0% found this document useful (0 votes)
2 views

Research Method, Handout

The document discusses the meaning and purpose of educational research, highlighting its role in generating knowledge and solving educational problems through systematic inquiry. It emphasizes the importance of scientific methods in research and outlines the major characteristics of educational research, including its focus on problem-solving and empirical evidence. The document also lists the main purposes of educational research, such as improving educational practices and developing effective theories in the field.

Uploaded by

afandiiabdallaa
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
2 views

Research Method, Handout

The document discusses the meaning and purpose of educational research, highlighting its role in generating knowledge and solving educational problems through systematic inquiry. It emphasizes the importance of scientific methods in research and outlines the major characteristics of educational research, including its focus on problem-solving and empirical evidence. The document also lists the main purposes of educational research, such as improving educational practices and developing effective theories in the field.

Uploaded by

afandiiabdallaa
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 66

UNI

TONE
TheFundame
ntal
sofEduc
ati
onal
Res
ear
ch

1.
1Meani
ngandPur
poseofEducat
ional
Resear
ch

1.
1.1 Meani
ng

Resear
ch i
s an i
ntel
l
ect
ualact
ivi
ty and r
esponsi
blef
orbr
ingi
ng l
i
ghtt
o cr
eat
e
knowl
edge,cor
rect
ingt
hepr
esentmi
stakes,r
emov
ingexi
sti
ngmi
sconcept
ionsand
addi
ngnewl
ear
ningt
otheex
ist
ingknowl
edge.I
tisal
soconsi
der
edascombi
nat
ionof
t
hoseact
ivi
ti
eswhi
char
eremov
edf
rom dayt
odayl
i
feandi
spur
suedbyt
hoseper
sons
whoar
egi
ft
edi
nint
ell
ectandsi
ncer
eincommi
tmentofpur
sui
tofknowl
edge;
alt
hough
i
tisnott
ruet
osayt
hatr
esear
chi
srest
ri
ctedt
osucht
ypeofper
sons.Ont
heot
herhand,
i
tist
ruet
osayt
hatmaj
orcont
ri
but
ionofr
esear
chcomesf
rom hi
ghl
ygi
ft
edand
commi
tt
ed wor
ker
s.Mor
eov
err
esear
ch i
s al
so consi
der
ed as t
he appl
i
cat
ion of
sci
ent
if
icmet
hodi
nsol
vi
ngt
hepr
obl
ems.I
tisasy
stemat
ic,f
ormalandi
ntensi
ve
pr
ocessofcar
ryi
ngont
hesci
ent
if
icmet
hodofanal
ysi
s.Resear
chi
sanat
temptt
o
di
scov
er,
dev
elopandv
eri
fyknowl
edge.I
tisani
ntel
l
ect
ual
processt
hatdev
elopedov
er
hundr
edsofy
ear
sev
erchangi
ngi
npur
pose,f
orm andal
way
sresear
chi
ngoft
rut
h.
Resear
chi
sal
sodef
inedasanhonest
,exhaust
ive,i
ntel
l
igentsear
chi
ngf
orf
act
sand
t
hei
rmeani
ngsori
mpl
i
cat
ionswi
thr
efer
encet
oagi
venpr
obl
em.I
namor
egener
alway
,
r
esear
chcanbedef
inedasasy
stemat
icandr
efi
nedt
echni
queoft
hinki
ng,empl
oyi
ng
speci
ali
zedt
ool
s,i
nst
rument
sandpr
ocedur
esi
nor
dert
oobt
ainamor
eadequat
e
sol
uti
onofapr
obl
em t
hatwoul
dbepossi
bleunderor
dinar
ymeans.I
tst
art
swi
tha
pr
obl
em,col
l
ect
sdat
aorf
act
s,anal
ysi
scr
it
ical
l
yandsear
chesdeci
sionsbasedont
he
act
ual
evi
dence.

Educat
ionalr
esear
chi
sast
udyandani
nvest
igat
ioni
nthef
iel
dofeducat
ionorbear
ing
uponeducat
ionalpr
obl
ems(
Good,1966)
.Int
hesameway
,resear
chi
neducat
ional
pl
anni
ngandmanagementi
sthatact
ivi
tywhi
chi
sdi
rect
edt
owar
dst
hedev
elopmentof
pol
i
cies,manager
ialf
unct
ions,andski
l
lsneededi
nthef
iel
dofar
ea,et
c.wi
thspeci
al
f
ocusonpedagogi
calandeconomi
caspect
s.Theul
ti
mat
eai
m ofsuchast
udyi
sto
pr
ovi
deknowl
edget
hatwi
l
lper
mitt
heeducat
ionalpl
annerandmanagert
oachi
eve
t
hei
rgoal
sbyt
hemostef
fect
ivesci
ent
if
icmet
hod.I
tist
hesy
stemat
icandschol
arl
y
appl
i
cat
ionoft
hesci
ent
if
icmet
hod,i
nter
pret
edi
nit
sbr
oadersense,t
othesol
uti
onof
pr
obl
emsi
ntheexi
sti
ngeducat
ionsy
stem.

1|
Page
I
ngener
al,onecani
nfert
hateducat
ionalr
esear
chi
sacar
eful
,cr
it
icali
nqui
ryor
exami
nat
ioni
nseeki
ngf
act
sorpr
inci
plest
hroughi
nvest
igat
iont
ofi
ndoutsol
uti
onsf
or
pr
obl
emsi
nandar
oundeducat
ionsy
stems.

1.
1.2 ThePur
poseofResear
chi
nEducat
ion

Themai
npur
posesofr
esear
chi
neducat
ion:

 Topr
ofi
tfr
om t
heexper
ienceoft
hepastsot
hati
twi
l
lbet
hesol
uti
onoft
hecur
rent
educat
ional
probl
ems;

 Todev
elopt
hebehav
ior
alsci
encei
neducat
ional
sit
uat
ions;

 Topr
ovi
deef
fect
iveknowl
edget
hatcanhel
ptheeducat
orst
oachi
eveeducat
ional
goal
sandobj
ect
ives;

 To hel
pthe cl
assr
ooms t
eacher t
o est
abl
i
sh f
aci
l
itat
ive condi
ti
ons i
nthe
cl
assr
oomsi
nor
dert
oachi
eveappr
opr
iat
eresul
tsandt
ouser
esour
cesopt
imal
l
y;

 Todev
elopsoundeducat
ionalt
heor
iesf
oll
owi
ngani
ntensi
vepr
ocessofsci
ent
if
ic
enqui
ryaboutphi
l
osophi
cal
,hi
stor
ical
,economi
calpsy
chol
ogi
cal
,soci
ologi
calet
c.
i
mpactonv
ari
ousaspect
sofeducat
ion;

 To f
ind sol
uti
ons f
oreducat
ionalpr
obl
ems sci
ent
if
ical
l
y so t
hatt
he comi
ng
gener
ati
oni
smadef
reef
rom t
radi
ti
onal
ignor
anceandpr
ejudi
ce;

 Todescr
ibeandanal
ysest
heexi
sti
ngsi
tuat
ioni
nor
dert
osear
chst
rat
egi
est
hat
wor
tht
heeducat
ionsy
stem;
and

 Topr
ovi
deunder
standi
ngofcont
empor
aryeducat
ional
probl
ems.

1.2Maj
orChar
act
eri
sti
csofEducat
ional
Resear
ch.

Educat
ionalr
esear
chi
sthenamef
orat
ypeofpr
ocedur
eempl
oyedi
nanswer
ing
r
efl
ect
ion quest
ions abouteducat
ion t
hati
s quest
ions f
orwhi
ch answer
s must
gener
atebyr
efl
ect
ivet
hinki
ng.Ref
lect
ivet
hinki
ngi
srequi
redi
nthedi
scov
eryofnew
pr
ocedur
es,r
ulesandpr
inci
plest
hathel
peducat
ionalmanagerandsuper
visor
sto
i
mpr
ovet
hesy
stem.Asr
esear
chmaybedef
inedast
heappl
i
cat
ionoft
hesci
ent
if
ic
met
hodi
nthest
udyofpr
obl
ems,i
tisusef
ult
oret
hinkt
hechar
act
eri
sti
csofsci
ent
ist

2|
Page
suchasani
nst
abl
edesi
ret
oknowt
het
rut
handanabsol
utehonest
yinr
epor
ti
ngwhat
he/
shef
inds(
Good,
2006)
.Ifsomebodyi
nvol
vesi
neducat
ionalr
esear
chhemusthav
e
t
hechar
act
eri
sti
csofsci
ent
ist
sment
ionedabov
e.

Accor
ding t
o Best& Kahn (
1992)
,the maj
orchar
act
eri
sti
cs ofr
esear
ch can be
descr
ibedasf
oll
ows:

 Resear
chi
sdi
rect
edt
owar
dst
hesol
uti
onofpr
obl
ems.

 Resear
chemphasi
zest
hedev
elopmentofor
gani
zat
ions,pr
inci
plesort
heor
iest
hat
wi
l
lbehel
pful
inpr
edi
cti
ngf
utur
eoccur
rences.

 Resear
chi
sbasedonobser
vabl
eexper
ienceorempi
ri
cal
evi
dence.

 Resear
chdemandsaccur
ateobser
vat
iondescr
ipt
ionanal
ysi
sandi
nter
pret
ati
on
pr
ocess.

 Resear
chst
ri
vest
oel
i
minat
efeel
i
ng,
emot
ion,
prej
udi
ce,
fav
orandpr
efer
ence

 Resear
chr
equi
resex
per
ti
se sy
stemat
icalappr
oachandaccur
acyofi
nvest
igat
ion
par
ti
cul
arpr
obl
ems

 Resear
chi
snotapur
elymechani
cal
appr
oach

 Resear
chmet
hodsar
einadequat
efort
hesol
uti
onofmanyofpr
obl
emswef
ace.

 Resear
chi
srel
ati
vel
ypr
osai
c,r
epet
it
iveandl
acki
ngi
nimpact
.

Basedont
heabov
echar
act
eri
sti
csofar
esear
chi
neducat
ion,i
tisev
identt
hatt
he
knowl
edgegai
nedbyt
her
esear
cheri
sofhi
ghestor
der
;because,
resul
tisnotbasedon
assumpt
ions,bel
i
efsandunt
est
edgener
ali
zat
ions.Fai
rl
yhe/
sheappl
i
est
hesci
ent
if
ic
met
hodofi
nvest
igat
ionwi
tht
ruechar
act
eri
sti
csofsci
ent
ist
.Howev
er,
resear
cher
sar
e
humanbei
ngs;i
tisdi
ff
icul
tfort
hem t
oreal
i
zeal
lst
atedchar
act
eri
sti
csofeducat
ional
r
esear
chcompl
etel
y.

1.
3TheNat
ureofResear
chi
nSpeci
alNeedsEducat
ion

I
nit
ial
l
yther
esear
ch met
hodsempl
oyed i
nthef
iel
dthatwast
o becomespeci
al
educat
ionr
esear
chwer
eder
ivedf
rom medi
cine.Manyoft
heear
lypi
oneer
sinser
vice
f
ori
ndi
vi
dual
s wi
th di
sabi
l
iti
es (
not
ethatI
tar
d,Segui
n,Mont
essor
i,Fer
nal
d,and
Gol
dst
ein)wer
ephy
sici
ans.Si
mil
arl
y,ear
lyser
vicesf
ori
ndi
vi
dual
swi
thdi
sabi
l
iti
es
occur
redi
nresi
dent
ialf
aci
l
iti
esandt
rai
ningschool
s,whi
chwer
ebasedont
hemedi
cal
t
radi
ti
onofcar
e(Medi
cal
Model
).
3|
Page
As Psy
chol
ogy
,Soci
ology
,and Ant
hropol
ogy become academi
c di
sci
pli
nes,t
hey
pr
ovi
dedmet
hodol
ogi
calt
ool
sforr
esear
chi
nSpeci
aleducat
ion.Asi
ndi
cat
edabov
e,
muchofSpeci
alEducat
ionr
esear
chder
ivedf
rom adef
ici
tper
spect
ive-t
hatl
ocat
edt
he
pr
obl
em i
nani
ndi
vi
dualandf
ocusedonhi
s/herdi
sabi
l
ityast
her
easont
hathe/
she
coul
d notper
for
m cer
tai
nfunct
ionsoract
ivi
ti
es.Mor
erecent
ly,speci
aleducat
ion
r
esear
cher
shav
eshi
ft
edt
oasoci
o-cul
tur
alper
spect
ive-t
hatf
ocusesont
hedy
nami
c
i
nter
act
ionbet
weent
hei
ndi
vi
dualandt
heenv
ironmentov
ert
hel
i
fespan.Mor
ethan
oner
esear
chmet
hodol
ogyi
simpor
tanti
nSpeci
alEducat
ionr
esear
ch.Ther
ati
onal
efor
mul
ti
ple sci
ent
if
icr
esear
ch met
hodol
ogi
esi
sbased on t
he compl
exi
tyofspeci
al
educat
ionasaf
iel
d&Educat
ional
Cont
ext
.

Compl
exi
tyofSpeci
alEducat
ionasaf
iel
d

Speci
alEducat
ionr
esear
ch,becauseofi
tscompl
exi
ty,maybet
hehar
destoft
he
har
destt
odosci
ence.

1.Var
iabi
l
ityofPar
ti
cipant
s

(
IDEA) i
dent
if
ies mor
ethan 12 el
i
gibi
l
ity
/di
sabi
l
ity cat
egor
ies and wi
thi
nthese
cat
egor
iesar
esev
eral
dif
fer
enti
dent
if
iabl
econdi
ti
ons.

2.Educat
ional
Cont
ext

Speci
alEducat
ionext
endsbey
ondt
het
radi
ti
onalconcept
ual
i
zat
ionof“
School
i
ng”f
or
t
ypi
calst
udent
s.Cur
rent
lymanyst
udent
swi
thdi
sabi
l
iti
esat
tendgener
aleducat
ion
cl
asses.Howev
er,
thecont
inuum ofspeci
aleducat
ioncont
ext
sisbr
oadert
hangener
al
educat
ion.Gi
veni
nav
ari
etyofset
ti
ngsot
hert
hanr
egul
arcl
assesHomeSpeci
al
educat
ioncl
assesoracombi
nat
ionofspeci
alcl
assesandgener
aleducat
ioncl
asses
andi
ncommuni
tyl
i
vingorv
ocat
ionalset
ti
ngs.Anewpar
adi
gm t
hatshi
ft
sthel
ocat
ion
oft
hepr
obl
em f
rom wi
thi
nthei
ndi
vi
dualt
otheenv
ironment
alr
esponset
othedi
sabi
l
ity
andway
stomodi
fyt
heenv
ironmentt
oremov
ebar
ri
ersandi
ncr
easet
hepr
obabi
l
ityof
access.Thi
svi
ewgui
deanddi
rectt
hinki
ng,
resear
ch,
theor
yandpr
act
icei
nthef
iel
dof
speci
alneedseducat
ion.Thehet
erogenei
tyofpar
ti
cipantchar
act
eri
sti
csposesa
si
gni
fi
cantchal
l
enget
oresear
chdesi
gns.Fur
ther
mor
e,l
i
keot
hersoci
alsci
encest
he
f
oll
owi
ngf
act
orsmayal
soaf
fectr
esear
chi
nspeci
alneedseducat
ion.Thesear
e:-

Subj
ect
ivi
ty:-aswest
udyhumanbehav
ior
,obser
vingandst
udy
ingbehav
iori
nnat
ural
cont
ext
.

Di
ff
icul
ti
esi
nrepl
i
cat
ionofbehav
ior
:-behav
ior
smaynotbeobser
vedaccur
atel
yto
r
epl
i
cat
einanot
herset
ti
ng.
4|
Page
Measur
ementPr
obl
em:-t
het
ool
sthatweusei
nsoci
alsci
encesar
elessaccur
atet
han
phy
sicalsci
ences.Ther
efor
e,mor
ethanoner
esear
chmet
hodol
ogyi
simpor
tanti
n
speci
alneedseducat
ionr
esear
ch.

1.
4Met
hodsofAcqui
ri
ngKnowl
edge

Koul(
1984)cat
egor
izet
hemet
hodsbywhi
chmanf
rom t
heear
li
ert
imessought
answer
stohi
spr
obl
emsasf
oll
ows:
Appeal
toaut
hor
it
y,t
radi
ti
on,
exper
ience,
deduct
ive
r
easoni
ng,
induct
iver
easoni
ng,
andsci
ent
if
icmet
hod.

Kahn(
2004)al
socat
egor
izedmet
hodst
oobt
ainknowl
edgei
nthebook“
Educat
ional
Resear
ch”t
hati
ncl
udecommonsense,appealt
oaut
hor
it
y,i
ntui
ti
on,r
evel
ati
onand
r
ati
onal
i
sm.Mostofar
esi
mil
art
othef
ormercl
assi
fi
cat
ionofKoul(
1984)
.Int
uit
ive
bel
i
efsar
e“hunches”or“
insi
ght
s”i
ntot
her
elat
ionshi
psbet
ween v
ari
abl
eswhi
ch
suggestcour
seofact
ionandwhi
char
eder
ivedf
rom one’
spastexper
ienceswi
th
si
mil
arphenomena.I
fthesear
etest
ed empi
ri
cal
l
ytheymaybef
ound cor
rector
i
ncor
rect
.Ift
heyar
efoundi
ncor
rectaf
terempi
ri
calt
est
,theymaynotbeaccept
edas
knowl
edge.Howev
er,r
evel
ati
onsmaybeconsi
der
edast
rue;t
hesour
ceasout
side
humanexper
ienceandt
hesecannotbeputt
oempi
ri
cal
test
.

The conceptof“
rat
ional
i
sm “
can be consi
der
ed as r
easonabl
e way ofder
ivi
ng
knowl
edge(
Deduct
iver
easoni
ng)
.Ther
efor
e,t
hemet
hodsofacqui
ri
ng knowl
edge
cat
egor
izedbyKoul
(1984)ar
emuchpr
efer
abl
etobedescr
ibednext
.

1.
4.1Aut
hor
it
y(Appeal
toaut
hor
it
y)

Theaut
hor
it
iescanbedef
inedasper
sonswho hav
ehadl
ongexper
iencewi
tha
pr
obl
em andwhohav
est
udi
edandt
houghtmuchaboutapar
ti
cul
arpr
obl
em.I
ngener
al,
i
fanyv
iew i
spr
esent
edbysomeoneaboutanyi
ssueandi
fiti
ssuppor
tedbysome
knownaut
hor
it
yineducat
ion(
C.VGood,M.
W.Tr
aver
s…)i
tisaccept
ed.Thef
oll
owi
ng
cr
it
eri
aar
eimpor
tantt
odeci
dewhet
hert
heper
sonmaybeconsi
der
edanaut
hor
it
yor
not
.Thesear
e:

I
.Thei
ndi
vi
dual
whoi
sbei
ngj
udgedasanaut
hor
it
yshoul
dbei
dent
if
iabl
e.

I
I.Theper
soni
srecogni
zedasanaut
hor
it
ybyt
hemember
soft
hepr
ofessi
onal
concer
ned.

I
II
.Theper
sonshoul
dbel
i
vingorhi
spoi
ntofv
iewshoul
dnothav
ebeenr
eject
edaf
ter
deat
h.

5|
Page
I
V.Theper
sonshoul
dnotbepr
ejudi
cedorbi
ased.

1.
4.2 Tr
adi
ti
on

Humanbei
ngssol
vemanyoft
hei
rpr
obl
emsusi
ngt
radi
ti
onalappr
oach.Theymay
acceptmanyt
radi
ti
onsoff
oref
ather
sorcul
tur
esunquest
ioni
ngl
y.I
nschool
s,t
eacher
s
of
tenr
elyont
radi
ti
onorexper
iencesasadependabl
egui
de.Al
thoughaccept
anceof
t
radi
ti
oni
sof
tennecessar
y,oneshoul
dnotal
way
sassumet
hatev
ery
thi
ngt
hathas
cust
omar
il
ybeendonei
sri
ghtandv
ali
d.

1.
4.3 Exper
ience

Per
sonalexper
ienceort
hatofot
her
sist
hemostpr
imi
ti
veandy
ett
hemostf
ami
l
iar
it
y
andf
undament
alsour
cesofknowl
edge.Anci
entpeopl
e(nomadsandv
ari
oust
ri
bes)
f
rom t
hei
rper
sonalexper
iencepr
obabl
yremember
edt
hatcer
tai
nwi
l
dfr
uit
sal
way
s
madet
hem i
l
l,t
hatgr
ainsmat
uredatpar
ti
cul
art
imesoft
hey
ear
,andt
hatsudden
f
loodsi
nther
iver
sdur
ingt
her
aini
ngseasonwer
eduet
othef
actt
hatwat
erdoesnot
gener
all
yst
ayonhi
l
ls.Whenmoder
nmanconf
ront
edwi
thapr
obl
em heof
tent
ri
est
o
seeki
tsanswer
sfr
om hi
sownper
sonalexper
ienceorf
rom ot
her
swhoar
efami
l
iar
wi
tht
hepr
obl
em.Pupi
l
sof
ten consul
tthei
rteacher
s,par
ent
sorev
en t
hei
rel
der
si
bli
ngst
oder
iveanswer
stot
hei
rquest
ions.Howev
er,per
sonalexper
ienceort
hatof
ot
her
sisausef
ulmet
hodt
oobt
ainknowl
edge;si
ncei
tisuncr
it
icalwemayl
eadt
o
i
ncor
rectconcl
usi
ons.

VanDal
en(
1973)i
nKoul(
1984)r
emar
kedf
ivepoi
ntshowaper
sonmaymakeer
ror
s
whenobser
vingorwhenr
epor
ti
ngwhathehasseenordone.Thesear
e;hemay
:

1.Omi
tev
idencet
hatdoesnotagr
eewi
thhi
sopi
nion.

2.Usemeasur
ingi
nst
rument
sthatr
equi
ressubj
ect
iveest
imat
es.

3.Est
abl
i
shabel
i
efoni
nsuf
fi
cientev
idence.

4.Fai
ltoobser
vesi
gni
fi
cantf
act
orsr
elat
ingt
oaspeci
fi
csi
tuat
ionor

5.Dr
aw i
mpr
operconcl
usi
onsori
nfer
encesowi
ngt
oper
sonalpr
ejudi
ces.Basedon
t
hement
ionedr
emar
ksoneshoul
dcaut
iousl
yandcr
it
ical
l
yuseexper
ienceasan
oppor
tuni
tyf
orobt
aini
ngr
eli
abl
eknowl
edge.

1.
4.4Deduct
iveReasoni
ng

TheGr
eekphi
l
osopherl
i
keAr
ist
otl
eandhi
sfol
l
ower
swer
eknownasasi
gni
fi
cant

6|
Page
cont
ri
but
ort
owar
dst
hedev
elopmentofasy
stemat
icmet
hodf
orobt
aini
ngr
eli
abl
e
knowl
edge.Ar
ist
otl
edev
elopedt
hesy
ll
ogi
sm,whi
chcanbedescr
ibedasat
hinki
ng
pr
ocessi
nwhi
chonepr
oceedsf
rom gener
alt
o speci
fi
cst
atement
sbydeduct
ive
r
easoni
ng.I
tpr
ovi
desameansoft
est
ingt
hev
ali
dit
yofanygi
venconcl
usi
onori
deaby
pr
oceedi
ngf
rom t
heknownt
otheunknown.Thesy
ll
ogi
sti
creasoni
ngconsi
stsof
:

 A maj
orpr
emi
sebasedonasel
f-
evi
dentt
rut
horpr
evi
ousl
yest
abl
i
shedf
actor
r
elat
ionshi
p.

 Ami
norpr
emi
seconcer
ningapar
ti
cul
arcaset
owhi
cht
het
rut
hfactorr
elat
ionshi
p
appl
i
esand

 A concl
usi
ons;i
famaj
orandmi
norpr
emi
sescanbeshownt
obet
ruei
fthe
concl
usi
on

 Ar
ri
vedati
snecessar
il
ytr
ue.

Exampl
e:

1.Al
lani
mal
sar
emor
tal
.

2.Li
oni
sanani
mal
.

3.Ther
efor
e,l
i
onwi
l
ldi
e.

Nev
ert
hel
esst
hemet
hodofsy
ll
ogi
sm ordeduct
ioni
susef
ul,i
thast
hef
oll
owi
ng
l
i
mit
ati
on

 Theconcl
usi
onsofasy
ll
ogi
sm ar
eal
way
sder
ivedf
rom t
hecont
entofpr
emi
ses.
Ther
efor
e,i
fthe

 Pr
emi
sesar
eunr
elat
edori
foneoft
hepr
emi
sesi
ser
roneoust
heconcl
usi
onar
ri
ved
atwi
l
lnotbev
ali
d.

 Thedependenceofdeduct
iver
easoni
nguponv
erbal
symbol
i
sm.

 Deduct
iver
easoni
ngcansy
stemat
izewhati
sal
readyknownandcani
dent
if
ynew
r
elat
ionshi
psasonepr
oceedsf
rom knownt
ounknown,buti
tcan’
tber
eli
edupon
asasel
f-
suf
fi
ci
entmet
hodf
orsecur
ingr
eli
abl
eknowl
edge.

1.
4.5I
nduct
iveReasoni
ng

The concl
usi
ons der
ived f
rom gener
ali
ti
es and f
rom st
atement
s of pr
esumed
aut
hor
it
iesbydeduct
iver
easoni
ngar
etr
ueonl
yift
heyar
ebasedupont
ruepr
emi
ses.
7|
Page
To det
ermi
newhet
hert
hepr
emi
sesar
etr
ue,t
her
eisaneed f
orbi
asi
ng gener
al
concl
usi
onsuponspeci
fi
cfact
sgat
her
edt
hroughdi
rectobser
vat
ionswhi
chi
sknown
as i
nduct
iver
easoni
ng.I
tis a goi
ng f
rom t
he par
ti
cul
arobser
vat
ions t
othe
gener
ali
zat
ion.

1.
4.6TheSci
ent
if
icMet
hod

Thesci
ent
if
icmet
hodi
samet
hodt
hati
susedt
oacqui
readv
ancedhumanknowl
edge
by t
he combi
ned ef
fect
s ofi
nduct
ive-
deduct
ive met
hod.I
tis a back and f
ort
h
mov
ementoft
houghti
nwhi
chmanf
ir
stoper
atesi
nduct
ivel
yfr
om pr
act
ical
l
yknownor
somet
imes conf
used i
nfor
mat
ion l
ear
ned f
rom exper
ience,pr
evi
ous knowl
edge,
r
efl
ect
ivet
hinki
ng,obser
vat
ionandsoon,t
owar
dsameani
ngf
ulwhol
eorhy
pot
hesi
s
andt
hendeduct
ivel
yfr
om suggest
edwhol
eorhy
pot
hesi
stot
hepar
ti
cul
arpar
tsi
n
or
dert
o connectt
hese wi
th one anot
heri
n a meani
ngf
ulpat
ter
ntof
ind v
ali
d
r
elat
ionshi
ps.I
ntheot
herway
,sci
ent
if
icmet
hodwor
ksi
nduct
ivel
yfr
om obser
vat
ions
t
ohy
pot
hesesandt
hendeduct
ivel
yfr
om t
hehy
pot
hesest
othel
ogi
cali
mpl
i
cat
ionsof
t
hehy
pot
hesesi
nrel
ati
ont
owhati
sal
readyknown.

