116

The Reading Matrix

Vol. 6, No. 2, September 2006

STUDENT WRITING, PERSONALITY TYPE OF THE STUDENT

AND THE RATER: ANY INTERRELATIONSHIP?
Fahimeh Marefat
[email protected]

Abstract

The way we learn is very much affected by our personality. Practitioners have
proposed that an understanding of personality type can help teachers explain why
students approach tasks differently: some are successful, while some fail to
participate in class activities (Oxford & Ehrman, 1990; Wilz, 2000). Meyers-
Briggs’s theory, anchored in Jung’s work, introduces four different character
types: Introvert/Extrovert, Sensitive/Intuitive, Thinking/Feeling, and
Judging/Perceiving. The Meyers-Briggs Type Indicator (MBTI), a 93-item paper-
and-pencil inventory, helps, as a reliable instrument, identify students’ personality
types. The current study aims at discovering the relationship, if any, between
learner personality type and his writing ability in the first place and then between
rater personality and his rating procedure. Eighty-six male and female graduate
and undergraduate EFL students and their teacher who rated their essays
participated in this study. The average of each learner’s scores on two in-class
writings, as well as midterm and final exams served as an index of his writing
ability. The participants were also asked to fill out the MBTI questionnaire with
two options for each item. Individuals were classified on the basis of their self-
reported preferences. Analysis of data indicated that the only dimension showing
significant impact across writing ability was the S/N preference. Surprisingly, a
link was observed between rater personality and her rating procedure.
__________________

"We, ignorant of ourselves, Beg often our own harms.” (Shakespeare, Anthony & Cleopatra)

Why bother learning about personality types? People differ from one another
depending on the way they perceive the world. In fact, our personality affects the way we
learn. Practitioners have proposed an understanding of personality type (how we interact
with the world and where we direct our energy, the kind of information we naturally notice,
how we make decisions) can help explain why we learn differently (Ehrman & Oxford,
1990; Ehrman & Oxford, 1995; Ehrman, 1994; Wilz, 2000). According to Ehrman and
Oxford, studies investigating psychological types are promising in that they offer “an
accessible conceptual framework for language trainers and learners …. greater self-
regulation and better learning performance” (1990, p. 324). Learners can actually move out
of their “comfort zone” and try other preferences, like hand preferences. In line with others,
Wilz (2000) expresses the dire need for personality type understanding on the part of the
teacher:

An awareness of student personality types allows teachers to have a better

understanding of the classroom dynamics and to be better able to determine what
 117

kinds of classroom activities and strategies would be most effective with a

majority of students in the class (p, 29).

Inspired by such studies that underscore investigation into learner characteristics, the
present study examines the relationship between psychological type as measured by the
Myers–Briggs Type Indicator (MBTI) to writing ability. This study aims to shed light on the
way personality preferences interact with writing ability. Would the preferences influence
the individual expressing himself? Would, for instance, someone connected with the
introvert pole write better than the person being classified under the extrovert pole? It is also
the aim of this research to see if any trace can be found in the rating process as regards
learner and rater personality type.

General characteristics of the MBTI Inventory: People are unique individuals and are
born with preferences. The MBTI instrument is used to understand personality differences;
it describes various behavior patterns that in turn affect the way we function in the world.
The intention, being noble, is to help people understand themselves and others better.
Primarily, the inventory was used in workplaces. It was reasoned if individuals learned what
vocation best matched their personality type, then they would enjoy their jobs, feel happier,
more productive, and more creative. And this would automatically make life more peaceful
(Myers & Briggs, 1998).
Depending on whether the individual shows a preference for the outer world or is energized
by an inner world of ideas, he is extrovert or introvert (E/I), respectively. Whether the
individual processes the data concretely or abstractly classifies him as sensing or intuitive
(S/N). Next comes decision making. If the method adopted is more of a logical and
objective fashion, then thinking is the dominant character type; if, on the other hand,
decision is made more subjectively, the personality type is feeling (T/F). It is interesting to
note that this dichotomy has been shown to be affected by gender. And finally the way you
organize the outside world determines whether you are a judging or perceiving person (J/P),
the former being more systematic while the latter more random.

An advantage of this inventory is that type is not “reductive”. The classification is

merely for the sake of practical purposes, it is not meant to “pigeonhole people.” Rather it
reminds one of “the increased respect for the complexity of human nature” (Callahan, 2000,
p. 61). In fact, no one type is preferred over the other. All are “equally valuable.” Ehrman
and Oxford (1995) report split-half reliability of .87 and test-retest reliabilities are, .70 - .85.
Construct validity of the inventory is supported by many studies of occupational preferences
and creativity (see Ehrman & Oxford, 1995, for details).

