ABSTRACT
/ E+uiy d i f f e r e n t methods of \laride e s t i m a t i o n i n
complex Surveys have been proposed by many r e s e a r a h e r s .
quenoville (1956) introduced t h e jackknife procedure as
a means of reducing bias of rgtio estimates. Hansen,
Hurvtitz, and Madow ( 1953) described t h e random groups
procedure b’jhich is c l o s e s t .LG the ori8inal concept of
saxpling, The b o o t s t r a p method, introduced
o r i g i n a l l y by jzfron (1979) for the independent and
identically distribute4 case, r e l i e s on recomputing t h e
es-Limzte 0 a laybe nuiiiber B of times by resampling t h e
ori,i.~al cample. simple r e p l i c a t i o n is both easy t o compute
and by r e p l i c a t i n g at both phases, t r u e l y independent
gamples ire foriiied,
ChapLer 1 UP t h i s p r o j e c t reviews the meaning of
~~rilpl-ii.::;s urvey arid sample survey d e s i g n . I t a l s o contai.
11~ the dei’iriicion of i n t e g r a t e d survey wiih its
advailtzbes and .J;i;;advantates, I have included b r i e f
notes on Plas-ker scimple and i-Ls usefulnessi
Chapter 2 include the d e s c r i p t i o n of the 1900 – 1986
FISH de5:ign and 1987 – 1992 NISI-: design, It a l s o
exarni~~etsh e o r e t i c a l l y t h e vvsights used f o r t h e
estimzcion i n the two designs. I have applied the
, ” estirAi;icr;; .o;: i~roceitureco f t h e two de s igns based 011 t h e
198?/88 survey ddta i n order t o comp&re t h e i r relatim
per f ori~iances~
iii
ch;i\gtY~-3; -= -*- -r r- . .- *..
f o r obtaining variallce i n lar$e scale sample surveys, .
They incluue t h e jackknife and b~otstrapm ethods, In
chapt e r 4, t h e method of u~ingt h e mean of t h e repli~aktre
r a t i o e:.ticuates has been considered ix order to compare
convenizntly trith the ,bootstrap hod, Finally chapter
5 concludes and brings out s stions based on my
obsorv~L.i ons,
CHAPTER ONE
1 .I ssili Survey_:Meanis
::;mpling Survey i s a sy~tema-ticc o l l e c t i c n of d a t a
from a p a r t of a population iiirough ,lie use of personal
interviews or other dsta &athering devices in order t o
make inference ahout the whole population. The need to
callect d a t a arises i:il eve?y conceivabl-e sphere of human
ac Livi. ty . exe oula knobsled,e of beha viour and
c h a r a c t e r i s t i c s of a parJcicul.zr set oP ul1i.L~i n a popul
a t i o n i s obtained. by F; tudyj..ng a r e p r e s e n t a t i v e crosss
e c t i o n of some of the u n i t s . We do not ritudy each
i n d i v i d u a l rile~nber of the mits because wc b e l i e v e t h a t
a r e p r e s e n t a t i v e i s cufficient for making i n f e r e n c e about
the whol-e set,>
k ,sample survey has noc come t o be con~idered as
a n organised f a c t – finding ins-irurfient, ~tirsng ortance
to modern c i v i l i z a t i o n l i e s i n the f a c t t h a t i t can be
used t o surninarize, for the guidance of admi.nistrat.ion,
f a c t s vihick would otherwise be i n a c c e s s i b l e owing t o the
remoteness a:ld obscurnity of the persons or other u n i t s
concerned, or .their numerousnesc. AS a fact- finding
i n s t r ume n t , a sai~~~sjulrev e y i s concerned wi t h t h e
a c c u r a t e ascertainment of t h e i n d i v i d u a l facts recorded
and with t h e i r cornpilL2tion and summari.zation,
1.2 zmple su-r veyJLesign
~osotf the s t e p s i n v o l v e d i n p l a n n i n g a sample
survey a r e common t o those for a complete enumeration,
Some of the important a s p e c t s r e q u i r i n g a t t e n t i o n at
the plaming staLe a r e the following; Formulation of
d a t a reqiiiremen-ts, adhoc or r e p e t i t i v e survey, methods
of d a t a c o l l e c t i o n , survey r e f e r e n c e and r e p o r t i n g
periods, pr ciSlem of sanip:Ling frame, planning of p i l o t
survey, choice of sampling desj. gn e t c. The p r i n c i p a l
airii of sampling desigil is t o choose the design with the
siaallest e r r or because cli f i’erent sampling designs
r e s u l t in d i f f e r e n t sa~lpling errors, However, i n
cnoosini, a design one h~,st o be si-tisfied wiLh a
ratioml sam~lingd e s i g n chosen from among a few
a1 tcrnacives. s p e c i a l attejkion should be $aid .to the
wo r k a b i l i t y o f t h e p a r t i c u l a r sampl ing des~gnu nder
the p r e v a i l i n g o p e r a t i o n a l conc?:i t i o n s . Generally a
stratf f i e d mu1 tistage desip is adopted for l a r g e s c a l e
sample sul-veys. Many surveys taken by sampling seek
information of obvious importance t o n a t i o n a l planning,
~ccht o p i c s a s agrl-cultural p r o d u c t i o n and l a n d us e ,
~nbour force, i n d u s t r i a l production wholesale and r e t a i l
p r i c e s , h e a l t h s t a t u s of’ the people and family incomes
and expenditure are normally covered, These t o p i c s a r e
based on 1:ousehold survey.
