...... _ ......

The Behavioral BEHAVIORAL Measurement MEASUREMENTS Database Letter Services ...... , Vol. 4, No.2 Enriching. . . the. health... and. .... behavioral. . sciences. by broadening instrument access ...... Spring.. .. 1997. . About This Newsletter ... behavioral scientists, and students. From widely Transitionally Speaking . recognized to obscure and unpublished, these instruments include questionnaires, interview Robert Perloff, Guest Newsletter Editor schedules, coding schemes, observation checklists, rating scales, tests, project techniques, Introduction and measures using vignettes or scenarios.

After I agreed to offer a segue or transition into Information Brought to Light: Most instruments the three innovative articles constituting the are "buried" in avalanches of published literature substance of and written especially for this issue and are hence difficult to discover. Worse still, of The Behavioral Measurements Letter (Volume scientists in one field (e.g., ) may be 1:, No.2, Spring 1997), a question leaped up at unfamiliar with instruments in other fields (e.g., me: what exactly did this issue segue from? How medicine, nursing, public health). The majority is this issue of the Behavioral Measurement of users-do not have access to instruments that Database Services (BMDS) semiannual newsletter either have been recently developed or are different from the issues preceding it? Raising this described in unpublished manuscripts. These question is no idle distraction from the business measures are generally known only by people in at hand, for this question mandates, necessarily a particular field or subspecialty. By maintaining and desirably, that the newsletter and its parent, information on instruments from these diverse . BMDS, be identified and explained for new sources, HaPI enables users to retrieve relevant readers as well as for our friends and colleagues measures about which they might otherwise be who have been with us since the newsletter's unaware. Thus, HaPI helps researchers avoid inception some four years ago. "reinventing the wheel." HaPI places existing information on measurement instruments at users' The principal offspring of BMDS is HaPI [the fingertips, no farther away than their keyboard. Health and Psychosocial Instruments (hence What and Why is the Behavioral Measurements HaPI, get it?) database]. HaPI's function, expressed in the sidebar to the newsletter title, Letter? above, is "Enriching the health and behavioral sciences by broadening instrument access," a The Behavioral Measurements Letter, a semi- double-duty enrichment serving as an icon both annual newsletter, is devoted to the exploration for HaPI and for the newsletter. of timely measurement topics. The Behavioral ' Measurements Letter is published by Behavioral HaPI, available online through OVID Measurement Database Services, producer of the Technologies (an international vendor of Health and Psychosocial Instruments (HaPI) databases) and now obtainable as a CD-ROM database. from BMDS, contains over 45,000 records of interest to psychologists, physicians, nurses, The impetus for this newsletter sprang from social workers, educators, evaluators, BMDS's belief in the paramount importance of sociologists, administrators, other health and Vol. 4, No.2, Spring 1997 The Behavioral Measurements Letter Introduction (Continued) explore that trait? In Fiske's own words from his article in this newsletter, "Are there one or more measurement. Just as in the physical sciences, methods that yield distorted patterns of advances in the health and behavioral sciences are correlations or correlational values that are much proportional to advances in measurement. As too low? Are there [rather] methods that generate Robert Pool stated in the case of the physical too high correlations, correlations that cannot be sciences, "These advances are vital, because accepted for the constructs as currently science's understanding of the physical world is construed?' , necessarily limited by the accuracy with which science can measure that world": (Science, 1988, The first "something new" is "Measuring 240, 604-605). Perceptions of Relational Communication," by Wisconsin communication professor Michael Earlier Editions of the Newsletter (Where we are Pfau. Like the article by Fiske, Pfau's i~ Seguing From) significant and instructive because, in identifying a new dimension in communication-' 'relational The first three volumes of the newsletter and No. messages, " he is suggesting that communication 1 of Volume 4, while not bereft of substantive behavior is well served by looking at implicit articles ("Finding the Right Measure," "Clinical (nonverbal) as well as explicit communication. Measurement for Primary Care," "Faith in As a matter of fact, I was instantly struck by the Measurement," "Ways to Measure Demographic importance and wide spread applicability of Variables," "In Memoriam-Donald T. "relational messages". Campbell," "Measuring Reminiscence in Research on Type A Behavior," "Beck In a separate communication, Pfau says that in Depression Inventory") were quite appropriately his article he "focused on the relational devoted to housekeeping items and articles communication measure, as opposed to a establishing the impetus for the newsletter in the combination of communication measures, because first place ["HaPlnings," "HaPI Thoughts" (the it has unique potential in the health lighter side), "Instrument Update," and related communication context to tap both verbal and developments pinpointing changes in and new nonverbal .components of person perception, and features of HaPI]. it is relatively unknown outside of the communication discipline." Seguing Into ... the Transition So here we have a powerful step up from what HaPI was a piddling three years ago and a step When the newsletter was inaugurated there were 18,000 records; this has increased 150%; in less up, also, in the context of the BMDS newsletter. than four years, to 45,000 records. First, a highly And this brings us to the second "something popular feature with users has now been new," a new feature of the newsletter: established as part of HaPI's repertoire: document illustrations provided in a new column, by Fred delivery of instruments. Next, probing in this Bryant, starting with his maiden column in this issue, beneath measurement's first layer of issue: "The Comparative Anatomy of Related information, are "something old" (Fiske's article) Instruments: an Emerging Specialty." Bryant's and "something new" (the articles by Pfau and column brings the newsletter to a higher level of by Bryant). The "old" is a review by Donald W. measurement breakthroughs and to a window of Fiske of the celebrated classic of his and Donald erudition providing more insight into measures Campbell's on the multitrait-multimethod (MT- and the concepts they seek to elucidate, where MM) matrix, a measurement breakthrough whose there is revealed "alternative measures of the significance is attested to by its thousands of same construct to determine conceptual overlap citations. The question these "hall of fame" and uniqueness." This brings added value to the measurement psychologists sought to ask is interpretation of measures as well as to decisions whether the determination of a psychological about what measures to use in a particular study. phenomenon or trait or behavior is due In this issue of the newsletter, Bryant's first fundamentally to that trait explicitly or is the column illustrates this procedure, using the finding an artifact, rather, of the method used to construct of optimism.

