The following article has been made available for private academic use only, on the understanding
that the reader respects the rules of copyright of the original publication. Please refer to
the printed version as slight differences do exist between that publication and the following.
Morris, Clifford & LeBlanc, Raymond
(1996, Spring). Multiple intelligences: Profiling dominant intelligences of grade eight
students. McGill Journal of Education, 31(2), 119-141.
Abstract
This paper outlines how verbal protocols were used to compare the self-perceived intelligences of students
to teacher nominations. Teachers nominated students considered dominant in one of Howard
Gardner’s seven theorized intelligences. The students then verbalized aloud while ranking
themselves using scenarios depicting each intelligence. Concurrent and retrospective verbal
protocols were taped, transcribed verbatim to text, coded, and analyzed. Results indicated a
strong agreement between teacher nomination and student identification of Gardner’s intelligences. More
detailed studies should be completed before determining the validity and reliability of profiling such
intelligences.
Introduction
Have you recently observed intellectual development in an
elementary school classroom only to realize that many of the students therein
appeared as intelligent as the ones formally identified as “gifted”? If your response was in the affirmative,
then what follows may interest you.
Historically, views about intelligence have been polarized. At the more conservative extreme, the
unitary view of intelligence reigns, a view depicting student potential as an
almost innate birthright, more as a result of nature than of nurture. Intelligence is here considered to be mainly
a unitary mental entity. This single
“general g” which is based, in the main, on reasoning and verbal abilities as
reflected by standardized Intelligence Quotient (IQ) tests supposedly comes in
varying degrees, with some possessing more of it than others (Eysenck, 1981;
Freeman, 1950; Jensen 1982; Piaget, 1971; Wechsler, 1958, 1974).
At the opposite end of the intellectual continuum, and more the
focus of this paper, lies a pluralistic, incremental, and dynamic notion of
intelligence, to be more precise, intelligences, or “frames of mind”. This “many kinds of minds” notion is
fostered by numerous theorists, in particular cognitive psychologists who
postulate the existence of a much broader intellectual spectrum (Carroll, 1982;
Davidson, 1990; Ceci, 1990; Das, 1992; Fodor, 1983, 1985; Gardner, 1983, 1987,
1990, 1993, 1995; Gould, 1981; Horne, 1979; Hunt, 1990; Resnick, 1976, 1979;
Sternberg, 1979, 1985a, 1985b, 1986a, 1986b, 1991; Sternberg & Salter,
1982; Sternberg & Smith 1988; Turnbull, 1979). These researchers posit that intelligence involves, in the main,
many “varieties of thinking” (Howard, 1990), and, according to Howard Gardner’s
Multiple Intelligences (MI) model, can be exemplified by all individuals in
many diverse and multiple ways.
However, little research currently exists relating such intellectual
conditions to public school students (Blythe & Gardner, 1990; Sternberg,
Okagaki, & Jackson, 1990; Olson, 1988), and virtually none exists as to how
grade 8 students and their homeroom teachers perceive dominant student
intelligences (from a perspective such as Gardner’s). In other words, the relationship between Gardner’s MI model and
the self-perceived dominant intelligences of grade 8 students and their
homeroom teachers has never been explored.
Multiple Intelligences
At the classroom level, there are several reasons why
teachers must foster the above broader approach to intellectual
functioning. Gone are the days when a
singular IQ score represents the sole sum of intellectual capabilities (Hatch
& Gardner, 1986; Sternberg, 1985a).
Due to space limitations, just two reasons shall be reiterated
here. First, as intellectual development
is dynamic and multi-faceted, it is misleading to quantify the process of
intelligence in the form of a single number (Armstrong, 1993, 1994; Blythe
& Gardner, 1990). Yet, many
schooling systems continue to do so by using the IQ score as the major
criterion for including or excluding students from access to special resources
(Armstrong, 1988; Kolligian & Sternberg, 1987; Maker, 1992; Maker, Nelison,
& Rogers, 1994; Matthews, 1988; Scherer, 1985; Sternberg, Okagaki, &
Jackson, 1990; Vail, 1987). And in
spite of constant references to domain specific intellectual competencies,
multiple talents, or multiple gifts, to name just three, the "bottom
line" for entrance into most special school options seems to be based
mainly on a single IQ score.
Second, the practice of using IQ scores exclusively as the
indicator of intellectual ability underestimates the tremendous potentiality
within all students. Not only do such
scores fail to "predict success in non-academic settings, but they also
are poor predictors of success in school" (Maker, 1992, p. 12). The many “strengths” of pupils can be
nurtured and cultivated, yet practice in most public schools continues to treat
"intelligence" as a rational or analytic mind (Faggella &
Horowitz, 1990; James, 1991). And
finally, over the years psychometricians, in attempting to analyze the raw data
behind the IQ test, have accumulated vast amounts of statistical data, in an
attempt to isolate "primary mental abilities" - verbal, numerical,
and spatial, among others. All
capacities that can not be factored out are supposedly allocated to a general
"g". However, the statistical assumptions
involved in reducing such data are so narrow that very little of theoretical
significance has ever come from them (Kornhaber, Krechevsky, & Gardner,
1990).
A more practical approach is Gardner's concept of “many kinds
of minds”: structured experiential learning that recognizes multiple
intelligences and based more on the psychology of holistic learning. While established schooling has focused
mainly on logical-mathematical and linguistic-verbal skills, Gardner (1982,
1987, 1989, 1990, 1991, 1993, 1995; see also Gardner, Kornhaber, & Wake,
1996) suggests that there are at least five additional “smarts.” His model thus can be readily blended into
programs, by providing teachers with a practical, thematic approach that will
challenge children's thinking.
