Elsevier

Measurement

Volume 42, Issue 9, November 2009, Pages 1270-1277
Measurement

Widely-defined measurement – An analysis of challenges

https://doi.org/10.1016/j.measurement.2009.03.009Get rights and content

Abstract

The paper examines fundamental problems of widely-defined measurement that lie outside the representation concerns of measurement theory. It is intended as a starting point of a research agenda. It shows that measurement is applied in a wide range of diverse domains of knowledge and enquiry for which a wide-sense definition of measurement is necessary. It examines philosophical objections to the application of measurement. It considers in particular problems of measurand concept formation, validity, verifiability and of theories for the measurand. It concludes that Measurement Science should address the whole range of applications of measurement and should endeavour to provide a universal framework of concepts and principles to address all applications of measurement.

Introduction

Measurement, that is the descriptive representation of the attributes of objects and events of the real world by symbols on the basis of an objective empirical process, is a basic tool of modern human thought. It is the way in which we describe and reason about the world.

Measurement has been developed through the physical sciences, which serve as a paradigm. From this basis its application has been extended to virtually all domains of human knowledge and discourse. However, the concepts and methods of measurement in this wider and more diverse range of disciplines offer significant conceptual problems, compared with measurement in the physical sciences that is the normative view of much metrological discourse.

Some of these problems will be outlined in the present paper as a starting point of discussion and an agenda for research.

Section snippets

Historical development

The present concepts and principles of measurement are the product of a long historical development. An understanding of this process of development is very necessary to help the extension of the application of measurement to new domains, or to areas where measurement is still problematic. The literature of the history of mathematics and science does not, in general, treat the development of measurement explicitly in its general account. Some critical historical philosophical studies of the

Measurement theory

Modern logical and philosophical understanding of the fundamental concepts of measurement is based on the representational theory. The theory is based on the viewing of the real world as empirical relational systems and measurement as a process of mapping them into symbolic relational systems.

The theory has been extensively presented in the literature [13], [14], [15], [16], [17], [18]. An outline has been presented in [19]. The theory has been extensively considered and important contributions

Properties of measurement

The properties of measurement arising from the above wide-sense definition will now be analysed and discussed to provide an explanation of the usefulness of its wide and diverse applications.

Measurement provides an objective description of the measurand. The description is not merely a matter of opinion or feeling. It is invariant in rational discourse.

Measurement is verifiable. Given a specification of the measurement process the same symbolic description of a measurand should in principle be

Applications of measurement

As was discussed above, measurement is extensively applied outside the physical sciences, in domains where widely-defined concepts of measurement are used. This outline of the range and diversity of applications illustrates the significance of measurement outside the physical sciences.

Measurement in psychology is the first example to be cited. It is the endeavour to apply measurement in psychology that first challenged the restrictive view of measurement of the physical sciences and led to the

Philosophical considerations

The wide application of measurement is driven by philosophical considerations. It is based on a belief that empirical observation represents the only secure basis of knowledge and that objectivity is possible.

In this outline it is not possible to review the relevant philosophical discussions adequately. Relevant entries in [54], [55], [56] are recommended for further reading. A good summary discussion is presented in [27].

There is a wide adoption of the programme of Galileo to measure what is

Measurand concept formation

The establishment of a process and scale of measurement requires the formation of the concept of the measurand. This has been a central problem of the establishment of measurement concepts of such attributes as temperature, or electrical current, which unlike length and weight are not simply perceived by the human senses.

The process of measurand concept formation has been discussed in the literature of logical positivism [63] and considered in [64].

The consideration of widely-defined

Experiment and observation

Measurement theory relies on an ability to perform experiments for the purpose of establishing the existence of empirical relational systems.

In many cases direct observation of the measurand is not possible and it must be indirectly observed. Such indirect experiments consist of exciting the system under test and observing the response. In many cases of widely-defined measurement, such experiments are not possible, because the system under observation cannot be significantly disturbed.

This is

Replicable relations

The establishment of a scale of measurement for a quality is based on the existence among the quality manifestations of empirical relations that can be observed to be replicable. Such relations are termed nomological, or law like.

In many systems, however, there are no replicable relations for the measurand.

As has been discussed in the section dealing with philosophical considerations humans are autonomous intelligent actors and may behave differently under apparently identical conditions. They

Reliability, validity and generalisability

The concepts of reliability, validity and generalisability are central to widely-defined measurement. A good summary account is given in [27].

Reliability is the term used to describe the extent to which a measurement procedure yields the same result on repeated trials. It is thus the same concept as that of uncertainty used in strictly defined measurement. This is well discussed in the literature [72]. Reliability may be difficult to achieve in practice in many applications of measurement, but

Theories

The aim of scientific inquiry is, in general, to establish a well-defined theory for a domain of knowledge. Such a theory consists of a system of quantities and relations between them. It provides descriptive, explanatory and predictive power. In the physical sciences such a system is well established.

Such theories are not well developed in the psychological and social sciences and, as been discussed above, the possibility of such theories for systems incorporating human actors is widely

Verifiability

One of the principal pragmatic strengths of measurement is that measurement statements are verifiable. A measurement statement consists of a symbol assigned to the measurand, together with the specification of the scale of measurement employed. Given such a statement the measurement can be replicated by any other observer.

In the strongly defined measurement in the physical sciences a measurement result consists of a number and the specification of the unit.

