1

Research Methodology
The study of conducting research is Research
Methodology.
Research: The word research is composed of two syllables
Re and Search.
Re is the pref meaning !"gain or o#er again or a
new$ and Search is the latter meaning !to eamine closely
and carefully$ or !to test and try$.
Together they form% a careful% systematic% patient
study and in#estigation in some feld of &nowledge
underta&en to establish principles ' policies.
Research can also be defned as
1. Search for &nowledge
(. Systematic and scientifc search for getting rele#ant
answers on any ta&en up specifc topic.
). Scientifc en*uiry into a sub+ect.
,. Research is a mo#ement from the un&nown to the
&nown.
-. .t is the #oyage of disco#ery
"cc to /ulmer%
Research is primarily committed to establishing
systematic% reliable and #alid &nowledge about the social
world.
"cc. To 0li1ord 2oody%
(
Research comprises of
3efning and redefning problems.
4ormulating hypothesis 5basic idea6
0ollecting
7rgani8ing
9#aluating datas
Ma&ing decisions
Suggesting solutions
Reaching conclusions
4inally% carefully testing the conclusions
To determine whether they ft the formulated :ypothesis.
Research Methods: May be understood as all those
methods or techni*ues that are used by a researcher for
conducting a Research depending upon the methods.
(1) Library Research: analysis of historical
records and documents.
; Statistical compilation% references% abstracts% guides
manipulation 5handle with s&ill6
(2) Field Research: 7bser#ation%
*uestionnaires% personal% <roup or telephonic
inter#iews% case study.
(3) Laboratory Research:
<roup 5team6 study% use of audio #isual tools.
)
Research Methodology: is the way do systematically
sol#e the research problem.
.n it we study the #arious steps that are generally
adopted by a researcher in studying his research problem
logically.
2hen we tal& of Research Methodology% we not only
tal& of research methods but also consider the logic behind
the methods we use in the contet of our research study
and eplain why we are using a particular method or we are
not using a particular method or techni*ue so that research
results are capable of being e#aluated either by the
researcher or others.
Steps:
16 2hy a particular research study has been underta&en=
(6 :ow the Research problem has been defned=
)6 2hat way and why the hypothesis 5basic idea6 has
been formulated=
,6 2hy a particular techni*ue of analy8ing data is used=
5or6 :ow the data were collected=
-6 :ow the collected data were interpreted=
>6 2hat deletion was made=
?6 2hat was the conclusion=
4inally what was the solution for the Research problem=
Importance of no!ing the s"b#ect $ research
Methodology:
,
16 " student preparing himself for a career of carrying out
research as his profession @
; 2ill be trained better to do research
; 2ill help him de#elop disciplined thin&ing
; 2ill help him obser#e the feld ob+ecti#ely.
; 2ill enable thoroughly to understand the logic
behind the research problem.
; 2ill increase the ability to e#aluate the results.
; 4ace the e#aluated results with confdence.
; Aseful in #arious felds such as <o#t. /usiness%
administration% community de#elopment B social
wor&.
%o &"alify a Research or st"dy:
To be a <ood or perfect one%
The Research adapted should process certain
characteristics%
.t must as far as possible be 16 0ontrolled
(6 Rigorous
)6 Systematic
,6 Calid
-6 Cerifable
>6 9mpirical
?6 0ritical
1' (ontrolled: The research problem should not be
a1ected or inDuenced by eternal factors 5i.e. #ariables
other than the participating facts6.
-
2' Rigoro"s: The procedures followed to fnd answers to
*uestions should be rele#ant% appropriate B +ustifed. /ut
the degree of rigiour may #ary from one problem to another
problem.
3' )ystematic: The in#estigation should follow a certain
logical se*uence 5Eot in a hapha8ard manner6
*' +alid , +erifable: The fndings should be #alid B
can be #erifed by you or others at any time.
-' .mpirical: The conclusions drawn should be based on
hard e#idence% gathered from real life eperiences or
obser#ations.
/' (ritical: The process of in#estigation must be
foolproof and free from drawbac&s. The process adapted
and the procedures used must be able to withstand any
critical scrutiny.
Types of Research
Research can be classifed from the #iew point or
perspecti#es as%
4rom the #iew point
0pplication ob#ecti1e
In&"iry mode

16 Fure Research 16 3escripti#e 16 Guantitati#e
Research
(6 "pplied Research (6 0orrelati#e (6 Gualitati#e
Research
>
)6 9ploratory
,6 9planatory
16 2"re Research: (3asic or F"ndamental Research)
<athering% &nowledge is termed as !pure$ or !basic$
research. Hust to gather &nowledge in order to formulate or
generali8e theories or policies.
9g6 Research on mathematics.
This types of research adds &nowledge to the already
eisting organi8ed body.
0pplied Research: To fnd an immediate solution for a
pressing practical problem.
9g: Social% economical and political trends pre#ailing in
a country.
0pplied +s F"ndamental 3ased on the ob#ecti1es of
Research:
1) 4escripti1e Research:
Sur#ey or fact fnding en*uires of di1erent
&inds. .t describes the actual pre#ailing
state of a1airs% eisting at present.
7therwise &nown as e post facts means
eisting position of facts ' issues.
:ere the #ariable inDuencing the research
has no control or the researcher has no
control o#er the #ariables.
9g: 4re*uency of shopping% customer preference
etc.
?
2) (orrelati1e Research:
<oes on to disco#er the eisting
relationship or interdependence between
two or more aspects ' #ariables.
7therwise &nown as comparati#e study.
.n#estigates association between #ariables.
9g: Sum of humour and +ob satisfaction%
5related #ariable6
Research problem is wor&ers turno#er
0nalytical Research:
The researcher has to use facts ' information already
eisting and analy8e these data to ma&e a critical
e#aluation.
9g: document study ' historical e#idence.
4escripti1e +s 0nalytical Research:
.5planatory Research:
"ttempts to clarify or eplain why and how% any
particular research problem arises and can be sol#ed.
*' .5ploratory Research: Study underta&e to eplore
a new area or an un&nown destination.
.... 3ased on the In&"iry Mode:
1) 6"antitati1e Research:
I
Relates to aspects that can be *uantifed and
epressed in terms of *uantity.
7therwise &nown as structured Research.
.n this type of Research% the ob+ecti#es% design%
sample and all the other factors inDuencing the
research is pre determined.
The research problem and its solution will be
epressed in terms of *uantity and hence statistical and
economic analysis is adapted in this type of Research.
6"antitati1e Research:
7therwise &nown as unstructured research.
The aspects related to *uality ' &ind or teture.
9g: /eha#iour science
0part from the abo1e7 other types of Research are%
(oncept"al Research: Research related to some abstract
idea or theory
Ased by philosophers or thin&ers for de#eloping new
concepts.
.mpirical research
5based on eperiments or eperience6
7therwise &nown as eperimental type of Research.
The result obtained by adapting 9mpirical Research is
considered to be most powerful 5e#idence enclosed6
/ased on the time consumed to complete a particular
research%
J
a6 one time Research:restricted to a single time period.
b6 Kongitudinal Research: 0onducted o#er se#eral time
period.
6"alities of a Researcher
%op 18 &"alities of a Researcher
1) 0nanalytical mind: 0onstant analysis on a
#ariety of factors.
2) 0 people person : 4or respondents to get the
best out of inter#iews ' focus groups.
3) %he ability to stay calm: especially when you ha#e
pressing deadlines. Leep well focused and thin&
logically there will always be an end point.
*) Intelligence : Researcher re*uires critical
analysis% but most of all common sense.
-) ("riosity::a#e curiosity and be passionate about
de#eloping deeper to unearth more insight.
/) 6"ic thiner: Things don$t always go as you plan% so
you need to be able to thin& fast.
9) (ommitment: Research is a tough +ob% the hours
may be long% the deadlines short. !
:) .5cellent !ritten and 1erbal comm"nication
sills: So that di1erent audience can clearly
understand the fndings.
;) )ympathetic: :a#ing a sympathetic ear when
listening to some respondents 5cry etc6 is a good s&ill%
to ha#e.
1M
18) )ystematic: 0hec&% chec& and chec& again.
Spending a proper amount of time for chec&ing always
pays.
"ccording to Micheal 4oster%
16 Truthful data ' facts @ desire for accuracy of
obser#ation.
(6 Eo epressions li&e approimately% almost or nearly.
)6 Should poccess alert mind. Eature is constantly
changing% be &een and watchful to notice such
changes% no matter how small or insignifcant they
may.
,6 Scientifc in*uiry @ desire for &nowledge @ it re*uires
moral courage% Steadfast 5constant ' not changing6
endurance 5to tolerate the diNculty% su1er patiently6
; 2hen a research scientist feel defeated or
completely lost% he needs immense courage and the
sense of con#iction 5found guilty6
)ignifcance or Importance of )earch
16 3oubt is better than o#er confdence for it leads to
in*uiry% in*uiry leads to in#ention. Frocess or the three
stages of research to bring out economic policies.
16 .n#estigation of pre#ailing economic structure with
the a#ailable facts.
(6 "nalyse or diagnose the data.
)6 Frediction for future de#elopments.
(6 Research encourages scientifc and inducti#e thin&ing.
11
9g:; Role of Research in :
16 .con omics:
Researches done on applied 5production and sales
of goods in a proftable manner6 economics is
increasing in modern days. <o#t. B business sectors
ha#e become more comple% they face se#eral
operational problems to sol#e this problems%
Research is carried on.
To frame <o#t. economic policies.
<o#t. budget a formulation depends on the
analysis of needs B desires of the people%
a#ailable of re#enues needs research.
3ecision ma&ing @ re*uires proper research.
"llocation of a countries scarce resource @ also
needs research.
