Mba ii rm unit-3.3 sampling methods a

Course: MBACourse: MBASubject: Research MethodologySubject: Research Methodology

Unit-3.3Unit-3.3

SAMPLING METHOD

• Sampling Concepts• Methods• Procedure • Sample size decisions

What is Sampling?What is Sampling?

• the process of selecting a small the process of selecting a small number of elements from a larger number of elements from a larger defined target group of elementsdefined target group of elements

• the information gathered from the the information gathered from the small group will allow judgments to small group will allow judgments to be made about the larger groupsbe made about the larger groups

Why sampling?Why sampling?

• Inability to analyze large quantities of Inability to analyze large quantities of data potentially generated by a data potentially generated by a population population

• Practical considerations such as cost Practical considerations such as cost and timeand time

• Sampling can produce sound results if Sampling can produce sound results if proper rules are followed for the drawproper rules are followed for the draw

Basic concepts of SamplingBasic concepts of Sampling

• PopulationPopulation• SampleSample• Sampling unitSampling unit• Sampling errorSampling error• Sampling frameSampling frame• Sampling sizeSampling size

Basic concepts of SamplingBasic concepts of Sampling

• Population:Population: baseline - baseline -the the entire group under study entire group under study as defined by objectivesas defined by objectives

• Sample:Sample: a subset of the a subset of the population that should population that should represent the entire represent the entire groupgroup

Basic concepts of SamplingBasic concepts of Sampling

• Sample unit:Sample unit: the basic level of investigation the basic level of investigation

• Sampling error:Sampling error:

same sampling methods, same population, the study same sampling methods, same population, the study with a larger sample size will have less sampling process with a larger sample size will have less sampling process error compared to the study with smaller sample sizeerror compared to the study with smaller sample size

Basic concepts of SamplingBasic concepts of Sampling

• Sampling frame:Sampling frame:

a master list of the a master list of the population (total or partial) population (total or partial) from which the sample will from which the sample will be drawnbe drawn

• Sampling size:Sampling size: number of number of samples to be drawnsamples to be drawn

Types of SamplingTypes of Sampling

• Probability samplingProbability sampling

in which members of in which members of the population have a the population have a known chance known chance (probability) of being (probability) of being selected selected

• Non-probability Non-probability samplingsampling

in which the chances in which the chances (probability) of selecting (probability) of selecting members from the members from the population are population are unknownunknown

Sampling methodsSampling methods

ProbabilityProbability• Simple random Simple random

samplingsampling• Stratified random Stratified random

samplingsampling• Systematic samplingSystematic sampling• Cluster samplingCluster sampling

Non-probabilityNon-probability

• Convenience samplingConvenience sampling• Judgment samplingJudgment sampling• Quota samplingQuota sampling• Snowball samplingSnowball sampling

Probability Sampling MethodsProbability Sampling Methods

Simple random methodSimple random methodevery unit has an equal non-zero every unit has an equal non-zero chance of being selectedchance of being selected

• Advantages: Advantages: – Known and equal chance of Known and equal chance of

selectionselection– Easy method when there is Easy method when there is

an electronic databasean electronic database• Disadvantages: Disadvantages:

– Complete accounting of Complete accounting of population neededpopulation needed

This method is the purest form of probabil i ty sampling

Probability Sampling MethodsProbability Sampling Methods

Stratified random Stratified random methodmethodthe population is separated the population is separated into homogeneous strata and a into homogeneous strata and a sample is taken from eachsample is taken from each

• Advantages:Advantages:

– More accurate overall More accurate overall sample of skewed sample of skewed populationpopulation

• Disadvantages: Disadvantages: – More complex sampling More complex sampling

plan requiring different plan requiring different sample sizes for each sample sizes for each stratumstratum

Often used when one or more of the stratums in the population have a low incidence relat ive to the other stratums.

