INSTRUCTOR: Welcome to Enrichment Presentation Number 4. In this presentation we're going to take a look at variables, because variables are the units that we study through measurement.

A variable is any attribute that can assume different values. Variables can be simple, such as height, weight, sales, or color. Or they can be more complex, such as health, intelligence or leadership. Variables can be classified as independent or dependent.

Independent variables can be further defined as either a treatment or a classification variable. A treatment variable is a variable that is controlled or manipulated by the experimenter or researcher.

Let's say I'm interested in how wall color affects mood. Wall color is the treatment variable that I will manipulate by placing people in rooms with three different colors. A classification variable, on the other hand, is some characteristic that was present prior to the study. So it's something that I'm going to gather data on.

Let's say that I'm interested in sales among retail stores. So I decide to compare four different geographic regions. The geographic region, then, is a classification variable.

Independent variables can be broken down further into levels or conditions. In the wall color study, the four colors that I use-- pink, white, and blue-- are considered the levels or conditions. In the study of retail sales, I can break down the geographic regions into large, medium, and small stores. Or I could choose inner city, suburbs, and small towns.

A dependent variable is the response to the independent variable. It is the measurement of the variable under study. Or, in other words, it is the thing that I'm studying, that I'm interested in. It may or may not change as a result of our manipulation, and this is what our data will tell us. In my wall color mood example, mood, then, is the dependent variable. In my retail example, sales is the dependent variable.

At times variables aren't straightforward and easily defined, like color or sales. Variables that could be considered complex include things like beauty, fear, intelligence, and even productivity. Because they have some subjectivity, they require us to define them further by providing what we term an operational definition. In other words, change them into something that is more easily measured. So for each complex variable that I study, I need to operationalize it.

An operational definition is a means for indirectly measuring and defining a variable. It specifies a defined procedure for measuring an external recordable behavior or event and uses these measurements as a hypothetical construct. In other words, I look at some behavior, and I use that to define the variable under study.

For example, I use an IQ test to measure intelligence. I might use heart rate or heart rate plus some behavior, such as running away, to measure fear. I might use a survey to measure happiness. The responses given on the survey, the heart rate, the answers given in the IQ test are observable and recordable and are understood and accepted as measures of the variables.

We can further classify variables by how they are measured. A nominal variable is used to name something or identify a characteristic. They're qualitative rather than quantitative. Examples might be geographic region, a political affiliation, gender, or even college major. We label these things and give them a name. So that's a hint on how to remember it. Nominal are names.

In ordinal measurements, these are measurements that are ranked sequentially and given an assigned meaning, such as low performing versus high performing, or small, medium, large. It's a name that has a bit more meaning to it.

An interval measure means that there are equal distances between the measures. Examples can include inches on a ruler, degrees of temperature, or minutes on a clock. A ratio measurement is similar to interval but includes a zero, and we can multiply and divide numbers on a ratio scale-- height, weight, temperature. So something can be twice as tall, twice as heavy, or twice as hot. That can't be done without a zero reference.

So what scale should we use for our study? Well, nominal scales can only tell us that a difference exists. Something can be pink, blue, or green. It's different. An ordinal scale can tell us the direction of the difference. Something is larger versus smaller or high versus low. But with an interval scale, we can determine the direction and magnitude of the difference.

However, a ratio scale allows us to determine direction, magnitude, and the ratio of the difference. For example, a 1% increase in exercise could lead to a 2% decrease in heart disease.

For this week's assignment, describe a simple experiment that includes an independent and a dependent variable. This can be anything-- work-related, school-related, something from home, something from nature. Describe where you will manipulate an independent variable and you hope to see a change in a dependent variable.

Then describe a complex variable and explain how you would operationalize it. Maybe you use leadership or productivity or intelligence. Provide a paragraph for each, and let us know if you have any questions.