Incontinence impact

Internal consistency is usually measured with Cronbach‟s
alpha, a statistic calculated from the pairwise correlations
between items. Internal consistency ranges between
negative infinity and one. It is expected that items
forming a domain of the questionnaire should correlate
moderately with each other, but should contribute
independently to the overall score in that domain. Very
high reliabilities (0.95 or higher) are not necessarily
desirable, as this indicates that the items may be entirely
redundant. Similarly very low reliability index suggests
that researcher is trying to assess different traits of the
condition which are not related to each other. An alpha
value of ≥0.7 is generally considered as acceptable in
reliability studies. These values can be easily derived
from SPSS software which is a very well-known
statistical package for medical professionals in
academics. Table 2 gives interpretation of Cronbach‟s
alpha values.
Table 2: Internal consistency measures.
Cronbach’s alpha Internal
More than 0.9 Excellent
0.8 to 0.9 Good
0.7 to 0.8 Acceptable
0.6 to 0.7 Questionable
0.5 to 0.6 Poor
Less than 0.5 Unacceptable
The next procedure is to carry out what test-retest
reliability. Test-Retest reliability means the study
participants are consistently giving the same score even
when the test conducted on two different occasions. In
order to measure the test-retest reliability, we have to
give the same test to the same respondents on two
separate occasions. Then the mean and deviation of each
item in the domain is calculated for two different
occasions. If the values lie close to each other, then it
would mean that the questionnaires are good and the test
results are reproducible. However one should know that
the time interval should be reasonably short, as longer
intervals may be associated with improvement in
symptoms (especially in follow-up studies after
therapeutic interventions for stress urinary incontinence)
and the values may differ significantly from each other.
Finally, one can test what is called as “criterion validity”,
which means whether the results of QoL measure in
question correlates well with other well established
scales. For example, we may want to compare KHQ with
other urinary symptom assessment tools (for example,