Вы находитесь на странице: 1из 2

Measurment in Research Written By Tasaddaq H. Qureshi (Pakistan) E-mail: tasaddaqqureshi@yahoo.

com

LEVELS OF MEASUREMENT
Researchers are supposed to choose measurement levels on the basis of their isomorphism (structural similarity) with the variables being measured. Because statistical operations (data analysis) depend on the level of measurement. 1: Nominal: In nominal measurement numerals are used to classify persons, objects, or characters. Example, classification of media users on the basis of the medium (radio, TV, newspaper) they use for news. (1) radio, (2) TV, (3) newspaper Nominal level is the weakest measurement level. Properties of nominal level are as: 1. Equivalence: Properties of objects in the one category are equal to each other. a = a, and if a=b then b = a. 2. All categories are exhaustive and mutually exclusive. Means that each measure accounts for every possible option, and that each measurement is appropriate to only one category. 2: Ordinal Level: Objects measured on ordinal level are ranked in a meaningful way, from smaller to greater. Example, categories of socio-economic status may be like (1) lower, (2) lower middle (3) middle, (4) upper middle, (5) upper. Here in the example a rank of 1 is assigned to lower, 2 to lower middle, 3 to middle, 4 to upper middle, and 5 to upper. The assignment of numerals has some mathematical meaning. 1 is lower than 2, 3, and 4 and 5. Similarly 2 is greater than 1 but smaller than 3, 4 and 5. However, ordinal level does not show the distance between two categories. It is therefore, compared with a horse race without a stopwatch. Properties of the ordinal level 1. Equivalence 2. Order among the categories objects in the same category are identical but the relation among them is asymmetrical. If a is greater than b, then b is not greater than a. However, if a is greater than b, and b is greater than c, then a is greater than c.

3.

Exhaustive and mutually excusive.

3: Interval level: When a scale has all the properties of an ordinal scale and the intervals between adjacent points on the scale are of equal value, the scale is at interval level. In addition to the intervals between two adjacent categories the interval scale contains properties of nominal and ordinal measurement levels. Example, temperature. The same amount of heat is required to warm an object from 30 to 40 degree as to warm it from 50 to 60 degrees. Interval scales lack a true zero point. 4: Ratio Level: In addition to possessing all the properties of an interval numerical system, the ratio measurement scales have a natural zero point. Properties that have natural zero points such as weight, length, area, etc can be measured on the ratio level of measurement. In short we can write it in short , as:

Levels
Nominal Ordinal Interval Ratio

Magnitude no yes yes yes

Interval no no yes yes

Absolute Zero no no no yes

RELIABILITY AND VALIDITY IN MEASUREMENT Reliability is equated with a measuring instruments consistency. If the same scale is administered repeatedly to the same individuals and it yields roughly the same set of responses, the scale is said to be reliable. Validity: measuring instrument is said to be valid if it fully and accurately measures the variable that it supposed to measure.

Вам также может понравиться