Scribd Logo
Upload

Welcome to Scribd!

  • 101 views

    Statistical Sampling and Sampling Techniques

    Uploaded by

    sumit71sharma
    Statistical sampling techniques allow conclusions to be drawn about a population by examining only a subset of the data. Common sampling methods include estimation sampling, acceptance sampling, discovery sampling, and judgement sampling. Key steps in designing a sample include determining the sample size, precision level, confidence level, and occurrence rate in the population. Sample results can then be projected back to the overall population within known confidence intervals.

    Copyright:

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

    Statistical Sampling and Sampling Techniques

    Uploaded by

    sumit71sharma
    101 views13 pages
    Statistical sampling techniques allow conclusions to be drawn about a population by examining only a subset of the data. Common sampling methods include estimation sampling, acceptance sampling, discovery sampling, and judgement sampling. Key steps in designing a sample include determining the sample size, precision level, confidence level, and occurrence rate in the population. Sample results can then be projected back to the overall population within known confidence intervals.

    Original Description:

    maths short

    Copyright

    © © All Rights Reserved

    Available Formats

    PPT, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Statistical sampling techniques allow conclusions to be drawn about a population by examining only a subset of the data. Common sampling methods include estimation sampling, acceptance sampling, discovery sampling, and judgement sampling. Key steps in designing a sample include determining the sample size, precision level, confidence level, and occurrence rate in the population. Sample results can then be projected back to the overall population within known confidence intervals.

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    101 views13 pages

    Statistical Sampling and Sampling Techniques

    Uploaded by

    sumit71sharma
    Statistical sampling techniques allow conclusions to be drawn about a population by examining only a subset of the data. Common sampling methods include estimation sampling, acceptance sampling, discovery sampling, and judgement sampling. Key steps in designing a sample include determining the sample size, precision level, confidence level, and occurrence rate in the population. Sample results can then be projected back to the overall population within known confidence intervals.

    Copyright:

    © All Rights Reserved
    Download as PPT, PDF, TXT or read online from Scribd
    Download as ppt, pdf, or txt
    You are on page 1of 13

    Statistical Sampling and Sampling

    Techniques

    Presented by Mr. Preman Dinaraj,


    ITEC Expert.
    INTRODUCTION

    • Statistical sampling techniques useful to draw conclusions based


    on inspection of part of a population. Procedure relies on less than
    100% population.
    • Relevant for- Survey sampling, quality control, accuracy of records,
    social opinion among others.
    • Practitioner must have some background in statistical course in their
    education.
    • Effective sampling requires more than objectivity. Requires means for
    establishing sample sizes. Also to apprise sample size mathematically.
    • Behaviour will be measurable with reference to Theory of Probability.
    • Samples obtained statistically- can state with degree of confidence that
    errors in sample applies proportionately to unsampled portion as well.
    SAMPLING WITH STATISTICAL
    MEASUREMENTS
    • Advantages- (a) Sample result will be objective and defensible. Bias
    arising out of judgemental sample is avoided here.
    • (b) Provides means of knowing before hand the maximum size of
    sample needed. Justify expenses or time frame w.r.to reasonable
    confidence level for the risk being evaluated.
    • (c) Provides estimate of degree of risk reg. non-rep character of sample
    to population. Limits deviations due to sampling variations.
    • (d) Can be more accurate than examining every item of population.
    Errors of omission or fact by the user.
    • (e) Saves time and money. Fewer items and one sample can test several
    characteristics of a population.
    • (f) Can objectively evaluate test results. Results can be projected
    within known limits of reliability.
    • (g) Data even from different users can be combined and evaluated.
    STATISTICAL TERMS

    • (A) Average- Primary measurement of central tendency of a


    variable. Calculated by sum of all variables divided by number of
    occurrences.

    • (B) Range- Difference between the most extreme values of a


    variable.

    • (C) Standard Deviation- Measure of the distance of all values


    from the arithmetic mean. Most useful measure of dispersion.

    • (D) Reliability or Confidence Level- Common sense notion of


    accuracy. To be used with concept of precision. Indicates probability of
    sample with the entire population.
    • (E) Precision- Common sense notion of accuracy. To be used with
    reliability. Another way of describing confidence level. Indicates range
    of values of a statistic measured by a sample, having a given probability
    of containing the true value of population’s statistics. Expressed as +/-.

