As stated in Section 1-2, a fundamental precept of science is empiricism: in order to test a claim, scientific researchers must make direct observations. In the previous section, you learned that researchers avoid subjective definitions of concepts because such definitions can cause problems for the reliability and validity of observation. For example, the concept of a hot classroom might be defined subjectively as a classroom in which a person feels hot. This definition probably will lead to observations of low reliability (repeatability). For example, a person might feel hot in a classroom after running there from her car on a warm day. However, after sitting in the room for several minutes, she probably will no longer feel hot because her body has cooled down. And her observation that the room felt hot also would have low validity (meaningfulness): it wasn’t the temperature of the room that she was observing, it was the temperature of her body that made her feel warm. An objective definition of a hot classroom would have avoided these problems.
Objective definitions specify phenomena that are publicly observable — phenomena that can be perceived in virtually the same way by anyone. Descriptions of publicly observable phenomena are relatively unaffected by a particular person’s beliefs, preferences, attitudes, emotional states, etc. Because the observations don’t depend on how a particular observer perceives a situation, they can be verified (shown to be true or correct) by others. For example, the concept of a hot classroom might be defined objectively as a classroom in which the temperature, as measured by a standard room thermometer, is ≥ 82º F (28º C). Or the concept of superior intelligence might be defined as a score of 116 or higher on the Stanford-Binet Intelligence Scale. Course grades are usually defined objectively. If the maximum number of points a student can earn is 1000, the cut-offs for the various grades might be objectively defined as follows: A (superior) = 900-1000, B (above average) = 800-899, C (average) = 700-799, and D (below average) = 600-669. When calculating final grades, therefore, an instructor doesn’t need to rely on subjective impressions of his students. Any competent observer can determine if an assigned grade is correct: the only information needed to do so is the total number of points earned by a student. Of course, the grading of individual assignments, such as term papers, might be influenced by an instructor’s subjective impressions; but once scores have been given to each assignment, the overall course grade is defined objectively.
It should be clear that objective and subjective definitions involve very different ways of using observations to give meaning to concepts. In the case of objective definitions, the observations made by individuals are not affected by their biases, preferences, attitudes, emotional states, prior experiences, etc., whereas they are in the case of subjective definitions. Objective definitions also differ from dictionary definitions. A dictionary definition is a statement that describes a concept in terms of other words that refer to the same thing, whereas an objective definition is a statement that describes a concept in terms of a specific set of observations. For example, the dictionary definition of empathy is, “the ability to share in another person’s emotions or feelings” (Webster’s New World Dictionary, Second College Edition). If you were performing a scientific study of empathy, this dictionary definition isn’t very helpful: it provides little information about how to measure empathy. An objective definition of empathy, on the other hand, would specify a set of observations that allow us to measure this concept. For example, we might objectively define empathy in infants as “the amount of time it takes to begin crying while listening to an audio-recording of another infant crying” (e.g., Martin & Clark, 1982).
Objective definitions, therefore, allow scientific researchers to empirically test claims in precise ways that can be repeated by other researchers. Because of this, objective definitions have two major advantages when compared to subjective definitions:
- Conceptual Clarity. If different researchers use the same objective definition of a concept, they must, by necessity, be studying the same thing. This advantage is especially important when researchers study complex and “fuzzy” concepts like intelligence.
- Verifiability. A verifiable observation is one whose accuracy can be checked by others. The results of studies that use objectively defined concepts are relatively easy to verify because the observations required to replicate the studies are specified in detail.
Objective Definitions and Validity
As you learned in the previous section, the term validity refers to the degree to which a set of observations allows us to say something meaningful about the concept being studied. For example, a scale for measuring weight provides valid observations if the number of pounds (or kilograms) indicated when a person steps on the scale closely reflects that person’s actual weight. That is, if the scale indicates that a man weighs 180 lbs. and he actually does weigh 180 lbs., then the scale allows us to say something meaningful about the man’s weight (i.e., its measures accurately what it was designed to measure). On the other hand, if random numbers appear whenever the man steps on the scale, the observations (the number of pounds indicated by the scale) are not valid.
But objective definitions also introduce some problems for validity:
- Problem 1. The observations specified by an objective definition measure not only the concept being investigated, but also other concepts.
