Performing and Using Assessment in Rehearsals for Growth
Daniel J. Wiener, PhD, RDT-BCT
March 3, 2021
Part One of this blog offers a primer for those less familiar with the general topic of assessment.
Part Two addresses the particular features of assessing individuals in Rehearsals for Growth.
Part Three addresses specific features of assessing couples in Rehearsals for Growth.
Part One: What is Assessment?
Generally speaking, we use the word “Assessment” in therapy or in psychology to refer to a kind or a process of testing, which may use a variety of different techniques. The broad purpose of assessment is to help us to arrive at a hypothesis about what’s going on– usually with an individual person, where we’re looking at personality traits or at behavioral capabilities.
In the context of therapy, most assessment is done in the early stages of therapy and usually is directed towards inquiring about some suspected dysfunction or limitation that a client has– we’re looking to see what they can’t do, or what they can’t do well, to identify gaps/deficiencies /limitations. Typically, the observations which you, as the therapist, make are informed by and directed toward certain concepts and certain rules that are being used in order to shape and narrow your focus on what you are looking for. Not that you’re incapable of noticing things other than what you started out as interested in, just that it takes a focusing of attention to really dive into what it is that constitutes a useful assessment.
So why do we bother with assessment? Well, the general answer is, we’re interested in having a better understanding in order to direct the making of changes in the client’s feelings and/or functioning. In our work as therapists we are assessing for the ultimate purposes of overcoming the client’s limitations and/or enhancing a working relationship with that client. Sometimes, though, people are sent to us– then, we are conducting assessment to serve the purposes of outside others who have an interest in our client– this happens, for example, when a judge orders an evaluation of our client.
I wish to make the distinction here between two kinds of assessment. The first relies on self-report, which is predominantly verbal, and which is directed toward things which are not easily observed by others and which are thought to be valuable to know. In self-report’s oral form, if I ask you “do you have a headache?” I am looking for information which is best gained by engaging you and inviting a response to that question. Your answer gives me something that I could put to use and that might also assist our relationship. Some self-report is volunteered by a client in contrast with the self-report which is elicited as responses to our questions. An interview is really a more formalized way of eliciting self-report, depending on how structured it is. When we have certain standardized questions which are being asked of all clients, such as: “have you ever thought of suicide? Or, “do you do use drugs?” we’re often comparing these responses to what are believed to be normative ones. In psychotherapy, of course, not only is the content of the answer of interest but also the paraverbal cues accompanying clients’ answers: does the client answer defensively, adding explanations or justifications? What affect accompanies their answer? Are they making or avoiding eye-contact as they reply? Attention to these cues may provide other valuable information for assessment purposes; sometimes, paraverbal cues convey more valuable information than does the content.
In self-report’s written form, personal essays on open-ended topics (e.g., “tell me/us about yourself”) or open-ended sentence-completion questions (e.g., “my father is someone who____”) may be used to gain information, form global impressions and assess cognitive skills. By far the most frequently-used written form is the written questionnaire, test or psychological inventory. The majority of these inventories consist of items that are scored numerically and generate combined measurements expressed as “scores.” Thousands of these inventories have been constructed; they vary not only in content and purpose but also in how thoroughly they have been standardized and validated. The advantages of written inventories are: convenience of administration, automation of test scoring and statistical analyses of subscales tied to empirical comparisons of subgroups and defined populations.
Psychometricians, professionals who construct, administer and study testing, regard two concepts, reliability and validity, to be of importance. A reliable test will produce a similar result every time. Because any measurement or data point is generated under unique circumstances at a single moment in time we cannot generalize if that measurement, when made under different circumstances and/or at a later time, gives a different result. An adult client’s height is a highly reliable measure, insofar as the resulting measurement over different circumstances and at different times will produce a similar result. On the other hand, asking a client how anxious (s)he feels at the present moment may produce quite different answers at different times and under varying conditions. Because of this, a single self-report of experienced anxiety is not a decisive measure that will allow us confidently to diagnose a client as having Generalized Anxiety Disorder (due to the low reliability of that single data point).