St
epsi
nSci
ent
if
icMet
hod
St
epsi
nsci
ent
if
icmet
hodconsi
stoff
iveel
ement
s:

1.
Ident
if
icat
ionanddef
ini
ti
onoft
hepr
obl
em.
2.
For
mul
ati
onofahy
pot
hesi
s.
3.
Impl
i
cat
ionofhy
pot
hesi
sthr
oughdeduct
iver
easoni
ng.
4.
Col
l
ect
ionandanal
ysi
sofev
idence.
5.
Ver
if
icat
ion,
rej
ect
ionormodi
fi
cat
ionofhy
pot
hesi
s.

1.
5Cl
assi
fi
cat
ionofResear
ch

I
ndeal
i
ngwi
thr
esear
chdesi
gnandpl
anni
ngpr
ocess,t
her
esear
cherhast
odeci
de
aboutr
esear
chmet
hodt
hatshe/
hecoul
dusei
nsol
vi
nghi
s/her
sresear
chpr
obl
em.
Resear
chmet
hodsar
eofut
mosti
mpor
tanti
nar
esear
chpr
ocess.Theydescr
ibet
he
v
ari
ousst
epsoft
hepl
anofat
tackt
obeadopt
edi
nsol
vi
ngr
esear
chpr
obl
em suchas
t
hemanneri
nwhi
cht
hepr
obl
emsar
efor
mul
ated,
thedef
ini
ti
onoft
erms,t
hechoi
ceof
subj
ect
sfori
nvest
igat
ion,
thev
ali
dat
ionofdat
agat
her
ingt
ool
s,t
hecol
l
ect
ion,
anal
ysi
s
andi
nter
pret
ati
onofdat
aandt
hepr
ocessofi
nfer
encesandgener
ali
zat
ions.

[
Accor
dingt
oGood,Bar
randscat
es(
1941)i
nKoul(
1984)
,resear
chmet
hodmaybe
cl
assi
fi
edf
rom t
hef
oll
owi
ngpoi
ntsofv
ieworont
hebasi
sof
:

1.Thef
iel
dtowhi
chi
tappl
i
es:
Educat
ion,
hist
ory
,phi
l
osophy
,psy
chol
ogy
,bi
ology
,et
c.

8|
Page
2.Pur
pose:
descr
ipt
ion,
predi
cti
on,
det
ermi
nat
ionofcauses,
det
ermi
nat
ionofst
atus,
et
c.
3.Appl
i
cat
ion:
pur
eresear
ch,
appl
i
edr
esear
choract
ionr
esear
ch.
4.Pl
acewher
eiti
sconduct
ed:
Resear
chi
nthef
iel
d,i
nthel
abor
ator
y.
5.Dat
agat
her
ingdev
icesempl
oyed:
test
s,r
ati
ngscal
es,
quest
ionnai
reset
c.
6.Nat
ureoft
hedat
acol
l
ect
ed:
obj
ect
ive,
subj
ect
ive,
quant
it
ati
ve,
qual
i
tat
ive,
etc.
7.Sy
mbol
sempl
oyedi
nreadi
ng,
descr
ibi
ng,
ort
reat
ingr
esul
ts:
mat
hemat
ical
symbol
s
orl
anguagesy
mbol
s,
8.For
msoft
hinki
ng:
deduct
ive,
induct
ive,
etc
9.Cont
rol
off
act
ors:
cont
rol
l
edanduncont
rol
l
edexper
iment
ati
on
10. Met
hodsempl
oyedi
nest
abl
i
shi
ngcasual
rel
ati
onshi
ps:
agr
eement
,di
ff
erence,
r
esi
duesandconcomi
tantv
ari
ati
on.Cl
assi
fi
cat
ionsoft
her
esear
chmet
hodsi
nthe
abov
emannergi
ver
iset
ocompr
ehensi
vel
i
stwi
thmuchofov
erl
appi
ngamongt
he
di
ff
erentmet
hods.Yetdespi
tet
hisl
ackofcl
earcutdi
sti
nct
ionamongt
hemet
hods,
mostaut
hor
sagr
eeont
hreebasi
ccat
egor
iessuchashi
stor
ical
,descr
ipt
ive,
and
exper
iment
almet
hod.

9|
Page
1.
6.1Cl
assi
fi
cat
ionofResear
chbyMet
hods
I
.Hi
stor
ical
Met
hod
Hi
stor
icalmet
hodi
sconcer
nedwi
tht
hepastandi
tat
tempt
sar
emadet
ofi
nd
outt
hepasti
ntheper
spect
iveoft
hepr
esent
.It
spr
ocessi
nvol
vesi
nvest
igat
ing,
r
ecor
ding,anal
yzi
ngandi
nter
pret
ingt
heev
ent
soft
hepastf
ordi
scov
eri
ng
gener
ali
zat
ion.Thehi
stor
icalr
esear
chmet
hodcanbest
udi
edunderhi
stor
ical
,
l
ogi
cal
anddocument
arycat
egor
ies.

I
I.Descr
ipt
iveSur
veyMet
hod

Descr
ipt
ivesur
veymet
hodi
sconcer
nedwi
tht
hepr
esentandi
nitat
tempt
sar
e
made t
ofi
nd outt
he pr
esentposi
ti
on oft
he phenomena whi
ch i
s bei
ng
i
nvest
igat
ed. I
ts pr
ocess i
nvol
ves descr
ipt
ion,r
ecor
ding,anal
yzi
ng and
i
nter
pret
ing condi
ti
ons t
hatnow exi
st.I
tof
ten i
nvol
ves some t
ypes of
compar
ison orcont
rast
sand mayat
temptt
o di
scov
eracauseand ef
fect
r
elat
ionshi
p.Thedescr
ipt
ivesur
veymet
hodscanbest
udi
edunderdescr
ipt
ive,
anal
yti
cal
,school
andgenet
icsur
vey
.

I
II
.Exper
iment
alMet
hod

Ex
per
iment
almet
hodi
sor
ient
edt dswhatwi
owar l
lbewhenat
tempt
sar
emade
t
ofi
ndoutcauseandef
fectr
elat
ionshi
p.Theexper
iment
almet
hodcanbe
st
udi
edunderpr
eexper
iment
aldesi
gns,t
rueexper
iment
aldesi
gns,f
act
ori
al
desi
gns,
quasi
-exper
iment
aldesi
gnsandt
imeser
iesdesi
gns.

1.
6.2Resear
chCl
assi
fi
cat
ionbyPur
posesorObj
ect
ives

The cl
assi
fi
cat
ion ofr
esear
ch based on t
he obj
ect
ives orpur
poses i
n whi
ch t
he
r
esear
cheri
ntendst
oaccompl
i
shar
ebasi
corf
undament
alr
esear
ch,appl
i
edr
esear
ch
andact
ionr
esear
ch.

I
.Basi
corFundament
alResear
ch
Accor
dingt
oTr
ver
s(1948)i
nKoul(
1984)
,basi
cresear
chi
sdesi
gnedt
oaddt
oan
or
gani
zedbodyofsci
ent
if
icknowl
edgeanddoesnotnecessar
il
ypr
oducer
esul
tsof
i
mmedi
atepr
act
icalv
alue.BestandKahn(
2003)def
inet
hef
undament
alr
esear
chas
t
hef
ormalandsy
stemat
icpr
ocessofdeduct
ive-
induct
iveanal
ysi
sandl
eadi
ngt
othe
dev
elopmentoft
heor
ies.Thi
sty
pe ofr
esear
ch i
s pr
imar
il
y concer
ned wi
tht
he
f
ormul
ati
onofat
heor
yoracont
ri
but
iont
otheexi
sti
ngbodyofknowl
edge.Themai
n
ai
m off
undament
alorbasi
cresear
chi
sto obt
ainand uset
heempi
ri
caldat
ato
f
ormul
ate,ex
pandorev
aluat
etheor
y.I
tisusedt
othedi
scov
eryofknowl
edgeandhas
10|
Page
l
i
ttl
econcer
noft
heappl
i
cat
ionoft
hef
indi
ngsorsoci
alusef
ulnessoft
hef
indi
ngs.
[
[

I
I.Appl
i
edResear
ch
Appl
i
edr
esear
chi
sunder
takent
osol
veani
mmedi
atepr
act
ical
probl
em andt
hegoal
of
addi
ngt
osci
ent
if
icknowl
edgei
ssecondar
y.I
tisdi
rect
edt
owar
dst
hesol
uti
onof
i
mmedi
ate,speci
fi
c and pr
act
icalpr
obl
ems.I
tis per
for
med i
nrel
ati
on t
o act
ual
pr
obl
emsandundert
hecondi
ti
onsi
nwhi
cht
heyar
efoundi
npr
act
ice.Appl
i
edr
esear
ch
hasmostt
hechar
act
eri
sti
csoff
undament
alr
esear
chi
ncl
udi
ngt
heuseofsampl
i
ng
t
echni
quesandt
hesubsequenti
nfer
encesaboutt
het
argetpopul
ati
on.Howev
er,i
ts
pur
posei
simpr
ovi
ngapr
oductorapr
ocesst
est
ingt
heor
eti
calconcept
sinact
ual
pr
obl
em si
tuat
ions(
Best& Kahn,2003)
.Appl
i
edr
esear
chi
sbasi
cal
l
yint
endedt
o
i
mpr
oveschoolpr
act
icesand t
o add gr
eat
ert
eacheref
fect
ivenessi
n apar
ti
cul
ar
manner
.Mostoft
hepr
obl
emsf
acedbyt
eacher
s,super
visor
s,pol
i
cymaker
s,pr
inci
pal
s,
and educat
ionalpl
anner
s and manager
s ar
e sol
ved t
hrough appl
i
ed r
esear
ch i
n
educat
ions.

I
II
.Act
ionResear
ch

Act
ionr
esear
chi
sat
ypeofappl
i
edordeci
sionor
ient
edr
esear
chbutwi
tht
he
st
ipul
ati
ont
hatt
her
esear
cheri
sthesameper
sonast
hepr
act
it
ionerwhowi
l
lmake
andl
i
vewi
tht
hedeci
si
on.Thi
sty
peofr
esear
chmost
lyconduct
edbyt
het
eacher
s,
super
visor
sandpr
inci
pal
sfordeal
i
ngwi
thcl
assr
oom pr
obl
emsandi
mpr
ovi
ng
cl
assr
oom pr
act
ices.Act
ionr
esear
chf
ocusesoni
mmedi
ateappl
i
cat
ion,
notont
he
dev
elopmentoft
heor
yorongener
ali
zat
ionofappl
i
cat
ions.I
tismost
lyemphasi
s
onpr
obl
emsi
nlocalset
ti
ngs.I
tsf
indi
ngsar
etobeev
aluat
edi
nter
msofl
ocal
appl
i
cabi
l
itynotuni
ver
salv
ali
dit
y.Thepur
poseofact
ionr
esear
chi
stoi
mpr
ove
schoolpr
act
icesandatt
hesamet
imet
oimpr
ovet
hosewhot
ryt
oimpr
ovet
he
pr
act
icessuchast
ocombi
net
her
esear
chpr
ocesses,habi
tsoft
hinki
ng,abi
l
ityt
o
wor
khar
moni
ousl
ywi
thot
her
sandpr
ofessi
onal
spi
ri
t(Best&Kahn,
2003).

Accor
dingt
oZei
chaerandNof
fkei
nGal
l
,etal(
2003)t
hepur
posesofact
ionr
esear
ch
hasbeenl
ookedbyt
hreedi
mensi
ons.Thesear
eper
sonal
,pr
ofessi
onal
,andpol
i
tical
pur
poses.

I
.Per
sonal
Pur
posesf
orAct
ionResear
ch:
Act
ionr
esear
chdonef
orper
sonal
pur
poses

i
sint
endedt
opr
omot
egr
eat
ersel
f-
knowl
edgef
ulf
il
lmentandpr
ofessi
onal
awar
eness

11|
Page
amongpr
act
it
ioner
s.

I
I.Pr
ofessi
onal
Pur
posesf
orAct
ionResear
ch:
Act
ionr
esear
chunder
takenf
or

pr
ofessi
onal
:

Purposes t
ypicall
yi nvolvet eachers’ef fort
st o extend theirresearch bey ond the
cl
assroom.Thi st ypeofact i
onr esearchcont r
ibutest obridget hegapt hatexists
between educationalt heory and educat i
onalpr acti
ce butwi thoutf undament all
y
al
teri
ngtherel
ationshipbet weent her esearchandpr act
icecommuni t
ies.

I
I
I.Pol
i
tical
Pur
posesf
orAct
ionResear
ch:Act
ionr
esear
chcanexpl
i
cit
lyaddr
ess
pol
i
tical
Pur
posesby
:

 Seeki
ngt
omakeone’
sownt
eachi
ngpr
act
icesmor
ehumanandj
ust
;
 Pr
ovi
dingf
ull
par
ti
ci
pat
ioni
nther
esear
chpr
ocessofal
lthosewhoar
eaf
fect
edbyi
t;
 Embr
aci
nganov
ertagendaofsoci
alchangewi
thacommi
tmentt
opr
omot
e
economi
c
andsoci
alj
ust
icet
hroughcol
l
abor
ati
veef
for
tst
oincr
easeeducat
ionaloppor
tuni
ti
es
andout
comesf
oral
lconst
it
uent
s.

I
ngener
al,t
heschoolsuper
visor
sandpr
inci
pal
smayconsi
dert
het
hreement
ioned
pur
posesofact
ion r
esear
ch (
per
sonal
,pr
ofessi
onaland pol
i
tical
)to car
ryoutt
he
r
esear
chpr
ojectt
osol
vel
ocalschoolpr
obl
ems.Theact
ionr
esear
cher
smayal
sogai
n
gr
eat
erknowl
edgeandr
esear
chex
per
iences.Someoft
hem maynar
rowt
hei
rpur
poses
whi
l
eot
her
smaybr
oadent
hei
rfocust
oaddr
essv
ari
edpur
poses.Deart
rai
nee,i
fyou
pl
ant
ocar
ryoutanact
ionr
esear
chpr
oposal
,wesuggestt
haty
oushoul
dref
lecton
y
ourownpur
posesear
lyi
nyourpl
anni
ng.Fur
ther
,thedet
ail
sofact
ionr
esear
chwi
l
lbe
caughtupi
nthecour
se“
Act
ionResear
chf
orEducat
ional
Manager
s”(
EdPM -212)

1.
6.3.Cl
assi
fi
cat
ionofResear
chbyTy
pesofDat
a

I
namor egeneralf
orm,resear
chdatacanbecl
assif
iedi
ntoqual
i
tati
veandquanti
tat
ive.
Thesetwotypesofdat
aar eusef
ult
oclassi
fyr
esearchi
ntot
wobroadcategor
ies.
Theseare:

I
. Quant
it
ati
ver
esear
ch
I
I.Qual
i
tat
iver
esear
ch
1.
6.3.
1.Quant
it
ati
veResear
ch
12|
Page
Quant
it
ati
ver
esear
chconsi
stsofr
esear
chi
nwhi
cht
hedat
acanbeanal
yzedi
nter
msof
number
s. A numer
ical met
hod of descr
ibi
ng obser
vat
ions of mat
eri
als or
char
act
eri
sti
csi
sknownasquant
if
icat
ion.Whenadef
inedpor
ti
onoft
hemat
eri
alor
char
act
eri
sti
cisusedasast
andar
dformeasur
inganysampl
e,av
ali
dandpr
eci
se
met
hodofdat
adescr
ipt
ioni
spr
ovi
ded.Ther
ear
efourl
evel
sofmeasur
ementscal
es.
Thesear
enomi
nal
scal
e,or
dinal
scal
e,i
nter
val
scal
eandr
ati
onscal
e.

1.A nomi
nalscal
e:i
sthe l
eastpr
eci
se met
hod ofquant
if
icat
ion and descr
ibes
di
ff
erencesbet
weent
hingsbyassi
gni
ngt
hem t
ocat
egor
ies.Nomi
naldat
aar
ecount
ed
dat
a.I
nthi
sscal
ecount
ingt
heonl
yfeasi
blemet
hodofquant
if
icat
ionandmaypr
ovi
de
an accept
abl
e basi
sforst
ati
sti
calanal
ysi
s.I
tis a non–or
der
abl
e quant
if
icat
ion.
Ex
ampl
e:Numberoff
emal
est
udent
sint
hecl
assr
oom,numberoff
ami
l
iesv
isi
tedt
he
school
,numberofr
esponsesi
nthecat
egor
yofy
es/
no,
li
ke/
disl
i
ke…

2.An or
dinalscal
e:a measur
ein whi
ch a numberr
epr
esenta r
ankor
der
ing of
i
ndi
vi
dual
sor
obj
ect
sonsomev
ari
abl
e.Thi
sscal
ehasnoabsol
utev
aluesandt
her
ealdi
ff
erences
bet
ween
adj
acentr
anksmaynotbeequal
.Exampl
e:1st, 3rd,
2nd, …

3.Ani
nter
valscal
e:ameasur
ethatl
acksat
ruezer
opoi
ntandf
orwhi
cht
hedi
stance
bet
ween any t
wo adj
acent poi
ntsi
sthe same.Exampl
e:t
her
momet
er scal
e,
psy
chol
ogi
calt
estandi
nvent
ori
esar
eint
erv
alscal
esandhav
ethi
sli
mit
ati
onal
though
t
heycanbeadded,
subt
ract
ed,
mul
ti
pli
edanddi
vi
ded.

4.A r
ati
oscal
e:ameasur
ethathasat
ruezer
opoi
ntandf
orwhi
cht
hedi
stance
bet
weenanyt
woadj
acentpoi
ntsi
sthesame.Exampl
e:Thezer
opoi
ntonacent
imet
er
scal
eindi
cat
est
hecompl
eteabsenceofl
engt
horhei
ght
.

I
nquant
it
ati
ver
esear
ch,
thesescal
espr
ovi
der
elev
anti
nfor
mat
iont
otestort
oinf
ert
he
hy
pot
hesesst
ati
sti
cal
l
y.I
ngener
al,quant
it
ati
ver
esear
chi
sani
nqui
ryt
hati
sgr
ounded
ont
heassumpt
iont
hatf
eat
uresoft
hesoci
alenv
ironmentconst
it
uteanobj
ect
ive
r
eal
i
tyt
hati
srel
ati
vel
yconst
antacr
osst
imeandset
ti
ngs.Thedomi
nantmet
hodol
ogy
i
stodescr
ibeandexpl
ainf
eat
uresoft
hisr
eal
i
tybycol
l
ect
ingnumer
icaldat
aon
obser
vabl
ebehav
ior
sofsampl
esandbysubj
ect
ingt
hesedat
atost
ati
sti
cal
anal
ysi
s.

Ther
ear
etwot
ypesofdat
aint
heappl
i
cat
ionofst
ati
sti
cal
treat
ment
.Thesear
e:

.Par
1 amet
ri
cdat
a:Thi
sty
peofdat
aisameasur
eddat
a,andpar
amet
ri
cst
ati
sti
cal
13|
Page
t
est
sassumet
hatt
hedat
aar
enor
mal
l
y,ornear
lynor
mal
l
y,di
str
ibut
ed.Par
amet
ri
c
t
est
sar
eappl
i
edt
obot
hint
erv
al-
andr
ati
o-scal
eddat
a.

2.Non-
par
amet
ri
cdat
a:Theset
ypesofdat
aar
eei
thercount
ed(
nomi
nal
)orr
anked
(
ordi
nal
).Nonpar
amet
ri
ctest
s,somet
imesknownasdi
str
ibut
ion-f
reet
est
s,donotr
est
ont
hemor
eri
gor
ousassumpt
ionofnor
mal
l
ydi
str
ibut
edpopul
ati
ons.

1.
6.3.
2Qual
i
tat
iveResear
ch

Qual
i
tat
iver
esear
chi
sar
esear
chappr
oachi
nsoci
alsci
enceandhumani
ti
est
hat
emphasi
zescol
l
ect
ingdescr
ipt
ivedat
ainnat
uralset
ti
ngs,usesi
nduct
ivet
hinki
ngand
emphasi
zesunder
standi
ngt
hesubj
ect
s’poi
ntsofv
iew.Accor
dingt
oPat
ton(
1990)i
n
Best& Kahn (
2003)qual
i
tat
iver
esear
ch met
hods consi
stoft
hree ki
nds ofdat
a
col
l
ect
ion.Thesear
ein-
dept
hopenendedi
nter
views,di
rectobser
vat
ionandwr
it
ten
document
s.

The dat
afr
om i
nter
view consi
stofdi
rectquot
ati
ons f
rom peopl
e aboutt
hei
r
exper
iences,
opi
nions,
feel
i
ngsandknowl
edge.Thedat
afr
om obser
vat
ionsconsi
stsof
det
ail
eddescr
ipt
ionsofpeopl
e’
sact
ivi
ti
es,act
ionsandt
hef
ullr
angeofi
nter
per
sonal
i
nter
act
ions and or
gani
zat
ionalpr
ocesses t
hat ar
e par
t of obser
vabl
e human
exper
ience.Documentanal
ysi
sinqual
i
tat
ivei
nqui
ryy
iel
dsexcer
ptquot
ati
onsorent
ir
e
passagesf
rom or
gani
zat
ional
,cl
i
nicalorpr
ogr
am r
ecor
ds,
repor
ts,
per
sonaldi
ari
esand
openendedwr
it
tenr
esponsest
oquest
ionnai
resandsur
vey
s.Qual
i
tat
ivest
udi
esuse
t
heabov
ement
ioneddat
acol
l
ect
ingt
echni
quesi
nisol
ati
onori
ncombi
nat
ion.

Fi
nall
y,i
tisadv
isabl
etot
rai
neest
ofocusont
hef
oll
owi
ngqual
i
tat
iver
esear
ch
assumpti
ons:

 i
tisdescr
ipt
ive

 i
tinv
olv
esf
iel
dwor
k;
 i
tisconcer
nedpr
imar
il
ywi
thpr
ocessr
athert
hanout
comesorpr
oduct
s;
 i
tisi
nduct
ivei
nthe r
esear
cher
s bui
l
d abst
ract
ions,concept
s,t
heor
ies and
hy
pot
hesesf
rom det
ail
s;
 t
her
esear
cheri
sthepr
imar
yinst
rumentf
ordat
acol
l
ect
ionandanal
ysi
sand
 i
tis pr
imar
il
yint
erest
ed i
n meani
ng how peopl
e make sense oft
hei
rli
ves
exper
iencesandt
hei
rst
ruct
uresoft
hewor
ld.

14|
Page
1.
7Resear
chPr
ocess
I
nthepr
evi
oussect
ionsoft
hisuni
twehav
eseent
hef
undament
alconcept
sofr
esear
ch
gener
all
yineducat
ionandpar
ti
cul
arl
yineducat
ionalpl
anni
ngandmanagement
.Nextt
he

Resear
chpr
ocess”wi
l
lbedeal
twi
tht
hef
oll
owi
ngsequent
ial
steps:
I
.Pr
obl
em i
dent
if
icat
ion
I
I.Obj
ect
iveoft
her
esear
ch
I
II
.Resear
chdesi
gn
I
V.Dat
acol
l
ect
ion
V.Dat
aanal
ysi
s
VI
.Int
erpr
etat
ionofr
esul
ts
VI
I.Repor
ti
ngt
her
esul
ts
1.
7.1Pr
obl
em I
dent
if
icat
ion
Ar
esear
chpr
obl
em mustbei
dent
if
iedanddef
inedwi
thoutanyambi
gui
ty.Pr
oduct
ivi
tyi
s
t
hecor
eobj
ect
iveofanyeducat
ionalor
gani
zat
ioni
ncl
udi
ngschool
s.Thi
sisbecauseofi
ts
sur
viv
ali
nthe pr
evai
l
ing compet
it
ive env
ironment
.So,each educat
ionalorschool
or
gani
zat
ionshoul
dat
temptt
oinv
olv
eini
mpr
ovi
ngi
tspr
oduct(
lear
nedst
udent
s)t
hrough
cont
inuousr
esear
chi
neachf
unct
ionalar
eaofeducat
ionalpl
anni
ngandmanagement
.In
anyschoolor
gani
zat
ion,
thebudgetal
l
ocat
ionf
orr
esear
chact
ivi
ti
esi
sli
mit
edornotatal
l
.
Thi
sfor
cet
her
esear
cht
eam t
osel
ectt
hebestcombi
nat
ionofr
esear
chpr
oject
soutof
t
hecompet
ingal
ter
nat
ivesi
ndi
ff
erentf
unct
ionalar
eas.Sot
her
esear
cht
eam shoul
d
i
dent
if
yresear
chpr
oject
swhi
chwi
l
lhav
emor
eval
ueaddi
ti
ont
otheor
gani
zat
ion.Al
so
eachsel
ect
edr
esear
chpr
ojectmustbecl
ear
lydef
ined.
Ther
emaybeasi
tuat
ioni
nwhi
cht
her
esear
cheri
sful
l
yawar
eoft
hesy
mpt
omsr
elat
ingt
o
cer
tai
ndef
ici
encyi
nachi
evi
nganor
gani
zat
ionalgoal
.Howev
er,she/
hemaynotbei
na
posi
ti
ont
ocl
ear
lyspel
loutt
hepr
obl
em whi
chi
scausi
ngsuchdef
ici
ency
.Unl
essi
tis
cl
ear
lyi
dent
if
ied,i
twi
l
lnotbepossi
blet
opr
oceedf
urt
hert
ocar
ryoutt
hepr
oject
.Ifa
r
esear
cherpr
oceeds wi
thi
l
l-def
ined pr
obl
ems she/
he may end up wi
th mi
sleadi
ng
concl
usi
onorabor
ti
ngt
her
esear
chpr
ojecti
nthemi
ddl
eduet
opoorpay
-of
fident
if
ied
t
hroughi
nter
im ev
aluat
ionoft
her
esear
ch.Hencet
her
esear
chpr
obl
em shoul
dbecl
ear
ly
def
ined.

