An Opportunity for Empiricism

Howard Gardner Review by Multiple Intelligences : The Theory in Practice David Lubinski and Camilla P. Benbow New York : BasicBooks, 1993 . 304 pp. ISBN 0-465-01821-1 (hardcover) ; 0-465-01822-X (paperback) . $30 .00 ($40.00, Canada) hardcover; $15 .00 paperback

Howard Gardner, professor ofeducation and codirector of Project Zero at Harvard University (Cambridge, Massachusetts), is recipient ofa MacArthur Prize Fellowship (1981-1986), the Grawemeyer Award in Education (1990), an American Psy- chological Association (APA) William James Award, and an APA National Psychology Awardfor Excellence in the Media. Gardner is author ofCreating Minds: An Anatomy of Creativity Seen Through the Lives of Freud, Einstein, Picasso, Stravinsky, Eliot, Graham, and Gandhi ; Leading Minds; and Frames of Mind and coauthor, with D. H. Feldman and M. Csikszentmihalyi, of Changing the World: A Framework for the Study of Creativity . 0 David Lubinski, associate professor ofpsychology, director ofthe Psychometrics and Applied Individual Differences Division, and codirector of the Study of Mathematically Precocious Youth (SMPY) at Iowa State University (Ames), is coeditor, with R. V. Dawis, of Assessing Indi- vidual Differences in Human Behavior : New Methods, Concepts, and Findings and coauthor, with R. V. Dawis, ofthe chapter "Aptitudes, Skills, and Proficiencies" in M. D. Dunnette and L. Hough (Eds .) Handbook of Industrial and Organizational Psychology (Vol. 3, 2nd ed.). a Camilla P. Benbow, distinguished professor, chair of psychology, and codirector ofSMPY and ofthe Office of Precollegiate Programs for Talented and Gifted (OPPTAG) at Iowa State University, is a member ofthe Johns Hopkins Society of Scholars and coeditor, with J. C. Stanley, of Academic Precocity and, with D. Lubinski, of the forthcoming Intellectual Talent: Psychometric and Social Issues .

ultiple Intelligences: The Theory in cal, spatial, linguistic, interpersonal, and lishing differential educational-voca- Practice is a distillation of work intrapersonal intelligences) . These intel- tional counseling expectations on the ba- in educational assessments and interven- ligences are construed as literally inde- sis of individual differences in abilities tions stimulated by Gardner's (1983) pendent from one another, both pheno- and interests, a perspective with a long theory of multiple intelligences (MI), typically and neurologically, and should and celebrated history in applied psy- which was first presented in his popular be, but are not, equally valued from a chology (Brayfield, 1950 ; Brewer, 1942 ; Frames ofMind: The Theory ofMultiple sociocultural point of view . We agree Williamson, 1939, 1965). It is a message Intelligences. Succinctly, Gardner has with Gardner that the differential validity that has been lost in today's school re- proposed that conventional general in- of ability dimensions beyond general in- form . Yet we feel obliged also to note telligence tests are narrow and one-di- telligence is underappreciated in edu- that these intelligences, while adding mensional whereas, in reality, there are cational settings (Humphreys, Lubinski, nuances to the psychology ofintellectual seven "intelligences" (i.e., musical, & Yao, 1993). We also applaud him for behavior, are not new . They possess bodily-kinesthetic, logical-mathemati- reminding us of the importance of estab- strong linkages to traditional psycho-

