What is Measurement?

For any rigorous, scientific investigation of phenomena (people, objects, events, attitudes, or behaviors) measurement is a central concern. Measurement entails developing rules and methods that guide researchers when they assign of numbers or labels to the phenomena under investigation. There is a wide variety of objects or things that researchers measure. The problem for social scientists is that the phenomena we seek to measure are generally too abstract to be classified has either objects or things. 

With the social sciences, measurement is the process of linking abstract concepts or constructs to empirical indicants.[1] It is a process involving both theoretical considerations about the concept under investigation and empirical observation or attitudes, behaviors, and events. From an empirical perspective, the researcher focuses his or her attention on an observable response. These responses can take the form of answers to a questionnaire or behavior observed in an observational study. From the theoretical perspective, the researcher focuses on the underlying unobservable—and, therefore, not directly measurable—construct that is represented by the response. When the relationship between the observable response and the unobservable concept is strong, we can make useful inferences about the underlying construct. But when the relationship between the construct and its presumed indicators is weak or faulty, the research could lead to incorrect inferences and misleading conclusions about the construct. 

Some things are relatively easy to measure because the rules used to create the measurement are simple. Age, for example is easily measured. We measure a person's age by looking at that persons recorded birth date and the calendar. Gender, on the other hand, is more complex. Historically gender was considered a dichotomous, or binary, variable. Researchers would ask respondents the following either/or question: Are you Male or Female? In recent years, however, there has been a growing awareness that some people are born with atypical gender characteristics that complicate gender identity. A person's gender identity is often no longer considered a simple dichotomous "either/or" or "yes/no" question. As a consequence, non-binary definitions of gender are used more frequently than ever before. Facebook, for example, has over 50 gender options from which users can choose.[2] Some researchers use a two-step protocol when asking questions of gender identity:[3]

Step 1:

What is your current gender identity?

O Male

O Female

O Trans male/Trans man

O Trans female/Trans woman

O Genderqueer/Gender non-conforming

O Different identity (please state): ____________________________

Step 2:

What gender were you assigned on your original birth certificate?

O Male

O Female


Many concepts of interest to researchers—brand loyalty, household income, purchase intent, voter intent—are difficult to measure because it is no easy to establish rules to measure the true value of these attributes.


 Toggle open/close quiz group

[1] Carmines, Edward G. and Richard A. Zeller. Reliability and Validity Assessment. Thousand Oaks, CA: Sage Publications Inc., 1979. p. 10.

[2] http://www.slate.com/blogs/lexicon_valley/2014/02/21/gender_facebook_now_has_56_categories_to_choose_from_including_cisgender.html

[3] Singer, T. B., M. Cochran, R. Adamec. Final Report by the Transgender Health Action Coalition (THAC) to the Philadelphia Foundation Legacy Fund (for the) Needs Assessment Survey Project (A.K.A. the Delaware Valley Transgender Survey). Transgender Health Action Coalition: Philadelphia, PA., 1997.

toc | return to top | previous page | next page