15|
Page
1.
7.2.Obj
ect
iveoft
heResear
ch
Theobj
ect
iveoft
her
esear
chmustbei
dent
if
iedbyt
aki
ngt
hesy
stem r
equi
rement
sint
o
account
.Int
hepr
ocessofi
dent
if
yingt
heobj
ect
ivesoft
her
esear
ch,t
her
esear
chermust
f
inal
i
zet
hef
oll
owi
ngst
eps:
1.Resear
chquest
ionswhi
char
enotr
esol
vedt
il
lthen.
2.Hy
pot
heseswhi
char
eanassumpt
ionaboutapopul
ati
onoft
hest
udy
.
3.Boundar
yoft
hest
udyshoul
dbecl
ear
lydef
ined.
Obj
ect
ivesoft
hest
udymaybecl
assi
fi
edi
ntoquant
it
ati
veandqual
i
tat
ive.Quant
it
ati
ve
obj
ect
ivesai
mtoopt
imi
zecer
tai
nmeasur
eofper
for
manceoft
hesy
stem oft
hest
udy
.
1.
7.3.Resear
chDesi
gn
Thi
spar
toft
her
esear
chpr
ocesspr
ovi
descompl
etegui
del
i
nesf
orcol
l
ect
ingdat
a.The
essencesofr
esear
chdesi
gnar
easf
oll
owsbel
ow:
I
. Sel
ect
ionofr
esear
chappr
oach
I
I. Desi
gnofsampl
i
ngpl
an
I
II
. Desi
gnofexper
imenti
fnecessar
y
I
V. Desi
gnofi
nst
rument
1.
7.3.
1.Sel
ect
ionofResear
chAppr
oach
I
nthi
spar
toft
her
esear
chdesi
gn,t
her
esear
chershoul
ddeci
det
het
ypeofst
udyt
obe
conduct
edbasedont
her
equi
rement
soft
hest
udy
.Al
thoughmosteducat
ionalr
esear
cher
used t
o descr
ipt
iver
esear
ch appr
oach,t
het
esthi
stor
icaland exper
iment
alr
esear
ch
met
hodsar
eal
sousedwher
etheyar
enecessar
y.Ther
efor
e,dependi
ngont
het
ypeoft
he
r
eal
i
ty,
thef
it
testr
esear
chappr
oachshoul
dbesel
ect
edbyt
her
esear
cher
.
1.
7.3.
2.Desi
gnofSampl
i
ngPl
an
ASampl
i
ngpl
ani
samechani
sm bywhi
cht
hesampl
i
nguni
tsofast
udyar
esel
ect
edf
rom
t
hesampl
i
ngf
rameoft
hepopul
ati
on.I
neducat
ionalr
esear
ch,t
hesel
ect
ionf
rom t
he
sampl
i
ngpl
anaf
fect
sthecostt
imeandener
gyt
oconductt
hest
udyandt
her
eli
abi
l
ityof
i
nfer
encesoft
he st
udy
.Consequent
ly,i
tshoul
d be sel
ect
ed wi
th ut
mostcar
e.The
sampl
i
ng pl
an can be cl
assi
fi
ed i
nto pr
obabi
l
ity sampl
i
ng pl
an and non-
probabi
l
ity
sampl
i
ngpl
an.
Pr
obabi
l
itysampl
i
ngpl
ani
ncl
udessi
mpl
erandom sampl
i
ng,sy
stemat
icsampl
i
ng,and
st
rat
if
iedr
andom sampl
i
ng,cl
ust
ersampl
i
ng,andmul
ti
-st
agesampl
i
ng.I
ntheot
herway
,

16|
Page
non-
probabi
l
itysampl
i
ngpl
ani
ncl
udesconv
eni
encesampl
i
ng,j
udgmentsampl
i
ng,quot
a
sampl
i
ngandsnowbal
lsampl
i
ng.Dependi
ngont
hepopul
ati
onsi
zer
equi
redpr
eci
sionand
av
ail
abl
eti
met
ocar
ryoutt
her
esear
chpr
oject
,asui
tabl
esampl
i
ngpl
ani
stobesel
ect
ed
byt
her
esear
cherorr
esear
cht
eam.
1.
7.3.
3.Desi
gnofexper
iment
Ast
udyi
nvol
vesdi
ff
erentr
esponsev
ari
abl
es.Eachr
esponsev
ari
abl
emaybeaf
fect
edby
sev
eralf
act
ors.To t
estt
heef
fectoft
hesef
act
orson r
esponsev
ari
abl
ea sui
tabl
e
exper
imenti
stobedesi
gnedsot
hatt
henecessar
ydat
afort
est
ingt
hesi
gni
fi
cancesof
t
heef
fect
sont
her
esponsev
ari
abl
ear
ecol
l
ect
edandt
hei
nfer
encesoft
het
estar
ehi
ghl
y
r
eli
abl
e.
Ther
ear
etwomai
nst
epsofdesi
gni
ngt
heexper
iment
I
. I
dent
if
yther
esponsev
ari
abl
esoft
hest
udy
I
I. Foreachr
esponsev
ari
abl
erepeatt
hef
oll
owi
ngst
eps:
a) I
dent
if
ythef
act
orsaf
fect
ingt
her
esponse.
b) Deci
deont
het
ypeofeachf
act
or(
afact
ormaybeei
therf
ixedorr
andom f
act
or)
.
c) Fi
xthenumberofl
evel
s(t
reat
ment
s)ofeachf
act
or.
d) For
mtheskel
etonoft
heexper
iment
.
e) Wr
it
ethemodel
oft
heexper
imentanddef
inei
tscomponent
s.
1.
7.3.
4Desi
gnofTool
s
Thesour
ceofdat
acanbecl
assi
fi
edi
ntot
wo:
1.Pr
imar
ysour
ceofdat
aand
2.Secondar
ysour
ceofdat
a
Pr
imar
ysour
ceofdat
aisadocumentorawr
it
tenwor
kbyani
ndi
vi
dualwhoact
ual
l
y
conduct
edt
her
esear
chst
udy
,dev
elopedt
het
heor
y,wi
tnessedt
heev
ent
sorf
ormul
ated
t
heopi
nionsdescr
ibedi
nthedocument
.Whi
l
esecondar
ysour
ceofdat
aisadocumentat
t
hedi
sposalofani
ndi
vi
dualwhodi
dnotact
ual
l
ydot
her
esear
ch,dev
elopt
heor
ies,
wi
tnesst
heev
ent
sorf
ormul
atet
heopi
nionsdescr
ibedi
nthedocument
.
Ther
ear
edi
ff
erentmet
hodsofcol
l
ect
ingpr
imar
ydat
a.Thesear
eobser
vat
ion,per
sonal
i
nter
view,t
elephonei
nter
view,mai
lsur
veyet
c.Thesuccessofsur
veymet
hodofr
esear
ch
t
hatmost
lyusedi
nbehav
ior
alandeducat
ionalr
esear
chdependsont
hest
rengt
hoft
he
quest
ionnai
re(
inst
rument
)used.Thei
nst
rumentusedmustcont
ainpr
ovi
sionst
ocol
l
ect
al
lthedat
ait
emswhi
char
erequi
redf
ort
est
ingdi
ff
erenthy
pot
hesesoft
heexper
imentas
wel
lasf
ort
est
ingt
hehy
pot
hesesofot
hert
est
srel
ati
ngt
ovar
iousr
esear
chi
ssues.
1.
7.4.Dat
aCol
l
ect
ion

17|
Page
Dat
aar
ethebasi
cinputt
oanydeci
sionmaki
ngpr
ocessi
neducat
ionalpl
anni
ngand
management
.Aswedi
scussi
nther
esear
chdesi
gncomponentt
hedat
acanbecol
l
ect
ed
f
rom pr
imar
ysour
ceandsecondar
ysour
ceofdat
a.Pr
imar
ydat
acol
l
ect
ionmet
hodsar
e:
1.Obser
vat
ionmet
hodi
nwhi
chdat
acanbecol
l
ect
edbyi
nvest
igat
ort
hroughper
sonal
obser
vat
ions.
2.Per
sonal
int
erv
iewi
ssur
veymet
hodofdat
acol
l
ect
ionwhi
chempl
oysaquest
ionnai
re.
3.Per
sonal
int
erv
iewcanbecl
assi
fi
edi
ntodoor-t
o–doori
nter
view,
execut
ivei
nter
view,
sel
f-
admi
nist
eredquest
ionnai
re,
etc.
4.Tel
ephonei
nter
viewi
sconsi
der
edt
obeacostef
fect
iveanddomi
nantdat
acol
l
ect
ion
met
hod.But
,thesecondar
ysour
ceofdat
acanbeobt
ainedf
rom i
nter
nalandext
ernal
sour
ceofdat
aused.
1.
7.5.Dat
aAnal
ysi
s
Af
tercol
l
ect
ing appr
opr
iat
e dat
a,pr
opert
ool
s and t
echni
ques shoul
d be used f
or
cl
assi
fi
cat
ionandanal
ysi
sofdat
a.Thet
ool
stocl
assi
fydat
aar
efr
equencydi
str
ibut
ion,
cumul
ati
vef
requencydi
str
ibut
ion,r
elat
ivef
requencydi
str
ibut
ionandchar
ts.Char
tsar
e
gr
aphi
calr
epr
esent
ati
onofdat
a.Di
ff
erentt
ypesofchar
tsar
epi
echar
t,barchar
t,st
acked
barchar
t,hi
stogr
am,
frequencypol
ygonandogi
vecur
ves.Thesecl
assi
fi
cat
iont
ool
sser
ve
asdat
apr
esent
ati
ont
echni
quesf
orcl
eari
nter
pret
ati
ons.Thedat
aanal
ysi
stechni
ques
i
nvol
veoneormor
eoft
hef
oll
owi
ngt
asks:
1.Comput
ati
onofst
ati
sti
cssuchasmean,medi
an,mode,st
andar
ddev
iat
ioncoef
fi
cient
of
2.v
ari
ati
on,
coef
fi
ci
entofskewer
s,et
c
3.Desi
gni
ngr
egr
essi
onequat
ionf
orest
imat
ingr
esponsev
ari
abl
easaf
unct
ionofsetof
i
ndependentv
ari
abl
es.
4.Per
for
mancecor
rel
ati
onanal
ysi
s.
5.Test
ingdi
ff
erenthy
pot
hesesr
elat
ingt
ovar
iousi
ssuesoft
her
esear
ch.
6.Fact
oranal
ysi
s.
7.Di
scr
imi
nat
eananal
ysi
s.
8.Conj
ointanal
ysi
s.
Bot
hthedat
apr
esent
ati
onandanal
ysi
stechni
queswi
l
lbet
reat
edl
ateri
nuni
tsi
xand
mor
especi
fi
cal
l
yint
hecour
se“
Int
roduct
iont
oSt
ati
sti
cal
Met
hodi
nEducat
ion(
EPsy
-211)
”.
1.
7.6I
nter
pret
ati
onofResul
ts
Af
tert
her
esear
chdat
ahasbeencol
l
ect
edandt
heanal
ysi
shasbeenmade,
ther
esear
cher
canpr
oceedt
othest
ageofi
nter
pret
ingt
her
esul
ts.I
nter
pret
ati
oncal
l
sforl
ogi
caland

18|
Page
cr
it
ical
exami
nat
ionsoft
her
esul
tsobt
ainedaf
teranal
ysi
s,keepi
ngi
nvi
ewt
hel
i
mit
ati
onof
t
hesampl
echosen,t
het
ool
ssel
ect
edandusedi
nthi
sst
udy
.Ini
nter
pret
ingt
her
esul
tin
educat
ional
resear
ch,
Koul
(1984)suggest
edt
hef
oll
owi
ngf
act
orst
obev
iewed:
1.I
gnor
ingunst
udi
edf
act
ors.
2.I
nter
pret
ingt
heexpect
edr
esul
ts.
3.I
nter
pret
ingnegat
iver
esul
ts.
4.I
nter
pret
ingr
esul
tswhent
henul
lhy
pot
hesi
sisr
etai
ned.
5.I
nter
pret
ingr
esul
tswhennul
lhy
pot
hesi
sisr
eject
ed.
6.I
nter
pret
ingnon-hy
pot
hesi
zedr
elat
ionshi
ps.
1.
7.7Repor
ti
ngt
heResul
ts
Thi
sist
hel
astandi
mpor
tantst
epoft
her
esear
chpr
ocess.I
tischar
act
eri
zedbycar
eful
l
y
f
ormul
atedi
nfer
ences,concl
usi
ons,orgener
ali
zat
ions.Ther
esear
chermustbeabl
eto
r
epor
ther
s/hi
spr
ocedur
es,f
indi
ngs,andconcl
usi
onswi
thut
mostobj
ect
ivi
tyt
oot
her
s
whomaybei
nter
est
edi
nther
esear
cher
’sst
udyandi
tsr
esul
ts.

19|
Page
UNI
TTWO
Dev
elopmentofResear
chPr
oposal
2.Pr
epar
ati
onofResear
chPr
oposal
Themai
npur
poseofpr
epar
ingr
esear
chpr
oposali
stodescr
ibet
hepr
obl
em y
ouwi
sht
o
st
udyandhowt
opl
ant
ost
udyi
t.Thi
spr
ocesscompel
syout
othi
nkt
hroughandr
ecor
d
onpapery
ourr
esear
chdesi
gnandpr
ocedur
est
haty
ouwi
l
luset
osel
ectasampl
e,col
l
ect
dat
aandanal
ysi
sthedat
a.Ther
esear
chpr
oposalshoul
dbeappr
ovedbef
orecol
l
ect
ing
dat
a.I
ngener
al,acar
eful
l
ypl
annedandwel
l
-desi
gnedr
esear
chpr
oposalshoul
dincl
ude
t
hef
oll
owi
ngcat
egor
iesofi
nfor
mat
ion:
1.Thet
it
leoft
her
esear
chpr
oposal
Thef
ir
stcomponentofanyi
ntr
oduct
orypar
toft
her
esear
chpr
oposali
sit
sti
tl
e.Thi
s
componentshoul
dbewor
dedi
nthati
tsuggest
sthet
hemeoft
hest
udy
.Goodt
it
leshoul
d
cl
ear
lyi
dent
if
yther
esear
chpr
oposal
andmustcl
ear
lyst
ateaboutt
hef
oll
owi
ng:
1. Thev
ari
abl
esi
ncl
udedi
nther
esear
chpr
oposal

sti
tl
e;
2. Ther
elat
ionshi
pbet
weent
hev
ari
abl
es;
3. Thepopul
ati
ont
owhi
cht
her
esul
tst
obegener
ali
zed.
I
nexper
iment
alr
esear
ch,
thev
ari
abl
esst
atedi
nther
esear
chpr
oposalt
it
lear
edependent
andi
ndependentv
ari
abl
eswher
easi
nnon-
exper
iment
alr
esear
chst
udi
est
heyar
evar
iant
s
andcr
it
eri
onv
ari
abl
es.I
nthi
sty
peofst
udy
,thet
it
leoft
her
esear
chpr
oposali
ssost
ated
t
hati
tshowt
heef
fect
sofi
ndependentv
ari
abl
eupondependentv
ari
abl
e.Thi
sty
peoft
it
le
i
ndi
cat
eswhi
chv
ari
abl
ewi
l
lbemani
pul
atedbyt
her
esear
chanduponwhi
chv
ari
abl
eit
s
ef
fectwi
l
lbeobser
ved.I
nnon-
exper
iment
alst
udyt
het
it
leshoul
dindi
cat
ether
elat
ionshi
p

20|
Page
bet
weent
hev
ari
antandt
hecr
it
eri
onv
ari
abl
e.I
nnon-
exper
iment
alst
udyt
hev
ari
abl
esar
e
notmani
pul
atedr
atheronl
yrel
ati
onshi
pbet
weent
hev
ari
antandcr
it
eri
onv
ari
abl
eis
st
ated.
Anot
herr
equi
rementf
oragoodr
esear
cht
it
lei
sthati
tshoul
dnotbet
ool
engt
hy.I
tshoul
d
bespeci
fi
ctot
hear
eaofst
udy
.Anadequat
eti
tl
efort
hest
udyi
sonewhi
chr
epr
esent
sa
r
easonabl
erest
ri
cti
on and i
mpl
i
esan obj
ect
iveappr
oach.I
tmustbenot
ed t
hatt
he
pr
oposedst
udycannotbecompl
etel
ypr
esent
edi
nthet
it
lebuti
tshoul
dbesowor
dedt
hat
i
twoul
dgi
vesuf
fi
ci
enti
nfor
mat
ionaboutt
henat
ureoft
hest
udy
.Inwr
it
ingpr
oposalt
it
le
f
if
teent
otwent
ywor
dsar
ethemaxi
mum t
obei
ncl
uded.

2.I
ntr
oduct
ionoft
heSt
udy
2.
1Backgr
oundoft
hest
udy
I
tist
hepar
toft
hepr
oposalt
hatpr
ovi
desr
eader
swi
tht
hebackgr
oundi
nfor
mat
ionf
ort
he
r
esear
chpr
oposal
.Thei
ntr
oduct
ionshoul
dci
tet
hosewhohadt
hei
deaf
ir
standshoul
d
al
soci
tet
hemostr
ecentandr
elev
antwor
k.Thenexpl
ainwhymor
ewor
kisr
equi
red.
Mor
especi
fi
cal
l
yitshoul
dincl
ude:
-
 Backgr
oundi
nfor
mat
ion
 Shoul
dident
if
yvar
iabl
esoft
her
esear
chpr
obl
em anddi
scusst
hem.
 St
ati
sti
csorcont
ext
ual
inf
ormat
iont
hatmayappl
ytot
het
opi
c
 Br
iefi
ntegr
atedsummer
yoft
hef
indi
ngsofpastr
esear
ch.
 Abr
iefr
ati
onal
forwhyt
hepr
oposedr
esear
chi
sneeded
 I
tmayal
soi
ncl
udea“
ver
bal
roadmap”–gui
dingt
her
eadert
owhatl
i
esahead.
3.St
atement
soft
hepr
obl
em
Thesecondpar
tofanyr
esear
chpr
oposali
sther
esear
chpr
obl
em.I
thasadef
ini
tepl
ace
i
nthei
ntr
oduct
orypar
toft
her
esear
chpr
oposal
andi
sanat
temptt
ofocusonacl
eargoal
.
St
atement
soft
hepr
obl
em shoul
dpr
imar
il
ybeanexpansi
ont
othet
it
le.I
tshoul
dbeei
ther
i
nquest
ionf
orm orasadecl
arat
ivest
atement
.
St
atement
soft
hepr
obl
em shoul
ddef
ineanddel
i
mitt
hespeci
fi
car
eaoft
hepr
oposed
r
esear
ch.I
tshoul
dbegi
nwi
tht
hegener
albackgr
oundoft
hepr
obl
em andwi
thaspeci
fi
c
st
atementoft
hepr
obl
em.Ther
esear
chpr
obl
em shoul
dbesost
ruct
uredt
hati
tshoul
d
begi
nwi
thabr
oadbaseofgener
alpr
obl
emsandexpl
anat
ions,f
oll
owedbyasur
veyof

21|
Page
r
elat
edr
esear
chl
i
ter
atur
eandi
tshoul
dendwi
thpr
obl
em st
atement
.
Thebackgr
oundoft
her
esear
chpr
obl
em shoul
dident
if
ythev
ari
abl
esoft
her
esear
ch
pr
obl
em;di
scusst
hev
ari
abl
eswhi
char
esel
ect
edf
orr
esear
chst
udyandot
heri
mpor
tant
v
ari
abl
eswhi
char
enoti
ncl
udedi
nther
esear
chst
udy
.Itshoul
dal
sobemadecl
eart
hat
whi
chcr
it
eri
awer
eusedf
ort
hesel
ect
ionoft
hev
ari
abl
es.
I
nwr
it
ingst
atement
soft
hepr
obl
em t
hel
anguageusedshoul
dbesi
mpl
eandpr
eci
se.
St
atement
soft
hepr
obl
em shoul
dal
sost
atet
hesi
gni
fi
canceoft
hepr
obl
em.Toj
ust
if
ythe
si
gni
fi
cantoft
her
esear
chpr
oposal
,thef
oll
owi
ngr
equi
rementshoul
dbegat
her
ed:
 Ther
esear
chpr
oposal
ist
imel
y(I
tiscr
uci
all
yimpor
tancei
ntheseday
s);
 I
t shoul
dber
elat
edwi
thapr
act
icalpr
obl
em andshoul
dpr
ovi
desol
uti
ont
oany
exi
sti
ngeducat
ional
probl
em;
 Ther
esear
chpr
obl
em shoul
dnothav
esomesmal
ltar
getofpopul
ati
ont
hati
tsr
esul
ts
cannotbev
eri
fi
ed;
 Tar
getpopul
ati
onofanyr
esear
chpr
obl
em shoul
dber
elat
edt
oapopul
arpopul
ati
on.
Mostoft
heaut
hor
it
iesi
neducat
ionalr
esear
chcommonl
yagr
eedt
hatast
atementoft
he
pr
obl
em gi
vesdi
rect
iont
other
esear
chpr
ocessandi
tisconsi
der
edast
hel
astst
onei
n
t
hepy
rami
doft
her
esear
chpr
obl
em (
Khan(
2004)
,Best&Kahn(
2003)
,Koul(
1984)
).
St
atement
soft
hepr
obl
em speci
fi
est
hev
ari
antandcr
it
eri
onv
ari
abl
esi
nnon-
exper
iment
al
st
udi
es ,i
ndependentand dependentv
ari
abl
es i
n exper
iment
alst
udi
es,t
he t
ype of
r
elat
ionshi
pbet
weenv
ari
abl
esandt
het
argetpopul
ati
onoft
hest
udy
.
I
nwr
it
ingst
atement
soft
hepr
obl
em anot
heri
mpor
tantcr
it
eri
ato bef
ocusedi
sit
s
measur
abi
l
ity
.Ifar
esear
chpr
obl
em i
nwhi
cht
her
elat
ionshi
pbet
weenv
ari
abl
escannotbe
measur
edempi
ri
cal
l
y,i
tcannotbecat
egor
izedasar
esear
chpr
obl
em.Thest
atement
whi
chseekst
oanswert
hequest
ionsofv
alue-
judgmentshoul
dnotbeal
soi
ncl
udedi
n
r
esear
chpr
obl
em.Suchquest
ionscannotbeanswer
edbyr
esear
chst
udy
4.Rev
iewofRel
atedLi
ter
atur
e
Thet
heor
eti
calandempi
ri
calf
ramewor
koft
hepr
obl
em ar
eadescr
ibedbyt
her
esear
cher
pr
ovi
desev
idencet
hatt
her
esear
cheri
sfami
l
iarwi
thwhati
sal
readyknownandwhati
s
st
il
lunknownandunt
est
ed.Becauseef
fect
iver
esear
chi
sbasedonpastknowl
edge,t
his
st
ep hel
pst
o el
i
minat
ethedupl
i
cat
ion ofwhathasbeen doneand pr
ovi
ded usef
ul
hy
pot
hesi
sandhel
pful
suggest
ionsf
orsi
gni
fi
canti
nvest
igat
ion.
Ci
ti
ngst
udi
est
hatshowsubst
ant
ialagr
eementandt
hoset
hatseem t
opr
esentconf
li
cti
ng
concl
usi
onshel
pto shar
pen and def
ine under
standi
ng ofexi
sti
ng knowl
edge i
nthe
pr
obl
em ar
ea.Mor
eov
er,i
tpr
ovi
desabackgr
oundf
orr
esear
chst
udi
esandmakest
he

22|
Page
r
eaderawar
eoft
hest
atusoft
hei
ssue.Par
adi
ngal
ongl
i
stofannot
atedst
udi
esr
elat
ing
t
othepr
obl
em i
sinef
fect
iveandi
nappr
opr
iat
e.Onl
ythosest
udi
est
hatar
epl
ainl
yrel
evant
,
compet
ent
lyexecut
ed and cl
ear
lyr
epor
ted shoul
d be i
ncl
uded.I
n sear
chi
ng r
elat
ed
l
i
ter
atur
ether
esear
chershoul
dnot
ecer
tai
nimpor
tantel
ement
s:
 Repor
tsofst
udi
esofcl
osel
yrel
atedpr
obl
emst
hathav
ebeeni
nvest
igat
ed;
 Desi
gnoft
hest
udyi
ncl
udi
ngpr
ocedur
esempl
oyedanddat
agat
her
ingi
nst
rument
s
used;
 Popul
ati
onst
hatwer
esampl
edandsampl
i
ngmet
hodempl
oyed;
 Var
iabl
est
hatwer
edef
ined;
 Ex
traneousv
ari
abl
est
hatcoul
daf
fectt
hef
indi
ngs;
 Faul
tst
hatshoul
dbeav
oidedand
 Recommendat
ionsf
orf
urt
herr
esear
ch.
Thi
spar
toft
hepr
oposali
sappr
opr
iat
etof
ormul
ater
esear
chhy
pot
hesi
s,t
ocl
ari
fyt
he
nat
ureoft
hepr
obl
em,r
ecogni
zi
ngt
hesi
gni
fi
canceoft
hest
udyandgi
vesdi
rect
iont
othe
dat
agat
her
ingpr
ocess.

5.Resear
chQuest
ionsand/
orHy
pot
hesi
s
Theot
herpar
tofr
esear
chpr
oposali
sst
atement
soft
hehy
pot
heses.Thebasi
cquest
ions
t
hatt
her
esear
chi
sdesi
gnedt
oanswerar
eusual
l
yfr
amedashy
pot
hesest
obet
est
edon
t
hebasi
sofev
idence.Thi
sst
epest
abl
i
shest
hepr
obl
em andt
hel
ogi
cunder
li
ngt
he
r
esear
chst
udy
.Itgi
vest
hedi
rect
iont
othedat
agat
her
ingpr
ocedur
e.I
tisal
soi
nthel
i
ght
oft
hehy
pot
hesest
hatt
her
elev
anceofdat
atobecol
l
ect
edi
sjudged.
Quest
ions ar
e mostof
ten used i
n qual
i
tat
ivei
nqui
ry.The r
esear
cherset
sal
i
stof
quest
ionst
hatf
ormt
hebasi
sforsur
vey
.Hy
pot
hesesar
erel
evantt
otheor
eti
calr
esear
ch
andt
ypi
cal
l
yusedi
nquant
it
ati
vei
nqui
ryandi
tissubj
ectt
otesti
.e.t
oacceptorr
eject
basedonempi
ri
cal
grounds.
Thef
ormul
ati
onofr
esear
chhy
pot
hesesi
sty
pical
l
ydet
ermi
nedbyt
hei
mpl
i
cat
ionsoft
he
r
elat
edl
i
ter
atur
eandt
hededuct
iver
easoni
ngoft
hepr
obl
em underi
nvest
igat
ion.Al
lthe
t
ermswhi
char
eusedi
nanyhy
pot
hesi
sshoul
dbe
1.Cl
ear
lyandpr
eci
sel
yst
ated
2.Test
abl
e
3.Depi
cti
ngt
heexpect
edr
elat
ionshi
pbet
weenv
ari
abl
es
4.Li
mit
edonscope

23|
Page
5.Consi
stentwi
thmostknownf
act
s
6.St
atedasf
araspossi
blei
nsi
mpl
eter
ms
7.Amenabl
etot
est
ingwi
thi
nar
easonabl
eti
mei
ngener
al,hy
pot
hesescanbecl
assi
fi
ed
i
nto:
1.Nul
lhy
pot
hesi
s(Ho:
X1-X2=0 )
2.Al
ter
nat
ivehy
pot
hesi
s(Ho:
X1-X2>0,
Ho:
X1-X2<0)
Exampl
esofHy
pot
hesi
s
Hy
pot
hesi
sforexper
iment
alr
esear
ch:
-
 “
Iti
shy
pot
hesi
zedt
hatt
hesci
enceachi
evementscor
esof10t
hgr
adewi
l
lbehi
gherf
or
gr
oupsusi
ngpor
tfol
i
ost
hanf
ort
hoseusi
ngsepar
ateassi
gnment
s.”
 Hy
pot
hesi
scanonl
ybef
ormul
atedaf
tert
her
esear
cherhasgai
nedenoughknowl
edge
r
egar
dingt
henat
ure,
ext
entandi
ntensi
tyoft
hepr
obl
em.
Exampl
esf
orResear
chQuest
ions
1.Howwel
comi
ngi
stheschool
tospeci
alneedsst
udent
s?
2.Doest
eacher
shav
eadequat
eski
l
linhandl
i
ng except
ionalchi
l
dreni
nther
egul
ar
cl
assr
oom?
3.Whatar
ethef
act
orst
hataf
fectef
fect
ivei
ncl
usi
onofst
udent
swi
thspeci
aleducat
ional
needs?

