Developmental Action Inquiry: A Distinct Integral Theory That Actually Integrates Developmental Theory, Practice, and Research

In this chapter, we offer an introduction to our work over the past 40 years of developing the distinctive integral theory, practice, and types of research known as Developmental Action Inquiry (DAI), including the most recent empirical and clinical
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
    17 1 Developmental Action Inquiry: A 2 Distinct Integral Theory That 3 Actually Integrates Developmental 4 Theory, Practice, and Research 5 William R. Torbert, Reut Livne-Tarandach, Aliki Nicolaides,David 6 McCallum, and Elaine Herdman-Barker 7 In this chapter, we offer an introduction to our work over the past 40 years of developing the distinctive 8 integral theory, practice, and types of research known as Developmental Action Inquiry (DAI), including 9 the most recent empirical and clinical findings using the psychometric instrument called  Harthill 10  Leadership Development Profile (LDP). 11 We will first describe and demonstrate three distinctive features of DAI: (1) its basis in an ontology and 12 epistemology that distinguish among four possible territories of experience; (2) its injunction to interweave 13 first-, second-, and third-person research in the midst of one’s practice with others; and (3) its recognition 14 of three different types of research findings (single-, double-, and triple-loop feedback) and their impact on 15 subsequent practices. We will follow this introductory section by describing the LDP, a third-person 16  psychometric, and the ongoing validity-testing research we are performing on it in the fields of our 17  practices. Then we will share some of the more provocative findings from recent research. In this section 18 we focus on the potentials and limitations we observe in the Expert and Strategist action logics. We also 19 introduce a few of the findings from two recently completed doctoral dissertations. (In the context of our 20 “in action” presentation at the conference itself, we interrupted the presentation at several points in order to 21 invite the audience into some first- and second-person research exercises together.) 22 Theoretically Distinctive Aspects of DAI: The Deep Four 23 In one of his early books,  Learning from Experience: Toward Consciousness   (1972), Bill Torbert critiqued 24 the positivist, objectivist theory-data model of modernist science and instead named four distinctive 25 territories of experience that humans seek to coordinate and align both in theory and in practice: 26 1. the outer world   (as seen and otherwise sensed by a person or measured by an instrument [this 27 includes others’ actions seen from the outside—altogether, what modern science calls “the 28 territory” 29 2. the self-sensing of one’s own embodiment  , breathing, moving, perceiving, etc.; 30 3. one’s own ongoing structures of thinking and feeling   (which in dialogue with others in a 31 scientific community of practice generate what modern science calls “the map”); and 32 4. the intentional attention  (which can be distinguished from the other three territories, can 33 experience all three simultaneously, and can be voluntarily cultivated in adulthood, but rarely 34 has been in our culture, up until the recent growth of adult development theories and practices 35 [Kegan 1994; Ouspensky 1949; Torbert 1976, 1987; Trungpa 1970; Wilber 2000). 36 In the  Power of Balance  (1991), Torbert described how anyone can confirm for him or herself the 37 reality of each of these four territories through thought and attention experiments, somewhat like Descartes’ 38 doubting procedure to establish the indubitable fact that we think. To highlight this difference between his 39 “theorizing about trans-theoretical experience” and Ken Wilber’s (2000) effort to create a “comprehensive 40    Alan 3/31/10 10:48 AM Deleted: ,    intellectual map of all experience,” Torbert sometimes calls these four, mutually orthogonal “territories of 1 experience” the  Deep Four   and Wilber’s four-quadrant AQAL model the  Flat Four   (since Wilber’s four 2 quadrants in our experience are often held by  proponents of Integral Theory as cognitive categories that 3 tend to keep our attention fixated within the single “thinking” territory (as most of you, our readers, likely 4 have been, as you have been reading this). The words for the Deep Four territories of experience are 5 obviously also cognitive categories, but, as the “deep” four, these terms invite us, not just to “think” the 6 categories, but also to experience—now, in each present moment—the pre- and postconceptual realities to 7 which they refer (e.g., the color and texture of the “outside world,” the “inner sensing” of our own 8  breathing, moving, and feeling, as well as the kind of “attending” that can “taste” this external text, one’s 9 own breathing, one’s thinking (about these words now), and one’s inquiring into the very source of 10 attention all at once, simultaneously, e.g., now) (see Table 17.1). 