Probability Sampling MethodsProbability Sampling Methods

Systematic methodSystematic methodthe defined target population is the defined target population is ordered and the sample is ordered and the sample is selected according to position selected according to position using a skip intervalusing a skip interval

• Advantages: Advantages: – Known and equal chance of Known and equal chance of

selected intervalselected interval– Less expensive…faster than Less expensive…faster than

Radom methodsRadom methods• Disadvantages: Disadvantages:

– Loss in sampling precisionLoss in sampling precision

Systematic sampling is frequently used to select a specif ied number of records from a computer f i le.

Probability Sampling MethodsProbability Sampling Methods

Cluster methodCluster methodthe population is divided into the population is divided into groups (clusters), any of which groups (clusters), any of which can be considered a can be considered a representative samplerepresentative sample

• Advantages:Advantages:

– Economic efficiency … Economic efficiency … faster and less expensivefaster and less expensive

– Does not require a list of all Does not require a list of all members of the populationmembers of the population

• Disadvantages:Disadvantages:

– Cluster specification error…Cluster specification error…the more homogeneous the the more homogeneous the cluster chosen, the more cluster chosen, the more imprecise the sample resultsimprecise the sample results

Non-probability Sampling MethodsNon-probability Sampling Methods

Convenience Convenience sampling methodsampling methodthe selecting on the basis of the selecting on the basis of convenience convenience the selection at familiar the selection at familiar locations and to choose locations and to choose respondents who are like respondents who are like themselvesthemselves

Judgment methodJudgment methodselecting samples that require selecting samples that require a judgment or an “educated a judgment or an “educated guess”guess”

often used during preliminary research efforts to get a gross estimate of the results

must be confident that the chosen sample is truly representative of the entire population.

Non-probability Sampling MethodsNon-probability Sampling Methods

Quota sampling methodQuota sampling methodsamples that set a specific samples that set a specific number of certain types of number of certain types of individuals individuals Often used to ensure desired Often used to ensure desired proportion of different proportion of different respondent classesrespondent classes

Snowball methodSnowball methodselecting samples which selecting samples which require respondents to provide require respondents to provide the names of additional the names of additional respondentsrespondents

special method used when the desired sample characterist ic is rare

Online Sampling Techniques

• Random online intercept sampling:Random online intercept sampling:relies on a random selection of Web site visitorsrelies on a random selection of Web site visitors

• Invitation online sampling:Invitation online sampling:is when potential respondents are alerted that they may fill is when potential respondents are alerted that they may fill out a questionnaire that is hosted at a specific Web siteout a questionnaire that is hosted at a specific Web site

• Online panel sampling:Online panel sampling:refers to consumer or other respondent panels that are set refers to consumer or other respondent panels that are set up by marketing research companies for the explicit purpose up by marketing research companies for the explicit purpose of conducting online surveys with representative samplesof conducting online surveys with representative samples

Developing a Sampling Developing a Sampling PlanPlan

• Step-1Step-1Define the relevant Define the relevant population (baseline)population (baseline)

• Step-2Step-2Identify sample frameIdentify sample frame

• Step-3Step-3Determine specific sampling Determine specific sampling method, all necessary steps method, all necessary steps must be specifiedmust be specified

• Step-4Step-4Determine sample size, Determine sample size, selecting samplesselecting samples

• Step-5Step-5Execute the samplingExecute the sampling

• Step-6Step-6SSampling validation ampling validation --compare sample profile compare sample profile with population profile Re-with population profile Re-sampling if necessarysampling if necessary

Factors to consider in sampling design

• Work objectivesWork objectives• Degree of accuracyDegree of accuracy• ResourcesResources• Time frameTime frame• Knowledge on populationKnowledge on population• ScopeScope• Statistical analysis needsStatistical analysis needs

How to determine sample size?

Common approaches for determining sample size

– Budget/time available– Executive decision– Statistical methods– Historical data/guidelines

ISBN: 8173196257

ConclusionsConclusions

• A well-designed sampling plan answers the following questions –

What will be learned?

How long will it take?

How much will it cost?

Sampling is important for a survey/research projectSampling is important for a survey/research project

Many sampling start with a general hope that something Many sampling start with a general hope that something interesting will emerge, and often end in frustrationinteresting will emerge, and often end in frustration

Reference

• http://www.surveysampling.com/

http://www.surveysystem.com/