6.Obj
ect
ivesoft
hest
udy
Theobj
ect
iveofar
esear
chpr
ojectsummar
izeswhati
stobeachi
evedbyt
hest
udy
.It
shoul
dbecl
osel
yrel
atedt
othest
atementoft
hepr
obl
em.Obj
ect
ivesshoul
dbesi
mpl
e,
speci
fi
c,st
atedi
nadv
ance(
notaf
tert
her
esear
chi
sdone)
,st
atedusi
ngact
ionv
erbst
hat
ar
etobemeasur
ed.
Gener
alandSpeci
fi
cObj
ect
ives
Commonl
yobj
.iscl
assi
fi
edi
ntot
wobutt
heyar
elogi
cal
l
yconnect
edt
oeachot
her
.
Speci
fi
c obj
.ar
e consi
der
ed as smal
l
erpor
ti
ons oft
he gener
alobj
ect
ives.Pr
oper
ly
f
ormul
atedobj
.wi
l
lfaci
l
itat
ethedev
elopmentofy
ourr
esear
chmet
hodol
ogyandwi
l
lhel
p
t
oor
ientt
hecol
l
ect
ion,
anal
ysi
s,i
nter
pret
ati
onandut
il
izat
ionofdat
a.
7.Si
gni
fi
canceoft
heSt
udy
Thewor
thandur
gencyoft
hest
udyshoul
dbecl
ear
lydescr
ibedi
nthesi
gni
fi
cantoft
he
r
esear
chpr
oposalorst
udypar
t.Thi
spar
tshoul
dindi
cat
ecl
ear
lyhow t
her
esul
tsoft
he

24|
Page
r
esear
chcani
nfl
uenceeducat
ionalt
heor
yorpr
act
ice.Theneedf
orunder
taki
ngar
esear
ch
st
udycanbeshowni
nsev
eral
way
ssuchas:
1.Toshowt
het
imel
apsebet
weent
heear
li
erst
udyandt
hepr
esentone,andt
her
efor
e,
t
henewknowl
edge,
techni
quesorcondi
ti
oni
ndi
cat
eaneedt
orepl
i
cat
ethest
udy
;
2.Toshow t
hatt
her
ear
egapsi
ntheknowl
edgepr
ovi
dedbypr
evi
ousr
esear
chst
udi
es
andt
oshow how t
hepr
esentst
udywi
l
lhel
ptof
il
lint
hesegapsandaddt
othe
quant
um ofexi
sti
ngknowl
edge;
3.To show t
he l
ack ofi
nfor
mat
ion abouta pr
obl
em bypr
esent
ing t
he suppor
ti
ng
st
atement
sofot
herr
esear
chst
udi
es;
4.Tor
elat
ethest
udyt
otheexi
sti
ngsoci
ali
ssuesandt
opr
ior
it
yar
easi
nthecont
extof
nat
ional
andi
nter
nat
ional
educat
ional
dev
elopment
s;
5.Thesear
esomeofi
mpor
tantt
hepoi
ntst
obeconsi
der
edt
ojust
if
ythesi
gni
fi
cantor
t
heneedf
orunder
taki
ngt
hest
udy
.
8.Del
i
mit
ati
onoft
heSt
udy
I
nthi
spar
toft
her
esear
chpr
oposal
,theboundar
iesoft
hest
udyshoul
dbemadecl
ear
wi
thr
efer
encet
o:
1.Thescopeoft
hest
udybyspeci
fyi
ngt
hear
east
owhi
cht
heconcl
usi
onswi
l
lbeconf
ined
and2.Thepr
ocedur
alst
atementi
ncl
udi
ngt
hesampl
i
ngpr
ocedur
es,t
het
echni
quesof
dat
acol
l
ect
ionandanal
ysi
s,t
hedev
elopmentofmeasur
ingt
ool
sandt
hei
rusesi
nthe
st
udy
.
9.Oper
ati
onal
def
ini
ti
onoft
ermsandconcept
s.
I
tisnecessar
ytodef
ineal
lunusualt
ermsandconcept
sthatcoul
dbemi
sint
erpr
eted.The
t
echni
calt
ermsorwor
dsand phr
aseshav
ing speci
almeani
ngsneed t
o bedef
ined
oper
ati
onal
l
y.Wesel
dom comeacr
ossunani
mousdef
ini
ti
onsormodesofmeasur
ement
f
orcer
tai
nconcept
sandt
ermssuchasi
ntel
l
igence,mot
ivat
ion,achi
evementet
c.andi
n
such si
tuat
ions i
tbecomes obl
i
gat
oryon t
he par
toft
he r
esear
chert
o pr
ovi
de an
oper
ati
onaldef
ini
ti
onofsucht
ermsorconcept
sbyst
ati
nghow t
hev
ari
abl
eswi
l
lbe
measur
edandi
nter
pret
ed.
10.Resear
chmet
hods,
mat
eri
alsandpr
ocedur
es
10.
1.St
udydesi
gn
10.
2.St
udyar
ea
10.
3.St
udysubj
ect
s/par
ti
cipant
s
10.
4.Sampl
esi
ze/sampl
i
ngmet
hod
10.
5.Met
hodsofdat
acol
l
ect
ion

25|
Page
10.
6.Dat
aanal
ysi
smet
hod
11.Ti
meSchedul
e
Pr
epar
ingt
imeschedul
einanyacademi
cresear
chpr
oposali
samust
,because,i
tshows
t
het
imei
nter
valt
ocompl
etet
her
esear
chst
udy
.Ithel
psal
sot
her
esear
chert
odi
vi
dea
st
udyi
ntophasesandt
ousehi
sti
mesy
stemat
ical
l
yfort
hecompl
eti
onofeachphase.I
t
i
susedt
omanaget
het
imepr
oper
ly.
12.BudgetSchedul
e
I
nthi
spar
toft
her
esear
chpr
oposal
,ther
esear
chershoul
dal
soi
ncl
udebudgetpr
oposal
est
imat
ion.Theest
imat
ingr
esear
chbudgetbasi
cal
l
yincl
udest
hef
undsr
equi
redf
ort
rav
el
expenses,t
ypi
ng,pr
int
ing and cy
clost
yli
ng,pur
chase of st
ati
onar
y mat
eri
als and
equi
pment
,tool
s,andbooksandot
herr
elev
antmat
eri
alset
c.
13.Bi
bli
ogr
aphi
es
I
neachr
esear
chpr
oposal
,ther
esear
chershoul
dgi
vel
i
stsofbooks,j
our
nal
sandot
her
document
sthathehasusedi
nsel
ect
ingt
hepr
obl
em andwhi
chhemayusedur
ingt
he
t
enur
eoft
hest
udy
.

UNI
TTREE
Met
hods,
Mat
eri
alandPr
ocedur
es
3.
1Met
hods
I
tist
hehear
toft
her
esear
chpr
oposal
.Itst
art
sonasepar
atepage,y
oumustdeci
de
exact
lyhow y
ouar
egoi
ngt
oachi
evey
ourst
atedobj
ect
ivesandhow y
ouar
egoi
ngt
o
26|
Page
col
l
ectandpr
ocessdat
a.I
tdescr
ibeshowt
her
esear
cherpl
anst
ocar
ryoutt
hest
udy
.B/
c
i
tdescr
ibeswhatAct
ivi
ti
esshoul
dbedescr
ibedwi
thasmuchdet
ailaspossi
bleandt
he
cont
inui
tyb/
nthem shoul
dbeappar
ent
.
Thi
spar
toft
her
esear
chpr
oposal
usual
l
yconsi
stsoft
hef
oll
owi
ngpar
ts.
A.St
udydesi
gn
B.Subj
ectoft
hest
udy[
par
ti
cipant
s]
C.Met
hodsofDat
acol
l
ect
ion
D.Met
hodofdat
aanal
ysi
syouwi
l
ldo,
thi
ssect
ioni
swr
it
teni
nthef
utur
etense.
3.
2St
udyDesi
gn
Thest
udyobj
ect
ivescanbeachi
eved t
hroughanumberofal
ter
nat
ivedesi
gns.The
r
esear
cherhav
etosel
ectt
hemostappr
opr
iat
eandmostf
easi
bledesi
gn.Thet
ypeof
r
esear
chdesi
gnchosendependson:Thet
ypeoft
hepr
obl
em;t
heknowl
edgeal
ready
av
ail
abl
eaboutt
he pr
obl
em;
ther
esour
ceav
ail
abl
efort
hest
udy
.
3.
3Subj
ect
soft
hest
udy
Thi
ssect
iondet
ail
sthepopul
ati
onf
rom whi
cht
her
esear
cherpl
anst
osel
ectt
hesampl
e
and sampl
e uni
t.Met
hods used t
o sel
ectt
he par
ti
cipant
s(sampl
i
ng t
echni
que)ar
e
di
scussedher
e.
Sampl
i
ngi
sthepr
ocessofsel
ect
inganumberofst
udyuni
tsf
rom adef
inedst
udy
popul
ati
on.
Youshoul
ddescr
ibet
het
ypeofdemogr
aphi
cdat
athaty
oupl
ant
ocol
l
ectandhow y
ou
wi
l
lobt
aini
t.
3.
4Sampl
i
ng
Sampl
i
ng i
sthe pr
ocess ofsel
ect
ing r
epr
esent
ati
ve member
s(sampl
es)f
rom t
he
popul
ati
on.
3.
4.1Sampl
i
ngLogi
c
The st
udyofa popul
ati
on oft
his si
ze woul
drequi
ret
he ser
vices oft
housands of
r
esear
cher
s,t
heexpendi
tur
esofmi
l
li
onsofbi
rrandmuchcl
asshour
s.For
tunat
ely
,the
pr
ocessofsampl
i
ngmakesi
tpossi
blet
odr
awv
ali
dinf
erencesorgener
ali
zat
ionsbased
oncar
efulobser
vat
ionofv
ari
abl
eswi
thi
nar
elat
ivel
ysmal
lpr
opor
ti
onoft
hepopul
ati
on
(
sampl
e).
Mor
eimpor
tant
ly,
theneedf
orsampl
i
ngdependsont
hef
oll
owi
ngpoi
nts.
 Usual
l
yacompl
eteenumer
ati
onofapopul
ati
oni
shar
dlypossi
ble.
 Resear
cher
sconduct
edonsampl
edemandsl
esst
imeandl
abor
.
 Sampl
i
ngi
sal
sor
elat
ivel
yeconomi
calsi
ncei
tcont
ainsf
ew peopl
eandl
esspr
int
ed

27|
Page
mat
eri
al,
fewercost
s,et
c.
 Sampl
i
ngpr
ovi
desmor
edet
aili
nfor
mat
ionbecausei
tdeal
swi
thsmal
lnumberof
uni
ts.
3.
5Popul
ati
onandSampl
e
Popul
ati
oni
sanygr
oupofi
ndi
vi
dual
sthathav
eoneormor
echar
act
eri
sti
csi
ncommon
t
hatar
eofi
nter
estt
other
esear
cher
.Thepopul
ati
onmaybeal
lthei
ndi
vi
dual
sofa
par
ti
cul
art
ype,
ormor
erest
ri
ctedpar
toft
hegr
oup.
Exampl
e:
 Al
lpubl
i
cschool
teacher
s,
 Al
lmal
esecondar
yschool
teacher
s,
 Al
lel
ement
aryschool
teacher
set
c.ar
eassampl
es.
Sampl
i
ngf
ramei
stheact
uall
i
stofsampl
i
nguni
tsf
rom whi
cht
hesampl
esel
ect
ed.The
t
argetpopul
ati
onf
rom whi
chi
nfor
mat
ioncol
l
ect
edshoul
dbewel
ldef
ined.I
tmaybeei
ther
t
ool
argeort
oosmal
l
.Ifi
tist
ool
arge,t
her
esear
chermustt
akeasmal
lpor
ti
onoft
he
popul
ati
ont
hatcoul
dbemanageabl
eandpr
ovi
dehi
m/heracompl
etepi
ctur
eoft
hei
ssue
(
Sar
ant
akos,
1993)
.
Sampl
eisasmal
lpr
opor
ti
onofapopul
ati
onsel
ect
edf
orobser
vat
ionandanal
ysi
s.By
obser
vingt
hechar
act
eri
sti
csoft
hesampl
e,onecanmakecer
tai
ninf
erencesaboutt
he
char
act
eri
sti
csoft
hepopul
ati
onf
rom whi
chi
tisdr
awn.Ameasur
edv
aluebasedupon
sampl
edat
aisst
ati
sti
cs.Apopul
ati
onv
aluei
nfer
redf
rom st
ati
sti
csi
spar
amet
ers.
Cont
rar
ytosomepopul
ati
onopi
nion,sampl
esnotsel
ect
edhaphazar
dly
,theyar
echosen
i
nasy
stemat
icr
andom waysot
hatchanceort
heoper
ati
onofpr
obabi
l
itycanbeut
il
ized.
Thebaseofr
andomnessi
stheassumpt
iont
hat
,whi
l
eindi
vi
dualev
ent
scannotpr
edi
cted
wi
thaccur
acy
,col
l
ect
iveev
ent
scoul
dcomewi
thi
t.Fori
nst
ance,al
thoughi
tmaynot
pr
edi
ctwi
thgr
eataccur
acyaboutani
ndi
vi
dual

sacademi
cachi
evement
,itwi
l
lpr
edi
ct
accur
atel
ythe av
erage academi
c per
for
mance ofa gr
oup.Randomi
zat
ion has t
wo
i
mpor
tantappl
i
cat
ionsi
nresear
ch.Thesear
e:
1.Sel
ect
ingagr
oupofi
ndi
vi
dual
sforobser
vat
ionwhoar
erepr
esent
ati
veoft
hepopul
ati
on
aboutwhi
cht
her
esear
cherwi
shest
ogener
ali
ze.
2.Equat
ingexper
iment
alandcont
rolgr
oupsi
nanexper
iment
.Assi
gni
ngi
ndi
vi
dual
sby
r
andom assi
gnmenti
sthebestmet
hodofpr
ovi
dingf
ort
hei
requi
val
ence.

28|
Page
3.
6Cr
it
eri
aforaGoodSampl
i
ngDesi
gn
1.Goal
ori
ent
ati
on:
Thesampl
i
ngdesi
gnbasedont
hest
udy
'sgoal
sandobj
ect
ives.
2.Measur
abi
l
ity
:Thesampl
i
ngdesi
gnpr
ovi
dest
hedat
anecessar
yforanal
ysi
s(enabl
es
v
ali
dinf
erencest
omakeaboutt
hepopul
ati
onf
rom t
hesampl
e).
3.Pr
act
ical
i
ty:Theact
ualact
ivi
ti
esofsampl
i
ng i
dent
if
ied ar
efeasi
blei
ntheact
ual
si
tuat
ion.
4.Economy
:Resear
chobj
ect
ivesmeetwi
thav
ail
abl
eresour
ces(
ti
me,
money
,et
c.)
3.
7Ty
pesofSampl
i
ng
Ther
ear
etwot
ypesofsampl
i
ng;pr
obabi
l
ityandnon-
probabi
l
itysampl
i
ng.Thegener
al
char
act
eri
sti
csofbot
hdi
scussedbel
ow.
3.
7.1Pr
obabi
l
itySampl
i
ng
Pr
obabi
l
itysampl
i
ngt
echni
queusesst
ri
ctpr
obabi
l
ityr
ulesi
nthesel
ect
ionpr
ocess.That
i
s,ev
eryuni
toft
hepopul
ati
onhasequal
,cal
cul
abl
echancet
obesampl
emember
.It
al
l
owsf
orahi
ghdegr
eeofr
epr
esent
ati
onoft
het
argetpopul
ati
onoft
hest
udy
.Pr
obabi
l
ity
sampl
i
ngi
swi
del
yusedi
neducat
ional
resear
chf
ort
hef
oll
owi
ngr
easons.
 I
tishi
ghl
yrel
i
abl
e.
 I
tsdegr
eeofr
epr
esent
ati
oni
shi
gh.
 I
tsf
indi
ngsar
ehi
ghl
ygener
ali
zabl
e.
Humanshav
elongpr
act
icedv
ari
ousf
ormsofr
andom sel
ect
ion,suchaspi
cki
nganame
outofahat
,orchoosi
ngt
heshor
tst
raw.Theseday
s,wet
endt
ousecomput
ersast
he
mechani
sm f
orgener
ati
ngr
andom number
s,whi
chf
ormt
hebasi
sforr
andom sel
ect
ion.
Somedef
ini
ti
ons
Bef
orewecanexpl
aint
hev
ari
ouspr
obabi
l
itymet
hods,wehav
etodef
inesomebasi
c
t
erms.Thesear
e:
 N=t
henumberofcasesi
nthesampl
i
ngf
rame
 n=t
henumberofcasesi
nthesampl
e
 NCn=t
henumberofcombi
nat
ions(
subset
s)ofnf
rom N
 f=n/
N=t
hesampl
i
ngf
ract
ion
Wi
tht
hoset
ermsdef
ined,wecanbegi
ntoexpl
aint
hedi
ff
erentpr
obabi
l
itysampl
i
ng
met
hods.

29|
Page
a.Si
mpl
eRandom Sampl
i
ng
Thesi
mpl
estf
orm ofr
andom sampl
i
ngi
ssi
mpl
erandom sampl
i
ng.Her
eist
hequi
ck
descr
ipt
ionofsi
mpl
erandom sampl
i
ng:
 Obj
ect
ive:
Tosel
ectnuni
tsoutofNsucht
hateachNCnhasanequal
chancet
obet
he
memberoft
hesampl
e.
 Pr
ocedur
e:Useat
abl
eofr
andom number
s,acomput
err
andom numbergener
ator
,or
amechani
cal
dev
icet
osel
ectt
hesampl
e.
Letusassumet
hatwear
edoi
ngsomer
esear
chont
heef
fectoft
rai
ningschoolt
eacher
s
ont
hequal
i
tyofeducat
ion.
Fi
rst
,wehav
etogett
hesampl
i
ngf
rameor
gani
zed.Then,wehav
eact
ual
l
ydr
awnt
he
sampl
e.
Fort
hesakeoft
heexampl
e,l
etussayt
haty
ouwantt
osel
ect100school
teacher
sto
sur
veyandt
hatar
e1000schoolt
eacher
sinacer
tai
nzone,wher
einy
ourr
esear
cht
akes
pl
ace.Thesampl
i
ngf
ract
ioni
sf=n/
N=100/
1000=.
10or10%.Now,dr
awt
hesampl
e,
y
ouhav
esev
eral
opt
ions.
1.Youcoul
dpr
intof
fthel
i
stof1000school
teacher
s,t
eart
heni
ntosepar
atest
ri
ps,
putt
he
st
ri
psi
nahat
,mi
xthem upr
ealgood,cl
osey
ourey
esandpul
loutt
hef
ir
st100.Butt
his
mechani
calpr
ocedur
ewoul
dbet
edi
ousandt
hequal
i
tyoft
hesampl
ewoul
ddependon
howt
hor
oughl
yyoumi
xedt
hem upandhowr
andoml
yyour
eachedi
n.
2.Per
hapsabet
terpr
ocedur
ewoul
dbet
ouset
heki
ndofbal
lmachi
net
hati
spopul
arwi
th
manyoft
hest
atel
ott
eri
es.Youwoul
dneedt
hreeset
sofbal
l
snumber
ed0t
o9,
onesetf
or
eachoft
hedi
git
sfr
om 000t
o999(
ifwesel
ect000wewi
l
lcal
lthat1000)
.Numbert
hel
i
st
ofnamesf
rom 1t
o1000andt
henuset
hebal
lmachi
net
osel
ectt
het
hreedi
git
sthat
sel
ect
seachper
son.
3.Nei
theroft
hesemechani
calpr
ocedur
esi
sver
yfeasi
bleand,wi
tht
hedev
elopmentof
i
nexpensi
vecomput
ers,t
her
eisamucheasi
erway
.Her
eisasi
mpl
epr
ocedur
ethati
s
especi
all
yusef
uli
fyouhav
ethenamesoft
hecl
i
ent
sal
readyont
hecomput
er.Many
comput
erpr
ogr
amscangener
ateaser
iesofr
andom number
s.
Thegai
nofsi
mpl
erandom sampl
i
ngi
s,i
tissi
mpl
etoaccompl
i
shandi
seasyt
oexpl
aint
o
ot
her
s.Becausesi
mpl
erandom sampl
i
ngi
saf
airwayt
osel
ectasampl
e,i
tisr
easonabl
e
t
ogener
ali
zet
her
esul
tsf
rom t
hesampl
ebackt
othepopul
ati
on.

30|
Page
Howev
er,si
mpl
erandom sampl
i
ngi
shar
dlyst
ati
sti
cal
l
yef
fi
cientmet
hodofsampl
i
ngand
y
oumay
,justbecauseoft
hel
uckoft
hedr
aw,
notgetgoodr
epr
esent
ati
onofsubgr
oupsi
n
apopul
ati
on.Todeal
wit
hthesei
ssues,
wehav
etot
urnt
oot
hersampl
i
ngmet
hods.

b.St
rat
if
iedRandom Sampl
i
ng
I
tal
sosomet
imescal
l
edpr
opor
ti
onalorquot
arandom sampl
i
ng.I
tinv
olv
esdi
vi
dingy
our
popul
ati
oni
ntohomogeneoussubgr
oupsandt
hent
aki
ngasi
mpl
erandom sampl
eineach
subgr
oup.
Obj
ect
ive:Di
vi
det
hepopul
ati
oni
ntonon-
over
lappi
nggr
oups(
i.
e.,st
rat
a)N1,N2,N3.
..Ni
,
sucht
hatN1+N2+N3+.
..+Ni=N.Thendoasi
mpl
erandom sampl
eoff=n/
Nineach
st
rat
a.
Ther
ear
esev
eralmaj
orr
easonswhyy
oumi
ghtpr
eferst
rat
if
iedsampl
i
ngt
o si
mpl
e
r
andom sampl
i
ng.
1.I
tassur
est
haty
ouwi
l
lbeabl
etor
epr
esentnotonl
ytheov
eral
lpopul
ati
on,
butal
sokey
subgr
oupsoft
hepopul
ati
on,especi
all
ysmal
lmi
nor
it
ygr
oups.I
fyouwantt
obeabl
eto
t
alkaboutsubgr
oups,
thi
smaybet
heonl
ywayef
fect
ivel
yassur
eyouwi
l
lbeabl
eto.
2.St
rat
if
iedr
andom sampl
i
ngwi
l
lgener
all
yhav
emor
est
ati
sti
calpr
eci
siont
hansi
mpl
e
r
andom sampl
i
ng.Thi
swi
l
lonl
ybet
ruei
fthest
rat
aorgr
oupsar
ehomogeneous.I
fthey
ar
e,weexpectt
hatt
hev
ari
abi
l
itywi
thi
n-gr
oupsar
elowert
hant
hev
ari
abi
l
ityf
ort
he
popul
ati
onasawhol
e.St
rat
if
iedsampl
i
ngcapi
tal
i
zesont
hatf
act
.
 Exampl
e:Supposewesayt
hat
,thest
udentcl
i
ent
sforacounsel
i
ngbur
eaudi
vi
ded
i
ntot
hreegr
oups:Fi
rsty
ear
,secondy
earandt
hir
dyearst
udent
s.Fur
ther
mor
e,l
et
usassumet
hatbot
hthesecondy
earandt
hir
dyearst
udent
sar
erel
ati
vel
ysmal
las
cl
i
ent
ele(
10%and5%r
espect
ivel
y).I
fwej
ustdi
dasi
mpl
erandom sampl
eofn=100
wi
thasampl
i
ngf
ract
ionof10%,wewoul
dexpectbychanceal
onet
hatwewoul
d
onl
yget10and5per
sonsf
rom eachofourt
wosmal
l
ergr
oups.And,
bychance,
we
coul
dgetf
ewert
hant
hat
!Ifwest
rat
if
y,wecandobet
ter
.
Fi
rst
,letusdet
ermi
nehowmanypeopl
ewewantt
ohav
eineachgr
oup.Assumewest
il
l
wantt
otakeasampl
eof100f
rom t
hepopul
ati
onof1000st
udentcl
i
ent
s.Howev
er,we
t
hinkt
hati
nor
dert
osayany
thi
ngaboutsubgr
oupswewi
l
lneedatl
east25casesi
neach
gr
oup.So,
letussampl
e50f
ir
sty
ear
,25-
secondy
earand25t
hir
dyearst
udent
s.
Weknow t
hat10% oft
hepopul
ati
on,or100cl
i
ent
s,ar
esecondy
earst
udent
s.I
fwe

31|
Page
r
andoml
ysampl
e25oft
hese,
wehav
eawi
thi
n-st
rat
um sampl
i
ngf
ract
ionof25/
100=25%.
Si
mil
arl
y,weknow t
hat5% or50cl
i
ent
sar
ethi
rdy
earst
udent
s.Soourwi
thi
n-st
rat
um
sampl
i
ngf
ract
ionwi
l
lbe25/
50=50%.Fi
nal
l
y,bysubt
ract
ionweknowt
hatt
her
ear
e850
f
ir
sty
earst
udentcl
i
ent
s.Ourwi
thi
n-st
rat
um sampl
i
ngf
ract
ionf
ort
hem i
s50/
850=about
5.
88%.Thet
otal
numberofsampl
ewi
l
lbe25+25+50=100.
Thebenef
itofst
rat
if
iedsampl
i
ngi
sbecauset
hegr
oupsar
emor
ehomogeneouswi
thi
n
t
hegr
oup t
han acr
osst
hepopul
ati
on asawhol
e,wecan expectgr
eat
erst
ati
sti
cal
pr
eci
si
on(
lessv
ari
ance)
.Inaddi
ti
on,becausewest
rat
if
ied,weknowwewi
l
lhav
eenough
casesf
rom eachgr
oupt
omakemeani
ngf
ulsubgr
oupi
nfer
ences.

c.Syst
emat
icRandom Sampl
i
ng
I
tisaneasi
erpr
ocedur
etouset
hansi
mpl
erandom sampl
i
ng,i
fthesampl
esel
ect
edi
s
v
eryl
argeandal
i
stoft
heaccessi
bleort
argetpopul
ati
oni
sav
ail
abl
e.Thepopul
ati
onuni
ts
ar
especi
all
ypr
epar
edandt
hesampl
esel
ect
edi
nasy
stemat
icwayusi
ngsampl
i
ng
f
ract
ionmet
hod,whi
chsy
mbol
i
cal
l
yrepr
esent
edasN/
n,wher
eNr
epr
esent
sthenumber
ofuni
tsi
napopul
ati
onand‘
n’i
sthenumberofdesi
redsampl
es.
Her
ear
ethest
epsy
ouneedt
ofol
l
owi
nor
dert
oachi
eveasy
stemat
icr
andom sampl
e:
 numbert
heuni
tsi
nthepopul
ati
onf
rom 1t
oN
 deci
deont
hen(
sampl
esi
ze)t
haty
ouwantorneed
 k=N/
n=t
hei
nter
val
size
 r
andoml
ysel
ectani
ntegerbet
ween1t
ok
 t
hent
akeev
erykt
huni
t
Exampl
e:Lookatt
heex
ampl
eabov
e,popul
ati
ont
hatonl
yhasN=100peopl
eini
tandt
hat
y
ouwantt
otakeasampl
eofn=20.Tousesy
stemat
icsampl
i
ng,
thepopul
ati
onl
i
stedi
na
r
andom or
der
.Thesampl
i
ngf
ract
ionwoul
dbef=20/
100=20%.I
nthi
scase,t
hei
nter
val
si
ze,
k,i
sequal
toN/
n=100/
20=5.Now,
sel
ectar
andom i
ntegerf
rom 1t
o5.
I
nourexampl
e,i
magi
net
haty
ouchose4.Now,
tosel
ectt
hesampl
e,st
artwi
tht
he4t
huni
t
i
nthel
i
standt
akeev
erykt
huni
t(ev
ery5t
h,becausek=5)
.Youwoul
dhav
esampl
i
nguni
ts
4,
9,14,
19,
andsoont
o100andy
ouwoul
dendupwi
th20uni
tsi
nyoursampl
e.
Thegai
nofsy
stemat
icsampl
i
ngi
sthati
tissi
mpl
ert
osel
ectoner
andom numberand
t
henev
ery‘
kth’(
e.g.5t
h)memberont
hel
i
st,t
hant
osel
ectasmanyr
andom number
sas
sampl
esi
ze.I
tal
sogi
vesagoodspr
eadr
ightacr
osst
hepopul
ati
on.Aweaknessi
sthat
y
oumayneedal
i
stt
ost
artwi
th,i
fyouwi
sht
oknowy
oursampl
esi
zeandcal
cul
atey
our
sampl
i
ngi
nter
val
.

