In
te
r
n
ation
a
l Jou
rn
al
o
f E
v
al
u
a
t
i
on
a
n
d
R
e
se
arc
h
in
Ed
u
c
ation
(
IJERE
)
V
o
l.7,
N
o.4,
D
ece
mbe
r
201
8,
pp.
3
1
3
~
3
1
6
IS
S
N
: 2252-
88
22
D
O
I
: 10.
1
1
591
/i
jere
.v7
i
4.15
1
4
6
313
Jou
rn
a
l
h
o
me
pa
ge
:
ht
tp:
//i
a
e
score
.
com
/
j
o
u
r
na
l
s
/
i
n
d
e
x
.
p
hp/IJ
ERE
/
Using P
a
rtial Credit Model t
o Im
prove the Quality of
an Instr
ument
Sa
r
a
s
K
r
i
s
h
n
a
n
1
,
N
o
r
a
ini Idris
2
1
I
N
T
I
I
n
te
r
n
a
t
i
o
n
a
l
U
n
iv
e
r
si
ty
,
M
a
la
y
s
i
a
2
U
n
iv
ersi
ti
M
ala
y
a
,
M
alay
s
i
a
Art
i
cl
e In
fo
ABSTRACT
A
r
tic
le hist
o
r
y
:
R
e
c
e
i
v
e
d
Au
g
2
1
,
2
018
Re
vise
d S
e
p 4,
201
8
Ac
ce
p
t
ed
Oc
t
1
0
,
2
018
In
u
si
ng
t
h
e
R
asch
m
od
e
l
t
o
i
m
prov
e
th
e
qu
ali
t
y
of
a
n
in
st
ru
men
t,
a
na
ly
si
s
pu
rpo
r
ts
t
o
d
e
term
in
e
if
t
h
e
s
amp
l
e
co
ll
abo
r
ates
w
ell
w
i
t
h
t
h
e
it
em
s
i
n
t
h
e
in
s
t
rum
e
n
t
s
u
c
h
that
t
h
e
r
esult
s
a
re
m
eas
uri
n
g
a
si
ngle
u
n
d
e
rl
y
ing
vari
abl
e
.
The
relevant
p
r
o
pert
ies
o
f
R
asch
a
nalysis
are
r
e
liabi
lit
y
and
v
ali
d
ity
w
hich
are
k
e
y
indi
cators
of
t
he
q
uality
o
f
a
measurement
ins
t
rument.
Thi
s
p
a
p
er
di
scu
sses
t
h
e
use
o
f
o
n
e
t
yp
e
o
f
R
asch
m
od
el
t
h
a
t
is
t
h
e
P
artia
l
C
r
edi
t
M
ode
l
to
i
nves
t
i
ga
te
r
eli
a
bil
i
ty
a
nd
v
al
idit
y
o
f
a
n
i
n
strument.
By
r
e
m
o
vi
ng
o
r
changi
ng
items
i
n
the
instrumen
t
w
hen
condition
s
o
f
reliabil
i
ty
a
n
d
v
a
l
i
d
i
t
y
are
no
t
m
e
t
,
t
he q
ual
i
ty
o
f
the
ins
t
rumen
t
i
s impro
v
ed.
K
eyw
ord
:
Cons
truc
t
va
lid
it
y
C
o
nt
en
t
v
a
li
dity
Fit sta
tis
tics
R
e
l
i
a
b
i
lit
y
S
e
para
t
i
on
Co
pyri
gh
t © 2
018 In
stit
u
t
e
of Advanced
En
gi
neeri
n
g
an
d
Scien
ce.
All
rights
res
e
rv
ed.
Corres
pon
d
i
n
g
Au
th
or:
Saras
K
r
ishnan,
F
a
cult
y
o
f
E
ngine
er
in
g a
n
d Q
u
an
t
i
t
y
S
ur
vey
i
ng,
IN
TI Inter
nat
i
o
n
a
l
U
n
i
vers
it
y,
Pe
rsiara
n Perdana
B
B
N
,
P
utra
N
il
a
i
,
71
80
0 N
i
la
i,
N
e
g
eri Sem
b
il
a
n
,
M
al
ay
si
a.
Em
ail:
sa
ras.
kris
hna
n@
new
i
n
t
i
.
ed
u.
my
1.
I
N
TR
OD
U
C
TI
O
N
A
sse
ssm
e
nt
i
s
an
i
mp
orta
nt
p
roc
e
ss
in
a
ny
c
u
rr
icu
l
um
p
rim
a
ril
y
be
c
a
use
a
s
se
ssm
e
nt
d
ri
ves
peda
g
o
g
y
.
It
i
s
t
hus
h
igh
l
y
c
r
ucia
l
t
o
d
eve
l
o
p
i
ns
trum
en
ts
t
h
a
t
a
re
a
b
l
e
to
acc
urate
l
y
r
e
p
r
e
s
ent
t
h
e
a
c
h
i
e
ve
d
l
e
a
r
ni
ng
o
ut
come
s.
D
e
s
p
i
t
e
t
h
e
n
e
e
d
to
c
o
ndu
ct
r
el
i
a
bl
e
a
n
d
c
o
n
s
i
ste
n
t
me
as
urem
ent
s
,
r
e
sea
r
che
r
s
had
p
o
i
n
t
e
d
out
t
he
l
a
c
k
of
h
igh
qu
al
it
y
i
n
st
ru
men
t
s
t
o
a
sse
ss
ach
i
e
v
e
d
l
ea
rn
in
g
o
u
tc
om
es
[
1]
.
The
m
a
j
o
r
c
o
nce
r
n
in
t
he de
v
el
o
p
m
e
n
t of an ins
t
rum
e
nt is i
f
i
t
a
c
t
ua
l
l
y
me
asure
s
w
h
at
the
r
es
ea
rche
r
se
t
s
o
ut to m
easure
[
2
]. The
tw
o
im
porta
n
t
m
e
a
surem
e
nts
of
t
he
qua
l
ity
o
f
a
n
i
nstrum
ent
ar
e
the
rel
i
a
b
i
lit
y
a
nd
val
i
d
it
y
of
t
he
i
ns
tr
um
ent.
