Sampling design 1216114348242957-8

SAMPLING SAMPLING DESIGNDESIGN

PROBABILITY PROBABILITY SAMPLING & NON-SAMPLING & NON-

PROBABILITY PROBABILITY SAMPLINGSAMPLING

Definition of Sampling:Definition of Sampling: Measuring a small portion of Measuring a small portion of

something and then making a something and then making a general statement about the whole general statement about the whole thing.thing.

Process of selecting a number of Process of selecting a number of units for a study in such a way that units for a study in such a way that the units represent the larger the units represent the larger group from which they are group from which they are selected.selected.

Why We Need Sampling (Purposes Why We Need Sampling (Purposes and Advantages of Sampling)and Advantages of Sampling)

1.1. Sampling makes possible the Sampling makes possible the study of a large, study of a large, population.population.

- The universe or population to be - The universe or population to be studied maybe too large or unlimited studied maybe too large or unlimited that it is almost impossible to reach that it is almost impossible to reach all of them. Sampling makes possible all of them. Sampling makes possible this kind of study because in sampling this kind of study because in sampling only a small portion of the population only a small portion of the population maybe involved in the study, enabling maybe involved in the study, enabling the researcher to reach all through the researcher to reach all through this small portion of the population.this small portion of the population.

Why We Need Sampling (Purposes Why We Need Sampling (Purposes and Advantages of Sampling)and Advantages of Sampling)

2.2. Sampling is for economy.Sampling is for economy.- Research without sampling - Research without sampling

may be too costly. Sampling may be too costly. Sampling reduces the study population to a reduces the study population to a reasonable size that expenses are reasonable size that expenses are greatly reduced.greatly reduced.

3.3. Sampling is for speedSampling is for speed..- Research without sampling - Research without sampling

might be too time consuming.might be too time consuming.

Why We Need Sampling (Purposes Why We Need Sampling (Purposes and Advantages of Sampling)and Advantages of Sampling)

4.4. Sampling is for accuracy.Sampling is for accuracy.

- If it takes too long a time to - If it takes too long a time to cover the whole study population, cover the whole study population, there maybe inaccuracy. The there maybe inaccuracy. The research must be finished within research must be finished within a reasonable period of time so a reasonable period of time so that the data are still true, valid that the data are still true, valid and reasonable.and reasonable.

Why We Need Sampling (Purposes Why We Need Sampling (Purposes and Advantages of Sampling)and Advantages of Sampling)

5.5. Sampling saves the sources Sampling saves the sources of data from being all of data from being all consumed.consumed.

- The act of gathering data - The act of gathering data may consume all the sources of may consume all the sources of information without sampling. information without sampling. In such a case, there is no more In such a case, there is no more data to apply the conclusion to.data to apply the conclusion to.

Disadvantages of Sampling Disadvantages of Sampling (Defective Sampling)(Defective Sampling)

1.1. If sampling is biased, or not If sampling is biased, or not representative, or too small, the representative, or too small, the conclusion may not be valid and conclusion may not be valid and reliable.reliable.

2.2. In research, the respondents to a In research, the respondents to a study must have a common study must have a common characteristics which is the basis of characteristics which is the basis of the study.the study.

Disadvantages of Sampling Disadvantages of Sampling (Defective Sampling)(Defective Sampling)

3.3. If the population is very large and If the population is very large and there are many sections and there are many sections and subsections, the sampling subsections, the sampling procedure becomes very procedure becomes very complicated.complicated.

4.4. If the researcher does not possess If the researcher does not possess the necessary skill and technical the necessary skill and technical knowhow in sampling procedure.knowhow in sampling procedure.

WHAT IS A GOOD WHAT IS A GOOD SAMPLE?SAMPLE? The sample must be valid.The sample must be valid.

