Trainer's Guide
Sampling to Study Drug Use

1. Describe the principles on which sampling is based.

2. Identify and describe different sampling methods.

3. Select a sample of health facilities from a list of such facilities using at least two different methods.

4. Compare the results of the methods and discuss when each method might be used in the field.

Instruct others how to sample when studying drug use.

Purchase in advance of the session, sweets, etc., which can be used in the different sampling exercises.

Prepare strips of paper with a different number on each strip repeated 7 times. Each participant will need to be able to tear off a number to use in the random sampling activities.

Have sampling tools available. These could include a random number table, a dice, a banknote, a coin, a bottle that can be spun and a calculator.

Write out a list of numbers equivalent to the class size on a flip chart or white board. You will use this list for the systematic sampling exercise.

VISUAL AIDS LISTING

1. Title Slide
2. Objectives
3. What is Sampling?
4. Sampling involves the selection of a number of units from a defined study population
5. Sampling Definitions
6. A Representative Sample
7. Sampling Methods
8. Non Probability Sampling Methods
9. Probability Sampling Methods
10. Simple Random Sampling
11. Systematic Sampling with Equal Probability
12. Systematic Sampling with Probability Proportional to Size
13. Stratified Sampling
14. Cluster Sampling
15. Multistage Sampling
16. Sample Size
17. Sample Size
18. Summary
19. Activity One: Sampling Selections

First Component

· (15 minutes) 2 VA 's 1-8

Introduction, Definitions and Sampling Methods

The key concept to convey is that whenever data are collected, there is an element of chance in what is collected. When learning about drug use, every effort needs to be made to ensure that the data collected are representative. No one method of sampling is always appropriate so drug use researchers must be aware of different sampling methods.

Second Component

· (45 minutes) 2 VA 9-14

Practical Demonstration of Simple, Systematic, Stratified, Cluster and Multistage Sampling

For this component use the class members as the population. Tell the group that you would like to distribute some sweets to them, but do not have enough for everyone. In order to distribute the sweets fairly, you will need to use the various sampling methods.

Start with simple random sampling in which each participant puts a number in a container and then draw numbers randomly and allocate the sweets.

For systematic sampling, divide the number of participants by the number of gifts to establish the sampling interval. Then use a random number table or a banknote to establish the starting point. Then on the flip chart or board add the sampling interval to the random start to determine the other recipients.

For cluster sampling, spin a bottle to identify one group, and then give presents to all group members. Point out that this is inherently less representative of the entire population.

For multistage sampling different methods can be demonstrated by tossing a dice or using a banknote to select a group, and then spinning a bottle or using a small simple random lottery to decide which three groups members receive gifts.

For non-probability sampling you can demonstrate convenience sampling and quota sampling. For convenience sampling you can write on a flip chart hidden from the participants "Young Males," then throw packets of sweets to the different groups. You will discover that the "young males" jump up to catch the sweets usually excluding the female class members.

Promise to demonstrate quota sampling after class. Do this by having an assistant give sweets to the first five males and the first five females exiting the class.

Note to trainers: Participants can get quite excited during this activity. Make sure that the learning objectives are not lost and that the VA's are used to distinguish the different methods. It is useful to take a break before the next component.

 

Third Component

· (15 minutes) 2 VA's 15-17

Sample Size and Summary

Stress that optimal sample size is usually a compromise between what is desirable and what is feasible. Do point out that there is no point collecting quantitative data if the sample is too small. Emphasize that in most situations, a sample must include at least 30 members to produce reliable estimates.

Use the summary slide to reinforce the point that everybody, including field staff, need to understand the principles of sampling.

Fourth Component

· (90 minutes) 2 VA's 18

Activity One: Sample Selection

Start by pointing out that real data will be used and that 20 out of 52 facilities will be chosen as would occur in an indicator study. Point out that each group will use not all methods during the class, but that it is possible to complete the activity as individual homework. Point out that the different methods will demonstrate different results, which can be compared with the real results on the summary table. Offer to mark the completed worksheets of individuals who want to do the extra work.

VISUAL AIDS WITH NOTES

Back to top | Back to Table of Contents