CONTEMPORARY PSYCHOLOGY, 1995, Vol. 40, No. 10 935 metric conceptualizations ofhuman abil- documentation of their efficacy (simply of external real-world relationships than ities, and nearly all have been identified because a problem exists) can make a more elaborate, expensive, and time- in the classic psychometric literature things worse. consuming system, then the former is (Carroll, 1993, p . 641 ; Messick, 1992). The driving force behind Gardner's certainly preferable. Moreover, their applied utility, in edu- agenda to change the way intelligences The importance of real-world criteria cational (Snow & Lohman, 1989) and vo- are assessed stems from his concerns, is, as a matter of fact, a recurring theme cational (Lubinski & Dawis, 1992) con- which are not clearly specified, about throughout Gardner's book. It is curious, texts, has been stressed by those who find traditional "decontextualized assess- therefore, that Schmidt and Hunter's scientific merit in conventional instruments," such as the Standard Achieve- (1981) work on the validity generaliza- ments. Nonetheless, they remain under- ment Test (SAT) and American College tion of "decontextualized" instruments used. Test (ACT) exams. He says that they is not cited with admiration . For over two These are some general comments on should be abandoned (p. 184). He rec- decades, these investigators have com- Gardner's theory and philosophy. Yet MI ommends instead contextualized assess- piled thousands of ability-performance theory is not the main thrust ofthis book, ments, which are conducted in more nat- validity coefficients on over 12,000 jobs although it is the conceptual foundation ural settings-where learning and cre- (over the full range of occupational pres- for the work presented. Discussed here, ativity actually happen. Contextualized tige, which they have classified into five in more detail, is the need to assess each assessments are predicated on the idea categories based on their nature and of these intelligences in an intelligence- that, to truly assess intelligence and its complexity). Through their extensive fair way by using portfolios and project many manifestations, we must do so while meta-analyses of literally hundreds of ratings, for example, in contrast to out- a person is operating on a meaningful task thousands of workers, using general in- of-context standardized assessments. This in "the real world." Why, we feel com- telligence and other conventional psy- book also attempts to capture how MI pelled to ask, do we need to engage in a chometric factors as predictors, they find theory is and can be translated into prac- zero-sum game? Is there not room in that, in intellectually demanding occu- tice through a collection of papers and psychology for both forms of assessment? pations, nearly half of the variance in chapters not necessarily written with this And should both approaches not be sub- performance criteria is accounted for by book in mind or solely by Gardner him- jected to empirical scrutiny to ascertain the general intelligence factor of tradi- self. Some chapters, for instance, include their comparative usefulness and incre- tional instruments.' Furthermore, the updates on Project Spectrum (a project mental validity? Moreover, there are general factor accounts for substantively aimed at assessing abilities in an intelli- concerns about contextualized assess- significant variance across all job classes gence-fair way) and the Key School in ment that are not addressed in this book. (cf. Schmidt, Ones, & Hunter, 1992, and Indianapolis (a K-6 elementary school The first involves the process of con- references therein). Surely these are real- based on MI theory). For the remainder struct validation itself. Scientific justifi- world criteria . 2 of this review, therefore, we have de- cation for the context of assessment, as Given their stinging criticism of the cided to focus on the conceptual and em- well as its medium, is obtained from the validity of psychometric measures, we pirical bases for Gardner's recommen- breadth and depth of a measuring instru- were surprised by the lack of concern for dations for school practice. ment's network of correlates and fore- the many technical reliability-validity We will examine in particular his as- casting capabilities . They document its problems one encounters with ratings sessment recommendations inasmuch as construct validity . The context of assess- (Landy & Farr, 1980) and portfolio as- Gardner has noted that, "while our ed- ment, like an instrument's public rela- sessments, particularly when done on a ucational work has ranged from curric- tions appeal or its face validity, is not a national level. We were left wondering ulum development to teacher education, central component to the construct val- whether Gardner fully appreciates these our primary point of leverage has been idation process. Gardner is skeptical of issues and what he anticipates the reli- the creation of new forms of assessment" assessments like reversed digit span, be- ability is of his proposed assessment which "allow individuals to demonstrate cause people are not often asked to per- scheme. Without reliability, we cannot their strengths and their understandings form such tasks in meaningful life situa- have validity in the real world or any- in ways that are comfortable for them yet tions (p. 241) . But in reality, most psy- where else. subject to public accountability" (p. xv) . chologists are no more intrinsically Moreover, the empirical work aimed These assessments have attempted to interested in digit span than a physician at evaluating Spectrum, Gardner's ap- "get away altogether from tests and cor- is intrinsically interested in oral temper- proach to assessment, would pass few relations among tests, and look instead at ature. What these scientific practitioners preliminary screening committees for a more naturalistic sources of information are interested in are the correlates and master's thesis in psychology . Too about how peoples around the world de- causes of individual differences assessed strong? Consider the following three velop skills important to their way of life" by these measures, because this network studies underscored in Chapter 6 . They (p. 7) . This certainly sounds interesting enables them to generate many more evidently constitute a critical evidential and important. Yet a logical-mathemat- valid inferences than if they were igno- base for the utility of Gardner's 15 sep- ical mind might be moved to ask: What rant of their client's status on these di- arate competencies, which he feels are is the empirical support for Gardner's mensions. From a scientific point of view, useful to assess in the schools. The sample proposals? From our perspective, the there is nothing inherently preferable sizes alone preclude meaningful conclu- answer to this question is important to about whether assessment occurs in one sions, yet they constitute the best em- examine. After all, we may all agree, ir- context versus another (and this pertains pirical work found in this book. The first respective ofour frame of mind, that our to the medium of assessment as well) . If, study was based on 13 four-year-olds and educational system needs improvement however, a context is efficient, easy to used 8 of Spectrum's 15 measures; the but, as the history of medicine reveals, adopt by other scientific practitioners, second involved 20 participants and used implementing changes without empirical and generates a more impressive network 10 of Spectrum's measures. Although the

