SAMPLING - PowerPoint PPT Presentation

1 / 33
About This Presentation
Title:

SAMPLING

Description:

EDUCATIONAL RESEARCH WHY SAMPLE? Not needed when can access entire population. Must sample when not feasible to access entire population. Sample Should Be ... – PowerPoint PPT presentation

Number of Views:137
Avg rating:3.0/5.0
Slides: 34
Provided by: John757
Category:

less

Transcript and Presenter's Notes

Title: SAMPLING


1
SAMPLING
  • EDUCATIONAL RESEARCH

2
WHY SAMPLE?
  • Not needed when can access entire population.
  • Must sample when not feasible to access entire
    population.

3
Sample Should Be Representative of the Larger
Population...
4
IMPORTANT TERMS
  • Sample--subset of people selected to study
  • Population--larger group from which sample comes
    to which we will infer

5
STEPS IN SAMPLING
  • ID ideal/preferred population
  • Identify accessible population
  • Use Principle of Sampling

6
Principle of Sampling
  • If sample is representative of accessible
    population, findings can be generalized to
    population
  • Generalizing from sample to population involves
    risks

7
TYPES OF SAMPLING PROCEDURES
  • Probability
  • Nonprobability

8
PROBABILITY SAMPLING
  • Participants drawn by chance (random)
  • Every member has known probability of being
    chosen (110 1100)
  • EPSEM--equal probability of selection method

9
Probability Sampling Types
  • Simple Random Sampling
  • Two-Stage Random Sampling
  • Stratified Sampling
  • Cluster Sampling
  • Systematic Sampling

10
SIMPLE RANDOM SAMPLING
  • All have equal and independent chance of selection

11
Performing Random Sampling
  • Define and list ALL population members
  • Assign each a from 0 to ????
  • Group columns of digits according to needed
  • Two digits for s up to 99
  • Three digits for s up to 999

12
Performing Random Sampling
  • Arbitrarily select a in the random number table
  • If selected corresponds to of member of
    identified population, member is in sample
  • go down list
  • repeat above selection method until desired of
    participants obtained

13
Two-Stage Random Sampling
  • Useful in large populations
  • Select random clusters
  • Select random individuals from random clusters

14
STRATIFIED SAMPLING
  • Information known about total population prior to
    sampling
  • Know population subgroups/strata
  • Distinguish all elements in population according
    to value on characteristic(s)

15
Performing Stratified Sampling
  • Identify subgroups/strata
  • Select randomly a specific of subjects from
    each stratum
  • Revisit random sampling

16
Cluster Sampling
  • Unit chosen is NOT an individual, but a group of
    individuals naturally together
  • Constitute a cluster
  • Are alike with respect to characteristics
    relevant to the study

17
Performing Cluster Sampling
  • Chosen at random from population of clusters
  • All members of clusters chosen must be included
    in sample
  • Using clusters as individuals, follow steps
    outlined in random sampling

18
Examples of Cluster Sampling
  • Schoolsclusters for sampling students
  • Blocks clusters for sampling residents
  • Counties clusters for sampling general
    populations
  • Businessesclusters for sampling employees

19
SYSTEMATIC SAMPLING
  • Convenient to draw a random sample when
    population elements are arranged sequentially

20
Systematic Sampling
  • Variation of simple random sampling
  • Different choices not independent
  • Yields simple random sample in most, but NOT all,
    sampling situations
  • If sequence varies in regular, periodic pattern,
    then will not have a random sample---rarely occur

21
Performing Systematic Sampling
  • Determine size of sample
  • Divide sample into population
  • Randomly select starting point
  • Select every nth subject
  • Need 5 people, have 45 in pop, select every 9th
    person once starting point chosen

22
Nonprobability Sampling
  • Probability of selection of population elements
    is NOT known.
  • Participants NOT chosen by chance
  • Cannot estimate likelihood selection
  • More convenient and economical

23
NONPROBABILITY SAMPLING TYPES
  • Accidental Purposive
  • Quota Snowball
  • Convenience
  • Cannot expect representative sample using
    nonprobability methods

24
ACCIDENTAL SAMPLING
  • Haphazard, availability, or convenience
  • Interviewing/surveying first X number of
    individuals encountered
  • EXTREMELY weak, but fairly popular
  • Most psych research is accidental

25
PURPOSIVE SAMPLING
  • Subjects judged to be representative are chosen
    from larger population
  • Doesnt produce representative sample
  • Results may be misleading

26
PURPOSIVE SAMPLING
  • Each sample selected for purpose, usually because
    of unique position of sample element
  • Used in national elections (referred to as
    bell-weather districts)

27
QUOTA SAMPLING
  • Selection of typical cases from diverse strata of
    a population
  • Must know characteristics of entire population to
    set right quotas
  • Approximation of population with respect to
    selected characteristics

28
SNOWBALL SAMPLING
  • Useful for hard to reach or identify, but
    interconnected populations
  • May consider when cannot think of another method
  • Generalizations are very tentative

29
Performing Snowball Sampling
  • Identify one member of a population
  • Speak with member
  • Ask member to identify others
  • Speak with those members
  • Ask those members to identify more
  • and so on and on and on....

30
Snowball Sampling Examples
  • drug dealers
  • prostitutes
  • practicing criminals
  • gang leaders
  • Alcoholics Anonymous members

31
CONVENIENCE SAMPLING
  • When random is impossible
  • Individuals who are available
  • Likely to be biased
  • Not representative of any population
  • Should be avoided if possible
  • Should be replicated

32
Sample Representativeness
  • Sampling Goal representativeness
  • Larger the sample, more confidence we have in
    representativeness of sample
  • More homogeneous the population, the more
    confidence of representativeness

33
(No Transcript)
Write a Comment
User Comments (0)
About PowerShow.com