11   [ Insert Table 17.1 ] 12 Described in these ways, the Deep Four territories of experience constitute what we think of as the 13 first-person field of action inquiry, bounded by the current limits and usually unexamined assumptions of 14 the self’s capacity for perception and sense-making (i.e., the specific developmental action-logic through 15 which one currently encounters the world, as will be described below. Given these limits, a more full 16 apprehension, understanding, and engagement of reality requires two additional aspects of all human 17 lives—the second- and third-person fields for first-person work and play. 18 First-, Second-, and Third-Person Fields of Inquiry 19 The process of inquiry or research in action is a means of attending to experience of the world, including 20 our dynamic sense-making of that experience, as well as our adaptive engagement. By adaptive 21 engagement, we mean those times when we (the person, team, or organization) are intentionally attempting 22 to assess whether our intentions, plans, actions, and outcomes are aligned with one another and with 23 changing conditions…and are responding accordingly. For some, this process is very rare, for others quite 24 regular. An important question is whether and how one tests the validity of the information one is 25  processing in real time. DAI is one of a broader family of recent efforts to begin describing and defining 26 this world of methods for valid inquiry in the midst of action (two others are action science [Argyris et al. 27 1985]) and Senge’s [1990] five disciplines). 28 The theory, practice, and research methods associated with Developmental Action Inquiry (DAI) all 29  point toward the capacity for individuals, teams or communities of practice, and larger organizations and 30 institutions to conduct interweaving first-person, second-person, and third-person research (studying 31 “myself,” studying “ourselves,” and studying “them”) in the midst of their daily practice  (Chandler and 32 Torbert 2003, Torbert 2000). In contrast to modern science—which offers a model of third-person research 33 “on” “subjects” (who are, ironically treated as objects)—DAI is a model of research on oneself and with 34 others that requires a high voluntary commitment by participants, as well as increasing mutuality and 35 collaboration among them (McGuire, Palus, and Torbert 2007). 36 A third distinctive feature of DAI is that progress occurs, not just by incremental single-loop hypothesis 37 or market testing, but also by double- and triple-loop learning and change, whereby the very assumptions of 38 one’s organization’s, team’s, or personal action-logic transforms (double-loop learning) or aligns across all 39 four territories for a moment (triple-loop learning). When, during our personal, relational, and collective 40 actions and inquiries, incongruities are found across the four territories of experience (e.g., an unintended 41 result, an ineffective performance, a strategy that feels inconsistent with one’s integrity, a lie, etc.), action 42 inquiry gradually generates, in the first-, second-, and third-person human system, the capacity for these 43 three distinct orders of change. 44 The fourth distinctive feature of DAI (the D in DAI) is the developmental theory used to map the 45 evolution of the action logics through progressive forms of increasing complexity, differentiation, and 46 integrity at the personal, team, organizational, and institutional scales (Torbert 2000a, 2000b). This theory 47 hypothesizes a specific sequence of action-logics (Opportunist, Diplomat, Expert, Achiever, Individualist, 48 Strategist, Alchemist, Ironist, Elder are the names of the personal action-logics) through which any human 49 system can (but may not) transform, as it gradually gains the capacity to monitor all four territories of its 50 activity and to develop greater congruity, integrity, and mutuality among them. Single-loop, incremental 51 learning is practiced increasingly regularly as a person masters the Achiever action-logic. Double-loop, 52 transformational learning is first explicitly recognized at the Individualist action-logic and becomes a 53  Alan 3/31/10 10:56 AM Deleted: -  Alan 3/31/10 10:56 AM Deleted:  promonents  Alan 3/31/10 10:57 AM Deleted: -  Alan 3/31/10 10:59 AM Deleted: -    touchstone of the Strategist action-logic. Triple-loop inquiry in-moments-of-action is increasingly practiced 1 in everyday life at the Alchemist and Ironist action-logics (Torbert et al. 2004). 2 Overall, DAI theory and method strike a different balance from Wilber’s Integral Theory and its AQAL 3 model. DAI theory and method put primary emphasis on the four attentional and experiential territories of 4 experience we can engage at each moment, rather than on four conceptual fields. DAI puts secondary 5 emphasis on interweaving first-, second-, and third-person and practice, rather than bifurcating attention 6  between the individual and the collective. And DAI puts tertiary emphasis on creating communities of 7 living inquiry with single-, double-, and triple-loop learning at their heart, rather than communities of 8 shared belief. 