936 CONTEMPORARY PSYCHOLOGY, 1995, Vol. 40, No . t o intercorrelations of these measures are position from empirical falsification. They tions in our educational system. But to not provided, Gardner contends that the also enable researchers interested in say that Gardner's ideas have not ade- second study reveals that Spectrum's multiple intelligences to bypass the pos- quately met meaningful scientific criteria, measures "identify a range of nonover- itive manifold observed in the full range over the same period, would be too gen- lapping capabilities in different content of talent (the positive intercorrelations erous. Based on our reading of the pres- areas" (p. 95), because only one pair of cutting across mechanical, quantitative, ent volume and published work in the scales manifested a significant correlation spatial, and verbal abilities) . most prestigious scientific outlets for ed- at p < .01. But it is difficult to observe a In conclusion, like Gardner's (1983) ucational research (e.g., Gardner & statistically significant correlation with 20 Frames ofMind, this volume is more of Hatch, 1989), we, like others, find little participants, given that a 95 percent a literary exercise than an exposition em- empirical support for or against the confidence interval for this sample size anating from sound scientific inquiry. unique features of Gardner's ideas. Be- spans .89 correlational units! Further- This is not only our view but also appears fore MI theory can be taken seriously by more, no reliability estimates of these to reflect the consensus of methodolog- the scientific community and policy measures are offered-but they are ically sophisticated reviewers of Gard- makers, Gardner's (1983) bold theoret- needed-because even with meaningful ner's work (Bouchard, 1984; Carroll, ical skeleton is in need of empirical flesh. sample sizes these scales could still 1993 ; Messick, 1992; Scarr, 1985; Snow, Only after this is obtained will we be in emerge as relatively independent be- 1985). Certainly, this volume does noth- a position to ascertain whether his cause of an overabundance of error vari- ing to alleviate Sternberg's (1994) con- framework ambulates scientifically with ance. cern with respect to Gardner's seven in- policy implications for our educational Finally, some validity data are pro- telligences: system. We hope that, if the next decade vided in the third study, in which 17 of generates half as much empirical inquiry the 20 children in Study 2 (discussed Curiously, to date, not only are the tests not as the past decade has generated literary previously) were assessed by the Stan- well underway, but they have not yet been prose in response to Gardner's seven in- ford-Binet. These scores were then cor- initiated. To my knowledge, there is not even telligences, we will be in an excellent po- related with 10 of Spectrum's 15 mea- one empirical test of the theory. . . . What sition for substantively appraising the sures. It is concluded that Spectrum's is clear is that the anticipated program of re- educational usefulness ofGardner's mes- measures are getting at something search has not been forthcoming, and may sage. To be sure, this volume is authored never be. (p. 561) unique, simply because the Stanford-Bi- by one of the great literary psychologists net only lightly covaries with them and of our generation. But, as its author is because they generate unique intellec- Yet MI theory prompts some other fully aware, literary skill is only one im- tual profiles in relation to four of the tough questions as well, which are in portant dimension ofthe human intellec- Stanford-Binet subscales. Yet the possi- need of answers. Are we truly willing to tual repertoire ; and often for establishing bility remains that