32|
Page
d.Cl
ust
er(
area)Random Sampl
i
ng
Thepr
obl
em wi
thr
andom sampl
i
ngmet
hodswhenwehav
etosampl
eapopul
ati
ont
hat
di
sbur
sedacr
ossawi
degeogr
aphi
cregi
oni
sthaty
ouwi
l
lhav
etocov
eral
otofgr
ound
geogr
aphi
cal
l
yinor
dert
ogett
oeachoft
heuni
tsy
ousampl
ed.
I
magi
net
aki
ngasi
mpl
erandom sampl
eofal
lther
esi
dent
sofSNNPRi
nor
dert
oconduct
per
sonali
nter
views.Byt
hel
uckoft
hedr
aw,y
ouwi
l
lendupwi
thr
espondent
swhocome
f
rom al
lov
ert
hest
ate.Youri
nter
viewer
sar
egoi
ngt
ohav
eal
otoft
rav
eli
ngt
odo.To
ov
ercomet
hispr
obl
em t
hemet
hodofcl
ust
erorar
ear
andom sampl
i
ngwasi
nvent
ed.

I
ncl
ust
ersampl
i
ng,
wef
oll
owt
hesest
eps:
 di
vi
depopul
ati
oni
ntocl
ust
ers(
usual
l
yal
onggeogr
aphi
cboundar
ies)
,
 r
andoml
ysampl
ecl
ust
ers,
 measur
eal
luni
tswi
thi
nsampl
edcl
ust
ers,
Exampl
e:ASuper
visormayr
equi
ret
ocar
ryoutast
udyont
heef
fectofl
i
brar
yfaci
l
iti
esi
n
st
rengt
heni
ngcl
assr
oom i
nst
ruct
ionatsecondar
yschooli
nAddi
sAbaba.He/
shesel
ect
s
t
hesampl
easf
oll
ows:
 Thenumberofsecondar
yschool
sinAddi
sisassumi
ngt
obe36
 Ther
equi
redsampl
eis9,
 Sui
tabl
ecl
ust
eri
ssub-
cit
y,
 Theav
eragenumberofsecondar
yschool
sineachsub-
cit
yassumedt
obe4
 Ther
equi
redsampl
ecl
ust
erwi
l
lbe9/
4=2.
25sub-
cit
ies,
 Thet
otal
numberofsub-
cit
iesi
s10,
 The2sub-
cit
ieswi
l
lber
andoml
ysel
ect
edf
rom t
he10sub-
cit
ies,
 Fi
nal
l
yal
lthe secondar
yschool
sfr
om t
he 2 sub-
cit
ies wi
l
lbe consi
der
ed (
i.
e.
2.
25X4=9)f
ort
hest
udy
.
Cl
ust
er sampl
i
ng has sev
eral r
etur
ns: r
educed cost
s, si
mpl
i
fied f
iel
dwor
k and
admi
nist
rat
ionar
emor
econv
eni
ent
.Inst
eadofhav
ing,asampl
escat
ter
edov
ert
heent
ir
e
cov
eragear
ea,
thesampl
elocal
i
zedi
nrel
ati
vel
yfewcent
ers(
clust
ers)
.Cl
ust
ersampl
i
ng’
s
weaknessi
sthatl
essaccur
ater
esul
tsof
tenobt
ainedduet
ohi
ghersampl
i
nger
rort
hanf
or
si
mpl
erandom sampl
i
ngwi
tht
hesamesampl
esi
ze.
e.Mul
ti
-st
ageSampl
i
ng
Mul
ti
-st
agesampl
i
ngi
sli
kecl
ust
ersampl
i
ng,buti
nvol
vessel
ect
ingasampl
ewi
thi
neach

33|
Page
chosencl
ust
er,r
athert
hani
ncl
udi
ngal
luni
tsi
nthecl
ust
er.Thus,mul
ti
-st
agesampl
i
ng
i
nvol
vessel
ect
ingasampl
einatl
eastt
wost
ages.I
nthef
ir
stst
age,l
argegr
oupsor
cl
ust
erssel
ect
ed.Thesecl
ust
ersdesi
gned t
o cont
ain mor
e popul
ati
on uni
ts t
han
ar
e r
equi
redf
or t
he f
inal sampl
e.I
nthesecondst
age,
popul
ati
onuni
tschosenf
rom
sel
ect
edcl
ust
erst
oder
iveaf
inal
sampl
e.I
fmor
ethant
wost
agesar
eused,
thepr
ocessof
choosi
ngpopul
ati
onuni
tswi
thi
ncl
ust
erscont
inuesunt
ilt
hef
inal
sampl
eachi
eved.
Anexampl
eofmul
ti
-st
agesampl
i
ngi
swher
e,f
ir
stl
y,el
ect
oralsub-
div
isi
ons(
clust
ers)
sampl
edf
rom aci
tyorst
ate.Secondl
y,bl
ocksofhousessel
ect
edf
rom wi
thi
ntheel
ect
oral
sub-
div
isi
onsand,t
hir
dly
,indi
vi
dualhousessel
ect
edf
rom wi
thi
nthesel
ect
edbl
ocksof
houses.
Ther
ewar
dsofmul
ti
-st
agesampl
i
ngar
econv
eni
ence,economyandef
fi
ciency
.Mul
ti
-
st
agesampl
i
ngdoesnotr
equi
reacompl
etel
i
stofmember
sint
het
argetpopul
ati
on,
whi
chgr
eat
lyr
educessampl
epr
epar
ati
oncost
.Thel
i
stofmember
sisr
equi
redonl
yfor
t
hosecl
ust
ersusedi
nthef
inalst
age.Themai
nweaknessofmul
ti
-st
agesampl
i
ngi
sthe
sameasf
orcl
ust
ersampl
i
ng:
loweraccur
acyduet
ohi
ghersampl
i
nger
ror
.
3.
7.2Non-
probabi
l
itySampl
i
ng
Non-
probabi
l
itysampl
escannotdependupont
her
ati
onal
eofpr
obabi
l
ityt
heor
y.That
meanswecannotbesur
ethatsampl
esi
nanon-
probabi
l
itysampl
ear
erepr
esent
ati
veof
t
hepopul
ati
on.
Thedi
ff
erencebet
weennon-
probabi
l
ityandpr
obabi
l
itysampl
i
ngi
sthatnon-
probabi
l
ity
sampl
i
ngdoesnoti
nvol
ver
andom sel
ect
ionandpr
obabi
l
itysampl
i
ngdoes.Atl
eastwi
tha
pr
obabi
l
ist
icsampl
e,weknow t
heoddsorpr
obabi
l
ityt
hatwehav
erepr
esent
ed t
he
popul
ati
onwel
l
.Wear
eabl
etoest
imat
econf
idencei
nter
val
sfort
hest
ati
sti
cinpr
obabi
l
ity
sampl
i
ng.Wi
thnon-
probabi
l
itysampl
es,wemayormaynotr
epr
esentt
hepopul
ati
onwel
l
andi
twi
l
lof
tenbehar
dforust
oknowhowwel
lwehav
edoneso.
I
n gener
al,r
esear
cher
s pr
eferpr
obabi
l
ist
ic orr
andom sampl
i
ng met
hods t
o non-
pr
obabi
l
ist
iconesandconsi
dert
hem mor
eaccur
ateandr
igor
ous.Howev
er,i
nappl
i
ed
soci
alr
esear
ch t
her
e may be ci
rcumst
ances wher
eiti
s notf
easi
ble,pr
act
icalor
t
heor
eti
cal
l
ysensi
blet
odor
andom sampl
i
ng.Her
e,weconsi
derawi
der
angeofnon-
pr
obabi
l
ist
ical
ter
nat
ives.
Tr
ochi
m M.
K.(
2006)di
vi
ded non-
probabi
l
ity
-sampl
i
ng met
hodsi
ntot
wo br
oad t
ypes:
acci
dent
alorpur
posi
ve,mostsampl
i
ngmet
hodsar
epur
posi
vei
nnat
urebecausewe
usual
l
yappr
oacht
hesampl
i
ngpr
obl
em wi
thaspeci
fi
cpl
ani
nmi
nd.Themosti
mpor
tant
di
sti
nct
ionsamongt
heset
ypesofsampl
i
ngmet
hodsar
etheonesbet
weent
hedi
ff
erent

34|
Page
t
ypesofpur
posi
vesampl
i
ngappr
oaches.
a.Acci
dent
alSampl
i
ng
I
tisal
sohaphazar
dorconv
ent
ionalsampl
i
ng.Oneoft
hemostcommonmet
hodsof
sampl
i
nggoesundert
hev
ari
oust
it
lesl
i
stedher
e.Wewoul
dincl
udei
nthi
scat
egor
ythe
t
radi
ti
onal"
manont
hest
reet
"(ofcour
se,now i
tispr
obabl
ythe"
per
sonont
hest
reet
")
i
nter
viewsconduct
edf
requent
lybyt
elev
isi
onnewspr
ogr
amst
ogetaqui
ck(
alt
houghnon-
r
epr
esent
ati
ve)r
eadi
ngofpubl
i
copi
nion.
Wewoul
dal
soar
guet
hatt
het
ypi
caluseofcol
l
egest
udent
sinmuchpsy
chol
ogi
calor
educat
ional
resear
chi
spr
imar
il
yamat
terofconv
eni
ence.
Cl
ear
ly,t
hepr
obl
em wi
thal
loft
heset
ypesofsampl
esi
sthatwehav
enoev
idencet
hat
t
heyar
erepr
esent
ati
veoft
hepopul
ati
onswear
eint
erest
edi
n,gener
ali
zi
ngt
oandi
nmany
cases,
wewoul
dcl
ear
lysuspectt
hatt
heyar
enot
.
b.Pur
posi
veSampl
i
ng
I
npur
posi
vesampl
i
ng,wesampl
ewi
thapur
posei
nmi
nd.Weusual
l
ywoul
dhav
eoneor
mor
especi
fi
cpr
edef
inedgr
oupswear
eseeki
ng.
Wemi
ghtsampl
eforspeci
fi
cgr
oupsort
ypesofpeopl
easi
nmodali
nst
ance,ex
per
tor
quot
asampl
i
ng.Wemi
ghtsampl
efordi
ver
sit
yasi
nhet
erogenei
tysampl
i
ng.Or
,wemi
ght
capi
tal
i
zeoni
nfor
malsoci
alnet
wor
kst
oident
if
yspeci
fi
crespondent
swhoar
ehar
dto
l
ocat
eot
her
wise,asi
nsnowbal
lsampl
i
ng.I
nal
loft
hemet
hodsweknowwhatwewant
,
wear
esampl
i
ngwi
thapur
pose.
i
.Modali
nst
ancesampl
i
ng:I
nst
ati
sti
cs,t
hemodei
sthemostf
requent
lyoccur
ri
ngv
alue
i
nadi
str
ibut
ion.I
nsampl
i
ng,whenwedoamodali
nst
ancesampl
e,wear
esampl
i
ngt
he
mostf
requentcase,ort
he"
typi
cal
"case.I
nmanyi
nfor
malpubl
i
copi
nionpol
l
s,f
or
i
nst
ance,
theyi
nter
viewa"
typi
cal
"vot
er.
i
i
i.Exper
tsampl
i
ng:Exper
tsampl
i
ngi
nvol
vest
heassembl
i
ngofasampl
eofper
sonswi
th
knownordemonst
rabl
eexper
ienceandexper
ti
sei
nsomear
ea.Of
ten,
weconv
enesucha
sampl
eundert
hesponsor
shi
pofa"
panelofexper
ts.
"Ther
ear
eact
ual
l
ytwor
easonsy
ou
mi
ghtdoexper
tsampl
i
ng.
1.I
nthi
scase,exper
tsampl
i
ng i
sessent
ial
l
yjusta speci
fi
csub-
caseofpur
posi
ve
sampl
i
ng.
2.Howev
er,
theot
herr
easony
oumi
ghtuseexper
tsampl
i
ngi
stopr
ovi
deev
idencef
ort
he
v
ali
dit
yofanot
hersampl
i
ngappr
oachy
ouhav
echosen.
Exampl
e:l
etussayt
haty
oudomodali
nst
ancesampl
i
ngandar
econcer
nedt
hatt
he
cr
it
eri
ayouusedf
ordef
ini
ngt
hemodali
nst
ancear
esubj
ectt
ocr
it
ici
sm.Youmi
ght

35|
Page
conv
eneanexper
tpanelconsi
sti
ngofper
sonswi
thacknowl
edgedexper
ienceandi
nsi
ght
i
ntot
hatf
iel
dort
opi
candaskt
hem t
oexami
ney
ourmodaldef
ini
ti
onsandcommenton
t
hei
rappr
opr
iat
enessandv
ali
dit
y.Theadv
ant
ageofdoi
ngt
hisi
sthaty
ouar
enotouton
y
ourownt
ryi
ngt
odef
endy
ourdeci
sions;y
ouhav
esomeacknowl
edgedexper
tst
oback
upt
o.
i
i
i.Quot
asampl
i
ng:Her
e,y
ousel
ectpeopl
enon-
randoml
yaccor
dingt
osomef
ixedquot
a.
Ther
ear
etwot
ypesofquot
asampl
i
ng:
Pr
opor
ti
onalquot
a sampl
i
ng:Wewantt
orepr
esentt
hemaj
orchar
act
eri
sti
csoft
he
popul
ati
onbysampl
i
ngapr
opor
ti
onal
amountofeach.
Exampl
e:I
fyouknowt
hepopul
ati
onhas40%womenand60%menandt
haty
ouwanta
t
otalsampl
esi
zeof100,y
ouwi
l
lcont
inuesampl
i
ngunt
ily
ougett
hoseper
cent
agesand
t
heny
ouwi
l
lst
op.So,i
fyouhav
eal
readygott
he40womenf
ory
oursampl
e,butnott
he
si
xtymen,y
ouwi
l
lcont
inuet
osampl
emenbutev
eni
flegi
ti
mat
ewomenr
espondent
s
comeal
ong,
youwi
l
lnotsampl
ethem becausey
ouhav
eal
ready"
mety
ourquot
a."
Non-
propor
ti
onalquot
asampl
i
ng:I
tisabi
tlessr
est
ri
cti
ve.I
nthi
smet
hod,
youspeci
fyt
he
mi
nimum numberofsampl
ed uni
tsy
ou wanti
n each cat
egor
y.Her
e,y
ou ar
e not
concer
nedwi
thhav
ingnumber
sthatmat
cht
hepr
opor
ti
onsi
nthepopul
ati
on.I
nst
ead,
you
si
mpl
ywantt
ohav
eenought
oassur
ethaty
ouwi
l
lbeabl
etot
alkaboutev
ensmal
lgr
oups
i
nthepopul
ati
on.
i
v.Het
erogenei
tysampl
i
ng:Wesampl
eforhet
erogenei
tywhenwewantt
oincl
udeal
l
opi
nions or v
iews and we ar
e not concer
ned about r
epr
esent
ing t
hese v
iews
pr
opor
ti
onat
ely
.Anot
hert
ermf
ort
hisi
ssampl
i
ngf
ordi
ver
sit
y.
I
nmanybr
ainst
ormi
ngornomi
nalgr
ouppr
ocesses(
incl
udi
ngconceptmappi
ng)
,wecan
usesomef
orm ofhet
erogenei
tysampl
i
ngbecauseourpr
imar
yint
eresti
singet
ti
ngbr
oad
spect
rum ofi
deas,noti
dent
if
yingt
he"
aver
age"or"
modali
nst
ance"ones.Het
erogenei
ty
sampl
i
ngi
s,i
nthi
ssense,
almostt
heopposi
teofmodal
inst
ancesampl
i
ng.
v
.Snowbal
lSampl
i
ng:Her
e,y
oubegi
nbyi
dent
if
yingsomeonewhomeet
sthecr
it
eri
afor
i
ncl
usi
oni
nyourst
udy
.Yout
henaskt
hef
ormer
stor
ecommendot
her
swhot
heymay
t
hinkmeett
hecr
it
eri
a.
Al
thought
hismet
hodwoul
dhar
dlyl
eadt
orepr
esent
ati
vesampl
es,
ther
ear
eti
meswheni
t
maybet
hebestmet
hodav
ail
abl
e.Snowbal
lsampl
i
ngi
sespeci
all
yusef
ulwheny
ouar
e
t
ryi
ngt
oreachpopul
ati
onst
hatar
einaccessi
bleorhar
dtof
ind.
Exampl
e:I
fyouar
est
udy
ingt
hehomel
esspeopl
e,y
ouar
enotl
i
kel
ytobeabl
etof
indgood
l
i
stsoft
hem wi
thi
naspeci
fi
cgeogr
aphi
calar
ea.Howev
er,i
fyougot
othatar
eaand

36|
Page
i
dent
if
yoneort
wo,y
oumayf
indt
hatt
heyknowv
erywel
lwhot
heot
herhomel
esspeopl
e
i
nthei
rvi
cini
tyar
eandhowy
oucanf
indt
hem.
3.
8Sampl
eSi
ze
Ther
eisusual
l
yat
rade-
offbet
weent
hedesi
rabi
l
ityofal
argesampl
eandt
hef
easi
bil
i
tyof
asmal
lone.Thei
dealsampl
eisl
argeenought
oser
veusanadequat
erepr
esent
ati
onof
t
hepopul
ati
onaboutwhi
cht
her
esear
cherwi
shest
ogener
ali
zeandsmal
lenought
obe
sel
ect
edeconomi
cal
l
y-i
nter
msofsubj
ectav
ail
abi
l
ity
,expensei
nbot
hti
meandmanyand
compl
exi
tyofdat
aanal
ysi
s.
Thei
dealmet
hodi
srandom sel
ect
ion,l
ett
ingt
hechanceorl
awsofpr
obabi
l
itydet
ermi
ne
whi
chmember
soft
hepopul
ati
onsel
ect
ed.Whenr
andom sampl
i
ngempl
oyed,whet
her
t
hesampl
eisl
argeorsmal
l
,theer
ror
sofsampl
i
ngest
imat
ed,gi
vi
ngr
esear
cher
sani
dea
oft
heconf
idencet
hatt
heymaypl
acei
nthei
rfi
ndi
ngs.
Todet
ermi
net
hesi
zeofanadequat
esampl
e,nof
ixedway
s.I
nmanycases,sampl
esi
ze
ofagi
venr
esear
chpr
ojecti
sinf
luencedbyf
act
orsl
i
ket
het
ypeofpopul
ati
on,
themet
hods
used,
theav
ail
abl
eti
meandr
esour
ces,
thepur
poseoft
her
esear
ch,
thei
nst
rument
sused,
andt
hecapaci
tyoft
her
esear
cht
eam.Howev
er,t
hef
oll
owi
ngconsi
der
ati
onmadei
n
det
ermi
ningt
hesi
zeofasampl
e:
 I
fthepopul
ati
oni
shomogenouswi
thr
espectt
othest
udy
,asmal
lsampl
emaybe
adequat
e.I
fitf
oundt
obehet
erogeneous,
alar
gesampl
emayber
equi
red.
 I
fahi
gh,
degr
eeofaccur
acyi
srequi
red,
alar
gesampl
edr
awn.
 I
fthenat
ureoft
hest
udyi
squal
i
tat
ive,
smal
lsampl
eissuf
fi
cient
.
 I
fthest
udyi
squant
it
ati
ve,
alar
genumberofsampl
esdemanded.

3.
9Sampl
eEr
rorandSampl
i
ngBi
as
Youhav
etoknowt
hatwhenev
erasampl
edr
awn,
bydef
ini
ti
on,
onl
ythepopul
ati
ont
hati
s
i
ncl
udedi
nthesampl
emeasur
ed,andusedt
orepr
esentt
heent
ir
epopul
ati
on.Hence,
t
her
emi
ghtal
way
sbesomeer
rori
nthedat
aresul
ti
ngf
rom t
hosemember
soft
he
popul
ati
onwhower
enotmeasur
ed,
thater
rori
scal
l
edsampl
i
nger
ror
.
I
fapopul
ati
onf
rom whi
chasampl
eisdr
awni
slar
ge,t
henpopul
ati
onv
alueswi
l
laf
fect
v
erymuchbyoneort
womember
soft
hepopul
ati
onwhohav
eext
remev
aluesofa
par
ti
cul
armeasur
e.Sampl
i
nger
rorofameanv
alueest
imat
edf
rom asampl
eisequalt
o
t
heest
imat
edst
andar
ddev
iat
ionoft
hev
ari
abl
edi
vi
dedbyt
hesquar
erootoft
hesampl
e

37|
Page
si
ze.I
tis,
ther
efor
e,notdependentont
hepopul
ati
onsi
ze,
butonl
yont
hev
ari
abi
l
ityoft
he
v
ari
abl
eofconcer
nandsampl
esi
ze.Ther
efor
e,sampl
i
nger
rori
sthev
ari
ati
onduet
o
r
andom f
luct
uat
ionwhenr
andom sampl
esusedt
orepr
esentpopul
ati
ons.Mat
hemat
ical
l
y,
St
andar
dEr
roroft
heSampl
i
ngMean(
Sampl
i
ngEr
ror
):
Sampl
i
nger
rorofameanv
alue sx=
Wher
e:S=t
hest
andar
ddev
iat
ionoft
hev
ari
abl
eN=t
hesi
zeoft
hesampl
e
Exampl
e:I
fonedr
awsasampl
eoff
ourobser
vat
ionsf
rom al
argepopul
ati
on,
thesampl
i
ng
er
rorwoul
dbeequalt
othest
andar
ddev
iat
iondi
vi
dedby2(
thesquar
erootoff
our
).To
hal
vet
hesampl
i
nger
roroft
hatv
ari
abl
e,oneneedst
oincr
easet
hesampl
eto16;i
tcoul
d
behal
vedagai
nbyi
ncr
easi
ngt
hesampl
esi
zet
o64,andhal
vedagai
nbyi
ncr
easi
ngt
he
sampl
eto256becausest
andar
ddev
iat
ionsonmanyv
ari
abl
esar
efai
rl
ysmal
lval
ues.Thi
s
r
epr
esent
sav
erysmal
ler
ror
.
I
tal
sof
oll
owst
hati
ncr
easesi
nsampl
esi
zear
enecessar
ytocont
inuet
odecr
easet
he
sampl
i
nger
ror
.
I
tisal
soi
mpor
tantt
ost
ress,
asi
snot
edabov
e,t
hatsampl
i
nger
rori
snotaf
unct
ionoft
he
per
cent
ageoft
hepopul
ati
oni
nthesampl
ewhent
hepopul
ati
oni
slar
ge,
buti
tisaf
unct
ion
oft
henumberi
nthesampl
eandt
hev
ari
abi
l
ityoft
hemeasur
eofconcer
nonl
y.Gener
all
y,
onl
ythr
eet
hingsneedt
oest
imat
esampl
i
nger
ror
:
1.Themet
hodusedt
odr
awt
hesampl
e.
2.Thesi
zeoft
hesampl
e.
3.Thev
ari
anceoft
hemeasur
eforwhi
chsampl
i
nger
rori
stobedet
ermi
ned.
Ani
mpor
tantpr
oper
tyofsampl
i
nger
rori
sthati
tcanal
way
scal
cul
atedi
fsampl
i
ngf
oll
ows
oneoft
hest
andar
dsampl
i
ngpr
ocedur
essuchasequalpr
obabi
l
itysampl
i
ng,st
rat
if
ied
r
andom sampl
i
ng,et
c.Anot
heri
mpor
tantpr
oper
tyofsampl
i
nger
rori
sthati
tconsi
der
s
hav
ingappr
oxi
mat
elyanor
maldi
str
ibut
ion.Thi
smeanst
hatwecanusepr
oper
ti
esoft
he
nor
mal
dist
ri
but
iont
ounder
standwhati
tmeansbyapar
ti
cul
arv
alueofsampl
i
nger
ror
.
Exampl
e:Thenor
maldi
str
ibut
iont
ell
sust
hatt
her
eisa95per
centpr
obabi
l
ityt
hatt
he
est
imat
edv
aluef
rom t
hesampl
eiswi
thi
npl
usormi
nusaboutt
wo(
1.96)st
andar
d
dev
iat
ionsoft
hepopul
ati
onv
alue.I
fwehav
ecomput
edasampl
emeanofhousehol
dsi
ze
of2.
6andwedet
ermi
net
hatt
hesampl
i
nger
roront
hisv
aluei
s0.
01,
thent
hismeanst
hat
wehav
e95per
centconf
idencet
hatt
heest
imat
edmeanv
alueof2.
6iswi
thi
n0.
02oft
he
popul
ati
onv
alue.
Ther
eisal
soa99per
centpr
obabi
l
ityt
hatt
heest
imat
edsampl
emeani
swi
thi
npl
usor
mi
nus2.
58st
andar
ddev
iat
ionsoft
hepopul
ati
onv
alue.I
tiscl
eart
hatt
hebi
ggert
he

38|
Page
sampl
i
nger
ror
,thel
esseroursur
ewear
eaboutt
hepopul
ati
onv
alue;
thi
sisdet
ermi
nedby
ei
theral
argest
andar
ddev
iat
ionf
ort
hev
ari
abl
eofi
nter
est
,orasmal
lsampl
e,orbot
h.
Thi
sil
l
ust
rat
edl
ateri
nthi
str
aini
ngmodul
e.
Sampl
i
ngbi
as,
howev
er,
isdi
ff
erentf
rom sampl
i
nger
rorandhast
hef
oll
owi
ngat
tr
ibut
es:
 Sampl
e-bi
asdoesnotdecr
easewi
thsampl
esi
zeandmayev
eni
ncr
ease,
dependi
ngon
t
hesour
ceoft
hebi
as.
 Sampl
ebi
ascanev
enbepr
esenti
nacensus(
a100% sur
vey
)ifi
tar
isesf
rom
measur
ementpr
obl
emsandi
nst
rumentpr
obl
ems.
 Sampl
e-bi
ascannotbecal
cul
atedi
nmostcasesandbear
snor
elat
iont
osampl
esi
ze,
popul
ati
onsi
ze,
orv
ari
abi
l
ityoft
hemeasur
es.
Sampl
e-bi
asmayar
isef
rom al
argev
ari
etyofsour
ces,
butnotl
i
mit
edt
o:
 Fai
l
uret
oadher
etot
her
andom sampl
i
ngpr
ocedur
es.
 Omi
ssi
onofspeci
fi
csubgr
oupsoft
hepopul
ati
onf
rom t
hesampl
i
ngf
rame.
 Faul
tymeasur
ingdev
ices(
thi
smaybei
nter
msoft
hespeci
fi
cquest
ionsusedi
na
quest
ionnai
re,and may al
so ar
ise i
n a sur
vey t
hat i
nvol
ves t
aki
ng phy
sical
measur
ement
swhent
hemeasur
ingdev
icei
sincor
rectet
c,sot
hatal
lmeasur
ement
s
ar
etoosmal
l
).
 Vi
olat
ionsofequalpr
obabi
l
ityofsel
ect
ionpr
inci
plesbecauseofdupl
i
cat
eli
stsi
nthe
sampl
i
ngf
rame,
orot
hercauses.
 Non-
responset
oasur
veybyspeci
fi
csubgr
oupsoft
hepopul
ati
onwhoar
erel
evantt
o
t
hemeasur
esofconcer
nint
hesur
vey
.
I
nsur
vey
sofhumanpopul
ati
ons,somebi
asi
sinev
itabl
ypr
esent
,becauseofsuchi
ssues
as non-
response and because i
tisi
mpossi
blet
o desi
gn a sur
veyquest
ionnai
re or
i
nter
viewt
hati
scompl
etel
yfr
eef
rom bi
as.Conf
idencel
i
mit
sar
eal
soani
mpor
tantpar
tof
t
heassessmentofbi
asandar
edet
ermi
nedbasedonhowmuchv
ari
abi
l
ityt
her
eisi
nthe
dat
aandhowsur
eorunsur
eonei
soft
hev
alueobt
ainedf
rom t
hesampl
e.Ast
hesi
zeof
t
hedat
aincr
eases,sot
heconf
idencel
i
mit
sbecomecl
oserandcl
osert
othev
alues
est
imat
edf
rom t
hedat
a.Usual
l
y,onecomput
eswhatei
thert
he95per
centor99per
cent
conf
idencel
i
mit
s.Thesel
i
mit
sindi
cat
erespect
ivel
yther
angeofv
alueswi
thi
nwhi
chone
hasei
ther95per
centor99per
centconf
idencet
hatt
het
ruev
alueofapopul
ati
onst
ati
sti
c
wi
l
lli
ebasedont
heest
imat
efr
om t
hesampl
e.Ther
efor
e,99per
centconf
idencel
i
mit
s
def
ineawi
derr
anget
han95per
centconf
idencel
i
mit
s.Forexampl
e:
 Ani
nter
pret
ati
onof95per
centconf
idencel
i
mit
sist
hatonehasasmal
l
ert
han1i
n20
chancet
hatt
het
ruepopul
ati
onv
aluewi
l
lbeout
sidet
hisr
ange.