Using an elaborate set of criteria, including evidence from
studies of brain damage, prodigies, developmental patterns, cross-cultural
comparisons, and various types of tests, Gardner identifies seven central
intelligences: (1) linguistic-verbal:
a language-based competence requiring listening, speaking, reading, and writing
skills, (2) logical-mathematical:
dealing with abstract concepts, patterns and symbols required for deriving
scientific proofs, (3) visual-spatial:
using mental imagery for discerning orientation in space, (4) bodily-kinesthetic: using physical body
movements to express emotion, (5) musical-rhythmic:
recognizing tonal and rhythmic patterns, and creating harmony, (6) social-interpersonal: a social type of
intelligence that operates primarily through person-to-person relationships and
communication, and (7) solidarity-intrapersonal:
an unaccompanied and intuitive style of learning useful in understanding the
self. Gardner believes that each of
these separate intelligences can be enhanced.
More importantly, he places those five latter ways, figuratively
speaking, on the same cerebral pedestal as the former two IQ-types. Gardner believes that each of the above
domain-specific faculties may develop almost independently; he places no order
of priority among them.
Research Objectives
This study represented an attempt to blend Gardner’s many kinds
of minds to grade eight students and their homeroom teachers. The students were asked to self-assess their
own intellectual capacities. The study
had three objectives i) to determine how grade eight students interpreted their
dominant intelligences as they interacted with characteristics representing
each of the seven intelligences, ii) to see if homeroom teachers had a high
level of agreement with the self-perceptions of grade eight students, and iii)
to investigate the usefulness of applying verbal data from student protocols as
a meaningful problem-solving avenue when assessing the intellectual profiles of
students. These objectives were operationalized through the following research question: How reliable are
teacher nominations of student’s dominant intelligences when compared with the
self-perceived three dominant intelligences of the same students?
Methodology
Subjects
The study involved 34 eight-grade students between the ages of
12 and 14 years (mean age = 13 years, 7 months). The students consisted of twenty-one males and thirteen
females. They were drawn from five
grade eight classes from one public (Carleton Board of Education) elementary
school. The subjects were selected in
the month of May, having been observed by their homeroom teachers since the previous
September. Twenty-nine were Caucasian,
four were Black, and one was Oriental.
All were naive with respect to the research task, and all students
voluntarily participated. An incentive
for participating was the opportunity of understanding more about their own
“intelligences.”
As an added inducement, a follow-up home visit was arranged,
eight months after the initial interview.
During this second one-hour session, subjects and their parents received
literature delineating the overall research, including a sheet containing a
vertical bar graph outlining their child's perceived intelligences (see Figures
1), and a verbatim copy of their protocols.
Materials
Two instruments were developed for this study: a
Teacher Nomination Checklist and a set of Gardner Scenario Cards. The Teacher Nomination Checklist (see
Appendix A) comprised intellectual characteristics grouped into seven
categories, each category reflecting one of Gardner’s seven dimensions. These characteristics were compiled from a
literature review on Gardner. As
described in more detail in the section that follows, the checklist was
submitted to grade 8 homeroom teachers to nominate as research subjects
students demonstrating a particular dominance in one of the Gardner
dimensions. The Gardner Scenarios (see
Appendix B) were developed from the Teacher Nomination Checklist and portrayed
in paragraph format. The sentences (in
the version shown to the subjects) were typed in large black print on white
sheets of 8.5" x 11.0" paper.
Unknown to the subjects, dispersed among the characteristics of
each scenario were at least two different characteristics borrowed from at
least one of the other six dimensions.
The reasoning for inserting these other characteristics into the
scenarios was twofold: to encourage subjects to verbalize more, and to make
their problem solving task (i.e., ranking their intelligences) more of an
ill-defined nature. The single
underlined sections (in the scenarios) highlight characteristics of a more
positive nature and borrowed from two of the other dimensions. The double underlined sections highlight
characteristics of a more negative nature and, in a similar fashion, were
borrowed from one of the other dimensions.
The decision to implement the scenario approach was considered
the foremost way to represent common life situations to the subjects. Gardner's MI model was best viewed as the
basis for educational reform, proposing reform that would immediately benefit
students. Also, his model posited that
individuals were somewhat intelligent in each intellectual dimension, stronger
in some dimensions and weaker in others.
Thus, the development of scenarios was considered a practical method for
representing real-life situations. To
sum, the Gardner scenarios attempted to ground a meaningful context, presenting
subjects with practical intelligent situations within the confines of their
lives, thus presenting them with a “gestalt” or holistic picture for the seven
intellectual dimensions.
Procedure
Subjects were selected on the basis of their performance on
three criteria: i) school grades, ii) achievement test scores, and iii)
judgments from their homeroom teachers. Five grade eight homeroom teachers were
provided with a checklist showing a series of intellectual characteristics for
each of Gardner's seven intelligences (see Appendix A). Participants selected were asked to have one
of their parents complete a statement of consent to their participation. Consent forms were collected by the school
principal, and individual times for each home interview were arranged.
Verbal Protocols
Verbal protocol analysis was selected as the methodology for
this research project. Hayes (1981)
defines a verbal protocol as a description of the activities, ordered in time,
in which a subject engages while performing a task. According to Ericsson and Simon’s theory (1980, 1984, 1993; see
especially Ericsson & Oliver, 1988), the conventional method of getting
subjects to verbalize their thought is to instruct them to “think aloud.” And, to remind the subject to think out
loud, the experimenter will often ask the subject to “keep talking.”
Two types of verbal protocol procedures
were used: concurrent verbalizations and retrospective comments. Concurrent verbalizations represent
information that is still being attended to in short-term memory. Retrospective comments consider the
information after the completion of the task-directed process. Ericsson and Simon (1993) argue that such
retrospective verbalizations represent “a subset of the sequence of thoughts
occurring during performance of a task [as] stored in long-term memory” (p.
xvi). To that end, the rationale for
using verbal protocol analysis as a method for assessing cognitive processing
represents a central thesis for what follows (De Groot, 1965; Newell &
Simon, 1972; Rowe, 1985; Schael & Dionne, 1991; Schoenfeld, 1985).
The following procedure was individually administered to the 34
subjects. Each interview lasted
approximately 55 minutes. Upon arriving
at the homes of the subjects, the investigator (the first author) informed
subject and parents of the nature of the experiment and the requirements and
benefits of participating in the study.