T
h
i
s
p
a
p
e
r
d
i
s
c
u
s
s
e
s
t
h
e
p
r
o
c
e
s
s
o
f
u
s
i
n
g
R
a
s
c
h
m
o
d
e
l
,
i
n
p
a
r
t
i
c
u
lar
t
h
e
R
a
sc
h
P
a
rtial
Cre
d
it
M
o
del
t
o
impro
v
e
the
qua
li
ty
o
f
a
n
i
n
s
tr
u
m
e
n
t.
P
rior
t
o
discus
si
n
g
t
he
r
el
ia
b
ili
ty
a
nd
va
li
dit
y
c
riter
i
a
of
t
he
P
ar
t
i
a
l
C
r
ed
it
M
o
d
e
l
(
P
CM)
i
n
t
he
n
ext
sec
t
i
on,
i
n
th
is
s
ec
tio
n
w
e
w
il
l
f
irst
h
ave
a
lo
o
k
a
t
the
g
e
ne
ral
c
o
n
d
iti
o
n
s
for
re
l
i
a
b
ili
t
y
a
nd
v
al
id
ity
o
f
R
a
sch
mo
d
e
ls.
A
n
i
nstrum
en
t
i
s
r
elia
bl
e
w
h
en
it
ma
i
n
ta
ins
c
ons
i
s
tenc
y
i
n
d
iff
e
r
en
t
set
tin
gs
a
nd
the
me
asurem
ent
resul
t
s
a
r
e
ge
n
e
raliza
b
le
t
o
ot
he
r
sa
mples
[2,
3].
U
s
ing
Ra
s
c
h
mode
ls,
t
h
e
m
easure
of
i
te
m
r
e
liab
ili
ty
i
s
in
t
he
range
o
f
0.0
0
t
o
1.0
0
w
h
e
re
by
t
h
e
h
i
gher
t
h
e
va
lue
,
t
he
h
i
gher
t
h
e
r
e
l
i
ab
ili
ty
i
s
[2]
.
A
p
o
i
nt
t
o
n
o
te
i
s
th
a
t
relia
bi
l
ity
doe
s
no
t
ne
cc
esar
il
y
i
m
p
l
y
va
lidi
t
y.
H
ow
eve
r
,
unr
el
i
a
b
ili
t
y
d
e
t
rac
t
s
fr
om
v
al
i
d
it
y
[3]
a
nd
s
o
relia
bi
l
ity
i
s
a
n
i
mp
ortan
t
m
ea
sure
o
f
t
h
e
qua
l
i
t
y
o
f
a
n
i
n
s
tru
m
e
nt.
A
n
othe
r
m
e
a
s
ur
e
o
f
r
el
ia
bi
l
ity
i
s
t
h
e
Cr
o
nbac
h
’s
a
l
pha
v
a
l
ue
w
h
e
reby
val
u
es
m
ore
tha
n
0
.7
0
ar
e
deem
ed
a
c
c
e
pt
abl
e
f
or
a
g
g
r
eg
at
e
d
a
t
a
[
4
]
.
H
o
w
e
ve
r,
i
t
ha
s
bee
n
a
rg
ue
d
t
h
a
t
e
xce
p
t
i
o
n
a
ll
y
h
i
g
h
C
r
o
n
b
ach’
s
a
l
pha
v
alue
i
n
d
i
c
a
t
es
r
e
d
u
n
da
nc
y
a
n
d
t
h
us
idea
l
l
y
t
h
e val
u
e s
hou
ld be b
e
t
w
een 0.7
0 an
d
0.
9
0
[3]
. Con
c
ept
u
a
l
l
y
,
i
t
em
a
nd
c
ase rel
i
a
b
i
l
i
t
ies are
e
q
u
i
va
l
e
n
t
to
C
ron
b
ac
h’
s
alp
h
a
[5].
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2252-
8822
IJERE
V
ol
.
7,
N
o.
4,
D
ece
mbe
r
20
18
:
31
3
–
316
31
4
The
tw
o t
y
pes
of va
l
id
it
y t
h
a
t
c
an be m
e
a
s
ur
ed usi
ng t
h
e
R
a
sch
mode
l
s
ar
e
the c
o
n
str
u
ct va
l
i
d
i
t
y an
d
the
c
o
n
t
e
n
t
va
lid
it
y.
C
on
str
u
ct
v
a
l
id
i
t
y
i
s
u
se
d
t
o
d
e
t
er
mine
i
f
an
i
ns
trum
en
t
m
e
a
s
u
r
es
t
he
u
n
o
b
s
e
r
vab
l
e
con
s
truc
t
tha
t
i
t
p
u
r
port
s
t
o
me
asure
[3]
,
a
nd
i
f
the
item
s
i
n
th
e
i
n
s
t
ru
men
t
a
re
s
y
s
t
e
ma
ti
cal
ly
a
nd
p
re
di
ct
abl
y
rela
t
e
d
t
o
e
a
c
h
o
ther
a
l
o
ng
the
c
o
nstr
uc
t
tha
t
i
s
be
i
ng
me
asure
d
[6].
T
h
i
s
is
a
c
h
ie
ve
d
b
y
e
xam
i
n
i
ng
the
rela
tio
ns
hip
be
tw
ee
n
the
m
e
a
s
ure
be
i
n
g
e
v
a
l
ua
te
d
a
n
d
t
h
e
var
i
a
b
l
e
s
k
n
o
w
n
t
o
b
e
r
e
la
t
e
d
or
t
he
ore
tic
a
l
l
y
rela
t
e
d
t
o
t
he
c
ons
truc
t
be
in
g
m
e
a
s
ured
b
y
t
h
e
i
n
s
t
rume
n
t
[
2].
O
n
t
h
e
o
t
h
e
r
h
a
n
d
,
c
o
n
t
e
n
t
v
a
l
i
d
i
t
y
d
e
t
e
r
m
i
n
e
s
if t
he
i
nstr
um
ent
me
asures
t
h
e
e
n
tire
d
o
m
a
i
n
re
l
ate
d
to
the
con
s
t
ruc
t
it
w
a
s de
signe
d t
o
m
easur
e [7].
2.
