08-Nov-19

Introduction

►Obtaining a portion, or sample, that is

Sampling and Sample representative of the whole population is referred
to as sampling
Preparation ►Total quantity from which the sample is obtained
Dr. Imran Pasha is called population
Associate Professor ►Selection of a limited number of samples for
National Institute of Food Science & Technology, analysis is of great benefit because
UAF • It allows a reduction in time, expense and personnel required to carry out
the analytical procedure

FST-511 winter 2019-20 1 FST-511 winter 2019-20 2

Introduction… Selection of sampling procedures

Why sample selection is important? ►Selection of an appropriate fraction of the whole material is
• One of the most important stages of food analysis procedures
• It can lead to large errors when not carried out correctly
►The primary objective of sample selection
• To ensure that the properties of the laboratory sample are representative of
the properties of the population, otherwise erroneous results will be
obtained

FST-511 winter 2019-20 3 FST-511 winter 2019-20 4

1
 08-Nov-19

Selection of sampling procedures… Selection of sampling procedures…

►Samples may be taken from two types of populations Sampling Sample preparation
• Finite population
►Size of a lot
• Infinite population
►Number of temperature observations made of a
lot over time Data obtained from an
analytical technique

►Regardless of population type the data obtained from Data interpretation Laboratory analysis
sampling are compared to a range of acceptable values
to ensure the population sampled is within
specifications
Data processing

FST-511 winter 2019-20 5 FST-511 winter 2019-20 6

Selection of sampling procedures… Accuracy vs Precision

►There is a potential for error at each step

►The uncertainty or reliability of the final result
depends on the cumulative errors at each stage
• Variance is an estimate of the uncertainty
• Precision is a measure of the reproducibility of the data
• Accuracy is a measure how close the data are to the true value
►The reliability of sampling is dependent more on
the sample size than on the population size

FST-511 winter 2019-20 7 FST-511 winter 2019-20 8

2
 08-Nov-19

Sampling plan Sampling plan…

►The International Union of Pure And Applied Chemistry (IUPAC) • A sampling plan should be a well organized document that establishes the
defines a Sampling plan as required procedures for accomplishing the program’s objectives
“a predetermined procedure for the selection, withdrawal, preservation, • It should address the issues of who, what, where, why and how
transportation and preparation of the portions to be removed from a lot of
samples”
►The primary aim of sampling is to obtain a sample,
subject to constraints on size, that will satisfy the
sampling plan specifications

FST-511 winter 2019-20 9 FST-511 winter 2019-20 10

Sampling plan… Sampling plan…

A sampling plan should be selected on the basis of The two primary objectives of sampling

1. To estimate the average value of a characteristic

►The sampling objectives
►The study populations 2. To determine the average value meets the specifications defined the
sampling plan
►The statistical unit
►The sample selection criteria
►The analysis procedure

FST-511 winter 2019-20 11 FST-511 winter 2019-20 12

3
 08-Nov-19

Sampling plan… Sampling for Attributes or Variables

Factors affecting choice of sampling plan Sampling plan are designed for examination of either Attributes or
1. Purpose of inspection Variables
• Is it to accept or reject the lot? 1. Attributes sampling
• Is it to measure the average quality of lot?
• Is it to determine the variability of the product? ► It is performed to decide on the acceptability of a
2. Nature of product population based on whether the sample possesses a
•
•
Is it homogenous or heterogeneous ?
What is the unit size?
certain characteristic, e.g. Clostridium botulinum
• How consistently have past population met specifications? contamination in canned goods
• What is the cost of the material being sampled?
► It provides data that are in dichotomous form that is,
3. Nature of test method
• Is the test critical or minor? data for which there exist two possible alternative, such
• Will someone become sick or die if the population fails to pass the test? as present or absent
• Is the test destructive or non destructive?
• How much the test cost to complete? ► The data gives hypergeometrical, binomial or poisson
4. Nature of population being investigated distribution
• Is the lot large but uniform?
• Does the lot consist of smaller,FST-511
easily identifiable sublots?
winter 2019-20 13 FST-511 winter 2019-20 14
• What is the distribution of the units within the population?

