DVISHORT FORM
The DVIShort Form is a brief, easily administered and automated
(computerscored) adult domestic violence offender screening instrument or
test. The DVIShort Form consists of 76 items and takes 15 to 20 minutes to
complete.
The DVIShort Form evolved from the Domestic Violence Inventory (DVI)
and contains variations of the six DVI scales. The DVIShort Form assesses
attitudes and behaviors important in domestic violence offender profiling.
Paperpencil test administration takes on average 17 minutes, and tests are
computerscored onsite within 2˝ minutes. The DVIShort Form was developed
specifically for domestic violence offender evaluation. It is much more than
just another alcohol and drug test; it measures important
attitudes and behaviors missed by other tests.
DVIShort Form Scales (Measures)

1. Truthfulness Scale 
3. Alcohol Scale 
5. Control Scale 
2. Violence Scale 
4. Drugs Scale 
6. Stress Coping Abilities 
At one sitting of approximately 15 to 20 minutes' duration, staff can acquire a
vast amount of information. The DVIShort Form collects information on offender
acting out (violence), controlling tendencies, substance abuse and stress
coping abilities. Early problem identification facilitates timely intervention
and improved outcome.
Reliability and Validity
The DVIShort Form has a proprietary builtin database that insures inclusion
of all administered tests in a confidential (no names) manner. Tens of
thousands of DVI and DVIShort Form tests' data are represented in the DVI
database. These reliability, validity and accuracy statistics are reported in
the document titled "DVI: An Inventory of Scientific Findings," which can be
read by clicking on the “DVI Research” link in the links table on the home
page. This research also presents the impressive correlation coefficients
between DVI scales and their analogous DVIShort Form scales.
In this analysis, Domestic Violence Inventory (DVI) scales were correlated with
analogous DVI ShortForm scales. Pearson’s r correlation coefficients can range
from zero to 1, where zero means there is no association between variables and
1 means that two variables are perfectly correlated. As shown in the following
table, correlation analysis of DVI scales with DVIShort Form scales all
yielded coefficients very close to 1.0. This is to be expected because only the
statistically strongest items in the DVI were selected for inclusion in the
DVIShort Form. Thus, the DVIShort Form scales are close approximations to
analogous DVI scales.
The DVIShort Form scales correlate highly significantly with DVI scale scores.
DVISHORT FORM CORRELATION
COEFFICIENTS (N=5,416) 2010

Scales:

Standard DVI

Truthfulness 
r = .95

Alcohol 
r = .97

Control 
r = .98

Drugs

r = .98

Violence 
r = .95

Stress
Coping Abilities

r = .98

For domestic violence offender assessment, test users now have a choice: DVI or
DVIShort Form.
Internal consistencies (coefficient alphas) for DVIShort Form scales are
reported in the following table for 5,416 domestic violence offenders assessed
in the year 2010.
RELIABILITY OF THE DVISHORT FORM
(N=5,416) 2010

DVIShort Form Scales

Coefficient
Alpha

Significance
Level

Truthfulness

.90

p<.001

Violence (Lethality) 
.88

p<.001

Control

.85

p<.001

Alcohol 
.91

p<.001

Drugs 
.90

p<.001

Stress Coping Abilities 
.90

p<.001

All DVIShort Form scales have alpha coefficients well above the professionally
accepted standard of .75 and are highly reliable. All coefficient alphas are
significant at the p<.001 level.
Early DVI studies used criterion measures and were validated with other tests,
e.g., Minnesota Multiphasic Personality Inventory (MMPI) LScale and FScale,
16PF, MacAndrews, MMPI Psychopathic Deviate, Treatment Intervention Inventory,
SAQAdult Probation III, etc. Much of this research is summarized in the
document titled "DVI: An Inventory of Scientific Findings." All DVI scales
correlate significantly with corresponding DVIShort Form scales.
It's reasonable to conclude that the Domestic Violence Inventory (DVI) and
DVIShort Form are evidence (reliability, validity, accuracy, scale
correlations, etc.) based.
Navigation links are at the top of this webpage.
* * *