PARTIAL
CREDI
T
MODEL (PCM)
O
f
t
he
d
iffe
re
nt
t
ype
s
o
f
R
a
s
ch
m
o
d
els,
t
h
e
P
a
r
ti
a
l
C
red
i
t
Mo
de
l
(P
CM
)
i
s
p
art
i
c
u
lar
l
y
u
s
eful
f
or
i
n
s
t
ru
me
nt
s
w
i
th
p
ol
yto
m
ous
s
co
rin
g
b
eca
u
s
e
PCM
a
l
l
o
ws
v
a
r
yi
ng
l
e
vel
s
o
f
in
t
e
r
m
ediat
e
c
orr
ectne
ss
betw
ee
n
c
o
m
p
l
e
t
e
ly
c
orrec
t
a
nd
c
omp
l
e
t
e
l
y
w
r
ong
a
n
sw
e
r
s
.
I
n
ot
h
er
w
or
d
s
,
differ
en
t
ite
m
s
i
n
t
h
e
in
st
r
u
me
nt
ca
n ha
ve
di
f
fe
rent
num
ber
of
r
e
s
po
nse
op
tio
ns. A
s suc
h
,
P
C
M i
s
an
i
de
al m
ea
sur
e
m
e
nt t
oo
l in t
he ed
u
c
a
tio
n
a
l
con
t
e
x
t
s
b
e
c
a
u
se
it
is
c
om
mo
n
t
o
a
w
a
rd
p
ar
t
m
a
rks
t
o
s
t
u
de
nt
s’
pa
rtly
c
or
rec
t
a
nsw
e
rs
w
here
by
h
i
g
h
er
m
arks
repr
esents
h
i
g
her
lea
r
ner
abi
l
i
t
y.
F
urtherm
o
re
,
from
t
he
m
easur
e
m
e
n
t
per
s
pec
t
i
v
e
poly
t
omou
s
da
ta
i
s
m
o
re
d
e
si
ra
b
l
e
th
an
d
i
c
ho
t
o
mou
s
d
at
a
b
e
c
a
u
se
t
he
e
rro
r
e
st
i
m
at
es
a
re
s
ma
l
l
e
r
f
or
p
o
l
yt
om
ou
s
da
ta
f
or
t
he
s
am
e
numbe
r of i
te
m
s
.
I
n
u
s
i
n
g
P
CM t
o de
ter
m
ine th
e
qua
l
ity of
a
n
i
n
s
trume
n
t,
w
e
i
n
ve
st
iga
t
e the
val
u
es
o
f
t
h
e
fit
sta
t
is
t
i
c
s
[8],
i
n
part
ic
ul
ar
t
he
m
ea
n
s
q
u
a
re
f
i
t
s
ta
ti
sti
c
s
a
n
d
the
sta
n
d
a
r
d
i
ze
d
fi
t
s
t
a
t
i
s
tics.
I
nfit
m
e
a
n
s
quare
i
s
us
e
d
t
o
in
ves
tiga
t
e
the
size
of
t
he
r
a
n
d
o
mne
s
s
or
t
he
a
m
o
u
n
t
of
d
i
s
tor
t
io
n
o
f
a
m
easure
m
e
n
t
sy
stem
.
Infi
t
m
ean
squ
a
re
m
ore
th
an
1
.2
0
s
u
gges
t
s
unus
ua
l
or
i
na
p
p
ro
pria
te
r
e
s
po
ns
e
p
a
t
t
e
rns
wh
il
e
i
n
fit
m
e
an
s
q
u
a
re
l
e
s
s
t
h
an
1.00
i
n
d
ica
t
e
s
t
o
o
l
itt
le
v
a
r
ia
t
i
o
n
i
n
the
res
p
ons
e
pa
tte
rns
t
h
at
f
urt
h
e
r
s
u
g
g
e
s
ts
p
r
e
senc
e
of
r
e
dun
da
n
t
i
t
e
ms
[
5
]. Th
e
st
a
nd
ard
d
e
vi
a
tio
n
of
l
e
ss th
a
n
2
.00
in
di
c
a
t
e
s a
n
o
ver
all fit
of the
ite
ms
[
5].
F
u
rther,
t
he
o
t
h
er
t
w
o
p
ro
pertie
s
used
t
o
i
n
v
e
st
i
g
a
t
e
t
h
e
q
u
a
li
ty
o
f
a
m
e
asur
em
ent
ins
t
rum
e
nt
a
re
t
he
it
e
m
s
e
p
ara
tio
n
a
n
d
t
h
e
me
a
n
m
ea
sur
e
.
Item
s
epar
ati
o
n
me
asures
the
spr
ead
o
f
it
e
m
s
in
s
ta
ndar
d
e
rr
or
u
n
its
[
9
]
.
I
t
r
e
f
l
e
c
t
s
t
h
e
e
x
t
e
n
t
t
o
w
h
i
c
h
t
h
e
i
t
e
m
s
a
r
e
s
e
p
a
r
a
t
e
d
a
l
ong
a
s
cal
e
in
t
erms
o
f
th
ei
r
d
i
f
f
i
c
u
l
ty
l
ev
e
l
s.
A
n
i
t
em
s
ep
a
r
a
t
i
o
n
of
m
o
r
e
th
an
1
.0
0
in
d
i
cat
e
s
t
ha
t
th
e
r
e
i
s
e
n
oug
h
sprea
d
o
f
i
t
em
s
a
n
d
t
hus
t
he
i
n
s
t
r
u
m
e
n
t
is
ac
cept
a
b
l
e.
M
e
a
n
m
ea
sur
e
i
s
a
n
i
ndic
a
t
or
o
f
t
h
e
d
i
ffic
ult
y
l
eve
l
o
f
t
he
i
n
s
trum
en
t
w
h
e
r
eby
a
p
o
s
iti
ve
v
al
ue
sug
g
es
t
s
t
ha
t
t
h
e
i
n
st
r
u
me
nt
i
s
ea
sy
w
hi
le
a
n
egat
i
v
e
v
a
lue
s
u
g
ges
t
s
tha
t
t
he
i
ns
t
r
u
m
e
n
t
is
d
i
f
fic
u
l
t
.
In
spec
ific
,
me
an
m
ea
sure
o
f
v
a
lue
s
1
.
0
0
a
n
d
2.0
0
i
nd
i
c
a
t
e
t
h
a
t
t
h
e
ins
t
rum
e
nt
i
s
t
o
o
easy
wher
eas
m
e
a
n
me
asure
s
-
1.
00
a
nd
-
2
.00
i
n
d
i
ca
te
t
ha
t
th
e
in
strum
e
n
t
i
s
to
o
di
ffic
u
l
t
.
