Once a researcher has identified a research question, conducted the literature review, and given some thought to what might be the most appropriate research method(s), they must then determine how they will measure the social phenomenon that is the focus of their research question. Measurement is a key part of research design and refers to the process of describing and ascribing meaning to key facts, concepts, or other phenomena under investigation. At its core, measurement is about defining one’s terms in as clear and precise a way as possible. Some constructs in social science research, such as a person’s age or the number of people in prison, may be easy to measure. Other constructs, such as creativity or prejudice, are considerably harder to measure. For these reasons, measurement in social science isn’t quite as simple as using some predetermined or universally agreed-upon tool. But no matter the topic, researchers must always think carefully and deliberately about how to measure the central components of their research questions.

This chapter covers the processes and concepts involved in empirical measurement: conceptualization and operationalization. We introduce you to variables and review the ways variables can be related to one another. This chapter also covers the four levels of measurement a variable can take as well as the link between the level of measurement of our variables and the types of analyses we will be able to conduct with our data in the latter stages of our research journey.

What Do Social Scientists Measure?

The question of what social scientists measure can be answered by asking what social scientists study. Think about the topics you have learned about in your criminology classes or the topics you have considered investigating yourself. Or, think about the many examples of research you have read about in this text so far.

Let’s take a look at a study on the fear of crime as an example of what types of concepts criminologists can measure. Kohm et al. (2012) studied how the fear of crime was related to coverage in news media. To conduct the study, they needed to be able to clearly define the fear of crime and identify how to measure it. How do we know whether or not someone is afraid of crime? How do we know that the person is not fearful of something else entirely? Is it a fear of property crime we want to examine or a fear of violent crime? Understanding how measurement works in research methods helps us answer these sorts of questions.

Social scientists will measure just about anything they have an interest in investigating. Let’s take a look at some other examples. Let’s say you are interested in learning something about the correlation between social class and levels of victimization. If these are the concepts in your study, you must develop some way to measure both social class and victimization. Or, if you wish to understand how a person’s gender shapes their workplace harassment experiences, then you must measure gender and workplace harassment experiences. You get the idea. Social scientists can and do measure just about anything you can imagine observing or wanting to study. Some things are easier to observe or measure than others, and the things we might wish to measure don’t necessarily all fall into the same category of measurables.

In 1964, philosopher Abraham Kaplan (1964) described three different categories of things that behavioural scientists observe: direct observables, indirect observables, and constructs. Direct observables are probably the simplest to measure in the social sciences. Direct observables are the sorts of things we can perceive ourselves with our senses. For example, broken windows in a car or graffiti on a building would be direct observables – you can see and touch them.

Indirect observables, on the other hand, are less straightforward to assess. Indirect observables require a degree of inference about the connection between the observation and what the observation represents (Babbie, 2007). If we conducted a study for which we wished to know if a person had ever committed a criminal offence, we would probably have to ask them about their behaviour, perhaps in an interview or a survey. Thus, we have observed criminal offending, even if it has only been observed indirectly. Birthplace might be another indirect observable. We can ask study participants where they were born, but it is unlikely that we have directly observed any of those people being born in the locations they report.

Sometimes, the measures we are interested in are more complex and more abstract than observational terms or indirect observables. Kaplan (1964) referred to these as constructs. Think about some of the constructs you’ve learned about in other classes—racism, for example. What is racism? One definition is prejudice or discrimination based on someone’s race or ethnicity. But how would you measure it? Another construct is patriarchy, which refers to the systemic power and authority that men hold over women. We know patriarchy influences society and social institutions, but measuring such a construct is more difficult than measuring a person’s income. In both cases, racism and patriarchy, these theoretical notions represent ideas whose meaning we have generally come to agree on. Though we may not be able to observe these constructs directly, we can observe the confluence of things they are made up of.

How Do Social Scientists Measure?