- Problem 2. The observations specified by an objective definition measure only some aspects of a concept.
The first problem for objective definitions is analogous to what happens to a person taking an antidepressant medication. By changing activity in particular areas of the brain, antidepressants improve mood; but they also affect other areas of the brain, which causes side effects. Just as an antidepressant is designed to improve people’s moods, the observations specified by objective definitions are designed to help researchers learn about some phenomenon. For example, if a teacher is interested in measuring students’ knowledge of the course material, she can objectively define this concept as scores on six multiple-choice tests. These observations should allow the teacher to get a good idea of how much each student knows. However, there are “side effects” of these observations: students’ test scores are affected by things other than their knowledge of the course material — things such as test anxiety, word comprehension, reading rate, distractibility, and “test-wiseness.” Because of these “side effects,” one person may get 65% correct because he suffers from severe test anxiety; another person may get 65% correct because she was distracted by a student who kept tapping a pencil; and a third person may get 65% correct because he’s a slow reader and ran out of time. Because test scores are affected by factors in addition to knowledge of the course material, the scores earned by these three students may not indicate well how much they really knew.
The second problem for objective definitions is most important for complex concepts, which are the norm in psychology. Ppsychological concepts (e.g., intelligence or empathy) refer to a complex cluster of cognitions, emotions, and behaviors that cannot all be measured by a single objective definition. For example, shyness is a personality trait characterized by avoidance of social situations due to a fear of being evaluated negatively by others. Assuming that shyness varies from severe (a person avoids interacting with anyone other than one or two intimates) to virtually absent (a person avoids few social interactions), researchers might objectively define shyness as the total amount of time a person avoids eye contact in a 5-minute conversation with a stranger. This definition would allow them to precisely measure a particular aspect of shyness — discomfort as indicated by lack of eye contact in one type of social interaction — but it should be obvious that shyness is a concept that is much more complex than this single definition can capture.
The best way to minimize the effects of these two problems is to use several objective definitions of a concept. Each objective definition, of course, measures (a) only some aspects of a concept and (b) concepts other than the one being studied. But by combining the results from several sets of observations, each using a different objective definition, we should get a better understanding of what concept or claim we’re investigating. For example, a teacher who wants to measure students’ knowledge of the course material, which is a complex concept, might choose a few measures from a list of possibilities (multiple-choice tests, essay tests, short quizzes, research papers, oral reports, class participation, reaction papers, etc.). In combining the results from several measures, the teacher should now get a good idea of how much each student has learned in the course.
Objective Definitions and Testable Claims
If a researcher is unable to develop a good objective definition of a concept, then any claim that includes that concept is untestable. In other words, a claim cannot be tested in a reliable and valid manner unless its concepts are linked to observable and verifiable events. The research on N-rays is a good example of this: the observations used by Blondlot to measure the presence of N-rays were subjective (he simply judged whether an object seemed brighter to him), which allowed personal factors (such as the desire to prove his claim about N-rays to be correct) to affect the results of his research. Most psychological concepts can, with some ingenuity, be objectively defined. Nevertheless, some concepts, such as evil or free will, probably can never be objectively defined and, therefore, any claims we might make about these concepts are likely to be affected by personal factors (e.g., prior beliefs, values, and desires).
Study Questions for Section 2-4
- How would you define “objective definition” in your own words?
- What is an example of an objectively defined concept from your everyday life?
- What is an objective definition of “knowledge of the course material” other than the ones mentioned above?
- What would be an example of an objective observation for each of the following:
(a) amount of intelligence
(b) amount of hunger
(c) degree of interest in a visual stimulus
(d) level of sexual desire
- What characteristics might an objective definition of “bad eyesight ” contain?
- What are the advantages of using objective definitions in research?
- What are the disadvantages of using objective definitions in research?
- How can researchers reduce problems introduced by these disadvantages?
- A scale for measuring weight would provide invalid observations if random numbers came up each time a person stepped on it. In this case, would the observations have high or low reliability? Why?
- What is an example of a testable psychological claim that you have learned in this course? What makes it a testable claim?
- What is an example of an untestable psychological claim that you have learned in this course? What makes it an untestable claim?
Practice Quiz for Section 2-4
Click HERE to find the details about the articles, books, etc., referred to in this section.