Validity, an essential part of a reliable psychometric test, helps to determine whether or not a test measures what we believe it to be measuring. Validity is not regulated by a single test or statistic but by a body of research aimed solely at establishing validity, a process that is costly in time and effort and is usually done only when the measure is thought to be of considerable importance. A common error is made when we overgeneralize from specific data about what we assume is an underlying construct or variable. If I ask Joe whether he has ever played basketball, it would be erroneous to infer that a “yes” answer tells me that he is taller than average; some taller people never had the opportunity or inclination to play, while some shorter people did. To know more about the validity of the “Have you ever played basketball?” question as an indicator of height it would be necessary to conduct research: first, collect data across a random or representative sample on both people’s height and their answers to the played-basketball question; then, perform statistical analyses of these data.
A second type of assessment we use is what’s called situational testing, which actually started in German psychology back in the 1920s, in contrast to the kind of psychometric testing that English and American psychologists became more specialized in. In situational or operational testing, you put people in a situation to test a skill directly, which builds in the test’s validity. Rather than rely on self-report (asking people how good a marksman they are), you create a situation to test the ability you’re interested in (hand them a pistol and ask them to shoot at a target). The result, their measured accuracy, is a very straightforward assessment (highly valid) of their competence or skill at being able to hit an intended target. The closer the behavior, in context, that you are measuring is to the variable you wish to assess, the greater the validity of the situational test.
Part II: RfG Assessment of Individuals
The Context of Assessment in RfG therapy
There’s always some purpose involved in assessment– broadly, assessment is a two-way street. Like the therapist, the client also has a purpose in assessing what’s going on in therapy, particularly during its early stage: is this therapist trustworthy? Is it worth my while to continue with her/him? Do I feel safe enough at this point to disclose certain things? Looked at from a systemic viewpoint, assessment is an ongoing process in any interpersonal system. In life generally– even though we don’t label it as an assessment as such—all of us are continually forming opinions, taking impressions about the capabilities or intentions of other people. Whenever we have direct encounters with other people, or have an interest in someone, we observe and form judgments, such as, “how intelligent is this person?” “do they mean well?” “are they trying to sell us something?” or, “is it likely to be productive or valuable for me to continue to interact with this person?” Seeking answers to these questions is part of our Survival Mind functioning, though of interest to Adventure Mind we might ask instead, “is this person playful?” “am I having a good time being with them?”
Each psychotherapy approach employs its own distinctive constructs and methods for clinical assessment, which are used for diagnosis, prognosis and treatment planning. Assessment (“seeing what is”) is generally used to guide interventions (“changing what is”), although the relationship between assessment and intervention is reciprocal in practice. That is, the effect of an intervention confirms or modifies prior assessment data as much as assessment guides intervention.
In Rehearsals for Growth, enactments of improv games and exercises serve as the primary assessment tools, applied on a number of levels. In improv terms, the therapeutic relationship is tested by the client’s willingness in the present moment to accept our offer of an enactment, where we invite their self-disclosure beyond the making of a mere verbal response. For purposes of assessment, an enactment is a situational test even though the behavior elicited (the performance) occurs in the playspace (i.e., the mutual understanding that we are engaging in pretense, not behaving for/as real).
From observing improv enactments we acquire data serving two broad purposes: giving us information about what people are capable of doing; and, providing data concerning what choices they opt for during enactments. As an example, when I observe client Joe, as a player in a couple of RfG enactments, blocking his scene partners repeatedly I get some information about how competitive Joe is, how well he’s listening to others, how important he thinks it is to cooperate, and/or how accommodating to other people Joe is. Yet, as with all clinical assessment, I can’t generalize and conclusively come up with a reliable evaluation of Joe’s style or personality without repeated observations and repeated observations across different situations. The way persons play one scene does not in itself tell you how they will play all the time, nor do their improv performances necessarily correspond to their behavior in everyday life. To generalize you’d have to observe them over time and see to what extent they are displaying a pattern. The most you can be sure of is that they’re showing themselves as capable of blocking.
