22 ) Sample and Sampling

insignificant, when the mumber in the sample is large. When large number
are considered the variations in the component parts tend to balance each
other and. therefore, the variation in the aggregate is insignificant. For
example, if delivery of 10 women is observed we would expect equal
number of boys and girls, i.e.,5 each. But since the sample is small in size
it is likely that we may not get exactly 5 boys and 5girls. The result may
be a combination of9 boys and I girl, or 8 boys and 2 girls, or 3 boys and
7 girls. If we take the example of larger sample i.e. 10,000 cases of delivery.
the chance of 5000 boys and 5000 girls would be very high i.e. the result
would be very near to 50% boys and 50% girls. The basic reason for such
likelihood is that the large size of sample has been taken and possibility of
variation in one direction compensating for others in a different direction
is greater.

Essentials of Sampling
For useful sample results, it is necessary that a sample possesses the following
essentials
1. Representativeness. Selected sample from the population should be
homogeneous so that it should not have much difference with the population.
To ensure representativeness the random method of selection should be used.
2. Adequacy. The size of sample should be adequate to make the result
more reliable.
3. Independence. The individual items composing the sample should be
selected independently of each other and all items of the universe should
have the same chance of being selected in the sample. By independence of
selection we mean that the selection of a particular item in one draw has
influence on the probabilities of selection in any other draw.
4. Homogeneity. The selected items in a sample should have the
similar characteristics of the universe from which it has been selected.
Two
samples taken from the same universe should give approOximately the same
result.

Methods of Sampling
There are various methods of sampling, All methods of sampling
can
grouped under the heads as (i) Random sampling and (ii) Non-randombe
sampling.
Sample and Sampling [23

1. Random Sampling
Random sampling is also referred to as probability sampling, since if the
sampling process is random, the laws of probability can be applied. In context
of random sampling, 'random is not used in the sense of haphazard. Random
sampling suggests that selection should be made without deliberate
diserimination.
A non-random sample is selected on a basis other than probability
considerations such as expert judgement, convenience or some other criteria.
For scientific research purposes, Random sampling is only applied.
Random sampling is of following two types
(1) Simple random sampling,
(2) Restricted random sampling.
IE Simple random sampling
Simple random sampling refers to the sampling technique in which each and
every item of the population is given an equal chance of being included in the
sample. The selection is thus free from personal bias because the investigator
does not exercise his discretion or preference in the choice of items. Since
selection of items in the sample depends entirely on chance. This method is
also known as the method of chance selection.
Samples chosen at random on line.

stratified random sampling

Samples chosen at random with strata on usuaily
different in size and based on a prior information
strata about variation and the site.

second stage unit

Primary units chosen at random. Samples chosen
at random within selected primary units.
primary (first stage) unit
clusture sampling
Allpossible samples taken within randomly
selected clusters.

two clustures
systemic sampling Sample evenly spaced after a random start
position A.

random sampling within segments

Segments may be of different size, not based on
a prior information about the size or variation
Fig. 4.1. Some random sampling plans for sampling over time or along a transect to
estimate a mean of total.
 24 ] Sample and Sampling

regular distribution random distribution contagious distribution

Fig. 4.2. Examples of Regular, Random and Contagious

distributions of individuals in an arca.

