UNIT 12 SAMPLING PROCEDURE*

Structure
12.0 Objectives
12.1 Introduction
12.2 Sampling Process
12.3 Types of Sampling
12.3.1 Probability Sampling
12.3.2 Non-Probability Sampling
12.3.3 Mixed Sampling
12.4 Selection of a Simple Random Sample
12.4.1 Lottery Method
12.4.2 Random Numbers Table Method
12.4.3 Steps in the Use of RNT
12.4.4 Advantages of SRS
12.4.5 Limitations of SRS
12.5 Selection of Systematic Random Sample
12.5.1 Advantages of Systematic Random Sampling
12.5.2 Disadvantages of Systematic Random Sampling
12.6 Selection of Stratified Random Sample
12.6.1 Proportional Stratified Sample
12.6.2 Disproportional Stratified Sampling
12.6.3 Advantages of Stratified Sampling
12.6.4 Disadvantages of Stratified Sampling
12.7 Selection of a Cluster Sample
12.7.1 Steps in Cluster Sampling
12.7.2 Advantages of Cluster Sampling
12.7.3 Disadvantages of Cluster Sampling

12.8 Multistage Sampling

12.9 Non-Probability Sampling Procedures
12.9.1 Convenience Sampling
12.9.2 Judgement Sampling
12.9.3 Quta Sampling
12.9.4 Snowball Sampling

12.10 Determination of the Sample Size

12.11 Let Us Sum Up
12.12 Answers/Hints to Check Your Progress Exercises

*
Prof. C G Naidu (retd.), School of Vocational Studies, IGNOU
12.0 OBJECTIVES Sampling
Procedure
On completion of this Unit, you should be able to
 explain different methods of drawing a sample;
 describe different types of samples;
 use random number tables to draw a sample; and
 determine the sample size.

12.1 INTRODUCTION
In Unit 2 of this course, you have learned about different types of data, namely,
primary data and secondary data. In that Unit, we also discussed the use of
different survey techniques like face-to-face interview, telephone survey, postal
survey, internet survey, etc. in collecting primary data. In Unit 15, you also have
learned the meaning of sampling, advantages of sampling, and sampling error.
In statistics, we often rely on a sample (that is, a small subset of a larger set of
data) to draw inferences about the population (that is, the larger set of data). For
example, you are interested to know the voting behaviour of Delhi people in the
next election. Who will you ask? Naturally, it is not possible for you to ask every
single Delhi voter how he or she is likely to vote. Instead, you may querry a
relatively small number of Delhi voters and draw inferences about entire Delhi
from their responses. In this case total voters of Delhi constitute the population
and the voters actually querried constitute your sample.
Ideally, the characteristics of a sample should reflect the characteristics of the
population from which it is drawn. In such cases, the inferences drawn from a
sample are probably applicable to the entire population.
In this unit you will learn how to draw a sample under different population
characteristics and how to determine the sample size.

12.2 SAMPLING PROCESS

In conducting a sample survey, the sampling process determines which sampling
units will be included in the survey. Sampling makes data process more
manageable and affordable. It enables the population characteristics to be
inferred with minimal errors on the basis of the sample. The sampling process
includes defining the target population from which we draw the sample,
identification of the sampling frame, selection of the sampling method, and
selection of the sampling units.
1) Survey Objectives: A sample survey begins with the specification of the
objectives. We should have a clear and un-ambiguous idea of the objectives

265
Sampling and of the survey, because all other steps – target population, sampling frame,
Statistical Inference
sampling procedure, etc. – are designed according to survey objectives.
2) Questionnaire Design: Keeping the objectives of the survey in view we are
required to design a questionnaire. We have already learnt the major steps
involved in designing of a questionnaire in Unit 1 of this course. In addition
to the questionnaire we need to develop training documents for the
investigators, particularly when the sample survey is conducted at a larger
scale involving a number of investigators.

3) Defining the Target Population: To draw the samples from a population we

must know the target population about which conclusions are to be drawn.
The target population is also referred to as the universe. The target
population is the group about which we wish to generalize or make
inferences from the sample. For example, you want to conduct a sample
survey on the family planning methods used by eligible couples in Delhi.
Here, all those couples in Delhi in the reproductive age group form the
target population.

4) Identifying Sampling Frame: The sampling frame is a list of cases from the
target population. The sampling frame is the actual operational definition of
the target population. In our earlier example of eligible couples in Delhi
using family planning methods, all the people in the reproductive age group
form the sampling frame. Many times we may not be able to list all the
cases in the target population for some reason or other. For example, we
want to list the people for a survey based on telephone directory. In this
case, certainly those people who do not have telephone numbers in the
directory will be excluded from the listing. This type of error is called
sampling frame error.
5) Selecting Sampling Procedure: Once the sampling frame is identified, we
select appropriate sampling procedure to select the sample for the survey.
We will discuss various sampling procedures in detail in the next section of
this Unit.
6) Selecting the Sampling Units: Sampling units are those cases from the
sampling frame which are included in the sample by using appropriate
sampling procedure. Essentially, a sampling unit is the case on which data
is collected. For example, you may decide to take 1000 sampling units from
the sampling frame (consisting of all the reproductive age group people in
Delhi) for your sample survey.

7) Survey data Processing: After selection of sampling units the next step is
data collection and processing. We need to check the incomplete
questionnaires and edit or cross-check the responses wherever there is a
doubt. Data entry and tabulation follows.

266
8) Analysis of Data: The next step in the sequence is analysis of data. Keeping Sampling
Procedure
in view our requirements we analyse the data by using various statistical
tools.

9) Publication and Dissemination of Results: On the basis of data analysis we

prepare technical and research reports. Finally, the socio-economic results
of the survey and their implications are discussed in seminars.

