Use of VIP test for persons with mental retardation
About the VIP Test | Review of the VIP Test
The following text comes from the VIP (Validity Indicator Profile) Manual, product number 51767.
Frederick, R. I. (1997). VIP (Validity Indicator Profile) Manual. Minneapolis: NCS Pearson, Inc.
Excerpted from the VIP (Validity Indicator Profile) Manual. Copyright © 1997, 2003 NCS Pearson, Inc. All rights reserved.
The 40 persons with mental retardation were included in the cross-validation sample to provide normative data and information about VIP® applicability in this clinically important population. Tables 1 and 2 below provide normative data on VIP scores and Performance Curve measures for this group. Tables 3 and 4 show the VIP response style classifications for this group on the Nonverbal and the Verbal subtest, respectively, as well as the mean Shipley IQ for each classification group.
As shown in Tables 1 and 2, on both VIP subtests the mean Total Score for those individuals with a bona fide history of mental retardation was higher than that for the computer-generated random protocols but lower than that for all other participant groups, including the "noncompliant" coached normals and patients at risk for malingering. Although there is no reason to believe that these individuals did not cooperate with the testing, only 20% received a Compliant classification on the Nonverbal subtest (Table 3), and only 25% on the Verbal subtest (Table 4). An Irrelevant classification was received by about a third of this sample on the Nonverbal subtest, and by half the group on the Verbal subtest.1
The Shipley IQs of these individuals ranged between 54 and 75. Despite this limited range, there was nevertheless a pronounced relationship between IQ and VIP response style classification. On both subtests, subjects classified as Compliant had a higher mean IQ than those classified as Inconsistent, who in turn had a higher mean IQ than those classified as Irrelevant. The rightmost columns of Tables 3 and 4 show the effect on VIP response style classification of dividing the overall sample at the median IQ of 64.5. On both subtests, the performance of those with IQs above the median was much more likely to be classified as Compliant, whereas the performance of those with IQs below the median was much more likely to be classified as Irrelevant.
The relationship between IQ and VIP response style classification in this sample of low-IQ individuals suggests that those with extremely low ability may stop trying after they encounter more difficult items mixed among easy ones. The organization of items on the VIP test is randomized with respect to difficulty, and some individuals may assume after trying the first few that they will not know the answers to any items. Accordingly, they may simply fill in the answer blanks randomly after a few items.
If this hypothesis is true, special test administration techniques may help individuals with low ability who want to perform well to persist in attempting items, thereby increasing their Total Score and their chances of receiving a Compliant response style classification. It is permissible to administer the Nonverbal items by individually pointing to each item and asking for a response, encouraging the individual to attempt each item. For the Verbal items, it is permissible to say, "Answer the ones you know; then try the other ones after that. It's okay to guess." Instructions such as these can help individuals who appear to have low ability maintain participation in the task.
In summary, the data presented here recommend against the use of the VIP test to evaluate the validity of cognitive testing for persons who are known from historical information to have bona fide mental retardation; too many cooperative individuals may be mistakenly classified as not cooperating fully with the testing. On the other hand, it is common in many examination settings to encounter individuals who perform extremely poorly on cognitive tests but who are not known from other evidence to have mental retardation. In these instances, malingering must be considered as a possible explanation for the low ability score, and use of the VIP test is appropriate. The interpretation of the VIP results in this situation is addressed in Chapter 7 of the VIP (Validity Indicator Profile) Manual.