Subjects were then interviewed in the absence of their parents who later
returned to be presented with the initial ranking results. Every semi-structured interview commenced
with a warm-up exercise, the purpose of which was to familiarize the subject
with the overall process of the methodology, in this case the verbalizing aloud
procedure. When the investigator felt
that the subject showed comfort with the underlying process and understood all
of the instructions, the interview commenced.
All verbalizations were audio taped onto cassettes and considered as the
primary source of field data.
During the interviews, the subjects were asked to complete
three out loud verbalization assignments: the initial two associated with
concurrent verbal protocols; the final task was of a retrospective nature. First, subjects read aloud the scenarios. Then, they were asked to rank the scenarios
by arranging (on a table) the seven (scenario) cards, from “most like me”
(indicating that they strongly agreed with the scenario’s characteristics) to
“most unlike me” (indicating that they strongly disagreed with the scenario’s
characteristics). The three most
dominant scenarios were placed in a row across the top of the table, with the
card containing the scenario perceived to be most like the subject, in the
upper-left hand corner. In the same manner,
the three least dominant scenarios were placed at the bottom of the table, with
the least dominant scenario located in the lower right-hand corner. The remaining scenario card, representing
the subject’s middle-of-the-road intelligence was centered between the other
two rows of cards. During this and the
subsequent two steps, subjects were encouraged to rearrange any or all of the
cards, and at any time. To conclude the
concurrent verbal protocol nature of the study, subjects were asked to verbalize,
one again, their feelings on individual characteristics in each scenario,
always commenting out loud either positively or negatively as to how they
perceived themselves in relation to the nature of that intellectual
characteristic.
Finally, the home interview concluded with the investigator
posing the following five retrospective-type questions. After being asked to point to their most
preferred and, in turn, to their least preferred scenario, subjects were asked:
“Why do you think that scenario best represents you?” and “Why do you think
that scenario least represents you?”
Then, in order to gain additional retrospective verbalizations that
would corroborate earlier concurrent protocols, subjects were asked two more
questions: “What additional characteristics would you add to your "most
like me" scenario in order to make that scenario more like you?”, and
“What additional characteristics would you add to your "least like
me" scenario in order to make that scenario more least like you?”. And to conclude, subjects were finally
asked: “How confident do you feel about your rankings of the seven
scenarios?” It was surmised that the
responses to these five final questions were of a retrospective verbal protocol
nature, and thus would delineate relevant supplementary field data thereby
advancing the extent of representativity and degree of confidence of the
overall interview process (Ericsson and Simon, 1993). The encoding technique and scoring procedure for each out loud
verbalization is described in the section that follows.
Data
Analysis
This section of the paper outlines how the raw field data was
gathered, analyzed, and interpreted.
First, protocols from the concurrent verbalizations will be
analyzed. Then, the retrospective
verbalizations, reinforcing the concurrent verbalizations, will be
investigated.
Concurrent Verbal Protocols
After the audio cassette tapes were transcribed verbatim to 34
individual word processing text files, the captured verbatim comments were line
numbered. Next, a coding grid was
developed to convert the verbatim transcripts into a format accessible for
verbal protocol analysis. The purpose
of the coding grid was to record and produce “objective traces of the sequence
of overt actions” (Schoenfeld, 1985, p. 283), in this case, every concurrent
and retrospective verbalization. The
coding grid was composed of a series of similar scales, each scale containing
six ranking cells, ranging from “unlike me” to “good at” (see Table 1). The remaining four ranking cells were “don’t
enjoy”, “not good”, “like me”, and “do enjoy.”
If a line number (from a verbatim transcript) was coded into a cell on
one of the grid’s 3 left-hand columns (see Table 2), the subject’s
verbalization to a characteristic was considered a negative statement;
conversely, each time a line number was coded into one of the 3 right-hand
columns, the verbalization was considered a positive statement. Each negative statement was given a
"-1" value and each positive statement was given a "+1"
value. These values (i.e., “-1” or
“+1”) were then totaled and considered as one frequency count for that
particular dimension. A separate coding
grid was used for each subject.
----------------------------------
In the
original, Table 1 was about here
----------------------------------
An example of a subject’s coding grid
containing the line numbers representing out loud verbalizations and including
borrowed characteristics, is outlined in Table 2. Here, six line numbers (“234”, “240”, “246”, “254”, “258”, and
“264”) were placed in the three left-hand columns, and five line numbers (i.e.,
“227”, “228”, “229”, “238”, and “261”) were placed in the three right-hand
columns. Since the columns to the left
housed only line numbers for negative-type verbalizations, the frequency count
for those six comments totaled “-6”; conversely, the three right-hand columns
housed five line numbers representing positive-type verbalizations, the total
being “+5”. Thus, the sum, including
any borrowed characteristics (“-6” + “+5”) was “-1”. In the same manner, when the borrowed characteristics were
removed, the sum (“-3” + “+5”) was 2.
As some scenarios contained more characteristics than others, all
frequency totals were converted to percent.
As outlined in greater detail below, these percent scores were used to
determine the dominance dimensions for each subject.
----------------------------------
In the
original, Table 2 was about here
----------------------------------
As the homeroom teachers considered only the most dominant
intelligence of each nominated student, just the subject’s three highest
rankings were analyzed and interpreted.
These rankings (see Table 3) were combined in the following way: the column heading “Rank 1” contained the
“most like me”, or most dominant student response; the heading “Ranks 1 and 2”
contained the “most like me”, and “second most like me” responses. Likewise, the column heading “Ranks 1, 2,
and 3” contained the top three dominant rankings, namely the “most like me”, the “second most like me”,
and the “third most like me” subject rankings.
During the interview, the more dominant a scenario appeared to represent
the subject's perceived intelligence, the higher the subject ranked that
scenario. Thus, what follows outlines
subject rankings by revealing two sets of results: a wide range of agreement
between homeroom teachers and their nominated students, and comparisons of
results between different aspects of the same studied group.