12.3 TYPES OF SAMPLING

The method of selecting a sample from a given population is called sampling.
Basically there are two types of sampling, viz., probability sampling and non-
probability sampling. In probability sampling the sampling units are selected
according to some chance mechanism or probability of selection. On the other
hand, non-probability sampling is based on judgement or discretion of the person
making a choice. Thus in non-probability sampling certain units may be selected
because of convenience or they serve a purpose or the researcher feels that these
units are representative of the population. No random selection on the basis of
chance mechanism is involved here.
12.3.1 Probability Sampling
It is also called random sampling. It is a procedure in which every member of the
population has a chance or probability of being selected in the sample. It is in this
probabilistic sense that the sample is random. The word ‘random’ does not mean
that the sample is obtained in a haphazard manner without following any rule.
Random sampling is based on the well-established principles of probability
theory. There are quite a few variants of the random sampling, viz., simple
random sampling, systematic random sampling and stratified random sampling.
We discuss these types below.
a) Simple Random Sampling
If there is not much variation in the characteristics of the members of a
population, we can follow the method of simple random sampling. In this
method, we consider the population in its entirety as a homogeneous group and
follow the principle of random sampling to choose the members for the sample.
There are two variants of simple random sampling, viz., simple random sampling
with replacement (SRSWR) and simple random sampling without replacement
(SRSWOR). This difference pertains to the way the sample units are selected.
According to the procedure of simple random sampling with replacement
(SRSWR), we draw one unit from the population, note down its features and put
it back to the whole lot in the sense that the unit again becomes eligible for
selection. In this way, the total number of units in the population always remains
the same. In other words, the composition of the population remains unchanged,
and each member of the population has the same chance or probability of being

267
Sampling and selected in the sample. In fact, if N is the size of the population, this probability is
Statistical Inference
1
. On the other hand, in the case of simple random sampling without
N
replacement, the unit once selected is not returned to the population in the sense
that it becomes ineligible for selection again. As a result, after each draw, the
composition of the population changes. Therefore, the probability of any
particular unit being selected also changes.
b) Systematic Random Sampling
In this variant of random sampling, only the first unit of the sample is selected at
random from the population. The subsequent units are then selected by following
some definite rule. For example, suppose, we have to choose a sample of
agricultural plots. In systematic random sampling, we begin with selecting one
plot at random and then every j th plot may be selected.
c) Stratified Random Sampling
Stratified random sampling is the appropriate method if the population under
consideration consists of heterogeneous units. Here, first we divide the
population into certain homogeneous groups or strata from each stratum.
Secondly, some units are selected by simple random sampling. Thirdly, after
selecting the units from each stratum, they are mixed together to obtain the final
sample.
Let us consider an example. Suppose, we want to estimate the per capita income
of Delhi by a sample survey. It is common knowledge that Delhi is characterised
by rich localities, middle class localities and poor localities in terms of the
income groups of the people living in these localities. Now, each of these
different localities can constitute a stratum from which some people may be
selected by adopting simple random sampling procedure.
d) Multi-Stage Random Sampling
Let us consider a situation where we want to obtain information from a sample of
households in a large city, say, Delhi. Sometimes, it may not be possible to
directly take a sample of households because a list of all the households may not
be easily obtained. In such a situation, one may resort to take samples in various
stages. Generally, the city is divided into certain geographical areas for
administrative purposes. These areas may be termed as city blocks. So in the first
stage, some of such blocks may be selected by random sampling. In the next
stage, from each of the selected blocks in the first stage, some households may be
selected again by the principle of random sampling. In this way, ultimately a
sample of households from a large city may be obtained. The above-mentioned
example is the case of a two-stage random sampling. However, if the nature of
the inquiry so demands, the method of sampling can be extended to more than
two stages.
268
12.3.2 Non-Probability Sampling Sampling
Procedure
We have considered the method of random sampling and some of its variants
above. It should be clear that the basic objective of the principle of random
sampling is to eliminate or at least minimise the effect of the subjective bias of
the investigator in the selection of the population sample. But for certain
purposes, there is a need for using discretion. For example, suppose a teacher has
to choose 4 participants from a class of 30 students in a debate competition. Here,
the teacher may select the top 4 debaters on the basis of her own conscious
judgement about the top debaters in the class. This is an example of purposive
sampling. In this method, the purpose of the sample guides the choice of certain
members or units of the population.
12.3.3 Mixed Sampling
In mixed sampling, we have some features of both non-probability sampling and
random sampling. Suppose, an institute has to send 5 students for managerial
training in a company during the summer vacation. Initially, it may shortlist
about 20 students who are considered to be suitable for the training by applying
its own discretion. Then from these 20 students, 5 students may finally be
selected by random sampling.
We will discuss the process of drawing various types of samples later in this
Unit.

12.4 SELECTION OF A SIMPLE RANDOM SAMPLE

Simple random sampling (SRS) is the basic sampling procedure where we select
a sample from the population. In this procedure each unit is chosen entirely
through chance mechanism and each unit of the population has an equal chance
of being included in the sample. Every possible sample of a given size has the
same chance of selection. That is, each unit of the population is equally likely to
be chosen at any stage in the sampling process and selection of one unit should in
no way influence the selection of another unit in the population.
Simple random sampling should be used with a homogeneous population. That
is, all the units in a population should possess the same attributes that we are
interested in measuring. The characteristics of homogeneity may include age,
sex, income, social status, geographical region, etc.
There are two most commonly used methods to extract a simple random sample.
The first is lottery method and the second is a random numbers selection method.
Irrespective of which method we decide to use, every element in the sampling
frame should be assigned an identifying number.
12.4.1 Lottery Method
The simplest method of selecting a random sample is drawing lottery. In this
method, the unit identification numbers are placed in a container and mixed
together. Finally, someone draws out numbers at random from the container until
269
Sampling and the desired sample size is attained. Suppose, we want to select n sample units out
Statistical Inference
of N population units. We assign the numbers 1 to N; one number to each of the
population unit; and write these numbers on N slips. The slips are made as
homogeneous as possible in shape, size, colour, etc. These slips are then put in a
container and thoroughly shuffled. Finally, n slips are drawn one by one. The n
units corresponding to numbers on slips drawn, will constitute a random sample.
Example 12.1
Let us assume that you are doing some research with a bank branch that wishes to
assess customers views on the quality of service in the bank branch. You are
asked to select 100 customers as the sample using simple random sampling
procedure with lottery method. First, you have to get the sampling frame
organized. For this, you will go through the bank records to identify the account
holders. You then assign the serial numbers to all the account holders. Suppose
there are 1000 account holders and you want to draw 100/1000=10% sample.
You could print the serial numbers, tear them into separate strips, put the strips in
a container, mix them up real good, close your eyes and pull out the first 100
strips.
The disadvantage of the lottery method is that it would be tedious and the quality
of the sample would depend on how thoroughly you mixed them up and how
randomly you picked them. Also, as the population size increases, it becomes
more and more difficult to draw samples using lottery method.