Measurement occurs at multiple stages of a research project, including planning, data collection, and sometimes even data analysis. A researcher begins the measurement process by describing the key ideas they hope to investigate, which are usually stated in the research question. For instance, let’s say our research question is: How do lawyers with different family backgrounds cope with the emotional demands of their job? Answering this question requires some idea about what coping means. We may come up with an idea about what coping means as we begin to think about what to look for (or observe) during data collection. Once we’ve collected data on coping, we also have to decide how to report on the topic. Perhaps, for example, there are different types or dimensions of coping, some of which lead to more successful emotional outcomes than others. The decisions we make about how to proceed and what to report will involve processes of measurement.

Measurement is a process in part because it occurs at multiple stages of conducting research. We could also think of measurement as a process because measurement in itself involves multiple stages such as identifying key terms, defining them, figuring out how to observe them as well as assessing whether our observations are any good. In the next sections of this chapter, we will elaborate on the task of identifying and defining the key terms/concepts and also figuring out how we will observe them. The assessment of whether our observations are any good will be explored in chapter 6c.

Conceptualization

A concept is the notion or image we conjure up when we think of some cluster of related observations or ideas. For example, masculinity is a concept. What do you think of when you hear that word? Presumably, you imagine some set of behaviours and perhaps even a particular style of self-presentation. Of course, we can’t necessarily assume that everyone conjures up the same set of ideas or images when they hear the word “masculinity.” In fact, there are many possible ways to define the term. And while some definitions may be more common or have more support than others, there isn’t one true, always-correct-in-all-settings definition. What is considered masculine may shift over time, from culture to culture, and even from individual to individual (Connell, 2005). Without understanding how a researcher has defined key concepts, it would be nearly impossible to understand the meaning of that researcher’s findings and conclusions. Thus, the process of measurement starts with conceptualization, which is the process of clearly defining our concepts and their components in concrete and precise terms.

Going back to our example of the concept of masculinity, think about what comes to mind when you see that term. How do you know masculinity when you see it? Does it have something to do with men? With social norms? If so, perhaps we could define masculinity as the social norms that men are expected to follow. That seems like a reasonable start, and at this early stage of conceptualization, brainstorming about the images conjured up by concepts and playing around with possible definitions is appropriate. But this is just the first step. It would make sense to consult previous research and theory (often identified in the literature review) to understand if other scholars have already defined the concepts we’re interested in. This does not necessarily mean we must use their definitions, but understanding how concepts have been defined in the past will give us an idea about how our conceptualizations compare with the predominant ones out there.

Understanding prior definitions of our key concepts will also help us decide whether we plan to rely on them for our own work, modify them slightly, or challenge them altogether. If we review the literature on masculinity, we will surely come across the work of Raewyn Connell, one of the world’s preeminent masculinity scholars. After consulting Connell’s prior work (1987; 2005; 2009) as part of our literature review, we might tweak our initial definition of masculinity just a bit. Rather than defining masculinity as “the social norms that men are expected to follow,” perhaps instead we’ll define it as “the social roles, behaviours, and meanings prescribed for men in any given society at any one time.” Our revised definition is both more precise and more complex. Rather than simply addressing one aspect of men’s lives (norms), our new definition addresses three aspects: roles, behaviours, and meanings. It also implies that roles, behaviours, and meanings may vary across societies and over time. Thus, to be clear, we’ll also have to specify the particular society and time period we’re investigating as we conceptualize masculinity.

As we can see with this example of masculinity, the conceptualization process is all the more important because of the imprecision, vagueness, and ambiguity of many social science concepts. After we have identified a clear definition we are happy with, we should make sure that every term used in our definition will make sense to others. Are there terms used within our definition that also need to be defined? If so, our conceptualization is not yet complete.

The option of challenging pre-existing conceptualizations may be most relevant when working with Indigenous communities. Rather than simply adopting colonial interpretations, it is critical to turn to Indigenous authors and your Indigenous community partners to see if certain conceptualizations should indeed be challenged. For instance, if you were examining the role of Indigenous spirituality in increasing resilience and decreasing involvement in criminal activity, you would not want to define or measure spirituality as “frequent church attendance.” Instead, a definition that pertains more to the inner nature of a person and their relationship to all that surrounds them would be a more appropriate reflection of how many Indigenous cultures conceptualize spirituality.