The Behavioral Measurements Letter 2 Vol. 4, No.2, Spring 1997 Introduction (Continued) The Method in MT-MM

Bryant seeks to determine whether and how Donald W. Fiske alternative measures of the same concept leave us with a level of confidence that the concept is The multitrait-multimethod (MT-MM) matrix is robust over different measures, in the same way now part of the common methodology for that Campbell and Fiske, a half century earlier, research psychologists studying individual sought to determine whether the results of differences. The inspiration of the late Donald T. research involving measured entities are Campbell,· it will be included in the training of attributable to the concepts underlying the psychologists for years to come. The basic idea measure or, instead, attributions to how the is straightforward: a researcher should design his concepts are measured-which, of course, is what studies so that he has measures of his essential multimethod and multi trait are all about. variables by more than one method. In other words, his research should involve the Thus, a more complex and sophisticated aspect correlations among several traits, each measured of measurement is represented in the articles by by the same set of methods. As the necessity of Fiske, Pfau, and Bryant: an understanding not using several methods becomes more generally - only of specific measures themselves-in an accepted, the field of individual differences should absolute and insulated way-but rather of where gain. There is some question, however, whether the measure stands with regard to other factors the published matrices are improving over the (the method used to measure concepts in the case dismaying set included in the original article of Fiske and of Bryant, and in the case of Pfau (Campbell & Fiske, 1959, Psychological Bulletin, a treatise on "relational communications," which 56, 81-105). can be communicated verbally as well as, and especially, nonverbally, along side of other verbal That paper has had a record-breaking citation rate contents. We are confident that you will enjoy and is still receiving more than a hundred citations these three articles-and feel that their perusal was a year. The wide range of journals in which it worthwhile. is being cited suggests that its message is spreading widely. Some of the citations are So, retreating full circle, this is what The probably in papers by methodologists who Behavioral Measurements Letter was and what it continue to work on the problem of statistically has evolved into. We hope that you will give a ~ analyzing the MT-MM matrix. "thumbs up" to the newsletter's modest contribution to the measurement enterprise and Apart from any such analyses, much can be will yourself be propelled to offer articles for learned about the constructs measured and the enriching The Behavioral Measurements Letter. methods used by a close examination of each matrix one obtains. Answers can be sought for Robert Perloff, PhD, is Distinguished a number of important questions. For each Service Professor Emeritus of Business construct (trait or other type of variable), is it Administration and of Psychology at the being measured differently by the several University of . He has been methods? Is the level of association compatible president of many national professional with the investigator's conceptualization of that and scientific societies including the variable or does. it indicate appreciable method American Psychological Association, the variance (unwanted variance due solely to the American Evaluation Society, and the particular methods)? Are there one or more Society of Psychologists in Management. methods that yield distorted patterns of Earlier in his career, Dr. Perloff was correlations or correlational values that are much Director of Research & Development at too low? Are there methods that generate too high Science Research Associates. He has just correlations, correlations that cannot be accepted concluded a 3-year term as a member of for the constructs as currently construed? the American Psychological Association's Board of Scientific Affairs. Of course, one's research field may not enable one to set up a design with several traits. or