12.4.2 Random Numbers Table Method

The random numbers are a collection of digits generated through a probability
mechanism. The random numbers have the following properties:

The probability that each digit (0,1,2,3,4,5,6,7,8 or 9) will appear at any place is
the same, that is 1/10.

The occurrence of any two digits in any two places is independent of each other.

In this method each unit in the population is assigned a unique number in a

sequence. To draw the sample we use a table of random numbers. You can find
the random number table (RNT), among other places, in Fisher and Yates (1963):
Statistical Tables for Biological, Agricultural and Medical Research. An example
of a random numbers table is shown in Table 12.1.

Table 12.1: Random Numbers Table

39634 62349 74088 65564 16379 19713 39153 69459 17986 24537

14595 35050 40469 27478 44526 67331 93365 54526 22356 93208

30734 71571 83722 79712 25775 65178 07763 82928 31131 30196
270
64628 89126 91254 24090 25752 03091 39411 73146 06089 15630 Sampling
Procedure
42831 95113 43511 42082 15140 34733 68076 18292 69486 80468

80583 70361 41047 26792 78466 03395 17635 09697 82447 31405

00209 90404 99457 72570 42194 49043 24330 14939 09865 45906

05409 20830 01911 60767 55248 79253 12317 84120 77772 50103

95836 22530 91785 80210 34361 52228 33869 94332 83868 61672

65358 70469 87149 89509 72176 18103 55169 79954 72002 20582

72249 04037 36192 40221 14918 53437 60571 40995 55006 10694

41692 40581 93050 48734 34652 41577 04631 49184 39295 81776

61885 50796 96822 82002 07973 52925 75467 86013 98072 91942

48917 48129 48624 48248 91465 54898 61220 18721 67387 66575

88378 84299 12193 03785 49314 39761 99132 28775 45276 91816

77800 25734 09801 92087 02955 12872 89848 48579 06028 13827

24028 03405 01178 06316 81916 40170 53665 87202 88638 47121

86558 84750 43994 01760 96205 27937 45416 71964 52261 30781

78545 49201 05329 14182 10971 90472 44682 39304 19819 55799

14969 64623 82780 35686 30941 14622 04126 25498 95452 63937

58697 31973 06303 94202 62287 56164 79157 98375 24558 99241

38449 46438 91579 01907 72146 05764 22400 94490 49833 09258

62134 87244 73348 80114 78490 64735 31010 66975 28652 36166

72749 13347 65030 26128 49067 27904 49953 74674 94617 13317

81638 36566 42709 33717 59943 12027 46547 61303 46699 76243

46574 79670 10342 89543 75030 23428 29541 32501 89422 87474

11873 57196 32209 67663 07990 12288 59245 83638 23642 61715

13862 72778 09949 23096 01791 19472 14634 31690 36602 62943

08312 27886 82321 28666 72998 22514 51054 22940 31842 54245

11071 44430 94664 91294 35163 05494 32882 23904 41340 61185

271
Sampling and
82509 11842 86963 50307 07510 32545 90717 46856 86079 13769
Statistical Inference
07426 67341 80314 58910 93948 85738 69444 09370 58194 28207

57696 25592 91221 95386 15857 84645 89659 80535 93233 82798

08074 89810 48521 90740 02687 83117 74920 25954 99629 78978

20128 53721 01518 40699 20849 04710 38989 91322 56057 58573

00190 27157 83208 79446 92987 61357 38752 55424 94518 45205

23798 55425 32454 34611 39605 39981 74691 40836 30812 38563

85306 57995 68222 39055 43890 36956 84861 63624 04961 55439

99719 36036 74274 53901 34643 06157 89500 57514 93977 42403

95970 81452 48873 00784 58347 40269 11880 43395 28249 38743

56651 91460 92462 98566 72062 18556 55052 47614 80044 60015

71499 80220 35750 67337 47556 55272 55249 79100 34014 17037

66660 78443 47545 70736 65419 77489 70831 73237 14970 23129

35483 84563 79956 88618 54619 24853 59783 47537 88822 47227

09262 25041 57862 19203 86103 02800 23198 70639 43757 52064

Source: Adapted from Table of Random Numbers at

http://www.mrs.umn.edu/~sungurea/introstat/public/instruction/ranbox/r
andomnumbersII.html
The above random numbers table contains 450 (5 digit) random numbers.

12.4.3 Steps in the Use of RNT

We need to follow the following steps while selecting a SRS by using RNT.

1. Determine the population size (N).

2. Determine the sample size (n).
3. List all the units of the population. Assign the numbers in a serial order.
Suppose there are 100 units in the population, assign the serial numbers from
00 to 99.
4. Determine the starting point of selecting the sample by picking up a page
from the random number tables and dropping your finger on a number in the
page blindly.

272
 5. Choose the direction in which you want to read the numbers (say Sampling
Procedure
from left to right or right to left or top to bottom or bottom to top).
6. Suppose you are looking for two digit numbers (00 to 99) you may
not get these numbers by direct reading from the tables since they are
5 digit numbers (see Table 12.1). You can either look at the last two
digits or first two digits of the numbers. For example, if the 5 digit
number you have chosen is 54245 (that is, the number in the 29th row
and 10th column of the random number table given at Table 12.1).
Then, the two digit number will be 45 if you chose the last two digits
of the number.
7. Look only at the numbers assigned to each population unit. If the
number represents one of the unit of the population it becomes part of
the sample. Suppose you want to select 10 sample units, the other
numbers you will be choosing are 71(11071), 30(44430),
64(94664),94(91294),63(35163),82(32883),04(23904), 40(41340),
85(61185). Observe that you have omitted 94(05494) since you have
already chosen this number.
8. Once a number is chosen, do not select it again.
9. If you reach the end point of the table before obtaining the required
sample, pick another starting point in the random number table and
select the remaining units for the sample.
Example 12.2

Suppose you have to select 100 account holders as a sample out of total 1000
account holders in the population using random numbers table.
628 126 254 090 752 091 411 146 089 630
831 113 511 082 140 733 076 292 486 468
583 361 047 792 466 395 635 697 447 405
209 404 457 570 194 043 330 939 865 906
409 830 911 767 248 253 317 120 772 103
836 530 785 210 228 869 332 868 672 358
469 149 509 176 169 954 002 582 249 037
192 221 918 437 571 995 006 694 692 581
050 734 652 577 631 184 295 776 885 796
822 973 822 467 013 072 942 917 129 624

Here, first you assign each account holder a number from 000 to 999. To draw a
sample of 100 account holders, you need to find 100 three digit numbers in the
range 000 to 999. Pick up any row or column in the random numbers table given
in Table 12.1. Suppose you have selected the fourth row and first column as
starting point to draw the sample, the first digit number is 628(64628) if you
chose last 3 digits as the number for your purpose. Here, you read the last 3 digits
273
Sampling and of the number. If the number is within the range (000 to 999) include the number
Statistical Inference
in the sample. Otherwise skip the number and read the next number in some
identified direction. If a number is already selected omit it. In this example since
you have started with fourth row and first column and moving from left to right
direction the following 100 numbers are selected for the sample.
If the number of units in the population is very large, neither of the above two
methods is feasible. These days by using a computer we can select a random
sample in a much easier way. There are many computer program which can
generate a series of random numbers if we have the units of the population listed
in a computer.