There is yet another aspect of conceptualization to consider: concept dimensions. In terms of social scientific measurement, concepts can be said to have dimensions when there are multiple elements that make up a single concept. Not all men are the same in their behaviour or attitudes, and these differences may be reflected in dimensions of the concept of masculinity. These dimensions could be regional (is masculinity defined differently in different regions of the same country?), age-based (is masculinity defined differently for men of different ages?), or perhaps power-based (are some forms of masculinity valued more than others?). In any of these cases, the concept of masculinity would be considered to have multiple dimensions. While it often is not necessary to spell out every possible dimension of the concepts you wish to measure, it may be important to do so depending on the goals of your research. The point here is to be aware that some concepts have dimensions and to think about whether and when dimensions may be relevant to the concepts you intend to investigate.

There is one last thing to consider in the process of conceptualization: reification. Although consulting scholarly definitions of our concepts is part of the process, we should also question prior definitions in the literature. How have these definitions changed over time? Are parts of these definitions missing? Are there discriminatory elements in the current definitions we can identify and revise in our own conceptualization? Think about the term “family.” Consider the definitions of family that were used in Western cultures 100 years ago, 50 years ago, and 5 years ago. How have our understandings of this concept changed over time? Challenging reification means evaluating and questioning these different definitions. Fifty years ago, family was conceptualized as an opposite sex couple with children; however, today our conceptualization of family is more inclusive of a variety of relationships and forms. But this conceptualization may vary based on social and geographical context, representing different dimensions of the concept.

As part of our work towards decolonizing research, we should pay specific attention to reification with regard to terms used with Indigenous peoples. Some have significantly changed over time, often as a result of colonization, cultural misunderstanding, or more positively via shifts in reclaiming empowerment. For instance, the term “Indian” was mistakenly used by European explorers who thought they had reached India. It is still widely used in the United States as a result of history and stereotyping. Currently, many Indigenous peoples prefer the term Indigenous, or First Nation, or their respective Tribal or First Nation affiliation.

The caution here is that our terms or concepts mean nothing more and nothing less than whatever definition we assign to them. It makes sense to negotiate social agreement about what various concepts mean. Without that agreement, it would be difficult to navigate through everyday living. But at the same time, we should not forget that we have assigned those definitions to that concept and that they are no more real than any other, alternative definition we might choose to assign. Conceptualization simply offers a clear and concrete definition of how we are using the concept within our own current research study.

Operationalization

Once a researcher has defined, or conceptualized, a concept, exactly how do they measure it? Operationalization refers to the process of explaining precisely how a concept will be measured. Operationalization flows from the conceptual definition by establishing a nominal definition, or dictionary definition, of the concept, then an operational definition, and finally identifying specific indicators, or empirical observations taken to represent the ideas we are interested in studying. Social scientists tend to measure most concepts using multiple indicators. For instance, if an indirectly observable concept such as socioeconomic status (SES)  is simply defined as the level of family income, it can be operationalized using an indicator (e.g., an actual survey question) that asks respondents to report their annual family income. However, if a researcher defines SES as a combination of elements (i.e. concept dimensions) including income along with level of education and occupation, as shown in Figure 6a.1, it would be measured with multiple survey questions that cover each of these elements.

Researchers using field research and other methodologies must also operationalize their concepts. For example, a researcher observing lawyers’ courtroom interactions might develop indicators of SES such as clothing styles, mannerisms, or patterns of speech. No matter what methodology a researcher chooses, they must always operationalize their concepts during the measurement process.

When operationalizing concepts in Indigenous-related research, the researcher needs to be careful not to adopt a colonial point of view as this may result in misdefining or oversimplifying the concept. For instance, if we were examining “well-being,” we might typically examine the elements of economic status, physical health, and individual happiness. From an Indigenous perspective, “well-being” might instead include the elements of spiritual balance, connection to the land or water, and cultural traditions as well as community harmony. Culturally grounded operationalizations are necessary to produce respectful and meaningful research outcomes.