Vol. 4, No; 2, Spring 1997 3 The Behavioral Measurements Letter

I The Method in MT-MM (Continued) it? The most convenient, most readily available context, such as a laboratory bench or a constructs, each being measured by several psychology classroom, may not be a wise choice. apparently highly diverse methods. Even so, one Mating behavior in the laboratory cage is quite can measure each central construct in the research different from the behavior observed in a plan by at least two methods. Correlations simulation of the natural environment in which between two methods of measuring can be rats mate (McClintock, 1981, New directions for illuminating. Correlations between two constructs methodology in social and behavioral science, San measured by just one method seem useless. If any Francisco: Jossey-Bass). of the constructs can be measured only by just one method, beware! The construct may be Our problem is not specific to psychology or the inextricably embedded in the method. social sciences more generally. Recently I learned that studies of the Golgi material in cells yield The MT-MM matrix can help us discard a method somewhat different results, depending upon the or a construct that is not useful. Weare quite discipline to which the researcher adheres, as a willing to discard a method when we find flaws consequence of training and practice with a given in it, when it generates weak:and unreliable data, set of tools. The two sets of results are simply when method variance overwhelms construct disparate, not contradicting each other but at the variance. But we sail along happily with our same time not mutually supportive of each other. constructs, until an MT -MM matrix shows that they are too specific to the method used to In other instances in the natural sciences, the measure them. Or we may find that our pet or situation is quite clear and clean: e.g., in the fledgling construct overlaps highly with another measurement of temperature, there are a score construct, for each of several methods. An of different methods, there are conceptual linkages interesting case study is "social ." In among them, and they agree with each other very the twenties, an effort was made to measure this well (especially in terms of our standards). So construct, which seemed like a reasonable one. when temperature is being measured, our problem But efforts to measure it yielded scores that were does not get in the way. so highly correlated with general intelligence that it could not be considered a separate variable. An examination of the measurement of [Editor's Note: One cannot help but wonder temperature may throw some light on our whether a similar fate awaits Daniel Goleman's problem. Temperature has a variety of effects and (1995) articulation of "emotional intelligence" there is some theory about each effect. So the (New York: Bantam Books). physicist can apply a piece of general physical theory to provide a conceptual basis for each The major focus of the "Convergent and method of measuring it. Most of the physical Discriminant Validation" paper and of this note theory relevant to temperature seems to be pretty is on method. We can reformulate constructs as well agreed upon by physicists. Unfortunately for our results require, but what can we do about our us, we do not have such an agreed upon general methods? Sometimes we can improve them, but theory in psychology, or at least not enough to more usually a bad method has to (and therefore provide a basis for a theory of measuring, a theory should) be discarded. Psychometricians are of method. (Psychometric test theory does not familiar with the several sources of possibly throw light on the underlying problems.) So we intrusive variance in ordinary testing, such as the need to develop conceptual formulations about items, the instructions, the examiner, and the measuring that we can test empirically to see reason for the testing session. All of these are which ones need to be modified. Until such within the testing room, the standard context well theorizing has been worked out and tested, we known to psychometricians. What about other can use the MT -MM matrix as a prop. If we contexts or situations? For each context, one can believe that a substantive construct should be prepare a list of components or features of that constant over a set of methods for measuring it, context, each of which could potentially affect the we can apply those methods to that construct and measurements obtained in the given context. For see what happens. If the methods agree, that's a construct that you have studied, what are the great. If they don't, then either the methods or one or more contexts in which you must study