We will explain one way of selecting a sample using computer generated random
numbers. In our example, let us assume we can copy and paste the list of account
holders into a column in an EXCEL spreadsheet. Then, in the column right to it
we paste the function =RAND() which is EXCEL's way of putting a random
number between 0 and 1 in the cells. Then, all we have to do is take the first 100
names in the sorted list. The entire process takes a minute if we are familiar with
using EXCEL program in computer.

12.4.4 Advantages of SRS

1) In simple random sample we assure population units to be homogeneous
and thus do not require additional information on the characteristics of the
population.
2) Using simple random sampling we can select an unbiased sample. This is
because, in SRS the chance of including each unit of population in the
sample is equal. The bias due to human preferences is completely
eliminated.
3) Through estimation of sampling error we can assess the accuracy of the
results.
4) If the population size is not too large, simple random sampling is a simple
and easily implementable sampling procedure.

12.4.5 Limitations of SRS

1) The greatest limitation of simple random sampling is that if the
population size is too large then we need to spend a lot of time in listing
the units of the population.
2) The simple random sampling procedure will be efficient only when we
have a homogeneous population. Suppose we have a population with
characteristics such as gender, age, social status, etc. Then, we need a
larger sample size to accommodate a representative sample with all
those characteristics of the population units. A better way to tackle this
issue is to use stratified sampling procedure which you will learn later in
this unit.
274
12.5 SELECTION OF SYSTEMATIC RANDOM Sampling
Procedure
SAMPLE
The systematic random sampling procedure is somewhat similar to the simple
random sampling procedure. In this sampling procedure, we select a starting
point at random and then systematically select the sample units from the
population units at a specified sampling interval.
The starting point and the sampling interval are based on the required sample
size. The sampling interval will be represented as K. The selection of a sample
using systematic random sampling procedure is very simple. Suppose the
population consists of n units and you have decided to select a sample of n units
using systematic random sampling procedure. Follow the following steps.

1. Number the units in the population from 1 to N (suppose you have

1000 units).
2. Decide the sample size n you need (suppose you want to select 100
units).
3. Determine the sampling interval by dividing the population by the
sample size. K = N/n = the interval size (here K=1000/100 = 10).
4. Select a unit at random from the first K units (1 to K) (suppose you
have selected unit number 5 as your first sample unit).
5. Then select the subsequent sample units by adding K to the previous
unit (the subsequent samples will be 15 (5+10), 25 (15+10), …, 995
(985+10).
Example 12.3

From a population consisting of 500 units draw a sample of 60 units using

systematic random sampling procedure.

To use systematic random sampling, the first thing we need to do is listing of the
population units in a random order by giving numbers from 1 to 500. The
sampling interval is K=500/60 = 8.3 or say 8. Now we select the first sample unit
at random from the first 8 population units. Suppose the first unit selected is 5.
The subsequent sample units selected are: 13, 21, 29, 37.........477. Therefore,
following are the population units selected in the sample.

5 13 21 29 37 45 53 61 69 77
85 93 101 109 117 125 133 141 149 157
165 173 181 189 197 205 213 221 229 237
245 253 261 269 277 285 293 301 309 317
325 333 341 349 357 365 373 381 389 397
405 413 429 429 437 445 453 461 469 477
275
Sampling and
Statistical Inference Thus, in the systematic random sampling procedure, the first sample unit is
selected at random and this sample unit in turn determines the subsequent sample
units to be selected. However, it is essential that the units in the population are
randomly ordered. In certain cases we prefer using systematic random sampling
procedure to simple random sampling procedure because it is easier to select
sample units. For example, if you want to find out the yield of coconut trees in a
field, select a tree at random, other trees are automatically selected at a gap
equivalent to sampling interval.

12.5.1 Advantages of Systematic Random Sampling

a) The main advantage of using systematic random sample is that the time
taken and work involved in systematic random sampling is less than
simple random sampling procedure. It is frequently used in exit polls on
voting behaviour and obtaining the opinions and views of consumers in
marketing research.
b) The other advantage of systematic random sampling procedure is that this
method can be used even when no formal list of the population units is
available. For example, suppose if we are interested in knowing the
opinion of consumers on improving the services offered by a bank, we
may simply choose every kth account holder visiting a bank branch,
provided that we know how many account holders are there : (1) For
example, there are 2000 account holders in the population and we want to
have 200 account holders as sample size. Then, K=2000/200=10) and we
select every 10th account holder visiting the bank.

12.5.2 Disadvantages of Systematic Random Sampling

a) The main disadvantage of systematic random sampling procedure is that
if there is a periodicity in the occurrence of units of a population, the use
of systematic random sampling procedure gives a highly
unrepresentative sample. For example, suppose you are interested in
obtaining the views/opinions of consumers of a store in your locality.
You may arrange all the consumers of the store according to their date of
visit and start selecting a sample of customers who visit the store on 1st
of every month. You know that the 1st day of every month cannot be
representative of the whole month.
b) The other disadvantage of systematic random sampling procedure is that
every unit of the population does not have an equal chance of being
selected. Rather the selection of population units in the sample depends
on the initial unit of selection. Regardless of how we select the first unit
of the sample, subsequent units are automatically determined. This lacks
complete randomness.

