LotsOfEssays.com

Basic Concepts of Data Collection and Analysis

Anastasi (1988) and Welch, and Comer (1983) report that there are several levels of measurement relative to data collection and analysis. These levels include, but are not limited to, nominal, ordinal, interval and ratio; and carry with them important limitations due to the need for differing statistical techniques. The first level, nominal, refers to a means of classifying the data. For this type of data collection and/or analysis, there is no order, in that the units of analysis are merely classified into two or more categories. As an example of nominal data, consider that all white houses would be categorized together, while all red barns would be placed in another category. The primary limitation of nominal data is that there is no order to the given units of .analysis; just classification. Further, there is no weight to nominal data. As a result, the researcher could not present data in relation to weight, or value; just classification. Statistical techniques are less sophisticated here. The second level of analysis is ordinal, which refers to ordered data. With this level of analysis, the units of analysis are placed in the indicated order. As one example or ordinal data, consider that attitudes about the use of nuclear plants would be placed in order, ranging from "strongly favor," to "favor," "depends," "do not favor," and to "strongly do not favor." With the use of ordinal data, there is both classification and order. The fact that there is order allows for the weighting of subject responses. Thus, ordinal data allows the researcher to state that one item is more than, or less than, another. The primary problem with ordinal data, however, is that the researcher is not allowed to assign weights that allow comparison based on intervals. This is because ordinal data do not facilitate intervals.

Anastasi (1988) and Welch and Comer (1983) report that interv...