In
b
o
t
h
c
a
ses,
i
t
is
a
dv
isa
b
l
e
t
o
ha
v
e
t
he
in
strum
e
n
t
r
evi
s
ed [5].
The
c
o
n
s
truc
t
val
i
d
it
y o
f
a
n i
n
strum
e
nt
i
s
de
term
ine
d
by i
n
spe
c
ti
ng t
h
e ta
bl
e
of m
isfit or
der of i
tem
s
.
Cons
truc
t
val
i
d
it
y
e
n
sures
th
at
a
s
in
g
l
e
c
o
n
s
t
r
uc
t
is
b
e
i
ng
m
ea
sur
e
d,
a
nd
tha
t
t
he
i
tem
s
a
re
s
yste
ma
tic
a
l
l
y
a
n
d
pred
ic
t
a
b
l
y
re
la
ted
to
e
ac
h
ot
her
[6
]
.
T
he
c
ond
i
t
i
ons
f
or
c
o
n
st
ruc
t
v
a
lid
i
t
y
a
r
e:
(
1)
i
nfi
t
m
e
a
n
squa
re
b
e
t
w
e
en
0.77
a
nd
1.3
0
[
6,
10],
a
nd
(
2
)
in
fit
z-sta
n
dar
d
i
z
ed
b
etwee
n
-
2.
00
t
o
+
2
.
0
0
[
9].
These
c
o
n
d
it
i
o
n
s
w
e
r
e
adop
te
d
in
ea
r
l
i
er pa
p
er
s as w
el
l (se
e
, [11,
12]).
The
c
on
te
nt
va
l
i
d
i
t
y
is de
t
e
rm
ine
d
us
i
n
g
t
he
p
o
i
n
t
m
easur
e c
o
rre
l
a
t
i
on
val
u
es
w
here
b
y
t
he
se
v
al
ues sho
u
l
d be
pos
it
ive.
3.
AN EX
EMPLAR
I
n
t
h
i
s
sec
t
io
n,
w
e
d
i
sc
uss
a
n
e
xam
p
le
o
f
a
n
i
ns
tru
m
ent
tha
t
w
a
s
re
fi
ne
d
usin
g
the
P
C
M
me
asure
m
e
n
t
s
(
se
e,
[
13]).
D
iscuss
i
o
n
is
s
u
p
por
ted
by
t
h
e
t
a
ble
s
o
f
sum
m
a
ry
o
f
sta
t
i
s
t
i
c
s
a
nd
m
i
s
f
it
orde
r
o
f
it
e
m
s
w
h
ic
h ha
ve be
e
n ge
nera
t
e
d u
s
i
n
g the
Wi
nste
ps
c
om
pan
i
on t
o
B
o
nd
and
F
ox (
2
007)
[9]
.
3.1.
Me
asu
r
e
m
en
t of R
e
l
iab
i
l
i
ty
Ta
b
l
e
1
d
i
sp
la
ys
t
he
r
el
i
a
b
i
l
i
ty
v
a
l
ue
s,
t
he
s
e
p
ara
tio
n
va
l
u
es,
t
h
e
i
n
f
i
t
m
e
a
n
s
q
u
a
r
e
,
t
h
e
s
t
a
n
d
a
r
d
dev
i
a
t
i
o
n, the
m
ean m
easure
a
nd Cro
n
b
ach’
s
alp
ha
v
al
ue
f
or t
he
t
h
ree
su
cc
essi
ve s
tage
s
of
d
eve
l
opm
e
n
t of a
n
in
strum
e
n
t
.
Tab
l
e1.
Sum
m
ary
of sta
tis
tic
s
R
e
l
i
a
bilit
y
S
e
p
a
r
a
tion
In
fit MN
SQ
S
.D
.
Me
a
n
C
ronbac
h
St
a
g
e
I
0.
97
5
.
4
8
1.
0
2
0
.
2
3
0.
15
0
.
8
5
Sta
g
e
II
0.
98
7
.
2
2
1.
0
0
0
.
0
9
-0.
2
2
0.
7
5
S
t
a
g
e
II
I
0.
98
6
.
4
8
1.
0
0
0
.
0
8
-0.
1
8
0.
7
5
Evaluation Warning : The document was created with Spire.PDF for Python.
IJERE
I
S
S
N
:
2252-
88
22
U
s
in
g PCM t
o
I
m
prove
t
h
e
Q
u
a
l
ity o
f
a
n
Instrum
e
n
t
(Sar
as
Kri
s
hn
a
n
)
31
5
The
ite
m
r
e
li
a
b
il
it
i
e
s
for
t
h
e
thre
e
sta
g
es
o
f
the
ins
t
rum
e
n
t
d
eve
l
o
p
m
e
n
t
a
re
v
ery
hi
gh,
t
ha
t
i
s
m
o
r
e
tha
n
0.
90, ind
i
c
a
tin
g h
i
gh rel
i
abi
l
i
t
y
.
Th
is is sup
p
o
rte
d
b
y C
r
on
bac
h
’s
a
lp
h
a
val
ues w
h
ic
h
a
r
e m
o
re t
ha
n 0.
70.
A
l
so, Cron
ba
c
h
’s
a
l
p
h
a
va
l
ue
s
be
tw
e
e
n
0.
7
0
a
n
d
0.90
i
ndic
a
t
e n
o
n
-
re
d
unda
n
c
y
of
it
e
ms
i
n
t
h
e
i
n
st
ru
men
t
[
3
]
.
In
a
d
d
i
t
i
on,
t
h
e
ite
m
sepa
rat
i
on
va
lue
s
o
f
m
o
re
t
han
1.
00
i
n
d
i
c
a
t
e
e
n
o
u
g
h
spr
ead
o
f
i
t
em
s.
W
i
t
h
re
fere
n
ce
to
F
i
sher
’s (200
7
)
instr
um
ent
qua
lit
y criter
i
a,
the
i
tem
re
l
i
a
b
i
l
i
ty va
l
ues of m
ore
tha
n
0
.
9
4 t
oge
t
h
er
w
it
h t
h
e i
t
e
m
separ
a
t
i
on va
lu
es of
m
o
re
tha
n 5.0
0
i
n
d
i
ca
t
e
t
ha
t the
i
n
s
t
r
u
me
n
t
ha
s exce
lle
nt qua
l
ity [1
4
]
,
as is the ca
se
i
n
a
l
l
three
sta
g
e
s
s
h
o
w
n
i
n
Ta
bl
e
1.