simple random stratified random systemic grid

sampling sampling sampling

two stage cluster random sampling

sampling sampling within block
Fig 4.3. Some two dimensional Probability Sampling designs for
sampling over space.
Random sampling is sometimes referred to as representative sampling.
If the sample is chosen at random and if the size of the sampie is
sufficiently large, it will represent all groups in the universe.
Random sampling is not always used as a primary sampling procedure.
However, it is necessary to introduce an element of randomness in the final
selection of items. For example, with each group the choice of cases to
constitute the sample should be based on chance selection. If the element of
randomness is not introduced, bias is likely to enter and make the sample
unrepresentative. For example, determining the number of cell X and its
dimension in a particular tissue, an experimenter might focus his microscope
held to a particular area of the slide and get the mean number of the cell X.
The result would be that the sample mean of X would tend
to exceed the
mean number of the actual mean of cell under study and the sample
be biased. would
Methods of obtaining a simple random sample. To
ensure
randomness of selection one may adopt any of the following
methods :
Sample and Sampling [25
1. Lottery method. This is the simplest and most popular method of
taking a random sample. Under this mnethod, all items of the universe/
population are numbered or named on separate slips of paper of identical
shape, size, and colour. These slips are then folded and mixed up in a
container or drum. A blindfold selection is then made of the number of slips
required to constitute the desired size of the sample. The selection of items
thus depends entirely on chance. The method would be quite clear with the
help of an example. If we want to take a sample of 100 patients out of a
population of 1000, the procedure is to write the names of all the 1000
patients on separate slips of paper, fold these slips, mix them thoroughly and
then make ablindfold selection of 100 slips.
While adopting lottery method it is absolutely essential to see that the
slips are of identical size, shape, colour, etc., otherwise there is a lot of
possibility of personal prejudice and bias affecting the results.
2. Table of random numbers. The lottery method discussed above
becomes dificult if the population size is very large. An alternative method
of random selection is that of using the table of random numbers. Three
such tables are available, namely, (i) Tippett's table of random numbers,
(ii) Fisher and Yate's numbers and (iii) Kendall and Babington Smith
numbers. Tippett's numbers are most popular. They consist of 41,600 digits
taken from census reports and combined by fours to give 10,400 four-figure
numbers. We give here the first forty sets as an illustration of the general
appearance of random numbers
3992 9792 7969 5911 3170 5624
2952 6641
1542 1396 7203 5356 1300 2693
4167 9524
3408 2762 3563 1089 6913 7691
2370 7483
1112 6107 6008 8126 4233 8776
0560 5246
1405 9025 7002 6111 8816 6446
2754 9143

numbers be
It is important that the starting point in the table of random
selected in some random fashion so that every unit has an equal chance of
being selected.
An example to illustrate how Tippett's table of random numbers may be
used is given below. Suppose we have to select 20 rats out of 6,000 kept in
6,000.
a big case. The procedure is to number all the 6,000 items from 1to
A page from Tippett's table may then be consulted and the first twenty
numbers up to 6,000 noted down. Rats bearing those numbers will be
Included in the sample. Making use of the portion of the table, given below,
the required numbers are :
26 ) Sample and Sampling
2952 3992 5911 3170 5624 4167
1545 1396 5356 1300 2693 2370

3408 2762 3563 1089 0560 5246

1112 4233

Rats bearing above number constitute the sample.

Universe size less then 1,000. If the size of universe is less than 1,000
the procedure will be different, as Tippett's numbers are available only in
four figures. Thus, for example, if it is desired to take a sample of 10 items
out of 400 all items from 1to 400 should be numbered as 0001 to 0400. We
may now select l0 numbers from the table which are up to 0400.
Universe size less than 100. If the size of universe is less than 100.
the table is used as follows : suppose ten numbers from out of 0 to 80 are
required. We start anywhere in the table and write down the numbers in
pairs. The table can be read horizontally, vertically, diagonally, or in any
other methodical way. Starting with the first and reading horizontally, we
obtain 29, 52, 66, 41, 39, 92, 97, 92, 79, 69, 59, 11, 31, 70, 56, 24, 41, 67
and so on. Ignoring the numbers greater than 80 we obtain for our purpose
ten random numbers, namely, 29, 52, 66, 41, 39, 79, 69, 59, 11 and 31.
Fisher and Yate's tables consist of 15,000 numbers. These have been
arranged in two digits in 300 blocks, each block consisting of 5 rows and 5
columns. However, this method cannot be followed in case of qualitative
data.
Merits of random sampling. (a) Less time. Sampling saves time
because experiment is performed with only a part of the population. Time is
saved not only in collecting data but also in processing it.
(b) Less cost. Sampling method reduces the cost of experiment because
only a few selected items are studied in sampling.
() Best for infinite population. In case of infinite population,
is the only method which can be used. sampling
(d) More reliable results. The result
obtained from sample is more
reliable than that obtained from a complete count.
There are several reasots
for it. First, it is always possible to determine
the extent of
Secondly, other types of errors to which a survey is subject sampling
are
errors.
more serious in a complete census than in a likely to be
sample survey.
(e) The destructive nature of certain tests. My
tests are of destrucuve
nature. Toxicity tests on organisms can be done only on a part of
population.
We cannot apply toxins to entire population.
Sample and Sampling