The Behavioral Measurements Letter 4 Vol. 4, No.2, Spring 1997 The Method in MT-MM (Continued) Measuring Perceptions of Relational Communication the conceptualization of the construct must be changed. Michael Pfau

Perhaps the solution to this problem is to label Relational communication is an intrinsic element each set of measurements not only by the in person perception. Judy Burgoon and Jerold substantive, conceptual core but also by the Hale, who did the pioneering work developing method by which they were obtained. We already and testing the relational communication scale, do this for self-ratings as opposed to peer-ratings describe relational communication as signifying and for intelligence as measured by the Wechsler "how two or more people regard each other, or as measured by the Stanford-Binet. regard their relationship, or regard themselves within the context of the relationship" (1984, The pervasiveness of method has been known for Communication Monographs, 11" 193-214). centuries. Shakespeare has Polonius say of Hamlet's strange behavior, "Though this be I called relational communication intrinsic because madness, yet there is method in't." it occurs in all instances in which a communication modality makes possible either Officially retired in 1986, Fiske is now real contact, as in interpersonal communication, almost completely retired from the or perceived contact, as in mediated frenetic world of actual research and communication. People form relationships with scholarship. At the University of Chicago others whom they encounter directly, and with from 1948, the time of his doctorate at those they experience indirectly (e.g., people they the University of Michigan, he worked on come to know via television). Horton and Wohl many problems, not solving any of them characterize the latter as "parasocial but contributing to our understanding of relationships" (1956, Psychiatry, 19,215-229), the problem and its consequent but to those who experience them, they are every reformulation: intra individual variability bit as real as direct encounters. over time; our overreliance on words as stimuli, responses, and in instructions; Relational messages can be communicated rater-judge effects; and - most central of verbally, as when one person says to another: "I all - the method problem. like you," "I care about you," or "I trust you." However, they always are communicated nonverbally, accompanying other verbal content. As a consequence, relational messages are present in all communication transactions. When a physician describes treatment regimens to patients; television spokespersons urge Staff adolescents not to smoke, drink, or use drugs; or Director Evelyn Perloff, PhD politicians espouse health care policy; their spoken Document Delivery words carry a substantive message, but their Coordinator Johanna Kehler, D Phil nonverbals simultaneously express a relational Technical Specialist Linda S. Perloff, PhD message. Our research in the political and health Editorial Specialist Philip Mabry, PhD contexts reveals that relational messages often Newsletter Editor Al K. DeRoy, PhD overpower substantive messages in impacting Computer Consultant Alfred A. Cecchetti, MS Computer Programmer/ receivers (e.g., see: Burgoon, Pfau, Parrott, Birk, Analyst lain F. Cameron, M Div Coker & Burgoon, 1987, Communication Measurement Consultant Fred B. Bryant, PhD Monographs, 54, 307-324; Pfau, 1990, Journal Information Specialist Kim Chrobak, MLS of Broadcasting & Electronic Media, 34, 195- Record Analysts Daniel Moore, MA 214; Pfau, Diedrich, Larson, & Van Winkle, Julia Peters, MA 1993, Journal of Broadcasting & Electronic Record Reviewer Jacqueline A. Reynolds, BA File Processor Betty Hubbard Media, 37, 275-292; Pfau & Kang, 1991, Business Services Manager .. Diane Cadwell Communication Studies, 42, 114-128).