Sampling for Attributes or Variables… Sampling for Attributes or Variables…

Sampling plan are designed for examination of either Attributes or
Variables… Sampling plan are designed for examination of either Attributes or
Variables…
2. Variable sampling
► It is performed to estimate on a continuous scale
• Quantitatively the amount of a substance, e.g. salt ►In general variable sampling requires smaller sample
• A characteristic, e.g. color size than attribute sampling
► The estimate obtained from the sample is compared
with an acceptable value (previously determined) and ►An attribute is something that a product either does or
the deviation measured does not have
► The data gives normal diatribution such as in the ►On the other hand, a variable is some property that
percent fill of a container and total solids of a sample can be measured on a continuous scale, e.g. the weight,
food fat content or moisture content of a material

FST-511 winter 2019-20 15 FST-511 winter 2019-20 16

4
 08-Nov-19

Types of sampling plans

Risks associated with sampling
There are three basic types of sampling plans
1. Single sampling plans
• Allow accept/reject decisions to be made by inspection of Two types of risks are there which must be
one sample of a specified size considered when developing a sampling plan
2. Double sampling plans 1. Consumer risks
• Require the selection of two sample sets ► Describes the probability of accepting poor quality
3. Multiple sampling plans population-happen rarely (<5%)
• Amount of samples depend upon the overall quality of the lot ► These may vary from major health hazards and
• Multiple sampling charts must be developed to relate the cumulative subsequent facilities to a lot being of slightly lower
number of defects to the number of samples taken from the lot quality than standard lots
• Chart consists of two parallel lines 2. Vendor risks
► A rejection line
When the cumulative number of defects lies above the rejection line the lot ► Describes the probability of rejecting an acceptable
should be rejected product happen usually (5-10%)
► An acceptance line
When the cumulative number of defects falls below the acceptance line the
lot can be accepted
FST-511 winter 2019-20 17 FST-511 winter 2019-20 18

Sampling procedures Sampling procedures…

There are following types of sampling A. Probability sampling

A. Probability sampling ► Provides a statistically sound basis for obtaining

1. Simple random sampling representative samples with elimination of
2. Systematic sampling human bias
3. Stratified sampling • Therefore most desirable
4. Cluster sampling ► The probability of including any item in the
5. Composite sampling sample is known and sampling error can be
B. Non Probability Sampling calculated
1. Quota Sampling ► Probability sampling is used when the research is conclusive in
2. Accidental Sampling nature. On the other hand, when the research is exploratory,
nonprobability sampling should be used

FST-511 winter 2019-20 19 FST-511 winter 2019-20 20

5
 08-Nov-19

Sampling procedures… Sampling procedures…

1. Simple random sampling 2. Systematic sampling

► Requires that the number of units in the ► Is used when complete list of sample units is not
population be known and each unit is assigned a available but when sample are distributed evenly
number over time or space, e.g. on a production line
► A specified quantity of random numbers between ► The first sample is selected at random and then
one and the total number of population units is every nth unit after that
selected
► The variance is difficult to determine
► Sample size is determined by lot size and the
potential impact of consumer or vendor error

FST-511 winter 2019-20 21 FST-511 winter 2019-20 22

Sampling procedures…
3. Stratified sampling

► Involves dividing the samples into overlapping

subgroups so that each subgroup is as homogenous
as possible
► Group differs from each other as much as possible
► From each subgroup random samples are taken
► The procedure provides a representative sample
because no part of the population is excluded and it
is less expensive than simple random sampling

FST-511 winter 2019-20 23 FST-511 winter 2019-20 24

6
 08-Nov-19

Sampling procedures…
4. Cluster sampling

► Involves dividing the population into subgroup, or

clusters, to make similar groups of clusters
• Clusters should be small with a similar number of units in each cluster
► Any heterogeneity occurs within each cluster
► The clusters are sampled randomly and may be either
totally inspected or subsampled for analysis
► More efficient and less expensive method than simple
random sampling, if populations can be divided into
homogeneous groups

FST-511 winter 2019-20 25 FST-511 winter 2019-20 26

Sampling procedures…
5. Composite sampling

► Used to obtain samples from bagged products, e.g.

