Sampling techniques

SAMPLING TECHNIQUESSAMPLING TECHNIQUES

Mukut HazarikaMukut HazarikaProfessorProfessor

Department of EducationDepartment of EducationDibrugarh UniversityDibrugarh University

Population and SamplePopulation and Sample

• PopulationPopulation: A population refers to any : A population refers to any collection of specified group of human collection of specified group of human beings or of non-human entities such as beings or of non-human entities such as objects, educational institutions, time objects, educational institutions, time units, geographical areas, prices of units, geographical areas, prices of wheat or salaries drawn by individuals.wheat or salaries drawn by individuals.

• SampleSample: The representative proportion : The representative proportion of the population is called a sample.of the population is called a sample.

SOME OTHER RELATED SOME OTHER RELATED TERMSTERMS• Sampling ElementSampling Element : Each entity (person, : Each entity (person,

family, group, organization) from the family, group, organization) from the population about which information is population about which information is collected is called a sampling element.collected is called a sampling element.

• Sampling UnitSampling Unit : This is either a single : This is either a single member (element) or a collection of member (element) or a collection of members (elements) subject to data members (elements) subject to data analysis (and selection) in the sample.analysis (and selection) in the sample.

• Sampling FrameSampling Frame : It refers to the complete : It refers to the complete list of all units / elements from which the list of all units / elements from which the sample is drawn.sample is drawn.

Steps Involved in the process Steps Involved in the process of Sampling:of Sampling:

• Defining the populationDefining the population

• Listing the population Listing the population

• Selecting a representative sampleSelecting a representative sample

• Obtaining an adequate sampleObtaining an adequate sample

Purposes of SamplingPurposes of Sampling

• Population in many cases may be so large Population in many cases may be so large and scattered that a complete coverage and scattered that a complete coverage may be not possible.may be not possible.

• In a short period of time valid and In a short period of time valid and comparable results can be obtained.comparable results can be obtained.

• Sampling is economical since it contains Sampling is economical since it contains fewer units.fewer units.

• Research projects sometimes may require Research projects sometimes may require the destruction of the units being tested. the destruction of the units being tested. So it saves destruction of units.So it saves destruction of units.

Principles of SamplingPrinciples of Sampling

The main principle of sampling is that the The main principle of sampling is that the researcher seeks knowledge about the population researcher seeks knowledge about the population by observing a fewer units and extend the by observing a fewer units and extend the inference about the sample to the entire inference about the sample to the entire population. population.

The principles of sampling are :The principles of sampling are :•Sample units must be chosen in a systematic Sample units must be chosen in a systematic

and objective manner.and objective manner.•Sample units must be clearly defined and Sample units must be clearly defined and

easily identifiable.easily identifiable.•Sample units must be independent of each Sample units must be independent of each

other.other.

Types of SamplingTypes of Sampling

• Probability Sampling Probability Sampling

• Non-Probability SamplingNon-Probability Sampling

Probability SamplingProbability Sampling

Probability Sampling is one in which every Probability Sampling is one in which every unit of a population has a fixed probability of unit of a population has a fixed probability of

being included in the samplebeing included in the sample..Conditions to be satisfied for Probability sampling :Conditions to be satisfied for Probability sampling :• Complete list of units to be studied is available.Complete list of units to be studied is available.• Size of the population must be known.Size of the population must be known.• Each element must have an equal chance of Each element must have an equal chance of

being selected.being selected.• The form of the distribution of the trait in the The form of the distribution of the trait in the

population can be reasonably be assumed to be population can be reasonably be assumed to be normal.normal.

•Same units of sample should be used Same units of sample should be used throughout the study.throughout the study.

•The selection process should be based on The selection process should be based on some sound criteria and should avoid errors, some sound criteria and should avoid errors, bias and distortions.bias and distortions.

Types of Probability Types of Probability SamplingSampling

• Simple or Unrestricted Random Simple or Unrestricted Random samplingsampling

• Stratified Random SamplingStratified Random Sampling

• Area or Cluster SamplingArea or Cluster Sampling

• Multi-stage SamplingMulti-stage Sampling

• Multi-phase SamplingMulti-phase Sampling

Simple or Unrestricted Random Simple or Unrestricted Random SamplingSampling

This technique is applied in a homogeneous This technique is applied in a homogeneous population. Each unit of the population has an population. Each unit of the population has an equal chance of being selected for the sample.equal chance of being selected for the sample.

