Professional Documents
Culture Documents
Evolution of Statistics in India
Evolution of Statistics in India
International Statistical Institute (ISI) is collaborating with JSTOR to digitize, preserve and extend access to
International Statistical Review / Revue Internationale de Statistique.
http://www.jstor.org
Evolution
of
Statistics
in
India
Summary
This is a briefhistoryof the evolutionof officialand academicStatisticsin Indiawhichfocusesmainly
on the period1930to 1960buttracesits originsin antiquityand recenthistory.Wealso commenton how
Statisticshas continuedto evolvesincethe 1960's.This is a historyof both institutionsand people,who
builtand shapedthem,and of ideas.
Statisticsas a key technology;Five YearPlans;Data
Key words:StatisticalSystemin India;Mahalanobis;
CentralStatistical
NationalSampleSurvey;StatisticsinIndianresearch
institutions
Requirements;
Organisation;
anduniversities.
1 Introduction
While statistics have been collected and used in the Indian subcontinentfrom antiquity,major
changes in collection and use took place during the British period (1757-1947) in Indian history.
Some of this change was due to new imperialneeds, but much of it occurredindirectlyas a result
of westerneducationand a spiritof scientificcuriosityand experimentation.Interestin rapidsocial,
economic and technological development added a new dimension after India's independencein
1947. Half a centuryafter that momentousevent seems a good time to take stock of how Statistics
has developed in India. The following account is meant to be a brief history ratherthan a current
assessment. To us the most importantperiod after independenceis the decade 1950 to 1960 when
so many things were happeningat the same time. Our account begins in antiquity,focuses on the
period 1930 to 1960 and ends with a brief sequel.
The architect of modern statistical methods in the Indian subcontinentwas undoubtedlyP.C.
Mahalanobis,but he was helped by a galaxy of very distinguishedscientiststhatincludedC.R. Rao,
R.C. Bose, S.N. Roy, S.S. Bose, K.R. Nair,D.B. Lahiriandmanyothers.Therewere also otherslike
P.V.Sukhatme,and V.G. Panse who workedindependentlyof Mahalanobis.Ourhistoryis a history
of some of these persons as well as a history of institutionsand interactionsbetween persons and
institutions.
In a concluding section, we try to assess what was unique aboutthe growthof Statistics in India
in the earlierpart of the centuryand what may have been the historicalreasons for what C.R. Rao
has called a golden periodfor Statisticsin India.
2
Historical Background
Early Origins
It is interesting and illuminatingto note that statisticalknowledge and probabilisticideas were
attributedto the kings and rulersmentionedin the greatIndianepic, the Mahabharata,as is evident
from the following quote of Ian Hackingfrom the Historyand PhilosophyScience Seminar(quoted
J.K. GHOSH,
P. MAITI,T.J.RAO& B.K. SINHA
14
15
belonged to the emperorand the cultivatorwas a tenantwith full liberty of exploiting his piece of
land. A proportionof produce, fixed from time to time, would have to be paid to the state as land
revenue.However,with the decline and fall of the Moghul empireshortlyafterAurangzeb,many of
the officials called Jagirdars,Inamdarsetc. appointedby the emperorbecame independentNawabs,
and Kings with no or nominalallegianceto a centralauthority.
3
16
first series of statistical papers on India. Census reportsof 1 January,1855 and 10 January,1868
were published.Earliercensuses of Calcuttatakenin the year 1822 andin 1847 by W.H.Careygave
a total count of 179,917 and 4,00000 respectively!(see Chaudhuri(1964)).
The censuses taken during 1769-1855 by the EIC or those taken by the Crown during 18581869 were fragmentary,hardly systematic and lacked any uniformity.The first systematic attempt
to ascertainthe whole populationof Indiaby 'actuallycountingheads' was made between 1867 and
1872. It was not a synchronouscensus for the whole country,nor was it complete.The operationof
a decennial census for the whole countrystartedin 1881 and is continuingever since. The reporton
the Census of BritishIndiatakenin 1881 was publishedin three volumes.
Kingsley Davis (1951) remarksthat 'the Indiancensuses are remarkablenot only for the information they reveal but for the special obstacles they had to overcome.... they (the census officials
in India) have enormouslyenrichedour knowledge of India in nearly every branchof scholarship,
from anthropologyand sociology to geographyand religion'.
Even thoughW. Hamiltonpublishedthe firstgazetterin 1815, he revisedit andpublishedEastIndia
Gazetterin 1828. However,Thornton'sgazetterin four volumes publishedin 1854 was considered
to be complete at thattime.
Impressedby the trendin statisticalactivities,the Secretaryof stateorderedthe Governor-General
in Council to preparea 'comprehensiveand coordinatedscheme of statistical survey' for each of
the twelve greatprovincesof the then BritishIndiaand Dr. W.W.Hunterwas appointedas DirectorGeneralof Statistics in Indiain 1869 to carryout this work.
