al data collection and/or analysis involves allowing the re searcher to make exact determinations of the distances between two points on a scale. This is because interval data collection and/or analysis is based upon points on a scale having fixed and equal intervals. For example, when an item costs $50.00, the re searcher can say that it is twice as expensive as an item costing $25.00. As another example, mortality rates, the number of pa tients treated, age, crime rates, years of education and income are all considered interval level measures. Of interval measures, Welch and Comer (1985) note the following.
These kinds of measures give us the most information be cause they allow us to classify, to order, and to array along a clearly delineated scale. Because they are pre cise numerical measures, the most powerful statistical techniques are appropriate to use with them (p. 36).
Welch and Comer (1983) continue that ratio data, which is quite similar to interval data, refers to the presence of absolute and fixed zero points, such as in percentage or proportion measures (e.g., percent urban versus percent rural, percent female crime rate per 1,000 population, proportion of green apples to all apples grown in a given geographic region, etc.). The use of ra tio data collection and/or analysis allows the researcher to clas sify/categorize, order, assume fixed intervals, and state an absolute zero point.
Criteria for Data Collection and/or Measurement
Welch and Comer (1987) report that there are several criter ia for the measurement of data whether it is collected or analyz ed. The first criterion is validity, which refers to whether or not the measure is measuring the target. Validity refers to the measure (e.g., item, question, behavior, etc.) measuring what it is supposed to measure. For example, a bathroom weight scale that accurately measures the weight of a person is valid. While the researcher's testing for vali...