Methods Adopted :Methods Adopted :•LotteryLottery : :

– Step 1 : Constructing the Sampling FrameStep 1 : Constructing the Sampling Frame– Step 2 : Writing numbers listed in the sampling frame on Step 2 : Writing numbers listed in the sampling frame on

small pieces of papers and placing them in some small pieces of papers and placing them in some

vessel.vessel.– Step 3 : Mixing all papers well and drawing out one piece Step 3 : Mixing all papers well and drawing out one piece of paper at a time from the vessel.of paper at a time from the vessel.

•Using Random Number tableUsing Random Number table..

•Using Electronic DevicesUsing Electronic Devices..

Stratified Random SamplingStratified Random Sampling

This technique is applied in a This technique is applied in a heterogeneous population.heterogeneous population.

• Step I : The population is divided into a Step I : The population is divided into a number number of homogeneous strata or sub-of homogeneous strata or sub-groups groups based on one or more criteria.based on one or more criteria.

• Step II : A sample is drawn from each stratum Step II : A sample is drawn from each stratum using simple random technique.using simple random technique.

Types of Stratified Random Sampling :Types of Stratified Random Sampling :– Proportionate Stratified Random SamplingProportionate Stratified Random Sampling– Dis-proportionate Sratified Random SamplingDis-proportionate Sratified Random Sampling

Proportionate Stratified Proportionate Stratified SamplingSampling

Class X Students

(1000)

Urban(400)

Rural(600)

Boys(250)

Girls(150)

Boys(400)

Girls(200)

25( 10%) 15(10%) 40(10%) 20(10%)

Disproportionate Stratified Disproportionate Stratified SamplingSampling

Class X Students

(1000)

Urban(400)

Rural(600)

Boys(250)

Girls(150)

Boys(400)

Girls(200)

50( 20%) 50(33.3%) 50(12.5%) 50(25%)

Area or ClusterArea or Cluster

• Step-I : The population is divided into a Step-I : The population is divided into a number of clusters each having a number of number of clusters each having a number of units.units.

• Step-II : Instead of drawing a single unit Step-II : Instead of drawing a single unit clusters of units are drawn using simple clusters of units are drawn using simple random technique.random technique.

• Step-III : Usually clusters are drawn in multi-Step-III : Usually clusters are drawn in multi-stages.stages.

• Step-IV : Usually all the units of the clusters Step-IV : Usually all the units of the clusters drawn are used as the sample.drawn are used as the sample.

Multi-stage Random Multi-stage Random SamplingSampling

In this method, sampling is done in various In this method, sampling is done in various stages, but only the sample of the last units is stages, but only the sample of the last units is

studiedstudied• Step-I : The population is divided into a number of areas Step-I : The population is divided into a number of areas

or clusters.or clusters.• Step-II : A sample of clusters is drawn (1Step-II : A sample of clusters is drawn (1stst Stage) Stage)• Step-III : Each selected cluster is divided into more sub-Step-III : Each selected cluster is divided into more sub-

groups ( clusters).groups ( clusters).• Step-IV : A sample of sub-groups is drawn randomlyStep-IV : A sample of sub-groups is drawn randomly

(2(2ndnd stage). stage).• Step-V : Sample of units are drawn randomly from the Step-V : Sample of units are drawn randomly from the

selected sub-groups (3selected sub-groups (3rdrd stage) stage)

Multi-phase Random Multi-phase Random SamplingSampling

The process in this type of sampling The process in this type of sampling is same as in Multi-stage sampling; is same as in Multi-stage sampling; however, in a Multi-phase sampling, however, in a Multi-phase sampling, each sample is adequately studied each sample is adequately studied

before another sample is drawn from before another sample is drawn from it.it.

Types of Non-probability Types of Non-probability SamplingSampling

• Purposive Sampling or Judgmental Purposive Sampling or Judgmental SamplingSampling

• Accidental/Incidental SamplingAccidental/Incidental Sampling

• Volunteer SamplingVolunteer Sampling

• Quota SamplingQuota Sampling

• Systematic SamplingSystematic Sampling

• Snow Ball SamplingSnow Ball Sampling

Non-Probability SamplingNon-Probability Sampling

Units do not have a fixed probability of Units do not have a fixed probability of being included in the sample. This is being included in the sample. This is usually used when sampling frame is usually used when sampling frame is

not available.not available.

