vious research title, the phrase “test anxiety” is the dependent variable.
3. Confounding or Extraneous Variables. Those variables are usually indicated in
experimental research. They are not included in the study but in one way or another cause an effect on the dependent variable. Most of the time, these are the variables that are minimized to lessen their impact on the expected response. Going back to the previous research example, the family background of the grade school student is an example of an extraneous variable. As you may notice, it is not included in the variables under the study. But each grade school student who will take part in the study has a family background that may likely affects the impact of the peace-loving learning environment variable. To lessen its effect, the researcher must control it. 4. Categorical Variables. Those that characterize and describe the quality of data. They are often classified into mutually exclusive categories and extensive categories. The former comprises those that take on specific values. Often, they do not follow a sequence. They just want to describe the data into the given options; thus, they are called nominal variables. For example, civil status can be classified only as single, married, widowed, or annulled. You can only be either one of these. Extensive categories, on the other hand, characterize data using a definite range. They consider logical order or rank and are therefore called ordinal variables. Description of data follows a sequence. For example, size is classified into small, medium, or large. When you state the classification sizes, you have to say it in order. Noticeably, the values of categorical variables are not numeric. 5. Continuous Variables. Those variables in which values are based on a given interval or continuum. These variables rely on numbers for description. For example, normal temperature ranges from 36°C to 37°C. So, the description of normal temperature is only within the given interval. Think about income example. When you state an income, you can just say the amount. It does not follow an interval, but it falls along a continuum. 6. Quantitative Variables. Those variables give details regarding the number or level of something. These variables count the frequency of responses or effects. An example is a popularity contest. The contestant who will get the highest number of votes as the most well- liked will be declared the winner. 7. Qualitative Variables. Those variables represent kinds or types of objects. They are synonymous with categorical variables. They are often categorized into names, labels, or groups. At times, numbers are used as codes to represent categorization to be able to measure the variable and describe its extent. Responses such as yes or no; few, many, too much; and agree or disagree, are some examples of qualitative variables. VARIABLES AND THEIR LEVEL OF MEASUREMENT As mentioned, variables must be measured based on the way they are defined operationally. Levels of measurement are the association of the values that are linked to a variable. Here are some reasons why studying them is important: 1. Proper interpretation of data related to the variable. For example, sex refers to male or female, and the study provided the number of male and female participants. Then, the idea being conveyed will be easily expressed. It is just about the quantity of the participants based on the two categories. 2. Decisions about the proper statistical analysis to be used. If the measure is the quantity of males vs. females, you do not need to test the relationship between them. Rather, you should just take note of the frequency and the average of males and females. The following are the four levels of measurement: 1. Nominal Scales. Concerned with the names and categories of responses. They do not intend to quantify the data. Qualitative and categorical variables are examples. Nationality and hair color are some examples. 2. Ordinal Scales. Used for data that intends to be ranked. This scale is expressed through sequential and numerical order and therefore allows a comparison of degree. Qualitative and sometimes quantitative variables are measured using this scale. Some examples are the following: first, second, and third; good, better, and best. 3. Interval Scales. Use equal units of measurement and intervals to know the distance between them more than the sequence. An interval scale does not use zero as its base point, though. 17
COX, Gary (2004) Lies Damned Lies and Rational Choice Analysis in SHAPIRO I SMITH R. and MASOUD T Eds Problems and Methods in The Study of Politics Cambridge PDF