Validity depends on 2 considerations:Validity depends on 2 considerations:1. 1. AccuracyAccuracy – bias is absent from the sample – bias is absent from the sample(ex. A company is thinking of lowering its (ex. A company is thinking of lowering its price for its soap bar product. After making a price for its soap bar product. After making a survey in the sales of their product in a known survey in the sales of their product in a known mall in Makati they concluded that they will mall in Makati they concluded that they will not cut down the price of the soap bar since not cut down the price of the soap bar since there was an increased in sales compared to there was an increased in sales compared to last year. Bias is present in this study since last year. Bias is present in this study since the company based its decision for the sales of the company based its decision for the sales of a known mall which have consumers who can a known mall which have consumers who can afford high price products. They did not afford high price products. They did not consider the sales of their products in other consider the sales of their products in other area wherein they have middle class or low area wherein they have middle class or low class consumers.) class consumers.)

WHAT IS A GOOD WHAT IS A GOOD SAMPLE?SAMPLE?2. 2. PrecisionPrecision – sample represents the – sample represents the

populationpopulation(ex. Customers who visited a particular (ex. Customers who visited a particular

dress shop are requested to log in their dress shop are requested to log in their phone numbers so that they will receive phone numbers so that they will receive information for discounts and new information for discounts and new arrivals. Management wish to study arrivals. Management wish to study customers satisfaction for that shop. By customers satisfaction for that shop. By means of interviewing thru phone they get means of interviewing thru phone they get comments and reactions of their client. comments and reactions of their client. Samples used are not an exact Samples used are not an exact representative of the population since it is representative of the population since it is limited only to those customers who log in limited only to those customers who log in their phone numbers and they did not their phone numbers and they did not consider customers without phone consider customers without phone numbers indicated. numbers indicated.

STEPS IN SAMPLING STEPS IN SAMPLING DESIGNDESIGN

1.1. What is the target population?What is the target population?- Target population is the aggregation of - Target population is the aggregation of elements (members of the population) from elements (members of the population) from which the sample is actually selected.which the sample is actually selected.

2.2. What are the parameters of interest?What are the parameters of interest?- Parameters are summary description of a given - Parameters are summary description of a given variable in a population. variable in a population.

3.3. What is the sampling frame?What is the sampling frame?- Sampling frame is the list of elements from - Sampling frame is the list of elements from which the sample is actually drawn. Complete which the sample is actually drawn. Complete and correct list of population members only. and correct list of population members only.

4.4. What is the appropriate sampling method?What is the appropriate sampling method? - Probability or Non-Probability sampling - Probability or Non-Probability sampling

methodmethod

STEPS IN SAMPLING STEPS IN SAMPLING DESIGNDESIGN

5.5. What size sample is needed?What size sample is needed?There are no fixed rules in determining the size of There are no fixed rules in determining the size of a sample needed. There are guidelines that should a sample needed. There are guidelines that should be observed in determining the size of a sample.be observed in determining the size of a sample.

When the population is more or less When the population is more or less homogeneous and only the typical, normal, or homogeneous and only the typical, normal, or average is desired to be known, a smaller average is desired to be known, a smaller sample is enough. However, if differences are sample is enough. However, if differences are desired to be known, a larger sample is desired to be known, a larger sample is needed.needed.

When the population is more or less When the population is more or less heterogeneous and only the typical, normal or heterogeneous and only the typical, normal or average is desired to be known a larger average is desired to be known a larger sample is needed. However, if only their sample is needed. However, if only their differences are desired to be known, a smaller differences are desired to be known, a smaller sample is sufficient.sample is sufficient.

STEPS IN SAMPLING STEPS IN SAMPLING DESIGNDESIGN

The size of a sample varies inversely as the The size of a sample varies inversely as the size of the population. A larger proportion is size of the population. A larger proportion is required of a smaller population and a smaller required of a smaller population and a smaller proportion may do for a bigger population.proportion may do for a bigger population.

For a greater accuracy and reliability of For a greater accuracy and reliability of results, a greater sample is desirable.results, a greater sample is desirable.

In biological and chemical experiments, the In biological and chemical experiments, the use of few persons is more desirable to use of few persons is more desirable to determine the reactions of humans.determine the reactions of humans.