First, when the Teacher Nomination (TN) was compared to the
Student Perceived Profile (SPP) on Rank 1 alone, the level of agreement (i.e.,
between the five teachers and the 34 students) was 50%. When a similar comparison was made between
TN and SPT+ (Student Profile Total including Borrowed Characteristics), the
level of agreement on the dominant dimension was 29%. When a similar nomination was compared to SPT- (Student Profile
Total excluding Borrowed Characteristics), the level of agreement increased to
38%. The difference in agreement level
suggested a distraction effect from the borrowed characteristics.
Correspondingly, when the two higher rankings (Ranks 1 and 2),
and the three highest rankings (Ranks 1, 2, and 3) were clustered together for
comparison, results indicates a somewhat similar pattern. The agreement levels increased when the conditions
for agreement were relaxed. For
instance, when TN was compared with SPP,
the increase in agreement grew to 74% when considering the relaxation
Ranks 1, 2 and 3. A similar improvement
pattern was shown when TN was compared with SPT+ and SPT- respectively. The relaxed condition appeared to favor the
profiles, moreso when the borrowed characteristics were excluded from the
findings.
Before describing below a similar comparison involving both
concurrent and retrospective verbal protocols, a short diversion is
necessary. During a home interviews, a
father of one of the female subjects expressed concern that some of the
scenarios contained bias against the female (n=13) population. More specifically, he felt that some of the
scenario activities fostered male-type cultural associations. An analysis of the results from the
concurrent verbal protocols showed no such marked difference. For example, when a comparison was made with
the male subjects (n=21), the level of agreement was 43% on the highest
ranking, 62% on the first two highest rankings, and 81% when the first three
subject rankings were considered.
Similarly, when the dominant dimensions of the female student (n=13)
were compared to the teacher nomination, the level of agreement moved from 62%
on the two higher rankings, to 69% when the three highest rankings were
considered. Again, and as expected, the
level of agreement increased when the agreement criteria was relaxed. To sum, the percent differences, while based
on a modest number of subjects (n=34), did not appear to detect any significant
gender bias.
Finally, Table 3 compares the initial
scenario rankings of the students (SPP) to two other results: first, when the
borrowed characteristics were included (STP+) in the final calculations, and
then when the borrowed characteristics were excluded (STP-). Stated slightly differently, SPP is compared
to the more detailed student profiles, SPT+ and SPT- respectively. The dominant dimensions of each profile are
compared to determine the agreement levels between three ways of identifying
student profile. For example, when the
dominant dimension of SPP was compared to the dominant dimension of SPT+, the
level of agreement was 44% on Rank 1.
When a similar comparison involved the top two rankings were compared,
the percentage of agreement was 65%.
When the top three rankings were included, the percentage of agreement
was 79%. In like manner, when the sum
of the borrowed characteristics were excluded from the results, the agreement
was 35% on the most preferred ranking, 59% on the two dominant rankings, and
85% on the first three dominant rankings, respectively. In both instances, the percentages improved
as the level of relaxation increased.
As with earlier results, these findings suggest that, in the main, the
presence of the borrowed characteristics may have aided the subjects as they
ranking the scenarios including their characteristics.
----------------------------------
In the
original, Table 3 was about here
----------------------------------
Concurrent and Retrospective Verbal Protocols
One objective of the study was to compare teacher
nominations to all of the dominant rankings as self-perceived by the
students. Restated slightly differently
and as a question, how did the retrospective comments reinforce the earlier
concurrent verbalizations? The results
are outlined in Table 4. When teacher
nominations (TN) were compared to the dominant dimensions (DD), the level of
agreement was 21% on the first ranking, 38% on the first two rankings, and 50%
when the initial three subject rankings were considered. When TN was compared to all dominant
dimensions, including the borrowed characteristics (DD(C+R)), the level of
agreement was 29% on the first ranking, 53% on the first two rankings, and 62%
on the first three rankings. When TN
was compared to the dominant dimensions including concurrent and
retrospective verbalizations (DD(C+R)), the level of agreement was 24% on the
most dominant rankings, 44% on the highest two rankings, and 62% on the first
three dominant rankings. These
statistics, seem to suggest that the retrospective comments supported the
concurrent verbalizations. And, as
expected, the percentage of agreement increased as the criteria was relaxed.
When TN was compared to DD(CAR) for the
male subjects (n=21), and for the female subjects (n=13), the levels of
agreement were 29% and 15% on the first ranking, 43% and 46% on the first two
rankings, and 62% and 62% on the first three rankings, respectively. Similar to before, the level of agreement
increased when the agreement criterion was relaxed. And, resembling the conclusions reached earlier, the difference
in percentage points between gender appeared inconsequential to justify any
significant gender bias, on any of the initial three rankings. Next, when DD was compared to DD(C+R), the
percentage of frequency agreement was 38% on the dominant ranking, 62% when the
first two rankings were considered, and 74% when the first three rankings were
considered. This increase suggests that
the verbal protocols may have reinforced the initial scenario rankings.
To conclude, the significance of the
retrospective verbalizations will be examined.
First, when the dominant dimensions, excluding the borrowed
characteristics (DDX), was compared to DD, the level of agreement was 35% on
the first ranking, 59% on the first two
rankings, and 79% when the first three most dominant rankings were
considered. Likewise, when the
retrospective verbalizations were included in the overall calculations, the
category rankings were 53%, 79%, and 85%.
Then, when DDI was compared to the dominant dimension, involving only
concurrent verbalizations, the level of agreement was 44% on the first ranking,
65% on the first two, and 79% on the first three rankings. And finally, when the retrospective comments
were added to the overall calculations, the rankings were 91%, 94%, and 97%,
respectively. Again, these latter
results suggest a pattern similar to previous conclusions, namely, that the
borrowed characteristics added stability in determining the dominant dimension
of Gardner’s intelligences. Of greater
relevance for this qualitative study was how the percentages increased when the
retrospective verbalizations were included with the earlier concurrent
comments. In other words, the three
highest rankings of the subjects represented a strong indicator when selecting
their overall dominant dimensions.