39|
Page
 Si
mil
arl
y,ani
nter
pret
ati
onof99per
centconf
idencel
i
mit
sist
hatonehasasmal
l
er
t
han1i
n100chancet
hatt
het
ruepopul
ati
onv
aluei
sout
side.
3.
9Dat
aCol
l
ect
ion
Af
teri
dent
if
yingt
hesampl
es,
thenextj
oboft
her
esear
cheri
stocol
l
ectt
her
esear
chdat
a
pr
oper
ly.To do so,dat
agat
her
ing i
nst
rument
sident
if
ied and pr
epar
ed.Becauset
he
i
nfer
ences,hy
pot
hesi
sorgener
ali
zat
ionshel
dident
if
iedasv
ali
d,v
eri
fi
edascor
rector
r
eject
edasanat
tai
nabl
e.
3.
9.1Ty
pesofDat
a
Thedat
acol
l
ect
edf
rom v
ari
oussour
ceswi
thdi
ff
erentt
ool
sandt
echni
quesgener
all
y
compr
iseofnumer
icalf
igur
es,r
ati
ngs,descr
ipt
ivenar
rat
ions,r
esponset
oopen-ended
quest
ions,f
iel
dnot
eset
c.I
neducat
ionalpl
anni
ngandmanagementr
esear
ch,usual
l
ytwo
t
ypesofdat
aar
erecogni
zed.Theyar
equant
it
ati
veandqual
i
tat
ivedat
a.
3.
9.1.
1Quant
it
ati
veDat
a
Quant
it
ati
vedat
aar
eget
ti
ngbyappl
yi
ngv
ari
ousscal
esofmeasur
ementsuchasnomi
nal
,
or
dinal
,and i
nter
valand r
ati
o-scal
es.Quant
it
ati
vedat
aar
eei
therpar
amet
ri
cornon-
par
amet
ri
cdat
a.
Par
amet
ri
cdat
aar
emeasur
eddat
aoni
nter
valorr
ati
oscal
esofmeasur
ement
,wher
eas,
non-
par
amet
ri
cdat
aar
eobt
ainedbyappl
yi
ngnomi
naloror
dinalscal
esofmeasur
ement
.
Themeasur
ementchar
act
eri
sti
csofquant
it
ati
ver
esear
chdat
aar
ethatt
henat
ureoft
he
dat
aexpr
essedi
nquant
if
iabl
ecodi
ng,quant
it
ati
vel
y,count
s,measur
es,oper
ati
onal
i
zed
v
ari
abl
es,
stat
ist
icset
c.
3.
9.1.
2Qual
i
tat
iveDat
a
Qual
i
tat
ivedat
aar
ever
balorot
hersy
mbol
i
cmat
eri
als.Thedet
ail
ed descr
ipt
ionsof
obser
ved,behav
ior
,peopl
e,si
tuat
ionandev
ent
sar
esomeexampl
esofqual
i
tat
ivedat
a.
The r
esponse t
o open-
ended quest
ions ofquest
ionnai
res ora schedul
e,f
ir
sthand
i
nfor
mat
ionf
rom peopl
eaboutt
hei
rexper
ience,i
deas,bel
i
efs,per
sonaldocument
s,f
iel
d
not
e,phot
ogr
aphs,of
fi
cialdocument
s,et
c.ar
edat
aofqual
i
tat
ivei
nit
snat
ure.Sel
ect
ed
cont
ent
s,casehi
stor
ies,
per
sonal
diar
iesandl
ett
ers,
unst
ruct
uredi
nter
viewandot
herdat
a,
whi
char
edescr
ipt
ivei
nnat
ure,
areexampl
esofqual
i
tat
ivedat
a.

40|
Page
3.
9.2Ty
pesofDat
aSour
ces
Cohen(
1994)i
ndi
cat
est
hatdat
asour
cesar
eei
therpr
imar
yorsecondar
y.
3.
9.2.
1Pr
imar
yDat
aSour
ce
Thepr
imar
ydat
asour
cecol
l
ect
edt
hroughl
abor
ator
ymeasur
ement
,fi
eldobser
vat
ion,
quest
ionnai
re,i
nter
view et
c.Mor
eov
er,somepr
imar
ysour
cedat
aforahi
stor
icalki
ndof
r
esear
chcanbeskel
eton,f
ossi
l
s,weapons,t
ool
s,pi
ctur
es,manuscr
ipt
s,char
ts,ar
chi
ves,
di
ari
es,
pai
nti
ngs,
let
ter
sandsof
ort
hwhi
chbel
ongt
oit
.
3.
9.2.
2Secondar
yDat
aSour
ces
Secondar
ysour
cesofdat
acol
l
ect
edf
rom sour
ces,whi
chal
readycr
eat
edf
ort
hepur
pose
off
ir
stt
imeuseandf
utur
euse.I
tcanbecol
l
ect
edf
rom t
echni
calpubl
i
cat
ionssuchas
manual
s,handbook,
dat
asheet
sandst
andar
ds,
booksandj
our
nal
s,of
fi
cial
publ
i
cat
ionsof
t
he cent
ralgov
ernment
,st
ate gov
ernment
s,l
ocalbodi
es,pr
ivat
e dat
a ser
vices,and
comput
erdat
abases.
I
ngener
al,dat
acol
l
ect
edf
rom pr
imar
yandsecondar
ysour
ces.Ther
esear
chershoul
dbe
awar
eoft
hef
oll
owi
nggener
alr
ulest
oal
lty
pesofdat
acol
l
ect
ion.Thesear
e:
1.Donotcol
l
ectmor
einf
ormat
iont
hani
srequi
redf
ort
her
esear
chpr
obl
em.
2.Makesur
ethewor
dingoft
hedat
acol
l
ect
ioni
nst
rumenti
scl
earandunambi
guous.
3.Usecl
earandexpl
i
citi
nst
ruct
ionsi
ndat
acol
l
ect
ioni
nst
rument
s.
4.Desi
gnt
her
esponseopt
ionsascar
eful
l
yast
hei
tem st
emst
hem.
5.Maker
espondi
ngt
othemeasur
ingi
nst
rumentasal
ter
nat
iveaspossi
ble.
6.Makesur
ethatt
hef
inal
product
slookspr
ofessi
onal
.
3.
10Rel
i
abi
l
ityandVal
i
dit
y
Whenev
erat
estorot
hermeasur
ingdev
iceusedaspar
toft
hedat
acol
l
ect
ionpr
ocess,
the
v
ali
dit
yandr
eli
abi
l
ityoft
hatt
esti
simpor
tant
.Justaswewoul
dnotuseamat
hemat
ics
t
estt
oassessv
erbalski
l
ls,
wewoul
dnotwantt
ouseameasur
ingdev
icef
orr
esear
cht
hat
wasnott
rul
ymeasur
ingwhatwecl
aimi
ttomeasur
e.Af
teral
l
,wear
erel
yi
ngont
he
r
esul
tst
oshow suppor
toral
ackofsuppor
tforourt
heor
yandi
fthedat
acol
l
ect
ion
met
hodsar
eer
roneous,
thedat
aweanal
yzewi
l
lbeer
roneous.
Wheny
ouwantt
odesi
gndat
acol
l
ect
iont
ool
s,i
tiscr
it
icalt
omeasur
ethei
rval
i
dit
yand
r
eli
abi
l
ity
.Theser
elat
edr
esear
chi
ssuescompelust
oconsi
derwhet
herwear
est
udy
ing
whatwet
hinkwear
est
udy
ingandwhet
hert
hemeasur
esweusear
econsi
stent
.
3.
10.
1Val
i
dit
y
Val
i
dit
yref
erst
othedegr
eet
owhi
chast
udyaccur
atel
yref
lect
sorassessest
hespeci
fi
c

41|
Page
conceptt
hatt
her
esear
cheri
sat
tempt
ingt
omeasur
e.Val
i
dit
yisconcer
nedwi
tht
he
successoft
hest
udyatmeasur
ingwhatt
her
esear
cher
ssetoutt
omeasur
e.Resear
cher
s
shoul
dbeconcer
nedwi
thbot
hext
ernalandi
nter
nalv
ali
dit
y.Ext
ernalv
ali
dit
yref
erst
othe
ext
entt
owhi
cht
her
esul
tsofast
udyar
etr
ansf
erabl
e.I
nter
nalv
ali
dit
yref
erst
o:f
ir
stt
he
st
ri
ctnesswi
thwhi
cht
hest
udywasconduct
ed(
e.g.
,thedesi
gnoft
hest
udy
,thecar
e
t
aken t
o conductmeasur
ement
s,and deci
sions concer
ning whatwas and was not
measur
ed)andsecondt
heext
entt
owhi
cht
hedesi
gner
soft
hest
udyhav
etakeni
nto
accountal
ter
nat
iveexpl
anat
ionsf
oranycausalr
elat
ionshi
pst
heyexpl
ore.I
nst
udi
est
hat
donotexpl
orecausalr
elat
ionshi
ps,onl
ythef
ir
stoft
hesedef
ini
ti
onsconsi
der
edwhen
assessi
ngi
nter
nal
val
i
dit
y.
Schol
arsdi
scusssev
eralt
ypesofi
nter
nalv
ali
dit
y.Thef
oll
owi
ngar
ebr
iefdi
scussi
onsof
t
het
ypesofi
nter
nal
val
i
dit
y;
a.FaceVal
i
dit
y:I
tisconcer
nedwi
thhowameasur
eorpr
ocedur
eappear
s.Doesi
tseem
l
i
kear
easonabl
ewayt
ogai
nthei
nfor
mat
iont
her
esear
cher
sar
eat
tempt
ingt
oobt
ain?
Doesi
tseem wel
ldesi
gned?Doesi
tseem ast
houghi
twi
l
lwor
krel
i
abl
y?Unl
i
kecont
ent
v
ali
dit
y,f
acev
ali
dit
ydoesnotdependonest
abl
i
shedt
heor
iesf
orsuppor
t.
b.Cr
it
eri
onRel
atedVal
i
dit
y:I
tal
sor
efer
redt
oasi
nst
rument
alv
ali
dit
y,whi
chusedt
o
demonst
rat
etheaccur
acyofameasur
eorpr
ocedur
ebycompar
ing i
twi
th anot
her
measur
eorpr
ocedur
e,whi
chhasbeendemonst
rat
edt
obev
ali
d.
c.Const
ructVal
i
dit
y:Const
ructv
ali
dit
yseeksagr
eementbet
weenat
heor
eti
calconcept
andaspeci
fi
cmeasur
ingdev
iceorpr
ocedur
e.Forexampl
e,ar
esear
cheri
nvent
inganew
I
Qtestmi
ghtspendagr
eatdealoft
imeat
tempt
ingt
o"def
ine"i
ntel
l
igencei
nor
dert
o
r
eachanaccept
abl
elev
elofconst
ructv
ali
dit
y.
Const
ructv
ali
dit
ycanbebr
okendowni
ntot
wosub-
cat
egor
ies:Conv
ergentv
ali
dit
yand
di
scr
imi
nat
eval
i
dit
y.Conv
ergentv
ali
dit
yist
heact
ualgener
alagr
eementamongr
ati
ngs,
gat
her
edi
ndependent
lyofoneanot
herwher
emeasur
est
heor
eti
cal
l
yrel
ated.Di
scr
imi
nat
e
v
ali
dit
yist
hel
ackofar
elat
ionshi
pamongmeasur
est
hatt
heor
eti
cal
l
yunr
elat
ed.
d.Cont
entVal
i
dit
y:Basedont
heext
entt
owhi
chameasur
ementr
efl
ect
sthespeci
fi
c
i
ntendeddomai
nofcont
ent
.
Exampl
e:Resear
cher
sai
mtost
udymat
hemat
icall
ear
ningandcr
eat
easur
veyt
otestf
or
mat
hemat
icalski
l
l.I
ftheser
esear
cher
sonl
ytest
ed f
ormul
ti
pli
cat
ionand t
hendr
ew
concl
usi
onsf
rom t
hatsur
vey
,thei
rst
udydi
dnotshowcont
entv
ali
dit
ybecausei
texcl
uded
ot
hermat
hemat
ical
funct
ions.
e.Pr
edi
cti
veVal
i
dit
y:I
nor
derf
orat
estt
obeav
ali
dscr
eeni
ngdev
icef
orsomef
utur
e

42|
Page
behav
ior
,itmusthav
epr
edi
cti
vev
ali
dit
y.TheSt
andar
dAchi
evementTest(
SAT)i
susedby
col
l
egescr
eeni
ngcommi
tt
eesasonewayt
opr
edi
ctcol
l
egegr
ades.Themai
nconcer
n
wi
tht
hisandmanyot
herpr
edi
cti
vemeasur
esi
spr
edi
cti
vev
ali
dit
ybecausewi
thouti
t,t
hey
woul
dbewor
thl
ess.
Wedet
ermi
nepr
edi
cti
vev
ali
dit
ybycomput
ingacor
rel
ati
oncoef
fi
cientcompar
ingSAT
scor
es,f
orexampl
e,andcol
l
egegr
ades.I
ftheyar
edi
rect
lyr
elat
ed,t
henwecanmakea
pr
edi
cti
onr
egar
dingcol
l
egegr
adesbasedonSATscor
e.Wecanshowt
hatst
udent
swho
scor
ehi
ghont
heSATt
endt
orecei
vehi
ghgr
adesi
ncol
l
ege.
3.
10.
2Rel
i
abi
l
ity
Rel
i
abi
l
ityi
stheconsi
stencyofy
ourmeasur
ement
,ort
hedegr
eet
owhi
chani
nst
rument
measur
est
hesamewayeacht
imeusedundert
hesamecondi
ti
onwi
tht
hesamesubj
ect
s.
I
nshor
t,i
tist
her
epeat
abi
l
ityofy
ourmeasur
ement
.Ameasur
econsi
der
edr
eli
abl
eifa
per
son'
sscor
eont
hesamet
estgi
vent
wicei
ssi
mil
ar.I
tisi
mpor
tantt
oremembert
hat
r
eli
abi
l
ityi
snotmeasur
edi
tisr
atherest
imat
ed.Rel
i
abi
l
ityi
ssuchani
mpor
tantconcept
t
hati
thasbeendef
inedi
nter
msofi
tsappl
i
cat
iont
oawi
der
angeofact
ivi
ti
es.For
r
esear
cher
s,keyt
ypesofr
eli
abi
l
itymeasur
esar
e:
a.Equi
val
encyRel
i
abi
l
ity
:Twoi
temsmeasur
eident
icalconcept
satani
dent
icall
evelof
di
ff
icul
tyt
otheext
ent
.Equi
val
encyr
eli
abi
l
ityi
sdet
ermi
nedbyr
elat
ingt
woset
soft
est
scor
est
ooneanot
hert
ohi
ghl
i
ghtt
hedegr
eeofr
elat
ionshi
porassoci
ati
on.I
nquant
it
ati
ve
st
udi
esand par
ti
cul
arl
yin exper
iment
alst
udi
es,a cor
rel
ati
on coef
fi
cient
,st
ati
sti
cal
l
y
r
efer
redt
oas“
r”,i
susedt
oshow t
hest
rengt
hoft
hecor
rel
ati
onbet
weenadependent
v
ari
abl
e(t
hesubj
ectunderst
udy
),andoneormor
eindependentv
ari
abl
es,whi
char
e
mani
pul
atedt
odet
ermi
neef
fect
sont
hedependentv
ari
abl
e.Ani
mpor
tantconsi
der
ati
oni
s
t
hatequi
val
encyr
eli
abi
l
ityi
sconcer
nedwi
thcor
rel
ati
onnotcausal
rel
ati
onshi
ps.
b.Test
-Ret
estRel
i
abi
l
ity
:Test
-Ret
estr
eli
abi
l
ityr
efer
stot
het
est
’sconsi
stencyamong
di
ff
erentadmi
nist
rat
ions.Todet
ermi
net
hecoef
fi
cientf
ort
hist
ypeofr
eli
abi
l
ity
,thesame
t
estgi
vent
oagr
oupofsubj
ect
sonatl
eastt
wosepar
ateoccasi
ons.I
fthet
esti
srel
i
abl
e,
t
hescor
est
hateachst
udentr
ecei
vesont
hef
ir
stadmi
nist
rat
ionshoul
dbesi
mil
art
othe
scor
esont
hesecond.Wewoul
dexpectt
her
elat
ionshi
pbet
weent
hef
ir
standt
hesecond
admi
nist
rat
iont
obeahi
ghposi
ti
vecor
rel
ati
on.
Onemaj
orconcer
nwi
tht
est
-r
etestr
eli
abi
l
ityi
swhathasbeent
ermedast
hememor
y
ef
fect
.Thi
sisespeci
all
ytr
uewhent
het
woadmi
nist
rat
ionsar
ecl
oset
oget
heri
nti
me.
Thi
scancr
eat
eanar
ti
fi
cial
l
yhi
ghr
eli
abi
l
itycoef
fi
cientassubj
ect
srespondf
rom t
hei
r
memor
yrat
hert
hant
het
esti
tsel
f.Whenapr
e-t
estandpost
-t
estf
oranexper
imenti
sthe

43|
Page
same,
thememor
yef
fectcanpl
ayar
olei
nther
esul
ts.
c.I
nter
-r
aterRel
i
abi
l
ity
:Int
er-
rat
err
eli
abi
l
ityi
stheext
entt
owhi
cht
woormor
eindi
vi
dual
s
(
Coder
s orr
ater
s) agr
ee.I
nterr
aterr
eli
abi
l
ity addr
esses t
he consi
stency of t
he
i
mpl
ement
ati
onofar
ati
ngsy
stem.
Thef
oll
owi
ngscenar
iocanexpl
ainat
estofI
nterr
aterr
eli
abi
l
ity
:Twoormor
eresear
cher
s
ar
eobser
vingahi
ghschool
classr
oom.Thecl
assi
sdi
scussi
ngamov
iet
hatt
heyhav
ejust
v
iewedasagr
oup.Ther
esear
cher
shav
easl
i
dingr
ati
ngscal
e(1bei
ngmostposi
ti
ve,5
bei
ngmostnegat
ive)wi
thwhi
cht
heyar
erat
ingt
hest
udent
'sor
alr
esponses.I
nterr
ater
r
eli
abi
l
ityassessest
heconsi
stencyofhowt
her
ati
ngsy
stem i
mpl
ement
ed.Forexampl
e,i
f
oner
esear
chergi
vesa"
1"t
oast
udentr
esponsewhi
l
eanot
herr
esear
chergi
vesa"
5,"
obv
iousl
ytheI
nterr
aterr
eli
abi
l
itywoul
dbei
nconsi
stent
.Int
err
aterr
eli
abi
l
ityi
sdependent
upont
heabi
l
ityoft
woormor
eindi
vi
dual
stobeconsi
stent
.Tr
aini
ng,educat
ionand
moni
tor
ingski
l
lscanenhanceI
nterr
aterr
eli
abi
l
ity
.
d.I
nter
nalConsi
stency
:Int
ernalconsi
stencyest
imat
esr
eli
abi
l
itybygr
oupi
ngquest
ionsi
n
aquest
ionnai
ret
hatmeasur
ethesameconcept
.Forexampl
e,y
oucoul
dwr
it
etwoset
sof
t
hree quest
ions t
hatmeasur
ethe same concept(
saycl
ass par
ti
cipat
ion)and af
ter
col
l
ect
ingt
her
esponses,
runacor
rel
ati
onbet
weent
hoset
wogr
oupsoft
hreequest
ionst
o
det
ermi
nei
fyouri
nst
rumenti
srel
i
abl
ymeasur
ingt
hatconcept
.
One common way ofcomput
ing cor
rel
ati
on v
alues among t
he quest
ions on y
our
i
nst
rument
sis byusi
ng Cr
onbach'
s Al
pha.I
n shor
t,Cr
onbach'
s al
pha spl
i
ts al
lthe
quest
ionsony
ouri
nst
rumentev
erypossi
blewayandcomput
escor
rel
ati
onv
aluesf
or
t
hem al
l(weuseacomput
erpr
ogr
am f
ort
hispar
t).I
ntheend,y
ourcomput
erout
put
gener
atesonenumberf
orCr
onbach'
sal
pha-andj
ustl
i
keacor
rel
ati
oncoef
fi
cient
,the
cl
oseri
tist
oone,
thehi
ghert
her
eli
abi
l
ityest
imat
eofy
ouri
nst
rument
.Cr
onbach'
sal
phai
s
al
essconser
vat
iveest
imat
eofr
eli
abi
l
ityt
hant
est
/ret
est
.
Thepr
imar
ydi
ff
erencebet
weent
est
/ret
estandi
nter
nal
consi
stencyest
imat
esofr
eli
abi
l
ity
i
sthatt
est
/ret
esti
nvol
vest
woadmi
nist
rat
ionsoft
hemeasur
ementi
nst
rumentwher
eas
t
hei
nter
nal
consi
stencymet
hodi
nvol
vesonl
yoneadmi
nist
rat
ionoft
hati
nst
rument
.
3.
11Desi
gni
ngDat
aCol
l
ect
ionTool
s
I
n conduct
ing r
esear
ch,r
esear
cher
s can use a v
ari
ety oft
ool
sto gat
herpr
imar
y
i
nfor
mat
ion.Among t
hese quest
ionnai
re,i
nter
view,obser
vat
ion,t
est
s,f
ocus gr
oup
di
scussi
on,r
ecor
dings,f
iel
d not
es ar
ethe common and wi
del
yused.Not
ethati
n
desi
gni
ngt
hedat
acol
l
ect
iont
ool
s,t
her
esear
chershoul
dcheckt
hef
ulf
il
lmentoft
he
f
oll
owi
ngcondi
ti
ons.

44|
Page
1.I
tinsur
edt
hatt
hesuppl
i
eddat
ameett
her
equi
rementofv
ali
dit
y.I
tist
osay
,thedat
a
shoul
dmeasur
e,whatt
heycl
aimedt
omeasur
e.
2.Speci
alat
tent
iongi
vent
omeasur
ementer
ror
ssuchaser
rorduet
omal
funct
ioni
ngof
measur
ingequi
pment
,er
rororbi
as,
int
ent
ional
li
e,di
stor
ti
onoff
act
s,andr
andom er
ror
.
3.I
tensur
edt
hatt
hedesi
gneddat
acol
l
ect
iont
ool
sar
esui
tabl
etosampl
edr
awnoutof
t
hepopul
ati
onsot
hatpr
opergener
ali
zat
ioncoul
dbemade.
4.Thet
ool
sar
echeckedf
inal
l
ywhet
heri
tmeasur
eswhati
tissupposedt
omeasur
eand
i
tst
rust
wor
thi
ness’
.

