Microsoft Word research method fm doc

 
In general, operational definitions of variables are used in order to: 
ƒ
Avoid ambiguity 
ƒ
Make the variables to be more measurable 
 

---

Research methodology 
39
Scales of Measurement
As part of the process of clarifying each of the variables to be studied, its scale of 
measurement should be specified. There are four types of scales of measurement: Nominal, 
Ordinal, Interval and Ratio. 
They are listed in ascending order of power and preference. 
1. Nominal Scale: This consists of two or more named categories (classes) which are 
qualitatively different from each other. 
E.g Sex: male (1); Female (2) 
Marital status: 1. Married 2. Single 3. Divorced 4. Widowed 
2. Ordinal scale: This has the additional quality that the categories are ranked and have 
implied order. However, the intervals between classes are not necessarily equal. 
Example 1. Severity of a disease: Severe (grade III); moderate (grade II);
mild (grade I); absent (grade 0). 
Example 2. Educational status: 0; 1-6; 7-8; 9 -12; more than 12. 
3. Interval scale: This has the additional quality that the intervals between classes are 
equal. 
Example
: Temperature (in Celsius) 
Equal differences between any pair of numbers in the scale indicate equal differences 
in the attribute being measured. The difference in temperature between 20
… C and 
25
…C is the same as the difference between 30…C and 35…C. The ratio between 
numbers in the scale is not, however, necessarily the same as that between the 
amounts of the attribute. That is, a room at 30
… C is not 'twice as hot' as one at 
15
…C. This is because the zero on the scale does not indicate absence of the 
attribute.