(6 /usiness 3ecisions:
.n business sectors there are both planning
and operational problems.
a6 Froblems Research: .n#estigation of
the present structure and de#elopment of
the mar&et @ relating to purchase%
production% promotion and sales.
b6 7perational Research: Relates to
application of logical% mathematical and
analytical techni*ues @ to sol#e mar&et
problems @ there by minimi8e cost and
proft maimi8ation.
1(
c6 Moti#ational Research: :elps to
determine people beha#ior or consumer
response.
"ll the abo#e three are responsible for business
decision ma&ing.
)6 Social sectors: To gain &nowledge on un&nown
aspects and do something better and more
eNciently.
Social scientist gain their &nowledge for their own sa&e
and for the de#elopment of the society.
1. 4ormulating the Research problem:;
a6 The formulation of a general topic into a specifc
Research problem thus constitutes the frst step in a
scientifc in*uiry.
Two steps are in#ol#ed in formulating the Research
problem%
a6 Anderstanding the problem thoroughly.
b6 Rephrasing the same into meaningful terms from an
analytical point of #iew.
1. .dentify a broad feld or sub+ect area of interest to you.
(. 3issect the broad area into small area.
). Select what is of most interest to you.
,. Raise Research *uestions.
-. 4ormulate ob+ecti#es
>. "ssess your ob+ecti#es
1)
?. 3ouble chec&
The best way to understand the problem is to
discuss with his own colleague or guide.
9amine all a#ailable literatures to get himself
ac*uainted 5get used to 6 with the selected
problem.
Re#iew two types of literature
0onceptual literature :
0oncerning concepts B theories
9mpirical Kiterature : 0oncerning studies made earlier
which are similar to the one proposed.
7utcome of the re#iew will be the &nowledge so as to
pre determine what data or materials are a#ailable for
operational purposes.
Eet step @ the Researcher rephrases the problem into
"nalytical or operational terms.
2<% %=. 2R>3L.M I?%> )2.(IFI( %.RM)
This step is of greatest importance in the entire
research process.
The problem to be in#estigated must be defned
unambiguously or clearly.
Frof 2.". Eeiswanger States%
The statement of the ob+ecti#e of the Research
problem is of basic importance because%
5i6 .t determines the data which are to be collected
1,
5ii6 0haracteristics of the rele#ant data
5iii6 0hoice of techni*ues to be used in these
eplorations
5i#6 4rame a 4inal report
Step ..
.5tensi1e Literat"re )"r1ey:
" brief summary of the problem should be written
down.
Ma&e etensi#e literature sur#ey
Sources of sur#ey can be% +ournals% bio;graphics% <o#t.
reports% boo&s% conference proceedings etc.
/ased on the nature of the problem.
9arlier study if any which is similar to the study in
hand should be carefully studied.
" good library will be a great help to the researcher at
this stage.
Stage ...:
4e1eloping =ypothesis :53e#elopment of wor&ing
:ypothesis6
State in clear terms the wor&ing hypothesis 5/asic .dea
of the Research problem6
.t is a tentati#e assumption in order to test to logical or
empirical conse*uences.
Fro#ide the focal point for research.
1-
:ypothesis should be #ery specifc and #ery well
limited to the place of research in hand because it has
to be tested.
:ypothesis guides the researched by limiting the area
of Research and &eep him on the right trac&.
.t sharpens his thin&ing and focuses attention on
important facets of the problem.
.t indicates the type of data re*uired for the study.
Type of methods of data analysis done.
=o! to de1elop !oring =ypothesis@
16 3iscuss with collogues ' eperts% about the problems%
its origin% its ob+ecti#es and solutions.
(6 9amination of data' records if a#ailable.
)6 Re#iew similar studies ' similar problems.
,6 To secure greater insight into the practical aspects of
the problem @ conduct personnel in#estigation or feld
inter#iews.
)%.2 I+:
2reparing the Research 4esign:
Research design is the conceptual structure within
which research is conduction. .t constitutes the blue print for
the collection% measurement and analysis of data.
The function of the Research design is to pro#ide
rele#ant e#idence with minimal ependiture of e1ort% time
and money. .t pro#ides an outline of what the researcher is
1>
going to do in terms of 16 4raming the hypothesis% (6 its
operational implications and )6 fnally data analysis
The Research design highlights certain decision%
16 The nature of the study
(6 Furpose of the study
)6 Kocation where the study would be conducted
,6 The nature of data re*uired
-6 4rom where the data would be collected
>6 The techni*ues of data collection that would be
used
?6 2hat time period the study would co#er
I6 The type of sample design that would be used
J6 The method of data analysis that would be adapted
1M6 The manner in which the report would be
prepared
%ype of Research 4esign : , types
16 )ampling Research 4esign : 3eal with selection of
rele#ant items
(6 >bser1ational Research 4esign: 3eals with the
obser#ations 5feld obser#ations6 that is to be made.
)6 )tatistical Research design: 3eals with the
information on the data collected B analysed.
,6 >perational Research 4esign: :ow the abo#e
three are carried out.
1?
4etermining sample 4esign:
"ll the items considered in any feld of in*uiry
constitutes a uni#erse or population. Study of the
entire population without lea#ing out a single item is
&nown as 0ensus Study
This type of census study is practically not possible.
So we select few items from the entire population for
our study purpose. The items so selected constitutes
what is technically called sample.
The way of selecting such a sample is &nown as the
Sample 3esign.
These samples can be either probability samples or non
probability samples.
Frobability: 9ach item in the population has on e*ual
chance of being selected for the study.
16 Simple random sampling
(6 Systematic random sampling
)6 Stratifed random sampling
,6 0luster ' area random sampling.
Eon Frobability sampling: "ll the items do not ha#e an
e*ual chance of being selected for the study.
The selection depends upon the con#enience B
+udgment of the Researcher.
Mied sampling: 2hen more than one type of sampling
techni*ue is used for a study% it is mied sampling.
1I
The sample design to be used in a Research study
must be decided by the researcher considering the nature of
the study.
>. (ollecting the 4ata:
Aathering appropriate dataB which are made use
in Research study.
3ata can be collected in se#eral ways either through
516 9periment 5or6 5(6 through sur#eys.
.n eperimental means% when a researcher
conducts a research% some *uantitati#e
measurements are obser#ed% based on which% he
eamines the truth of the underlying hypothesis.
.n case of sur#eys% data are collected by
16 /y obser#ations
(6 Through personnel inter#iew
)6 Through telephone inter#iews
,6 /y mailing of *uestionnaires
-6 Through schedules ' enumerators
The Researcher should select one of these methods of
collecting the data ta&ing in account the
1) Eature of in#estigation
2) 7b+ecti#e B scope of .n*uiry
3) 4inancial Resources
*) Time frame
-) 3esired degree of "ccuracy.
1J
/) .5ec"tion of the 2ro#ect: (2"tting a plan)
.mportant step in Research study.
See that the pro+ect is eecuted in a systematic
manner and in time.
9g6 .f the sur#ey done in a pro+ect is #ia
Guestionnaire the answers can be machine
coded ' processed
.f inter#iew were conducted% ma&e sure that the
inter#iewers is well trained @ to &eep the sur#ey as
much as realistic as possible.
I. 0nalysis of 4ata :
"fter the data are collected the researcher turns to
the tas& of analy8ing the data the analysis of data
re*uire closely related operations% li&e !coding%
9diting B Tabulation$.
The wide data collected should be condensed into
small manageable groups% for easy analysis.
0oding: The collected data are transformed into
symbols that may be tabulated or counted.
9diting: Anwanted B irrele#ant data will be
remo#ed.
Tabulation: Technical procedure where the data
are put in the form of tables.
Research 4esign:
The most important step after defning the !Research
problem$ is preparing the Research 3esign
(M
Research design is the conceptual structure within
which the research is conducted.
.t constitutes the !/KA9 FR.ET for collection%
measurement and analysis of data.
Research design pro#ides an answer to the *uestion%
what the Researcher is going to do with regards to
framing hypothesis% its operational implications and
how to analyse the data=
Research 4esign: C 4ecisions
:ighlights certain decisions%
16 Eature of the study
(6 Furpose of the study
)6 Kocation where the study would be conducted
,6 Eature of 3"T" re*uired
-6 4rom where the 3"T" can be collected
>6 Time period of the study
?6 Type of sample design to be used
I6 Techni*ues of data collection
J6 Methods of 3ata "nalysis
1M6 Freparation of Report.
Research 4esign
May be sub di#ided into%
16 Sampling design: 3eals with% the method of !selecting
items$ for the study.
(1
(6 7bser#ational design: Relates to the condition under
which the obser#ations are to be made.
)6 Statistical 3esign: 3eals with the no of items
selected or the study and how the selected data will be
analysed.
,6 7peration design: The techni*ue by which the
sampling% obser#ational and statistical designs can be
carried out.
Research 4esign $ Feat"res :
16 :elps to identify the type and source of information
needed for the study.
(6 Specifes the methods to be adopted in collecting B
analy8ing data.
)6 Specifes the time schedule of the research and the
monetary budget in#ol#ed.
(oncepts Relating to Research 4esign
16 4ependent and Independent 1ariables :
+ariables : " magnitude that #aries is &nown as
#ariable
(ontin"o"s 1ariable : Calues that can be epressed
e#en in decimal poins are &nown as continuous #ariables
9g: age 5, years ) months6
:eight 5-.( cm6
2eight 5,-.) &g6
((
?on contin"o"s +ariables: Calue that can be
epressed only in integer #alues are called Eon continuous
#ariables
9g: Eo. of students in a class 5 ,-6
Eo. of children in a family 5)6
Statistically &nown as discrete #ariables
4ependent or .ndogeno"s 1ariables :
2hen the change in one #ariable depends on the
change in other #ariable% it is &nown as dependent or
9ndogenous #ariable.
3emand ;;;;; Frice 5independent6
Independent or .5ogeno"s 1ariable
The #ariable that causes the change in the dependent
#ariable is &nown as independent or eogenous #ariable.
3emand 53ependent6 ;;;;;;; Frice %.ncome
:ere demand is a dependent #ariable while price '
income is an independent #ariable.
