Page 108 - SREENARAYANAGURU OPEN UNIVERSITY
P. 108

easy data analysis.

                                             Ordinal scale
                                             The ordinal scale works on the concept of the relative position
                                           of objects or labels according to the choice or preference of the
                                           individual. It ‘orders’ the members of a population in ranks, from
                                           highest to lowest or lowest to highest. Although each value is
                                           ranked, there is no information that specifies what differentiates

                                           the categories from each other. Range values  do not possess the
                                           arithmetic  property  of  equal  intervals  or  true  zero  points. This
                  ◆ Description and        means that, with ordinal data, we only know that one member of a
                order
                                           population has more or less quality than another member, but we
                                           do not know how much more or less. Ordinal scaling maintains
                                           descriptive qualities along with an intrinsic order, but it does not
                                           have a scaling origin and therefore the distance between variables
                                           cannot be calculated. The descriptive qualities indicate labelling
                                           properties similar to the nominal scale, in addition to which, the
                                           ordinal scale also has a relative position of variables.

                                             In nutshell ordinal scale has the following properties:

                                                ◆ A rating scale in which numbers are assigned to objects to
                                               indicate how well objects possess certain characteristics.
                                                ◆ Can determine if an object has more or less of a characteristic
                                               than another  object,  but not how much more  or less (1 ,
                                                                                                        st
                                                nd
                  ◆ No distance and            2 ….).
                origin
                                                ◆ Any series of numbers can be assigned preserving ordered
                                               relationships between objects.

                                                ◆ In addition to the counting  operation  allowed for nominal
                                               scale data, ordinal scales allow the use of statistics based on
                                               percentiles, for example, percentile, quartile, median (such as
                                               top ten percentiles or the 1  decile).
                                                                         st
                                             For example: A software company may need to ask its users:

                                             How would you rate our application?

                                                   ◆ Excellent
                                                   ◆ Very Good

                                                   ◆ Good
                                                   ◆ Bad
                                                   ◆ Very bad


                                             Interval scale

                                             Interval  scales incorporate  all the properties of nominal  and
                                           ordinal scales and also indicate the distance or interval between




                 102     SGOU - SLM -MCom Research Methodology
   103   104   105   106   107   108   109   110   111   112   113