Session 8: Measurement and Scaling
Target Population
Collections of elements or objects that possess the information sought by researcher.
Double Sampling (PT)
Randomly select each element one time and then repeat for a two-times sample.
Cluster Sampling (PT)
Randomly select from specific clusters. On each selected cluster, randomly select one or two elements.
Non-Probability Sampling
- Rely on researchers personal judgement - They do not permit an objective evaluation - Not possible to project in a statistical way
Probability Sampling
- Sampling units are selected by chance = random. - Every element can be pre-specified - Precision and interferences can be detected
Nominal Scales (1)
A labeling scheme where the numbers used to serve as 'tags' or labels for identifying of classifying objects. Always one-to-one correspondence.
Ordinal Scales
A ranking scale of numbers assigned to objects to indicate a relative position regarding a particular characteristic.
Distance
Absolute separation between the scale descriptors are known and can be expressed in units.
Ordinal Scales (2)
Allows for rank order by which data can be sorted, but still does not allow for relative degree of difference between them.
Nominal Scales (2)
Also called qualitative type, differentiates between items or subjects based only on their names or categories and other qualitative classifications they belong to.
Measurement
Assignment of numbers or other symbols to characteristics of objects according to certain pre-specified rules.
Ratio Scales
Contains all the characteristics of the previous nominal, ordinal and interval scales pus a fixed zero point. It also has the characteristic of Origin.
Scaling
Creating a continuum upon which measures objects are located or plotted.
Stage 4:
Fieldwork or Data Collection
Convenience Sampling (NPT)
Groups happen to be chosen by convenience of a characteristic or a setting.
Sampling
Key component of any research design. Its design involves a marketer to ask oneself a series of questions regarding the characteristics of the sample.
Interval scales
Numerical equal distances on a scale represent equal values in the characteristic being measured. This scale does not have a fixed 'zero' point.
Quota Sampling (NPT)
One participant is chosen from each group. Within each of these groups one element is selected based on judgment or convenience.
Sampling Frame
Representation of the elements in a target population. Is a list or directions for identifying the target population.
Judgmental Sampling (NPT)
Researcher assumes the correct characteristics within specific groups from the participants.
Systematic Random Sampling (PT)
Select numbers generated by basing on a mathematical system or an equation.
Stratified Sampling (PT)
Select one element randomly from each group
Simple Random Sampling (PT)
Select random numbers from the entire population.
Snowball Sampling (NPT)
Specific participants are selected at random. The participants then refer to other participants in the group.
Sample
Subgroup of the individuals of the population
Ratio Scales (2)
Takes its name from the fact that measurement is the estimation of the ration between a magnitude of a continuous quantity and a unit magnitude of the same kind.
Census
The complete enumeration of the individuals of a population
Order
The relative size or position of the descriptors. No absolute values, just relative ones.
Origin
The scale has a unique and fixed beginning or true zero point.
Description of Scale
These are the unique labels we use to identify/designate each value on a scale
Nominal Scales (3)
Used for identifying respondents, brands, attributes and other objects.