.5traneo"s 1ariable :
The independent #ariable which is not directly related
to the purpose of the study but a1ects the dependent
#ariable is &now as 9traneous #ariables.
The inDuence caused by the etraneous #ariable on
the dependent #alue is technically &nown as
9perimental 9rror
" research study or a Research design should always
be framed in such a manner that the inDuence of
()
!9traneous #ariables$ on the dependent #ariable is
completely controlled and the inDuence of the
independent #ariable is clearly e#ident.
(ontrol:
<ood Research design should minimi8e the e1ect for
9traneous #ariables.
(onfo"nded Relationship
The relationship between dependent and independent
#ariable is said to be confounded by an etraneous
#ariables.
Research =ypothesis:
2hen the formulated hypothesis is tested by adopting
scientifc methods% it is &nown as Research :ypothesis.
.5perimental , ?on .5perimental =ypothesis testing:
2hen the ob+ecti#e of the Research is to test the
hypothesis% it is Research hypothesis.
Research in which the independent #ariable are
5handled with s&ill6 manipulated% it is eperimental
hypothesis testing.
2hen the #ariables are not manipulated% it is non
eperimental hypothesis testing.
.5perimental , (ontrol Aro"ps:
2hen a group is eposed to usual conditions in an
eperimental hypothesis% research it is control <roups.
2hen the group is eposed to special or certain new
conditions% it is eperimental groups.
(,
I. %reatments:
The di1erent conditions to which the eperimental B
control groups are sub+ect to is &nown as treatments.
J' .5periment:4ertili8ers and crops6
Frocess of #erifying the truth.
0bsol"te .5periment:
3etermine the fact
(omparati1e .5periment:
3etermine the impact in comparison with
another fact.
1M. .5perimental "nits
Fre;determined bloc& to which di1erent
treatments are applied.
9g : animal testing
%ypes of Research 4esign
There are three di1erent types of Research design%
1) .5ploratory Research 4esign:
.s a 4ormulati#e Research design
Main purpose is the disco#ery of ideas B insights
Should be Deible enough considering di1erent
dimensions of the problem under study.
2) 4escripti1e and 4iagnostic Research 4esign:
(-
3escripti#e Research 3esign is concerned with
describing the characteristics of a particular
indi#idual or a group.
Study concerned with narration of facts or
characters related to an indi#idual% group or
institution are descripti#e research studies.
3iagnostic Research design determines the
fre*uency with which a #ariable occurs or its
relationship with another #ariables.
/oth the Research designs should be planned
carefully.
Research design should be Rigid 5Eo Deibility6
3) =ypothesis testing Research 4esign:
Test the hypothesis of causal relationship
between two or more #ariables.
"dopt procedure that not only reduce bias but
enhance reliability @ and facilitates deri#ing
.nferences 5results6 about the Research problem.
Importance of Research 4esign:
4acilitates the smooth Dow of the #arious stages of
Research.
:elps yield maimum information with minimum
e1ort% time and money.
:elps to plan in ad#ance data collecting and analysis
techni*ues.
Frepare with utmost care to a#oid errors.
(>
"ttain reliability
(haracteristics of a Aood Research 4esign
Fosses the *ualities of being Deible% suitable eNcient
B economical.
Should minimi8e !bias$ and maimi8e reliability of data
collection B "nalysis.
Eo eperimental error should be allowed
Should yield maimum information
Research problem should be #iewed from di1erent
angles or dimensions.
The choice of Research design depends on%
Eature of the Research problem
7b+ecti#es of the Research problem
S&ills ' ability of the Researcher
Methods of gathering information
"#ailability of monetary support
Time schedule
=D2>%=.)I)
" Research hypothesis is a predicti#e statement% which
is capable of being !tested$ using scientifc methods% which
in#ol#es independent and dependent #aluables. 5eg6 the
female students perform as well as the male students.
(?
This statement is a hypothesis that can be ob+ecti#ely
tested and #erifed.
.t is a proposition that can be put to test in order to
eamine its #alidity.
(haracteristics of =ypothesis
16 " hypothesis should be precise and clear. .f not clear%
the inferences will not be reliable.
(6 .t must be capable of being put to test.
)6 .t should state the relationship between the #ariables%
in case relational hypothesis.
,6 .t should be stated in a simple language.
-6 .t should be consistant and deri#ed from all &nown
facts.
>6 :ypothesis must be amenable to testing within a
reasonable period of time
?6 :ypothesis should eplain what it actually to eplain.
5the solution for the Research problem6. The
eplanation should be on empirical reference.
(oncepts Relating to %esting of =ypothesis
16 Eull :ypothesis B "lternati#e
:ypothesis 5Statistical "nalysis6
Eull :ypothesis: 3enoted by :M. .f both the #ariables 5say
male or female6 or 5:ead or Tail6 are e*ually good% it is Eull
:ypothesis.
(I
"lternati#e :ypothesis: 3enoted by :a or :1. .f one
#ariable is considered superior to other or #ice #ersa or if
there is a di1erence% it is alternati#e hypothesis.
Mean 2op"lation (") or (p)
Total ' Eo. of #ariables
?"ll =ypothesis
:o : u O 1MM
0lternati1e =ypothesis
:a : u O 1MM
:a : u P 1MM
:a : u Q 1MM
0spects to be considered !hile form"lating ?"ll
=ypothesis
16 The researcher always tries to re+ect Eull hypothesis
since "lternati#e :ypothesis should be pro#ed.
(6 Eull hypothesis when it is actually true% when re+ected
in#ol#es great ris&% the le#el of signifcance should be
considered.
)6 Eull hypothesis should be #ery specifc 5Eo
approimation6
The le#el of signifcance:
.mportant concept of hypothesis testing.
.t is a certain percentage chosen with great !care%
reason and thought$
(J
5eg6 let us consider the le#el of signifcance to be -R.
.t means the Researcher ta&es a ris& of re+ecting Eull
hypothesis 5:o6 by -R when :o is actually true.
3' 4ecision R"le
The researcher should ma&e a decision% if to accept or
Re+ect :o.
The decision rule should be decided on the number of
items to be tested and the basic of which to accept or
re+ect.
,. %ype I and %ype II .rrors
5i6 Researcher may re+ect :o% when it is true @ Type . 9rror
5which must ha#e been accepted6.
5ii6 Researcher may accept :o% when it is false @ Type ..
9rror 5which must ha#e been re+ected6
-. >ne tailed and %!o tailed %ests:
5i6 7ne tailed test re+ects the Eull hypothesis when the
sample mean is either greater or lower than the
hypothesi8ed #alue of the population mean.
%!o tailed %est: 2hen the sample mean is both
greater and lower than the hypothesi8ed #alue of the
population mean.
2roced"re for =ypothesis %esting:
1. Testing hypothesis refers whether the formulated
hypothesis is #alid or not
(. 2hether to "ccept or Re+ect Eull :ypothesis.
5i6 Maing a formal statement:
)M
Ma&ing a formal statement of the null hypothesis
and alternati#e hypothesis.
5ii6 Selecting a signifcant le#el of testing
" pre;determined le#el of signifcance should be
specifed.
9ither -R or 1R le#el can be considered for the
purpose.
5iii6 3eciding the 3istribution to use:
0hoice should be made generally relates to Eormal
distribution or t;distribution.
5i#6 Selection of random sample B computing an
"ppropriate #alue
Selection of Random sample
0omputing suitable #alue
3rawing a sample for furnishing 9mpirical data.
5#6 0alculation of Frobability:
The di#erged results from the epected results%
when :o is true.
5#i6 0omparing the probability:
/y ma&ing a comparison with the assumed
signifcance le#el.
.f the #alue is less than or e*ual to :o% in case of
one;tailed test% :o is re+ected. :ere type . error is
committed.
)1
.f the #alue is greater than the mean% :o is
accepted. 2ere type;. error is committed.
compile% compare B compute the data and come
out with the inference.
?"ll =ypothesis: The null hypothesis is the proposition
or proposal that implies no e1ect on the phenomena.
0lternati1e =ypothesis: is the one predicti#e statement
that implies some e1ect on the phenomena.
(oncepts Relating to %esting of =ypothesis:
)0M2L. )<R+.D
)ampling 4esign:
"n art of obtaining a sample from a gi#en population.
The techni*ue or procedure the researcher adopts for
selecting items for the sample from the population or
uni#erse.
)teps in sampling 4esign
%ype of <ni1erse:
16 The frst step in sampling design% is to clearly defne
the total number of items ' cases to be studied% which
is technically &nown as AE.C9RS9
4inite
Ani#erse
.nfnite
Finite <ni1erse: The number of items is certain.
)(
9g: Eo. of students in a class.
Eo. of wor&ers in a factory.
Infnite <ni1erse: The number of items is infnite. 5Eo
idea about the number of items6
9g: 0hennai population% Eo. of stars
(6 Sampling Anit:
" geographical area li&e a state% district or
#illage.
4amily% religious community or a school.
.ndi#idual 5Researcher can select one or two such
units6.
)6 Source Kist: 7therwise &nown as Sampling
4rame
0onsists of names of all items of a uni#erse.
.f not a#ailable the researcher has to prepare a
Source list.
.t must be reliable% comprehensi#e% correct and
appropriate.
.t should be the representati#e of the
population ' uni#erse.
,6 Si8e of sample:
Refers to the number of items to be chosen
from the uni#erse.
Si8e of sample must be optimum. "n optimum
sample may defned as the one that satisfes.
))
The re*uirements of representati#es.
4leibility
9Nciency
Reliability
0osts or budget should be considered.
Factors InE"encing siFe of sample :
2arameters of Interest:
The items or parameters are selected based on the
researchers own interest.
3"dgetary constraint :
0ost consideration eercises a ma+or inDuence.
a6 )ampling 2roced"re:
The type or techni*ue used by the researcher to select
the items.