flour, seeds and larger items in bulk
► Two or more samples are combined to obtain one
sample for analysis that reduces differences between
samples
► e.g. FDA and FSIS composite 12 and atleast six
subsamples, respectively, for the sample to be
analyzed for compliance with nutrition labeling
regulations
FST-511 winter 2019-20 27 FST-511 winter 2019-20 28

7
 08-Nov-19

B. Non-Probability Sampling Non-Probability sampling

• nonprobability sampling does not involve random selection

A. Accidental, Haphazard or Convenience Sampling
B. Purposive Sampling
1. Expert Sampling
2. Quota Sampling
3. Heterogeneity Sampling
4. Snowball Sampling

FST-511 winter 2019-20 29 FST-511 winter 2019-20 30

1. Expert Sampling
involves the assembling of a sample of persons with known or demonstrable
experience and expertise in some area
2. Quota Sampling
• Accidental, Convenience Sampling select people nonrandomly according to some fixed quota (40% women and
• the traditional "man on the street" interviews conducted frequently by 60% men)
television news programs to get a quick (although nonrepresentative) 3. Heterogeneity Sampling
reading of public opinion
when we want to include all opinions or views, and we aren't concerned
• Purposive Sampling about representing these views proportionately
• sample with a purpose in mind What we would like to be sampling is not people, but ideas
• people in a mall or on the street who are stopping various people and asking
if they could interview them? (market research) 4. Snowball Sampling
• begin by identifying someone who meets the criteria for inclusion in your
study. You then ask them to recommend others who they may know who
also meet the criteria

FST-511 winter 2019-20 31 FST-511 winter 2019-20 32

8
 08-Nov-19

Problems in sampling…
Problems in sampling
6. Preservatives can be used to stabilize certain food
1. Poor sample storage results in sample degradation substances
• (e.g. mercuric chloride, potassium dichromate, chloroform)
2. Sample should be stored in containers (air tight)
• To protect samples from moisture and other environmental
7. Mislabeling of samples must be avoided
factors i.e. heat, light and air • Mark samples in such a way that the markings will not be
removed or damaged during storage or transport
3. Light sensitive samples should be stored in
• Containers made of opaque glass
8. If the sample is an official or legal sample the container
• Containers wrapped in aluminium foil
must be sealed to protect against tampering and the
seal mark easily identified
4. Oxygen sensitive samples should be stored under
nitrogen or an inert gas 9. Official samples also must include the date of sampling
with the name and sign of the sampling agent
5. Refrigeration or freezing may be necessary to protect • The chain of custody of such samples must be identified
chemically unstable samples clearly
• Freezing should be avoided when storing unstable emulsions
FST-511 winter 2019-20 33 FST-511 winter 2019-20 34

Preparation of samples Preparation of samples…

General size reduction consideration
► If the particle size or mass of the sample is too large
for analysis, it must be reduced in bulk or particle size General size reduction consideration…
1. To obtain smaller quantity for analysis the sample can
be spread on a clean surface and divided into quarters ►Once we have selected a sample that represents
• The two opposite quarters are combined
• If the mass is too large for analysis, the process is repeated
the properties of the whole population, prepare it
untill an appropriate amount is obtained for analysis in the laboratory
2. For homogeneous liquids ►The preparation of a sample for analysis must be
• Pour liquid into four containers done very carefully in order to make accurate and
• The samples are then homogenized to ensure negligible precise measurements
differences between each portion

FST-511 winter 2019-20 35 FST-511 winter 2019-20 36

9
 08-Nov-19

Preparation of samples…
Making Samples Homogeneous
► A number of mechanical devices have been developed
for homogenizing foods
► The type of homogenization method used depends on
the properties of the food being analyzed
• Solid
• Semi-solid
• Liquid
► Homogenization can be achieved using
a. Mechanical devices
(Grinders, mixers, slicers, blenders)
b. Enzymatic methods
(Proteases, cellulases, lipases)
c. Chemical methods
(Strong acids, strong bases, detergents)
FST-511 winter 2019-20 37

10