----------------------------------
In the
original, Table 4 was about here
----------------------------------
Graphing Intellectual Profiles
During the second home visit, the first author presented a
graphic profile of the above results to each subject. Due to space limitations, only one profile (see Figure 1) will be
shown. The percentages, located over
and under the vertical bars, indicate a relative frequency within each
intellectual dimension. For example,
subject 30 perceived the Solidarity-Intrapersonal dimension (see Appendix B) as
the dominant (75%) intelligence; conversely, the Musical-Rhythmic dimension was
perceived as the least preferred (-54%) intelligence.
-----------------------------------
In the
original, Table 5 was about here
-----------------------------------
Discussion
In this concluding section, four items will be discussed. Following remarks addressing the research
question, comments will be offered on the advantages of conducting interviews
in the homes of the subjects. Then,
some educational implications will be examined. The paper concludes by suggesting additional research in this
area.
Research Objectives
This qualitative study involved homeroom teachers who nominated
students considered dominant in one of Gardner’s seven theorized
intelligences. A series of verbal
reports, considered as field data, were gathered as the subjects self-assessed
their intellectual dimensions. In
short, they compared themselves to a series of scenarios. The scenarios utilized a common framework,
presenting seven idealized grade eight students within the confines of typical
school life. An analysis of their out
loud verbalizations suggested that they could produce their intellectual
profile, in fact, with considerable ease.
They enjoyed and appeared confident when asked about the ranking
process; they found the exercise to be quite a challenging task.
An analysis of both concurrent and retrospective out loud
student verbalizations indicated a strong level of agreement between the
teacher nominations and the three dominant intelligences as self-perceived by
the students. In other words, the
homeroom teachers seemed able to pinpoint in their students one of Gardner’s
intelligences. Verbal protocol analysis
appeared to be a meaningful way of assessing intellectual profiles. Or restated as a question, did the
retrospective out loud comments from the subjects reinforce the earlier and
concurrent verbal protocols voiced during the home interview? As outlined, an analysis of the field data
from the transcripts provided evidence that such retrospective statements
seemed to reinforce initial student comments.
Having stated such tentative conclusions, the jury is still out
... and may be out for some time on the complex ways in which student
intelligences might be organized within different intellectual domains, and on
the interaction between such possible domains.
Nevertheless, the attractiveness of Gardner’s MI model is that it
provides educators, especially classroom instructors, with a theoretical basis
for stating something that they may have already hoped would be true: that their students exemplifying different
skills, regardless of the frame of mind: be it language or logic, art or sport
or music, communication with others or self-reflection, are really exemplifying
just one of their “intelligences.”
Home Interviews
Most subjects commented that they felt privileged to participate
in the pilot study. Specifically, they
seemed to enjoy solving the ranking problem, in that it was not a test and
that, for once, there were “no wrong answers.”
Some did however state that it was not an easy task to “look inside
themselves” as they attempted to self-rank their intellectual profile. Others had difficulty with the “borrowed”
characteristics. As was expected, the
insertion of additional characteristics forced the subjects to qualify their
rankings with additional verbalizations.
All subjects expressed a strong interest in this intellectual form of
self-analysis. This was the first time
that they were ever exposed to this interview format, or to paraphrase Das
(1992), Sternberg (1985a), or Gardner (1987), that they were possibly ever
exposed to an intellectual approach beyond IQ.
The parents of the subjects were most interested in the
research objectives, and more importantly, quite pleased that this form of
intellectual investigation was being conducted at this point in time. In particular, many mothers and fathers
commented on the inferior quality of current state-funded public schooling,
especially assessment tools used to tap intellectual accomplishments. They questioned service delivery models
currently in practice for the intellectually gifted child and were in agreement
that Gardner’s model showed intellectual “promise”. Simply put, it was an enjoyable task for the first author to sit
down in the homes of the research subjects and, in many cases, to be present
with both parents, in an informal and relaxing manner, responding to their
valid cerebral concerns. Of greater
interest was the fact that almost all of the parents were pleased to see that
something was being done in the line of seeking broader assessment instruments
to measure intellectual potential within the educational community.
Twelve parents suggested that they wished that they had been
permitted to complete a nomination checklist, similar to the checklist
completed by their son’s / daughter’s grade eight homeroom teacher. They felt that such input might have
improved the validity to the overall exploratory project. On a more personal note, all parents were
most hospitable to the first author who invited himself into their Kanata,
Ontario homes. To sum, the home
interview procedure was considered a positive medium for gathering raw field
data. Subsequent qualitative studies of
this nature would benefit from similar home involvement.
Educational Implications
Many students are unaware of their inner talents, skills,
competences, or to coin Gardner’s term, “intelligences.” They often perceive themselves as
educational “washouts.” Such “talented”
youngsters sometimes fail to realize that they may indeed have a learning
strength in, at least, one of Gardner’s dimensions. Such students frequently appear discouraged and withdrawn. At times, they even seem to become
aggressive and rebellious, perhaps to mask their low self-esteem, due, in part,
to their ignorance of their other "intelligences". It seems that, in many cases, public
schooling practices tend to focus on the academic weaknesses of students at the
expense of developing some of their other strengths, or talents ... a process
so often witnessed over the years by both authors.
Educators, especially those directly responsible for assessing
intellectual performances, must retrain themselves into rethinking that
intellectual functioning in students is not a unitary “frame of mind” as
traditionally assumed. Instead, they
must intrinsically feel and believe that the minds of their students may have
many “frames”, or pluralistic as to domain-specific contents. Then, these educators must continue research
studies, more advanced than what was described in this paper. They must, in close collaboration with their
classroom colleagues, devise statistically reliable and valid assessment
instruments which will measure and train broader forms of intelligences.
Conclusion
This investigation was the first, to the knowledge of the
authors, to conceptualize Gardner’s MI model as the source for a
problem-solving activity for grade 8 students and utilizing verbal protocols as
a methodology. A thorough search of the
literature associated with the above descriptors failed to reveal any research
studies similar to what was discussed.
Or, to put it another way, the methodology was highly exploratory in
nature. Thus any conclusions presented
here are highly speculative. They must
be treated with extreme suspect, until replication with different samples and
other populations have been clarified.
Nonetheless, Gardner’s conceptualization of intelligence deserves
consideration.