The techni*ue should be selected so that for a gi#en
sample si8e B budget% the sampling error must be #ery
small or negligible.
)ampling .rror: may be caused 5.n case of Eon
probability sampling6 due to
516 .nter#iewer /ias
5(6 Mista&es
5)6 Eon response problems
5,6 Guestionnaire design Daws
),
5-6 3ata processing B analysis errors
.n case of probability sampling% 5homogenous items 6
the sampling error is negligible since the sample is more
accurate.
0haracteristics of a <ood sample :
Should bind a truly representati#e sample.
Small sampling error
Should ft into the budgetary constraints.
Result should be applicable in general.
0haracteristics of sample techni*ues :
16 Much cheaper
(6 Sa#es time
)6 Much reliable
,6 Suitable for carrying out di1erent sur#eys
-6 Scientifc in Eature
"d#antages of sampling:
16 Cery accurate
(6 9conomical in Eature
)6 Cery reliable
,6 Suitable for di1erent sur#eys
-6 Kess time consumption
>6 .n case of large uni#erse% sampling method is the
only practical method for collecting the data.
)-
4iGerent types of sample 4esign:
0lassifed under two general categories.
16 Frobability sampling
(6 Eon @ Frobability sampling.
Frobability sampling:
7therwise &nown as !choice sampling$ or !random
sampling$.
9#ery item has an e*ual chance of being included in
the sample.
9g: Kotteries 5or6 subscribers
2hen done property% probability sampling ensures that
the sample has a similar composition and profle as
that of the entire population.
? di1erent types of probability sampling%
16 Simple Random sampling
(6 Stratifed Random sampling
)6 0luster sampling
,6 Systematic sampling
-6 "rea sampling
>6 Multi stage sampling
?6 Sampling with probability propotional to si8e
516 Sample Random Sampling
The sample is drawn so that each person or item has
an e*ual chance of being drawn during the selection.
)>
9g: Kotteries 5in a ball bo6
5(6 Stratifed Random sampling: 5strata @ layers6
; Stratifed sampling techni*ue is generally used
when the population is heterogeneous.
; The entire population is di#ided into sub population
5sub groups6 5i.e. the sub population being
homogenous6.
; .tems are selected from each stratum.
; This method is more reliable B accurate.
; 9g: -M students of a school ha#ing 1MMM
students on a total were selected B inter#iewed on
the interest in music.
The students were grouped based on their age% ?
years% I years% J years% 1M years B 11 years.
4rom each age group% - students were chosen and
totally -M students were inter#iewed.
5)6 0luster sampling:
The entire population is sub di#ided into mutually
eclusi#e groups or clusters
Simple Random sampling is applied and the need
clusters are selected for the study.
.f all the elements found in the selected cluster is
ta&en for the study% then it is one stage cluster
sampling.
)?
.f random sampling is applied in selecting elements
found within the clusters% it is two stage cluster
sampling.
5,6 Systematic sampling:
; Selecting e#ery nth element for the study.
1
st
% -
th
% 1M
th
% 1-
th
% (M
th
SSSSSSSSSSSSS.
5-6 "rea sampling: when the clusters are in the form of
some geographical sub di#isions.
5>6 Multi stage sampling:
; .f the researcher selects elements or items at di1erent
stages% it is multi stage sampling.
; 9g: Sur#ey of wor& eNciency in nationali8ed baul&s
; Stage le#el
; 3istrict le#el
; Towns
; Cillages
.tems are selected at four stages ' le#els.
5?6 Sampling with probability proportional to si8e:
; The probability of inclusion into the study is directly
proportional to the si8e of the clusters.
; This techni*ue is used when the no. of elements
present in each cluster #aries.
; 3epending on re*uirement of the researcher the
cluster which is more appropriate can be selected for
the study.
)I
... ?on 2robability sampling:
; 9ach item does not ha#e an e*ual chance of being
included in the sample.
; ) types of non probability samplings are%
16 0on#enience sampling
(6 Guota sampling
)6 Hudgmental sampling
16 (on1enience sampling:
; 0hoosing items at the con#enience of the
Researcher.
9g: Street inter#iews 5sampling of people who
are at easy access6
3rawbac&: Kac& of accuracy
(. 6"ota sampling
The researcher simply assume *uotas% with certain
restrictions imposed on how they should be selected.
9g: 0aste basic
/enefts: less epensi#e% #ery con#enient
). H"dgment sampling : 5otherwise &nown as purposi#e
sampling6
Researcher employs his own 9pert +udgment about
who to include in the sample frame.
4isad1antages of sampling:
16 .nade*uacy of samples
)J
(6 0hances of bias
)6 Froblems of accuracy
,6 3iNculty of getting the representati#e sample.
-6 Antrained Manpower.
>6 "bsence of .nformants
?6 0hances of committing errors.
)ampling .rror: Sampling error is the de#iation of the
selected sample from the true characteristics% traits%
beha#ior% *ualities or fgures of the entire population.
4ata (ollection
3ata 0ollection is the systematic gathering of
information 5data6 for a particular purpose from #arious
sources. 5Carious sources can be *uestionnaires% inter#iews%
obser#ations eisting records and electronic de#ices6.
Two .mportant sources of .nformation are%
5i6 Frimary 3ata
5ii6 Secondary 3ata
2rimary 4ata:
Refers to the data collected for the frst time 57riginal
data6
.5ample: Froceedings from conferences meetings%
Students records.
)econdary 4ata:
Refers to the data that ha#e already been collected
and used earlier by somebody or some agency.
.5ample: 7nline database% :istorical boo&s etc.
.5ample: Ta&ing census 5total population6 of <o#ernment
of .ndia @ Frimary 3ata.
,M
2hen research is done by some other scientist on the
basis of this data% it is secondary data.
Selecting " particular source of 3ata depends on%
516 Furpose B Scope of the study.
5(6 "#ailability of time.
5)6 "#ailability of Resources.
5,6 The degree of "ccuracy desired.
5-6 Statistical tools to be used.
5>6 Sources of .nformation 5data6.
1' 2"rpose , )cope of 4ata (ollection:
Should be clearly stated at the #ery beginning of the
study.
" statement indicating the Research problem and the
type of information needed for the study in order to
sol#e the Research Froblem% is needed.
.ts purpose is to establish a factual information for
ma&ing decisions.
Scope of the en*uiry means the co#erage with regard
to the type of information% the sub+ect matter and
geographical area.
(b) 01ailability of %ime:C
The in#estigation should be done within a reasonable
period of time.
Ta&ing which the information collected may become
outdated.
9ample: 3emand of a new product launched is
studied% if the result comes out after ( years% by the
time the producer may attain great lose.
,1
So% ma&e sure the in#estigation is carried out within a
reasonable period of time.
(c) 01ailability of Reso"rces:
.n#estigation or data collection greatly depends% on
resources a#ailable li&e%
1. Eo. of s&illed personnel
2. The 4inancial Fosition.
.f the no.of s&illed personnel to carry our the en*uiry is
suNcient and the a#ailability of funds is not a problem%
the datas can be collected from a big area co#ering a
good number of samples.
(d) %he desiredIe5pected 4egree of 0cc"racy:C
3eciding the degree of accuracy is a must for the
in#estigator.
/ecause "bsolute "ccuracy is not possible in
statistical wor&s.
Since Statistics is based on estimates% tools used for
measurements in not always perfect B there may be
unintentional bias on the part of the in#estigator%
enumerator or informant.
<enerally the degree of accuracy depends upon the
ob+ections of the en*uiry.
9ample: 3uring purchase of <old% e#en 1'1M
th
gram in its
weight is signifcant.
/ut it is not the same in case of purchasing rice or
wheat.
,(
(e) )tatistical %ools to be "sed:
2ell defned or identifable ob+ect or group of ob+ects
that can be measured or counted in a statistical
in#estigation is called statistical Anit.
.n the absence of a clear and precise Statistical Anit
Serious errors may be committed by collecting
irrele#ant data. This will ultimately lead to fallacious
52rong6 conclusions.
(f) )o"rces of Information:
The researcher has to decide about the source from
which the information can be obtained or collected.
1. 4irst hand data.
(. The data from other published sources.
5Fublications% Hournal% Maga8ines etc.6
(g) Method of 4ata (ollection:C
1. Frimary 3ata.
(. Secondary 3ata.
2rimary 4ata:C
4irst hand data.
9ither !0ensus$ or !Sample$ techni*ue is to be used.
(ens"s: Total no.of items ha#e to be in#estigated.
)ample: Selected representati#es from the total population
ha#e to be in#estigated.
,)
(.?)<) )0M2L.
Total Eo. of items% 1MMR
"ccuracy attained% Time
consuming% 9pensi#e More
Kabour.
Selected Representations%
1MMR "ccuracy cannot be
attained% Kess time ta&en% Kess
epensi#e% Kess Kaborious.
Should be #ery cautious and careful while choosing a
particular method.
Methods of (ollecting 2rimary 4ata:
May be obtained by applying any of the following
methods%
1. 3irect Fersonal .nter#iews.
(. .ndirect 7ral .nter#iews.
). .nformation from 0orrespondents.
,. Mailed *uestionnaire methods%
-. Schedule sent through 9numerators.
1' 4irect 2ersonal Inter1ie!s:
" face to face contact is made with the informants.
.nter#iewer as&s them *uestions regarding the study
and tries to get the desired information.
The information thus collected is frst hand and
original.
Merits:
5i6 Response is encouraging when personally
contacted.
5ii6 .nformation are more accurate 5if not found
accurate% he can be reeamined or 0ross;eamined%
there by try to obtain the information6.
,,
5iii6 Fro#ides hope for getting supplementary
information% which may be of greater use latter.
5i#6 " delicate situation 5Some Fersonnel Guestions6 can
usually be handled more e1ecti#ely by a personnel
inter#iew than other sur#ey techni*ues.