276
12.6 SELECTION OF STRATIFIED RANDOM Sampling
Procedure
SAMPLE
In some cases the population may not be homogenous, that is, all the units may
not be equal with respect to the characteristic we intent to survey. The
characteristics of the population under study may be male/female, rural/urban,
literate/illiterate, high income/low income groups, etc. In situations where these
units vary widely, the simple random sampling procedure or the systematic
random sampling procedure will not provide us with a representative sample. In
such situations by using stratified random sampling we can obtain a
representative sample.
In stratified sampling, we divide the population into different strata in such a way
that units are homogenous within each stratum. Moreover, each stratum is
different. Suppose we want to stratify the population on the basis of gender
distribution then we list the population units separately according to males or
females. Subsequently, we decide the sample size to be drawn from each stratum.
There are two approaches to decide the sample size from each stratum. These are:
(a) proportional stratified sample, and (b) disproportional stratified sample. We
will discuss these two procedures below.
12.6.1 Proportional Stratified Sample
When we take a sample from a population with several strata, we are required to
take samples from each stratum. Such sample could be in proportion of the
stratum population size to the total population size. Suppose we divide the
population (N) into K non-overlapping strata N1, N2, N3..... NK such that
N1+N2+N3+...........+NK = N. We decide to draw a sample of the size n. Then the
sample proportions of different strata are given by:
n1 n2 n3 n
   ...  12
N1 N 2 N3 NK
Example 12.4
Suppose we want to draw a sample of 200 units from a population consisting of
1000 units. The population is heterogeneous in nature in terms of high income or
low income and rural or urban. The strata population sizes are given as follows:

High income - urban = 200

Low income - urban = 400
High income- rural = 100
Low income-rural = 300
To have a representative sample each stratum in the sample should represent the
corresponding stratum in the population. For this we should take a different
sample size from each stratum depending upon the stratum size. The deciding
factor in each of the stratum is same as the proportion of total sample to the

277
Sampling and population. In our example, to have a sample of 200 units, the proportion of
Statistical Inference

sample to the population in each stratum is

n 200
  0.2
N 1000
Observe that we are considering the same proportion for each stratum. Then the
sample from each stratum will be as follows:

Stratum Stratum Sample to Stratum sample

population
Category Population size
size( Ni ) proportion

(1) (2) (3) (4)=(2)  (3)

High income - urban 200 0.2 40
Low income -urban 400 0.2 80
High income-rural 100 0.2 20
Low income-rural 300 0.2 60
Total 1000 0.2 200
There are several advantages of stratified sampling over simple random
sampling. The stratified sampling ensures sample representation of not only the
entire population, but also each stratum. This is important where the stratum size
is small. Moreover, stratified sampling generally has more statistical precision
than simple random sampling.

12.6.2 Disproportional Stratified Sampling

In proportional stratified sampling, we assumed that each stratum in the

population is homogeneous. Consequently, we expect that variability within
stratum is lower than the variability for the population as a whole. On the other
hand, if the variability within each stratum is not small then we use
disproportional stratified sampling. In disproportional stratified sampling, the
strata allocation is based on size and variability (that is, the standard deviation of
the characteristic under study). In this procedure a larger sample is drawn from
the stratum having higher variability. This procedure is also sometimes called
double weighing scheme and provides the most efficient sample and most
precise/reliable estimates for a given sample size. The only requirement is that
we should have knowledge/estimate of the standard deviation of the
characteristic under study within each stratum.

Follow the steps given below for using disproportional stratified sampling.

278
1) Divide the population into strata based on the chosen characteristic Sampling
Procedure
(example, Rural/Urban, Male/Female, etc.)
2) The number of units taken from each stratum is directly proportional to the
relative size of the stratum and standard deviation i of the characteristic
under consideration. Suppose, if 1, 2, 3, …k are the standard
deviations of k strata and P1, P2, P3, …Pk are the stratum proportions to
the total population, and n ( n1  n2  ......  nk ) is the sample size
required. Then the stratum sample size using disproportional stratified
sampling procedure is
Pi   i  n
ni 
 Pi i
3) Choose the sample from each stratum using either simple random
sampling or systematic random sampling.
Let us go back to Example 12.4, where we have divided the population into 4
strata. We observe that there are small number of households in high income
strata and large number of households in low income strata. Assume that
the variance of income among higher income groups is higher than the
variance among the lower income groups. Therefore, in order to avoid under-
representation of higher income groups in the sample, a disproportional
sample is taken in each stratum. That means, if the variability within the stratum
is higher, we must have larger sample size of that stratum to increase the
precision of the estimates. Similarly, if the variability within the stratum is lower,
we must have smaller sample size of that stratum. That is, higher the stratum
variance larger the stratum sample size and lower the stratum variance smaller
the sample size. This is in addition to the fact that larger stratum size requires a
larger sample size.

Example 12.5
Consider Example 12.4 again. Suppose the stratum variances are given as
follows:

Stratum Variance (2)

High income urban 6.5
Low income urban 2.5
High income rural 4.5
Low income rural 2.0
Use the disproportional stratified sampling procedure to choose a sample of size
200 from the four strata.
For this example the disproportional sample size for each stratum is given below:
279
Sampling and Stratum Stratum Stratum Stratum Stratum Pi  Sample
Statistical Inference population population Variance standard i size
proportion (i2) deviation Pi   i  n
(Pi) (i =i2)  Pi i
High 200 0.20 6.5 2.5 0.50 56
income-
urban
Low 400 0.40 2.5 1.6 0.64 72
income-
urban
High 100 0.10 4.5 2.1 0.21 24
income-
rural
Low 300 0.30 2.0 1.4 0.42 47
income-
rural
Total 1000 1.77 200

12.6.3 Advantages of Stratified Sampling

a) In stratified random sampling the sample is drawn from each stratum of the
population. Therefore, the stratified random sampling procedure is more
representative.
b) The stratified random sampling procedure is more precise than
simple random sampling. Therefore, to a great extent this procedure
avoids sample selection bias.

c) As we have seen in simple random sampling and systematic random

sampling procedures, when there is heterogeneity of population we
need to have a large sample size to have a fairly representative
sample. However, in stratified random sampling this objective can be
achieved with a smaller sample size. This saves a lot of time, money
and other resources for data collection.
12.6.4 Disadvantages of Stratified Sampling
a) The main disadvantage of stratified random sampling procedure is that we need
a detailed knowledge of the distribution of the characteristics in the population.
If we cannot accurately identify the homogeneous groups, it is better to use
simple random sample since improper stratification can lead to serious error.
b) The other disadvantage of stratified random sampling is that we need to prepare
a list of population units for each stratum separately. As the list of population
units may not be readily available for each characteristic, the preparation of
lists may be a very difficult task.