T
he
i
nfi
t
m
ea
n
squa
re
a
re
b
e
t
w
e
e
n
1.00
a
nd
1.
2
0
i
nd
ica
tin
g
t
h
at
t
he
r
es
po
nse
pat
t
ern
is
s
u
i
t
a
ble,
t
her
e
i
s
e
nou
gh
var
i
a
t
i
on
i
n
t
he
r
e
s
p
o
ns
e
s
a
nd
t
here
a
re
n
o
redu
nd
an
t
it
e
m
s
i
n
t
h
e
in
strum
e
n
t
.
T
h
e st
a
n
d
a
rd
d
ev
i
a
tio
n
val
u
es
f
o
r
the
i
t
e
ms are
les
s
than
2
.00,
an
indi
cation
of
an
o
verall f
it of the
it
e
m
s
[5].
F
urt
h
er
,
the
ite
ms
a
re
n
o
t
t
o
o
e
as
y
or
t
o
o
d
iffic
u
lt
,
as
i
nd
ica
t
ed
by
the
val
u
es
o
f
t
h
e
me
an
m
easure
,
and
the
r
ef
ore
t
h
e
in
st
r
u
me
n
t
i
s
ac
cepta
ble
fo
r me
asure
m
e
n
t
purp
oses.
3.2.
Me
asu
r
e
m
en
t of V
alid
ity
Ta
b
l
e
2
d
i
s
p
la
ys
t
he
i
nf
it
m
e
a
n
s
quar
e
(
MN
S
Q
),
t
he
i
n
f
it
z
-
sta
n
da
rd
i
zed
(
ZSTD)
a
n
d
t
h
e
poi
n
t
me
asure
cor
r
ela
t
i
o
n (
P
MC)
value
s
w
hic
h
w
e
r
e
used
t
o
esta
bl
i
s
h
t
he
va
l
i
d
i
t
y
o
f
t
he
ins
t
r
um
ents.
Ta
ble
2.
M
i
s
f
i
t
or
der
of item
s
M
NS
Q
ZS
TD
P
M
C
S
t
a
g
e
I
0.
80
<
M
N
SQ <
1
.92
-
1
.
10
<
Z
S
TD
<
3
.50
0.
27
<
P
M
C
<
0.
7
0
S
t
a
g
e
I
I
0
.8
2
<
MNS
Q
<
1
.
1
9
-
1
.6
0
< ZS
T
D
<
1
.
5
0
-
0
.1
9
< PM
C
< 0
.5
7
Sta
g
e
II
I
0.
86
<
M
N
SQ <
1
.17
-
1
.
60
<
Z
S
TD
<
1
.40
0.
04
<
PMC
<
0
.
52
The
i
n
s
t
r
u
me
n
t
i
n
S
t
age
I
did
n
o
t
me
et
t
he
c
on
di
tio
ns
o
f
con
s
tr
uc
t
va
li
d
ity
s
i
n
c
e
s
om
e
of
t
he
i
nf
it
me
an
s
q
u
are
s
a
re
m
ore
tha
n
1
.
30.
B
y
re
movi
n
g
o
r
mod
i
fy
in
g
s
o
m
e
item
s
,
the
i
n
strum
e
n
t
i
n
S
t
age
II
f
u
l
f
ille
d
the
c
o
nd
i
tio
ns
o
f
c
ons
t
r
uc
t
va
li
d
i
t
y
w
he
reb
y
0.
77
<
MN
S
Q
<
1
.3
0
a
n
d
-
2.00
<
ZS
TD
<
2
.
0
0.
H
ow
e
v
er,
the
in
strum
e
n
t
i
n Stage
II
di
d
n
o
t
fu
l
f
ill t
h
e co
nd
it
ion
for co
n
t
e
n
t va
li
d
ity bec
ause the
re are
ne
g
a
t
i
v
e
P
M
C
v
a
l
u
es.
F
u
r
t
he
r
refine
m
e
nt
o
f
the
ite
ms
l
ed
t
o
th
e
ins
t
r
u
me
nt
i
n
S
t
age
III
w
h
i
c
h
have
b
e
t
ter
quality
t
han
the
in
strum
e
n
t
s
i
n
S
tage
I
a
n
d
S
tage
I
I
in
t
erm
s
o
f
re
lia
bi
li
t
y
a
n
d
v
a
l
i
d
it
y
base
d
on
the
criteria
d
i
sc
usse
d
in
t
h
i
s
p
ap
e
r
.
4.
DISC
USSION
The
use
o
f
f
i
t
s
ta
tist
i
cs
i
s
i
n
flue
nc
e
d
b
y
i
t
s
in
t
e
nde
d
a
ppl
i
c
a
t
i
o
n
w
he
reb
y
f
or
a
h
ig
h
s
t
a
k
es
t
est
i
ng,
the
rea
s
on
a
b
le
s
ti
p
u
la
te
d
ran
g
e
o
f
t
he
f
it
sta
tist
i
cs
a
re
b
e
t
w
e
en
0
.
80
a
n
d
1.
20,
f
or
a
s
u
r
ve
y
t
h
e
val
u
e
s
a
r
e
betw
ee
n
0.60
a
nd
1.4
0
a
n
d
f
or
c
l
i
n
i
c
a
l
o
bs
erva
tio
n
t
h
e
r
ecom
m
e
n
d
e
d
val
u
e
s
o
f
th
e
fit
st
at
i
s
ti
c
s
a
re
b
et
wee
n
0.50
a
n
d
1
.7
0
[15]
.
Li
nac
r
e
(
2
0
0
2
)
,
i
n
d
i
s
c
uss
i
ng
the
impl
ica
t
i
o
n
s
o
f
u
si
ng
d
i
f
f
er
en
t
val
u
es
o
f
the
me
an
s
q
u
a
r
e
f
i
t
s
t
a
t
i
s
t
i
c
s
a
n
d
s
t
a
n
d
a
r
d
i
z
e
d
f
i
t
s
t
a
t
i
s
t
i
c
s
s
t
a
t
e
d
t
h
at
t
he
m
e
a
n
sq
ua
re
f
i
t
s
t
a
t
i
s
t
ic
s
betw
e
e
n
0
.