5#6 The inter#iewer can ad+ust the language according
to the status and educational le#el of the person
inter#iewed% thereby can a#oid incon#enience and
misinterpretation.
4emerits:
5i6 9pensi#e @ when the no. of informants is large.
5ii6 <reater chance of personnel bias and pre+udice.
5ta&ing a decision before fnding the full facts6.
5iii6 .nter#iewer should be thoroughly trained B
eperienced. 5untrained personal will spoil the
wor&6
5i#6 Time consuming 5.nter#iewers can be contacted
only at the con#enience of the informants6
(oncl"sion:
3irect personal .nter#iews can be used in .ntensi#e
4ield Sur#ey rather than 9tensi#e 4ield Sur#ey.
The present day of etreme ad#ancement in
communication system% a good number of sur#ey% can
be conducted by Eews papers B tele#ision channels by
replying through e.mails B SMS.
They are less epensi#e B etremely *uic&.
3efects @ Eo Fhone or Tele#ision% delicate B sensiti#e
Guestions cannot be as&ed% #alue answers.
2' Indirect >ral Inter1ie!s:
,-
The in#estigator contact a third party called
2itnesses who is capable of supplying necessary
information.
<enerally adopted when the information to be
obtained is comple or the informer is not willing to
re#eal the answers.
9ample: 2hen a drug addicted person is inter#iewed% he
will not be willing to response directly% so the information
are gathered #ia agents 5may be relati#es6.
The accuracy of this method depends on%
5i6 The pro#en integrity of the "gency'Ferson.
5ii6 "bility of the inter#iewer to act the right information
from the witness.
5iii6 /ribery or other reasons may twist the witness to
gi#e false information there by bringing a wrong
conclusion.
(oncl"sion:
5i6 Ket more care should be ta&en in the selection of
!2itness$ because it is on their #iews% the fnal conclusion is
reached.
3' Information From (orrespondents:
The in#estigator appoints local agents or
correspondents in di1erent places to collect
information under this method.
These correspondents collect the information and
transfer or transmit the information to the central
oNce where the data are processed.
.5ample: Eews paper "gencies.
,>
<enerally these "gencies are paid sta1% sometimes
honorary.
0d1antages:
1. 0heap
(. 0an be used to get regular information at regular
inter#als 5daily% wee&ly or monthly6.
4isad1antages:
5i6 Kac&s "ccuracy.
*' Mailed 6"estionnaire Method:
" list of *uestions pertaining to the sur#ey is &nown as
Guestionnaire. .t is prepared and sent to #arious
informants by post.
The *uestionnaire contains *uestions and pro#ides
space for answers.
" re*uest is made to the informants through a
co#ering letter to fll up the *uestionnaire and sent it
bac& within a specifed time.
The *uestionnaire studied can be classifed as%
16 The degree to which the *uestionnaire is formali8ed or
structure.
(6 The disguise'lac& of disguise of the *uestionnaire.
)6 The communication method used.
,6 2hen no formal *uestionnaire is used% inter#iewers
adopt other tactics li&e showing pictures on which
respondents comments.
2hen a research follows a prescribed se*uence of
Guestions it is !structured study$.
,?
2hen no prescribed se*uence of Guestion eists% the
study is Eon;Structured.
2hen the *uestionnaire is constructed in such a way
that the 7b+ecti#e is clear 5the *uestionnaire6 it is
&nown as Eon;disguised.
2hen the ob+ecti#e is not clear% the *uestionnaire is a
3isguised one.
7n this basis% , types of studies can be distinguished%
5i6 Eon;disguised Structured.
5ii6 Eon;disguised Eon;structured.
5iii6 3isguised Structured.
5i#6 3isguised Eon;Structured.
Merits: Guestionnaire Methods%
516 9asily adopted in large populations and when the
informants are wide spread o#er a large geographical
area.
5(6 Relati#ely cheap B timely.
5)6 .nformation pertaining to personnel life or family or
confdential matters will be re#ealed in written rather
than personal inter#iews or telephones.
4emerits:
516 "pplicable only among literate people.
5(6 Ancertainty of the Respondents lac&s co;
operation.
5)6 Kac&s "ccuracy because the information may not
be correct.
A"ideline to mae this method more .Gecti1e:
5i6 Frepaid postage stamp should be aNed.
5ii6 Sample should be large.
5iii6 Guestionnaire should be interesting.
,I
5i#6 Kegal 0ompulsion should be made to pro#ide in
formations.
-' )ched"les sent thro"gh .n"merators:C
Sending Schedules through enumerators or
inter#iewers.
The enumerators contacts the informants% gets replied
to the *uestions contained in the schedule and fll
them in their own handwriting.
:ere the *uestions are as&ed face to face and the
response is papered.
Merits:
516 "pplicable among illiterates.
5(6 Cery little scope for Eon;response% as the
enumerators go personally.
5)6 .nformation are more Reliable B "ccurate.
4emerits:
516 9pensi#e @ Since enumerators are paid
personals.
5(6 Success @ depends on the eNciency of the
enumerators.
5)6 .nter#iewer re*uires training and eperience.
5,6 Cariations in answers must be remo#ed to a#oid
#ariations.
)econdary 4ata:
"re those data which ha#e already been collected and
analy8ed by some earlier agency for its own use and later
the same data is used by a di1erent agency.
Sources of Secondary 3ata:
516 Fublished Sources.
,J
5(6 Anpublished Sources.
1' 2"blished )o"rces:
The <o#ernment% .nter Eational and local agencies
publish statistical data.
0hief "mong them%
(i) Inter ?ational 2"blications:C
.nter Eational .nstitutions B bodies li&e ..M.H.
5.nternational Monetary 4und6% ../.R.3. 5.nternational /an& of
Reconstruction and 3e#elopment6% ..0.".4.9. 5.nternational
0onference on "griculture B 4ood 9ngineering6 and A.E.7.
Anited Eations 7rgani8ation publish regular B occasional
reports on 9conomics B Statistical matters.
(ii) >Jcial 2"blications of (entral , )tate
Ao1ernments:
Se#eral departments of the 0entral and State
<o#ernments publish reports on di1erent sub+ects.
.5ample:C Fublications are%
5a6 Reser#e /an& of .ndia /ulletin.
5b6 0ensus of .ndia.
5c6Statistical "bstract of the states.
5d6 "gricultural Statistics of .ndia.
5e6 .ndian Trade Hournal.
(iii) )emi >Jcial 2"blications:
5a6 .ndian Statistical .nstitute 5..S..6
5b6 .ndian 0ouncil of "gricultural Research 5..0.".R.6
5c6.ndian "gricultural Statistics Research .nstitute
5..".S.R...6
Fublish the fndings of their research programs.
-M
(i1) 2"blications of 1ario"s commercial and fnancial
instit"tions'
(1) Reports of 1ario"s committees , commissions
appointed by the Ao1ernment'
5a6 Ra+ 0ommittee$s Report on "gricultural Taation.
5b6 2anchoo 0ommittee$s Report on Taation B /lac&
Money.
(1i) Ho"rnals , ?e!s 2apers:C
Fowerful source of secondary data.
0urrent B important matter can be obtained.
4rom Hournals B Eews papers li&e 9conomic Times%
0ommerce 0apital% .ndian 4inance% etc.%
<np"blished )o"rces:
Records maintained by <o#ernment B Fri#ate 7Nces.
Theses of Research Scholars from uni#ersities B
institutions.
2reca"tions in the "se of secondary 4ata:
Froper scrutiny is made before they are used by
in#estigator.
/e 9tra;0autious while using secondary data.
Should not be accepted as such because the
secondary data may process%
5/ias% .nade*uate Si8e% Substitution% errors of defnition
or arithmetical errors6
Factors to be (onsidered before "sing the secondary
4ata:
-1
(i) )"itability of 4ata:
ma&e sure that the data a#ailable is suitable for the
purpose of en*uiry.
(ii) 0de&"acy of 4ata:
Ma&e sure that the data are suNcient or ade*uate for
the present analysis.
(iii) Reliability of 4ata:
The reliability of data is must% without which there is
no meaning in the research.
The reliability of data can be tested by fnding the
agency that has collected the data% if the agency has
used proper methods for collection the data.
.4I%I?A:
7nce data ha#e been obtained from primary or
secondary sources the net step in a statistical in#estigation
is to edit the data. 5to Scrutini8e6.
7b+ecti#e% editing is to detect possible errors and
irregulations.
9diting needs great care and attention.
9diting Secondary data is simple but the data
collected from sur#ey 5Frimary 3ata6 need ecessi#e
editing.
.diting 2rimary 4ata7
5i6 The data should be complete in e#ery respect.
5ii6 The data should be accurate.
5iii6 The data should be consistant.
5i#6 The data should be homogenous
1' .diting For (ompleteness:
-(
The editor should see that each schedule or
*uestionnaire is complete in all respects.
"nswers to e#ery *uestions is furnished.
.f not answered% try to meet them in person to get the
answers.
.f not +ust mar& Eo "nswer.
2' .diting For 0cc"racy:
The reliability of conclusions depends on the
correctness of in formations.
.f the information is wrong% the conclusion can ne#er
be #alid.
9ditor should see that the in formations are accurate
in all respects.
"rithmetic errors can be detected easily B corrected.
.f the error is due to fault information supplied% it may
be diNcult to #erify.
3' .diting For (onsistency:
9ditor should see that the answer to *uestions are not
contradictory in nature.
.5ample: "re you a student= Eo which class do you study=
T
The answers are contradictory and such answers
should be classifed.
*' .diting For =omogeneity:
Anderstand the *uestions in the same sense.
0hec& uniform interpretation and ma&e sure the
information supplied by the #arious informants are
homogenous B uniform.
.5ample: .ncome 5Uearly% Monthly% 2ee&ly% 3aily6
-)
(hoice 3et!een 2rimary , )econdary 4ata:
" proper choice between the type of data 5Frimary or
Secondary6 needed for a particular statistical in#estigations
is to be made by considering the nature% ob+ecti#e% scope of
the study% time frame B fnances and the degree of
precision aimed at% and the status of the "gency.