The process of coming up with indicators must not be too arbitrary or casual, and reviewing prior theoretical and empirical work in the topic area can help. Theories point toward relevant concepts and possible indicators; published empirical studies give some very specific examples of how others have defined the important concepts in an area and what sorts of indicators they have used. It might make sense to use the same indicators as other researchers have, or it might make sense to develop new indicators that improve upon previous ones.

Another important aspect that influences the choice of indicators is your methodological approach. A quantitative survey implies one way of measuring concepts – specific, direct written questions with limited response options, while qualitative field research implies a quite different way of measuring concepts – observing people interact in certain ways and watching for certain specific behaviours. As you can see, your data collection strategy – i.e. the specific method you select, whether it be interviews or surveys, or field research – will play a significant role in shaping how you operationalize your concepts.

Figure 6a.2 is an example of the process of operationalization, and its connection to conceptualization. While the research question is very general – how does crime vary across Canada?  – the process of operationalization narrows the focus of the research. This figure does make the measurement process look like a set of linear stages through which a researcher neatly progresses before beginning data collection; however, it doesn’t necessarily always work that way, especially in inductive research. It is normal to revise indicators as the study progresses. For example, in the proposed study in Figure 6a.2, it is unlikely that the indicators will change, but the focus may narrow further to only look at property crimes. If the study was instead exploring how people across Canada perceive crime in their communities, the indicators may evolve as part of the research based on observation or listening to what participants themselves share as indicators of crime.

Variables and Levels of Measurement

Now that we have introduced you to the process of measurement and how we conceptualize and operationalize our concepts of interest, let us move on and talk about the indicators a researcher develops to measure abstract concepts pertinent to their study, which are often called variables. A variable is something with a quantity or quality that can vary (e.g., from low to high, negative to positive, or from one colour/feature to another), in contrast to constants, which do not vary (i.e., remain constant and do not change). The options a variable can take are called attributes. For instance, if our variable is eye colour, the attributes would be the options this variable can take (brown, blue, green, grey, etc.). Variables are essentially groupings of relevant attributes (Maxfield & Babbie, 2018).

The fact that there are changes in the variable enables a researcher to measure those changes and use the differences to address their research question. Returning to the example in Figure 6a.2 about how crime varies across Canada, the variable in this study would be the crime rates for specific offences and the provinces. For instance, we would expect that the rates for robbery would not be the same in each province, and we can compare the differences in robbery between the provinces to help address our research question.

Levels of Measurement

Based on their characteristics, variables are generally categorized into one of four possible levels of measurement: nominal, ordinal, interval, and ratio.

The nominal level is the most basic level of measurement. Variables at the nominal level of measurement are both exhaustive and mutually exclusive, meaning that all possible categories or attributes are included and no one can belong to more than one category, respectively. Relationship status, gender, and province of birth are examples of nominal-level variables. For example, to measure relationship status, we might ask respondents to tell us if they are currently partnered or single. These two attributes exhaust the possibilities for relationship status (i.e., everyone is always one or the other of these), and it is not possible for a person to simultaneously occupy more than one of these statuses (e.g., if you are single, you cannot also be partnered). Thus, this measure of relationship status meets the criteria for nominal-level attributes to be exhaustive and mutually exclusive. One unique feature of nominal-level measures is that they cannot be mathematically quantified. We cannot say, for example, that being born in Ontario has more or less quantifiable value than being born in Nova Scotia – you are not more provincial if you are born in one province or another. Nominal-level variables simply categorize or name differences.

