Scientific Variables

Scientific Variable

Introduction

In general, all of the features of the observation units are too large and not entirely known to be completely described. Moreover, man has limited power of observation. Thus, it is necessary to limit what must be observed to the relevant characteristics of the units of interest (Hoover & Donovan, 2010). Some characteristics can be observed with the use of simple tools to aid the senses, like a scale, a ruler and a test tube. However, many others require the use of more sophisticated instruments of observation, such as a stethoscope, a microscope, a telescope and other instruments of the vast and formidable scientific paraphernalia. The instrumentation of the observation process was fundamental to the development of the scientific method.

The representation (or modeling) of units (systems), requires the quantification and description of its relevant features and its registration in written form (Kuhn, Pease & Wirkala, 2009). For descriptions have the same meaning for all scientists should be adopted definitions, notations and conventions appropriate and accurate. For this reason, the various sciences create their own language.

It sometimes depends on the observation of some degree of subjective judgment. For this reason, it is important to establish the conditions of observation to avoid bias of the observer. Often elaborate strategies should be established to allow the observer to avoid his own trend and get the correct record of the facts.

The measurement of a feature is an attempt to find a matching one-to-one between the alternative and the characteristic values of a given set, that takes into account the relationships between these alternatives and the operations that can be performed on them. Rarely complete matching can be obtained; usually obtained by a partial match (Ynalvez & Shrum, 2011). Thus, variables and features are not synonymous, the concept of feature has a higher abstraction since there is no need to refer to the scale of measurement or the measurement units. In other words, the same characteristic can be measured in different ways resulting in different variables.

A scientific variable associated to each alternative a characteristic value within a predefined scale of measurement. The rule of correspondence between the alternatives of a feature and set values that defines a scientific variable determines a measurement scale (Leek, et. al., 2010). The setting range depends on the assumptions regarding the rules of this transformation. A scientific variable whose values designate categories, or classes, in which the units are classified is called categorized scientific variable. These variables are, by definition, a finite number of values.

Numeric variables are variables whose values are numbers on which you can perform at least some arithmetic operations. Usually variables with interval scale and variables with rational scale fall into this category. Discrete variables are variables such that all values can be placed in correspondence one-on-one with a subset of the set of integers (Malenovský, et. al., 2012). Continuous variables are variables where this correspondence is not possible. A continuous scientific variable, therefore, can take any value in the valid ...