Basic HTML Version
336
F.
J.
EVANS
AND
D. SCHMEIDLER
x
7.15
6.25
3.55
TABLE
1
Distribution
of
Scores
for
SHSS:C
(N
=
60)
S.D.
--
2.22
2.28
1.80
.
N
. -
20
- 2 0
20
-
HGSHS:
A'
High
Medium
Low
SS
divided into subgroups from
HGSHS:A
performance. High
Se
scored
8
or
above on HGSHS:A, medium Ss scored between
7
and
5,
low
Ss
scored
4 or
below. Heavy lines indicate
HGSHS:A
subgroups.
N
=
20
in
each group.
Weitzenhoffer, Landes, and Moore (1961) and Hammer et al. (1963)
found little evidence of practice effects on similar items. The common
items are embedded within items of relatively different difficulty
;
however, Field et al. (1965) found that order of item administration
had no apparent effect on item scores. I n spite of such findings, items
do not necessarily maintain similar characteristics in the two scales.
Thus, the number of
Ss
passing the two common challenge suggestions:
Arm Rigidity and Arm Immobilization, is higher in SHSS:C (57% and
62%
respectively) than in HGSHS:A
(38%
and 30%).
Two
common
items, Fly Hallucination and Amnesia, correlate poorly between scales
(.33 and .37 respectively).
The correlation between summed scores on common items is .64.
When corrected by the Spearman-Brown formula for equivalent 12-
item length, this value is
.78.
This value is comparable to the Kuder-
Richardson reliabilities of both scales
(.74
and .75 respectively). The
summed scores on noncommon items correlate .41, which, when cor-
rected for a 12-item test, is .60. The corrected correlations between
scores
on
common and noncommon items differ significantly
( p
<
.05),
Item Discrimination between HGSHS
:A
Groups. Table
2
presents
the frequency
of
Ss
passing a given item (item difficulty)
of
HGSHS:A
within the high, medium, and low subgroups. Similar distributions
of
SHSS:C scores
for
these same subgroups are presented in Table 3.
Two HGSHS:
A
items fail to discriminate between the subgroups
a t the
.05
level of significance: Head Lowering and Fly Hallucination.
Neither item discriminates between high, medium, and low subgroups
in
the expected order. Biserial correlations
for
these items with cor-
rected total scores are low (.28 and .23 respectively). Item
1:
Head
Falling, item 3: Hand Lowering, and item
7:
Moving Hands Together
do not discriminate between high and medium groups, while item 2:
Eye Closing does not discriminate well. The biserial correlations for
SHSS:
c
Score
0
1 2
3
4
5
6
7
8
9 10
--
-_-------
- 1
- -
1 1 3 5 5 2 1
- 1
- 2 1 2
5 3
3
1 2
2
1 1 6
4
2
4
- - - -
Downloaded by [University of Macedonia] at 04:28 30 March 2012