Do Better Trainees Make Worse Family Therapists? A Followup Study of Client Families

• Author: YVONNE STOLK,AMARYLL J. PERLESZ
• Date: 01 March 1990
• DOI: http://doi.org/10.1111/j.1545-5300.1990.00045.x
• Published date: 01 March 1990

Fam Proc 29:45-58, 1990

Do Better Trainees Make Worse Family Therapists? A Followup Study

of Client Families

YVONNE STOLK, M.A.a

AMARYLL J. PERLESZ, M.A.b

aResearch Psychologist, Psychiatric Epidemiology & Services Evaluation Unit, Parkville, Victoria, Australia.

bSenior Clinical Psychologist, Bouverie Family Therapy Center, 35 Poplar Road, Parkville, Victoria 3052, Australia.

This study was designed to evaluate the effects of training on family therapy outcome and to validate measures of

cognitive/perceptual and executive skills. Followup data on family satisfaction were obtained from 176 individuals (73

families) treated by 12 trainee family therapists in their first or second year of training. The prediction that clients'

scores on a scale of family satisfaction with therapy (FSS) would increase with training was not confirmed. Instead, there

was a tendency for FSS scores to decrease with training. However, trainees' ratings of outcome did increase significantly

as training progressed. Cognitive/perceptual and executive skill measures were not validated by the FSS, but negative

associations indicated a tendency for trainees with high course-skills to receive poor FSS scores. The results suggest that

trainees became more confident as they acquired technical skills, but that they put aside their nonspecific therapist skills

as training progressed, to the dissatisfaction of families.

While a number of studies have recently tested the effectiveness of family therapy training by developing measures to

assess gains in cognitive skills and executive skills in simulated situations (20, 24), we are not aware of any studies to date

that have attempted to respond to Kniskern and Gurman's (17) challenge almost a decade ago:

[T]he most important, sensitive and difficult [area to consider when assessing whether a trainee has changed as a

result of training], is that of the outcomes of the families treated by the trainee. No training program can

responsibly be said to be effective unless its graduates can be demonstrated to produce more positive effects and

fewer negative effects with the families they treat after receiving training than before receiving training. [p. 91]

At best, some studies have used a reported increase in the number of families seen by trainees following training as an

indicator of the effectiveness of their training programs (2, 3). Crane, Griffin, and Hill (5) did attempt to analyze clients'

perceptions of the skill levels of their trainee therapists 6 and 18 months after therapy. However, in the absence of pre- and

post-course trainee measures, the authors' focus was on the "insiders'" (clients') perspective of therapy, rather than on the

effectiveness of the training process itself.

We believe the current study to be the first attempt in the field both to test the effects of training on families' satisfaction

with therapy and to validate measures of cognitive/perceptual and simulated executive skills against family oucome. It also

reports on the complexities of designing an adequate study for this purpose, which may explain why it has taken so long for

such a validation to be attempted.

Trainees who completed a 2-year training program in family therapy were assessed, before and after training, in the areas

of perceptual/conceptual skills, executive skills, and personal development. The results of this first phase of the study

showed that trainees, as a group, performed significantly better in all areas assessed at the end of training (19). In the

second phase of the study, as an integral part of the evaluation of training, treatment outcome data on families were

gathered from the perspective of the trainee therapist and from the perspective of families treated during training.

Outcome Methodology and Training Evaluation

The primary purpose of the current study was not to explore issues of family therapy outcome, but to evaluate the

effectiveness of the training course in preparing trainees for successful work with families. It is impossible, however, to do

the latter effectively without being drawn into the methodological minefield of outcome research (10, 13, 15, 26). The

initial hurdle to be surmounted is that there are two levels of outcome to considerthat of the trainee and that of the

familyand, hence, two levels at which experimental controls should be instituted.

A number of reviews in recent years have enumerated the factors that must be controlled before conclusions can be

drawn about the overall effectiveness of family therapy and the factors that contribute to effectiveness. Wells, Dilkes, and

Trivelli (26) and Jacobson (15) consider the use of control groups and random assignment to groups a minimal requirement

in family therapy outcome research. Control groups should receive either an alternative treatment to test the comparative

effectiveness of treatments or no treatment. Families in the two groups should be matched on key criteria, and measures of

______________________________________________________________________________________________________________

1