5
0
an
d
1.50
i
s
pr
od
uc
t
i
ve
f
or
m
e
a
sure
me
nt
w
here
as
s
ta
n
d
ardiz
e
d
f
it
s
t
a
tis
t
i
cs
b
et
w
e
e
n
-
1.
90
a
n
d
1.90
i
n
d
ica
t
e
s
t
ha
t
data
h
a
v
e
rea
s
ona
b
l
e pr
ed
icta
bi
l
i
t
y
[
16].
Ov
er
t
h
e
y
ea
rs,
d
i
f
f
e
re
nt
s
e
t
s
of
c
ri
teria
of
r
e
l
ia
b
i
li
ty
a
nd
v
a
lid
i
t
y
h
a
v
e
b
e
e
n
ad
opt
ed
.
Fo
r
in
st
a
n
ce,
A
oyam
a
(
20
0
7
)
set
t
h
e
co
nd
i
t
i
o
n
of
c
o
n
te
n
t
v
al
i
d
it
y
to
b
e
0.7
7
<
MNSQ
<
1
.
30
[6]
wher
eas
P
ada
,
K
a
r
t
ow
a
g
i
r
an
a
nd
S
uba
l
i
(
2
0
16)
u
se
d
t
h
e
c
o
n
d
i
t
i
on
b
e
t
w
e
en
0
.
70
<
M
N
S
Q
<
1
.
3
0
[
1
7
]
.
O
n
t
h
e
o
t
h
e
r
h
a
n
d
,
Zu
bairi
a
nd
K
a
ssim
(2
01
6)
u
se
d
t
h
e
c
ond
i
t
i
on
0.4
0
<
MN
S
Q
<
1
.60
[
1
8
]
f
or
a
cce
p
t
a
b
le
f
it.
A
lso,
d
iffe
rent
rese
arc
h
ers have repor
t
e
d d
i
ffe
r
ent me
as
ures in t
h
eir disc
u
ssi
o
n. F
or i
ns
ta
nc
e,
Wa
t
s
o
n a
nd Ca
ll
i
n
g
h
a
m
(200
3)
[1
0]
d
id
n
o
t
e
xa
mine
t
he
outfi
t
m
e
a
n
s
qua
r
e
s
ta
t
i
st
i
c
s
i
n
t
he
ir
s
t
u
d
y
b
ut
S
a
b
ud
in,
Mans
or,
Mee
r
ah
a
n
d
Mu
ham
m
a
d
(201
8)
[
1
9
]
did.
5.
CONCL
U
S
ION
Rasc
h
m
o
de
l
i
s
a
v
a
l
ua
ble
tool
t
o
in
ve
st
i
g
at
e
t
h
e
q
u
al
i
t
y
o
f
a
n
a
s
s
e
s
s
m
e
n
t
i
n
s
t
r
u
m
e
n
t
i
n
t
e
r
m
s
o
f
i
t
s
relia
bi
l
ity
a
n
d
v
al
i
d
it
y.
I
nstea
d
o
f
ha
vin
g
t
o
c
o
mple
te
l
y
r
ev
is
e
th
e
ins
t
rum
e
nt,
Rasc
h
m
ode
l
i
n
form
s
us
w
hic
h
it
e
m
s in
t
he i
ns
trum
en
t
sp
ec
i
f
i
call
y
v
io
la
tes the
co
n
d
i
t
i
ons o
f
r
e
l
ia
bil
ity
a
nd
v
ali
d
it
y
.
B
y
re
mo
v
i
ng
o
r
re
vi
si
ng
the
i
t
e
m
s
in
c
once
r
n,
t
he
q
u
a
li
t
y
o
f
t
h
e
a
s
se
ssm
e
nt
i
ns
t
r
ume
n
t
i
s
i
m
p
rove
d
so
t
ha
t
the
i
n
strum
e
n
t
f
i
t
s
t
h
e
pur
pose
of
a
sse
ssm
e
nt.
Re
se
arc
h
ers
have
u
sed
an
d
re
p
o
rt
ed
d
iffe
re
nt
m
ea
su
res
o
f
R
asch
a
n
a
ly
sis
in
t
h
e
i
r
stud
ies
as se
e
n in
t
he
d
isc
u
ss
ion
ab
o
v
e.
I
n
c
o
n
cl
us
io
n,
t
her
e
a
re
no
ha
rd
a
n
d
f
a
s
t
r
u
l
e
s
on
t
h
e
a
cce
pt
a
b
l
e
r
ang
e
o
f
t
h
ese
me
as
u
r
e
s
n
ei
th
e
r
a
con
d
i
t
i
on
o
f
w
h
i
c
h
o
f
t
hese
m
ea
sur
e
s
ar
e
o
b
l
i
ga
tor
y
.
Esse
nt
ia
ll
y,
t
he
r
e
s
e
a
rc
her
ma
kes
the
u
l
t
i
m
a
te
d
e
c
isio
n
on
h
ow
t
o
us
e
the
pr
oper
t
i
e
s
o
f
R
asc
h
a
nal
y
s
i
s
t
o
i
mp
r
o
ve
t
he
q
ua
li
t
y
o
f
a
m
easu
r
e
m
e
n
t
ins
t
ru
m
e
nt.
Howeve
r,
r
e
g
ardless
o
f
t
he
m
ea
sures
a
n
d
cri
t
eria
a
d
o
p
t
e
d
,
t
h
e
R
a
s
ch
P
arti
al
C
re
dit
Mo
d
e
l
whi
c
h
i
s
e
sp
eci
a
l
l
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N: 2
2
5
2
-
88
22
I
J
ERE
Vo
l
.
7
, No
.
4
, Decemb
e
r
2
0
1
8
:
3
13 –
3
1
6
31
6
usef
u
l
w
he
n
de
ali
n
g w
i
th
p
o
l
y
t
omo
u
s
d
a
ta i
s
a
n
e
xce
l
le
nt
t
o
o
l
t
o i
d
e
n
ti
fy
poss
i
b
l
e thre
a
t
s to t
he re
liab
i
li
ty an
d
va
li
di
t
y
o
f
an
i
ns
tr
um
ent
a
nd
th
us
i
mpr
o
v
i
n
g
the
qua
l
i
t
y
o
f
t
he
i
n
s
tr
u
m
e
n
t.