Eow% Secondary data are generally used from fairly
reliable published data by <o#ernment% Fri#ate
organi8ations and research agencies% periodicals
maga8ines etc.%
.n fact% primary data are collected only if there do not
eist any secondary data.
.n some cases both Frimary B Secondary data may be
used.
6<.)%I>??0IR.
Guestionnaire can be defned as a group of *uestions
designed to collect information from a specifc sub+ect.
5or6
Kist of *uestions sent to a number of persons for
getting answers and which obtains standardi8ed results that
can be tabulated and treated statistically.
5or6
Media of communication between the in#estigator and
the respondent.
-,
<enerally used in social research when the population
is #aried% large% di#erse B Scattered.
Should be designed with utmost care B caution so that
all the needed information are collected without any
diNculty.
3rafting a <ood *uestionnaire Re*uires @ 0are% S&ill%
2isdom% eNciency and eperience.
2oints to Remember: Khile 4rafting 0 6"estionnaire
(1) )iFe of the 6"estionnaire:
5a6 Eo. of *uestions should be as small as possible
depending on the nature% ob+ecti#es B scope of the study.
5b6 Karge no.of *uestions may irritate the informants and
may be diNcult to 9dit or Scrutini8e by the in#estigator.
5c6 "#oid irrele#ant and unimportant *uestions.
5d6 "#erage Eo. of *uestions should be 1- to (- 5at the
most6.
5e6 .f it is more than (-% di#ide it into #arious sections.
2' %he 6"estions sho"ld be (lear:
Should be 9asy% /rief% Anambiguous50lear in
meaning6% Eon o1ending% courteous in tone%
corroborati#e 5Supporti#e6 in nature B to the point.
3' %he 6"estions sho"ld be arranged in a Logical
)e&"ence:
2hen arranged Kogically% the answers can be easily
tabulated or coded @ and does not lea#e any chance of
omissions.
--
.5ample: To fnd if a person owns a tele#ision.
*' 6"estions sho"ld be simple to "nderstand:
Cague B 3ouble meaning words should be a#oided.
.5ample: Frice'0ost'Rate' 0apital .ncome'Salary.
-' 6"estions sho"ld be (omprehensi1e(Incl"des
.1erything) and .asily 0ns!erable:
Guestions should be comprehensi#e 5i.e.6 it should
include all the needed in formations.
9asy to be answered. "#oid mathematical calculations
li&e Ratios% Fercentages etc.%
/' 6"estions of 2ersonal 0nd )ensiti1e ?at"re )ho"ld
?ot 3e 0sed:
"#oid personal *uestions which the respondent may
feel shy or irritated to answer.
.5ample: 3o do drin& =
.f such *uestions are una#oidable% a highest amount of
politeness should be used.
9' %ypes of 6"estionnaire:
(a) )h"t 6"estions:
2here possible answers are suggested by the frames
B the respondents are re*uested to the tic& one of
them.
Two types of shut Guestions.
(i) )imple 0lternati1e 6"estions:
57therwise &nown as 3ichotomous *uestions6 0hoose
from two clear cut alternati#es Ues or Eo' Right or 2rong.
(ii) M"ltiple (hoice 6"estions:
->
2hen it is diNcult to defne a clear cut alternati#e%
additional Guestions between Ues B Eo is inserted%
.5ample: To fnd if a person smo&es= 3o you smo&e=
5a6 Ues% Regularly 5 6
5b6 Eo% Ee#er 5 6
5c6 7ccasionally 5 6
5d6 Seldom 5rarely6 5 6
9asy B 0on#enient to answer.
Sa#e Time.
9asy to tabulate.
:' Leading 6"estions sho"ld be 01oided:
Kong Guestions leading to se#eral answers should be
a#oided. .t should be framed into short *uestions.
.5ample: 2hy do you use a particular type of car% say
Maruti 0ar% "#oid this continuous *uestions.
.nstead as&%
2hich car do you use =
2hy do you prefer it =
;' (ross (hecs:
Should be designed to pro#ide internal chec&s on the
accuracy of the in formations gi#en by the
respondents.
18' 2reC%esting the 6"estionnaire:
Try out the Guestionnaire on a small scale before using
in a large scale.
-?
The drawbac&s% short comings and problems faced in
the small scale informants can be impro#ed or
modifed when used in large scale.
11' 0 (o1ering Letter:
" 0o#ering Ketter should be enclosed for the purpose
regarding defnitions% concepts B purpose.
"ttach a self addressed en#elope in case of mailed
*uestionnaire.
Mentions about "wards or incenti#es for *uic& reply.
Fromise to send a sur#ey copy of the report.
;;;;;;;;;;;;;;
)0M2LI?A
" carefully designed sample may actually be better
than a poorly planned and eecuted census.
Merits:
1' It sa1es time:
Sa#es time because fewer items are collected and
processed.
2' It Red"ces (ost:
Since only few items are studied% there is reduction in
cost B reduction in man power.
3' More Reliable Res"lts can be >btained:
Sampling is more Reliable because
5i6 fewer chance of sampling error.
-I
5ii6 9perience% Trained B Technical people can be employed
to process B analy8e the data.
*' It pro1ides more detailed in formations:C
More detailed information can be obtained by sample
sur#ey.
-' >nly )ampling Method to depend "pon:
2hen the population is large and fnite% the only
method applicable is sampling.
/' 0dministration (on1enience:
The organi8ation and administration is easy in sample
sur#ey.
9' More )cientifc:
Results can be tested since more scientifc.
)hortcomings (or) 4emerits
1' Ill"sory (False) (oncl"sion:
.f sampling is not carefully planned B eecuted% the
conclusions may be false.
2' )ample ?ot Representati1e:
.f the sample ta&en from the population is not the right
representati#e% the result may be false or misleading.
3' Lac of .5perts:
-J
.f there is a lac& of eperts to plan% eecute and
analy8e the samples% the result would be unsatisfactory.
*' 2ersonal 3ias:
There may be personal bias B pre+udice in choosing
the sampling techni*ues.
-' )iFe of )ample:
.f the si8e of the sample is not appropriate% it leads to
untrue conclusions.
.ssentials of )ampling:
1' It m"st be the right representati1e:
The Sample selected should process the similar
characteristics of the original uni#erse.
2' =omogeneity:
Selected samples should be homogenous with the
samples B the uni#erse.
3' 0de&"ate )amples:
" good number of items should be included for the
study.
*' >ptimiFation:
" proper si8e of sample should be maintained to ha#e
optimi8ed results in terms of cost B eNciency.
)tatistical La!s:
>M
Ind"ction:
The logical process of drawing a general conclusion
from the study of representati#e items is called .nduction.
Sampling is based on two fundamental principles of
Statistics theory
5i6 Kaw of Statistical Regularity
5ii6 The Kaw of .nertia of Karge Eumbers.
%he La! of )tatistical:
Reg"larity: 5Mathematical Theory of Frobability6% States%
" moderately large number of items chosen at
random from a #ery large group are almost sure to ha#e the
characteristics of the large group.
2roof:
1. "#erage income of 1%MMM people is to be found out%
(. 2e ta&e a sample of 1MM people B fnd the a#erage.
). "nother person ta&es 1MM people B fnd the a#erage.
The "#erage income found by both the persons will
ha#e least di1erence.
.f the a#erage income of the same 1MMM people is
found out by census% the result will be more or less the
same.
2' La! of Inertia of Large ?"mbers:
.t States% 7ther Asing being e*ual% as the sample si8e
increases% the results tend to be more accurate B reliable.
)ampling .rror:
>1
The de#iations or di1erence between the actual
population and the sample representati#e.
Reasons For )ampling .rror:
1. 4aulty Selection of Sample
(. Substitution
3' Fa"lty 4emarcation (4emographic limit) of
sampling <nit:
.n case of "rea sampling the sampling units at the
borders should be decided if to ta&e it or re+ect it.
*' Fa"lty estimation techni&"es:
2rong selection of sampling techni*ues.
%ypes of )ampling .rrors:
(i) 3ias .rror:
0aused due to bias or pre+udice on the part of the
informant.
(ii) <nbias .rror:
9rror caused due to the Eormal 0ourse of
in#estigation.
Red"cing )ampling .rror:
Sampling 9rror can be reduced by increasing the si8e
of the sample.
>(
;;;;;;;;
.L2.RIM.?%)
9periment is the process of eamining the truth of a
statistical :ypothesis related to some research problem.
9periments are of two types%
1. "bsolute 9periment.
(. 0omparati#e 9periment.
"bsolute 9periment:
2hen a researcher wants to determine the impact of a
fertili8er on the yield of a crop% it is a case of "bsolute
9periment.
0omparati#e 9periment:
2hen a researcher wants to determine the impact of
one fertili8er as compared to the impact of some other
fertili8er% it will be called as 0omparati#e 9periment.
Research 4esign are of three types7
1. Research design in case of descripti#e B diagnostic
studies.
(. Research design in case of eploratory Research studies.
). Research design in case of :ypothesis Testing Research
Studies.
Research 4esign In case of =ypothesis %esting
Research )t"dies:
>)
:ypothesis testing research studies are generally
&nown as 9perimental studies.
The researcher test the casual relationship between
the #ariables.
Frofessor 4isher is considered as the pioneer of this
type of studies. 59perimental Studies6
:e performed this study when he was wor&ing at a
"gricultural Research Station in Kondon.
:is found out that% by di#iding plots into di1erent
bloc&s and then by conducting eperiments in each of
these bloc&s% whate#er in formations is collected and
inference drawn from them can be more reliable
Frofessor 4isher laid three principles of 9perimental
3esigns%
1. The Frincipal of Replication
(. The Frincipal of Randomi8ation.
). The Frincipal of Kocal 0ontrol.
%he 2rincipal of Replication :
The 9periment should be repeated more than once.