In physical and chemical measurements

Conclusions

Measurement is applied in a wide range of human inquiry and discourse. Measurement in the physical sciences is the dominant paradigm. However, in significant areas of application of measurement that paradigm is inapplicable and a wide-sense definition of measurement is necessary.

Measurement science should address the whole range of applications of measurement. It should endeavour to provide a universal framework of concepts and principles to address all applications of measurement.

Measurement

Acknowledgements

The author thanks his colleagues Professors K.T.V. Grattan and S.H. Khan for their support. He is also grateful to Professors D. Hofmann, R. Morawski and L. Mari for creating the environment in which fundamental measurement science ideas can be discussed.

References (86)

  • J.L. Heilbron

    The Oxford Companion to the History of Modern Science

    (2003)
  • J.L. Heilbron

    The Oxford Guide to the History of Physics and Astronomy

    (2005)
  • A.W. Crosby

    The Measure of Reality: Quantification and Western Society, 1250–1600

    (1997)
  • J.L. Heilbron

    Weighing imponderables and other quantitative science around 1800

    (1993)
  • M.S. Morgan et al.

    Models as Mediators: Perspectives on Natural and Social Science

    (1999)
  • J.J. Roche

    The Mathematics of Measurement: A Critical History

    (1998)
  • J. Michell

    Measurement in Psychology: Critical History of a Methodological Concept

    (2005)
  • F.S. Roberts

    Measurement Theory: With Applications to Decision Making, Utility, and the Social Sciences

    (1984)
  • J. Pfanzagl et al.

    Theory of Measurement

    (1971)
  • D.H. Krantz et al.

    Foundations of Measurement Volume I: Additive and Polynomial Representations

    (1971)
  • D.H. Krantz et al.

    Foundations of Measurement Volume II: Geometrical, Threshold, and Probabilistic Representations

    (1989)
  • P. Suppes et al.

    Foundations of Measurement Volume III: Representation Axiomatization, and Invariance

    (1990)
  • L. Narens

    Abstract Measurement Theory

    (1985)
  • L. Finkelstein

    Foundational problems of measurement

  • A. Anastasi et al.

    Psychological Testing

    (1997)
  • J. Rust et al.

    Modern Psychometrics: The Science of Psychological Assessment

    (1999)
  • G.A. Gescheider

    Psychophysics: The Fundamentals

    (1997)
  • L. Cohen et al.

    Research Methods in Education

    (2007)
  • P. Abell

    Model Building in Sociology

    (1971)
  • H.M. Blalock

    Measurement in the Social Sciences: Theories and Strategies

    (1975)
  • R.A. Zeller et al.

    Measurement in the Social Sciences: The Link Between Theory and Data

    (1980)
  • K. Kempf Leonard

    Encyclopedia of Social Measurement

    (2005)
  • P. Hudson, History by Numbers: An Introduction to Quantitative Approaches, Arnold/co-published in the USA by Oxford...
  • D.E. McNabb

    Research Methods for Political Science: Quantitative and Qualitative Methods

    (2004)
  • M. Boumans

    How Economists Model the World into Numbers

    (2005)
  • B.E. Needles et al.

    Principles of Financial Accounting

    (2008)
  • M.J. Mard et al.

    Valuation for Financial Reporting: The Determination of Fair Value for Audited Intangible Assets

    (2007)
  • J. Berry

    Tangible Strategies for Intangible Assets: How to Manage and Measure Your Company’s Brand, Patents, Intellectual Property, and Other Sources of Value

    (2005)
  • J.R. Hitchner

    Financial Valuation: Applications and Models

    (2006)
  • T.R. Bielecki et al.

    Credit Risk: Modeling, Valuation and Hedging

    (2002)
  • J.F. Dean

    The Art & Science of Financial Risk Analysis: The Analyst’s Guide to Determining a Company’s Financial Strength and Creditworthiness

    (2006)
  • R. Franzosi

    Content Analysis

    (2008)
  • K. Johnson

    Quantitative Methods in Linguistics

    (2008)
  • Cited by (41)

    • Measurements as the basis for interpreting the content of emotionally coloured acoustic signals

      2022, Measurement: Journal of the International Measurement Confederation
    • A structural model of direct measurement

      2019, Measurement: Journal of the International Measurement Confederation
    • Reflections on and test of the metrological properties of summated rating, Likert, and other scales based on sums of ordinal variables

      2019, Measurement: Journal of the International Measurement Confederation
      Citation Excerpt :

      Finally, omitting items in a sumscale seriously affects resulting scores (no test-free measurement). But perhaps Thurstone’s criteria are simply too rigorous in the realm of measuring cognitive states, and we had better stick to Steven’s (1951) less demanding definition, where “measurement is the attaching of numbers to objects in a meaningful way” [5,8,16,17]. While researchers should keep validity and reliability on their mind when they shorten scales, it remains an issue as to what extent shortened sumscales preserve the metrological properties of the parent sumscale [19].

    • Intersubjectivity of measurement across the sciences

      2019, Measurement: Journal of the International Measurement Confederation
    • Phatic systems in digital society

      2016, Technology in Society
      Citation Excerpt :

      That measurement is a fundamental source of examples of symbolic tokens follows from reflecting on the characteristics of measurements themselves. Again, guided by Finkelstein’s [39] perceptive analysis, these characteristics can be summarised using the following three categories of nature, use and symbolism: Nature: Measurement provides a description of some attribute of an object or process.

    • "measurement" and related concepts. Their interpretation in the VIM

      2014, Measurement: Journal of the International Measurement Confederation
    View all citing articles on Scopus
    View full text