Vol. 4, No.2, Spring 1997 5 The Behavioral Measurements Letter Measuring Perceptions (Continued) representing these six relational dimensions: composure (physician was relaxed, calm, and We have known about the presence of relational poised while talking with patient), immediacy (the messages in human communication since Ruesch physician communicated warmth and and Bateson's classic book in 1951 involvement, and found the conversation (Communication: The social matrix of psychiatry, stimulating), lack of dominance (physician didn't New York: W. W. Norton). Later, when dominate conversation), informality (physician Watzlawick, Beavin, and Jackson observed that, made the interaction very informal), receptivity "One cannot not communicate" (1967, (the physician was open to patient concerns, Pragmatics of human communication: A study of interested, and willing to listen), similarity interactional patterns, pathologies, and paradoxes, (physician made patient feel similar, and treated New York: W. W. Norton), they were stressing them as an equal). Due to unreliability, lack of that people always communicate relational dominance and informality were measured as messages, regardless of the substance of their single items due to unreliability. Coefficient alpha verbal message. However, precisely because reliabilities of the remaining dimensions ranged relational messages contain both verbal and respectively from .69 to .80, indicating that nonverbal components, the construct proved relational communication, particularly the difficult to measure. dimension of receptivity, exerted significant influence on patient satisfaction with physicians. Burgoon and Hale (1987, Communication Monographs, 51, 193-214), conceptualized When employed in a mediated campaign's relational communication in a seminal article. context, Pfau (1990) used 26 items which- They derived 12 relational communication represented seven relational communication dimensions based on a review of the extant dimensions. Dimensions and scale items were: literature on interpersonal and relational immediacy/affection (the communicator was communication. Subsequently, they constructed -involved, enthusiastic, interested, and warm), and factor-analyzed 64 Likert items using 5-point similarity/depth (communicator revealed depth, strongly-disagree and -agree scales. Results similarity, caring, and friendliness), composure revealed an 8-factor solution as optimal. The eight (communicator was relaxed, comfortable, poised, dimensions were: similarity/depth, composure, and lacked tension), receptivity/trust (communi- formality, immediacy/affection, receptivity/trust, cator was sincere, honest, interested in com- equality, dominance, and task orientation. municating, and seemed willing to listen), However, the results indicated that more informality (the communicator displayed abbreviated versions of the measure also were informality, equal status, and casualness), lack satisfactory . of dominance (communicator didn't come off as taking the upper hand, manipulative, controlling, Initially, the researchers used Likert scales or pressing influence), and equality (the because the relational communication instrument communicator was cooperative, displayed was intended as a self-report measure. However, equality, and didn't convey a superior attitude). the measure also has been employed as an other- The informality measure was dropped due to report (Burgoon, et aI., 1987; Burgoon, Olney, inadequate reliability. Reliability ratings of & Coker, 1987, Journal of Nonverbal Behavior, remaining dimensions ranged from .60 to .83. 11, 146-165), an observer- report measure The results of the study of communication (Burgoon & Newton, 1991, Southern Communi- modalities in campaign influence revealed that cation Journal, 56, 96-113), and has been adapted relational communication in general, and the for use as a 7-interval adjective-opposite measure. dimensions of receptivity/trust and similarity/ depth in particular, exercised considerable When employed in a health context, the measure influence across modalities. was adapted, with less relevant dimensions removed, and the number of items for each The relational communication instrument is quite dimension reduced. Burgoon and colleagues promising. It uniquely taps both verbal and (1987) used 14 items to tap relational nonverbal components in person perception. It is communication in a physician-patient context, adaptable to both direct and mediated contexts.

The Behavioral Measurements Letter 6 Vol. 4, No.2, Spring 1997 · Measuring Perceptions (Continued) The Comparative Anatomy of Related Instruments: Finally, the relational communication measure appears to be both valid and reliable. Burgoon An Emerging Specialty and Hale (1987) conducted a number of tests Fred B. Bryant demonstrating validity for their relational communication instrument. Further, the relational An exciting, new approach to construct validation communication instrument has achieved improved , reliability ratings in recent studies. is being pioneered quietly in the professional journals these days. Far 'away from the limelight, quantitative specialists labor silently in relative Michael Pfau is Professor and Director of obscurity, systematically comparing alternative Graduate Studies, School of Journalism measures of the same construct to determine and Mass Communication, the University conceptual overlap and uniqueness, using state- of Wisconsin-Madison. He has authored of-the-art multivariate statistical tools. In this or co-authored four books and more than column, I briefly consider the wide-ranging 50 articles and book chapters. His benefits of this basic measurement thrust, and then research interests involve mass media spotlight illustratively a recent empirical example influence, particularly in the health and involving the construct of optimism as an political contexts. illustration. In future issues of The Behavioral Measurement Database Letter, I will highlight this important measurement initiative from other research areas. HaPI Advisory Board