RfG enactments can serve as a special sort of situational testing; these enactments provide us with opportunities to observe how people behave in non-routine, unrehearsed situations. From them we don’t learn how Joe behaves in routine life situations but rather how he does in some contrived or non-ordinary, non-realistic ones where his life habits and routines do not dominate his choices.
In a study of patterns of improvising by individual clients in group therapy over a series of RfG E/Gs with different partners, it was found that there were distinct deficiencies in their improvising that corresponded to limitations in their psychosocial functioning (Wiener, 1999). A Dramaturgic Model was proposed that identified five role functions hypothesized as necessary both for adequate improvising and for competent psychosocial functioning in life.
Individual clients, observed in repeated enactments, display patterned or even predictable behaviors in their improvising with unaffiliated others (Wiener, 1994, p.154), often making repeated use of specific characters, mannerisms, or plot devices rather than behaving spontaneously. Status performances, especially when playing opposite to that client’s habitual status position, tend to be enacted characteristically; for specific emotions, the expressed intensity will be predictably limited in specific clients. Also, clients habitually approach improv situations with attitudinal sets (such as “not losing” in competitive scenarios or wanting to be seen as smart or ‘cool’) that bear directly on their relationship functioning outside of improvising.
Realize, however, that observed functional constraints in improv performances should not lead us to label these as fixed personality traits. For one thing, a client’s skill level of improvising curtails his/her full range of choices, though that range generally increases with practice. For another, the character interpretations that show up across repeated improvs are indicative of “trial sub-personalities” that offer the client/player an opportunity to practice role expansion (indeed, “rehearsing for growth”!). In clinical practice, RfG assessment of individuals consists initially of observing improv performance behavior as it bears upon broad and clinically significant issues (such as self-esteem, qualities of self-image, expectations from others, need for success, need for control, or need for the therapist’s approval). Whenever these clinical issues are viewed as constraints that the therapy targets for removal, the therapist follows exploratory enactments with selected or designed interventions that empower clients to overcome these assessed limitations through safe practice.
Part III- RfG Assessment of Dyadic Relationship Functioning
Most of what has been learned regarding RfG relationship assessment stems from clinical experience working with couples. When client couples in a significant relationship improvise together, patterns observed in their repeated improv interactions are clinically useful for assessing specific aspects of their relationship (information on which aspects are presented further below). In contrast to the considerable variability in improv performance features by individual clients playing opposite unaffiliated partners, couples display striking reliability in their performance dynamics with their own relationship partners. For example, A’s anger is regularly met by B’s fearful attempt to placate, leading to A’s magnanimous soothing of B, followed by B’s offer of humor or affection, etc. Most likely, the reason for such greater reliability is that how we perform with a relationship partner is strongly influenced by the context of the relationship outside of therapy. To speculate on the causes of this phenomenon, as a significant relationship deepens, the interaction between partners becomes stable and more predictable. One could also say that relationship development increases the “scripting” and decreases the spontaneity of responses that partners make with one another. That couples show dynamic variability at all in interacting within relationships is due chiefly to two factors: (1) our significant relationships are sufficiently complex to encompass a considerable variety of responses, triggered by variations in conditions and contexts; and (2) the play context, when evoked, which offsets the tendency to remain confined entirely within habitual patterns.
RfG clinical practice grew out of the insight that both good relationship functioning and good improvising shared three specific attitudinal and behavioral characteristics:
- Attending to others/Accepting offers (from others as well as from one’s imagination);
- Validating others (or, ‘making others look good’);
- Going on Adventures together (“Yes AND” rather than mere “Yes”).