When subjects are likely to be destroyed When subjects are likely to be destroyed during experiment, it is more feasible to use during experiment, it is more feasible to use non-humans.non-humans.

STEPS IN SAMPLING STEPS IN SAMPLING DESIGNDESIGN

Example: Example: A Company would like to make a study in the quality of A Company would like to make a study in the quality of

digital cameras it manufactured. digital cameras it manufactured. 1.1. Target population – consumers of digital camerasTarget population – consumers of digital cameras2.2. Parameters of interest – quality of digital cameras Parameters of interest – quality of digital cameras

(scale of 1 to 5 , 5 being the most satisfactory)(scale of 1 to 5 , 5 being the most satisfactory)3.3. Sampling frame – database of stores in which digital Sampling frame – database of stores in which digital

cameras are sold, usually customers gives cameras are sold, usually customers gives information about them for warranty purposesinformation about them for warranty purposes

4.4. Sampling method – Probability sampling (Stratified Sampling method – Probability sampling (Stratified sampling). sampling).

5.5. Size of sample – it is more on heterogeneous Size of sample – it is more on heterogeneous population, average responses would like to know by population, average responses would like to know by the manufacturer, so large proportion will be needed the manufacturer, so large proportion will be needed from the population. from the population.

STEPS IN COMPUTING STEPS IN COMPUTING THE SIZE OF A SAMPLETHE SIZE OF A SAMPLE

1.1. Determine the size of the target population. Determine the size of the target population. 2.2. Decide on the margin of error. As much as Decide on the margin of error. As much as

possible the margin of error should not be possible the margin of error should not be higher than 5%. Probably 3% is an ideal one.higher than 5%. Probably 3% is an ideal one.

3.3. Use the formula n = Use the formula n = N N 1 + Ne1 + Ne2 2 (pagoso , et al. (pagoso , et al.

p.46)p.46)n = sample sizen = sample size

N = the size of the populationN = the size of the population e = the margin of errore = the margin of error4.4. Compute the sample proportion by dividing the Compute the sample proportion by dividing the

result in number 3 by the population.result in number 3 by the population.

STEPS IN COMPUTING STEPS IN COMPUTING THE SIZE OF A SAMPLETHE SIZE OF A SAMPLE

1.1. Population is 5,346Population is 5,3462.2. Margin of error is 3%Margin of error is 3%3.3. Using the formulaUsing the formula n = n = ___5,346_ ___5,346_ 1+ 5346(.03)1+ 5346(.03)22

n = 920n = 9204.4. Sample proportion (%) = 920 / 5346 Sample proportion (%) = 920 / 5346 = 17%= 17%

General Types of SamplingGeneral Types of Sampling

1.1. Probability samplingProbability sampling

the sample is a proportion (a certain the sample is a proportion (a certain percent) of the population and such percent) of the population and such sample is selected from the population sample is selected from the population by means of some systematic way in by means of some systematic way in which every element of the population which every element of the population has a chance of being included in the has a chance of being included in the sample.sample.

PROBABILITY SAMPLINGPROBABILITY SAMPLING

randomization is a feature of the randomization is a feature of the selection process rather than an selection process rather than an assumption about the structure of assumption about the structure of the population.the population.

more complex, time consuming and more complex, time consuming and more costlymore costly

General Types of SamplingGeneral Types of Sampling

2.2. Non-probability samplingNon-probability sampling

the sample is not a proportion of the sample is not a proportion of

the population and there is no the population and there is no system in selecting the sample. system in selecting the sample. The selection depends upon the The selection depends upon the situation.situation.

NON-PROBABILITY SAMPLINGNON-PROBABILITY SAMPLING

no assurance is given that each no assurance is given that each item has a chance of being included item has a chance of being included as a sample as a sample

there is an assumption that there is there is an assumption that there is an even distribution of an even distribution of characteristics within the characteristics within the population, believing that any population, believing that any sample would be representative.sample would be representative.