there may be little re- agree that all seven intelligences possess the verisimilitude of novel scientific the- liable variance in these measures to ap- comparable social utility? Should more ories (even in the social sciences), it is preciably correlate with any external in- resources be devoted to the development frequently not the most important. dicator.' And, if so, when contrasting of some intelligences in relation to oth- Spectrum's measures with well-validated ers? How relevant is this taxonomy to References conventional instruments, unique profiles people with IQs below 90 (over 25 per- Bereiter, C. (1976) . IQ and elitism. are essentially guaranteed . Such empir- cent of the population)? How do we deal Interchange, 7, 36-44. Betz, N. E. (Ed.). (1986) . The g factor in ical results are further ensured when with the fact that, ifwe paid more atten- employment [Special issue] . Journal of samples are highly restricted in range. tion to general intelligence for allocating Vocational Behavior, 29 (3). These 17 participants all resided in up- educational opportunity resources, social Bouchard, T. J., Jr. (1984, July 20). per-middle-class homes, and only <_ 4 had mobility would be more fluid (Bereiter, [Review of Frames ofmind : The theory of multiple intelligences] . American journal IQs below 100 (cf. p. 95)! Later in the 1976; Humphreys, 1992)? That is, will a ofOrthopsychiatry, 54, 506-508 . volume, this series of studies is referred shift in emphasis, from conventional Brayfield, A. H. (1950) . Readings in to as a "sustained effort to evaluate Spec- ability assessment to portfolio assess- modern methods of counseling. New trum" (p. 223) . ment, give us a more or less fluid society? York: Appleton-Century-Crofts. vocational There are, however, advantages to And finally, is Gardner's supposition Brewer, J. M. (1942) . History of guidance. New York: Harper & Brothers. conducting evaluations using small sam- about human development really accu- Carnap, R. (1950) . Logical foundations of ples with multiple measures having du- rate? Namely, "My own observations probability . Chicago: University of bious psychometric properties, especially suggest that rarely in life are the fates of Chicago Press. if one "begins with the assumption that individuals determined by what they are Carroll, J. B. (1993) . Human cognitive abilities: A survey offactor-analytic every child has the potential to develop unable to do" (p. 205) . studies . New York: Cambridge strength in one ofseveral areas" (p. 89). To say that Gardner has had an impact University Press. If, for research purposes, one defines on the educational community would be Gardner, H. (1983). Frames of mind: The strength in a given area as scoring at or an understatement . In the decade be- theory of multiple intelligences . New beyond one standard deviation above the tween publication York: Basic Books. of this volume and Gardner, H., & Hatch, T. (1989) . Multiple mean on an experimental measure (as Frames of Mind, literally tens of thou- intelligences go to school: Educational Gardner does in the studies discussed sands of pages of text have appeared (see implications of the theory of multiple previously), researchers are almost sure Appendix C for a small fraction of pub- intelligences. Educational Researcher, to find that everyone in their sample is lished references), which frame and re- 18 1 4-10 . Humphreys, L. G. . Commentary: "talented" or "at for some- frame (1992) promise" the educational implications of What both critics and users of ability thing. Unreliable measures with low in- Gardner's model. It has provided a fruit- tests need to know. Psychological tercorrelations help to protect this sup- ful challenge to our thinking and opera- Science, 3, 271-274 .

