ABSTRACT
It is important to consider the types of data your data are. There are four general data types: nominal, ordinal, interval, and ratio. Nominal data are distinct categories. Ordinal data are ordered categories: they have a name and an order. Interval data are ordered and have meaningful distances (i.e., equal spacing between intervals). Ratio data are ordered, have meaningful distances, and have a true (absolute) zero that represents absence of the construct. The types of data restrict what options are available to analyze the data. Many analyses approaches assume data that are interval or ratio.
It is also important to consider whether the data were transformed because score transformations are not neutral—they can impact the results. Norm-referenced scores, such as percentile ranks, standardized scores, T scores, z scores, scaled scores, standard scores, scaled scores, and stanine scores are scores that are referenced to some norm or standard of comparison.