12.7 SELECTION OF A CLUSTER SAMPLE

Very often population units are spread over a vast geographical area. In that case
collection of data through simple random sampling requires a lot of time, money
280
and manpower as we have to cover the entire geographical area for collecting data Sampling
Procedure
on the selected units. Imagine taking a sample of respondents spread all over Uttar
Pradesh in order to conduct personal interviews. Using simple random sample, the
respondents will be spread all over the state and you have to travel and spend lot
of money meeting the respondents. In such situation cluster sampling will be
much useful.
The basic principles of cluster sampling are:
i) The differences or variability within a cluster should be as large as possible.
As far as possible the variability within each cluster should be the same as that
of population.
ii) The variability between clusters should be as small as possible. Once the
clusters are selected, all the units in the selected clusters are included in the
sample for obtaining data.
In cluster sampling we divide the population into groups called clusters. Then we
select a sample of clusters using a simple random sampling.
The population units in each of the clusters are assumed to be as heterogeneous as
those in the total population. That is, each cluster itself is a representative of the
population.
12.7.1 Steps in Cluster Sampling
In cluster sampling, we follow the steps given below:
1) Divide the population into a number of clusters.
2) Determine the number of clusters needed for your sample.
3) Randomly select the sample of clusters.
4) Survey all units within the sampled clusters.

Suppose the division of clusters is based on the geographical boundaries of the

population, then it is called area sampling. You have observed that in the case of
cluster sampling the clusters are selected using random sampling method.
Subsequently all the population units within each sampled cluster are included in
the sample. Suppose instead of including all the population units within each
selected cluster you chose to include only a sample of units within each cluster.
Then you can clearly understand that there are two stages.

In the first stage you select the clusters and in the second stage you select the
sample units within each sampled clusters. This sampling procedure is called two-
stage sampling. Here, the clusters are called primary units and the units within the
sampled clusters are called secondary units.
Example 12. 6
Suppose we are interested in finding the options of ATM customers of a Bank in
Uttar Pradesh state. We can divide the state into say 30 clusters (may be we can
281
Sampling and consider district as a unit and include one or two districts in one cluster). Here, we
Statistical Inference
assume that each of these clusters will represent the opinions of the ATM
customers of Uttar Pradesh as a whole. We then select a sample of clusters and
obtain the opinion of all the ATM customers in each of the cluster.

12.7.2 Advantages of Cluster Sampling

a) The main advantage of cluster sampling is that it takes less travel time and
related data collection costs.
b) Since the researcher need not cover all the clusters and only sample of clusters
are covered, it is a more practical method which facilitates fieldwork.

12.7.3 Disadvantages of Cluster Sampling

a) In cluster sampling we assume that each cluster represents the heterogeneity of

the population units of all clusters. However, this assumption may not be true
in many cases, because often the tendency is that the units in the clusters are
more homogeneous than the units of the entire population. That means it is
difficult to form heterogeneous clusters.

b) The cluster sampling has a lower sampling efficiency for a given sample
size than random sampling and stratified sampling. This method is cost
ffective not statistically efficient.

12.8 MULTISTAGE SAMPLING

We have seen in cluster sampling that when we select a sample instead of
covering all the units from each cluster, we call it two-stage sampling. The
multistage sampling is as extension of two-stage sampling.

The four methods we have covered so far, namely, (a) simple random sampling,
(b) systematic random sampling, (c) stratified sampling, and (d) cluster sampling
are the simplest probability (or random) sampling procedures. However, in real-
life, we use sampling methods that are more complex than the above four
methods. The basic principle in multistage sampling is that we can combine these
simple methods in a variety of useful ways to address our sampling needs. We call
it multistage sampling when we combine two or more of the above sampling
methods.
Example 12.7
Consider the case of interviewing school students in Haryana in order to grade the
schools according to socio-economic background of the parents. For this problem,
in the first stage we need to apply cluster sampling. We divide the state of
Haryana into a number of clusters, say districts. Then we select a sample of
districts (clusters) using simple random sampling method. In the second stage we

282
divide the schools using stratified sampling. Here the strata may be government Sampling
Procedure
schools, government-aided schools, central schools, and public schools. We select
a sample of schools in each stratum using either a simple random sampling or a
systematic random sampling. In the third stage we again use simple random
sampling and select a sample of classes in each sampled school for face-to-face
interviews with the students. In the fourth stage of sampling we consider selecting
a sample of students from each sampled class using simple random sampling or
systematic random sampling.

In multi stage sampling it is possible to consider as many stages as necessary to

achieve a representative sample. In each stage a suitable method of sampling is
used. Each stage results in a reduction of the sample size.
Advantages
a) Multistage sampling procedure provides cost gains by reducing the data
collection costs.
b) Multistage sampling is more flexible and allows us to use different sampling
procedures in different stages of sampling.
c) If the population is spread over a very wide geographical area, multistage
sampling is the most appropriate sampling method.
Disadvantages
If the sampling units selected at different stages are not representative, multistage
sampling becomes less precise and less efficient.
Check Your Progress 1
1) Which of the following is a procedure of selecting samples from a population?
a) Random sampling
b) Non-random sampling
c) Stratified sampling
d) All the above
2) Suppose you are applying a stratified random sampling procedure on a
population. How do you make your sample selection?
a) Select at random an equal number of units from each stratum
b) Draw equal number of units from each stratum and weigh the results
c) Select the sample at random from each stratum proportional to the
population
d) b. and c. both
e) a. and c. both
283
Sampling and 3) State whether the following statements are ‘True’ or ‘False’.
Statistical Inference
a) A sampling procedure that selects units from a population at uniform
intervals is called simple random sampling.
b) A sampling procedure that divides the population into well-defined
groups from which random samples are drawn is known as
stratified sampling.
4) A population is made up of groups that have wide variation between groups
but little variation within each group. In this situation the appropriate type
of sampling procedure to use is
a) Cluster sampling
b) Systematic sampling
c) Stratified sampling
d) Multistage sampling

12.9 NON-PROBABILITY SAMPLING

PROCEDURES
There are different types of non-probability sampling such as
1. Convenience Sampling
2. Judgment Sampling
3. Quota Sampling
4. Snowball Sampling
We discuss the procedure of drawing a non-probability sampling below.
12.9.1 Convenience Sampling
This is one of the most commonly used methods of non-probability sampling. In
this method the researcher’s convenience forms the basis for selection of the
sample. Especially for a exploratory research there is a pressing need for data. In
such situations the selection of sampling units is left to the interviewer. The
population units are included in the sample simply because they are in the right
place at the right time. This method is often used during preliminary research
efforts to get a gross estimate of the results, without incurring the cost or time
required to select a sample. For example, during the budget session or when the
price of a product is increased or a new government is formed, convenience
samples are used by the researchers/journalists to reflect public opinion.
Convenience samples are extensively used in marketing research.
The advantage of convenience sampling is that it is less expensive and less time-
consuming. The limitations of convenience sampling are: (a) it involves sample

284
selection bias, and (b) it does not provide a representative sample of the Sampling
Procedure
population and therefore we cannot generalise the results.