In 1870, Huntergave a plan for an ImperialGazetterof India.The local governmentshadplanned
differentlyfor conductingthis work with bigger budgetsand manpowerwhile severalpublic bodies
such as the Asiatic Society insisted on a systematic and coordinatedeffort so that the work was
executed under a uniformplan. It was agreed upon to provide a data base collected by each local
governmentas a common basis for comparisonof statistics of the country and to suggest quick
compilation methods from the data collected on a uniform plan. Thus the Statistical Account of
Bengal (the presentBangladesh,West Bengal, Biharand Orissa)was publishedin 20 volumes under
Hunter'ssupervision.
For each district there were details on topographicaldata, ethnic divisions and creeds, agricultural situation, commerce, working of district administrationand finally the sanitaryand medical
aspects and such meteorologicaldata as could be procured.Statisticalaccountsfor the provincesof
Assam, N.W. Provinces, Punjaband othersfollowed. Thus about 100 printedvolumes aggregating
to 36,000 pages covering 240 districtscomprising 15 British Indianprovinces were publishedand
later condensed in the ImperialGazetterof Indiawhich was releasedin 1881 in nine volumes.
S.B. Chaudhuri(1964) in his comprehensivework 'Historyof Gazettersof India' commentsthus:
"No comparablearea of the world has anything like this prodigious compilation of
statisticaldata and demographicand historicalmaterialas a countrywhich is almost a
continentin the immensityand diversityof its character."
The need for timely and accuratecollection of agriculturaldata was felt by the Indian Famine
Commission and agriculturaldepartmentswere organisedin variousprovinceswhich resultedin the
publicationof 'AgriculturalStatistics of British India' in 1886. To scrutinizeand summarizethese
data collected by the agriculturaldepartments,a statisticalBureauwas formedat the centrein 1895
to coordinate the agricultural,foreign trade, prices, wages and industrialstatistics. The Director
General(DG) of Statisticswas in chargeof this operation.
17
18
the Departmentof Statistics, including the CSO and NSSO, has been transferredto the Ministry
of Planningand ProgrammeImplementationfrom February,1973. B. Ramamurti,S. Subramanian,
P.C.Mathew,and K.R. Nair headed the CSO in the early years.
National Sample Survey
The StandingCommitteeof the DepartmentalStatisticiansas well as the National Income Committee (NIC), established in 1949, felt that there was an urgentneed for improvingthe quality of
statisticalinformation.In 1950, the NIC recommendedthe use of samplingmethodsto fill the gaps
in the estimationof nationalincome.
The Indian StatisticalInstitute(ISI) which had acquiredexperienceand expertise in large scale
surveys since 1935 was approachedby the Governmentof Indiato preparethe design with detailed
plans and estimatesfor a comprehensivesocio-economicnationalsamplesurveycoveringruralareas
of India. The first round of data collection startedin October 1950 and was completed by March
1951 with a sanctionedstrengthof 607 personnel.The Directorateof the National Sample Survey
was transferredfrom the Ministry of Finance to the Cabinet Secretariatin 1957. As mentioned
earlier, it has been in the Departmentof Statistics since 1961. While the field work was done by
the Directorate,technical design includingdrawingup of questionnaires/schedules,instructionsto
field workers,details of scrutiny,dataprocessingand tabulationwere entrustedto the ISI underthe
general direction of Mahalanobis.In January1971, the design and analysis wing was reorganised
by shifting it from the Indian StatisticalInstituteto the Departmentof Statistics and forming the
NationalSample SurveyOrganisationwhich also includedthe Field OperationsandData Processing
Divisions. The NSS is the largestmulti-purposesocio-economic surveyin the world. The activities
were coordinatedby four main divisions, viz., SurveyDesign and Research,Field Operations,Data
ProcessingandEconomicAnalysis. Currently,besides socio-economicsurveys,datais also collected
throughAnnual Survey of Industries,CropYield and Area EstimationSurvey,UrbanFrameSurvey
and Prices Survey.J.M. Sengupta,D.B. Lahiri,S. RajaRao, M.N. Murthywere associatedwith NSS
since the early rounds.
According to D.B. Lahiriwho is one of the chief architectsof the NSS right from the beginning
"theNSS chose a multi-purposeand multi-subjectframework,and the surveydesign was gradually
so evolved as to permitstudy of the inter-connectionsbetweenthe variouscomponentsof the socioeconomic picture of the country and its constituentregions and states. The NSS has in the main
been a population survey in a comprehensivesense, althoughthere has been a sizeable effort on
the estimation by an area survey of crop acreage and productionto which Mahalanobisattached
great importancebecause of chronic food shortage".Commentingon the complexity and scope
of Mahalanobis'splans for the National Sample Surveys of India, Deming (1973) remarked:'No
country,developed, under-developedor over-developed,has such a wealth of informationabout its
people as India .........'