3.
11.1Quest
ionnai
re
Documentconsi
stsofawr
it
tenl
i
stofquest
ionspr
ovi
dedt
oindi
vi
dual
sint
hesampl
esas
t
oel
i
citi
nfor
mat
ion.Respondent
srecor
dwr
it
tenort
ypedr
esponset
oeachquest
ionnai
re
i
tem.Respondent
scanf
il
loutt
hequest
ionnai
reatt
hei
rconv
eni
ence,answert
hei
tem i
n
anyor
der
,takemor
ethanoneset
ti
ngt
ocompet
eit
,makemar
ginalcomment
sorski
p
quest
ions.
Hopki
ns(
1993)i
ndi
cat
ed t
hati
n educat
ionalr
esear
ch,quest
ionnai
resar
e qui
ckand
si
mpl
ewayofobt
aini
ngr
ichi
nfor
mat
ionf
rom pupi
l
s.Theyaskspeci
fi
cquest
ionsabout
aspect
soft
hecl
assr
oom cur
ri
cul
um ort
eachi
ngmet
hod.
Gal
lD.
,etal(
2003)suggest
edt
hef
oll
owi
ngst
epsi
nconst
ruct
ingandadmi
nist
eri
nga
r
esear
chquest
ionnai
re:
1.Def
ini
ngr
esear
chobj
ect
ives:I
tisi
mpor
tantt
hatt
odef
iney
ourr
esear
chpr
obl
em and
l
i
stt
hespeci
fi
cobj
ect
ivest
obeachi
evedorhy
pot
hesi
stobet
est
edbyt
hequest
ionnai
re.
Youmi
ghtst
artwi
thabr
oadt
opi
cbuty
oushoul
dshar
peni
tsf
ocusbef
orebegi
nni
ngon
t
hedesi
gnoft
hequest
ionnai
re.
2.Sel
ect
ingasampl
e:Oncey
ourr
esear
chobj
ect
ivesorhy
pot
hesesar
ecl
ear
lyst
ated,
you
shoul
dident
if
ythet
argetpopul
ati
onf
rom whi
chy
oursampl
ewi
l
lbesel
ect
ed.I
fyoudonot
hav
eknowl
edgeoft
hesi
tuat
iont
hor
oughl
y,y
oumi
ghtmakeami
stakeofsendi
ngy
our
quest
ionnai
ret
oagr
oupt
hatdoesnothav
ethedesi
redi
nfor
mat
ion.I
tmayhel
pyout
obe
appr
opr
iat
elyf
il
ledandr
etur
nedquest
ionnai
re.
3.Desi
gni
ngt
hequest
ionnai
re:Somer
esear
chquest
ionnai
resappeart
othr
ownt
oget
her
i
nanhourort
wo.Theexper
ienceofr
ecei
vi
ngt
hishaphazar
dquest
ionnai
rehasl
edmany

45|
Page
educat
orst
odev
elopnegat
iveat
ti
tudet
owar
dst
hequest
ionnai
reasar
esear
chappr
oach.
You wi
l
lneed t
o ov
ercome t
his negat
ive at
ti
tude by car
efulconst
ruct
ion and
admi
nist
rat
ionofy
ourquest
ionnai
re.Bel
ow ar
esummar
iesofgui
del
i
nesf
ordesi
gni
ng
quest
ionnai
re.
 Eachquest
ionshoul
dber
elev
antt
other
esear
cht
opi
c.
 Keept
hequest
ionnai
resasshor
taspossi
ble.
 Youdonotuse,
compl
exort
echni
cal
ter
mst
hatr
espondentmaynotunder
stand.
 Or
gani
zet
hei
temssot
hatt
heyar
eeasyt
oreadandcompl
ete.
 Numbert
hepagesandi
temst
hequest
ionnai
re.
 I
ncl
udebr
iefandcl
eari
nst
ruct
ionpr
int
edi
nbol
dfacei
nupperorl
owercase.
 Or
gani
zet
hequest
ionnai
rei
nlogi
cal
sequence.
 Begi
nwi
thaf
ewi
nter
est
ingandnon-
thr
eat
eni
ngi
tems.
 Donotputi
mpor
tanti
temsatt
heendofl
ongquest
ionnai
re.
 Av
oidnegat
ivel
yst
atedi
tems.
 Av
oidbi
asedorl
eadi
ngquest
ions.
4.Pi
l
ot-
test
ing t
he quest
ionnai
re:bef
ore di
str
ibut
ing t
he quest
ionnai
ret
othe act
ual
sampl
e,i
tshoul
d bet
est
ed bygi
vi
ng i
tto asampl
eofi
ndi
vi
dual
sdr
awn f
rom t
he
popul
ati
onwhom y
oupl
ant
osel
ecty
ourr
espondent
s.Pr
ovi
despacef
orr
espondent
sto
getcr
it
ici
sm andr
ecommendat
ionf
ori
mpr
ovi
ngt
hequest
ionnai
re.I
tisal
soadv
isabl
eto
askr
espondent
stost
atei
nthei
rownwor
dswhatt
heyt
hinkeachquest
ionmeans.The
quest
ionsshoul
dber
evi
sedandr
etest
edunt
ilal
lormostmember
soft
hepi
l
ot-
test
sampl
eaccur
atel
yunder
standt
hem.
5.Pr
e-cont
act
ingt
hesampl
e:I
tinv
olv
est
her
esear
cher
sident
if
yingt
hemsel
ves,di
scuss
t
hepur
poseoft
hest
udyandr
equest
ingcooper
ati
on.I
tcant
aket
hef
orm ofl
ett
er,
post
car
d,andt
elephonecal
l
.Ital
ert
sther
espondentandusedt
oreducet
hechanceof
bei
ngt
hrownoutbyt
her
espondent
s.I
tal
sohel
pst
oknow how muchr
espondent
sar
e
cooper
ati
ve.
6.Thecov
erl
ett
er:accompany
ingt
hequest
ionnai
rest
rongl
yinf
luencest
her
etur
nrat
e,
whi
chi
sthemai
nobj
ect
iveofdoi
ngaquest
ionnai
resur
vey
.Consequent
ly,i
tshoul
dbe
desi
gnedcar
eful
l
y.Thel
ett
ershoul
dbebr
ief
,buti
tmustconv
eycer
tai
ninf
ormat
ionand
i
mpr
essi
on.Thepur
poseoft
hest
udyi
stoexpl
aint
oper
suadet
her
espondent
sthatt
he
st
udyi
ssi
gni
fi
cantandt
hatt
hei
ranswer
sar
eimpor
tant
.Thei
nvest
igat
oral
soshoul
d
assur
econf
ident
ial
i
tyf
ort
her
espondent
s.
I
tisal
sodesi
rabl
etoassoci
atey
ourst
udywi
thapr
ofessi
onalor
gani
zat
ionwi
thwhi
ch

46|
Page
pr
ospect
iver
espondent
smi
ghti
dent
if
y.
7.Fol
l
ow-
upwi
thr
espondent
s:Af
ewday
saf
tert
het
imel
i
mitspeci
fi
edi
nthecov
erl
ett
er,
i
tisdesi
rabl
etocont
actnon-
respondent
sbysendi
ngaf
oll
ow-
upl
ett
eral
ongwi
thot
her
copyoft
hequest
ionnai
reandanot
hersel
f-
addr
essedenv
elope.I
tist
otr
yot
herappr
oach
i
nyourappeal
tocooper
ati
ont
hant
hepr
evi
ousmet
hod.
Thef
ormatofaquest
ionnai
remaybeopenendedorcl
oseended.I
nthef
ir
stcase,t
he
r
espondentat
tempt
stogi
vet
hef
ullaccountoft
her
esponse.Howev
er,
int
hecl
ose-
ended
f
ormatt
hesubj
ectwi
l
lsel
ectr
esponses,whi
chhe/
shef
eel
sasappr
opr
iat
eanswerf
rom
t
hegi
vensetofal
ter
nat
ives.
Adv
ant
agesandDi
sadv
ant
agesofQuest
ionnai
re
 Adv
ant
ages
 I
thel
pst
o obt
ainmassofi
nfor
mat
ionf
rom r
espondent
swi
del
ydi
str
ibut
edi
n
di
ff
erentset
ti
ngs.
 I
tisf
reef
rom t
hebi
asofi
nvest
igat
or.
 I
tenabl
est
ocol
l
ecti
nfor
mat
ionf
rom r
espondent
sthr
oughmai
lwi
thoutgoi
ng
per
sonal
l
y.
 I
thel
pst
her
espondent
stohav
eadequat
eti
met
ogi
vewel
lthoughtanswer
.
 I
tenabl
est
oreachi
naccessi
bler
espondent
s.
 I
tisdependabl
eduet
othef
actt
hatl
argeamountofi
nfor
mat
ioncol
l
ect
ed.
 Di
sadv
ant
ages
 I
thas l
ow r
ate ofr
etur
n ofappr
opr
iat
elyf
il
led quest
ionnai
re because t
he
i
nvest
igat
ordosnothav
econt
rol
overt
her
espondent
s.
 Usedonl
ywheneducat
edr
espondent
sar
efound.
 I
thasl
esscont
rol
overt
hequest
ionnai
reonceoutofhand.
 I
tbui
l
dsi
nfl
exi
bil
i
tyr
egar
dingt
oit
em cor
rect
ion.
Not
e:Despi
teal
lthesepr
obl
ems,quest
ionnai
rei
sthemosti
mpor
tantandwi
del
yused
met
hodofdat
acol
l
ect
ion.
3.
11.2I
nter
view
I
tis a popul
armet
hod i
n educat
ionalr
esear
ch and soci
alsci
ence (
Robson,1993)
.
I
nter
viewconsi
stsofor
alquest
ionsaskedbyt
hei
nter
viewer
sandor
alr
esponsesgi
venby
t
her
esear
chpar
ti
ci
pant
s.I
nter
viewt
ypi
cal
l
yinv
olv
esi
ndi
vi
dualr
esponsesbutt
her
eisan
i
nter
esti
nconduct
inggr
oupi
nter
view.Respondent
sty
pical
l
yspeaki
nthei
rownwor
dsand
t
hei
nter
viewer
sonei
theraudi
otapeorv
ideot
aper
ecor
dthei
rresponses.
Ty
pesofI
nter
view

47|
Page
Thecommont
ypesofi
nter
viewar
est
ruct
uredandunst
ruct
uredi
nter
views.
a.St
ruct
uredi
nter
view:I
tinv
olv
esaser
iesofcl
ose-
endedquest
ionst
hatei
therhav
eyes-
noansweroranswer
edbysel
ect
ingf
rom amongasetofshor
tanswerchoi
ces.The
r
espondent
s’answer
sar
enotf
oll
owedupt
oobt
aingr
eat
erdept
handt
heset
aket
hef
orm
ofcl
osedendedquest
ionnai
re.Theadv
ant
ageofani
nter
viewov
erquest
ionnai
re,
howev
er,
i
sthatt
her
esponser
atei
ncr
eased.Mor
eov
er,
thei
nter
viewercani
nter
actwi
thi
ndi
vi
dual
s
t
oreducet
henumberofunusabl
eor“
donotknow”r
esponses.
b.Unst
ruct
ured i
nter
view:I
tdoesnoti
nvol
ve a det
ail
ed i
nter
view gui
de i
nst
ead t
he
i
nter
viewerasksquest
ionst
hatgr
adual
l
yleadt
her
espondent
stopr
ovi
det
hedesi
red
i
nfor
mat
ion.Usual
l
ythet
ypeofi
nfor
mat
ionsoughti
sdi
ff
icul
tfort
her
espondentt
o
expr
ess ori
s psy
chol
ogi
cal
l
ysensi
ti
ve.Fort
hisr
eason,t
he i
nter
viewermustadapt
cont
inual
l
ytot
her
espondent
sst
ateofmi
nd.Thei
nfor
mat
ionsoughti
shi
ghl
ysubj
ect
ive
andt
imeconsumi
ng.
I
nter
viewerRol
e
Resear
cher
shav
edi
scov
eredmanyi
nter
viewerbehav
ior
sthataf
fectt
hequal
i
tyofdat
a
y
iel
dedbyt
hei
nter
viewmet
hod.Li
stsofbehav
ior
scompi
l
edf
rom v
ari
oussour
cesabout
t
her
oleoft
hei
nter
viewerpr
esent
edasf
oll
ows:
 Assur
erespondent
sofabsol
uteconf
ident
ial
i
tybef
orest
art
ingt
hei
nter
view.
 Bui
l
drappor
tbyengagi
ngasmal
ltal
kbef
orest
art
ingt
hei
nter
view.
 Sav
ecompl
exorcont
rov
ersi
alquest
ionatt
heendoft
hesessi
on.
 Expl
aint
hepot
ent
ial
benef
itoft
hest
udy
.
 Tal
klesst
hant
hei
nter
vieweedoes.
 Askquest
iont
hatcont
ainsonl
yasi
ngl
eidea.
 Usel
anguaget
hati
scl
earandunder
standabl
etot
her
espondent
.
 Av
oidl
eadi
ngquest
ions.
Adv
ant
agesandDi
sadv
ant
agesofI
nter
view
 Adv
ant
ages
 Adequat
einf
ormat
ionwi
thgr
eatdept
hobt
ained.
 Ther
eisapossi
bil
i
tyofgr
eat
erf
lexi
bil
i
tyi
nrest
ruct
uri
ngt
hequest
ionsespeci
all
yin
caseofunst
ruct
uredt
ypeofi
nter
view;t
her
eisar
oom t
ochangest
ruct
ureoft
he
quest
ions.
 Usedascompl
ement
.
 Usedt
ogetper
sonal
inf
ormat
ion.
 Sampl
escont
rol
l
edmor
eef
fect
ivel
yandwecani
dent
if
ythewi
l
li
ngnessandr
efusal

48|
Page
ofr
espondent
s.
 Usedt
oadaptt
hel
anguageofi
nter
viewt
othel
evel
ofr
espondent
s.
 Di
sadv
ant
ages
 I
tisv
eryexpensi
vei
nter
msoft
imeandot
herr
esour
ces.
 I
thast
hepossi
bil
i
tyofbi
asbyt
hei
nvest
igat
or.
 I
thasadef
ectt
hatcer
tai
nty
pesofr
espondent
sar
einaccessi
blef
ori
nter
view
becauseoft
hei
rpol
i
cyorbusi
ness.
 I
tdemandsgoodski
l
lsofi
nter
viewi
ng.
 I
tdoesnothel
ptogett
hought
sandf
eel
i
ngsofy
oungchi
l
dren.
3.
11.3FocusGr
oupDi
scussi
on(
FGD)
Af
ocus gr
oup di
scussi
on i
s a st
ruct
ured gr
oup pr
ocess used t
o obt
ain det
ail
ed
i
nfor
mat
ionaboutapar
ti
cul
art
opi
c.I
tispar
ti
cul
arl
yusef
ulf
orexpl
ori
ngat
ti
tudesand
f
eel
i
ngsandt
odr
awoutpr
eci
sei
ssuest
hatmaybeunknownt
other
esear
cher
.
Af
ocusgr
oupi
scomposedofsi
xtoni
nepar
ti
cipant
swhoar
ebr
oughtt
oget
hert
odi
scuss
acl
ear
lydef
inedt
opi
c.Ty
pical
l
y,f
ocusgr
oupsar
ecomposedofhomogeneouspeopl
e,al
l
r
epr
esent
ingapar
ti
cul
arsegmentoft
hepopul
ati
on.Af
ocusgr
oupsessi
onshoul
dlast
about1-
2hour
swi
tht
wohour
sbei
ngt
heabsol
utemaxi
mum t
ime.Agr
oupf
aci
l
itat
or
keepst
hedi
scussi
onont
rackbyaski
ngaser
iesofopen-
endedquest
ionsmeantt
o
st
imul
atedi
scussi
on.
Adv
ant
ages
 I
tisr
elat
ivel
yeasyt
ounder
take.
 I
tsr
esul
tobt
ainedi
nashor
tper
iod.
 Soci
ali
nter
act
ioni
nthegr
ouppr
oducesf
reerandmor
ecompl
exr
esponses;
 Ther
esear
chercanpr
obef
orcl
ari
fi
cat
ionandsol
i
citdet
ail
.
 Responseshav
ehi
ghf
acev
ali
dit
yduet
othecl
ari
tyoft
hecont
extanddet
ailoft
he
di
scussi
on.
Di
sadv
ant
ages
 I
trequi
reshi
ghl
yski
l
ledmoder
ator
.
 Gr
oupsar
eof
tendi
ff
icul
ttoassembl
e.
 I
ndi
vi
dual
responsesar
edependentofoneanot
her
.
 Becauset
hegr
oupi
shand-
sel
ect
ed,t
her
esul
tsmaynotber
epr
esent
ati
veoft
he
gener
alpopul
ati
on.
3.
11.
4.Obser
vat
ion
Anal
ter
nat
ivet
osel
f-
repor
tist
oobser
vet
hebehav
ioroft
hei
ndi
vi
dual
sst
udi
edi
nthei
r

49|
Page
nat
uralset
ti
ng.Resear
cher
sinbot
hqual
i
tat
iveandquant
it
ati
vet
radi
ti
onhav
edev
eloped
sy
stemat
icpur
posef
ort
hiscase.I
fusedpr
oper
ly,i
tav
oidst
hei
naccur
acyandbi
asof
somesel
f-
repor
tdat
a.I
templ
oyssenseofv
isi
onasi
tsmai
nsour
ce(
RobsonandHopki
ns,
1993)
.
Ther
ear
esomanyobser
vat
ionmet
hods;howev
er,i
nthi
stextt
hef
oll
owi
ngmet
hodsar
e
t
reat
ed:
I
.Par
ti
cipantobser
vat
ion:Togeti
nfor
mat
ionf
rom pat
ient
sint
hehospi
tal
,sl
eepasa
pat
ientandobser
veal
ltheact
ivi
ty.Her
ether
esear
cherbecomesamemberoft
hegr
oups,
whi
chhe/
shest
udi
es.Ther
esear
cherf
ull
yengagesi
nandobser
vest
hesi
tuat
ionf
rom
i
nsi
de.
I
I.Non-
par
ti
cipantobser
vat
ion:Ther
esear
cherobser
vest
heev
entf
rom out
side.He/
she
doesnoti
nvol
vei
nthel
i
feoft
hei
ndi
vi
dual
sthatar
eint
hest
udi
es.
Adv
ant
agesandDi
sadv
ant
agesofObser
vat
ion
 Adv
ant
ages
I
tpr
ovi
desi
nfor
mat
ionwhenot
hermet
hodsf
ail
.
 I
tisbet
terf
orl
esscompl
i
cat
edpr
obl
ems.
 I
tof
fer
sfi
rst
-handi
nfor
mat
ionwi
thoutdependi
ngonr
epor
tsofot
her
.
 I
tconsi
der
sreal
i
tyi
nit
snat
ural
set
ti
ng.
 I
tal
l
owst
hecol
l
ect
ionofwi
dei
nfor
mat
ion.
 Di
sadv
ant
ages
 Notusedwhent
hesampl
est
udi
edi
slar
gei
nnumber
.
 I
tpr
ovi
desnoi
nfor
mat
ionaboutpastorf
utur
e.
 I
tcannotst
udyi
ndi
vi
dual

sat
ti
tudedi
rect
ly.
 I
tex
posedt
oobser
ver
s’bi
as.
3.
11.5DocumentExpl
orat
ion
Document
s,whi
ch dealwi
th cur
ri
cul
um orot
hereducat
ionalconcer
n,can pr
esent
backgr
oundi
nfor
mat
ionpl
ainl
yandar
eimpor
tantsour
cesofdat
a(Hopki
ns,
1993)
.
Document
s Exampl
e
1.Publ
i
cdocument
s Yearbooks,ar
chi
ves,censusst
ati
sti
cs,mass-
medi
a,nov
els
andpoet
ry
2.Ar
chi
val
recor
dsSer
vicer
ecor
dsofhospi
tal
s,cl
i
nics,doct
ors,soci
alwor
ker
s,school
document
s,andr
ecor
dsoft
heor
gani
zat
ion.
3.Per
sonal
document
sLi
fehi
stor
ies,
diar
ies,
memor
anda,
aut
obi
ogr
aphi
esandl
ett
ers.
4.For
mal
studyr
epor
tear
chr
epor
tsofdi
ff
erentki
nds.

50|
Page
Adv
ant
agesandDi
sadv
ant
agesofDocumentAnal
ysi
s
 Adv
ant
ages
 I
tpr
ovi
des a cont
extf
orunder
standi
ng a par
ti
cul
arcur
ri
cul
um i
n educat
ional
r
esear
ch.
 I
tenabl
est
eacherr
esear
cher
sorsuper
visor
stost
udypastev
ent
sori
ssues.
 I
tgi
vesoppor
tuni
tyt
ofi
ndot
her
sper
cept
ionseasi
l
y.
 I
tiseconomi
cal
compar
edt
oot
her
s.
 Di
sadv
ant
ages
 Somedocument
smaynotbecompl
eteandupt
odat
e.
 Ther
eli
abi
l
ityofsomedocument
sisdoubt
ful
duet
owr
it
er’
sper
sonal
bias.
3.
11.
6Fi
eldNot
es
Wr
it
tenr
ecor
dofwhatt
her
esear
cherhear
sseesexper
iencesandt
hinksi
nthecour
seof
col
l
ect
ingdat
ainaqual
i
tat
ivest
udy
.Bogdan,etal
.
,(1992)descr
ibedt
hatf
iel
dnot
es
consi
stoft
wot
ypesofmat
eri
als.
1.Thedescr
ipt
ivepar
t:I
tist
hel
ongestpar
twher
ether
esear
cherobj
ect
ivel
yrecor
dst
he
det
ail
sofwhatoccur
redi
nthef
iel
d.I
tinv
olv
est
hef
oll
owi
ngpoi
nts.
i
.Por
trai
tsoft
hesubj
ect
sincl
udest
hei
rphy
sical
appear
ance,
dress,
manner
ism andst
yles
oft
alki
ngandact
ing:
i
i
.Reconst
ruct
ionofdi
aloguebet
weensubj
ect
srecor
ded.
i
i
i.Descr
ipt
ionofphy
sicalset
ti
ngl
i
kev
erbalsket
chesoft
hingssuchast
hebl
ackboar
d,
cont
ent
sofbul
l
eti
nboar
ds,
thef
urni
tur
eandt
hei
rar
rangement
set
c.
i
v.Apar
ti
cul
arev
ent
,whi
chi
ncl
udesal
i
stofi
ndi
vi
dual
s,i
nvol
vedi
ntheev
ent
,themanner
,
andt
henat
ureoft
heact
ionanddet
ail
descr
ipt
ionofbehav
ior
.
2.Ther
efl
ect
ivepar
toff
iel
dnot
es:I
naddi
ti
ont
othedescr
ipt
ivepar
t,f
iel
dnot
escont
ain
sent
encesandpar
agr
aphst
hatr
efl
ectt
her
esear
cher
smor
eper
sonalaccount
soft
he
cour
seoft
hest
udy
.Mor
eov
er,
Hopki
ns(
1993)i
dent
if
iedf
ouri
mpor
tantr
olesoff
iel
dnot
es
i
neducat
ional
resear
ch.
 Theycanf
ocusonapar
ti
cul
ari
ssueoft
eachi
ngbehav
iorov
eraper
iod.
 Theycanr
efl
ectgener
ali
mpr
essi
onsoft
hecl
assr
oom andi
tscl
i
mat
e.
 Theycanpr
ovi
deanongoi
ngdescr
ipt
ionofani
ndi
vi
dualchi
l
dthatcoul
dbeusedi
na
casest
udyand
 Theycanhel
pust
orecor
dourdev
elopmentast
eacher
s.
3.
11.
7Test
s
I
neducat
ionalr
esear
ch,t
est
sar
eusef
ulandusual
l
yappl
i
edi
nst
rument
.Theymeasur
e

51|
Page
aspect
sofhumanbehav
iorori
nnerper
sonalqual
i
ties.Psy
chol
ogi
calt
est
scanbeof
di
ff
erentt
ypes(
i.
e.achi
evementt
est
s,i
ntel
l
igencet
est
s,apt
it
udet
est
sandper
sonal
i
ty
i
nvent
ori
es)
.
Not
eInor
dert
oenhancet
hev
ali
dit
yandr
eli
abi
l
ityofourr
esear
chdat
a,combi
ningand
usi
ngdi
ff
erentdat
agat
her
ingt
ool
s,sayobser
vat
ionandi
nter
view,i
scal
l
edt
ri
angul
ati
on.
Ther
efor
e,i
tappear
slogi
calt
oempl
oydi
ff
erentmet
hodsofdat
agat
her
ingi
nst
rument
sin
school
basedr
esear
ch.

UNI
TFOUR

Dat
aPr
ese
ntat
i
on,
Anal
ysi
sandI
nte
rpr
etat
i
on

4.
1Dat
aPr
esent
ati
on

Thenextst
epaf
tercol
l
ect
ingdat
amustbecl
assi
fyi
ngandpr
esent
ingi
tinameani
ngf
ul
f
ormt
ohav
ebet
teri
nsi
ghti
nar
esear
chpr
obl
em.Di
ff
erentt
ool
sofcl
assi
fi
cat
ionand
pr
esent
ati
onofr
esear
chdat
aar
efr
equencydi
str
ibut
ion,
cumul
ati
vef
requencydi
str
ibut
ion,
r
elat
ivef
requencydi
str
ibut
ionandchar
ts.

4.
1.1Fr
equencyDi
str
ibut
ion
Ar
awdat
aar
eacol
l
ect
eddat
athathav
enotbeenor
gani
zednumer
ical
l
y.Anar
rayi
san
ar
rangementofr
awnumer
icaldat
ainascendi
ngordescendi
ngor
derofmagni
tude.The
di
ff
erencebet
weent
hel
argestandsmal
l
estnumberi
scal
l
edt
her
angeoft
hedat
a.I
fdat
a
ar
eofr
epeat
ingnat
ure,
thent
heyshoul
dbepr
esent
edi
nfor
msoft
hedat
aofapar
ti
cul
ar
t
ype.Whensummar
izi
ngl
argemassesofr
awdat
a,i
tisof
tenusef
ult
odi
str
ibut
ethedat
a
i
ntocl
assesorcat
egor
iesandt
odet
ermi
net
henumberofi
ndi
vi
dualbel
ongi
ngt
oeach

52|
Page
cl
ass,cal
l
edt
hecl
assf
requency
.At
abul
arar
rangementofdat
abycl
assest
oget
herwi
th
t
hecor
respondi
ngcl
assf
requenci
esi
scal
l
edf
requencydi
str
ibut
ionorf
requencyt
abl
e.
Thef
requencydi
str
ibut
ioncanbecl
assi
fi
edi
ntodi
scr
etef
requencydi
str
ibut
ionand
cont
inuousf
requencydi
str
ibut
ionwhi
char
edemonst
rat
edi
ntabl
e6.
1andt
abl
e6.
2
respect
ivel
y.
Table4.
1.Discr
etef
requencyDi
str
ibut
ion
I
ncomecat
egor
y Numberofr
espondent
s(f
requency)
Lowi
ncome 300
Medi
um i
ncome 200
Hi
ghi
ncome 100

Tabl
e4.
2.Cont
inuousf
requencyDi
str
ibut
ion

Monthlyincomesi
n(Bi
rr) Numberofr
espondent
s(f
requency)
cl
assinterval
s
0-
500 100
500-
1000 170
1000-
1500 30
1500-
2000 100
2000-
2500 60
2500-
3000 40
3000-
3500 40
3500-
4000 30
4000-
4500 30

Tot
al 600

Thedat
aor
gani
zedandsummar
izedasi
ntheabov
efr
equencydi
str
ibut
ionar
eof
tencal
l
ed

53|
Page
gr
oupeddat
a.Asy
mbol
def
ini
ngacl
asssuchas(
0-500)i
ntabl
e6.
2iscal
l
edcl
assi
nter
val
.
Theendnumber0and500,ar
ecal
l
edcl
assl
i
mit
s;t
hesmal
l
estnumber(
o)i
sthel
ower
cl
assl
i
mit
,andt
hel
argernumber(
500)i
stheuppercl
assl
i
mit
.Thet
erm cl
assandcl
ass
i
nter
valar
eof
tenusedi
nter
changeabl
yal
thought
hecl
assi
nter
vali
sact
ual
l
yasy
mbolf
or
t
hecl
ass.Acl
assi
nter
val
,atl
eastt
heor
eti
cal
l
y,hasei
thernouppercl
assl
i
mitornol
ower
cl
assl
i
miti
ndi
cat
edi
scal
l
edanopencl
assi
nter
val
.Forex
ampl
e,r
efer
ri
ngt
oincome
sour
cesofi
ndi
vi
dual
s,t
hecl
assi
nter
val
“4500Bi
rrandabov
e”i
sanopencl
assi
nter
val
.

Cl
assBoundar
ies

I
fthei
ncomesour
cesar
erecor
ded t
othenear
estbi
rrt
hecl
assi
nter
val4000-
4500
t
heor
eti
cal
l
yincl
udesal
lval
uesf
rom 3999.
5-4500.
5bi
rr
.Thesenumber
s,i
ndi
cat
edbr
ief
ly
byt
heexactnumber
s3999.
5and4500.
5,ar
ecal
l
edcl
assboundar
ies,
ort
ruecl
assl
i
mit
s.
Thesmal
l
ernumber(
3999.
5)i
sthel
owercl
assboundar
yandt
hel
argernumber(
4500.
5)
i
stheuppercl
assboundar
y.I
npr
act
ice,t
hecl
assboundar
iesar
eobt
ainedbyaddi
ngt
he
upperl
i
mitofonecl
assi
nter
valt
othel
owerl
i
mitoft
henext
-hi
ghercl
assi
nter
valand
di
vi
dingby2.Somet
imescl
assboundar
iesar
eusedt
osy
mbol
i
zecl
asses..

Thesi
zeorwi
dthofacl
assi
nter
val

Thesi
zeorwi
dthofcl
assi
nter
vali
sthedi
ff
erencebet
weent
hel
oweranduppercl
ass
boundar
iesandi
sal
sor
efer
redt
oast
hecl
asswi
dth,
classsi
ze,
orcl
assl
engt
h.I
fal
lcl
ass
i
nter
val
sofaf
requencydi
str
ibut
ionhav
eequal
widt
hs,
thi
scommonwi
dthi edbyC.
sdenot
nsuchcaseCi
I sequalt
othedi
ff
erencebet
weent
wosuccessi
vel
owercl
assl
i
mit
sort
wo
successi
veuppercl
assl
i
mit
s.