9 In attempting to assess empirically who acts regularly at which action-logic and how persons at a given 10 action-logic fare in the rough and tumble world of everyday life, let us look next at the historical evolution 11 from the Loevinger SCT instrument to the Harthill LDP and then at some of the field research using the 12 Harthill LDP. 13 The Evolution from Loevinger’s SCT to the LDP 14 Loevinger’s Sentence Completion Test (SCT) is a language-based instrument, which delineates 15  preconventional, conventional, and postconventional stages of self-development srcinally designed to 16 address women’s issues. An example of this gender focus is evident in a handful of items included in the 17 srcinal test such as “The worst thing about being a woman…,” “A good mother,” and “For a woman a 18 career is…” (Loevinger, Wessler, and Redmore 1970). This test includes 36 open-ended sentences that aim 19 to measure a broad range of content: “moral development, interpersonal relations and conceptual 20 complexity” (Loevinger 1998, 3) and correlate with eight stages of ego development (Loevinger and 21 Wessler 1970, 186–87). Loevinger called her latest stage of ego development “Integrated self-actualizing 22 identity” and suggested, quite peculiarly, that “because this stage is rare in most samples and there are 23 major differences among qualified raters both as to the description of this level and application of the 24 description in specific cases, under most circumstances it is best combined with the (prior) Autonomous 25 stage” (7). She did not imagine, theorize, or research the possibility that integrity may be the fruit of 26 developing a postcognitive, observing, listening attention that registers ongoing transformation across all 27 four territories of experience and all action-logics. Thus, for example, she offers neither theory nor method 28 for scoring this developmental stage. 29 Working in tandem for many years, Torbert (1987) and Cook-Greuter (1999) set out to modify the SCT 30 in several ways. First, they sought to increase its face validity and usefulness as a work-related, leadership 31 development instrument by creating and validating work-related stems, starting from Molloy (1978) “A 32 good boss…” Harthill has continued and extended this work, so that the Harthill LDP now has nine 33 work-related stems, including stems about power, time management, and teams, that replace prior stems 34 emphasizing gender or that had the lowest correlations to overall protocol ratings. This is a meaningful and 35 unique contribution because it expands the scope of generalizability of the Harthill LDP test to better map 36 management-oriented topics that are at the heart of developmental action-logic implementations in 37 organizational contexts. 38 Another unique feature of the Harthill LDP is that the LDP has now twice refocused the Loevinger 39 definitions and scoring manuals for the later action-logics. Cook-Greuter (1999) undertook the first 40 revision. Herdman-Barker and Torbert have undertaken a second revision, aligning the scoring criteria 41 more closely with the Alchemist and Ironist constructs found in Torbert’s (1987, 1991, 2004) work and 42 increasing the degree of difficulty of rating an overall profile as showing regular mastery of the Alchemist 43 action-logic. One of the new criteria is whether the sentence completion “treats 44 attention/conscience/consciousness as a process distinct from thinking and acting.” For example in a 45 response to the “I am” sentence stem we see nuances of this criterion: 46 “I am, therefore I think”—Descartes got it the wrong way round. Our thoughts and 47 emotions are an inevitable aspect of our being—delightful, painful, exciting, infuriating—  48 but can hide our inner depths from us. 49 In the above example we are illustrating how one person differentiates the thinking territory from the more 50 inclusive attention/conscience/consciousness territory. 51  Alan 3/31/10 11:07 AM Deleted: s  Alan 3/31/10 11:09 AM Deleted: -  Alan 3/31/10 11:09 AM Deleted: -    Alan 3/31/10 11:11 AM Deleted: thirty-six  Alan 3/31/10 11:13 AM Deleted: -  Alan 3/31/10 11:15 AM Deleted: action    Another new criterion of an Alchemist sentence completion is whether it conveys “passionate, artistic 1 self-expression, not hyper-rational.” For example, in the following example of a response to the “I am” 2 sentence completion stem we illustrate what we mean by the passionate and not hyper-rational. 3 I am… a riot of differing roles and impulses held together in a loose alliance by 4  something I call me, I am mostly happy and amazed, by any rational analysis my 5 existence is such a staggering improbability that delighted laughter is the only possible 6 response. 