Table 1: Nonverbal subtest statistics by participant group
|
Compliant groups |
Noncompliant groups |
Persons with mental retardation |
|||||||||||||||
Honest normals |
Brain injury patients |
Patients at risk for malingering |
Coached normals |
Random protocols |
||||||||||||||
n |
M |
SD |
n |
M | SD | n |
M |
SD |
n |
M |
SD |
n |
M |
SD |
n |
M |
SD |
|
Total score |
100 |
87.6 |
9.8 |
61 |
78.6 |
11.8 |
49 |
74.2 |
14.2 |
52 |
63.7 |
16.2 |
50 |
50.8 |
5.1 |
40 |
56.7 |
8.3 |
Adjusted score |
100 | 75.5 | 17.9 | 61 | 57.3 | 23.5 | 49 | 49.1 | 27.0 | 52 | 34.6 | 24.4 | 50 | 8.0 | 6.4 | 40 | 16.4 | 13.7 |
Performance curve measuresa |
||||||||||||||||||
| Sector 1 Distance | 100 | 70.3 | 20.1 | 59 | 52.6 | 21.5 | 45 | 47.8 | 27.0 | 33 | 36.2 | 26.6 | 1 | _b | _b | 26 | 18.0 | 16.5 |
| Sector 2 Distance | 88 | 13.8 | 10.1 | 59 | 18.9 | 13.7 | 42 | 17.4 | 11.0 | 32 | 23.6 | 15.0 | 1 | _b | _b | 26 | 18.7 | 13.2 |
| Sector 3 Distance | 53 | 18.9 | 20.6 | 48 | 26.2 | 20.9 | 36 | 35.9 | 21.5 | 31 | 36.0 | 25.1 | 1 | _b | _b | 26 | 56.3 | 17.7 |
| Sector 1 Residual | 100 | .012 | .010 | 59 | .018 | .015 | 45 | .029 | .029 | 33 | .044 | .030 | 1 | _b | _b | 26 | .060 | .036 |
| Slope | 100 | -.004 | .002 | 61 | -.005 | .002 | 49 | -.005 | .003 | 52 | -.004 | .003 | 50 | .000 | .002 | 40 | -.004 | .003 |
| Point of Entry | 100 | .99 | .02 | 61 | .96 | .09 | 49 | .93 | .13 | 52 | .78 | .22 | 50 | .51 | .13 | 40 | .77 | .21 |
| Peak Performance Interval | 94 | 36.3 | 23.0 | 49 | 28.8 | 20.6 | 34 | 26.9 | 15.2 | 14 | 14.0 | 9.4 | 0 | _ | _ | 9 | 9.4 | 11.2 |
| Suppression Sector Distance | 1 | _b | _b | 0 | _ | _ | 4 | 22.8 | 1.7 | 4 | 44.3 | 15.0 | 1 | _b | _b | 0 | - | - |
| Suppression Sector Starting Point | 1 | _b | _b | 0 | _ | _ | 4 | 65.8 | 6.7 | 4 | 23.3 | 18.2 | 1 | _b | _b | 0 | - | - |
a See Chapter 4 of the VIP (Validity Indicator Profile) Manual for an explanation of the rules that underlie the varying sample sizes for different Performance Curve measures.
b Statistics are not reported for Performance Curve measures that apply to fewer than two cases.
Table 2: Verbal subtest statistics by participant group
Compliant groups |
Noncompliant groups |
Persons with mental retardation |
||||||||||||||||
Honest normals |
Brain injury patients |
Patients at risk for malingering |
Coached normals |
Random protocols |
||||||||||||||
n |
M |
SD |
n |
M |
SD |
n |
M |
SD |
n |
M |
SD |
n |
M |
SD |
n |
M |
SD | |
Total score |
100 |
64.5 |
4.2 |
60 |
58.1 |
7.5 |
48 |
59.5 |
8.0 |
52 |
49.4 |
12.3 |
50 |
39.3 |
4.3 |
40 |
45.6 |
7.2 |
Adjusted score |
100 |
51.0 |
8.5 |
60 |
39.2 |
12.0 |
48 |
41.5 |
14.7 |
52 |
27.3 |
16.8 |
50 |
6.4 |
5.6 |
40 |
15.1 |
12.4 |
Performance curve measuresa |
