
Manipulation
Time
In some cases the difference in a metric value between two entities
(authors for example) might not reflect a difference in performance.
Instead the difference can be caused by non-performance variables.
| 55
Publications types receive differing levels of citations
Because some publication types are cited more often than others, we
should not compare different types without applying normalization.
| 56
Citation practices differ between disciplines
Because some subject areas are cited more often than others, we should
not compare papers from different fields without applying normalization.
| 57
Size and Time
Citation count is
directly related to
volume
Do you think the declining
trend is indicative of a
decrease in quality or is it
caused by a non-
performance variable?
CPP normalizes for
differences in
publication volume
FWCI normalizes for
subject area,
document type, and
time
| 58
Reporting Requirements
Please think about what kinds of reports you aim to
produce, for what audience, and how frequently etc.
Draft a short outline of a report, and what it should include,
and please explain it to me.
| 59
Notes
| 60
Generating Reports
| 61
Approach writing a report similarly to writing a
scientific paper.
Be clear about the questions, scope and methods
Guide the reader through the results of your analysis
Summarize the findings in an executive summary (like an abstract)
Practical tips for using SciVal for reporting
Take screenshots of the relevant outputs in SciVal
Export data from SciVal and creating charts in Excel
Writing text which describes and interprets the data
Keeping it simple and get to the point
| 62
Exporting Data from SciVal
| 63
Use Screenshots to capture charts
| 64
Report Template
A good looking report reflects well on your skills as an analyst
It helps to prepare a template such that the report is well structured.
| 65
Final Assignment
Assignment: Group into pairs and together use SciVal to
create a report.
The report can be about anything relevant to your job: a journal
analysis, institutional analysis, department analysis etc. Think
of something you might actually have to produce in future.
The report should be in Power Point format and must include:
Cover page with title and your names job titles and email
addresses
Key Findings
Charts visualizations
A few groups can volunteer to share their report in 45
minutes from now. Feel free to ask me for help in working
on your report.
| 66
Notes
| 67
Thank you very much
Alexander van Servellen
[email protected]www.elsevier.com/research-intelligence
