Scribd Logo
Upload

Welcome to Scribd!

  • 29 views

    Stat-304 Lecture 2

    Uploaded by

    ahmed.iqbal2907

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Stat-304 Lecture 2

    Uploaded by

    ahmed.iqbal2907
    29 views14 pages

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    29 views14 pages

    Stat-304 Lecture 2

    Uploaded by

    ahmed.iqbal2907

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    You are on page 1of 14

    STAT-304

    SAMPLING TECHNIQUE II
    Lecture 2
    Reference

     Sampling: Design and Analysis (Advanced Series) 2nd Edition (2010) by


    Sharon L. Lohr. Chapter 5, Page# 182-185

     Singh, R. and Singh N, (1996), “Elements of Survey Sampling”, Kulwar,


    Dodrecht. Chapter 11, Page # 283-313
    Multistage Sampling
     With cluster sampling, the smaller the size of the clusters the better is.
    When (1) the units are distributed over a larger number of clusters and (2)
    instead of completely enumerating all the units in each selected clusters,
    only a sample of units is observed.
     For example, in a High School example, the students could be grouped by
    grade levels or classes; here grade levels are too large to serve as clusters
    for sampling purposes, and classes are the obvious choice.
     The problem with cluster sampling is that, because clusters usually
    comprise existing groupings that were formed for other purposes, the
    lowest level of clustering still often yields clusters that are too large to be
    used efficiently in cluster sampling.
     The solution to this problem is to divide the clusters into sub-clusters for
    sampling purposes; essentially this is what is done in multistage sampling.
    Two-Stage Cluster Sampling
     ‘The procedure of sampling , which consists in first selecting the clusters
    and then randomly choosing a specified number of units from each
    selected cluster, is known as two stage sampling

     There are two desirable conditions for selecting appropriate clusters: –


    Geographic proximity (Closeness) of the elements within a cluster –
    Cluster sizes that are convenient to handle.

     Example: Consider an university student opinion poll. – If the students in


    an university hold similar opinions but differ widely from university to
    university, the sample should contain few representatives from many
    different universities. – If the students in an university vary greatly, the
    sample should contain many representatives from each of a few
    universities.
    Cont.
     In two-stage cluster sampling, the sample of elements is obtained as a
    result of two stages of sampling.

     The population elements are first grouped into disjoint


    subpopulations, called primary sampling units (PSU). Then, in a first-
    stage sampling, a sample of PSU is drawn.

     In the second-stage sampling units (SSU) may be clusters of elements,


    for each PSU in the first-stage sample.

     A sample of SSU is drawn (second-stage sampling) from each PSU in


    the first-stage sample. When the SSU are clusters, every element in
    the selected SSU is surveyed.
    Estimator 1

    (Singh, R. and Singh N, (1996), “Elements of Survey Sampling”) P#285)


    Estimator 2

    ((Singh, R. and Singh N, (1996), “Elements of Survey Sampling”) P#288)


    Estimator 3

    (Solved Examples Related to all (estimators 1,2,3) from Book)

    ((Singh, R. and Singh N, (1996), “Elements of Survey Sampling”) P#289)


    Estimator of Population Total

    (Solved Examples Related to all estimator from Book)

    ((Singh, R. and Singh N, (1996), “Elements of Survey Sampling”) P#294)


    Exercise Questions
     11.4: A district is running 110 nursery schools (anganwadi) in
    the rural area. The total number of children in these schools, is
    knowns to be 8040. The Department of Foods and Nutrition of a
    university has undertaken a project to determine the quality of
    food intake by the children in these schools. For this purpose, a
    sample of 10 was selected using WOR equal probability
    sampling. From each selected school, about 5 percent of
    children were selected using same procedure.

    (Singh, R. and Singh N, (1996), “Elements of Survey Sampling”)


    Exercise Page # 308-310.)
    Elaborate diet records were kept for each selected child, and the average daily
    calorie intake was determined. The information thus collected, is presented in
    the following table.

    School Average calorie intake for sample children


    1 100 5 650 680 712 766 770
    2 93 5 745 690 703 740 671
    3 40 2 736 692
    4 80 4 703 777 687 714
    5 42 2 699 724
    6 62 3 760 704 680
    7 96 5 715 660 690 670 650
    8 46 2 668 701
    9 99 5 704 712 692 697 687
    10 104 5 714 716 704 725 730

    Estimate unbiasedly the average daily calorie intake for the


    children of all 110 schools, and also obtain the confidence
    interval for it.
     11.7: The excessive rains have caused damage to cotton crop in a certain
    area consisting of 132 villages. Before deciding on the extent of relief to be
    given to cultivators, the administration has decided to estimate the average
    per hectare loss for the area. For this purpose, a sample of 10 villages was
    selected using without replacement SRS, and about 5 percent of cultivators
    from the sample villages were selected using the same sampling scheme. Per
    hectare loss (in '00 rupees) incurred by each selected cultivator was assessed
    by visiting his fields. The results are as follows:

    Carton Loss per hectare


    1 43 2 8 12
    2 76 4 16 7 11 13
    3 46 2 14 9
    4 67 3 17 10 12
    5 97 5 14 10 15 17 11
    6 75 4 16 9 7 14
    7 83 4 9 13 11 13
    8 54 3 15 15 13
    9 69 3 12 9 11
    10 58 3 14 15 14

    Work out the estimated average per hectare loss, and also
    construct the confidence interval for it.

    You might also like