CONTEMPORARY PSYCHOLOGY, 1995, Vol. 40, No. 10 937 Humphreys, L. G., Lubinski, D., & Yao, G. ability versus nonability factors? The familiar the general factor are fine for predicting (1993). Utility of predicting group ".50 barrier" in applied psychology may not school performance "but are only an indif- membership : Exemplified by the role of be a barrier at all, but rather the asymptotic ferent predictor of performance in a profes- spatial visualization for becoming an limit of what the ability domain has to offer; sion after formal schooling" (p. 14) is dated. engineer, physical scientist, or artist . the remaining criterion variance might be This years ago, but Journal ofApplied Psychology, 78, 250- was believed to be true 261 . accounted for by a host of disparate, nona- as the quality of criterion assessments im- Landy, F. J., & Farr, J. L. (1980). bility variables each adding a small, but proved and sample sizes became more re- Performance rating. Psychological unique, increment of validity. spectable, the forecasting efficiency of the Bulletin, 87, 72-107. z The error ofomission that Gardner com- general factor in industrial and vocational Lubinski, D., & Dawis, R. V. (1992). mits by not citing and substantively apprais- settings unequivocally revealed itself. In the Aptitudes, skills, and proficiencies. In ing Hunter and Schmidt's work is not a small words of Paul E. Meehl (1990), "Almost all M. D. Dunnette & L. M. Hough (Eds.), one, given the agenda of this volume. He human performance (work competence) dis- Handbook ofindustrial/organizational purports to sample widely from psychology positions, if carefully studied, are saturated psychology (2nd Vol. ed., 3, pp. 1-59). and multidisciplinary contexts to marshal to some extent with the general intelligence Palo Alto, CA: Consulting Psychologists for his Press. support theoretical ideas. Hunter and factor g, which for psychodynamic and ideo- McNemar, Q. (1964). Lost: Our Schmidt's work on validity generalization logical reasons has been somewhat neglected intelligence? Why? American (VG) has appeared frequently in high visi- in recent years but is due for a comeback Psychologist, 19, 871-882. bility outlets and, over the last 10 years, VG (Betz, 1986)" (p. 124). Yet Gardner's analysis , Meehl, P. E. (1990). Appraising and has been one ofthe most frequently discussed motivates him to conclude that, "a focus on amending theories : The strategy of topics in the Annual Review ofPsychology (cf. testing for an allegedly general ability is no Lakatosian defense and two principles Lubinski & Dawis, 1992; & Schmidt et al., longer tenable" (p. 242) . that warrant it. Psychological Inquiry, 1, 1992, and references therein) . There is a rule 3 This is the same point McNemar (1964) 108-141 . in philosophy of science called the Total Ev- made in his American Psychologist classic, Messick, S. (1992). Multiple intelligences "Lost: Our Intelligence? Why?," with respect or multilevel intelligence? Selective idence Rule. It was given to us by Carnap emphasis on distinctive properties of (1950), and it holds that when evaluating a to early attempts at measuring creativity . hierarchy: On Gardner's Frames ofMind substantive hypothesis or theoretical conjec- Some builders ofthese early measures argued and Sternberg's Beyond IQ in the ture investigators are required to assimilate that they were getting at something unique, context of theory and research on the and weigh all available evidence that speaks because their scales correlated so lightly with structure of human abilities. to the scientific significance of the concept general intelligence . What they failed to Psychological Inquiry, 3, 365-384. under analysis . For anyone interested in demonstrate, however, was whether their Scarr, S. (1985). An author's frame of school and work performance, it behooves new assessment tools tapped meaningful mind: Review of Frames ofmind : The them to amass the findings on VG. Gardner's psychological phenomena beyond general theory ofmultiple intelligences . New intelligence-it not. Ideas in Psychology, 3, 95-100 . comment that IQ tests and other indices of turns out that they did Schmidt, F. L., & Hunter, J. E. (1981) . Employment testing: Old theories and new research findings. American Psychologist, 36, 1128-1137 . Schmidt, F. L., Ones, D. S., & Hunter, J. E. A Response on Four Fronts (1992) . Personnel selection. Annual Review ofPsychology, 43, 627-670. is an invited response by Howard Gardner to David Lubinski and Snow, R. E. (1985) . [Review of Frames of The following mind: The theory ofmultiple Camilla P. Benbow's review of his book Multiple Intelligences : The Theory in intelligences] . American Journal of Practice . Education, 88, 109-112. Snow, R. E., & Lohman, D. F. (1989) . Implications of cognitive psychology for appreciate the opportunity to respond participants running the gamut from the educational measurement. In R. L. Linn to the review by Lubinski and Ben- arts to study skills, as well as essays about (Eds.), Educational measurement (3rd bow. I cannot say "review of my book," schools in the future . ed., pp. 263-331) . New York: Collier- because their essay constitutes an at- One section, covering about a sixth of Macmillan. Sternberg, R. J. (1994) . Commentary: tempt by the authors to grind their own the book, is directed toward assessment, Reforming school reform : Comments on axes, or, to twist the metaphor, to execute and even that section stresses the impor- Multiple intelligences : The theory in a hatchet job. I reply on four fronts. tance ofgoing beyond assessment per se. practice . Teachers College Record, 95, The key chapter, "Assessment in Con- 561-569 . text : The Alternative to Standardized Williamson, E. G. (1939) . How to counsel Scope students . New York: McGraw-Hill. A review should convey to the reader the Testing," discusses a number of the issues Williamson, E. G. (1965). Vocational intention and scope ofthe book. Multiple that also concern Lubinski and Benbow. counseling : Some historical, philosophical, Intelligences: The Theory in Practice is and theoretical perspectives . New York: McGraw-Hill. not an extension of the theoretical or Accuracy empirical work summarized in Frames of My attention was caught by Lubinski and Footnotes Mind: The Theory of Multiple Intelli- Benbow's remark that, on page 184, I 1 It is possible that we have an appreciable gences, but rather a report on various ef- recommend the abandonment of Stan- understanding of the relationship between forts to reform precollegiate education dard Achievement Test (SAT) and Amer- abilities and performance for most educa- along lines suggested by multiple intel- ican College Test (ACT) exams. I quote tional-vocational domains. To the extent that ligences (MI) theory. Included in the my actual words: nonintellectual factors are relevant to performance, and they undoubtedly are (e.g., collection are answers to frequently energy, health, interest, motivation, person- posed questions about the theory, de- In my view, there is little need and little ad- ality, chance factors, and so on), an important scriptions of several programs imple- vantage to be gained by continuing to require question becomes: How much of individual mented at levels ranging from kinder- the Scholastic Aptitude Test (I have fewer differences in performance is attributable to garten to high school and encompassing reservations about the achievement tests).