TYPES OF TYPES OF PROBABILITY PROBABILITY

SAMPLINGSAMPLING

A. PURE RANDOM A. PURE RANDOM SAMPLINGSAMPLING

this type of sampling is one in which this type of sampling is one in which every one in the population of the inquiry every one in the population of the inquiry has an equal chance of being selected to has an equal chance of being selected to be included in the sample. be included in the sample.

also called the lottery or raffle type of also called the lottery or raffle type of sampling.sampling.

this may be used if the population has no this may be used if the population has no differentiated levels, sections, or classes. differentiated levels, sections, or classes.

done with or without replacementdone with or without replacement

PURE RANDOM SAMPLINGPURE RANDOM SAMPLING

main advantage of this technique main advantage of this technique of sampling is that, it is easy to of sampling is that, it is easy to understand and it is easy to apply understand and it is easy to apply too.too.

disadvantage is that, it is hard to disadvantage is that, it is hard to use with too large a population use with too large a population because of the difficulty because of the difficulty encountered in writing the names encountered in writing the names of the persons involved. of the persons involved.

PURE RANDOM SAMPLINGPURE RANDOM SAMPLING Steps in selecting sample using a table of random Steps in selecting sample using a table of random

numbers:numbers: Define the populationDefine the population Determine the desired sample sizeDetermine the desired sample size List all the members of the populationList all the members of the population Assign each of the individuals on the list a Assign each of the individuals on the list a

consecutive number from zero to the required consecutive number from zero to the required number, ex. 01-89 or 001-249number, ex. 01-89 or 001-249

Select an arbitrary number in the table of Select an arbitrary number in the table of random numbers (Close your eyes and point)random numbers (Close your eyes and point)

For the selected number, look at only the For the selected number, look at only the appropriate number of digitsappropriate number of digits

If the selected number corresponds to the If the selected number corresponds to the number assigned to any individual in the number assigned to any individual in the population, then that individual is in the population, then that individual is in the samplesample

Repeat the steps until the desired sample size Repeat the steps until the desired sample size is reached.is reached.

B. SYSTEMATIC B. SYSTEMATIC SAMPLINGSAMPLING a technique of sampling in which every a technique of sampling in which every

kth name (old system of counting off) in kth name (old system of counting off) in a list may be selected to be included in a a list may be selected to be included in a sample. sample.

also called as interval sampling, there is also called as interval sampling, there is a gap or interval, between each selected a gap or interval, between each selected unit in the sample.unit in the sample.

used when the subjects or respondents used when the subjects or respondents in the study are arrayed or arranged in in the study are arrayed or arranged in some systematic or logical manner such some systematic or logical manner such as alphabetical arrangement and as alphabetical arrangement and geographical placement from north to geographical placement from north to south. south.

SYSTEMATIC SAMPLINGSYSTEMATIC SAMPLING

Steps in systematic sampling:Steps in systematic sampling: Define the populationDefine the population Determine the desired sample sizeDetermine the desired sample size Obtain a list (preferably randomized) of the Obtain a list (preferably randomized) of the

populationpopulation Determine what K is equal to by dividing the size Determine what K is equal to by dividing the size

of the population by the desired sample sizeof the population by the desired sample size Select some random place at the top of the Select some random place at the top of the

population listpopulation list Starting at that point, take every Kth name on Starting at that point, take every Kth name on

the list until desired sample size is reachedthe list until desired sample size is reached If the end of the list is reached before the desired If the end of the list is reached before the desired

sample is reached, go back to the top of the list.sample is reached, go back to the top of the list.