Like nominal-level measures, variables at the ordinal level are also exhaustive and mutually exclusive. The difference is that at the ordinal level, variables can be ordered by rank, meaning we can say that one attribute of the variable is more or less than another. Examples of ordinal-level measures include social class, degree of support for policy initiatives, and satisfaction. For example, using a Likert-scale question in a quantitative survey to measure the variable from very supportive to very unsupportive, we can say that one person’s support for a new crime prevention policy may be more or less than their neighbour’s level of support. However, there is no way to calculate or quantify the distance between the levels of support, meaning we cannot say exactly how much more or less supportive one neighbour is than the other.

At the interval level, measures meet all the criteria of nominal and ordinal levels; the distinguishing characteristic is that the distance between attributes is known and equal. Temperature and age are good examples of interval-level variables. The distance between 5 years and 10 years is the same as between 15 years and 20 years; because a year is a consistent length, it allows for the mathematical calculation of precise differences.

Finally, at the ratio level, attributes are mutually exclusive and exhaustive, attributes can be rank ordered, the distance between attributes is equal, and attributes have a true zero point. In fact, it is only with ratio-level variables that we can say there is a true zero point; therefore, it is only with these variables that we can say what the ratio of one attribute is in comparison to another. Examples of ratio-level variables include time incarcerated, number of offences, and years of education. For instance, say you were reviewing sentencing decisions in Alberta and measuring total time incarcerated in days as one of the variables. A person cannot be sentenced to both 11 and 21 days of incarceration in total (mutually exclusive), all number of days can be included in the response (exhaustive), days are a consistent unit of measure (the distance between days is equal), and there is the potential that a person will not have been sentenced to incarceration at all (true zero point). Finally, this true zero point allows us to say that one person has a sentence that is “x” times longer than another person’s sentence.

It is important to be aware of the level of measurement for your variables, especially for quantitative research analysis. You can always collapse higher levels of measurement, but you cannot increase the detail in basic levels of measurement. For instance, you can break your interval-level variable of age into categories and create an ordinal-level variable (e.g., age 25 and under, age 26 to 50, and age 51 and over). But you cannot turn the nominal-level variable of province of birth into an ordinal- or interval-level variable, as the detail simply is not there to do so. Certain statistical analyses require variables to have specific levels of measurement to be valid, so it can be good to have these potential analyses in mind when determining your variables to ensure that you have the correct levels of measurement. This will be discussed more in the quantitative analysis chapter (chapter 14). For now, keep in mind these levels of measurement and the differences between them, as shown below in Figure 6a.3.

Conclusion

In this chapter, we delve into the important process in research design that we call “measurement .” It is a process that starts with conceptualization – the assigning of meaning to the concepts we include in our studies and being as clear and precise as possible when we do so. Conceptualization is followed by operationalization, which involves explaining exactly how a concept will be measured, such as what exact survey question will be asked to measure a concept, such as SES, level of education, or number of crimes committed in the past year.

This chapter also discusses the terms “variable” and “attribute” and outlines how variables are related to one another. Next, we discussed the four levels of measurement that a variable can take: nominal, ordinal, interval, or ratio. The importance of recognizing the precise level of measurement of the variables of our study is highlighted, particularly the fact that there is a direct connection between this level and the types of analyses we will later be able to conduct, and, subsequently, conclusions we will later be able to make in the final phases of our research journey.

References

Babbie, E. (2007). The Practice of Social Research (11th ed.). Wadsworth.

Connell, R. W. (1987). Gender and Power. Stanford University Press.

Connell, R. W. (2005) Masculinities (2nd ed.). University of California Press.

Connell, R. W. (2009). Gender: In World Perspective (2nd ed.). Polity.

Kaplan, A. (1964). The Conduct of Inquiry: Methodology for Behavioral Science. Chandler Publishing Company.

Kohm, S. A., Waid-Lindberg, C. A., Weinrath, M., Shelley, T. O. C., & Dobbs, R. R. (2012). The impact of media on fear of crime among university students: A cross-national comparison. Canadian Journal of Criminology and Criminal Justice, 54(1), 67-100. https://doi.org/10.3138/cjccj.2011.E.01

Maxfield, M. G., & Babbie, E. R. (2018).  Research methods for criminal justice and criminology (8th ed.). Cengage Learning.