7 In other words, our scoring criteria for the Alchemist are emerging from a theoretical perspective that 8 treats it as critical to distinguish simultaneous experiential contact with the four territories of experience 9 from sheer cognitive complexity and clever, fashionable, postmodern wordsmithing. We are looking for a 10 weave between cognitive and relational strands and for a unique and glimmering oddness that shines 11 through Alchemists we meet. In addition, based on our coaching, consulting, and workshop contacts with 12  people scored at postconventional action-logics (to be described below), we have raised the number of 13 Alchemist stems and the number of categories those stems must fall into for a total protocol to be scored 14 early Alchemist or full Alchemist. All this, we believe, increases the construct validity of the Alchemist 15 designation when it is used to summarize the center of gravity action-logic represented by a person’s LDP. 16 Over the past twenty years of ongoing research and validity testing of the LDP in field conditions (e.g., 17 executive coaching and organizational change projects), Harthill has shifted the focus from using the 18 measure only for third-person adult development assessment to using it in the context of first- and 19 second-person adult development and transformation as well. In doing so, Harthill reinvented the key 20 names and descriptions of the Harthill LDP to become more descriptive and actionable and less evaluative 21 and abstract than those associated with the Loevinger SCT. Thus, we reconstructed such terms as “from 22 lower to higher stages” to “from earlier to later action-logics.” We changed particular “stage” names from 23 Conformist to Diplomat, from Self-aware to Expert, etc. Likewise, we fully rewrote and continue to amend 24 the 20-plus pages of action-logic-related feedback that anyone taking the Harthill LDP today receives as 25  part of the feedback package (along with introducing a 2/300-word commentary written for that particular 26  protocol). All these changes make it feasible and effectual to use the instrument and to offer feedback on 27  people’s performance on it in action research situations. These changes seem to us an improvement in 28 accuracy and objectivity. They also make it feasible to for us to give individual participants and the 29 institutions involved in the action research useful feedback. This feedback at times generates or supports 30 double-loop, developmentally transformational learning, in addition to testing the external validity of the 31 LDP. We will discuss this in greater detail below. To further establish the value inherent in using Harthill 32 LDP for research and adult development tasks, we explore the reliability and validity of the LDP below.   33 Reliability Assessment of the LDP 34 To ensure the validity testing of this instrument, Harthill has invested in reliability assessment, which is an 35 important first step in testing the psychometric value of the LDP, a prerequisite for measurement validity 36 (Schutt 2004). To assess the reliability of the LDP, we conducted three reliability tests. First, we tested the 37 extent to which the set of 36 stems can be aggregated to one score reflecting profile action logic. To do so 38 we tested the internal consistency  by calculating the Cronbach’s alpha score for the 36 stems. Cronbach’s 39 alpha values range from zero to one, one indicates perfect internal consistency and zero indicates no 40 internal consistency. In general Cronbach’s alpha values higher than .8 are considered satisfactory as 41 indicators of internal consistency (Schutt 2004). Our analysis of the internal consistency of Harthill LDP 42 was built on 891 profiles and generated a Cronbach’s alpha of .906, a relatively high value indicating good 43 internal consistency that justifies the aggregation of stems into one score reflecting a single action logic. 44 Another dimension of reliability testing concerns the assessment of inter-rater reliability. The most 45 recent reliability test between the two authorized Harthill scorers covers 805 profiles, with 13 levels of 46 scoring (from Expert through Alchemist, with three  possible levels at each action-logic [e.g., Early 47 Achiever, Achiever, Late Achiever]). The test shows perfect matches for 72 percent of the profiles and 48 agreement within one part-stage in another 22 percent. Hence, there is either perfect agreement or 49 agreement within one  part-stage in 94 percent of the cases. There was a difference of more than one 50 action-logic between the two scorers in only one case of 805. Moreover, it is important to note that any 51 difference in the two scores is adjudicated prior to sending respondents their feedback packages. 52    Alan 3/31/10 11:36 AM Deleted: -  Alan 3/31/10 12:25 PM Deleted: 3  Alan 3/31/10 12:26 PM Deleted: %  Alan 3/31/10 12:26 PM Deleted: %.  