The treatment is applied to many eperimental units.
The information collected and the inference drawn
from these eperimental units will be more reliable
and statistically accruable.
"im: To eamine the e1ect of two #arieties of paddy.
.5ample: " paddy feld is di#ided into ( parts. <row one
#ariety in one part and the other #ariety in the other. Then
we compare the yield of the two parts.
3raw conclusion on that basis.
>,
?o 2rinciple of Replication is
4ield
Fart . Fart .
Treatment 0ompare the yield of
the two parts.
7ne #ariety "nother
of Faddy Cariety of Faddy.
Khen 2rinciple of Replication is "sed:
4irst di#ide the feld into se#eral parts.
<row one #ariety in half of the parts and the other
#ariety in the remaining parts.
0ollect the information of the two #arieties and draw
the conclusion by comparing both.
The Results so obtained will be more reliable and
accurate compared to the results drawn without using the
principle of Replication.
The 9periment can be repeated se#eral times.
4ield
Treated

0onclusion drawn.
Farts
2' %he 2rinciple of RandomiFation:
rr rr R r r
rr rr R r r
rr rr R r r
rr rr R r r
rr rr R r r
rrrrr
rrrrr
rr
rrrrr
r
rrrrr
rrrrr
rr
rrrrr
r
>-
Frinciple of Randomi8ation Fro#ides us a protection
against the e1ects of 9traneous Cariables.
The #ariations or e1ects caused by these etraneous
#ariables can be combined under the heading
0hance.
.5ample: 2hen the researcher grows one #ariety of paddy
in the frst half of the feld and the other #ariety in the net
half of the feld% there may be a possibility or chance that
the soil fertility of the frst half of the feld may be di1erent
in comparison to the net half.
.n this case% he may go on to culti#ate the two
#arieties of paddy in di1erent parts of the feld on the
basis of some random sampling techni*ue.
5i.e.6 :e may apply Randomi8ation principle and protect
himself from the e1ects of the 9traneous 4actors.
/y using Randomi8ation Frinciple a better estimate
can be drawn.


rr rr r r r
rr rr r r r
rr rr r r r
rr rr r r r
rr rr r r r
>>
2hen treated
0onclusion
drawn is more
accurate.
50an protect 91ects of 9traneous
Cariables6
3' %he 2rinciple of Local (ontrol:
The etraneous #ariable which is a &nown source of
#ariability can be made to #ary etensi#ely or
deliberately o#er a wide range.
Eow the #ariability it causes can be measured and
eliminated.
.n short% through the principle of Kocal 0ontrol% we can
eliminate the #ariability due to etraneous factors
from the eperimental error.
The etraneous #ariable is brought to a control.
Minds of .5perimental 4esign:
9perimental 3esign refers to the framewor& of the
structure of an eperiment.
>?
0lassifed into ( /road 0ategories%
1. .nformal 9perimental designs.
(. 4ormal 9perimental designs%
Informal .5perimental 4esigns:
3esigned based only on the di1erence between the
magnitudes or performance.
Three Types%
1. /efore and after without control design.
(. "fter 7nly with 0ontrol design.
). /efore and after with control design.
3efore and after !itho"t control design:
0onsider a test group%
)tep'1: The dependent #ariable is measured before
introduction of the treatment.
)tep'2: The treatment is introduced.
)tep'3: The dependent #ariable is measured after the
treatment has been introduced.
)tep'*: .nference:
%he eGect of the %reatment : The le#el o the
phenomenon after the treatment.
The le#el of the phenomenon before the treatment.
Test "rea Ke#el of Fhenomenon Treatment Ke#el
of Fhenomenon
>I
.ntroduced
/efore Treatment 5T6 "fter Treatment
5U6
V 91ect of the Treatment O 5U6 @ 5T6
4emerit:
2ith the passage of time% se#eral etraneous #ariable
may be there in the treatment e1ect.
(2) 0fter only !ith control 4esign:
)tep'1: Two "reas are selected% the control "rea B the test
area.
)tep'2: The treatment is introduced in the test area alone.
)tep'3: The dependent #ariable in both the areas are
measured% at the same time.
)tep'*: Treatment 91ect is calculated by subtracting the
#alue of the dependent #ariable in the control area from its
#alue in the test area.
>J
%reatment: 1
.Gect O Calue of dependent #ariable in the
control "rea @ Calue of 3ependent Cariable in the test area.
Test "rea Treatment .ntroduced Ke#el of
Fhenomenon 5U6
0ontrol "rea Eo Ke#el of
Fhenomenon 5W6
Treatment
(3) 3efore 0nd 0fter !ith (ontrol 4esign:
)tep'1: .n this design% two areas are selected and the
dependent #ariables in both the areas are measured for an
identical time period before treatment.
)tep'2: Treatment is introduced only in the test area.
)tep 3 : The dependent #aluable is measured on both the
areas 5control
area B test area6 for an identical time period.
)etp* : The e1ect of the treatment is determined by
subtracting the change in the dependent #aluable in the
control area with the charge in the dependent #aluable of
the test area.
?M
%reatment :C
91ect O 55U6 @ 5T66 @ 55W6;5"66
Merits: "#oids 9traneous #ariables resulting from passage of
time and non
comparability of control and test areas.
II Formal .5perimental 4esign:
?1
71er relati#ely more control and use specifc statistical
procedures for analysis types%
16 0omplete Randomi8ed design 5<enerally called 0.R.
3esign6
(6 Randomi8ed /loc& 3esign 5R./. 3esign6
)6 Katin S*uare 3esign 5K.S. 3esign6
,6 4actorial 3esigns .
516 0ompletely Randomi8ed 3esign :;
.n#ol#es two principals% the principle of replication and the
principle of Randomi8ed of the eperimental designs.
The items are randomly assigned do eperimental
treatments.
This design is simpler and easier.
9ample: .t the research has ( items of (M parts and if he
wishes to test to under treatment /% this completely
randomi8ed design gi#es e#ery possible group of 1M items
selected from a set of (M% an e*ual chance of being
assigned to treatment " B treatment /.
7ne way analysis of #ariance 5one way "E7C"6 is used to
analy8e such a design.
?(
2' RandomiFes 3loc 4esign:C
The sub+ects or items are frst di#ided into groups% &nown as
N3locsB
See that% the items in each group or blac& is homogenous.
Randomly select items from each gi#en bloc& and assign
treatment.
9traneous #ariables can be fed and can be measured.
The main feature of this study is% each treatment appears
the same no of times in each bloc&.
This design is analy8ed two way analyses of #ariance 5two
way "E7C"6 techni*ue.
3' Latin )&"are 4esign:C
Ased in "gricultural Research.
K.S 3esign is used when two or more etraneous
#ariables is found.
.5ample: .Gect of fertiliFer on the feld of !heat is do be
determined'
:ere along with the e1ect of fertili8er% the fertility of the
soil must be considered.
.f the facility of the soil is not considered along with the
fertili8er the result obtained may be dependable.
?)
Similarly the impact of the #arious seeds used many also
#ary the yield.
To o#er come this diNculty K.S design is used.
9ach fertili8er 5T1% T(% T)% T,% T-6 will appear - items
but will be used only once in each row and in each
coloumn.
.5ample: Eo treatment occurs more than once.
. .. ... .C C
T1 " / 0 3 9
T( / 0 3 9 "
T) 0 3 9 " /
T, 3 9 " / 0
T- 9 " / 0 3
(oncl"sion :
The feld is di#ided into se#eral bloc&s 5.% ..% ...% .C B C6 and
there are #ariety of fertili8er 5T1% T(% T)% T,% T-6.
/ut each fertili8er is used in each bloc& only once.
a two way 0?>+0 techni*ue.
*' Factorial 4esign :
"re used in eperiments where the e1ect of the depended
#ariable% when a1ected by more than one #ariable is to be
determined.
Ased in social B economic studies where usually large no of
factors a1ect a particular problem.
?,
4actorial design are of two types:
. . Simple 4actorial design
... 0omple 4actorial designs.
)imple Factorial 4esign :
2hen the e1ect of the dependent #ariable is a1ect by only
two factors% it is simple factorial designs.
7therwise &nown as T27 factors 4actorial 3esign.
(omple5 Factorial 4esign :
This design is used when more than two factors at a time
a1ects the dependent #ariable.
7r the design considers three or more independent #ariable.
The greater the no of independent #ariable% the higher the
order of interaction% analysis possible.
0d1antage :
"ccuracy
0an determine the e1ects of more #ariable in a single
eperiment.
?-
>3).R+0%I>?
7bser#ation is defned as a planned method of !atching
that in#ol#es constraints 5steps6 to impro#e accuracy.
(haracteristics of >bser1ation :
i. 7bser#ation are direct
ii. ta&es place in natural situations
iii. Kess Structured.
i#. Ma&es only *uantitati#e study.
0pplicable in :
1. Kife styles
(. 9ncounters ' Settlement.
). Relationships
,. <roups ' 7rgani8ation.
0cc' to 3loc , (amion :
1. 7bser#ed in natural surroundings
(. Anderstands e#ents a1ecting social relations.
). .dentifes regulati#es in social life.
,. :ypothesis free en*uiry.
?>
-. "#oids manipulations of independent #ariable.
>. Recording is not selecti#e.
4iGerentiate .5perimental %echni&"e , >bser1ation
%echni&"e :
.5perimental %echni&"e >bser1ation %echni&"e
Eo such controls
Eot "lways
0onducted in smaller units
Eo so. Study is towards
sharpening the +udgment.
Eo so.
4ew 0ontrols.
7bser#ation is natural
conducted in large unit. 4ewer
sub+ects are watched for long
period .Study is directed
towards sensiti8ing the
obser#er ' e#ents.
/eha#ior obser#ed is more
di1erent.
)%0%I)%I(0L 0?0LD)I)
??
2robability: .S a measure of the epectation that an certain
e#ent will occur.
