Journal of Rehabilitation Medicine 51-8 | Page 36

572
L. Turner-Stokes et al.
Table V. The UK FIM+FAM conversion table based on raw and interval scores
Interval
Raw Score Scale Interval
Raw Score Scale Interval
Raw Score Scale Interval
Raw Score Scale Interval
Raw Score Scale Interval
Raw Score Scale
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59 60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89 90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119 120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149 150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179 180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
30.0
30.0
60.0
74.8
82.1
89.2
90.8
93.5
95.8
97.6
99.2
100.6
101.9
102.9
103.9
104.8
105.6
106.3
107.0
107.7
108.3
108.8
109.4
109.9
110.3
110.8
111.2
111.6
112.0
112.4
113.0
113.4
113.7
114.0
114.3
114.5
114.9
115.1
115.3
115.6
115.8
116.0
116.2
116.4
116.6
116.8
117.0
117.2
117.4
117.5
117.7
117.8
118.0
118.2
118.3
118.5
118.6
118.8
118.9
119.0
119.1
119.2
119.3
119.4
119.6
119.7
119.8
119.9
120.0
120.1
120.2
120.3
120.5
120.4
120.6
120.7
120.8
120.9
121.0
120.9
121.1
121.2
121.3
121.3
121.5
121.6
121.5
121.6
121.8
121.8
excellent targeting of the sample by the items thresholds
with the person mean of –0.04 (SD 0.52). This justifies
using a smaller sub-sample (n = 320) to estimate a χ 2
statistic sensitive to the sample size. Table V shows the
conversion table for the full-scale model. We then cor-
related the interval scores from Table V with the interval
scores obtained for the same FIM+FAM raw score in
Fig. 3. Scatterplot of interval vs ordinal data. A scatterplot illustrating
the relationship between the UK Functional Assessment Measure (UK
FIM+FAM) ordinal summed raw scores and Rasch interval scores
comparing traumatic brain injury (TBI) with left and right strokes.
www.medicaljournals.se/jrm
121.9
121.9
122.2
122.1
122.3
122.4
122.3
122.5
122.6
122.7
122.8
122.9
122.9
122.9
123.0
123.2
123.1
123.3
123.3
123.4
123.5
123.6
123.7
123.8
123.9
124.0
124.1
124.2
124.3
124.4
124.6
124.6
124.8
124.9
124.9
125.0
125.2
125.3
125.5
125.6
125.7
125.9
126.0
126.2
126.3
126.5
126.6
126.9
127.0
127.2
127.4
127.6
127.9
128.0
128.3
128.5
128.8
129.1
129.4
129.7
130.0
130.4
130.7
131.1
131.5
131.8
132.3
132.8
133.2
133.8
134.4
135.0
135.6
136.3
137.0
137.9
138.7
139.8
140.8
142.0
143.3
144.9
146.6
148.7
150.4
154.0
157.8
162.9
170.5
184.5
210.0
our previous Rasch analysis with a UKROC stroke
dataset to determine whether a single conversion table
could suffice. Fig. 3 shows a scatterplot illustrating the
relationship between ordinal summed raw scores and
Rasch interval scores comparing TBI with left and right
strokes. To investigate whether separate conversion
tables for TBI and stroke patients were necessary, we
pooled our previously reported stroke samples (n = 680
left; nn = 638 right stroke) (17) with a randomly selected
TBI sample of comparable size (n = 680) and conduc-
ted DIF analysis for 3 super-item solution. Significant
DIF was identified between left stroke and TBI for all
3 super-items: Motor (F(1, 1316) = 152.03, p < 0.001),
Communication (F(1, 1316) = 59.84, p < 0.001) and
Psychosocial (F(1, 1316) = 137.94, p < 0.001). Sig-
nificant uniform DIF was also identified between
right stroke and TBI samples for the Communication
(F(1, 1358) = 8.64, p = 0.003) and Psychosocial (F(1,
1358) = 19.40, p < 0.001) super-items. DIF plots (A-
F) are included in Fig. S2A–F. These would suggest
that separate conversion tables for stroke and TBI are
necessary. However, as demonstrated by Fig. 3, the
differences lie predominantly in the middle range of
the scale and further work is required to determine
whether they are sufficiently meaningful at a clinical
level to justify separate transformation tables for the
different conditions.