– – 2
~ousel-iold surveys on these v a r i e t y of t o p i c s form a
sizeade and indispei~sablc p a r t of the o f f i c i a l s t a t i s t i c a l
operatioris of rriany c o u n t r i e s . Since. f i r s t 5mbarkFng on
household surveys tilose sta-kistlcal oi’fices have been
conductrin~ these surveys on an adhoc b a s i s . Recently
t h e r e have been need t o in.tegrate t h e s e surveys t o obtain
an i n t e r r e l a t e d s t a t i s t i c s on ~ontinuing b a s i s . For
example, according t o I(ordos (l983), Poland, a f t e r t r y i n g
other methodology of household ml’veys f’rom 1957 up t o
1981 decided i n 1962 t o ernbark on the i n t e g r a t e d system
of household surveys, The implementation wac doile i n
two sjlasas. The fj-rst ghase span 3 982 – 1985 whereby
the so-called minimum progrmme was implemented, The
sec2rid phase covered the period ? 986 – 1 9909 d.uring which
househulci surveys methodology w2s improved, u n i f i c a t i o n
of conce yts, d.e f i n i t i o n s slid classific&tions which
r e l a t e t o hnuse~icjlci surveys were f i n a l i s e d ,
A. number o f developing c o u n t r i e s are developing
integrz’ced on-gorixg houseiiold surveys under ‘il-ie ~~lspices
of t h e u n i t e d Nations National Household Survey Capability
programme (~?\TH,scpa) s a ra-tionaliza+Son of existing
survey programmec;. ‘!’he Kenya Bureau of statistic^ s e t
up a Ns~tl~liailn teqrated sample s u r v e y progralllnie (NIssP)
i n 1974 (lciregyera and philip, 1985), The Federal
o f f i c e of sta’cistics (FO,s) Nigeria, has been involved
i n household survey since t h e ecrljr f i f t i e s , but r e s o r t e d
t o a:l integrated system of household surveys i n 1980,
The implernent~,tionw as t o be done i n two p h a s e s , t h e
j:i l, t ~hases san t h e p e r i o d 1380-1 386 and t h e second
phase from 1987 – 1992.
An integrated survey my be defined as a survey
i n which continuous s e r i e s of d a t a on a wide v a r i e t y of
t o p i c s are c o l l e c t e d f o r t.,,e same s e t of u l t i m a t e
sa!nplil?g u n i t s or wl. tnin the sampling a r e a uiiits. ~hus
accordi-ng; t o Fozeman (1903), the casual arrangemeiits
t h a t had earlier on s u f f i c e d for adhoc surveys could
hardly be effective nor econoinical i n handling frequent
household surveys. This l e d to the i d e a of an i n t e g r a t e d
rnuliis’ub ject survey which makes i t p o s s i b l e f c r data on
se~e~2s.uib Jcct!~t: o be collzcted f o r t h e same s e t o f
ul-timzte samplin; u n i t s ,
An integrsted survey is u s e f u l i n the maximum and
effective uti%iza.tion of the availade f i e l d s t a f f on a
2errnancnt bdsis; helps i n studying the interrelc.:i-l;ionships
amon; i t e m s belongLng t o difilerent s u b j e c t f i e l d s
or t o p i c s . ~t is c o s t e f f e c t i v e and convenient,
I n t e g r a t i o n rnzkes c r o s s – c l a s s i f i c a t i o n using r e l e v a n t
itenis of informat,ion belonging t o d i f f e r e n t s u b j e c t
f i e l d p o s c i b l e and aims at u n i f i c a t i o n of’ b a s i c concepts,
d e f i n i t i o n s and. c l a s s i f i c a t i c n used i n d i f f e r e n t surveys.