The results as outlined might appear to the critical reader
simply as bold generalizations. True, they are subject to additional
empirical validation. Nevertheless, the above approach represent a
step towards profiling student intelligences. Perhaps more important
than any formal empirical findings, whether statistically reliable and valid, is
that perhaps for the only time in their lives, these grade 8 students were
singled out and respected for things they liked doing, their supposedly dominant
intelligences (Morris, 1992a, 1992b, 1993a, 1993b,1993c). Such attention
to students in the long run may contribute more to their success in life than
equal efforts to remediate their learning limitations (Sternberg,
Okagaki, & Jackson,1990). The qualitative research study outlined
in this report represented the initial step in providing students with a chance to
see their own possible intelligences. In the words of the late and
great Zurich psychologist, Dr. Carl Gustav Jung, "The creative mind plays with
things it loves" (Lincoln & Suid, 1986).
Acknowledgement
Clifford Morris wishes to acknowledge the cooperation of the (former) Carleton
Board of Education in the pursuit of this Ph. D. (pilot) study. Moreover,
he wishes to thank the parents of the children (n=34) from Kanata, Ontario,
who made their homes available for interviewing their children who
were the research subjects in this multiple intelligences (MI) investigation.
References
Armstrong, T.(1988, September). Learning Differences - Not Disabilities. Principal, 68(1), 34-36.
Armstrong, T.
(1993). Seven Kinds of Smart:
Identifying and Developing Your Many Intelligences. New York: Penguin Group.
Armstrong, T.
(1994). Multiple
intelligences in the classroom.
Alexandria, Virginia: Association for Supervision and Curriculum
Development.
Blythe, T. &
Gardner, H. (1990, April). A school for all intelligences. Educational Leadership, 47(3),
33-37.
Carroll, J.
B. (1982). The measurement of intelligence.
In R. J. Sternberg (Ed.), Handbook of human intelligence (pp.
29-120). New York: Cambridge University
Press.
Ceci, S. J. (1990).
Framing intellectual assessment in terms of a person-process-context
model. Educational Psychologist,
25(3 & 4), 269-291.
Das, J. P. (1992, April-June). Beyond a unidimensional scale of merit. Intelligence, 16(2), 137-149.
Davidson, J. E. (1990).
Intelligence recreated. Educational
Psychologist, 25(3 & 4), 337-354.
De Groot, A. D.
(1965). Thought and choice in chess. The Hague: Mounton.
Ericsson, K.
A., & Oliver, W., L,
(1988). Methodology for
laboratory research on thinking: Task selection, collection of observations,
and data analysis. In R. J. Sternberg
& E. E. Smith (Eds.),
The psychology of human thought (pp. 392 - 428). Cambridge: Cambridge University Press.
Ericsson, K. A., & Simon, H. A. (1980, May). Verbal reports as data, Psychological review, 87(3),
215-251.
Ericsson, K. A., & Simon, H. A. (1984).
Protocol analysis: Verbal reports as data, Cambridge, MA: MIT Press.
Ericsson, K. A., & Simon, H. A. (1993).
Protocol analysis: Verbal reports as data: Revised Edition. Cambridge, MA: MIT Press.
Eysenck, H. J. (1981).
The intelligence controversy.
New York: John Wiley.
Faggella, K. &
Horowitz, J. (1990, September). Different child, different style. Instructor, 100(2), 49-54.
Fodor, J. (1983).
The modularity of mind: An essay on faculty psychology. Cambridge: MIT Press.
Fodor, J. A.
(1985, March). Précis of the
modularity of mind. Behavioral and
Brain Sciences, 8(1), 1-42.
Freeman, F. (1950).
Theory and Practice of Psychological Testing. New York: Henry Holt and Company.
Gardner, H.
(1982, September). Giftedness:
Speculations from a biological perspective.
New Directions for Child Development, 17, 47-60.
Gardner, H.
(1983, 1993). Frames of mind:
The theory of multiple intelligences.
New York: Basic Books.
Gardner, H, (1987, May). Beyond the IQ: Education and human development. Harvard Education Review, 57(2),
187-193.
Gardner, H.
(1989). To open minds:
Chinese clues to the dilemma of contemporary education. New York: Basic Books.
Gardner, H.
(1990, Spring). The difficulties
of school: Probable causes, probable cures.
Dædalus: Journal of the american academy of arts and sciences. 119(2), 85-113.
Gardner, H. (1991).
The unschooled mind: How children think and how schools should teach. New York: Basic Books.
Gardner, H.
(1993). Multiple
intelligences: The theory in practice.
New York: Basic Books.
Gardner, H. (1995).
Reflections on multiple intelligences: Myths and messages. Phi Delta Kappa, 77(3),
200-209.
Gardner, H.,
Kornhaber, M. L., & Wake, W. K.
(1996). Intelligence:
Multiple Perspectives. Orlando,
FL: Harcourt Brace College Publishers.
Gould, S.
J. (1981). The mismeasure of man.
New York: W. W. Norton.
Hatch, T., & Gardner, H. (1986, February). From
testing intelligence to assessing competencies: A pluralistic view of
intellect. Roeper Review, 8(3),
147-150.
Hayes, J. R.
(1981). The complete problem
solver. Philadelphia: Franklin Institute
Press.
Horne, J. L. (1979).
Trends in the measurement of intelligence. Intelligence, 3, 229-240.
Howard, V. A. (Ed.).
(1990). Varieties of
thinking: Essays from Harvard’s Philosophy of Education Research Center. New York: Routledge.
Hunt, E.
(1990). A modern arsenal for
mental assessment. Educational
Psychologist, 25(3 & 4), 223-241.
James, G. A.
(1991). Different strokes for
different folks: Personal learning styles.
Today's Parent, 8(1), 19-21.
Jensen, A. R.
(1982). The chronometry of
intelligence. In R. J. Sternberg (Ed.),
Advances in the psychology of human intelligence (Vol. 1). Hillsdale, NJ: Erlbaum.
Kolligian, J. Jr., & Sternberg, R. J. (1987).
