OMBC 106 Unit 5: Attitude Measurement and Scaling

Measurement

• Measurement is the process of assigning numbers or symbols to characteristics of objects according to certain pre-specified rules.
• It involves measuring characteristics of objects, not the objects themselves.
• Numbers are used to provide an accurate description and facilitate statistical analysis and communication of results.

Scaling

• Scaling is an extension of measurement that creates a continuum on which measurements are located.
• It involves creating a scale to measure characteristics, such as satisfaction levels.

Types of Measurement Scales

• There are four types of measurement scales: nominal, ordinal, interval, and ratio.
• The choice of measurement scale has implications for the statistical technique used for data analysis.

Nominal Scale

• Nominal scale is the lowest level of measurement.
• Numbers are assigned for identification purposes only.
• Objects assigned a higher number are not superior to those with a lower number.
• Each number is assigned to only one object, and each object has only one number assigned.
• Examples of nominal scale include:
  • Religion (Hindu, Sikh, Christian, etc.)
  • Gender (male, female, other)
  • Caste
  • Respondents
  • Marital status
  • Brands
  • Attributes
  • Stores
  • Players of a hockey team
• Assigned numbers cannot be added, subtracted, multiplied, or divided.
• Only arithmetic operations that can be carried out are count of each category.
• Frequency distribution tables can be prepared, and mode of distribution can be worked out.
• Chi-square test and contingency coefficient can be used with nominal scale variables.

Operations with Nominal Measures

• Percentage of numbers can be calculated (e.g., 40% Male and 60% Female).
• Cannot be ranked (e.g., First Turk, Second Arab, Third German, etc.).
• Arithmetic means cannot be calculated.