The TREND TOWER displays average responses by individual question on the survey.
The survey questions are numbered down the left side of the "tower." The
response values on a modified Likert Scale from zero to nine are listed across the
top of the "tower." The "0" indicates "Total Disagreement"
while the "9" signifies "Total Agreement" with a specific statement.
The MEAN is a mathematical average of the data element. In compiling results
for this survey the MEAN is calculated from all respondents in the survey pool. Those
who chose not to respond to a specific question are statistically insignificant for
our purposes--however, when non-responses occur the mathematical package used to
process this survey treats that response as total disagreement. The MEAN is calculated
using the quantity of responses to each possible selection on the zero to nine scale
as a basis. On the Trend Tower, the MEAN line zig-zags down the tower on the left-it
is the first data line running down the tower, when reading from left to right.
The
MODE is determined by the quantity of respondents who selected a specific number
on the zero to nine scale. The scale number selected most frequently is the MODE.
In cases of more than one mode, the lowest mode is plotted. On the Trend Tower, the
MODE line often falls to the right of the MEAN line. When the MODE line falls to
the left of the MEAN line, a cultural inversion has occurred (a signal that many
people were more pessimistic than the average rating for the particular question
might indicate).
The AVERAGE is calculated to provide a middle point of all
the combined MEAN points plotted. One can thus determine above or below average plots
easily. It is this average calculation that becomes the overall Quality of WorkLife
Index (QWLI) for the organization or organizational unit. The AVERAGE appears on
the Trend Tower as a large-dashed line.
The STANDARD ERROR indicator is plotted
on either side of the AVERAGE and indicates the range within which the true AVERAGE
of the means will statistically occur, accounting for potential chance errors in
data collection methodologies. On the Trend Tower the standard error lines are the
small-dashed lines to the left and the right of the AVERAGE.
The MAXIMA and
MINIMA identify the "best" and "worst" survey item responses.
The five highest rated items appear as a rank ordering from one to five on the right
side of the Trend Tower. The five lowest rated items appear as a rank ordering from
one to five on the left side of the Trend Tower. Successes worth publicizing and
issues worth addressing can emerge from those maxima and minima responses that the
Trend Tower identifies.
1. Check for anomalies. Look for both positive and negative plots that stand out
from what appears to be a normal range.
2. Check for inversions. When searching
for trends, identify those areas where the Mode is "lower" than the Mean.
This occurrence, called an "inversion," may indicate a negative critical
mass in the organization. These inversions must be taken seriously and prioritized
when making changes or improvements.
3. Check for wide diversity of opinion.
Search also for those instances where the difference between the Mode and Mean is
greatest. Where opinion is varied on an important subject (wide dispersion of opinion)
it may be difficult to change because people see the situation differently. Normally,
it is easier and more productive to unite opinion before undertaking a change effort.
4.
Check for relationships between similar questions. In all cases, use common sense.
Organization Diagnostics, Inc. gathers data using at least two methods--numeric scaled
responses, and one-on-one interviews, for example. Information produced in these
two modalities ought to correlate. Responses to like questions ought to produce the
same or similar results.
5. Check the UHS element associated with low ranking
questions. The specific question asked may help identify action/solution needs. Also,
the UHS element category (e.g.: Leadership and Management, People, Skills, etc.)
may provide a lens through which to look for other related issues that might be associated
with the low-rated question.
6. Schedule a Coaching Lab for a Focus on a Particular
Department or Work Group.