Other Statistical Divisions and Activities in the Government
Among the otherimportantstatisticalwings in the Governmentof India,the office of the Director
General of CommercialIntelligence and Statistics is one of the oldest establishments.It continues
to be responsiblefor commercialintelligenceand foreign tradestatistics.The office of the Registrar
General which was createdin 1948 carriesout its decennial Census Operationsas well as Sample
RegistrationSystem and publicationof otherdemographicand vital statistics.Duringthe 1951 and
1961 Censuses, several major changes were adoptedeither in collection or analysis of data under
the leadershipof the RegistrarsGeneral,R.A. GopalaswamiandAshok Mitra.Forprovidingreliable
vital rates to meet the needs of planningand policy decision, the Office of the RegistrarGeneral,
19
20
J.K. GHOSH,
P. MAITI,T.J.RAO& B.K. SINHA
21
2)'-((1
A -
L2)
where Ati is the mean vector for the ith population, i = 1, 2 and E-1 is the inverse of E.
Mahalanobis (1930) considers only the case of a diagonal E. In Mahalanobis(1936), he considers also the correlatedcase, i.e. general E and also introducedthe familiarstudentizedversion
where the parametersA and E are replacedby theirestimates.He also introducedwhat amountsto
graphicalcluster analysis.
There is evidence (Pearson,1928, quotedin Rudra,1996) thatPearsonhad expressedreservations
about the D2-statistic and did not want to publish Mahalanobis'spaper on the D2-statistic which
had been originally submittedto Biometrika.Mahalanobiswas disappointedbut did not give up
(Mahalanobis, 1929, quoted in Rudra, 1996). He published his paper elsewhere (Mahalanobis,
1930). Mahalanobis'sconfidence in this work has been amply justified in subsequenttheory and
applications.One of the firstmajortheoreticalcontributionsof the IndianSchool was the proof that
the studentizedD2-statistichas a non-centralF distribution(Bose & Roy, 1938). The D2-statistic
remainsa powerfulandfundamentaltool in multivariateanalysis,classificationproblemsandcluster
analysis.
Some of the conclusions of Mahalanobisin his anthropologicalpapershave also stood the test of
time. He was right in claiming that the Bengali Brahminsresemble other Bengali castes far more
closely thanthey resembleBrahminselsewherein India.However,in some othercases, laterevidence
points in a differentdirection.For example, as far as the Anglo-IndianCommunityis concerned,it
is now believed that Mahalanobishad probablyconfinedhis study to a sample from upperstratum
of the community and hence his conclusion of resemblanceto uppercaste Hindus is applicableto
the upperclass Anglo-Indiansonly.
With this backgroundin analysis of anthropologicaldata, it is not surprisingthat Mahalanobis
(1933) turnedhis attentionto Risley's (1891) famousdataset whichwas collected on 5784 individuals
belonging to 87 castes and tribes of NorthernIndia summarizedby 11 means and 8 indices. In a
total of 20797 values, he found 142 serious discrepanciesof which 133 were correctedby 'crossexaminationof data' and 'internalconsistency checks'. Also the importanceof standardizationin
measurementswas stressedby him (Mahalanobis,1928).
During the early twenties, an officer of the Indian Civil Service, J.A. Hubback observed that
the crop cutting system was quite inadequateand defective and conductedextensive crop cutting
experimentson the paddycrop, which he called a 'randomsamplingmethod', by demarcatingareas
by a specially devised detachabletriangularmetal frame as distinct from the traditionaltapes to
demarcatethe rectangularareas.This firstexperimentin 1923 relatedto the GoddaThanaof Santal
Parganasof Bihar State where 400 samples were harvestedover an areaof 100 squaremiles. Later,
in 1925 he extendedhis methodto 8 subdivisionseach of about 1000 squaremiles in SantalParganas
District and also in the State of Orissa.Mahalanobisgot interestedin the use of randomsample cuts
for the estimation of areas under crop and crop yield. Hubback'swork (Hubback, 1925) seems to
have had influence on Fisher,Yates,Cochranand othersof thattime in Britain.
On the suggestion of Sir Gilbert Walker,the Director General of Observations,Mahalanobis
(1923) looked at the correlationsbetweenupperair variables.He got appointedas a Meteorologistin
22
J.K. GHOSH,
P. MAITI,T.J.RAO& B.K. SINHA
Calcutta,besides his usual duties as Professorof Physics at the PresidencyCollege. Thus duringthe
late twenties, Mahalanobisgot involved in variousdirectionsof the growthof the new discipline of
Statistics-the conceptualdevelopmentsin multivariateanthropometricdataanalysis,the acceptance
of sample surveys as a methodof datacollection as stressedby Kiaer(1895, 1897), Bowley (1906),
Jessen (1926) among others, meteorologicalstudies, crop cutting experimentsto name a few. It is
only naturalthatthe 'StatisticalLaboratory',which was being runas a 'workshop'at the Presidency
College, Calcutta,should be given a statusof an institutionof researchand higherlearning.
Two young colleagues, SubhenduSekharBose and HarishChandraSinha urgedMahalanobisto
approachsome importantpersons of the countryto starta StatisticalSociety. On the fourteenthof
December 1931, ProfessorPramathaNathBanerjea,Nikhil RanjanSen andP.C.Mahalanobisissued
the following notice: (cf. Sankhya,1, p. 124, AnnualReport)
"A meeting will be held to consider steps to be taken towardsthe establishmentof an
IndianStatisticalInstituteon Thursday,the Seventeenthinstantat 2.30 p.m. in the Board
Room of M/s. Martin & Company, 12, Mission Row, Calcutta.Sir R.N. Mookerjee,
K.C.I.E, K.C.V.Owill preside."