REFERE
NC
E
S
[1]
Garf
i
e
ld
J
,
Ben‐Zv
i
D.
“
Ho
w
s
t
u
d
en
ts
l
earn
statisti
cs
r
ev
is
ited
:
A
curren
t
r
ev
iew
of
r
es
earch
o
n
teach
in
g
a
n
d
learning statist
i
c
s
”.
Inter
natio
nal S
t
ati
s
t
i
ca
l
R
eview
. 75
(
3)
:3
72
-96
;
2
00
7.
[2]
Hagan
T
L
.
“Me
a
surement
s
i
n
qua
ntitative
research:
How
t
o
s
el
ect
a
nd
report
on
r
esearch
i
n
s
t
r
u
m
ent
s
”.
O
n
co
logy
Nur
s
i
ng Fo
ru
m
.
41(4
)
:4
31
-4
33
;
201
4.
[3]
Pesudovs
K
,
B
u
rr
JM
,
Harl
ey
C
,
E
lliott
DB
.
The
devel
o
p
m
ent,
a
ss
ess
m
en
t
,
a
nd
s
e
l
ecti
o
n
o
f
q
uesti
o
n
n
aires
.
Optome
try
an
d
Visio
n
S
c
ie
nc
e
.
8
4
(8
):66
3-7
4
;
2007
.
[4]
Ism
a
il
W,
Z
ail
a
ni
M
A
,
A
w
a
d
ZA,
Hussin
Z,
F
a
i
sal
M
,
S
aad
R
.
Acad
e
m
i
c
a
nd
S
o
c
ial
M
e
d
i
a
P
r
acti
c
es
o
f
Arabic
La
ng
ua
g
e
a
mon
g
Ma
la
y
s
ia
n
St
ud
e
n
ts.
T
u
r
k
is
h
O
n
line
Jou
r
na
l
of Ed
ucatio
na
l T
ech
nol
og
y
(
T
OJE
T
)
.
16(3):
1-
0;
20
17
.
[5]
Green
K
E,
F
ran
t
o
m
C
G
.
“
S
u
rv
ey
d
ev
elo
p
m
e
nt
a
nd
v
ali
d
at
ion
w
ith
t
he
R
as
c
h
m
od
el”.
I
n
International
Confer
ence
on
Q
u
estio
nn
a
i
re D
eve
l
o
p
m
en
t,
E
v
al
uati
o
n
,
an
d
T
e
s
t
i
n
g
,
Ch
arl
e
s
t
o
n
, SC
, 20
0
2
No
v 1
4
.
[6]
Aoy
a
ma
K
.
“In
v
e
stiga
t
in
g
a
hie
r
a
r
c
h
y
of
s
tu
de
n
t
s’
i
n
t
e
r
pre
t
a
tio
n
s
o
f
graph
s
”.
In
te
rn
atio
na
l E
l
e
c
t
ro
n
i
c
J
o
u
r
na
l of
Ma
the
m
at
ic
s Ed
u
c
atio
n.
1
2
;
2
(
3):29
8
-31
8
;
20
07.
[7]
Heal
e R, T
wy
cross
A
.
“
Va
l
i
dit
y
a
n
d
r
el
iabili
ty
i
n
q
u
ant
i
t
a
ti
ve
studi
es”.
E
v
id
ence-ba
sed
Nur
s
i
n
g
: e
b
n
u
r
s
-2
01
5.
[8]
Callingham
R
,
C
a
rmi
c
hael
C
,
W
a
ts
on
J
M
.
“Exp
l
a
i
n
ing
studen
t
achi
ev
emen
t:
T
he
i
n
f
l
u
en
c
e
o
f
teach
ers
’
p
edago
g
i
cal
con
t
en
t
kno
w
l
edg
e
i
n
statisti
cs”.
Inter
nati
o
n
a
l
J
o
u
r
n
a
l
of S
c
ien
ce a
nd
M
a
th
e
m
at
ics E
duca
t
i
o
n
.
14(7):1339-135
7;
20
16
.
[9]
B
o
n
d
T
G,
F
o
x
C
M
.
A
p
p
l
yi
ng th
e
Ra
sch
mo
del
:
Fu
nd
am
enta
l M
e
asur
emen
t
i
n
the Hu
man
S
c
iences
.
2
nd
e
d
i
t
i
o
n
.
N
ew
Jers
ey:
Law
r
ence E
r
lbaum
;
2
00
7
.
[10]
W
a
t
s
o
n
J
,
Calling
h
am
R
.
“S
ta
t
i
s
t
ical
l
iteracy:
A
com
p
lex
h
i
erar
chi
cal
c
o
n
st
ruct
”.
St
a
t
istics
Ed
ucat
io
n
Res
e
arch
J
o
urna
l
.
2(
2):3-
4
6; 20
0
3
.
[11]
Kri
s
h
n
an
S, Id
ris N. “The Use of
a
Hierarchical C
onstruct
t
o
In
ves
t
i
g
a
t
e S
t
u
d
ents’ Learni
ng
o
f
Inf
e
rent
ial S
t
atis
tics”.
In
Pro
ceedi
n
g
s
o
f
t
h
e
Jo
int
IA
S
E
/IAOS
Satellit
e
Con
f
er
ence
, M
acao
, Chi
na 2
01
3 A
u
g
.
[12]
Kri
s
hnan
S
,
Idri
s
N
.
“
Inves
t
i
g
ati
n
g
reli
ab
i
l
ity
a
n
d
v
al
idi
t
y
f
o
r
the
con
s
t
r
uct
o
f
i
n
f
eren
tia
l
stati
s
tics
”
.
International
J
o
urna
l o
f
Le
a
r
n
i
n
g
,
Te
a
c
hing
an
d Edu
c
at
ion
a
l
R
e
se
arc
h
. 4(1):51
-60;
2
014.
[13]
Kri
s
h
n
an
S
,
Id
ris
N
.
“
Im
pro
v
i
n
g
t
h
e
Q
u
ality
o
f
a
M
easu
r
em
en
t
In
strum
e
nt
u
sing
t
he
P
art
i
al
C
red
it
M
o
del
”
.
In
Int
e
rn
atio
na
l
Confer
ence o
n
L
e
a
r
n
i
ng a
nd
T
e
ach
i
ng
,
S
i
n
g
a
p
o
r
e
, 2
01
5 Ma
rc
h 25
&
26
.