4,
-.__’,
It is becauce of t h e s e advan-es that the United Nations
la~mched the NHsCP i n 1979 to assist the developing
c o u n t r i e s t o undertake systematic progrmrnes of household
surveys and t o develoy t h e i r survey c a p a b i l i t i e s .
since an integrated survey c o n s i s t s of a number of
succe~sive surveys on the same or s e v e r a l t o p i c s , i t
becomeb advantageous: t o hevre a s e t of sample units from
which subs ample^ can be s e l e c t e d t o s u i t each successive
survej-s or t o . il~ve a nuibber of independent subsamples
such t h a t any given stlrvey is undertaken over one or
more subsamgles, such a sample is c a l l e d a rqasker Saraple,
A Master Sample is t h e r e f o r e defined as a general
purpose sample from which subsamples can be drawn to serve
the needs of any s p e c i f i c survey or survey rounds conducted
over a period of time,
I n a multis-Lage sample, the same u l t i m a t e smple
u n i t s could be used f o r more than one survey, Altern
a t i v e l y , u l t i m a t e sample u n i t s could be r o t a t e d
s y s t e m a t i c a l l y for a s e r i e s of surveys or a new u l t i m a t e
u n i t s s e l e c t e d f o r each survey. For the sane s e t of
primaries. Another method is t o have a number of
independent subsample of p r i m a r i e s , i n such a way t h a t
any given survey is c a r r i e d out over one or more
subsamples dependiug on its sample size requirments.
Each aliernhtivo has i t s own m e r i t s and demerits,
1LA.- 7~ mnr-i tr; of a ass- ->be ara that it a13nw$
consideraole f1exibilj.t~ i n the sample s e l e c t i o n , f o r t h e
most frequent r e c u r r i n g surveys. It is c o s t e f f e c t i v e i n
c o n s t r u c t i n g frarne for su,bsanpling ~incel i s t i n g i s done
only once t llr oughout the auration of the survey programme.
Other advantages as given by Torene et a1 (1987) a r e :
( i ) A master sam~le of p r i m a r i e s i n t e g r a t e s a l l
the surveys in the survey programme.
( i i ) l’TJlumerators iivuld become more familiar ivith
the iTlas’ier swail~le oi’ p r i m a r i e s than i f new
and d i f f e r e n t pri~lsriesw ere s e l e c t e d f o r
eacil of tlie v~riouss urveys .
(iii) Future surveys could be placed i n the fielci
with more convenience and speed s i n c e he
would already have a f i r s t s t a g e sample.
he i,lain limitation of a master sample design is
-chat its dasigii is usually t a i l o r e d towards the core
survey, The sampling requiremen-ts for the other surveys
i n t h e progarme a r e merely accornrnodated i n t h i s ol?e
master sainb~led esign. The ma s t e r sample de s ign i s n o t
an opti~lal design for a l l the surveys i n t h e programle.
Even at that, the master sample is still a valuable
elelL.ent for i n t e g r a t e d household survey programmes. The
liiaster. san,ple desihn should be determined by the availhble
ccst i,r:j I.ersor~ne;l which a l s o a f f e c t the d e c i s i o n on the
sampl~ &ze, The desidn adopted should be c a r e f u l l y ,
considerea sirice according t o Foreman (1 983) any inade-
%..
qu~ciesi n ihe master samde selecti.on would be r e f l e c t e d
i n all octirnateo QQ i t , Fin~llyt h e desikn should
be such L!~c. t the sarupling errors a r e easily colliputed f o r
any of the survey est~-mtes, e.g. by replication.
The ]?OE s.dcpced n r.otdtional ma s t e r sarfi~~loef p r ima r i e s
with a nei, selec tioil 01 :louseholds for each survey, ‘.?he
ma s t e r sample p r ima r i e s w2re selected 1~~9cli.e ~,<~A frzl;e
created ucring the 1973 popul-z tion cencus, Tao d i f f e r e n t
desi~ns:! ere u s e d i n u, ch of t h e 19 s t a t e s thzt rnade up
Nigeixia, one for the urPar: s e c t o r and the other f o r t h e
rural secior. of the Sttte. urban sectorL c o n s i s t e d of
thcse LownG, wiLh 20 EA’S or more, while the rural sector
consi~ted he tcxns with lesc than 20 KA~,s,
l,3 -&–i-.-r n~ and Ob-j e c t i v e s ..–7
~riie o b j e c t i v e s of this tvorli is t o study criticc;tlly
the two NI$H designs used by FO,S during the 1980-1 986
and 1987-1 992 cnvey periods i n order t o ascertaic t h e i r
relati ue per i or~na:lce@,
~lsot h e cstl~nationp roc edur e used by the T~fso r
( i ) l?c\mlolr~ group procedure
( iij Group Jackknife
(iFi) 50~li;s~-r ~npa i v e and improved procel i-.es
Do you need help? Talk to us right now: (+234) 08060082010, 08107932631 (Call/WhatsApp). Email: [email protected].
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
