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

Module 7.2

SAMPLING TECHNIQUE

Learning objectives At the end of this session, students will be able to  Explain use of sampling  List the type of sampling methods  Select appropriate sampling method

List of Topics

 Definition and use of Sampling

 Sampling methods

 Sampling error

Definition of sampling  Procedure by which some members of the population are

selected as representatives of the entire population Study population The study population is the population to which the results of the study will be inferred

The study population depends upon the research question  How many injections do people receive each year in India?  Study population: Population of India

 How many needle-sticks to health care workers experience each year in India?  Study population: Health care workers of India

 How many hospitals have a needle-stick prevention policy in India?  Study population: Hospitals of India

SAMPLE – Representative of time  Seasonality  Day of the week  Time of the day SAMPLE – Representative of place  Urban  Rural SAMPLE – Representative of persons  Age  Sex  Other demographic characteristics

TERMINOLOGY

 Sampling unit (Basic Sampling Unit, BSU)  Elementary unit that will be sampled  People  Health care workers  Hospitals

 Sampling frame  List of all sampling units in the population

 Sampling scheme  Method used to select sampling units from the sampling

frame

WHY TO SAMPLE

 Obtain information from large populations

 Ensure the efficiency of a study

 Obtain more accurate information

Population

 Infinite/finite size  Characterized by unknown

parameters

Sample

• Finite size • Characterized by

measurable parameters (e.g., mean, standard dev.)

A sample is a part of the population, selected by the investigator to gather information (measures) on certain characteristics of the original population

SAMPLING  Non – probability  Probability

Non-probability samples  Probability of being selected is unknown  Convenience samples  Biased  Best or worst scenario

 Subjective samples  Based on knowledge  Time/resource constraints

Probability samples  Every unit in the population has a known probability of being

selected  Only sampling method that allows to draw valid conclusions

about population

Probability samples  Removes the possibility of bias in selection of subjects  Ensures that each subject has a known probability of being

chosen  Allows application of statistical theory

Methods used in probability samples 1. Simple, random sampling 2. Systematic sampling 3. Stratified sampling 4. Cluster sampling 5. Multistage sampling

Simple, random sampling  Principle  Equal chance for each statistical unit

 Procedure  Number all units  Randomly draw units

 Advantages  Simple  Sampling error easily measured

 Disadvantages  Need complete list of units  Does not always achieve best representatively

METHODS OF SRS  Lottery method  Random number table

EXAMPLE OF SRS 1 Albert D. 2 Richard D. 3 Belle H. 4 Raymond L. 5 Stéphane B. 6 Albert T. 7 Jean William V. 8 André D. 9 Denis C. 10 Anthony Q. 11 James B. 12 Denis G. 13 Amanda L. 14 Jennifer L. 15 Philippe K. 16 Eve F. 17 Priscilla O. 18 Frank V.L. 19 Brian F. 20 Hellène H. 21 Isabelle R. 22 Jean T. 23 Samanta D. 24 Berthe L.

25 Monique Q. 26 Régine D. 27 Lucille L. 28 Jérémy W. 29 Gilles D. 30 Renaud S. 31 Pierre K. 32 Mike R. 33 Marie M. 34 Gaétan Z. 35 Fidèle D. 36 Maria P. 37 Anne-Marie G. 38 Michel K. 39 Gaston C. 40 Alain M. 41 Olivier P. 42 Geneviève M. 43 Berthe D. 44 Jean Pierre P. 45 Jacques B. 46 François P. 47 Dominique M. 48 Antoine C.

Systematic sampling  Principle  A unit drawn every k units  Equal chance of being drawn for each unit

 Procedure  Calculate sampling fraction/interval (k = N/n)  Draw a random number (≤ k) for starting  Draw every k units from first unit

 Advantages  Ensures representatively across list  Easy to implement

 Disadvantage  Dangerous if list has cycles

Example of systematic random

Stratified sampling  Principle  Classify population into homogeneous subgroups (strata)  Draw sample in each strata  Combine results of all strata

 Advantage  More precise if variable associated with strata  All subgroups represented, allowing separate conclusions

about each of them  Disadvantages  Sampling error difficult to measure  Loss of precision if small numbers sampled in individual

strata

Cluster sampling

 Principle  Random sample of groups (“clusters”) of

units  All or proportion of units included in selected

clusters  Advantages  Simple: No list of units required  Less travel/resources required

 Disadvantages  Imprecise if clusters homogeneous (Large

design effect)  Sampling error difficult to measure

Cluster sampling  The sampling unit is not a subject, but a group (cluster) of

subjects.  It is assumed that:  The variability among clusters is minimal  The variability within each cluster is what is observed in the

general population

The two stages of a cluster sample

1. First stage: Probability proportional to size • Select the number of clusters to be included • Compute a cumulative list of the populations in each unit

with a grand total • Divide the grand total by the number of clusters and obtain

the sampling interval • Choose a random number and identify the first cluster • Add the sampling interval and identify the second cluster • By repeating the same procedure, identify all the clusters

The two stages of a cluster sample 2. Second stage

• In each cluster select a random sample using a sampling frame of subjects (e.g. residents) or households

Self-weighting in cluster samples  Stage one: The larger units are more likely to be

selected in the first round  Unit B twice as large as unit A, hence unit B will have twice

the chance of being selected  Stage two: Individuals in larger unit selected are less

likely to be selected in the second round  Individual in unit B will have half the chance of being

selected within the unit  The two effects cancel each other and each person in

the population has the same probability of being sampled

WHO - 30 x 7 cluster sampling  Procedure: list of all villages (areas) with total population

 Village Inhabitants Cumulative  1 34 34  2 60 94  3 30 124  4 76 200  5 315 515

4,715

 Divide the cumulative total by 30 clusters we wish to select

 4,715 : 30= 157.1

Find a random number with three digits (= Sampling interval) e.g. 123

Choose from the cumulative distribution the clusters by adding 157 (sampling interval)

3 124 124 * 1st cluster 4 76 200 5 315 515 ** 2nd

123+157=280

 In each village (area) choose 7 children, randomly

 Total sample 30 X 7= 210

Example of cluster sampling

Village 4

Village 5

Village 3

Village 2Village 1

Multistage sampling  Principle  Several chained samples  Several statistical units

 Advantages  No complete listing of population required  Most feasible approach for large populations

 Disadvantages  Several sampling lists  Sampling error difficult to measure

Non-probability samples  Probability of being selected is unknown  Convenience samples  Biased  Best or worst scenario

 Subjective samples  Based on knowledge  Time/resource constraints

 Purposive sampling, Quota sampling

Sampling errors  We observe a sample instead of the whole population.  If we take repeated samples from the same population, the

results obtained from one sample differ from the results of another sample.

 This type of variation is called sampling error.

Sampling error  No sample is a perfect mirror image of the population  Magnitude of error can be measured in probability samples  Expressed by standard error of mean, proportion, differences…  Function of:  Sample size  Variability in measurement

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.

• 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.