ECON1203: Statistics
Chapter 5: Data collection and
sampling
1. Methods of collecting data
Personal
interview
Telephone
interview
Self-administered
survey
Advantages
High response rate
Interviewer can clarify
questions
Less expensive
Disadvantages
Costly
Interviewer can bias
Less expensive
Questionnaire design
1. The questionnaire should be short.
2. The questions should be short, clear and unambiguous (e.g. what is
family?)
3. Start with simple demographic questions to help respondents get started
and comfortable.
4. Dichotomous (e.g. yes/no) questions are simple but limiting.
5. Open-ended questions give more detailed answers but are difficult to
tabulate.
6. Avoid leading questions because they suggest an answer.
7. If there is time, pretest to uncover potential problems.
8. Consider the type of data (interval/ordinal/nominal) & techniques
(descriptive/inferential).
ECON1203: Statistics
2. Sampling
3. Sampling plans
1. A simple random sample is selected so that every possible sample with
the same number of observations is equally likely to be chosen (e.g.
raffle).
2. A stratified random sampling separates the population into mutually
exclusive sets, or strata, and then draws simple random samples from
each stratum (e.g. by income).
3. A cluster sample is a simple random sample of groups or clusters of
elements (e.g. blocks in a city).