12.9.2 Judgment Sampling

This is another commonly used non-probability sampling procedure. This
procedure is often referred to as purposive sampling. In this procedure the
researcher selects the sample based on his/her judgment. The researcher believes
that the selected sample elements are representative of the population. For
example, the calculation of consumer price index is based on judgment sampling.
Here the sample consists of a basket of consumer items and other goods and
services which are expected to reflect a representative sample.
The prices of these items are collected from selected cities that are viewed as
typical cities with demographic profiles matching the national profile.
The advantage of judgment sampling is that it is low cost, convenient and quick.
The disadvantage is that it does not allow direct generalisations to population.
The quality of the sample depends upon the judgment of the researcher.
12.9.3 Quota Sampling
In this procedure the population is divided into groups based on some
characteristics such as gender, age, education, religion, income group, etc. A
quota of units from each group is determined. The quota may be either
proportional or non-proportional. The proportional quota sampling is based on
the proportion of each characteristic in the population so that the proportion in
the sample represents the population proportion. For example, if you know that
there are 80% of the households whose income is below say Rs.100000 per
annum and 20% households have income above Rs.100000 per annum in a city.
You want to take a sample of size 100 households. Then you include 80
households from below Rs.100000 income and 20 households from above
Rs.100000 income. The objective here is to meet the proportional quota of
sampling from each characteristic in the population.
The non-proportional quota sampling is a bit less restrictive. In this procedure,
you specify the minimum number of sampled units from each group. You are not
concerned with having proportions in the population. For instance, in the above
example you may simply interview 50 households from each income group
instead of 80% and 20%. The interviewer is instructed to fill the quota for each
group based on convenience or judgment. The very purpose of quota sampling is
that various groups in the population are represented to the extent the investigator
desires.
Do not confuse the quota sampling with stratified sampling that you have learned
earlier. In stratified sampling you select random samples from each stratum or
group whereas in quota sampling the interviewer has a fixed quota. For example,
in a city there are five market centres. A company wants to assess the demand for
its new product and sends 5 investigators to assess the demand by interviewing
285
Sampling and 50 prospective customers from each market. It is left to the investigator whom
Statistical Inference
he/she will interview at each market centre. If the product is targeted to women,
this way you cannot elicit the information among various groups of women
customers like housewives or employed women or young or old. In this sampling
you are simply fixing a quota for each investigator.
The quota sampling has the advantage over others if the sample meets the
characteristics of the population that you are looking into. In addition, the cost
and time involved in collecting the data are greatly reduced. However, there are
many disadvantages as well. In quota sampling, the samples are selected
according to the convenience of the investigator instead of selecting random
samples. Therefore, the selected samples may be biased.
If there are a large number of characteristics on the basis of which the quotas are
fixed, then it becomes very difficult to fix the quotas/sub-quotas for each
group/sub-group. Also the investigators have the tendency to collect information
only from those who are willing to provide information and avoid unwilling
respondents.
12.9.4 Snowball Sampling
In snowball sampling, we begin by identifying someone who meets the criteria
for inclusion in our study. We then ask him/her to recommend others who also
meets the criteria. Although this method would hardly lead to representative
samples, there are times when it may be the best method available. Snowball
sampling is especially useful when we are trying to reach populations that are
inaccessible or hard to find. For example, if we are studying the homeless, we are
not likely to find good lists of homeless people within a specific geographical
area. However, if we go to that area and identify one or two, we may find that
they know very well who the other homeless people in their vicinity are and how
we can find them.

12.10 DETERMINATION OF THE SAMPLE SIZE

The use of appropriate sampling procedure is necessary for a representative
sample. However, this condition is not sufficient. In addition to the above, we
should determine the sample size. The question of how large a sample should be
is a difficult one. Sample size can be determined by various considerations. The
following are the some of the considerations in determining the sample size:
Sampling error
Number of comparisons to be made
Response rates
Funds available
Sampling Error: In Unit 13 you will learn that smaller samples have greater
sampling error than large samples. On the other hand, larger samples have larger
non-sampling errors than smaller samples. The sampling error is a number that
286
describes the precision of an estimate of the sample. It is usually expressed as a Sampling
Procedure
margin of error associated with a statistical level of confidence. For example, for
a prime minister preferential poll you may say that the incumbent is favored by
65% of votes, with a margin of error (precision) of plus or minus 5 percentage
points at a 95% confidence level. This means that if the same surveys were
conducted with 100 different samples of voters, 95 of the surveys would be
expected to show the incumbent favoured by between 60% and 70% of the voters
(65%  5%). Remember as you increase the precision level of your results you
need larger sample size.
Number of Comparisons to Make: Sometimes we may be interested in making
comparisons of two or more groups (strata) in the sample. For example, we may
want to make the comparison between male and female respondents or between
urban and rural respondents. Or we may want to compare the results for 4
geographical regions of the country say north, south, west and east. Then we
need an adequate sample size in each region or stratum of the population.
Therefore, the heterogeneity of population characteristics plays a significant role
in deciding the sample size.
Response Rates: In mail surveys, we know that all those questionnaires mailed
to the respondents may not reach us back after filling the questionnaires. As per
the experiences on mail survey, the response rate ranges between 10% and 50%.
Then, if you are expecting a 20% response rate, for example, you will have to
mail 5 times the number of sample size required.
Funds Available: The funds available may influence the sample size. If the funds
available for the study are limited then you may not be able to spend more than a
certain amount of the total money available with you on collecting the data.
It is even more difficult to decide the sample size, when you use the non-
probability sampling procedures. This is because there are no definite rules to be
followed in non-probability sampling procedures. It all depends upon on what
you want to know, the purpose of inquiry, what will be useful, what will have
credibility and what can be done with available time and resources. In purposive
sampling, the sample should be judged on the basis of purpose. In non-
probability sampling procedures, the validity, meaningfulness, and insights
generated have more to do with the information-richness of the sample units
selected rather than the sample size.
Some Formulae to Determine the Sample Size
Technical considerations suggest that the required sample size is a function of the
precision of the estimates you wish to achieve, the variance of the population and
the confidence level you wish to use. If you want more precision and confidence
level then you may need larger sample size. The more frequently used confidence
levels are 95% and 99%. And the more frequently used precision levels are 1%
and 5%. There are different formulae used to determine the sample size

287
Sampling and depending upon various considerations discussed above. In this section we will
Statistical Inference
discuss three of them.
If we wish to report the results as percentages (proportions) of the sample
responding, we use the following formula:
Pi (1  Pi )
ni 
A 2
P (1  Pi )
2
 i
Z Ni

Where, ni = sample size of the ith attribute required

Pi = estimated proportion of the population possessing ith attribute of interest (for
example, proportion of males, females, urban, rural, etc.)
A = precision required (0.01, 0.05 etc.)
Z = standardized value indicating the confidence level (Z=1.96 at 95%
confidence level and Z=2.58 at 99% confidence level)
Ni = population size of the ith attribute (known or estimated)

Example 12.8: A population consists 80% rural and 20% urban people. Given
that the population size is 50000, determine the sample size required. Assume
that the desired precision and confidence levels are 1% and 99% respectively.
In this example,
P1 = proportion of rural people = 0.80
P2 = proportion of urban people = 0.20
N1 = rural population size = 50000  0.80 = 40000
N2 = urban population size = 50000  0.20 = 10000
A = 0.01
Z = 2.58 (at 99% confidence level)
The required sample size is
P1 (1  P1 )
n1 = rural sample =
A 2
P (1  P1 )
2
 1
Z N1
0.80(1  0.80)
=
0.012 0.80(1  0.80)
2

2.58 40000

0.80(0.20)
=
0.0001 0.80(0.20)

6.6564 40000
0.16
=
0.16
0.000019 
40000

288
 0.16 Sampling
=
0.000019  0.000004 Procedure

0.16
= = 8410.8 or say 8411
0.000023

P2 (1  P2 )
n2 = urban sample =
A 2
P (1  P2 )
2
 2
Z N2
0.20(1  0.20)
=
0.012 0.20(1  0.20)
2

2.58 10000

0.20(0.80)
=
0.0001 0.20(0.80)

6.6564 10000
0.16
=
0.16
0.000019 
10000
0.16
=
0.000019  0.000016
0.16
= = 4568.4 or say 4568
0.000035

Therefore we need to have a sample of size 8411+4568 = 12979 units.

If we wish to report the results as means (averages) of the sample responding, we
use the following formula:
2
Pi
ni  2
A 2 Pi

Z 2 Ni

Where, ni = sample size of the ith attribute required

Pi = estimated standard deviation of the ith attribute of interest (for example,
average income of high income group, low income group etc.)
A = precision required (0.01 or 0.05 as the case may be)
Z = standardized value indicating the confidence level (Z=1.96 at 95%
confidence level and Z=2.58 at 99% confidence level)
Ni = population size of the ith attribute (known or estimated)

Example 12.9: It is planned to conduct a study to know the average income of

households. Given that the standard deviation of households is 2.5 and the
population size is 10000, determine the sample size required. Assume that the
desired precision and confidence levels are 5% and 95% respectively.
In this example,
P1 = standard deviation of income = 2.5
289
Sampling and N1 = number of households = 10000
Statistical Inference
A = 0.05
Z = 1.96 (at 95% confidence level)
The required sample size is
2
P1
n1  2
A 2 P1

Z 2 N1

2.5 2
=
0.05 2 2.5 2

1.96 2 10000
6.25
=
0.0025 6.25

3.8416 10000
6.25
=
0.000651  0.000625
6 .25
= = 4898
0.001276

If we wish to report the results in a variety of ways or we have the difficulty in

estimating the proportion or standard deviation of the attribute of interest, we use
the following formula:
0.25
n 2
A 0.25

Z2 N

Where, n = sample size required

A = precision required (0.01 or 0.05 as the case may be)
Z = standardized value indicating the confidence level (Z=1.96 at 95%
confidence level and Z=2.56 at 99% confidence level)
N = population size (known or estimated)

Example 12.10: Given that the population size is 10000, determine the sample
size required when desired precision and confidence levels are 5% and 99%
respectively.
In this example,
N = 10000
A = 0.05
Z = 2.58 (at 99% confidence level)
The required sample size is
290
 0.25 Sampling
n
0.05 2
0.25 Procedure

2.582 10000

0.25
n
0.0025 0.25

6.6564 10000
0.25 0.25
n   624
0.0003756  0.000025 0.000401

Check Your Progress 2

1) Say whether the following statements are ‘True’ or ‘False’.
a) When the units included in the sample are based on judgment of the
investigator, the sampling is said to be random.
b) With increasing sample size the sampling error decreases.
c) Convenience sampling has the disadvantage that it may not be
representative sample.
2) One of the major disadvantage of judgment sampling is
a) The procedure is very cumbersome
b) The sample selection depends on the individual judgment of the
investigator
c) It gives small sample size
d) It is very expensive

12.11 LET US SUM UP

The most commonly used probability sampling procedure is the simple random
sampling which allows a chance to all population units to be included in the
sample. The sample units are chosen using random number tables. A systematic
random sample uses the first sample unit at random as a starting point and the
subsequent sample units are chosen systematically. A stratified sample
guarantees inclusion of units from each stratum. A cluster sample involves
complete enumeration of one or more randomly selected clusters.
The non-probability sampling procedures include convenience sampling,
judgment sampling, quota sampling and snowball sampling. These sampling
procedures are not independent from sampling bias but still popular in some
situations particularly marketing research.
A number of factors decide the sample size. It may be the number of groups in
the population, the heterogeneity of population, funds and time available, etc.

291
Sampling and Using a sample saves a lot of money, time and manpower. If a suitable sampling
Statistical Inference
procedure is used in selecting units, appropriate sample size is selected and
necessary precautions are taken to reduce sampling errors, then a sample should
yield a valid and reliable information about the population.

12.12 ANSWERS/HINTS TO CHECK YOUR

PROGRESS EXCERCISES
Check Your Progress 1
1) d
2) c
3) a) False
b) True
4) c
Check Your Progress 2
1) a) False
b) True
c) True
2) b

292