TheCl
assMar
k

Thecl
assmar
kist
hemi
dpoi
ntoft
hecl
assi
nter
valandi
sobt
ainedbyaddi
ngt
hel
ower
anduppercl
assl
i
mit
sanddi
vi
dingbyt
wo.Thus,
forexampl
ethecl
assmar
kofi
nter
val
(0-
500)i
s
f
0+500^
1
=250.Thi
scl
assmar
kisal
socal
l
edt
hecl
assmi
dpoi
nt.
V 2 J

Gener
alRul
esf
orFor
mingFr
equencyDi
str
ibut
ions

1.
Det
ermi
net
hel
argestandsmal
l
estnumber
sint
her
awdat
aandt
husf
indt
her
ange.

54|
Page
2.
Div
idet
her
angei
ntoconv
eni
entnumberofcl
assi
nter
val
shav
ingt
hesamesi
ze.I
fthi
s
i
snotf
easi
ble,usecl
assi
nter
val
sofdi
ff
erentsi
zesoropencl
assi
nter
val
s.Thenumber
ofcl
assi
nter
val
sisusual
l
ytakenbet
ween5and20,dependi
ngont
hedat
a.Thecl
ass
i
nter
val
sar
eal
sochosensot
hatt
hecl
assmar
kscoi
nci
dewi
tht
heact
ual
l
yobser
veddat
a.

3.
Det
ermi
net
henumberofobser
vat
ionf
all
i
ngi
ntoeachcl
assi
nter
val
,thati
s,f
indt
he
cl
assf
requenci
es.Thi
scanbedonebyusi
ngat
all
y,orscor
esheet
.

4.
1.2Cumul
ati
veFr
equencyDi
str
ibut
ion

Thecumul
ati
vef
requencydi
str
ibut
ioni
samodi
fi
edf
orm off
requencydi
str
ibut
ion,as
shownbel
ow.Thenew f
requencyobt
ainedbyaddi
ngi
ndi
vi
dualf
requenci
esofcl
ass
i
nter
val
sar
ecal
l
edcumul
ati
vef
requenci
es.Cumul
ati
veper
cent
agef
requencycanbe
obt
ainedbymul
ti
ply
ingcumul
ati
vef
requencyby100/
N,wher
eN=numberofscor
es.

Tabl
e6.
3.Cumul
ati
vef
requencydi
str
ibut
ionoft
het
estscor
esof100AMUst
udent
s

Cl
assi
nt val Exactuni
er t
sof Mi
dpoi
nt Fr
equency Cumulati
ve Cumul ati
ve
cl
assinter
val
s fr
equency percentage
fr
equency

72-
74 71.
5-74.
5 73 8 92+8=100 100%
69-
71 68.
5-71.
5 70 27 65+27=92 92%
66-
68 65.
5-68.
5 67 42 23+42=65 65%

63-
65 62.
5-65.
5 64 18 5+18=23 23%

60-
62 59.
5-62.
5 61 5 5=5 5%

Tot
alN=100

6.
1.3Rel
ati
veCumul
ati
veFr
equency

Ther
elat
ivef
requencydi
str
ibut
ioni
samodi
fi
edf
orm off
requencydi
str
ibut
ion,
asshown
i
ntabl
e6.
3.Thev
alueofr
elat
ivecumul
ati
vef
requencydi
str
ibut
ioncanbedet
ermi
nedby
55|
Page
di
vi
dingeachf
requencyt
othet
otalf
requencyorbydi
vi
dingcumul
ati
veper
cent
age
f
requencyt
o100.

4.
1.4Char
ts
Resear
chdat
acanbepr
esent
edi
ngr
aphi
calf
orm(
orchar
ts)whi
char
ecl
assi
fi
edas:pi
e
char
t,barchar
t,st
ackedbarchar
t,hi
stogr
am,
andf
requencypol
ygonandogi
vecur
ve.

Pi
e char
t:i
s a char
tin whi
ch t
he ar
ea ofa ci
rcl
eis di
vi
ded i
nto di
ff
erentsect
ors
r
epr
esent
ingdi
ff
erentcat
egor
iessucht
hatt
heangl
eofeachsect
ori
spr
opor
ti
onalt
othe
f
requencyoft
hecor
respondi
ngcat
egor
y.Lookatt
hepi
echar
t,i
nfi
g.6.
1basedont
he
i
nfor
mat
ionont
abl
e6.
1,t
hepr
opor
ti
onsi
nthepi
echar
t:

Fi
g.4.
1Pi
echar
tofi
ncomedi
str
ibut
ion

Barchar
t:i
sagr
aphi
cal
viewoft
hegi
vendat
asucht
hatt
hef
requencyofeachcat
egor
yis
shownasav
ert
ical
str
ipagai
nstt
hatcat
egor
yinpr
opor
ti
ont
othehei
ghtofot
hersuch
v
ert
ical
str
ips.

Fi
g.6.
2.Barchar
tofi
ncomedi
str
ibut
ion

56|
Page
4.
2Quant
it
ati
veAnal
ysi
sandI
nter
pret
ati
onofResear
chDat
a

The r
esear
ch dat
a ofa gi
ven si
tuat
ion shoul
d be char
act
eri
zed bysome st
ati
sti
cal
measur
esf
ort
hepur
poseofest
imat
ionorcompar
isonwi
thsi
mil
ardat
aormaki
ng
i
nfer
enceaboutt
hesampl
epopul
ati
ont
owhi
cht
hedat
abel
ongs.Toanal
ysi
squant
it
ati
ve
dat
ast
ati
sti
cal
l
y,t
her
esear
chmayuseei
therdescr
ipt
ivest
ati
sti
csori
nfer
ent
ial
stat
ist
ics.

4.
2.1Descr
ipt
iveSt
ati
sti
cs

Decept
ivest
ati
sti
calmeasur
esar
eusedt
odescr
ibet
hechar
act
eri
sti
csoft
hesampl
eor
popul
ati
oni
ntot
ali
ty.I
tsgener
ali
zat
ioni
sli
mit
edt
othepar
ti
cul
argr
oupofi
ndi
vi
dual
s
obser
vedorst
udi
ed.Noconcl
usi
onsar
eext
endedbey
ondt
hisgr
oup.Thest
ati
sti
cal
anal
ysi
sbasedont
hecomput
ati
onofdescr
ipt
ivest
ati
sti
calmeasur
esi
smost
lyappl
i
edi
n
act
ionr
esear
ch(
teacherr
esear
ch)
,andpr
ovi
dev
aluabl
einf
ormat
ionaboutt
henat
ureofa
par
ti
cul
argr
oupandt
hatgr
ouponl
y.Thedescr
ipt
ivest
ati
sti
cst
hatcommonl
yusedi
n
educat
ional
resear
char
e:

I
.Measur
esofcent
ral
tendencyorav
erage;
I
I.Measur
esofspr
ead,
disper
sionorv
ari
abi
l
ity
;
I
II
.Measur
esofr
elat
iveposi
ti
on;
I
V.Measur
esofr
elat
ionshi
p.

4.
2.1.
1Measur
esofCent
ral
Tendency
Thet
hreemostcommonl
yusedmeasur
esofcent
ralt
endencyar
ethemean,t
hemedi
an
hemode.
andt

a)Themean(
M):Themeanofadi
str
ibut
ioni
scommonl
yunder
stoodast
hear
it
hmet
ic
meanorav
erage.I
tist
hemostwi
del
yusedmeasur
eofcent
ralt
endency
.Iti
sdef
inedas
t
her
ati
obet
weent
hesum oft
heobser
vat
ionsandt
henumberofobser
vat
ions.

Ar
it
hmet
icMeanf
orungr
oupeddat
a:Thef
ormul
ausedt
ofi
ndt
hemeanf
orungr
ouped
dat
acanbe:

M =i1,Wher
e:M=mean
N
^=sum of
57|
Page
t
h
Xi
=thei obser
vat
ion

N=t
otal
numberofobser
vat
ion

4.
2.1.
2Measur
esofVar
iat
ion

The measur
es ofcent
ralt
endency ar
ethe est
imat
es oft
he av
erage ofa setof
obser
vat
ions.The av
erage v
alue does not expl
aint
he di
ff
erences bet
ween t
he
obser
vat
ions.Thi
sgapcanbef
ulf
il
ledbyappl
yi
ngmeasur
ementofv
ari
ati
on.Thev
ari
ati
on
(
spr
ead)i
s a measur
e oft
he ext
entt
o whi
ch t
he i
ndi
vi
dualobser
vat
ions v
arywi
th
r
efer
encet
otheav
erageoft
hegi
vensetofobser
vat
ion.Thedi
ff
erentmeasur
ersof
v
ari
ati
ons,whi
ch ar
e di
scussed i
nthi
s modul
e,ar
erange,quar
ti
le dev
iat
ion,mean
dev
iat
ion,
standar
ddev
iat
ion,
andcoef
fi
cientofv
ari
ati
on.

a)Range:Rangei
sthedi
ff
erencebet
weent
hehi
ghestv
alueandt
hel
owestv
alueoft
he
gi
vensetof
obser
vat
ions.

b)Quar
ti
leDev
iat
ion:Quar
ti
ledev
iat
ioni
sthehal
foft
hedi
ff
erencebet
weent
het
hir
d
quar
ti
le
(
Q3)andt
hef
ir
stquar
ti
le(
Q1)
.Thet
hir
dquar
ti
lei
sthev
alueoft
hev
ari
abl
eofi
nter
est
cor
respondi
ngt
othef
ir
st75%oft
hef
requencyofgr
oupeddat
a,wher
east
hef
ir
stquar
ti
le
i
sthe
v
alueoft
hev
ari
abl
eofi
nter
estcor
respondi
ngt
othef
ir
st25% oft
het
otalf
requencyof
gr
oupedof
t
hedat
a.

c)Av
erageDev
iat
ion:
Meandev
iat
ionorav
eragedev
iat
ion(
AD)i
sthemeanoft
he
absol
utedev
iat
ionoft
hei
ndi
vi
dual
obser
vat
ionf
rom ei
therar
it
hmet
icmeanor
medi
anoft
hegi
vensetofobser
vat
ions.

d)St
andar
dDev
iat
ion:
Standar
ddev
iat
ioni
sthewi
del
yusedmeasur
eofv
ari
ati
on.
Al
ter
nat
ivel
yiti
scal
l
edr
ootmeansquar
edev
iat
ionandi
tsnot
ati
oni
s“a”f
orpopul
ati
on
dat S”f
aor“ orsampl
edat
a.I
tist
hesquar
erootoft
hev
ari a2or
ance, S2 St
andar
ddev
iat
ion
f
orungr
oupeddat
a:

e)Coef
fi
cientofVar
iat
ion:Coef
fi
cientofv
ari
ati
oni
sanot
hermeasur
eofv
ari
ati
on,
whi
chi
sani
mpr
ovedmeasur
eov
erst
andar
ddev
iat
ion.I
tist
her
ati
obet
weent
he

58|
Page
st
andar
ddev
iat
ionandmean,andi
tisusual
l
yrepr
esent
edbyper
cent
age.Thi
s
measur
eindi
cat
es t
he consi
stencyoft
he obser
vat
ions wi
thi
n a sampl
e and
di
scr
imi
nat
esagi
vensetofsampl
esshar
plyi
nter
msofconsi
stency
.

4.
2.1.
3Measur
eofr
elat
iveposi
ti
ons

Ar
aw scor
e on t
est
,taken byi
t,hasno meani
ng.I
tget
smeani
ng onl
yby
compar
isonwi
thsomer
efer
encegr
ouporgr
oups.Thecompar
isonmaybedone
wi
tht
hehel
poft
hemeasur
essuchassi
gmascor
e(<rscor
es)
,st
andar
dscor
es(
Z
orTscor
es)
,per
cent
il
es,
andper
cent
il
eranks.

a)Si es(
gmascor ascor
es)
:Indescr
ibi
ngascor
einadi
str
ibut
ion,i
tsdev
iat
ion
f
rom t
hemeanexpr
essedi
nst
andar
ddev
iat
ionuni
tsi
sof
tenmor
emeani
ngf
ul
t
hant
hescor
eit
sel
f.Theuni
toft
hemeasur
ementi
sthest
andar
ddev
iat
ion.

b)St
andar
dScor
es(
T)orT-
scor
es:St
andar
dscor
esi
sat
ypeofder
ivedscor
ethat
usesst
andar
ddev
iat
ionuni
tst
oindi
cat
eani
ndi
vi
dual

sper
for
mancer
elat
ivet
othe
nor
m oft
hegr
oup’
sper
for
mance.

c
)Per
cent
il
es:Per
cent
il
es ar
ethe poi
nts whi
ch di
vi
de t
he ent
ir
e scal
e of
measur
ementi
nto100equalpar
ts.Thet
ent
hper
cent
il
emaybedef
inedast
hat
poi
nti
naf
requencydi
str
ibut
ionbel
ow whi
ch10per
centoft
het
otalmeasur
eof
scor
esf
all
.Iti
sal
soev
identt
hatt
hemedi
an,
expr
essedasaper
cent
il
eisP50.

d)Per
cent
il
eRanks:Theper
cent
il
eranki
sthepoi
nti
nthedi
str
ibut
ionbel
owwhi
chagi
ven
per
cent
ageofscor
esf
all
.I he90thper
ft cent
il
eranki
sascor
eof70,
90%oft
hescor
esf
all
bel
ow70.Themedi
ani
sthe50th per
cent
il
erank,f
or50% t
hescor
esf
allbel
ow i
t.The
cal
cul
ati
onoft
heper
cent
il
eranksi
sther
ever
sepr
ocessofcal
cul
ati
onofPer
cent
il
eto
pi
nts.Thef
ormul
atoobt
aint
heper
cent
il
eranki
s.

4.
2.1.
4Measur
esofRel
ati
onshi
p:Cor
rel
ati
on

Cor
rel
ati
oni
sther
elat
ionshi
pbet
weent
woormor
epai
redv
ari
abl
esort
woormor
eset
sof
dat
a.The degr
ee ofr
elat
ionshi
ps i
s measur
ed and pr
esent
ed byt
he coef
fi
ci
entof
59|
Page
cor
rel
ati
on.Most
ly,i
tisdenot
edby“
r”.I
nresear
ch,v
ari
abl
escanbecor
rel
atedposi
ti
vel
y,
negat
ivel
yand t
her
ear
eot
hert
rai
tst
hatpr
obabl
yhav
eno cor
rel
ati
on.Thepai
rsof
v
ari
abl
est
hatusual
l
ycor
rel
atedposi
ti
vel
y(asonei
ncr
easet
heot
hert
endst
oincr
eases)
ar
edescr
ibedbel
owi
nthet
abl
e.

4.
2.2I
nfer
ent
ial
Stat
ist
ics

I
nfer
ent
ialst
ati
sti
calanal
ysi
sal
way
sinv
olv
est
hepr
ocessofsampl
i
ngandt
he
sel
ect
ionofasmal
lgr
oupassumedt
ober
elat
edt
othepopul
ati
onf
rom whi
chi
tis
dr
awn.Thel
argegr
oupi
sknownast
hepopul
ati
on,andt
hesmal
lgr
oupi
sthe
sampl
e.St
ati
sti
cisameasur
ebasedonobser
vat
ionsoft
hechar
act
eri
sti
csofa
sampl
e.Ast
ati
sti
ccomput
edf
rom asampl
emaybeusedt
oest
imat
eapar
amet
er,
t
hecor
respondi
ngv
aluei
nthepopul
ati
onf
rom whi
cht
hesampl
eissel
ect
ed.

The mai
n pur
pose of i
nfer
ent
ialst
ati
sti
calanal
ysi
sther
efor
e,i
sto dr
aw
concl
usi
onsaboutpopul
ati
onsbasedonobser
vat
ionsofsampl
es.Est
imat
ingor
i
nfer
ri
ngapopul
ati
onchar
act
eri
sti
c(par
amet
er)f
rom ar
andom sampl
e(st
ati
sti
c)
i
snotanexactpr
ocess.I
thasbeennot
edt
hatsuccessi
vemeansofr
andoml
y
sel
ect
edsampl
esf
rom t
hesamepopul
ati
onar
enoti
dent
ical
.Thus,
ift
hesemeans
ar
enoti
dent
ical
,itwoul
dbel
ogi
calt
oassumet
hatanyoneoft
hem pr
obabl
y
di
ff
ersf
rom t
hepopul
ati
onmean.Thi
swoul
dseem t
opr
esentani
nsur
mount
abl
e
obst
acl
eto sat
isf
act
ions,f
ort
heyhav
eonl
yasampl
eto useasabasi
sfor
gener
ali
zat
ions about a popul
ati
on. St
ati
sti
cs (
sampl
e dat
a) ar
e usual
l
y
r
epr
esent
edbyl
ett
er phabetsuchasx,s,andr
sofRomanal .Ont
heot
herhand,
par
amet
ers(
popul
ati
on dat
a)ar
eusual
l
yrepr
esent
ed byl
ett
ersoft
heGr
eek
phabetsuchasj
al u,
a,
andp.

I
nfer
ent
ialst
ati
sti
calanal
ysi
s may be i
n any one oft
he f
oll
owi
ng f
orms:
si
gni
fi
cance ofdi
ff
erence bet
ween means,anal
ysi
s ofv
ari
ance,anal
ysi
s of
cov
ari
ance,
cor
rel
ati
onmet
hod,
chi
-squar
etest
,theMann-Whi
tneyt
estet
c.

60|
Page
4.
3Qual
i
tat
iveAnal
ysi
sandI
nter
pret
ati
onofResear
chDat
a

The t
ype oft
ool
s used byt
he r
esear
cher
sforcol
l
ect
ing dat
a det
ermi
nes t
he dat
a
anal
yzi
ngt
echni
ques.Thedat
agat
her
edt
hroughopenendedquest
ionnai
res,i
nter
views,
obser
vat
ions,document
s,et
c.ar
emost
lyusedt
o anal
ysi
squal
i
tat
ivel
y.Her
e,abr
ief
descr
ipt
ionofqual
i
tat
ivedat
aanal
ysi
sispr
ovi
ded.Cur
rent
ly,comput
ersof
twar
epr
ogr
am
cal
l
ed NUD*
IST (
Non-
Numer
icalUnst
ruct
ured Dat
aIndexi
ng,Sear
chi
ng and Theor
y-
Bui
l
ding)i
sdesi
gnedt
obeusedt
oor
gani
zeandanal
yzequal
i
tat
ivedat
a.Howev
er,t
her
e
ar
enot
est
sofr
eli
abi
l
ityandv
ali
dit
yandt
her
ear
enocommonl
yagr
eed-
uponpr
ocedur
es.
Thedat
aanal
ysi
stechni
quet
hroughcomput
ersof
twar
e(f
orbot
hquant
it
ati
ve&qual
i
tat
ive
dat
a)i
snoti
ncl
udedi
ntheobj
ect
iveoft
hiscour
se.Onewhoi
sint
erest
edi
nthi
scomput
er
sof
twar
ecanl
ear
nindependent
ly.

Thef
ir
stst
epi
nanal
yzi
ngqual
i
tat
iver
esear
chr
ecor
dsi
nvol
vesor
gani
zi
ngt
hedat
a.The
met
hodofor
gani
zi
ngt
hesedat
awi
l
ldi
ff
erdependi
ngont
her
esear
chmet
hodanddat
a
col
l
ect
iont
echni
quesused.Oncet
hedat
ahav
ebeenor
gani
zed,t
her
esear
chercanmov
e
t
odat
aanal
ysi
sst
age,
descr
ipt
ion.Ther
esear
cherdescr
ibest
hev
ari
ousper
ti
nentaspect
s
oft
hest
udyi
ncl
udi
ngt
heset
ti
ng,bot
htempor
all
yandphy
sical
l
y;t
hei
ndi
vi
dual
sbei
ng
st
udi
ed,t
hepur
posesofanyact
ivi
tyexami
ned,t
hev
iewpoi
ntsofpar
ti
cipant
s,andt
he
ef
fectofanyact
ivi
tyont
hepar
ti
cipant
s.

Asaf
inalpoi
nt,t
heor
gani
zeddat
amustbei
nter
pret
ed.I
nter
pret
ati
oni
nvol
vesexpl
aini
ng
t
hef
indi
ngs,answer
ing”why
”quest
ions,at
tachi
ngsi
gni
fi
cancet
opar
ti
cul
arr
esul
ts,and
put
ti
ngpat
ter
nsi
ntoananal
yti
cfr
amewor
k.I
tisappeal
i
ngt
orushi
ntot
hecr
eat
ivewor
kof
i
nter
pret
ingt
hedat
abef
oredoi
ngt
hedet
ail
edhandwor
kofput
ti
ngcoher
entanswert
o
maj
ordescr
ipt
ivequest
ions.Nev
ert
hel
essdescr
ipt
ioncomesf
ir
st.Thedi
sci
pli
neandr
igor
ofqual
i
tat
iveanal
ysi
sdependonpr
esent
ingsol
i
ddescr
ipt
ivedat
a.Thei
nter
pret
ati
onof
qual
i
tat
iver
esear
chdat
aismor
edependentont
her
esear
cher
sbackgr
ound,
ski
l
ls,
biases,
andknowl
edget
hanonconcl
usi
onsdr
awnf
rom quant
it
ati
ver
esear
ch,t
hatar
eder
ived
mor
edi
rect
lyf
rom t
henumer
icalanal
ysesoft
hedat
a.I
nsuchaway
,ot
her
swhoar
e
r
eadi
ngt
heser
esul
tscanunder
standanddr
awt
hei
rowni
nter
pret
ati
ons.

Last
ly,t
her
esear
chershoul
dkeepi
nmi
ndt
hei
ssuesofi
nter
nalv
ali
dit
yandext
ernal
v
ali
dit
yofqual
i
tat
iver
esear
chj
ustast
hiswasdonei
nexper
iment
alr
esear
ch.I
nter
nal

61|
Page
v
ali
dit
yisconcer
nedwi
tht
heaccur
acyoft
hei
nfor
mat
ionandhowi
tmat
cheswi
thr
eal
i
ty.
I
ntheot
herway
,ext
ernalv
ali
dit
yofqual
i
tat
iver
esear
chdat
aanal
ysesi
sconcer
nedwi
th
r
esear
cher
s’needst
odi
scusst
hel
i
mit
edgener
ali
ze-
abi
l
ityoft
hef
indi
ngsandt
heneed;
i
f
possi
blet
orepl
i
cat
ethest
udyandi
tsf
indi
ng.

WhatI
sEt
hicsI
nResear
chandWhyI
sItI
mpor
tant
?

Whenmostpeopl
ethi
nkofet
hics(
ormor
als)
,theyt
hinkofr
ulesf
ordi
sti
ngui
shi
ng
bet
weenr
ightandwr
ong.Butt
hemostcommonwayofdef
ini
ng"
ethi
cs"i
sthat
nor
msf
orconductt
hatdi
sti
ngui
shbet
weenaccept
abl
eandunaccept
abl
ebehav
ior
.
Ther
ear
esev
eral
reasonswhyi
tisi
mpor
tantt
oadher
etoet
hical
nor
msi
nresear
ch.
Fi
rst
,nor
mspr
omot
etheai
msofr
esear
ch,
suchasknowl
edge,
trut
h,andav
oidance
ofer
ror
. Second,si
ncer
esear
chof
teni
nvol
vesagr
eatdealofcooper
ati
onand
coor
dinat
ionamongmanydi
ff
erentpeopl
eindi
ff
erentdi
sci
pli
nesandi
nst
it
uti
ons,
et
hicalst
andar
dspr
omot
ethev
aluest
hatar
eessent
ialt
ocol
l
abor
ati
vewor
k,such
ast
rust
,account
abi
l
ity
,mut
ual
respect
,andf
air
ness

Thi
rd,manyoft
heet
hicalnor
mshel
ptoensur
ethatr
esear
cher
scanbehel
d
account
abl
etot
hepubl
i
c.Four
th,
ethi
cal
nor
msi
nresear
chal
sohel
ptobui
l
dpubl
i
c
suppor
tforr
esear
ch.Peopl
emor
eli
kel
ytof
undr
esear
chpr
ojecti
ftheycant
rustt
he
qual
i
tyandi
ntegr
it
yofr
esear
ch.Fi
nal
l
y,manyoft
henor
msofr
esear
chpr
omot
ea
v
ari
etyofot
heri
mpor
tantmor
alandsoci
alv
alues,suchassoci
alr
esponsi
bil
i
ty,
humanr
ight
s,andani
malwel
far
e,compl
i
ancewi
tht
hel
aw,andheal
thandsaf
ety
.
Et
hicalconcer
nsshoul
dbeatt
hef
oref
rontofanyr
esear
chpr
ojectandshoul
d
cont
inuet
hrought
othewr
it
e-upanddi
ssemi
nat
ionst
ages.Thef
oll
owi
ngi
sar
ough
andgener
alsummar
yofsomeet
hical
pri
nci
plest
hatv
ari
ouscodesaddr
ess:

 Honest
y,

 Obj
ect
ivi
ty,

62|
Page
 I
ntegr
it
y,

 Car
eful
ness,

 Openness,

 Respectf
orI
ntel
l
ect
ual
Proper
ty,

 Conf
ident
ial
i
ty,

 Responsi
blePubl
i
cat
ion,

 Responsi
bleMent
ori
ng,

 Respectf
orcol
l
eagues,

 Soci
alResponsi
bil
i
ty,

 Non-
Discr
imi
nat
ion,

 Compet
ence,
Legal
i
ty,

 Ani
mal
Car
e,

 HumanSubj
ect
sPr
otect
ion,

 Pl
agi
ari
sm

Gener
alFor
matoft
heResear
chRepor
t

63|
Page
SomeUniver
sit
ies,r
esearchorganizat
ionsorj
our
nal
boardshaveest
abl
ishedt
hei
rown
for
mattowhichthei
rresearchpapermustconfor
m,butall
for
matsaresomewhatsi
mil
ar
tothef
oll
owingsecti
ons(Koul,1997).

 Thepr
el
imi
nar
ysect
i
on
 Thebodyoft
her
epor
tort
extand
 Ther
efer
encesect
i
onPr
eli
minar
y
Sect
ion
1
.Ti
tl
epage
2.
Pref
ace(
Acknowl
edgement
)
3
.Tabl
eofcont
ent
s
4.
Listoft
abl
es
5
.Li
stoff
igur
es
Bodyoft
heRepor
torText
a)I
ntr
oduct
ion(
noheadi
ng)
1
.Backgr
ound
2.
Stat
ementoft
hepr
obl
em
3.
Anal
ysi
sofpr
evi
ousr
esear
ch
4.
Signi
fi
canceoft
hepr
obl
em
5
.Del
i
mit
ati
onoft
hest
udy
6.
Stat
ementofhy
pot
hesesandassumpt
ionsunder
lyi
ngi
t
7.
Oper
ati
onal
def
ini
ti
onoft
erms
b)
Desi
gnoft
hest
udy
1
.Subj
ect
s
2.
Dat
acol
l
ect
ioni
nst
rument
s
3.Pr
ocedur
es

c
)Anal
ysi
sandi
nter
pret
ati
onoft
hedat
a
1
.Tex
t
2.Tabl
e(i
fany
,ar
eusual
l
yincl
udedi
ntot
het
ext
)
3.Fi
gur
es(
ifany
,ar
eusual
l
yincl
udedi
ntot
het
ext
)
d)
.Summar
yandconcl
usi
on

64|
Page
1
.Br
iefr
est
atementofpr
obl
em andpr
ocedur
es
2.
Pri
nci
pal
findi
ngsandconcl
usi
onswi
tht
hei
rpr
act
ical
impl
i
cat
ions(
ifany
)
3
.Suggest
ionf
orf
urt
herr
esear
ch
Ref
erenceSect
ion
1
.Ref
erences
2.
Appendi
c

65|
Page
66|
Page

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy