Variables, Marker

Marker variables are variables that are used to indicate some other feature. Often, the variable of interest is not directly observable; instead, a marker believed to indicate the existence or level of the variable is used. For example, often the use of biological gender (male and female) is used to represent differences in socialization that a child experiences. Consider a person stating that he or she has the “flu.” If one asks them, “How do you know that?” the person may list a number of symptoms (markers)—raised temperature, body aches, fatigue—all of which are indicative of someone infected with an influenza virus. The person has not directly tested or provided direct proof that the condition is the flu, but the set of [Page 1850]markers that exist permit a reasonable inference that the person is suffering from the flu.

The reason that biological gender serves as a “marker” variable is that no direct measurement exists for the process of socialization. Instead, the assumption made about gender involves differences between how little girls and little boys are socialized. The belief that the process is different creates an expectation that the mark or measure of the difference can be inferred or realized on the basis of measuring the biology of the person. The assumption of the feminization of girls and the masculinization of boys by the social system serves as the reason for expecting gender differences between men and women.

A similar event occurred when considering the original measurement of human immunodeficiency virus (HIV). The measures for the existence of the virus did not measure the level of virus in the body; instead, the measures tested for the existence of the antibodies to the virus found in the blood. The argument for the use of this measure is that only a person exposed to the virus, with the virus in the blood, would have generated any antibodies to the virus. The use of a marker variable is a means to indicate a feature not capable of direct measurement but believed to exist. According to this theory, the presence of particular antibodies in a person indicates exposure to a particular virus. HIV is a virus for which there is no known cure, so a person is considered actively infected with the virus if antibodies are present.

Marker variables also act as a means to make inferences, much like tracking wild animals whereby a person tracking animals relies on marks left by the animals to determine what kind of animal, in which direction it is moving, how recently it moved through the area, how big it is, and other features of the animal and the circumstances. The question is always how indicative or accurate are the marks left as interpreted by the person making the observation. The determination of whether the marker indicates the outcome remains an important element of examining the underlying relationship. The less accurate the marker, the less is one’s ability to make accurate claims about the relationship that is directly sought for measurement.

...