Alan 3/31/10 12:26 PM Deleted:  part  Alan 3/31/10 12:26 PM Deleted: %    With regard to the question of whether the six novel business-related stems introduced to this version 1 yield consistent scores, there has not yet been a definitive reliability test between scorers because scoring 2 manuals for each stem are still being constructed. In the interim, however, we can examine whether the 3 new sentence stems are adding to, or subtracting from, the overall reliability of the profile scores. Of 4 course, one of the criteria for eliminating six prior sentence stems was that their scores showed among the 5 lowest correlations with profile scores. The important finding here is that the six new stems show higher 6 correlations with the final protocol ratings than the stems they replaced and, indeed, higher than the average 7 correlations among all the remaining stems. Hence, short of a direct reliability test between the two scorers, 8 the new stems seem to be adding to, not subtracting from, the reliability of total profile scores. 9 Validity Assessments   of the LDP   10 What follows is a brief discussion of the value of the LDP based on several research studies with 11  participants who were profiled using the LDP and were engaged in the practices of DAI. To further 12 establish the psychometric value of the LDP tool we present key findings concerning construct validity , 13 criterion validity , and external validity  that emerged from the data. 14 C ONSTRUCT V ALIDITY B ASED ON F ACTOR A NALYSIS   15 Recent efforts to establish measurement validity included a construct validity study of 891 profiles. This 16 study compared a factor analysis among all 36 items in preconventional and conventional protocols 17 (Achiever and earlier; n=830) with a factor analysis of all 36 items in protocols scored as postconventional 18 (Individualist or later; n=61). Not only are the factors in the two sets different from one another, but the 19 structure of the factors is different as well. For the earlier action-logics, stems load on eight factors, mostly 20 each stem on a single factor (only two stems load on two factors). For the postconventional action-logics 21 (Individualist and later), stems loaded on 11 factors, but loadings were not confined to one factor per stem. 22 More than half (52%) of the stems loaded on two factors of more (nine stems loaded on two factors, seven 23 loaded on three factors, and three loaded on four factors). These results illustrate the fundamental 24 difference between the conventional and postconventional action-logics and also echo the adult 25 developmental theoretical foundation on which the Harthill LDP is built. The limited but focused loadings 26  presented by the conventional sample represent a relatively simple mental map, with Aristotelian-ly 27 distinct, independent categories (“nothing can be both A and not-A”), as one would theoretically expect of 28 action-logics up through the conventional. In contrast, fewer factors along with complex sets of loadings 29 suggest a systems-oriented, inter-independent mental map. Such a movement between personal 30 “independence” and “inter-independence” (McGuire, Palus, and Torbert 2007) occurring in the movement 31  between the conventional and postconventional action-logics is predicted by all the 32 currently-in-ongoing-revision developmental theories (Kegan 1982, 1994; Torbert 1999, 2004; Wilber 33 2000, 2006). 34 Criterion and External Validity of the LDP 35 As readers familiar with validity research on the Loevinger SCT know, there is an extensive body of 36 internal validity research about it (Westenberg et al. 1998), but there is still very little research that shows: 37 (1) whether people measured at different action-logics act differently in their everyday settings according to 38 criterion variables (such as whether they seek feedback, or whether they successfully lead organizational 39 transformation), and how generalizable that result is; or (2) what contexts reliably generate transformation 40 of a person’s action-logic. Our action research orientation integrates our own first-person efforts to use the 41 theory to make our own leadership more timely, effective, and transformational and in our second-person 42 teaching, coaching, and consulting efforts to help others become more effective. For these reasons most of 43 our validity assessment efforts have been dedicated to generating and measuring the criterion validity and 44 external validity of the LDP measure. What follows is a brief illustration of how we approach this criterion 45 validity and external validity testing in the action of our research. 46 Offering feedback and coaching to people who take the measure has permitted one set of criterion 47 validity tests. Developmental theory suggests that people at earlier action-logics are more likely to avoid 48 feedback, especially of a double-loop nature that questions their current action-logic, whereas people at 49 later action-logics will increasingly seek out such feedback and associated transformational opportunities. 50    Alan 3/31/10 12:27 PM Deleted: 6  Alan 3/31/10 12:29 PM Deleted: -  Alan 3/31/10 12:29 PM Deleted: -  Alan 3/31/10 12:30 PM Deleted: 2  Alan 3/31/10 12:30 PM Deleted: 9  Alan 3/31/10 12:30 PM Deleted: 2  Alan 3/31/10 12:30 PM Deleted: 7  Alan 3/31/10 12:30 PM Deleted: 3  Alan 3/31/10 12:30 PM Deleted: 3  Alan 3/31/10 12:31 PM Deleted: 4  Alan 3/31/10 12:32 PM Deleted: -
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks