2) Chapter 2: Sampling and Measurement Flashcards
What is “Variable”?
Any characteristic we can measure for each subject.
Variable can vary in value among subjects in a sample or population.
What is measurement scale?
Values that the variable can take form the measurement scale.
When variable is called “Quantitative”?
When the measurement scale has numerical values.
When variable is called “Categorical”?
Also so called qualitative.
Qualitative = Categorical
When the measurement scale is a set of categories.
Why it is important to distinguish between categorical/qualitative and quantitative variables?
Because different statistical methods apply to each type.
Example average for quantitative.
What is an Interval Scale of Measurement? For what type of variables is it used?
An interval scale of measurements is a scale which have a specific NUMERICAL distance or interval, and is used for Quantitative Variables.
What type of scales are applied for categorical values?
Nominal scale where categories do not have “high” or “low” end.
Ordinal scale where categories have a natural ordering of values.
Quantitative aspects of ordinal data, and sensitivity analysis?
Course grades (ABCD) are ordinal, but we treat them as interval when we assign numbers to them.
Number of values in the measurement scale or:
- Discrete variables?
- Continuous Variables?
Discrete - if it is separate number (1;2;3;4;5;6…)
Contentious - can take an infinite continuum of possible real number values (21.3851)
Simple random Sampling?
Simple random sampling is a method of sampling for which every possible sample has equal chance of selection.
Subjects of population can be sampled.
Why it is good to use random sampling?
To prevent sample become biased.
How to select simple random sample?
We need a list of all subjects in population. (Sampling frame).
- Number the subjects in the sampling frame.
- Generate a set of these numbers randomly.
- Sample the subjects whose numbers were generated.
Probability and non-probability sampling?
Probability: when probability any particular sample selected is known.
Non-probability sampling: not possible to determine the probabilities of the possible sampling.
Three types of bias?
- Sampling bias:
Volunteer sampling, when the subject of sampling are unlikely to be a representative cross section. - Response Bias: When the subject gives an incorrect response (or lying), or the question i wrong.
- Non-response bias: when sampled subjects cannot be reached or refuse to participate.