SYSTEMATIC SAMPLINGSYSTEMATIC SAMPLING

k = skip interval = k = skip interval = population sizepopulation size

sample sizesample size population size = 64population size = 64

sample size = 8sample size = 8

k = 8k = 8

SYSTEMATIC SAMPLINGSYSTEMATIC SAMPLING

main advantage is that it is more main advantage is that it is more convenient, faster, and more convenient, faster, and more economicaleconomical

disadvantage is that the sample disadvantage is that the sample becomes biased if the persons in the becomes biased if the persons in the list belong to a class by themselves list belong to a class by themselves whereas the investigation requires whereas the investigation requires that all sectors of the population are that all sectors of the population are to be involved.to be involved.

C. STRATIFIED C. STRATIFIED SAMPLINGSAMPLING the process of selecting randomly, samples the process of selecting randomly, samples

from the different strata of the population from the different strata of the population used in the study.used in the study.

advantage is that it contributes much to the advantage is that it contributes much to the representative of the samplerepresentative of the sample

Steps involves in stratified sampling:Steps involves in stratified sampling: Define the populationDefine the population Determine the desired sample sizeDetermine the desired sample size Identify the variable and subgroups (strata) for which Identify the variable and subgroups (strata) for which

you want to guarantee appropriate representation you want to guarantee appropriate representation (either proportion or equal)(either proportion or equal)

Classify all members of the population as members of Classify all members of the population as members of one of the identified subgroupsone of the identified subgroups

Randomly select (using table of random numbers) an Randomly select (using table of random numbers) an appropriate number of individuals from subgroups.appropriate number of individuals from subgroups.

STRATIFIED SAMPLINGSTRATIFIED SAMPLING

Example: A call center company wants to seek suggestions of Example: A call center company wants to seek suggestions of their agents for a new marketing strategy for their new their agents for a new marketing strategy for their new services.services.

1. Population 5,000 agents.1. Population 5,000 agents. 2. Desired sample size 5002. Desired sample size 500 3. Variable of interest is age and there are three subgroups 3. Variable of interest is age and there are three subgroups

under 30, 30 to 45 and over 45under 30, 30 to 45 and over 45 4. We classify the agents into the subgroups4. We classify the agents into the subgroups 20% or 1,000 are under age 3020% or 1,000 are under age 30 65% or 3,250 are age 30 to 4565% or 3,250 are age 30 to 45 15% or 750 are over age 4515% or 750 are over age 45 5. We want 500 agents. Since we want proportional 5. We want 500 agents. Since we want proportional

representation.representation. 20% of the sample (100) under age 3020% of the sample (100) under age 30 65% (325) should be age 30 to 4565% (325) should be age 30 to 45 15% (75) should be over age 4515% (75) should be over age 45 Therefore, using table of random numbers,Therefore, using table of random numbers, 100 of the 1000 under age 30 are selected100 of the 1000 under age 30 are selected 325 of the 3250 age 30 to 45 are selected325 of the 3250 age 30 to 45 are selected 75 of the 750 over age are selected75 of the 750 over age are selected

D. CLUSTER SAMPLINGD. CLUSTER SAMPLING also called as multistage cluster samplingalso called as multistage cluster sampling

used when the population is so big or the used when the population is so big or the geographical area of the research is so large.geographical area of the research is so large.

Advantage : EfficiencyAdvantage : Efficiency

Disadvantage: reduced accuracy or Disadvantage: reduced accuracy or representativeness, on the account of the fact representativeness, on the account of the fact that in every stage there is a sampling error.that in every stage there is a sampling error.

CLUSTER SAMPLINGCLUSTER SAMPLING

Steps in cluster sampling:Steps in cluster sampling: Define the populationDefine the population Determine the desired sample sizeDetermine the desired sample size Identify and define a logical clusterIdentify and define a logical cluster Obtain, or make a list of all clusters in the Obtain, or make a list of all clusters in the

populationpopulation Estimate the average number of population Estimate the average number of population

members per clustermembers per cluster Determine the number of clusters needed by Determine the number of clusters needed by

dividing the sample size by the estimated size dividing the sample size by the estimated size of the clusterof the cluster

Randomly select the needed number of Randomly select the needed number of clusters (using a table of random numbers)clusters (using a table of random numbers)