Past Empirical Research: In a qualitative study in 1990 and using the MBTI,
Ehrman and Oxford worked with 20 Foreign Services Institute (FSI) students. Their study
showed “some language learning advantage for introverts, intuitives, feelers, and
perceivers” (p.323). In a follow-up study in 1994 on 831 FSI students, Ehrman found that
“introverts, intuitives, and thinkers were better readers. Sensing types were disadvantaged
for both reading and speaking.” A subsequent study by Ehrman and Oxford (1995)
suggested that extroverts are good candidates for good language learners as they speak out
and interact.
From among the studies aiming at highlighting teacher role in strategy training two are
reported here. Hismanoglu (2000) emphasizes the importance of language learning
strategies in foreign language learning and teaching. He defines the concept of language
learning strategy, gives a summary background on language learning strategies, , and
 118

outlines the taxonomies of language learning strategies proposed by several researchers. He

also takes into account the teacher's role in strategy training.
Zhenhui (2001) highlights the significance of the degree of congruence between
teacher and learner styles. The study suggests the teacher to take into consideration learner
preferences and to teach learners to employ different and various learning styles,
“encouraging students to diversify their learning style preferences.” This way the learners
will become more effective.
Giving it a Computer Mediated Communication (CMC) flavor, Beauvois and Eledge
(1996) were interested to qualitatively and quantitatively analyze the attitude of the
university students, as determined by the MBTI, toward the use of CMC in their French
course. Both introvert and extroverts, the study describes, found the experience of using
local area network (LAN) beneficial.

Personality type and writing preference: To delve into the relationship between
student reflective writing and teacher feedback, Callahan (2000) conducted a study with
student teachers. Using the MBTI inventory, she chose as raters three students whose types
completely differed from hers. They read other participants’ reflective writings and tried to
identify the writers’ types. Analysis of the data revealed that as writers students need to go
beyond their own preferences and familiarize themselves with other options. The readers
who comment on student writing with eagerness can have a pivotal role in helping students
“build upon their won preferences and develop their less preferred approaches” (Callahan,
2000, p. 72).

As the current study is concerned with student writing and his MBTI index, below
comes an interesting brief analysis of the relationship between reflective writing and
character types as depicted by Callahan (2000).

Those with E preferences best respond to reflecting about the outer world. As one
might expect, they are better talkers than writers and so do not go for keeping journals or
preparing portfolios, where metacognitive processes are involved. In a word, for these
students, reflecting on their writing processes seems “awkward.” Moreover, this camp
wishes the teacher to set goals for them. Reflection for students with I orientation is
pleasant and quite “natural.” Their journals tend to be quite “voluminous.” As opposed to
their E counterparts, the I group finds setting goals and standards an interesting task. As for
the teacher, the dynamics of class work is greatly affected by teacher personality type.
Teachers with E preferences would not choose to assign reflective writing, while the
teachers with I preference would include and even emphasize such assignments.

The written product of students with an S preference is lengthy and detailed. Such
individuals find reflective writing an opportunity to go back and see if they have missed
anything. In contrast, Ns find “reading between the lines” and metaphoric use of language a
fun. Interestingly, N writers put their readers in situations to discover the unstated details.
An S teacher would certainly reward specific elements, while a teacher with too much
reliance on intuition must appreciate concrete thinking, too.

As regards the T/F dichotomy, it can be said that the T group is interested in
describing their strengths and weaknesses in writing. This group, as opposed to the F group,
reveals rich notions in their protocols that would otherwise remain hidden. If asked to name
some elements of successful writing, the T group’s response would be “organizational
 119

patterns” and “rhetorical features” while the F people would be excited by a piece that
evokes a strong feeling. Teachers with a dominant T preference would welcome impersonal
reflective writings whereas those teachers with F preferences would comment more on the
thinking quality of the writer.

With J/P preferences in mind, the reader would agree that the first group would set
goals for future improvement easily; they offer tidy, organized projects. In contrast, the
latter would resist explorations on their future planning and find it difficult to draw
conclusions. In fact, their work is always in progress. A judging teacher would certainly
appreciate an organized, neat portfolio handed to her as an end-of-the-semester assignment;
a teacher with a P preference would find a last-minute reflective writing invaluable.

Research Questions: A study attempting to characterize student and teacher character

types would clearly shed light on aspects that are not directly visible to the teacher. Aiming
at discovering these new horizons, the present research addressed the following questions:

1. Is there any relationship between student personality type and his writing ability?
2. Is there any relationship between rater and student personality type?

