Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)



Department name when degree awarded


Committee Chair(s)

Elwin Nielsen


Elwin Nielsen


Statement of the problem

Psychology is currently unable to demonstrate that psychotherapy is effective because it has no reliable no-inference objective measures of behavior and behavior change. Reliable measures of therapeutic outcomes are currently nonexistent. The problem with which this study was concerned was one of reliably evaluating psychotherapy outcomes and behavioral change.


The objective of this study was to develop and assess new measures of behavioral change. The assessment was directed toward those distressing behaviors which cause adult clients to seek psychotherapy.

Methods and procedure

Development of a Behavioral Checklist which would assess the distressing behaviors which brought clients to therapy required a list of specific behaviors which may cause a client stress. A questionnaire which would measure stress required not only a list of behaviors, but also a measure of the stress caused by each behavior. A Behavioral Checklist which would measure the degree that specific behaviors were representative of the client's behavior was developed and each behavior was quantified by the S's self-report, as to degree of stress. When the degree that the behavior represents the client was multiplied times the amount of stress caused by that behavior, the result was a multiplicative measure of the behavior and the distress concomitant with that behavior which brought the client to therapy.

The test-retest reliability of the Behavioral Checklist for a one week period was obtained from a University class of 21 undergraduate students. This reliability study revealed that 40 of the questionnaire's original 203 items had low test-retest reliability. After the 40 unreliable questions were deleted, the questionnaire's reliability ranged from .816 to .865 depending upon which of the first three scoring techniques were used.

Following its development, the Behavioral Checklist was used to evaluate a "clinical" population to determine if it would measure change due to psychotherapy.

Findings of the Study

Using the therapist's ratings of improvement, the Ss were divided into an improved and an unimproved group. At test for each of the three scoring techniques used on the Behavioral Checklist revealed that Ss judged improved had significant gains on each scoring technique post therapy while Ss not so judged did not have significant gains.

The Behavioral Checklists, pre less post scores, yield correlations varying from .54 to .78 when correlated with the therapist's rating of therapy outcomes. Specific Target Behaviors, which are operational definitions of the reasons clients came to therapy, were scored and evaluated in the same manner as the Behavioral Checklist. The Specific Target Behaviors yielded correlations ranging from .33 to .407 when correlated with the therapist's rating of therapy outcomes.

Summary and Conclusions

The current study indicates that a client's self-report of no-inference objective behaviors will measure change resulting from psychotherapy. Self-report no-inference behaviors will measure change due to psychotherapy whether they are from a prepared list of no-inference behaviors (Behavioral Checklist) or operational definitions of the client's own statements about the reasons that he came to therapy {Specific Target Behaviors). The results of the current study indicate that using no-inference, overt, quantifiable behavior is a fruitful area in which to study therapy outcomes with a clinic population.



Included in

Psychology Commons