Fromthe minutesof the meeting, we note that 'it was unanimouslyresolvedthatan IndianStatistical
Institutebe startedand that Sir R.N. Mookerjeebe requestedto accept the office of the Presidentof
the Institute ..'.
The Indian Statistical Institutewas founded as a society on 17 December, 1931. Sankhya,the
IndianJournalof Statistics,was foundedtwo years later.
The Second Period (1931-1950)
The second periodis markedby the emergenceof samplesurveys,multivariateanalysisanddesign
of experimentsas majorstatisticaltools for practicalwork. They were also subjectsfor researchat
cutting edge. Another notable feature of the period was the introductionof undergraduateand
postgraduatecourses in Statistics.Last butnot least, trainingprogrammesand practiceof Shewart's
StatisticalQualityControlwere introducedduringthis period.
Startingwith exploratorysurveysconfinedto a few squaremiles in Bengal in 1937, Mahalanobis
was perhaps the first person to organise and carry out an objectively defined large scale survey
covering the whole of Bengal (about 59000 squaremiles) in 1941. This scheme was designed to
estimate the yield of jute crop and acreageunderjute in Bengal. The survey in its earlieryears was
able to provideimportantinformationon the variabilityof the characteristicsunderstudyandcosts of
differentsurvey operations.He recognisedthe need for assessing and controllingthe non-sampling
errors.
In his reportto the Indian CentralJute Committeeafter a careful assessment of Mahalanobis's
method H. Hotelling states:
"... no technique of random sample has, so far as I can find, been developed in the
United States or elsewhere, which can comparein accuracyor in economy with that
describedby ProfessorMahalanobis...."
Also, Fisher commented:
"... The ISI has takenthe lead in the originaldevelopmentof the techniqueof Sample
Surveys, the most potentialfact findingprocess availableto the administration."
Mahalanobis's(1946) Sample Survey of Jute productionin Bengal gave a figure of 7540 bales (1
bale = 400 lbs.) while the plot to plot enumerationby the Governmentwhich was ten times more
costly and had a fifty fold manpowercomparedto the samplingmethod gave an underestimateof
23
6304 bales. The customs and tradefigure which independentlytakes into account nearly all of the
producegave an answer of 7562 bales.
D.B. Lahiri collaboratedwith Mahalanobison the analysis of errorsin Censuses and Surveys
in the Indian context (Mahalanobis& Lahiri, 1961). According to Lahiri (1973) the three notable
contributionsto sample survey techniquesby Mahalanobisare "pilot surveys, concept of optimum
survey design, and interpenetratingnetwork of subsamples (IPNS)". All the three concepts are
forerunnersof importantpracticalstatisticalcontributionsthat emerged later-'pilot surveys' as a
preludeto Wald's 'sequentialanalysis', 'optimumsurveydesign' stressingthe philosophythatall the
resourcesprovidedfor a surveyshouldbe used optimallyas a precursorto 'operationsresearch'and
'IPNS technique' as one of the curtainraisersfor 'resamplingprocedures'like Bootstrap.Edward
Deming (1964) acknowledgedthus ... ". .. for 14 yearsI haveused only interpenetratingnetworkof
samples (IPNS), initiatedby him (Mahalanobis),as everyoneknows, about 1936. .. " ". .. The main
featureof the IPNS is simplicity in the calculationof the standarderrorof an estimate.It also enables
one to estimaterapidlythe mathematicalbias, if any,in the formulaof estimation.... It helps to detect
gross blunders in selection, recordingand processing. It permits evaluationof variancesbetween
investigators,coders and otherworkersin the variousstatisticalstages of processing".Mahalanobis,
(Mahalanobis, 1938) was aware of the probabilityproportionalto size (pps) selection (Hansen &
Hurwitz (1943)) even in 1937. He realised that, in agriculturalsurveys, it would be necessary to
select plots using the cumulativetotals of theirareas,since these areasvary considerably.However,
he assumedthatexcessive workload wouldmakethe selection impracticableandtakinginto account
the high costs of travelbetween plots which are widely scattered,he recommendedthe use of 'grid
sampling'. On a different level, he had considered the possibility of air surveys 'using specially
sensitised films' for estimationof crop acreagein 1937 itself-a techniquewhich has now become
popularas 'Remote Sensing'.
Mahalanobis's work on D2 and studentized D2 led to very innovative use of matrix and ndimensional geometric methods for derivationof the distributionof these statistics undera multivariateNormalmodel. The identificationof the studentizedD2 as a non-centralF by R.C. Bose and
S.N. Roy was the firstmajorbreakthroughin theoreticalstatisticsin the thirtiesby the Indianschool.