Aaron T. Beck, MD Although largely ignored in favor of more University of School of Medicine glamorous substantive research, this ground- Timothy C. Brock, PhD breaking measurement contribution strengthens , Psychology the foundation of empirical inquiry in several vital William C. Byham, PhD ways. First, comparative studies of related Development Dimensions International instruments fine-tune our understanding of exactly Donald Egolf, PhD what our research instruments are measuring. , Communication Although multiple measures share the same Sandra 1. Frawley, PhD conceptual "label" on the surface, they may well Yale University School of Medicine, Medical Informatics tap different aspects of the same construct or even different constructs altogether. Despite our David F. Gillespie, PhD George Warren Brown School of Social Work tendency to judge books by their covers, the only Washington University definitive way to determine the degree of Robert C. Like, MD, MS functional and conceptual equivalence across University of Medicine and Dentistry of New Jersey instruments is this emerging type of psychometric Robert Wood Johnson Medical School comparison. Such basic quantitative work better Joseph D. Matarazzo, PhD enables us to choose the most appropriate Oregon Health Sciences University instruments for our research purposes. Next, Vickie M. Mays, PhD another vital contribution of this basic University of California at Los Angeles, Psychology measurement work is the improvement of Michael S. Pallak, PhD conceptual clarity by identifying constructs that Behavioral Health Foundation are truly unitary and by decomposing Kay Pool, President multidimensional constructs into their constituent Pool, Heller & Milne, Inc. parts. This conceptual dissection explicates the Ellen B. Rudy, PhD, RN, FAAN University of Pittsburgh School of Nursing meaning of research constructs empirically, documenting how respondents react explicitly to Gerald Zaitman, PhD Harvard University Graduate School of the instrument instead of relying implicitly on the Business Administration instrument's theoretical or intended structure. Stephen 1. Zyzanski, PhD Multiple facets of conceptual variables can thus Case Western Reserve University School of Medicine be identified and better understood, and gaps in

Vol. 4, No.2, Spring 1997 7 The Behavioral Measurements Letter The Comparative Anatomy (Continued) Administering the three instruments to a sample of 389 college students, Chang et al. (1994) used measurement coverage can be highlighted for confirmatory factor analysis to test whether future research. Finally, this type of measurement responses to each instrument were more foray often leads to refinements in existing accurately represented by a single, total score or instruments, creating psychometrically-purified by separate subscale scores. Analyses indicated forms of the original measures for future use. that: (a) the LOT had separate optimism and These modified measures offer greater conceptual pessimism subscales that correlated -.52; (b) the precision and better reliability. HS was most accurately represented by a single, total score that assesses pessimism; and (c) the A 1994 article by Edward Chang, Thomas OPS had multiple subscales that confound D'Zurilla, and Albert Maydeu-Olivares, optimism and pessimism with several other (Cognitive Therapy and Research, 18, 143-160) overlapping constructs. To improve measurement illustrates the benefits of a comparative anatomy precision for the OPS, Chang et al. omitted all of measurement instruments. Chang et al. (1994) OPS questions that did not clearly reflect either sought to improve understanding of the construct positive or negative expectancies (leaving only 14 of optimism by comparing responses items), and found that responses to the remaining systematically to three measures designed to tap items could be accurately represented by separate this construct: (a) the Life Orientation Test (LOT; optimism and pessimism subscales that correlated Scheier & Carver, 1985, Health Psychology, ~. -.45. 219-247); (b) the Hopelessness Scale (HS; Beck, Weissman, Lester, & Trexler, 1974, Journal of Chang et al. found additional differences when Consulting and Clinical Psychology, 42, 861- they compared the psychometric properties of the 865); and (c) the Optimism and Pessimism Scale three instruments. The distributions of scores (OPS; Dember, Martin, Hummer, Howe, & obtained from the different measures were not Melton, 1989, Current Psychology: Research and equivalent. Nor did the instruments show Reviews, ~, 102-119). That is, it was equivalent levels of reliability-specifically, the -hypothesized that optimism consists of life OPS optimism subscale was less reliable than any orientation, hopelessness, and optimism itself and of the other .subscales and than the HS total score. pessimism. Intercorrelating the various scale scores from these three instruments, Chang et al. found that These instruments differ in at least four ways. the subscales of optimism and pessimism showed First, they each reflect different conceptual good convergent validity and modest discriminant definitions of optimism. Specifically, life validity across the instruments. These results orientation (LOT) defines optimism as both clearly indicate that the three instruments do not positive and negative expectancies about future yield equivalent information. outcomes. Hopelessness (HS) in contrast, considers only negative expectancies about self As another means of assessing the validity ofthe and future, while optimism and pessimism (OPS) distinction between optimism and pessimism, define optimism as both positive and negative Chang et al. examined the relations between _ views of life in general. Second, the three (a) the various scale scores from the three instruments are usually scored differently. measures, and (b) external criteria of grade-point Specifically, the LOT and HS are typically average (GPA) taken from college records and summarized in terms of a total score; whereas the self-reported level of psychological stress as OPS is typically summarized in terms of scores measured by the Derogatis Stress Profile (DSP; on separate optimism and pessimism subscales. Derogatis, 1980, The DSP: A summary report, Third, the instruments differ in their length. Towson, MD: Clinical Psychometric Research). Specifically, the LOT consists of 12 items, the Whereas none of the optimism or pessimism HS of 20 items, and the OPS of 56 items. Fourth, subscales were significantly correlated with GPA, the instruments have different response-formats. all correlations with DSP total score were Specifically, the LOT and the OPS use a Likert statistically significant (i.e., higher optimism response-format, whereas the HS uses a true-false related to lower stress, higher pessimism, to format. Thus, although the instruments have higher stress). Subsequent analyses revealed that similar titles, they take very different forms. scores on the LOT optimism subscale had more