Limitations of random sampling, Despite the various sdvantages of

sampling. it is not altogether free from imitations, Some of the limitatioS
involved in sampling are stated below:
() Sometimes the population may be s0 small that it may be impossible to
draw arepresentative sample from it.
(2) The result computed from sanmple may be falsc, inaccurate and
misleading if the sample has not been drawn properly.
(3) There may be personal bias with regard to the choice of technique and
drawing of sampling units.
(4) At times the sampling plan may be so complicated that it requires more
time, labour and money than a complete count.
(5) If the information is required for each and every unit in the domain of
study, a complete enumeration survey is necessary.
Although there may be some shortcomings in sampling techniques, yet,
it is a very useful method for biostatistical investigation. According to
FF. Stephen "Samples are like medicines. They can harm only when they
are taken carelessly." Every good sample should have a proper label with
instructions about its use. Professor Chou States, "Sampling is a simple
process of learning about the population on the basis of a sample drawn
from it."

|II Restricted random sampling

Restricted random sampling are of three types :
1. Systematic sampling. This method is used in those cases where a
complete list of the population from which sampling is to be drawn is
available. This method is to select every Kn item from the list where 'K'
refers to the sampling interval. The starting point between the first and the
k item is selected at random. K is calculated by the formula :
Size of the universe N
K=
Sample size
For example, if a pond consists of 1,000 fishes of a species and if we
Want to draw a sample of 200 fishes this means we must take every fifth
item (i.e., k= 5).
1000
K= =5
200
The first item between one and five shall be selected at random.
Suppose it comes out to be three. Now we shall g0 on adding five and
obtain numbers of the desired sample. Thus, the second item would be the
Sth fish; the third, 13 th; the fourth, 18th fish; and so on.
 28 ) Sample and Sampling

Meris of ywtematic sampling. (1) The systematic sampling is more

convenient to adop than the random sampling.
(2) The time and labour involved by this sampling are relatively smaller.
(3) The results obtaincd are also found to be generally satisfactory.
(4) If populations are sufficiently large, homogeneous and each unit is
mumbered, then, this method can give accurate result.
Liìmitations of systematic sampling. (1) Systematic sampling becomes a
less representative design than simple randomn sampling if we are
dealing with populations having hidden periodicities.
2. Stratified sampling. This method is followed when the population is
not homogeneous. The population under study is first divided into
homogeneous groups or classes called stratas. Then the sample is drawn
from each stratum by simple random method in proportion to its size. The
resulting sample is called a stratified sample.
This method of sampling is for giving representation to all strata of
population such as selecting sample from defined areas, classes, ages, sexes
etc. This technique gives more representative sample than simple random
sampling in agiven large population. For example, if population is divided
into four stratas, their respective sizes being 15, 10, 20, 55% of the
population and a sample of 1,000 is to be drawn, the desired proportional
sample may be obtained in the following manner:
From stratum one 1,000 (0.15) = 150 items
From stratum two 1,000 (0.10) = 106 items
From stratum three 1,000 (0.20) = 200 items
From stratum four 1,000 (0.55) = 550 items
Sample size = 1,000
Proportional stratification yields a sample that represents the universe
with respect to the proportion in each stratum in the population. This
procedure is satisfactory if there is no great difference in variation from
stratum to stratum. But it is not efficient procedure, especially when there is
considerable variation in different stratas.
Merits of stratified sampling. (1) Proportionate representative sample
from each group is secured and it gives greater accuracy.
(2) Stratified samples can be more concentrated
geographically.
Limitations. Utmost care must be exercised in dividing the
into various stratas. Failing the point the results may not population
be reliable.
Sample and Sampling [29

3. Cluster sampling. Acluster is randomly selected group. This method

is used when units of population are natural groupS such as schools of
fishes, hospital ward, slums of a town etc. The technique of cluster sampling
allows small number of the target population to be sampled while the data
nrovided statistically valid at 95% confidence limits.
As per module approved by W.H.O. it is most often used to evaluate
vaccination.

focus point under microscope

magnified

cell X
cell Y

Fig. 4.4. Showing biased sampling for determining

the mean number of cell X in a particular tissue
structure. Focusing microscopic field to the
aggregated area of cell X, as a biased result of no.
of cell X is found in greatest number.

2. Non-Random Sampling
Non-random sampling methods are those which do not provide every item in
the universe with a known chance of being included in the sample. The
selection process is partially subjective. The choice of sample items depends
exclusively on the judgement of the investigator. For scientific research
purposes non-random sampling is not applied.
Choice of sampling methods
The different methods discussed above can be used in different situations,
Wherever they are appropriate. However, it is difficult to say that a particular
method would always be better than the rest. No one method can be
Fegarded as the best under all circumstances. A number of factors such as
the size of the sample, the size and nature of population, time, finance for
Sampling etc. would influence the selection of a particular method of
Sampling. The choice of sampling plan must be decided according to the
different factors.