Mahalanobis'sworkwith S.S. Bose on cropcuttingexperimentsandyield estimatespavedway for
fundamentaldiscoveriesin constructionof designof experimentsby R.C.Bose using finitegeometries
and Galois fields. R.C. Bose derived new methods of constructionfor balancedincomplete block
designs, orthogonallatin squares,confoundedfactorialdesigns andmuchelse. Formanyyears,India
remaineda leading contributorin this area.Otherswho madeimportantcontributionswere K.R. Nair
(Mahalanobis& Nair, 1940; Nair, 1992) and C.R. Rao. It was C.R. Rao who introducedthe notion
of orthogonalarrays(Rao, 1947) which in the handsof Taguchihad a profoundeffect on industrial
experimentation.
At Mahalanobis'sinitiative,the firstpost graduatecourse in Statisticswas introducedat Calcutta
Universityin 1941. The firstbatchof studentsincludedC.R. Rao, who influencedthe growthof the
ISI more than anyone else except Mahalanobisand, along with Mahalanobis,is the most famous
statisticianto come out of the Indiansubcontinent.This is how he (Rao, 1992) recalls his early years
at the ISI:
"I passed the M.A. degree examinationwith a first class, securing the first rankand a
high percentageof marks.I was thus among the firstfive to receive the M.A. degree in
Statisticsfrom any IndianUniversity.The Professorofferedjobs to all of us in the ISI as
technical apprenticeson a salaryof Rs.75 a month.I joined the ISI in December 1943."
(At currentexchange rates Rs. 75 amountsto abouttwo U.S. dollars).
Prior to this, C.R. Rao had an M.A. degree in Mathematicsfrom AndhraUniversity in 1940.
He received his Ph.D. from Cambridgein 1948 underthe guidance of R.A. Fisher and Sc.D. from
24
the same University in 1965. When Rao returnedfrom Cambridgein 1948 he took charge of the
Researchand TrainingSchool (RTS) which was one of the Divisions of the Institutewith the sole
responsibilityof organisingthe researchand trainingactivities.D. Basu was amongthe firstbatchof
scholars,who joined in September,1950. Graduallywith freshadmissionsof traineesevery year,the
RTS expandedits activitiesfurther.C.R. Rao held the post of Professorand Head of the Division of
the TheoreticalResearchandTrainingin the ISI from 1949 to 1963. In 1963 he became the Director
of the RTS. After Mahalanobis'sdeathin 1972, Rao became the Secretaryand Directorof the ISI,
the designationswhich Mahalanobishad.
During the mid-forties Mahalanobisforesaw the need for introducingQuality Control (QC) in
Indian industries and later C.R. Rao also had been associated with the QC movement in India.
In 1945-46 a special course on QC was organisedby the ISI which was attendedby 12 persons.
There was encouragementfrom a few men like C. Tattersallof the OrdnanceTesting Laboratory
who fully realized the importanceof using QC in industry.But governmentdepartmentswere
apathetic.Influencedby the pioneeringworkdone by WalterShewartin StatisticalQualityControl,
Mahalanobisinvited him to India. Shewartarrivedin Calcuttaon December 22, 1947 and took the
lead in organizing a one-week conference on 'Standardizationin IndustrialStatistics' in Calcutta
from 8th to 14th February1948 under the auspices of ISI and Indian StandardsInstitution.This
was attendedby 190 persons. All these efforts finally culminatedin startingthe first SQC Unit at
Bombay in 1953 followed by two units at Bangaloreand Calcuttain 1954. The main objective of
these unitsinitially was promotional-to visit industriesandact as consultants.This service was later
on extendedto manyprincipalindustrialcities in the country.Todaythe Division, knownas the SQC
and OR Division, is engaged in both teaching and researchbesides consultationand promotional
activities.
Afterdevastatingfloods in the Brahmaniriverin the stateof Orissain 1926, an expertcommitteeof
engineersattributedthis to the risingriverbed andrecommendeda correspondingrise in the height of
embankments.Whenthe problemwas referredto Mahalanobis,he (Mahalanobis,1931, Mahalanobis
& Chakravarti,1931) studiedthe dataon rainfallin the catchmentareasof the riverduringthe period
1868-1928 and related this to the level of the rivers. Contraryto the engineers' suggestion, he
recommendedconstructionof dams in the upperreaches of the river to stop the excessive rainfall
from flooding the plains. Based on his calculationsfor a multipurposescheme of flood control,
hydroelectricpower generationand irrigationfacilities, the Hirakuddam was constructedin 1957.
The Chief Ministerof Orissa wrote a letterthankingMahalanobisfor his efforts.This work may be
regardedas one of the earliest case studies in Systems Analysis and OperationsResearch,subjects
which flourishedafterthe second world war.
Fifties and Early Sixties
The substantialcontributionsof the Instituteto theoreticaland applied work, its trainingand
promotionalactivitiesculminatedin recognitionby the Governmentof India.The parliamentpassed
the Indian StatisticalInstituteAct, 1959 which declaredthe Instituteas an "Institutionof National
Importance"and empowered it to awarddegrees and diplomas in Statistics. The Prime Minister
of India, PanditJawaharlalNehru who piloted this bill in Parliament,made a speech that remains
relevantas one of the most inspiringdefences of science and academic freedom ever made by the
head of a Government:
... "Now we wantscience to grow, andI thinkit is quiteessentialthatwe shouldaccept
this broadapproachto this question that scientific work should have a certainlatitude.