Purposive or Judgmental SamplingPurposive or Judgmental SamplingThe researcher purposively selects those individuals The researcher purposively selects those individuals

who are representative of the population. It is also known as who are representative of the population. It is also known as judgmental sample because the investigator selects the judgmental sample because the investigator selects the sample on the basis of his/her own judgment.sample on the basis of his/her own judgment.Advantages:Advantages:

(i) It is less costly and more readily accessible to the (i) It is less costly and more readily accessible to the researcher.researcher.

(ii) Purposive sampling guarantees that those (ii) Purposive sampling guarantees that those individuals will be included in the sample that are individuals will be included in the sample that are representative of the population.representative of the population.Disadvantages:Disadvantages:

(i) Purposive sampling introduces subjectivity in the (i) Purposive sampling introduces subjectivity in the sampling because too much emphasis is given on the ability sampling because too much emphasis is given on the ability of the investigator to assess which element is typical of the of the investigator to assess which element is typical of the population.population.

(ii) In the case of purposive sampling the inferential (ii) In the case of purposive sampling the inferential statistics can’t be used legitimately, because, under all statistics can’t be used legitimately, because, under all inferential statistical techniques, there is an assumption of inferential statistical techniques, there is an assumption of randomness.randomness.

Accidental / Incidental SamplingAccidental / Incidental Sampling

It refers to a sampling procedure It refers to a sampling procedure in which the investigator selects the in which the investigator selects the persons who are most conveniently persons who are most conveniently available or who accidentally come in available or who accidentally come in contact during a certain period of contact during a certain period of time in the research. time in the research.

Quota SamplingQuota Sampling

Steps followedSteps followed::(i) The population is divided into (i) The population is divided into

different strata.different strata.(ii) From each stratum a quota is (ii) From each stratum a quota is

selected.selected.Advantages:Advantages:

(i) Quota samples are the most (i) Quota samples are the most satisfactory means when quick and crude satisfactory means when quick and crude results are desired.results are desired.

(ii) It can guarantee the inclusion of (ii) It can guarantee the inclusion of individuals from different strata of individuals from different strata of population.population.

(iii) It is less costly. (iii) It is less costly.

Disadvantages:Disadvantages:

(i) The selected sample is not (i) The selected sample is not representative of the population because representative of the population because there is no means of establishing there is no means of establishing randomness.randomness.

(ii) In quota sampling, the investigator (ii) In quota sampling, the investigator selects the most accessible units/ individuals selects the most accessible units/ individuals which may not be typical of the population.which may not be typical of the population.

Systematic SamplingSystematic Sampling

This sampling is done by selecting or This sampling is done by selecting or drawing every nth person from a drawing every nth person from a predetermined list of elements or predetermined list of elements or individuals.individuals.

Advantages:Advantages:

(i) Systematic sampling is relatively a (i) Systematic sampling is relatively a quick method of obtaining a sample of quick method of obtaining a sample of elements.elements.

(ii) Systematic sampling makes it (ii) Systematic sampling makes it very easy to check whether every nth very easy to check whether every nth person has been selected.person has been selected.

Disadvantages:Disadvantages:

(i) Systematic sampling ignores (i) Systematic sampling ignores all persons between every nth all persons between every nth element chosen.element chosen.

(ii) In systematic sampling, the (ii) In systematic sampling, the sampling error increases if the list is sampling error increases if the list is arranged in a particular order, say, arranged in a particular order, say, the list increases or decreases with the list increases or decreases with respect to some trait such as age, respect to some trait such as age, education, income, caste, etc.education, income, caste, etc.

Snowball SamplingSnowball SamplingSome selected behaviour is usually used as Some selected behaviour is usually used as

the basis of contact and/or association in this type the basis of contact and/or association in this type of sampling. Snowball sampling is used for of sampling. Snowball sampling is used for obtaining an impression of informal social relations obtaining an impression of informal social relations among individuals.among individuals.

Steps followed:Steps followed:(i) At first few respondents are selected (i) At first few respondents are selected

who are known and available to the researcher.who are known and available to the researcher.(ii) These respondents give other (ii) These respondents give other

names who meet the criteria of research.names who meet the criteria of research.(iii) This process is continued until (iii) This process is continued until

adequate number of persons are interviewed or a adequate number of persons are interviewed or a constellation of friendships converges into some constellation of friendships converges into some type of a definite social pattern.type of a definite social pattern.