Include in the sample all population members Include in the sample all population members in selectedin selected clustercluster

CLUSTER SAMPLINGCLUSTER SAMPLING

Same example in the stratified sampling:Same example in the stratified sampling:1.1. Population 5,000 agentsPopulation 5,000 agents2.2. Desired sample size 500Desired sample size 5003.3. Logical cluster is a branchLogical cluster is a branch4.4. 50 branches all over the country50 branches all over the country5.5. Although the branch vary in number of agents , Although the branch vary in number of agents ,

there is an average of 100 agents per branch.there is an average of 100 agents per branch.6.6. The number of clusters (branch) needed equals The number of clusters (branch) needed equals

the desired sample size, 500 divided by the the desired sample size, 500 divided by the average size of a cluster, 100. Thus, the average size of a cluster, 100. Thus, the number of branch needed is 5.number of branch needed is 5.

7.7. Therefore, we randomly select 5 of the 50 Therefore, we randomly select 5 of the 50 branchbranch

8.8. All the agents in each of the 5 selected branch All the agents in each of the 5 selected branch are in the sample.are in the sample.

TYPES OF NON-TYPES OF NON-PROBABILITY PROBABILITY

SAMPLINGSAMPLING

A. ACCIDENTAL A. ACCIDENTAL SAMPLING SAMPLING

/CONVENIENCE /CONVENIENCE SAMPLINGSAMPLING no system of selection but only no system of selection but only

those whom the researcher or those whom the researcher or interviewer meet by chance are interviewer meet by chance are included in the sample.included in the sample.

process of picking out people in the process of picking out people in the most convenient and fastest way to most convenient and fastest way to immediately get their reactions to a immediately get their reactions to a certain hot and controversial issue.certain hot and controversial issue.

ACCIDENTAL / CONVENIENCE SAMPLINGACCIDENTAL / CONVENIENCE SAMPLING

not representative of target not representative of target population because sample are population because sample are selected if they can be accessed selected if they can be accessed easily and conveniently.easily and conveniently.

Advantage : easy to useAdvantage : easy to use

Disadvantage: bias is presentDisadvantage: bias is present

it could deliver accurate results when it could deliver accurate results when the population is homogeneous.the population is homogeneous.

ACCIDENTAL / CONVENIENCE SAMPLINGACCIDENTAL / CONVENIENCE SAMPLING

Examples:Examples: the female moviegoers sitting in the the female moviegoers sitting in the

first row of a movie theatrefirst row of a movie theatre the first 100 customers to enter a the first 100 customers to enter a

department storedepartment store the first three callers in a radio the first three callers in a radio

contestcontest use of volunteers use of volunteers

B. PURPOSIVE B. PURPOSIVE SAMPLINGSAMPLING

the respondents are chosen on the the respondents are chosen on the basis of their knowledge of the basis of their knowledge of the information desired.information desired.

TYPES OF PURPOSIVE TYPES OF PURPOSIVE SAMPLINGSAMPLING

1. QUOTA SAMPLING1. QUOTA SAMPLING specified number of persons of certain specified number of persons of certain

types are included in the sample.types are included in the sample.

advantage over accidental sampling is advantage over accidental sampling is that many sectors of the population are that many sectors of the population are represented. But its represented. But its representativeness is doubtful because representativeness is doubtful because there is no proportional representation there is no proportional representation and there are no guidelines in the and there are no guidelines in the selection of the respondents.selection of the respondents.

PURPOSIVE SAMPLINGPURPOSIVE SAMPLING

2. 2. JUDGEMENT SAMPLINGJUDGEMENT SAMPLING sample is taken based on certain sample is taken based on certain

judgements about the overall judgements about the overall population.population.

Critical issue: Objectivity “how much Critical issue: Objectivity “how much can judgement be relied upon to can judgement be relied upon to arrive at a typical sample?”arrive at a typical sample?”

Advantage: reduced cost and time Advantage: reduced cost and time involved in acquiring the sampleinvolved in acquiring the sample