Therefore,we have decided that in this particularmatter,this should continueto be an
autonomousorganization..."
25
Soon after, in June 1960, the Instituteintroducedthe Bachelor of Statistics (B.Stat.), Master of
Statistics (M.Stat.) and Ph.D. degree courses. In view of its currentexpertise in the related areas
of QuantitativeEconomics, Mathematicsand ComputerScience, the Act has been amendedby the
Parliamentin September1995 permittingthe Instituteto give degrees in these relateddisciplines as
well.
Fisher's view that "teaching,instructionor trainingin Statistics, at whateverlevel, is bound to
be, on the one side with fact finding projectsin the traditionalstatisticalfields of demographyand
economics, and on the other side with opportunitiesto gain firsthand familiaritywith at least some
field in naturalsciences... " was also sharedby Mahalanobis.The syllabusfor the B.Stat.degreethus
includes an acquaintancewith biological, physical and geological sciences. With ProfessorJ.B.S.
Haldane as a regularstaff member,the Biometricunit expandedfurther.This unit had been set up
earlier under the leadershipof Masuyamabefore his departurein August, 1954. The Institutenow
has a Biological Sciences Division consisting of 37 scientific workers.
As early as in 1950, the ComputingMachines and ElectronicsLaboratory(CMEL) was started
in the ISI. In March, 1956, HEC-2M was installed while a Russian ComputerURAL was gifted
in 1959. Lateron the Institutepossessed IBM 1401 and Honeywell systems. A joint projectof the
Indian Statistical Instituteand the JadavpurUniversity,Calcuttaresulted in the first Indian-made
solid state generalpurposedigital electroniccomputerISIJU-1 in April 1966. Furtherimprovements
had to be abandonedon the advice of the Government.However, the Instituteremains a leading
centre of researchin Image Processing,PatternRecognitionand variousotheraspects of theoretical
and appliedComputerScience.
Mahalanobissaw the need for trainingstatisticalofficers from the MiddleEast, South and South
East Asia, the Far East and from the CommonwealthCountriesof Africa. Under the auspices of
UNESCO and the Governmentof India, the InternationalStatisticalEducationCentre (ISEC) was
opened in 1950 and is jointly operatedby the InternationalStatisticalInstituteand the ISI, Calcutta.
Since its inception the centrehas providedtrainingto 1239 traineesfrom over 50 countries.
After Mahalanobis'sdeath,C.R. Rao continuedas Secretaryand Directorof the ISI from 1972 to
1976. In 1976, the Institutegot a distinguishedprobabilistas its new Director-Gopinath Kallianpur.
Kallianpurwas a Professorat the ISI in the fifties, buthad left permanentlyto workin the U.S.. After
thirteenyears at the Universityof Minnesota,Kallianpurreturnedto the ISI as the Directorin 1976,
a post he held until 1979.
The ISI also had a new Constitutionin 1976. The new Constitutionputs more stress on teaching
and research and less on societal activities. The post of Secretarywas abolished, but the Director
was to functionin futurealso as the Secretaryof ISI. The ISI was to remainboth as an Institutionof
National Importancegovernedby the Act of Parliamentand a Society governedby the Registration
of Societies Act. Currently,the Institute has its headquartersin Calcutta and two other centres
at Delhi and Bangalore. It also has a network of service units of the SQC and OR Division at
Vadodhara,Mumbai,Thiruvananthapuram,
Pune,Coimbatore,Chennai,Hyderabad,Calcutta,Delhi
and Bangalore.
Arguablythe golden period of the ISI was the fifties (cf. Rao, 1973). In additionto Mahalanobis
and Rao, the faculty included R.R. Bahadur,D. Basu, G. Kallianpur,D.B. Lahiri, M. Mukherjee,
R. Mukherjeeand many other distinguishedluminaries.Bahadurwas a professor at the ISI from
1956-1961, when he returnedto the Universityof Chicago.D. Basu left Dhaka(now in Bangladesh)
after independenceto join the ISI as a studentand laterbecame a Professoras well as the firstDean
of Studies. Between them Rao, Bahadur,Basu and Kallianpurand a new groupof brilliantstudents
R. Ranga Rao, V.S. Varadarajan,
S.R.S. Varadhanmade fundamental
including K.R. Parthasarathy,
contributionsto probabilityand classical inferenceduringthis periodwhich were as importantas the
earlier contributionsof Bose and Roy to design of experimentsand multivariateanalysis. Among
the other earlier studentsof the Institutewho achieved internationalreputationare G.P. Patil, T.N.