||||||||||||||||||
Sector 1 Distance |
100 |
51.5 |
9.1 | 57 | 40.5 | 14.6 | 46 | 45.1 | 13.7 | 40 | 26.5 | 19.2 | 1 | _b | _b | 19 |
20.2 |
12.3 |
Sector 2 Distance |
100 |
10.0 | 6.6 | 57 | 13.8 | 10.0 | 46 | 10.5 | 6.9 | 40 | 19.2 | 11.8 | 1 | _b | _b | 19 |
14.8 |
9.3 |
Sector 3 Distance |
44 |
8.5 |
4.7 | 45 | 12.0 | 9.4 | 29 | 14.1 | 9.9 | 33 | 21.8 | 15.8 | 1 | _b | _b | 19 |
28.1 |
15.0 |
Sector 1 Residual |
100 | .008 |
.007 | 57 | .015 | .012 | 46 | .014 | .009 | 40 | .034 | .029 | 1 | _b | _b | 19 |
.041 |
.037 |
Slope |
100 | -.008 | .003 | 60 | -.009 | .003 | 48 | -.009 | .003 | 52 | -.006 | .006 | 50 | .000 | .003 |
40 |
-.004 |
.005 |
Point of Entry |
100 | 1.00 |
.02 | 60 | .97 | .13 | 48 | .97 | .13 | 52 | .85 | .24 | 50 | .46 | .16 |
40 |
.74 |
.23 |
Peak Performance Interval |
97 | 34.0 | 11.5 | 54 | 23.6 | 12.8 | 44 | 25.0 | 14.6 | 27 | 15.4 | 14.3 | 0 | - | - |
12 |
12.3 |
8.0 |
Tail Score |
100 |
1.7 | 1.2 | 60 | 1.7 | 1.6 | 48 | 1.5 | 1.1 | 52 | 2.5 | 1.5 | 50 | 4.0 |
1.3 |
40 |
3.4 |
1.6 |
| Suppression Sector Distance | 0 | - | - | 1 | _b | _b | 0 | - | - | 5 | 41.4 | 14.7 | 1 | _b | _b | 0 |
- |
- |
Suppression Sector Starting Point |
0 |
- | - | 1 | _b | _b | 0 | - | - | 5 | 9.6 | 17.1 | 1 | _b | _b | 0 |
- |
- |
a See Chapter 4 of the VIP (Validity Indicator Profile) Manual for an explanation of the rules that underlie the varying sample sizes for different Performance Curve measures.
b Statistics are not reported for Performance Curve measures that apply to fewer than two cases.
Table 3: Performance of persons with mental retardation on the nonverbal subtest
Total group |
IQ < 64 or = 64 |
IQ = 65 or > 65 |
|||||
Response style |
n |
% |
Mean IQ |
n |
% |
n |
% |
Compliant |
8 | 20 | 68.9 | 1 | 5 | 7 | 35 |
Inconsistent |
18 | 45 | 63.6 | 10 | 50 | 8 | 40 |
Irrelevant |
14 | 35 | 62.6 | 9 | 45 | 5 | 25 |
Suppressed |
0 | 0 | - | 0 | 0 | 0 | 0 |
Total |
40 | 100 | 64.3 | 20 | 100 | 20 | 100 |
Table 4: Performance of persons with mental retardation on the verbal subtest
Total group |
IQ < 64 or = 64 |
IQ = 65 or > 65 |
|||||
Response style |
n | % | Mean IQ | n | % | n | % |
Compliant |
10 | 25 | 70.8 | 0 | 0 | 10 | 50 |
Inconsistent |
9 | 23 | 65.4 | 4 | 20 | 5 | 25 |
Irrelevant |
20 | 50 | 60.7 | 15 | 75 | 5 | 25 |
Suppressed |
1 | 3 | 60.0 | 1 | 5 | 0 | 0 |
Total |
40 | 100 | 64.3 | 20 | 100 | 20 | 100 |
Footnotes
1 The one person with mental retardation who received a Suppressed response style classification on the Verbal subtest almost certainly employed random responding rather than deliberate suppression. As noted in Chapter 4 of the VIP (Validity Indicator Profile) Manual, purely random protocols will be misclassified as Suppressed about 6% of the time.