The Behavioral Measurements Letter 8 Vol. 4, No.2, Spring 1997 The Comparative Anatomy (Continued) choices about instrumentation. Clearly, this type of "behind the scenes" measurement work shows to do with self-reported stress than did scores on great promise for enhancing conceptual and the OPS optimism subscale-a finding that seems psychometric precision in the social and health to reflect the lower reliability of the latter sciences. measure.

Chang et al. 's findings challenge the traditional When science, art, literature, and philosophy theoretical view of optimism and pessimism as are simply the manifestation of personality polar opposites of a single continuum; and they they are on a level where glorious and demonstrate that the data from anyone particular dazzling achievements are possible, which instrument should not simply be integrated with can make a man's name live for thousands the data from a seemingly related instrument, of years. But above this level, far above, unless there is evidence that the two measures are separated by an abyss, is the level where the factorially congruent, that is, that they assess the highest things are achieved. These things are same concept. Although all three instruments essentially anonymous. might appear at first glance to be equivalent, Chang et al. 's data indicate otherwise, and their Simone Weil results better enable researchers to make informed

HaPI Thoughts

• • •• • • • • • ••••• • • •••• • ••••• • •• • •• • • • •• ••• ••• • • • • • • • • • • • • • ? Is the glass half empty or • • half full or could it be both? • • • • • • • Folk wisdom suggests that • • • • how you see the glass reflects • • your level of dispositional • • optimism. Such face validity • • • • is often accepted. Without • • empirical evidence, belief in • • the accuracy of the contents • • • • in the glass remains a • • matter of "faith, validity." • • • • • • Give us your opinion! • • • • • • • • •• • • • •••• • • • • ••• ••• • •••• •• • • • • •••• • • ••• • • •••

Vol. 4, No.2, Spring 1997 9 The Behavioral Measurements Letter ......

The Behavioral BEHAVIORAL Measurement MEASUREMENTS Database Letter Services ......

In This Issue: • About This Newsletter ... Transitionally Speaking- Robert Perloff, Guest Newsletter Editor. . . .. 1 • The Method in MT -MM- Donald W. Fiske ' : . . ... 3 • Measuring Perceptions of Relational Communication-Michael Piau ...... 5 • The Comparative Anatomy of Related Instruments: An Emerging Specialty- Fred B. Bryant : , 7 • HaPI Thoughts 9 ......

.'lVI~3.LVW a3.LVa

L8LO-Z£Z~I va 'q~Jnqsn!d • L8Z0II XOS: Od

'v'd '46mqSH!d SgJ!AlgS g£c:~ 'ON l!wJad gSeqelea CI"d lUglli~UnSegw SaWH a6"E1jSOd's'n lelO!ABqgs: al"E1t1)tlnS JdUd'] SlUgUIglnSU~rw:lUlO!AUqgg d1iL