Sampling

A sampling frame covers exactly or bigger than the target population so every unit has a chance to be selected.

Unit -> The subject (e.g every adult age 15 or over who resides in Singapore)
Population -> Collection of Units ( collection of all the units abov)
Sample -> Subset ( a portion of adults would be selected to provide their employment status)

Census-> Measurement would be taken from every unit in the population
Sample -> Measurement would be taken from some selected units in the population


1) Probability Sampling plan (good)
Evert unit in the population must have a known probability of being selected into the sample without any biasness

Frame must have good coverage and up to date (complete)

• Simple random sampling (Every sample has an equal chance of selection)
I. Assign a number from 1 to N to each sampling unit in the sampling frame, where N is the
the number of sampling units.
II. Use the function randbetween (1,N) in Microsoft Excel to generate random numbers.
III. The sampling unit that has the assigned number the same as the generated number is
selected into the sample.

• Systematic sampling (when size of the population is not known when planning)
Selecting units from a list through intervals
Example:
Conducting a survey in the mall, for every k=10 people, we will survey one.
Probability is 1/10
Its systematic because we systematically take every 10 people.

Cyclic effect: Undesirable sample where the arrangement of the sampling units and the value k=10 have the same cyclical effect

• Stratified Sampling
Take a sample from each separate group

• Multistage Sampling
Several stages of selection. At each stage, probability sampling plan is implemented

• Cluster Sampling
Each cluster is a small representation of the population.
Use simple random sampling to choose the cluster. All units in that cluster will form sample

2) Non-Response
- Results are distorted.
- Do not remove from data
- Not all selected units are contactable or willing to take part in the study.

3) Convenience
Survey anyone that is "free" at that particular time.
The information that we can gather from respondents who are
easily available is normally different from the hard ones to get.

4) Volunteer sample

- Self Selected to form sample

- Biased as its for people with the strong view on the issue

5) Judgement sample
- Biased as selection is based on opinions of experts
Sample units are chosen from the population by
interviewers using their own discretion about which informants are “typical” or “representative”

6) Interviewing (Quota)
Selection in which the elements are chosen in the field by interviewers
They can choose anyone they like

More

7) Response
- The phrasing of the question, tone or attitude of the interviewer can affect the results

8) Parameter
A numerical fact about population usually unknown to us
It is estimated from a sample
Our parameter of interest is from a defined population. We can use a sample to give us an estimate about that parameter.

Sample - Estimate                   Population-parameter

e.g Unemployment rate
Sample:                                    Population:
A portion of adults                  The collection of every adult age 15 above
would be selected                     who resides in Singapore
to provide their                       
employment status.                  Parameter:
                                                 Singapore's unemployment rate
Estimate:
Percentage of unemployed
adults in the sample

Estimate of parameter = parameter + random error + bias(Hard to quantify)
Assuming we are using simple random sample


< Prev Measurements                                       Next Probability>