Advantages:Advantages:

(i)(i) Snowball sampling is important and Snowball sampling is important and helpful in studying small social helpful in studying small social group and its impact upon formal group and its impact upon formal organizational structure.organizational structure.

(ii)(ii) This method is employed when the This method is employed when the target population is unknown or target population is unknown or when it is difficult to approach the when it is difficult to approach the respondents in any other way.respondents in any other way.

Disadvantages:Disadvantages:

(i)(i) Snowball sampling becomes Snowball sampling becomes cumbersome and difficult when N is cumbersome and difficult when N is large.large.

(ii)(ii)The elements included in sample The elements included in sample are dependent on the subjective are dependent on the subjective choices of the originally selected choices of the originally selected respondents and thereby respondents and thereby introduces some bias in the introduces some bias in the sampling.sampling.

Volunteer SamplingVolunteer Sampling

It refers to those sampling in It refers to those sampling in which the respondent himself/herself which the respondent himself/herself volunteers to give information.volunteers to give information.

SOME RESEARCH PROBLEMS:SOME RESEARCH PROBLEMS:

• Mathematical creativity and some socio-Mathematical creativity and some socio-psychological factors of Hindus and Muslims.psychological factors of Hindus and Muslims.

• Construction and Standardization of a Construction and Standardization of a Multiple-choice Reading Comprehension test.Multiple-choice Reading Comprehension test.

• Achievement Level of Primary School Students Achievement Level of Primary School Students at the end of Class IV.at the end of Class IV.

• Effects of intervention training on some Effects of intervention training on some cognitive abilities.cognitive abilities.

• A study to determine the effects of exposure A study to determine the effects of exposure to pre-school education on cognitive abilities.to pre-school education on cognitive abilities.

Data-Gathering ToolsData-Gathering Tools

The major data-gathering tools of The major data-gathering tools of research are:research are:

• Psychological TestsPsychological Tests

• Inquiry FormsInquiry Forms

• ObservationObservation

• InterviewInterview

• Sociometric TechniqueSociometric Technique

Psychological TestsPsychological Tests

Psychological tests are devised to evaluate Psychological tests are devised to evaluate or measure behaviour in a standardized way or measure behaviour in a standardized way for the purpose of selection, classification, for the purpose of selection, classification, prediction and guidance as well as for the prediction and guidance as well as for the evaluation of educational programmes. evaluation of educational programmes. They are designed to measure general They are designed to measure general mental ability or intelligence, special mental ability or intelligence, special abilities or aptitudes, creativity, abilities or aptitudes, creativity, achievement, personality traits and achievement, personality traits and adjustment, interests and values.adjustment, interests and values.

What are tests ?What are tests ?

• Tests consist of a series of tasks Tests consist of a series of tasks which the subject is required to which the subject is required to perform.perform.

• Cronbach (1964) defines a test as a Cronbach (1964) defines a test as a systematic procedure for comparing systematic procedure for comparing the behaviour of two or more persons the behaviour of two or more persons at a particular time ; or one or more at a particular time ; or one or more persons at different time.persons at different time.

Standardized Test:Standardized Test:

A standardized test is one that has A standardized test is one that has specific directions for administration specific directions for administration and scoring, a fixed set of test items, and scoring, a fixed set of test items, and has been administered to and has been administered to representative samples taken from representative samples taken from the population for whom the test is the population for whom the test is intended for the purpose of intended for the purpose of establishing norms (Koul, 1997). establishing norms (Koul, 1997).

Classification of TestsClassification of Tests

• On the basis of Time limit in producing On the basis of Time limit in producing the response:the response:

a) Power Testa) Power Test

b) Speed Testb) Speed Test

• On the basis of Administrative On the basis of Administrative conditions:conditions:

a) Group testa) Group test

b) Individual testb) Individual test

• On the basis of Purposes:On the basis of Purposes:

a) The Tests of General mental a) The Tests of General mental ability or Intelligenceability or Intelligence

b) The Tests of Special ability or b) The Tests of Special ability or AptitudesAptitudes

c) The Tests of Creativityc) The Tests of Creativity

d) The Tests of Attainment or d) The Tests of Attainment or AchievementAchievement

e) The Personality measurese) The Personality measures

• On the basis of Standardization:On the basis of Standardization:

a) Standardized Testsa) Standardized Tests

b) Non-Standardized Teacher b) Non-Standardized Teacher made testsmade tests

• On the basis of Nature of Items:On the basis of Nature of Items:

a) Verbal Testa) Verbal Test

b) Nonverbal Testb) Nonverbal Test

c) Performance testc) Performance test

• On the basis of Scoring:On the basis of Scoring:

a) Objective Type Testa) Objective Type Test

b) Essay Type Testb) Essay Type Test

• Power Test:Power Test: A power test is one which has a A power test is one which has a generous time limit so that most generous time limit so that most examinees are able to attempt every examinees are able to attempt every item.item.

• Speed Test:Speed Test:A speed test is one which has severe A speed test is one which has severe time limit so that no examinees are time limit so that no examinees are able to attempt all items. Speed test able to attempt all items. Speed test measures how rapidly or with what measures how rapidly or with what speed the examinees can respond speed the examinees can respond within a given time limit.within a given time limit.

Characteristics of TestsCharacteristics of Tests

• ValidityValidity• ReliabilityReliability• UsabilityUsability

a) Objectivitya) Objectivityb) Economy of Timeb) Economy of Timec) Costc) Costd) Ease of Administrationd) Ease of Administratione) Ease of Scoringe) Ease of Scoringf) Ease of Interpretation f) Ease of Interpretation

ValidityValidity

Meaning: The term ‘Validity’ means Meaning: The term ‘Validity’ means truthfulness or fidelity.truthfulness or fidelity.

Definition:Definition:

• Validity refers to the degree to which Validity refers to the degree to which a test measures what it claims to a test measures what it claims to measure.measure.

• The validity of a test concerns what The validity of a test concerns what the test measures and how well it the test measures and how well it does so (Anastasi, 1988)does so (Anastasi, 1988)

• Validity of a test is defined as the Validity of a test is defined as the accuracy with which it measures that accuracy with which it measures that which is intended to measure or as which is intended to measure or as the degree to which it approaches the degree to which it approaches infallibility in measuring what it infallibility in measuring what it purports to measure (Lindquist, purports to measure (Lindquist, 1951)1951)

Types of Validity:Types of Validity:

• Content ValidityContent Validity

• Criterion-Related ValidityCriterion-Related Validity

a) Concurrent Validitya) Concurrent Validity

b) Predictive Validityb) Predictive Validity

• Construct ValidityConstruct Validity

Content validity:Content validity:

Content validity is the Content validity is the representativeness or sampling representativeness or sampling adequacy of the content – the adequacy of the content – the substance, the matter, the topics – of substance, the matter, the topics – of a measuring instrument. It is a measuring instrument. It is estimated by evaluating the estimated by evaluating the relevance of the test items, in relevance of the test items, in relation to instructional objectives relation to instructional objectives and actual subject matter studied.and actual subject matter studied.

Criterion-Related Validity:Criterion-Related Validity:

Criterion related validity is one which Criterion related validity is one which is obtained by comparing (or is obtained by comparing (or correlating) the test scores with correlating) the test scores with scores obtained on a criterion scores obtained on a criterion available at present or to be available at present or to be available in the future.available in the future.

a) Concurrent Validitya) Concurrent Validity::

The degree to which a test The degree to which a test is correlated with a criterion which is is correlated with a criterion which is available at the present time.available at the present time.

b) Predictive Validity:b) Predictive Validity:

The degree to which a test is The degree to which a test is correlated with a criterion which will correlated with a criterion which will be available in the future. It refers to be available in the future. It refers to the association between present the association between present results as indicated by a test and results as indicated by a test and future behaviour.future behaviour.

Construct ValidityConstruct Validity::

The degree to which certain psychological The degree to which certain psychological traits or constructs are actually traits or constructs are actually represented by test performance. Construct represented by test performance. Construct validation is an analysis of the meaning of validation is an analysis of the meaning of test scores in terms of psychological test scores in terms of psychological constructs (Cronbach and Meehl , 1955).constructs (Cronbach and Meehl , 1955).

Example: If a test is designed to measure Example: If a test is designed to measure emotional stability, what kinds of evidence emotional stability, what kinds of evidence are necessary before we can have are necessary before we can have confidence that the information provided confidence that the information provided by the test reflects this factor ?by the test reflects this factor ?