Intelligence, information processing, and specific learning
disabilities: A triarchic synthesis. Journal
of Learning Disabilities, 20, 8-17.
Kornhaber, M.,
Krechevsky, M., & Gardner, H.
(1990). Engaging
intelligence. Educational
Psychologist, 25(3 & 4), 177-199.
Lincoln, W. & Suid, M. (1986). Quotations for
teachers. Palo Alto, CA: Dale Semour.
Maker, C. J. (1992, Fall). Intelligence and creativity in multiple intelligences:
Identification and development. Educating
Able Learners. 17(4), 12-19.
Maker, C. J., Neilson,
A. B., & Rogers, J. A. (1994,
Fall). Giftedness, diversity and
problem-solving: Multiple intelligences and diversity in educational
settings. Teaching exceptional
children. 27(1), 4-19.
Matthews, D. (1988, December). Gardner's multiple intelligence theory: An evaluation of relevant
research literature and a consideration of its application to gifted education.
Roeper Review, 11(2), 100-104.
Morris, C. (1992a, Sept.
/ Oct.). Gardner’s multiple
intelligences in our classrooms: Our students are smarter than we think: Part
I Teaching Today, 11(1),
25-26.
Morris, C. (1992b, Nov.
/ Dec.). Gardner’s multiple
intelligences in our classrooms: Our students are smarter than we think: Part
II Teaching Today, 11(2),
27-28.
Morris, C. (1993a, Jan.
/ Feb.). Gardner’s multiple
intelligences in our classrooms: Our students are smarter than we think: Part
III Teaching Today, 11(3),
25-26.
Morris, C. (1993b,
June). A qualitative analysis of
Gardner’s dimensions for intermediate students. Paper presented at the annual meeting of The Canadian Society for
the Study of Education, Carleton University, Ottawa.
Morris, C. (1993c, December). Perceived Cognitive Interests by Grade
Eight Pupils within Howard Gardner’s Framework. Paper presented at the 35th Annual Conference of The Ontario
Educational Research Council, Ontario Institute for Studies in Education,
Toronto, Ontario.
Newell, A., & Simon,
H. A.
(1972). Human problem solving.
Englewood Cliffs, N. J.: Prentice-Hall.
Olson, L. (1988, January 27). Children flourish here: 8 teachers and a
theory changed a school world, Education
Week, 7(18), 1, 18-19.
Piaget, J.
(1971). The psychology of
intelligence. London: Routledge & Kegan.
Resnick, L. B. (1976).
Changing conceptions of intelligence.
In L. B. Resnick (Ed.), The
nature of intelligence (pp. 1-10).
New York: John Wiley and Sons.
Resnick, L. B. (1979).
The future of IQ testing in education.
Intelligence, 3, 241-253.
Rowe, H. A. H.
(1985). The development of a
classification system of problem solving strategies. In H. A. H. Rowe (Ed.), Problem
solving and intelligence (pp.
118-148). Hillsdale, NJ: Erlbaum.
Schael, J., & Dionne, J. P. (1991, April). Judges’ agreement and disagreement patterns when encoding
verbal protocols. Paper presented
at the Annual Meeting of the American Educational Research Association,
Chicago, IL. (ERIC Document
Reproduction Service No. ED 334 209)
Scherer, M. (1985, January). How many ways is a child intelligent? Instructor, 94(5), 32-35.
Schoenfeld, A.
H. (1985). Verbal data, protocol analysis, and the
issue of control. In A. H. Schoenfeld
(Ed.), Mathematical problem solving
(pp. 270-344). New York:
Academic Press.
Sternberg, R. J. (1979).
Six authors in search of a character: A play about intelligence tests in
the year 2000. Intelligence, 3,
283-293.
Sternberg, R. J.
(1985a). Beyond IQ: A triarchic
theory of human intelligence,
Cambridge: Cambridge University Press.
Sternberg, R. J. (1985b, December). Human intelligence: The model is the message. Science, 230, 1111-1118.
Sternberg, R. J. (1986a, February). Identifying of the gifted through IQ: Why a little bit of
knowledge is a dangerous thing. Roeper
Review, 8(3), 143-147.
Sternberg, R. J. (1986b, March-April). Inside intelligence. American Scientist, 74(2),
(137-143).
Sternberg, R. J. (1991).
Death, taxes, and bad intelligence tests. Intelligence, 15(3), 257-270.
Sternberg, R. J.,
Okagaki, L., & Jackson, A. S.
(1990, September). Practical
intelligence for success in school. Educational
Leadership, 48(1), 35-39.
Sternberg, R.
J., & Salter, W., (1982).
Conceptions of intelligence. In
R. J. Sternberg (Ed.), Handbook of
human intelligence (pp. 3-28).
Cambridge: Cambridge University Press.
Sternberg, R. J., &
Smith, E. E., (Eds.). (1988). The psychology of human thought. Cambridge: Cambridge University Press.
Turnbull, W. W. (1979).
Intelligence testing in the year 2000.
Intelligence, 3, 275-282.
Vail, P. L. (1987). Smart kids
with school problems: Things to know and ways to help. New York: Penguin Books.
Wechsler, D.
(1958). The measurement and
appraisal of adult intelligence.
(5th ed.). Baltimore: Williams
& Wilkins.
Wechsler, D.
(1974). Wechsler intelligence
scale for children -- revised. New
York: Psychological Corporation.