938 CONTEMPORARY PSYCHOLOGY, 1995, Vol. 40, No. 10 Most colleges are not selective enough to in, say, an assessment center, and then much of The Bell Curve, but I was cer- warrant such an instrument, and those that examine actual productivity on the job. tainly not impressed, particularly when are have sufficient additional sources of in- What of efforts to assess multiple in- I considered how many of the signers formation about their candidates . (p. 184) telligences themselves, along the lines have a vested interest in the survival of suggested by standard psychometrics? psychometrics. Although I do not object to point out, I go on Such efforts certainly can be conducted; to criticism, however, I frankly resent the indeed, Multiple Intelligences cites a snide tone of the review written by Lu- there are risks and expenses involved in a shift number of individuals who have devel- binski and Benbow . to a wider and more flexible set ofinstruments other Rhetoric is one thing, reality is an- but to my mind these are worth taking. The oped MI instruments, and I know of very fact that some schools have already taken more recent efforts. I have personally other. As I pointed out in my review of them-and these are among the very best been leary of these efforts, for I feel that The Bell Curve (Gardner, 1995), much schools-shows that my vision is not utopian. they run counter to the contextualized psychometric work has been carried out (p . 186) bases of the theory (so I have often ar- with a callous disregard for the uses to gued "intelligences should be assessed which such instruments have typically Such misleading paraphrases lead me to directly and not through the lens of a been put, and most especially, to those suggest that readers consult the book it- standard test" [p. X]); putative "MI tests" uses that represent a disservice to the self, where they will find that a - zero- are as likely as not to repeat the sins of broader society. sum game" occurs in the minds of Lu- traditional testing to which I have already MI theory has garnered unexpectedly binski and Benbow rather than in the alluded. The few empirical demonstra- wide appeal in part because it offers a words of Howard Gardner. tions reported on in the book are just more generous-and, I fervently be- that: efforts to show that one can make lieve-a more accurate view of the hu- Substance rough-and-ready performance-based as- man mind. My own turn to the classroom In the aforementioned chapter, and in sessments of the intellectual strengths of has been motivated by a desire to im- other passages (as in the one just cited), young children and that the results of prove the performances (and hence the I consider some of the issues involved in these surveys are consistent with the ma- lives) of school children . Naturally, I a shift to performance-based assessments. jor claims of the theory. They are not in- would like to obtain the most convincing Lubinski and Benbow show scant aware- troduced as ersatz national exams but evidence for any effects that might be ness that the current short-answer ex- rather as useful classroom assessments . achieved, and Ijoin Lubinski and Benbow aminations, emphasized far more in the Contrary to what Lubinski and Ben- in hoping that the best measures can be United States than in other industrialized bow say, much current experimental and devised, even if none of us turn out to be societies, have had quite destructive in- empirical work bears on the claims of MI the devisers . In the meantime, I am con- fluences . Such tests all too often constrain theory: As instances I can mention Ros- tent to let MI theory-and its rivals-be what goes on in American classrooms; now, Skleder, Jaeger, and Rind's (in assessed by two criteria : concordance and, as dramatized by the content ofand press) work on personal intelligence and with the accumulating information about the response to The Bell Curve, they can Rauscher, Shaw, and Ky's (1993) work the human brain and the human mind and also inflict pain on larger society as well. on the relation between musical and spa- contributions to the educational welfare The continued reliance on subtests like tial abilities . But as I indicated, review of of school children . Reverse Digit Span does send out an un- such work was not within the scope of Multiple Intelligences. References helpful signal; we should be looking at Gardner, H. (1993) . Frames of mind: The abilities that truly matter (like under- theory of multiple intelligences. New standing the relations among numbers) Rhetoric and reality York: Basic Books. rather than abilities that might correlate I have no objection to criticism; I can give Gardner, H. (1995) . Cracking open the IQ with abilities that we truly value. If we receive it. Lubinski and Benbow box. American Prospect, 20, 71-80. it and Rauscher, F., Shaw, G. L., & Ky, K. N. want to have schools in which youngsters correctly cite individuals who have cri- (1993) . Music and spatial task learn what is worth knowing, and strive tiqued MI theory and I can add to their performance. Nature, 365, 611 . to use their minds well, and if we want a list! However, the argument from au- Rosnow, R., Skleder, A., Jaeger, M., & society in which individuals are valued thority is a weak one at best, and theory Rind, B. (in press) . Intelligence and the epistemics of interpersonal acumen: for what they can actually do, then an evaluation is not a bean-counting game. Testing some implications of Gardner's exploration of new forms of assessment The Wall Street journal was able to find theory. Intelligence . becomes not an option, but an impera- 52 psychologists who would endorse tive. I am of course aware of the work of Hunter, Schmidt, Bishop, and others, which I review in a forthcoming textbook An Opportunity for "Accuracy" on intelligence. That work is interesting but controversial and limited (see forth- Thefollowing is a rejoinder by David Lubinski and Camilla P. Benbow. coming critiques by Richard Murnane, Frank Levy, Earl Hunt, and others). In a T n a section titled Accuracy, Gardner communicate a more neutral stance) ; but sense the Hunter-Schmidt line of re- begins, "My attention was caught by these are not the words we had in mind. search bypasses performance at both Lubinski and Benbow's remark that, on They come at the end of the same para- ends : The correlations are typically be- page 184, 1 recommend abandonment of graph. "I would like to see leading col- tween test scores and supervisers' ratings. Standard Achievement Test (SAT) and leges follow the example of Bates College A more genuine performance-based in- ACT exams" (p. 938) . He goes on to and Franklin and Marshall College: They strument would look at a sample of work quote his "actual words" (words that should dispense with the requirement of