From the beginning of RfG in 1984, thinking along these lines has guided clinicians to look for problematic patterns in a couples’ improvisational enactments that indicated problematic relationship functioning. This has led to assessing the dyad’s improvisational process via what are termed the Seven Signs of Good Improvising (Wiener, 1994, pp. xix-xx and pp. 154-156). An abbreviated version of these Signs is given in Table 1 below.
Table 1. Assessment Criteria for Seven Signs of Good Improvising
|Description of Good Improvising
(a) Clear Boundaries
Players are clear themselves and with each other regarding the distinction between player-as-person and player-as-character.
(b) Balanced contribution
|There is frequent contribution from each player, and a balanced degree of participation, equality of give-and-take. Players are observant and accept the offers of one another; they listen and don’t talk over one another.
(c) Character acceptance
|Players give and fully accept character, making others look good without imposing conditions for how they appear (e.g., smart, heroic, sexy, high-status, central to the scene, etc.). They put developing the scene ahead of showing off or hiding out.
(d) Wide expressive range
|Players “physicalize” in a grounded way corresponding to the story; they fully use their expressive range, according to the spirit of the situation.
(e) Strong character
|Players stay in the present moment when they don’t know what is happening or when their imagination is blank. They act un-self-consciously and non-defensively, except when they playfully incorporate such actions into the scene.
(f) Positive outcome
|Players are often surprised and pleased by the results of the scene; they enjoy having co-created and shared an adventure and like each other at the end of the enactment. They accept and learn from what occurred and quickly let go of judgments, both of self and their partner.
(g) Spontaneous idea development
|Players are not pre-planning but are making it up as they go; they remember where they have been and reincorporate previously-used story elements.
The one Sign of greatest predictive utility regarding good relationship functioning is (f) Positive Outcome—tested by asking, post-enactment, “did (each of) you experience your enactment together as enjoyable?”
A more standardized version of Couples assessment using these Seven Signs can be accomplished by employing a battery of the following 4 RfG dyadic enactments (described in Wiener, 1994): the two Exercises, “One-word-at-a-time Story (p. 65),” and “Tug-of-War” (121); and the two Games, “Poet’s Corner” (p. 84) and “Status Transfer” (p. 116). Each of these four enactments has the advantages of: Frequent prior clinical use, establishing a broad range of possible outcomes; relatively simple instructions used; and, modest prior interpersonal/improv skills required of the couple.
Four Additional Domains of Couple System Assessment
In addition to assessing their improv performances as described above, RfG therapists also attend to four descriptive lenses of the couples’ functioning outside of performance:
- fit between the operating styles of partners—Are these symmetrical (similar) or complementary (opposite)? Functional fit may be examined in such areas as: work (result-oriented vs. process-driven); thought process (metaphorical or literal); and emotional expressiveness (intense or subdued).
- style of narrative conflict—parallel narratives, in which partners mutually invalidate some of the facts asserted by the other in their accounts of events; or, intersecting narratives, in which partners agree on the facts but differ in their interpretations or significance of these facts.
- balance of functioning—rough equivalence in contribution to their partnership across all areas vs. extreme over/under-functioning.
- in-session emotional range—the range of emotions exhibited in the couple’s interchanges with one another, compared with the individual ranges of expression shown when each partner interacts with the therapist.
These domains are ordinarily assessed from observation and information verbally gathered outside of enactments. To shift dysfunctional patterns the RfG therapist may construct and offer Proxy Scenes (Wiener, 2016) to explore experimenting with altering dysfunctional patterns.
Wiener, D. J. (2016). Removing personal constraints via proxy scene enactment. Drama Therapy Review. 2, (2), 183-193.
Wiener, D. J. (1999). Using theater improvisation to assess interpersonal functioning. International Journal of Action Methods, 52, (2), 51-69.
Wiener, D. J. (1994). Rehearsals for Growth: Theater Improvisation for Psychotherapists. New York: W. W. Norton.