Data Collection Quantitative Methods

Quantitative data is data in number form. Deciding on a method of data collection requires knowledge of the data type you're collecting. If you know beforehand what data analysis you will use, you can design an appropriate method of quantitative data analysis. By first deciding the data type, you will guarantee that both your data collection method and your post-collection analysis are appropriate.

Nominal data takes the form of categories, even though it may seem as if categories in the form of words is not quantitative data. These categories should not have a hierarchical relationship. For example, race is a form of nominal data that takes on values such as “Caucasian,” “Asian” and “Hispanic.” Because there is no hierarchical relationship between these races, race is nominal data. Nominal data collection takes place through categorization, which may be done by the researcher or the subjects themselves. Any method of categorizing that yields well-defined categories may be used in nominal data collection.

Ordinal data are similar to nominal data in that they fall in categories. The difference is that ordinal data has a hierarchy. For instance, weight categories in boxing or other competitive sports are examples of ordinal categories. The collection of ordinal data is similar to that of nominal data -- researchers must employ classification methods. However, because of clear “value” differences between categories, researchers may also employ other methods of analysis, such as weighing a person on a scale.

Interval data takes the form of numbers. These numbers can be integers or decimals, positive or negative. However, for data to be true interval data, data points with value “zero” must not be special in the data set. That is to say, “zero” means nothing for interval data. One example of such data is temperature measured in Fahrenheit. In Fahrenheit, “zero” implies nothing. Researchers measure interval data through direct measurement with the proper tools. Psychological assessments, such as IQ tests, also belong to the field of interval data collection.

Ratio data is interval data extended to a number line in which “zero” has meaning. An easy example of such data is temperature measured in Celsius. Fahrenheit measurements are directly mapped onto the Celsius scale. However, now the number “zero” has meaning: it is the temperature at which water freezes. The measurement of ratio data is similar to that of interval data. However, researchers must modify or create new scales if they want the number zero to have interpretable meaning.