CONTEMPORARY PSYCHOLOGY, 1995, Vol. 40, No. 10 939

the Scholastic Aptitude Test and its counterpart instruments" (p. 184) . Gardner then goes on to say, "Such On Accuracy misleading paraphrases lead me to suggest that readers consult the book itself, where they will find that a 'zero-sum game' occurs in the minds of Lubinski Thefollowing is a final response by Howard Gardner. and Benbow rather than in the words of Howard Gardner" (p. 939). We agree with Gardner that readers should consult the book itselfrather than simply relying ubinski and Benbow are dismissive and later in the response, I favor making on "misleading paraphrases" in our ex- L about "literary" matters, but literacy the SAT optional (p. 186, line 10). The change. We are extremely comfortable matters. I did not recommend "abandon- zero-sum remains just where I originally with readers deciding for themselves . We ment" of the Standard Achievement Test placed it-in the imaginations of the re- stand behind our review. (SAT); as indicated in the quoted passage viewers.

Call for Nominations

The Publications and Communications Board has opened nominations for the editorships of the Journal of Experimental Psychology: Animal Behavior Processes, the "Personality Processes and Individual Differences" section of the Journal ofPersonality and Social Psychology, the Journal of Family Psychology, Psychological Assessment, and Psychology and Aging for the years 1998-2003. Stewart H. Hulse, PhD; Russell G. Geen, PhD; Ronald F. Levant, EdD; James N. Butcher, PhD; and Timothy A. Salthouse, PhD, respectively, are the incumbent editors.

Candidates should be members of APA and should be available to start receiving manuscripts in early 1997 to prepare forissues published in 1998. Please note that the P&C Board encourages participation by members of underrepresented groups in the publication process and would particularly welcome such nominees.

To nominate candidates, prepare a statement ofone page or less in support ofeach candidate and send to the attention of the chair of the appropriate search committee. Search committee chairs are

" Joe L. Martinez, PhD, for JEP: Animal Behavior Processes. Members of the search committee are Russell M. Church, PhD; Michael Domjan, PhD; Michael S. Fanselow, PhD; and William D. Timberlake, PhD.

David L. Rosenhan, PhD, for the "Personality Processes and Individual Differences" section ofthe Journal ofPersonalityandSocial Psychology. Members ofthe search committee are Nancy E. Cantor, PhD; Susan Fiske, PhD; Oliver John, PhD; and Mark Snyder, PhD.

Carl E. Thoresen, PhD, for the Journal ofFamily Psychology. Members of the search committee are Arthur M. Bodin, PhD; James H. Bray, PhD; Lucia Gilbert, PhD; John M. Gottman, PhD; and Howard A. Liddle, EdD.

Hans H. Strupp, PhD, for Psychological Assessment. Members of the search committee are Lee Anna Clark, PhD; Ken Pope, PhD; M. Tracie Shea, PhD; and Auke Tellegen, PhD.

Lyle E. Bourne, PhD, forPsychology andAging. Members of the search committee are Carol Barnes, PhD; Alfred Kaszniak, PhD; M. Powell Lawton, PhD; Michael A. Smyer, PhD; and Larry Thompson, PhD.

Address all nominations to the appropriate search committee at the following address:

Lee Cron P&C Board Search Liaison Room 2004 American Psychological Association 750 First Street, NE Washington, DC 20002-4242

First review of nominations will begin December 11, 1995.

940 CONTEMPORARY PSYCHOLOGY, 1995, Vol. 40, No. 10