26
On the recommendationof the Royal Commissionof Agriculture,a small Statistics section was
set up in the Indian Council of AgriculturalResearch(ICAR) in 1929. P.V. Sukhatmejoined the
section in 1940. After completingschool educationin Pune, P.V.Sukhatmegraduatedin 1932 from
Fergusson College of the same city with Mathematicsas the principalsubject and Physics as the
subsidiary.During 1933-36, he studiedat the UniversityCollege of Londonandwas awardeda Ph.D.
in 1936 and a D.Sc. in 1939 for his work on bipartitionalfunctions. Before joining the Statistics
section in ICAR, he was a Professor at the All India Instituteof Hygiene and Public Health at
Calcuttaduring 1939-40. Towardsthe end of 1943, the enquirycommitteeset up by the Government
of India to look into the causes of the devastatingBengal famine observed that one of the main
factors responsible for the famine was the defective statistics of crop productionavailable at that
time. The Statistics section, under the guidance of Sukhatme,began researchin the methods of
collection of yield statistics of crops by developing survey techniques of yield estimation under
randomsampling.Official forecastswere also being releasedfrom the resultsof sample surveys by
1949. The workdone by the section was recognizedby the UnitedNationsandthe FAOand a special
trainingprogrammewas organisedfor the statisticalofficersfromthe South-EastAsian governments
27
28
J.K. GHOSH,
P. MAITI,T.J.RAO& B.K. SINHA
circularcuts in the National Sample Survey and other surveys. As againstthis, the ICAR had been
using the rectangularcuts of size 33' x 16.5' in the crop yield surveys conductedthroughthe state
agency.
Keen interestin resolving this technicalcontroversyon this issue was shown by Mahalanobisand
he suggestedjoint studies both by the ICAR and the ISI (Mahalanobis,1946):
"I may mention, however, that for some considerabletime, I have been pressing on
ICAR authoritiesthe need for carryingout crop cutting work by both ISI and ICAR
methods in the same region with a view to studyingthe relativeefficiencies of the two
systems."
Adhikariindicatesthatthese studies did not reveal significantdifferencesbetweenthe two methods.
However,the controversycontinued.
Both the ISI under the leadershipof Mahalanobisand the ICAR under the leadershipof Panse
differedsignificantlyaboutthe investigatingagencies in conductingthe field work also.
In the scheme of Panse, it was emphasizedthat any sampling methodmust fit into the existing
administrativestructure.Mahalanobis,on the other hand, advocatedthat the field work should be
done by well trainedinvestigatorsrecruitedfor the purposeof the survey.
Adhikari(1990) relates this to the fact that Panse and Mahalanobiscame from differentparts of
India with very differentsystems of collecting revenue-one going back to the Moghul period and
the otherintroducedby the British.
"The Moghul Emperorappointedofficials called Jagirdars,Inamdarsover large tracts
responsiblefor the collection of land revenue.At the level of the village, there was the
village accountant(now called a patwari), also a state employee who would actually
determinethe land revenue of each cultivatorand collect it on behalf of the officials.
"Atthe end of the 18thcentury,the Britishgovernmentintroduceda new systemin a fairly
large partof India, particularlyEasternIndia, called the permanentsettlement.In this
system the intermediatetax collectors,called the Zamindars,were made responsiblefor
the paymentof revenueof the largetractsunderthemto the Britishtreasury.The amount
of revenue was made permanent"... "... The system of patwaries as functionaries of
Governmentceased to exist..."
What appearedto be a scientific controversywas rootedin the social backgroundof the scientists
and, in a way, both were right and both were wrong.
Sastry (1977) refers to joint studies of the Ministry of Agriculture,CSO and ISI conductedin
1960-61 and the studies by a TechnicalCommitteeset up by the PlanningCommission conducted
on 4 crops during 1963-66. No markeddifferencesin the yield rateswith the two types of cuts were
observed.
7 Statistics in some Indian Universities
The Departmentof Statistics of CalcuttaUniversityis the oldest in the country and one of the
pioneering Departmentsin the world which introducedStatistics as a separate and full-fledged
discipline and not as a partof Mathematicsor Economics. It was startedin 1941 at the initiationof
Mahalanobisin collaborationwith the IndianStatisticalInstitute,which was at thattime functioning
fromthe premisesof the PresidencyCollege, Calcutta.The post-graduateMaster'sdegreeprogramme
in Statisticsof CalcuttaUniversitywas the firstof its kind introducedin Asia. Initially,Mahalanobis
was the honoraryHeadof the Departmentof Statisticswhich consistedof one Lecturer,one Assistant
Lecturerand some part-timelecturers.As regardsresources,it had a few hand-operatedcalculating
29
Most of the earlierdiscussion centeredon the period before the sixties. In this section, we take a
quick look at the sequel.
One of the reasons for rapidgrowthof Statisticsin Indiawas the close interactionbetween the ISI
and various technical wings of the Governmentof India.It did not last very long after 1960. Nehru
died in 1964, Mahalanobisin 1972, ending an era of close cooperationthatbegan in 1947.
30
J.K. GHOSH,
P. MAITI,T.J.RAO& B.K. SINHA
There were other historical reasons for fewer interactions.The Planning Commission of the
Governmentof India was strengtheningits in house research.Also, the PlanningUnit of ISI Delhi,
which was stationedin the same buildingas the PlanningCommissionmoved to a new campus, far
from the PlanningCommission. Finally, differencesin perceptionabout what would be an optimal
path of developmentled to partingof ways.
