Chapter 3 Operational Definitions & Measurement
3.1 Designing Research
We saw from the last section that conducting a research study involves forming a hypothesis, collecting evidence to confirm or disconfirm the hypothesis, and then interpreting the evidence. Imagine you wanted to see if a placebo (a treatment with no effect) would cause people to experience less pain. This was the question of David J. Scott and his colleagues (2007 ). The study involved injecting participants (with their informed consent) with a saline solution that caused pain. Participants were given either fake pain reliever or no treatment. To support the claim that the placebo reduces pain, the placebo participants should report lower pain than the non-placebo participants. Pain was measured using self-report surveys. Let’s look at the building blocks of this study.
3.2 Constructs versus Measures
The first concept is what the research is about. There is an important distinction between constructs and measures. A construct is a “concept, model, or schematic idea” (Shadish, Cook, & Campbell, 2002, p. 506). Constructs are the big ideas that researchers are interested in measuring: depression, patient outcomes, prevalence of cumulative trauma disorders, or even sales. For constructs in the social sciences, there is often disagreement and debate about how to define a construct. To do science, we must be able to quantify our observations (collect data) on the constructs. To go from a construct (the idea) to a measure requi res an operational definition. An operational definition describes how a construct is measured.
Constructs are what the study is about. The example study is about placebos and the reduction of pain. It isn’t really about saline solution or the Total Mood Disturbance measure as described in the article (Scott et al., 2007). The constructs of interest are placebos and pain. Pain was measured using the Total Mood Disturbance measure. Placebos were manipulated (the researcher controlled which participants were given a placebo and which were not).
3.3 IVs and DVs: Variables in Your Study
Another term for the measure in a study is the dependent variable (DV). Researchers look for a change in the DV that is due to a manipulation (the administration of the placebo or none). We call the manipulation the independent variable (IV). A quick mnemonic (memory aid) for the IV is that it is the variable that “I control”. The IV is also sometimes called the treatment. Researchers look for IVs (the causes) that cause changes in DVs (the effects). Thus, if you are designing a strong study, you want your IV and DV to be strongly related to each other.
So far, we have seen that studies have constructs, at least an IV and a DV. Another term for DV is dependent measure or outcome. All studies need an operational definition that explains how the DV construct is represented as a measure.
But what about the IV? The researcher manipulated the IV; they did not measure it. The construct behind the IV in this example is the placebo. Studies also need an operational definition that explains how the IV construct is represented as a manipulation. Here, the placebo was manipulated by creating two groups; one received the placebo and the other one did not.
Do you see the pattern? Studies exist at two levels. The construct level describes the themes of the study. Constructs are how researchers tie studies together. If you were reading research reports on this topic, you would probably look for “placebo” and “pain.” You would not search for “sugar pill” and “Total Mood Disturbance Measure.” The second level is the measurement level (more generally, the operation level). The operation level is exactly what happened in the study. Constructs are what we investigate, operations are what we do.
Psychologists are operationalists because they use study operations to represent constructs of interest. Is it possible for two psychologists to disagree on the link between study operations and constructs? Yes, this happens all the time. What if participants did not believe they were taking a “real” pain pill? Or, what if the sugar pill actually had effects on pain? Psychologists do argue about whether study operations are a good match for study constructs (this concept is called construct validity, and we’ll revisit it later). But psychologists understand that there is no way to perfectly capture a construct using a measure. If we had to perfectly agree on all measures for all constructs, we would be essentialists. Psychologists also understand that we do not have access to constructs except through study operations. Thus, we don’t argue about the “true nature” of constructs (which would be essentialism). We define constructs based on the measures we use to capture them (which is operationalism).
3.4 Other Variables: Samples and Populations
What is the role of the cause of the pain in this study? You’ll notice it is neither a DV nor an IV. It is best described as part of the study’s setting. Researchers must also make decisions about the settings they represent in their study. Therefore, the setting of the study is another source of constructs. Finally, the participants in the study are also a construct. Who is the study about? This is the population of interest. Because most studies are about large populations, the study is conducted with a sample, a subset of the population. Again, researchers draw conclusions about the study constructs (the population) through observation of study operations (the sample).
Now that you can see the difference between constructs and operations, we will look closer at how we measure.
3.5 Classifying Measurement Scales
We can classify measures in three ways: according to their level of measurement, whether or not they are continuous or discrete, and whether they represent qualitative or quantitative data.
3.5.1 Level of Measurement
A stair diagram is used because higher levels of measurement satisfy all the requirements of the levels below.
Ratio scale/ratio measurement. Examples: weight, length
Interval scale/interval measurement. Example: Fahrenheit temperature
Ordinal scale/ordinal measurement. Example: the order in which people finish a race Nominal scale/Nominal measurement. Example: gender
Notice that these levels are stair steps. Each level has all the characteristics of the level below it. So interval scales meet all the requirements of ordinal and nominal scales as well (plus they meet the additional requirement for interval scales).
To determine the level of measurement, ask yourself these questions:
- Can you rank/order the numbers? (if no, nominal scale. if yes, keep going) example: kinds of fish. can you rank halibut and mullet? (no, nominal scale) example: Olympic medals, can you rank gold, silver, and bronze? (yes, keep going)
- If you add/subtract the numbers, does the result have meaning? (if no, ordinal scale. if yes, keep going) example: 30 degrees F plus 10 degrees equals 40 degrees (yes, keep going) example: 1st place plus 2 equals 3rd place? (no, this doesn’t make sense, ordinal scale)
- Does the score have a value of 0 that means ‘none’ or ‘nothing’? (if no, interval scale. if yes, ratio scale) example: counting people; 0 people means no people (yes, ratio scale) example: 0 degrees F means no heat? (no, interval scale)
Continuous or Discrete
Separately, decide if your variable is continuous or discrete. If you can have an infinite number of fractions of a value, it’s continuous. If you cannot, the measure is discrete. example: 5 yards, 5.0005 yards, 5.5 years, and 5.500001 yards are all valid measurements (continuous) example: Olympic medals; the measurement between gold and silver does not exist (discrete)
There may be instances where a grey area exists; at some level, all variables are discrete. For example, you could subdivide a measurement of length down to the molecule. At that point, you cannot have fractional values. Try to avoid over-thinking this issue. If you can reasonably talk about fractional values (half seconds; twenty-five cents are a fraction of a dollar) then the measure is continuous. If you cannot (there is no such thing as half a dog or an eighth of an employee), then the measure is discrete.