Frobability #alue ranges from M;1.
%he Main terms pertaining to probability theory:C
1'Random .5periment: "n eperiment which can be repeated
under the same conditions and the outcome cannot be prod iced
is &nown as Random 9periment.
.5ample:C 2hen a coin is tossed7 we cannot predict whether
head or tail is going to occur.
2')ample )pace:C " set of possible outcomes of a random
eperiment is &now as sample space.
.5ample: 2hen a coin is tossed twice the possible outcomes are
::% :T% T: B TT% .T is represented%
) O (==7 =%7 %=7 %%)
3'0n .1ent: "ny possible outcome of an eperiment is &now as
an e#ent.
9ample: 2hen a coin is tossed twice% :: is an e#ent.
"n e#ent can be classifed as%
a' )imple .1ent: 2hich has only one sample point
.5ample: ::% TT.
b' (ompo"nd .1ent: 2hich has more than one sample point.
?I
.5ample : T:.
d' (omplimentary .1ent : " and "$ are complementary e#ents%
if "$ consists of all the sample point which is not included in "$
than the sum of the probability of the sample space is e*ual to
1.:ence F5"$6;F5"$6 F5"$6XF5"$6 O 1 F5"$6 O 1 @ F5"$6.
9ample: 2hen a dice is thrown% the probability of odd Eumbers
turn up are complimentary to e#en number turn up.
" O 1%)%-
"1 O (%,%>
e' M"t"ally .5cl"si1e .1ent: 0 and / two mutually eclusi#e
e#ent% if " pre#ents the occurrence of /.
9ample: 2hen a coin is tossed once the occurrence of :ead
pre#ents the occurrence of Tail.
.f " and / are mutually eclusi#e e#ents than the probability of
occurrence of " or / is the sum of their indi#iduals probability.
F5"A/6 O F5"6 X F5/6.
.f " and / is +oint sets% than the addition theorem of probability
can be stated as.
F5"A/6 O F5"6 X F5/6 @ F5"/6
?J
f' Independent .1ent: " and / are in dependant e#ent% if the
occurrence of " does not inDuence the occurrence of / these two
e#ents are called independents e#ents.
.5ample: 2hen a coin is tossed twice the occurrence of head in
the frst toss dose not inDuence the toss in the second coin.
Froduct of " and / is the product of their indi#iduals probability
F5"/6 O F5"6 T F5/6.
Frobability 3istribution: .4 !T$ is total no of success discrete
Random #ariables which ta&es the #alue of T1%T(% T)........... Tn%
F1%F(%Fn then follows the probability distribution.
2roperties of probability 4istrib"tion I 2 Main properties:
1. F5T16 is greater than or e*ual Wero and less than or e*ual '
one 59pressed in fractions6.
(. The sum of all be probability distribution will be e*ual to
one.
.5ample: 2hen a coin is tossed Twice the probability
distribution is% T5 probability of obtaining hed6 O M%1%( F5T16 O Y%
Z% Y O Y.
.5pectation of 2robability: Ket !T$ be the discrete Random
Cariable% which ta&es the #alue T1% T(% T)..........Tn
Respecti#e probability is F1% F(% F) ............ Fn
IM
9pectation of Frobability distribution: F1T1 X F(T(............FnTn.
Frobability 3istribution O 5M 1',6 X 51 1'(6 X 5( Y6 O 1.
/inomial 3istribution: " /inomial eperiment is a statistical
eperiment. .t has the following properties%
5i6 The eperiment consists of PnQ repeated trials.
5ii6 9ach trail can result in +ust two possible outcomes. 2e
call one of this outcomes as N)"ccessB and the other
NFail"reB .
5iii6 The probability of success is denoted by P2Q% in e#ery trail
and the probability of failure is denoted by l @ p or !*$.
5i#6 The trials are independent.
5#6 The outcome on one trail dose not a1ect the outcome of
other trials.
?otations: T @ The number of successes that result from the
/inomial 9periment.
n ; The no of trail in the binomial eperiment.
F @ The probability of success on an indi#idual trial.
0 ; Eo of coeNcient .
/inomial 4ormula : Suppose a /inomial eperiment consists of !n$
;;;;;;; and result in !T$ successes B if the probability of success on
an indi#iduals trial is F% then the /inomial probability is 3 (57 n7
p) O n(5 L 25 L (1 $ 2)
,
I1
0oeNcient of /inomial 3istribution: The binomial coeNcients are
the coeNcients in the epansion of two terms 5 X 16n.
The coeNcient of r in 5X16n is denoted by n0r or 5n'r6.
9ample: The coeNcient of 5X16( O 5X16 5X16
O 1( X XX1
O 1( X ( X1
The 0onfdents are O 1X(X1
The coeNcient of 5X16) O 5X16 5X16 5X16
O 5(X(X16 5X16
O ) X ( X(( X( XX1
O ) X )( X) X1
The 0oeNcient is O 1. ) .) 1
The Fascal triangle lists out all the /inomial 0oeNcient.
1
1 1
1 ( 1 5X16(
1 ) ) 1 5X16)
1 , > , 1 5X16,
1 - 1M 1M - 1 5X16 -
1 > 1- (M 1- > 1 5X16>
I(
1 ? (1 )- )- (1 ? 1 5X16?
9ample: " coin is tossed 4our Times what is the probability of
obtaining two or more heads= .
Step @. : 2hen a coin is tossed 7ne time% the probability of !:ead$
of tail is e*ual%
5ie6 F O * O1'(.
Step @ .. : The #arious possibilities of :ead and tail e#ents will be%
5pX*6, O 1p, X ,p)* X>p)* X >p(*( X ,p*) X1*,
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
1p, O 51'(6 where F O 1'(6 5* O 1'(.6
O Z Z Z 1'(
O 1'1>
,p)* O ,51'(6) Z O , Z Z Z Z
>p(*( O > 51'( Z Z 1'( O )'I
Therefore% the probability of obtaining ( or more heads is%
)'I X Y X 1'1> O 11'1>
(. F7SS.7E 3.STR./AT.7E:
2hen !F$ is #ery small 5Success rate is etremely small6 and
!n$ is #ery large 5total no of trail items of e#ents is large6 then
F7.SS7E distribution is used.
F @ 7 5Successes approach 8ero6 5np O m is fnite6
I)
The eperimental result in outcomes that can be classifed
as Successes or failure6.
T @ The "ctual Eumber of successes that occurs in a specifc
region.
F 5% u6 O 5eu6 5u6
;;;;;;;;;;;;;;;;;;;;;;;;
1
This is Foisson formula here we conduct a Foisson
eperiment in which the a#erage number of successes within a
gi#en region is !A$. Then the Foisson probability is as abo#e .
2here is the "ctual number of successes.
!e$ is e*ual to (.?1I(I.
!u$ is the mean of the distribution.
The a#erage number of successes 5u6 that occurs in a
specifed region is &nown.
The probability of success is proportional to the si8e of the
region.
I,
The probability of successes occurring in an etremely
small region #irtually 8ero.
?otations:
!e$ O " constant e*ual to approimately (.?1I(I.
u O The mean 5a#erage6 no of success that occurs in a specifc
region.
9ample: The a#erage no of homes sold by " really company is (
:omes per day. 2hat is the probability that eactly ) homes will
be sold tomorrow=.
u O ( 5"#erage of ( homes sold per day6
O ) 5) homes will be sold tomorrow6
e O (.?1I(I 50onstant6
2oisson Form"la:
F 5% u6 O 5eu6 5u6
;;;;;;;;;;;;;;;;;;;;;;;;

F5)% (6 O 5(.?(I(I[6 5(\6
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
1 ( )
O 5M.1)-),6 5I6'>
O M.1IM
I-
?ormal 4istrib"tion and its properties:
Eormal distribution refers do a family of continuous
probability distribution described by the normal e*uation.
?ormal .&"ation:
The #alue of the random #ariable !U$ is ; 5 @u6 ['([
1
U O ] 0 50oeNcient6
^(
2here ! T$ is a normal Random #ariable.
!u$ is the mean
! $ is the standard de#iation
!_ ! is constant 5).1,1-J6
!e$ is constant 5(.?1I(I6
)tandard 4e1iation:
Represented by ! $ it means how much #ariance
5di1erence6 or dispersion eist from the a#erage.
I>
9ample: O Cariance 5Root of its #ariance let us consider a
population has eight #alues% (%,%,%,%-%-%?%J.
Mean: (X,X,X,X-X-X?XJ
;;;;;;;;;;;;;;;;;;;;;;;;;
I
O ,M'I O - 16 di1erence between the mean B the mean B the
#aliance #alue is ta&en.
Standard 3e#iation (- s*uare the #alue.
5( ; -6 [ O 5;)6 [ O J 5 , @ -6 [ O 5;16 [ O1 5- @ -6 [ O M[ O M
5, @ -6 [ O ;1[ O 1 5 , @ -6 [ O 5;16 [ O1 5? @ -6 [ O ([ O ,
5- @ -6 [ O 5M6 [ O M
Rate of ^ JX1X1X1XMXMX,X1> )I
;;;;;;;;;;;;;;;;;;;;;;;;;;;; O ] O ,';
I I
.5ample: "n a#erage light bulb manufactured by "cme
0orporation lasts )MMdays with a standard de#iation of -M days.
"ssuming that the bull life is normally distributed what is the
probability that an "cme light bulb will last at most )>- days=
"nswer: The #alue of the normal Random #aluable 5T6 O )>-
days.
The mean 5u6 O )MM days standard de#iation O -M days.
Eormal 3istribution U O 1 e5u6 [
I?
^(
1
O ]
-M^( ).1,1-J (.?1I(I @ 5)>- @ )MM6 ['( -M

1
O ]
-M >M1) (.?1I(I'1MM O 1')M>.- (?1 O II,''
U O .II or IIR
:ence there is an IIR chance that the bulb will last most
)>- days.


II