Statistical analyses were used to test for changes in
scores on the IDQOL, RSMB and POS of both
groups over time. Given the relatively small sample
sizes (SFBT: 18 clients; CAU: 18 clients, two dropouts)
non-parametric tests (the Wilcoxon signedrank
test and Mann–Whitney test) were used to
analyse the data rather than parametric tests. The
Wilcoxon test was used to investigate differences
within the SFBT group and within the CAU group.
The changes in scores on the IDQOL-16, RSMB
and POS before SFBT and after SFBT, and before
SFBT and at follow-up, were analysed for statistical
significance. The direction of the difference (positive
or negative change) was calculated using the
Sign test. The Mann–Whitney test was used to
investigate differences between the SFBT group
versus the CAU group. The non-parametric tests
were performed on the key variables of goal attainment
(via SQP), quality of life (i.e. psychological
functioning, social functioning; via IDQOL-16),
maladaptive behaviour (via RSMB, and resilience
(i.e. autonomy and social optimism; via POS). To control the problem of multiple comparisons, the
Dunn–Bonferroni correction (Dunn 1961) was used
by dividing the P-value by the number of variables:
p/n = 0.05/5 = 0.01. Scaling (SQP) is not an intervention
in CAU, thus no SQP data were available
for the CAU group. Additionally, the number of
clients in both groups who showed improvements
on the outcome measurements are given