Page 210 - Becoming Metric Wise

CHAPTER 7

Indicators

We will illustrate how most indicators introduced in the previous chapter
can be applied to any set of articles, not just those published in the same
journal. Also, collaboration indices are introduced. Moreover, some indi-
cators which are less suitable for journals are discussed. We begin with a
discussion of the h-index and h-type indices as these were originally
introduced outside the context of journals. Among other things, we show
how time series of indicators can be constructed.

7.1 INTRODUCTION
7.1.1 Definitions
Numerical data are facts expressed as numbers. Numerical data can be
communicated and transformed into other data.
An indicator is a mathematical expression.
Applying this mathematical expression to numerical data leads to an
indicator value.
Here is an example. The number of articles published by a university
research group in a given year and the number of FTEs (full-time person
equivalents) in that year are numerical data. Dividing the number of pub-
lications by the number of FTEs is an indicator (a mathematical expres-
sion). The resulting value of this indicator is the productivity of this
research group in that year.

7.1.2 Some Remarks on Indicators
Although strictly speaking not correct, we will follow the general custom
to consider the words index and indicator as synonyms. Moreover, people
often refer to the value of an indicator or index as the indicator or the
index, for instance one says: “This journal’s h-index is 25,” although
strictly speaking they should say, “The value of this journal’s h-index is
25.” Although somewhat sloppy, we will often follow this habit.
Indicators are often used as proxy variables. This means that they are
used for something else. The journal impact factor (JIF), for instance, is

Becoming Metric-Wise © 2018 Elsevier Ltd.
DOI: http://dx.doi.org/10.1016/B978-0-08-102474-4.00007-8 All rights reserved. 201