NMQ 745 Chapter 10 Sampling
10.1 INTRODUCTION
Impossible to include the entire population
Restrictions being time and cost
Objective of a survey (uses the sample to learn about the population
Sample to be drawn in such a way that it would be valid to generalize its results to the
population
10.2 SAMPLING METHODS
two major classes (probability methods and non-probability methods,
probability methods are based on the principles of randomness and probability theory
non-probability methods are not
probability samples (accurately generalize to the population
10.2.1 Probability sampling methods
non zero probability of being selected
objective mechanism is used in the selection procedure
Four methods
1. simple random sampling
2. systematic sampling
3. stratified sampling
4. cluster sampling
appropriate choice of one
nature if the research probel
availability of a good sampling frame
money time and the characteristics of the population
10.2.1.1 simple random sampling
complete and up to date sample frame available
on this list each population element has to be numbered sequentially such that each
element can uniquely be identified
the actual drawing of the sample involves the generation of a predetermined number
of random numbers
10.2.1.2 systematic sampling
systematic sample is drawn is by systematically moving through the sample frame and
selecting every KTH element
1. calculate the sampling interval K as the nearest integer (whole number) to the ratio
(proportion) N/n (see practical example below)
2. get a random integer number between 0 and k, and call it S.
3. the sample consist of the units s, s+k, s+2k, s+3k, s+(n-1) k in the sample frame
population size is not known and the population elements arrive at a certain location over
time
10.2.1.3 stratified sampling
population is divided into a number of homogeneous, non-overlapping groups, called strata
10.1 INTRODUCTION
Impossible to include the entire population
Restrictions being time and cost
Objective of a survey (uses the sample to learn about the population
Sample to be drawn in such a way that it would be valid to generalize its results to the
population
10.2 SAMPLING METHODS
two major classes (probability methods and non-probability methods,
probability methods are based on the principles of randomness and probability theory
non-probability methods are not
probability samples (accurately generalize to the population
10.2.1 Probability sampling methods
non zero probability of being selected
objective mechanism is used in the selection procedure
Four methods
1. simple random sampling
2. systematic sampling
3. stratified sampling
4. cluster sampling
appropriate choice of one
nature if the research probel
availability of a good sampling frame
money time and the characteristics of the population
10.2.1.1 simple random sampling
complete and up to date sample frame available
on this list each population element has to be numbered sequentially such that each
element can uniquely be identified
the actual drawing of the sample involves the generation of a predetermined number
of random numbers
10.2.1.2 systematic sampling
systematic sample is drawn is by systematically moving through the sample frame and
selecting every KTH element
1. calculate the sampling interval K as the nearest integer (whole number) to the ratio
(proportion) N/n (see practical example below)
2. get a random integer number between 0 and k, and call it S.
3. the sample consist of the units s, s+k, s+2k, s+3k, s+(n-1) k in the sample frame
population size is not known and the population elements arrive at a certain location over
time
10.2.1.3 stratified sampling
population is divided into a number of homogeneous, non-overlapping groups, called strata