Something similar happened to weaken relations with the NSS. During the final years of
Mahalanobis'slife, ISI's managementof the Design Division of the NSS had come in for criticism. Reports of surveys were far behind schedule. A year before Mahalanobisdied this section
was taken out of the ISI and made a part of the NSS Organisation.In a sense this was a natural
administrativestep leadingto the unificationof the field operationsandtheoreticalwings of the NSS.
But it did not bringaboutany improvementin publicationof reports.The relationwith the NSS was
neverthe same again even thoughISI is representedon the GoverningCouncil of NSS.
Globally too, specially in the United States, theoreticaland applied statistics had been drifting.
This had its effect on Indianstatisticiansmany of whom were trainedin the U.S. or had interactions
throughvisiting assignments.
Therewere manyachievementson the theoreticalside to compensatefor thisisolation.Indeedthere
was an exponentialgrowthin fundamentaltheoreticalworkby Indianstatisticiansin Indiaandabroad.
Some of the notable contributionsby the post-C.R. Rao generationsinclude developmentof new
tests in multivariatenon-parametricinferenceby S.K. Chatterjee,M.L. Puriand P.K.Sen, a critical
evaluation of classical survey sampling from differentpoints of view by V.P. Godambe,D. Basu
and J.N.K. Rao and introductionof a new class of designs called searchdesigns by J.N. Srivastava.
Moreover,R.N. Bhattacharya,J. Sethuraman,M. Ghosh,J.K. Ghoshandmany otherscontributedto
variousaspects of asymptoticsand inferencerangingfromrigorouslyderivedEdgeworthexpansions
and higher order asymptotics to sequentialanalysis, reliability and life testing. A paradigmshift
involving Bayesian Analysis had arrivedin India.
The scenario on the applicationsside showed less activity.It was here that the effects of breach
between academiaand government,the prime user of Statisticsin India, is most visible. But there
have been outstandingexceptions as well as modest but sustainedgrowthin certainareas.
We would place in the first category Sukhatme's(1965) new hypotheses about nutritionwhich
suggest that an individualcan adjustto varying intakes of calories and that the usually stipulated
norms may be higher than necessary.A bivariateanalysis of availabledata on protein and calorie
intake for Asian countries, particularlyfor India, revealed (Sukhatme, 1965) that when diet is
adequatein energy,the proteinintakeis usually satisfactory.It was shown thatproteindeficiency is
the indirectresult of inadequateenergy in the diet. His scientific beliefs as well as philanthropyled
to the establishmentof 'IndiraCommunityKitchen' in Pune. Based on a survey carriedout by the
NSS, Minhas analyzedthe datawhich seems to lend empiricalsupportto this hypothesisbut results
of similarenquirieslaterhave been ambiguous.
Othersubstantialcontributionshave been in the assessmentof the extentof povertyandits decline
after independence.Contributorsinclude Nikhilesh Bhattacharya,S.D. Tendulkarand others. (See
for example, Pal, Chakravarty& Bhattacharya(1986), Tendulkar(1989), Minhas,Jain& Tendulkar
(1991)). The monographby Bhattacharya,Coondoo, Maiti & Mukherjee(1991) deals with time
trendsin povertyand inequalityin ruralIndiausing NSS budgetdatafrom 1952-53 to 1983. In this
workan econometricmodel is fittedto explainthe observedvariationin the incidenceof this poverty.
Therehave also been extensivestudiesof officialstatisticsanddatacollectedby NSS. An innovative
study by Minhas (1988) shows how estimates obtainedfrom these two sources can be reconciled.
Among other importantcontributionswe would include a definitivestudy of one of India's largest
dams by Minhas et al. (1972).
We now leave aside these outstandingbut mostly individualachievementsand turnto areas that
saw a modest but sustainedgrowth.These includemanyimprovementsin the preparationof national
31
So far its influence on the StatisticalSystem in India is not properlyfelt. However,it is beginning
to take an active step in encouragingnew initiatives.The IndianOfficial StatisticalSystem itself is
receiving a great deal of critical attentionboth from scholarsand from the media. We would regard
this too as a positive sign of the relevanceof Statisticsand the need to change with times. One of the
importantnew prioritiesis to provideshorttermestimatesof the nationalincome and industrialand
agriculturalgrowth.The CSO's shorttermestimateshave not been as reliableas its annualestimates.
In any case a right frameworkis now in place for interactionbetween academia, industryand
governmentas well as interdisciplinaryresearchinvolving Statistics and other sciences. Only the
J.K. GHOSH,
P. MAITI,T.J.RAO& B.K. SINHA
32
Conclusion
33
34
J.K. GHOSH,
P. MAITI,T.J.RAO& B.K. SINHA
Resume
'
Ceci est une courte histoire de l'hvolutionde la Statistiqueofficielle et acad6miqueen Inde. Nous retragonsl'histoire
'
partirdes origines de l'antiquit6jusqu'h l'histoire plus r6centemais portonsplus d'attentionsur la periodedes annees 1930
1960. Ceci et l'histoire d'institutions,des gens qui les bitirentet les formhrentet d'id6es.