    • (F) Confidence Interval- Is the plus or minus interval about the


    sample static. Is another way of expressing precision.

    • (G) Frequency Distribution- Classification of the elements of a set


    of data by a quantitative characteristic. More the classes in a frequency
    distribution, more the details shown. Too much details hinder
    summarisation.
    SAMPLING PLANS
    •Basic Hypothesis- (a) Random selection-Population is a homogenous group
    •(b) Non-biased selection of sample items. All items of population have equal
    chance of being selected in the sample.
    •(c ) Estimate how many departures from prescribed procedure, or estimate a
    quantity like the value of errors in the population. Manner of population
    distribution will determine the Plan. The selection of Plan be precisely
    documented.
    •Some commonly used Sampling Plans are-

    •(1) Estimation Sampling- Most widely used. are of 2 types-


    • (a) Attribute Sampling- To be used when question of “how many” is
    pertinent. Used to determine the characteristics or attributes of population.
    Results expressed as % of the type of event and each observation can fall only in
    1 category.
    • Stop or Go Sampling- Extension of attributes sampling. Used to
    find out upper precision limit of attribute sample. Possible to reach
    objective with a smaller sample size than attributes model.
    • (b) Variable sampling- Used to answer “how much”. Useful for
    populations like dollars, days etc. Provides estimate of average or total
    value of population.
    • (2) Acceptance Sampling- Sample size predetermined by random
    sampling method. If not more than a given number of errors is found,
    the sample is accepted. Allows only for acceptance or rejection of
    sample.
    • (3) Discovery sampling- A.k.a exploratory sampling. Useful
    where evidence of single error or irregularity calls for intensive
    investigation like fraud avoidance, evasion of regulation or for quality
    control issues.
    •
    • (4) Judgement Sampling- User determines sample size and method
    of selection of the sample. For best results, user must have previous
    background of sampling methods as well as the samples. Caution- One
    can’t make an inference to the population as a whole using judgement
    sampling.
    COMMON SAMPLING SELECTION
    TECHNIQUES
    • 1. Unrestricted Random Numbers- Each item in the sample has
    an equal chance of being included in the sample. Most commonly used
    method.
    • 2. Interval Sampling- Sample selected with uniform interval
    between each sample item after random start.
    • 3. Stratified Sampling- Items in the population segregated into 2
    or more classes or strata. Each strata then sampled independently.
    Results of several strata can be combined to obtain results or be
    considered separately as per requirement.
    • 4. Cluster Sampling- Form the population into groups or clusters
    of items. Examine the items within the selected clusters in entirety.
    • 5. Multistage Samples- Involves sampling on several levels. The
    user takes a sample from several locations and then takes another
    sample from within the sampled items.
    DESIGNING A SAMPLE

    • Decide the sampling method. Thereafter the stages will be-


    • (a) Determine the sample size. First step- clearly define the target
    population and the error or exception being tested.
    • (b) Determine the tolerable error or the maximum error acceptable to
    test that the procedures are being followed.
    • (c) Decide the required Precision level. Estimate for Precision Level by
    finding the variation between sample estimate and actual population.
    Also decide the Tolerable Error Level.
    • (d) Define the Confidence Level of Assurance Level. If sampling
    process preceded by risk analysis, then Confidence Level will be (1-
    Detection risk).
    • (e) Test the Occurrence Rate or population proportion i.e. the
    proportion of items in the population having the error or exception. If
    Occurrence Rate in population becomes larger, the sample size to be
    examined has to be increased.
    • (f) Select the sample and perform substantive test on it.
    PROJECTING THE RESULTS

    • Last stage in evaluating test results.


    • (a) Evaluate findings for each characteristic being tested separately.
    Each characteristic represents a distinct and independent sample.
    • (b) Based on determination of “Acceptable Error Rate” test the results.
    • (c) Never hesitate to take a fresh look at the sample when significant
    matters are revealed by the tests. In case of critical exception being
    detected, determine the cause and effect of the exception.

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy