Order 995769: Capstone project on physical health, levels of prevention and sampling in research

Module 7.1

Sampling

SAMPLE SIZE DETERMINATION

Learning Objectives

At the end of this module, the students will be able to describe :

 sampling and its purpose.

 need of Sample size in research .

List of Topics

 Purpose of sampling

 Sampling in Descriptive survey

 Sample size determination

 Adequacy of sample size

WHAT IS THE PURPOSE OF SAMPLING?

 The purpose of sampling is to examine some portion of the population and

 To extend the knowledge obtained from the sample to the population at large.

RELATIONSHIP BETWEEN POPULATION AND SAMPLE

PLACE OF SAMPLING IN DESCRIPTIVE SURVEYS  Define objectives

 Identify study population

 Establish Plan of Analysis

 Identify variables to study

 Choose sampling frame

PLACE OF SAMPLING IN DESCRIPTIVE SURVEYS CONT...  Select sample

 Design questionnaire

 Collect data

 Analyse data

 Communicate results

 Use results

SAMPLE SIZE DETERMINATION

 No universal answer

 Assumption dependent (and therefore partly subjective)

 Other considerations (cost, time-frame, feasibility)

NEED FOR ADEQUATE SAMPLE SIZE  Studies on small number of subjects

- May fail to detect important effects - May estimate effects imprecisely

 Studies on larger number of subjects - Waste of resources - May lead to a loss in accuracy

ADEQUACY OF SAMPLE SIZE  Depends on the scope of the study

 Curative effect of a drug for AIDS

- Single positive result is important

- Sample size may be irrelevant

ADEQUACY OF SAMPLE SIZE CONT …  Testing new malaria vaccine

- Several known preventive measures of varying effectiveness

- Vaccine should be tested on adequate number of subjects

TYPES OF OUTCOME MEASURES

Proportion, Mean

• Proportion of subjects who develop

malaria after entering high risk area.

• Mean Cholesterol level among the

diabetic individuals

HOW LARGE A SAMPLE DO I NEED?

 Depends on the aim(s) of the study

 Reasonable guess of the expected result

 How precise the result should be?

 Operational constraints

 Relevant information

SELECTION OF SAMPLE SIZE

1) If we take the small samples < 30 % our study is not valid

2) If we take large samples > 75 % it is laborious to do

3) We need optimum size which gives reliable results minimum sample size advised is 30 %

80%

20%

25%

70%

30%

75%

HOW TO FIND SUITABLE SAMPLE SIZE

151 June 2016

DataQualitative Quantitative

Find Permissible or Allowable error L

Error Not Exceed 10% Or 20 %

Error Not Exceed 1 % Or 2 %

Find 30 % Incidence or Prevalence

10 % L = 30 / 100 * 10 = 3

20 % L = 30 / 100 * 20 = 6

1 % L = 30 / 100 * 1 = 0.3

2 % L = 30 / 100 * 2 = 0.6

n = 4 * p * q / L 2 n = 4 σ 2 / L 2

10 % n = 4 * 30 * 70 / 3 * 3 = 933

20 % n = 4 * 30 * 70 / 6 * 6 = 233

1 % n = 4 * 8 * 8 / 0.3 * 0.3 = 2844

2 % n = 4 * 8 * 8 / 0.6 * 0.6 = 711

References  Sandelowski, M. (1995). Sample size in qualitative

research. Research in Nursing & Health, 18, 179–183.

 Emmel, N. (2013). Sampling and choosing cases in qualitative

research: A realist approach. London: Sage.

References

 NIST/SEMATECH, "7.2.4.2. Sample sizes required", e-

Handbook of Statistical Methods.

 Kish, L. (1965). Survey Sampling. Wiley. ISBN 0-471-48900-X

Epidemiology workshop series Community Medicine