Appendix A: Teacher Nomination Checklist
A Verbal / Linguistic Student
- is acutely sensitive to the
meaning of words, their order, sounds, rhythms, and inflections
- can freely grasp different
functions, phonology, syntax, semantics, and pragmatics of language(s)
A Logical / Mathematical Student
- enjoys responding to
inductive and deductive thinking and reasoning exercises
- can easily recognize
complex scientific patterns, including abstract mathematical relationships
A Visual / Spatial Student
- uses initial perceptions
accurately when perceiving the visual world accurately
- can easily re-create
aspects of one’s visual experience, even in the absence of relevant stimuli
A Bodily / Kinesthetic Student
- uses the body in highly
differentiated and skilled ways, for goal-directed and expressive purposes
- works well with objects
needing fine-motor finger and hand movements to gross-motor actions
A Musical / Rhythmic Student
- is most sensitive to
melody, rhythm, timbre, and the emotional aspect of sounds
- can easily recognize beats,
pitch, tonal patterns, including various environmental sounds
A Social / Interpersonal Student
- knows others well by
recognizing faces, voices, and persons; can react quickly to their wants
- reads easily the social
signals of others, and comprehend their motives, feelings, and intentions
A Solidarity / Intrapersonal Student
- is most sensitive to one’s
own feelings, wants, fears, and personal history
- is aware of and can respond
to one’s strengths and limitations with considerable grace and ease
Note:
Due to space limitations, the
characteristics from the original “Teacher Nomination Checklist” were
summarized. Interested readers may
contact the first author for the complete list.
Appendix
B: Gardner Scenario Cards
Scenario 1 Verbal / Linguistic
Susan,
the Word Player
Susan is a grade 8 student in a French
Immersion program. She really likes
reading and writing about the biographies of famous people. During many of her English classes, she
often tells jokes and repeats tricky tongue twisters. She is an excellent piano player who enjoys daily practicing
(+5) (see Note 1 below). Also, Susan likes
to play word games that involve many directions. She does not enjoy playing any kind of sports (-4). She likes to memorize poems and tell
long detailed stories. On many
occasions, Susan will not listen to what her teacher tells her to do (+7).
Scenario 2 Logical / Mathematical
William,
the Questioner
William likes to solve difficult arithmetic
problems and figure out complicated designs in geometry. He is not very good in music class, as he
can not remember how to sing songs (-5).
He is always asking those very difficult questions. He is very good at doing arithmetic in his
head, and he is one of the best students in the computer class. In Science class, he is often the first student to understand and complete
the science experiment. William
also likes to draw and sketch scenes
(+3). During gym classes, he is
known by his classmates and teacher as having strong leadership skills; he is
often selected to be a team captain. William seems to have a special way of knowing how to handle the players on his team
(+6).
Scenario 3 Visual / Spatial Chris,
the Visualizer
Chris is the school's chess champion. He often daydreams in class. He appears happiest when his science teacher
allows him to go to his own quiet corner (+7) to design and then invent
gadgets using construction sets such as "Leggo", "Meccano",
or "Gearopolis". He also
likes "to express himself" through dance, drama, and movement
(+4). Also, he seems to learn best when
allowed to sketch or draw cartoon characters.
At times, he seems to learn more through visuals, that is, when pictures
accompany words, as in the game of "Pictionary". Chris finds it easier to understand the
information in a subject when his teachers writes short notes on the
chalkboard. He does not like to read
books, tell stories, or write short stories (-1).
Scenario 4 Bodily / Kinesthetic Paul,
the Mover
Paul plays the center position on the
school hockey team and the end position on a community football team. He also is good at tennis, skiing, and
swimming. He enjoys figuring out how
things work with numbers (+2).
During a recent Christmas concert, he performed a break dance. However, he does not like to memorize
poems. He finds it difficult to discuss
out loud any topics in English class; he has problems remembering
(-1). A member of the Innovator's Club,
he enjoys inventing gadgets by taking apart and reassembling small machines
such as food grinder, old clocks and broken radios. Finally, Paul seems to be able to understand how many of the
other students in his class "think" (+6).
Scenario 5 Musical / Rhythmic Nicola,
the Music Lover
Often referred to by her classmates as a
"music lover", Nicola learns best when surrounded by melody and
rhythm. She can be found singing,
humming, or whistling tunes to herself.
She often gets into fights with other grade 8 students; thus, she has
a real problem getting along with others (-6). She likes poetry, songs, and composing melodies. She enjoys the printed word, as she loves
to read books and play word games such as Scrabble (+1). She is most sensitive to sounds around her
-- sounds such as distant bells ringing or crickets chirping. In fact, she often hears things missed by
her classmates, and even by her music
teacher. She often daydreams
in class, pretending that she's a magical being in fabulous adventure stories
(+3).
Scenario 6 Social / Interpersonal Sara,
the Socializer
Elected by her classmates as the class
president, Sara recently become absorbed with her older brother's computer;
she especially loves to experiment with the word processor (+1 &
+2). She enjoys many team and
individual sports; in fact, she was just
selected by her coach as captain of the community's softball team
(+4). She seems to be so
"street-smart". She loves to
be involved in class plays, skits, and classroom planning committees. When someone cries during class, Sara can
usually be found at that person's side correctly identifying and soothing her
emotions. She does not like to
participate in art class, as her sketches, drawings and paintings are usually
poorly-formed (-3).
Scenario 7 Solidarity / Intrapersonal Jeff, the Individual
Jeff is that type of grade 8 student
who prefers being left alone to work; however, he is not afraid to ask others
for help. He is a quiet student, has
few friends and seems to know himself very well. He enjoys drawing sketches of imaginary characters
(+3). He has difficulty figuring out
the answers to arithmetic problems (-2).
He often dresses, acts and behaves in his own way. He enjoys a private place to pursue his own
personal hobbies and ongoing interests.
In fact, Jeff's teacher was amazed when he recently took apart the clock
that was not working and put it back together. She was even more surprised when it worked (+4).
Note 1
All
underlines, brackets and positive and negative numerals were not in the
version that the first author showed the research subjects.
They have been deliberately inserted here to show the more interested reader how the
process of “borrowed characteristics”
was operationalized. For example, in
Scenario 1, the “+” in the “+5” indicates that that characteristic
was of a positive nature; the “5” in
the “+5” indicates that the meaning
for that added characteristic was borrowed from Scenario 5.
About the Authors
At the time of this writing (Spring of 1996), Clifford Morris was a doctoral candidate in the Faculty of
Education at the University of Ottawa. He was also a special education teacher at Century Public
School with the former Carleton Board of Education. His area of interest was student intellectual
development. Raymond LeBlanc was a professor in the Faculty of Education at the University of
Ottawa. His domain of expertise was special education.
Revised on
Sunday, 19 March, 2006. ~