[14]
Fisher
W
P
.
“
Rating
sca
l
e ins
t
ru
ment qu
a
l
i
ty cr
i
t
e
ria”.
Ras
c
h M
e
as
ur
ement
T
r
an
sact
io
ns
. 2
1
(1),
1
09
5; 20
0
7
.
[15]
Wri
ght B.
“
R
eason
abl
e
m
ean-s
quare
f
it
va
lu
es”.
Ra
sch M
e
as
u
r
emen
t
T
r
ans
ac
t
i
o
n
s
. 8
:
3
7
0
;
19
94
.
[16]
Li
nacre
JM
.
“Wh
a
t
do
i
n
f
it
and
o
u
tf
it,
m
ean-s
qu
are
an
d
st
anda
rd
ized
m
ean”.
R
a
sch
M
e
as
ur
e
m
ent
T
r
an
sa
c
tio
n
s
.
16
(2):
87
8;
2
0
0
2
.
[17]
P
a
da
AU,
K
artow
a
giran
B,
S
u
b
al
i
B.
“
Separat
i
on
i
ndex
a
nd
f
it
it
e
m
s
of
c
reati
v
e
t
h
i
n
king
s
ki
ll
s
as
sessm
en
t”.
Res
e
ar
ch a
n
d
Eva
l
u
a
ti
on in E
duc
a
t
i
on
(
R
EiD)
.
1
;
2(1
)
:1
-2;
20
16
.
[18]
Z
u
b
a
i
r
i
A
M
,
K
a
s
s
i
m
N
L
.
“
C
l
a
s
s
i
c
a
l
a
n
d
R
a
s
c
h
a
n
a
l
y
s
e
s
o
f
d
i
c
h
o
t
o
m
o
usly
s
co
red
re
ad
ing
co
mp
rehen
s
io
n
test
it
e
m
s”.
M
a
laysia
n Jo
ur
na
l o
f
ELT
Res
e
ar
ch
. 7
;2(
1
):2
0
;
20
16
.
[19]
S
a
bu
di
n
S,
M
an
s
o
r
AN,
M
eerah
S
M
,
M
u
h
amm
a
d
A.
“
Validity
a
nd
Rel
i
ab
ilit
y
of
S
t
u
d
e
nts
’
S
ci
enc
e
a
n
d
T
echn
o
l
o
g
y
Cul
t
u
r
e
In
st
rument
(
BS
T-M)
u
sin
g
R
asch
M
easu
r
em
e
n
t
M
o
d
e
l”
.
I
n
te
r
n
a
t
io
na
l J
o
u
r
n
a
l
o
f
A
c
ad
e
m
i
c
R
e
s
e
ar
c
h
in
Bu
si
nes
s
a
nd Socia
l S
c
ien
c
e
s
.
8
(5):
98
6-99
5;
2
0
1
8
.
BIOGRAPHI
E
S
OF
AUT
HORS
S
a
ras
Krishn
an
o
btai
ned
her
do
ctorat
e
degree
in
M
athem
a
tics
E
d
u
c
a
t
io
n
from
t
h
e
U
nive
rs
ity
o
f
M
a
laya,
M
a
lay
s
i
a
i
n
2
014.
S
he
h
as
a
B
S
c
i
n
Indu
st
rial
S
cien
c
e
m
ajo
r
ing
i
n
M
a
t
hem
a
t
i
cs
a
nd
M
S
c
i
n
A
p
p
l
i
ed
M
at
hem
a
tics,
bot
h
f
r
o
m
t
he
U
n
i
v
e
rsiti
T
e
kno
logi
M
a
la
y
s
i
a
,
M
a
la
ys
ia
.
S
h
e
i
s
a
fu
ll
t
i
m
e
te
a
c
hi
ng
s
ta
f
f
a
t
th
e
I
N
TI
I
nt
e
r
n
a
ti
on
a
l
U
n
i
v
e
r
s
it
y
i
n
M
a
la
ysia
a
nd
o
ve
r
th
e
y
e
ars
h
a
ve
tau
ght
d
iff
e
rent
c
ou
rses
i
n
c
lu
d
i
n
g
calcu
lus
,
e
n
g
ineeri
ng
m
a
them
at
ics
and
m
a
th
e
m
atical
s
tud
i
es
f
o
r
vari
ou
s
pro
g
ram
s
.
Her
resea
r
ch
i
nt
erests
i
n
c
lu
de
t
each
in
g
a
nd
l
earni
ng
,
and
asses
s
m
e
n
t
s,
part
icul
arly in the h
i
gh
er learn
i
n
g
i
n
s
titu
tio
n
s.
E-mai
l: saras.
kri
s
hnan@newi
n
ti.
edu.my
Prof
D
at
o’
D
r
N
o
r
a
i
n
i
Idris
holds
a
BSc
E
d
.
(H
o
n
s)
m
ajor
ing
in
Mathema
tics
and
Physics
an
d
M
.
Ed.
i
n
M
ath
e
matics
E
d
u
cation
f
r
o
m
t
he
U
niversi
t
i
M
a
lay
a
,
Mal
a
ys
ia.
S
h
e
w
e
n
t
o
n
to
obtai
n
a
P
h
D
in
M
a
t
h
e
m
a
ti
c
s
E
d
u
c
a
t
i
o
n
f
r
o
m
t
h
e
Oh
io
S
t
a
t
e
U
ni
ve
r
s
it
y
in
th
e
U
n
ited
States
o
f
A
m
eri
ca.
S
h
e
w
a
s
aw
arde
d
th
e
Di
stin
gu
ished
Di
versity
E
nhan
cem
ent
Aw
ard
w
h
e
n
she
wa
s
in
U
SA
f
or
org
a
nizi
ng
w
ork
s
h
ops
t
o
im
prove
p
erf
o
rm
an
ce
o
f
t
h
e
min
o
rit
y
g
ro
ups.
S
he
i
s
also
a
g
old
m
e
dal
recipi
ent
at
t
he
I
nt
ernat
i
on
al
I
nnov
ati
on
and
In
ve
n
t
i
on
Com
p
eti
tion
h
eld
at
G
enev
a
,
S
wit
zerl
a
nd
in
2
0
05.
E-mai
l:
n
.
i
d
ri
s07@gmail.com
Evaluation Warning : The document was created with Spire.PDF for Python.