Observed Ward Behavior Strongly Associated with Real World Living Skills: An Analysis of Concurrent Validity between NOSIE and ILSI Charlie A. Davidson, Elizabeth A. Cook, Jeffrey R. Nolting, M.A., William D. Spaulding, Ph.D. University of

1
Observed Ward Behavior Strongly Associated with
Real World Living Skills An Analysis of
Concurrent Validity between NOSIE and
ILSICharlie A. Davidson, Elizabeth A. Cook,
Jeffrey R. Nolting, M.A., William D. Spaulding,
Ph.D.University of Nebraska-Lincoln 
There were significant correlations
between the NOSIE Total Assets score and the mean
of the ILSI scores at 3 months, r .60, p
lt.0001, and at 6 months, r .52, p lt.0001. In
addition, there were a number of significant
correlations between most of the NOSIE and ILSI
subscales at both time points (see Figures 1 and
2 for significant r values). Without taking
covariates into account, NOSIE subscales with the
highest number of significant correlations with
ILSI subscales at both 3 and 6 months were Daily
Schedule Competence and Neatness. The NOSIE
subscale with the fewest significant correlations
with ILSI subscales at both times was
Psychoticism. However, correlations were not
always stable across time. The final
models accounted for 30.8 (at 3 months) and
28.1 (at 6 months) of the variance in ILSI
mean, and NOSIE Daily Schedule Competence was the
best and only predictor included in these models.
The standardized regression coefficients show a
positive relationship between ILSI mean and daily
schedule competence, which indicates that greater
ability to adhere to a daily schedule is
associated with higher ILSI overall mean. Factors
such as years of education, number of previous
hospitalizations, age of onset, premorbid
behavioral problems, highest occupational
functioning, and all other NOSIE subscales did
not enter into the final models.
Introduction
Figure 1
Results
Figure 2
As hypothesized, ward behaviors appear to be
significantly related to real world living
skills. NOSIE total and ILSI mean had moderate
correlations at both 3 and 6 months, and there
were a number of moderate correlations between
NOSIE subscales and ILSI subscales at both 3 and
6 months. However, contrary to our hypothesis,
it does not appear that demographic variables (at
least the variables entered into the regression
analyses in this study) significantly mediate the
relationship between NOSIE subscales and ILSI
mean. Daily Schedule Competence appears to be a
strong predictor of ILSI mean, despite the
inclusion of demographic variables in the
analyses. This study demonstrates concurrent and
construct validity of the NOSIE-30 and ILSI,
which increases the clinical value of these
measures. The concurrent and construct validity
of the NOSIE-30 and ILSI appear to be relatively
strong given that ratings of similar behaviors
made by different clinical staff were moderately
and significantly related. This study also has
important implications for future analysis of
ward behaviors in relation to real world living
skills. Future studies should examine the
relationship between NOSIE and ILSI subscales,
totals, and means using a larger sample size. In
addition, other possible covariates, such as
symptom severity level, should be entered into
regression analyses to determine if there are
other variables that mediate the relationship
between these variables. Finally, NOSIE and ILSI
subscales, totals, and means should be entered
into regression analyses as independent factors
to predict outcome measures such as length of
stay.
Discussion
NOSIE Variables ILSI Variables
Positive Dsc Daily Schedule Competence Soc Social Interest Neat Neatness Negative Irr Irritability Psyc Psychoticism Mot Motor Retardation Total Tot Total (Positive Negative) Ilsipe Personal Management Ilsihy Hygiene and Grooming Ilsicl Clothing Ilsiin Interpersonal Skills Ilsiho Home Maintenance Ilsico Cooking Ilsire Resource Utilization Ilsiba Basic Skills Ilsige General Occupational Ilsimo Money Management Ilsimean Mean of ILSI scales
This study examined archival longitudinal data
collected at the Lincoln Regional Centers
Community Transition Program, a 40-bed inpatient
unit. Participantsdiagnoses were generally
those of schizophrenia spectrum disorders,
bipolar disorder, and other severe mood or
personality disorders. Data consisted of 3 and 6
month administrations of the NOSIE-30 and ILSI.
The NOSIE-30 was collected weekly by direct care
staff and reported monthly. The ILSI was
completed at 3 and 6 months post-intake and then
semi-annually, reflected performance during the
most recent month, and was collected by multiple
inpatient staff. Data were cleaned using square
root and log transformations, and were winzorized
whenever possible to minimize skewness. Data
were analyzed using a series of Pearsons
correlations and stepwise regressions to assess
full and partial correlations. Demographic
covariates consisted of the following variables
years of education, number of previous
hospitalizations, age of symptom onset, highest
occupational functioning, and premorbid
behavioral problems. These covariates were
chosen because they had a large number of strong,
significant correlations with NOSIE and ILSI
variables.
Method