Participants: In the spring semester of 2003-4, and fall semester 2004-5, 42 male and
44 female EFL students at undergraduate and graduate levels doing their advanced writing
or essay writing courses took part in this study.
Writing ability: The average of each learner’s scores on two in-class writings as well
as midterm and final exams was used as an index of writing ability. The writing scores
ranged between 12 and 19.25.
Instrument: The Myers-Briggs Type Indicator Form M is a 93-item, paper-and-pencil
inventory. There are two options for each item. Individuals are classified on the basis of
their self-reported preferences. The Persian version of the questionnaire as translated and
validated by Hoseini (2003) was used in this study. Below are offered two sample items (the
original and the Persian translation) to give the reader a feeling of the questionnaire.
PART I. Which answer comes closest to describing how you usually feel or act!
1. When you go somewhere for the
day, would you rather …
_ plan what you will do and when, or
_ just go?
Part III. Which answer comes closest to describing how you usually feel or act!
59. When you start a big project that is due in a week, do you …
_ take time to list the separate things to be done and the order of doing them,
_Or plunge right in?

Procedure: The first phase of the survey was completed after the final exam scores
were announced. As it was summer time and there was no access to students, the inventory
was sent to them via email along with a letter giving them some tips as to how to fill out the
questionnaire.
For the second phase, the students were asked to fill out the forms in the class. The
researcher was available for any question. Based upon the directions available in the MBTI
form, the students were instructed not to spend too much time on one question if they were
not sure about it. They could skip the question and return to it later. Also, they were assured
that the information about their character type would be treated as confidential.
 120

When I reported them their type along with a short descriptor interpreting the four-letter
types, a considerable percentage, via email or in person, expressed their surprise at how
closely the profile had reflected their real selves.

Data Analysis: The personality inventories completed, the next step was to determine
which dimensions of personality were more dominant. The independent variables were the
four indices of the MBTI: extroversion-introversion (E/I), sensing-intuitive (S/N), thinking-
feeling(T/F), and judging-perceiving(J/P). Each individual was assigned to either level of
each independent variable based upon the scoring instructions contained in the MBTI
manual.

Findings: Firstly, a summary of descriptive statistics that would help the reader to
appreciate the coming tables is offered. The frequencies for 8 possible types are depicted in
table 1. Tendencies in this sample are towards I, S, T, and J.

Table 1. Frequencies of personality types

Type E I S N T F J P
Frequency 38 48 48 38 55 31 65 21

Surprisingly, with this small population, all 16 conceivable type combinations were
observed, though the frequency in two cases was as low as 1. The majority of the
participants (19.8%) had ISTJ preference and ESTJ students ranked second with a
percentage of 12.8%. ISTPs and INFPs had the lowest frequency, 1.2 percent. It is
interesting to note that the highest mean (17.41) was observed among the INTJs, with the
third highest frequency of 9.3%. Table 2 summarizes the type combinations in the study.

Table 2. Descriptive statistics for the writings of the 16 type combinations

Type Mean SD Number Percent

ESTJ 15.57 2.04 11 12.8
ESTP 15.26 1.41 2 2.3
ESFJ 15.54 2.47 6 7.0
ESFP 16.50 4.24 2 2.3
ENTJ 15.63 2.32 8 9.3
ENTP 16.87 1.24 2 2.3
ENFJ 16.62 .17 2 2.3
ENFP 16.60 1.31 5 5.8
ISTJ 16.26 1.61 17 19.8
ISTP 12.50 0 1 1.2
ISFJ 15.46 2.66 8 9.3
INFP 17.00 0 1 1.2
INTJ 17.41 1.53 8 9.3
INTP 16.37 1.73 6 7.0
INFJ 16.65 1.54 5 5.8
ISFP 17.13 2.30 2 2.3
 121

Addressing the first research question, whether any relationship can be established
between the personality types and writing ability, a factorial analysis was conducted with the
writing score as the dependent variable and the four MBTI indices as the independent
variables. Results, as shown in table 3, revealed no main effect for any of the factors. Though
not significant, sensing/intuitive is the best candidate for the likely effect. In fact, sensing
students had different writing abilities than those with intuition, the means being 15.79 and
16.58, respectively.

Table 3. Factorial ANOVA Results

Source SS DF MS F Sig.
EI .0072 1 .007 .002 .966
SN 14.661 1 14.661 3.732 .057
TF 5.537 1 5.537 1.409 .239
JP .148 1 .148 .038 .846
EI*SN 2.076 1 2.076 .528 .470
EI*TF .531 1 .531 .135 .714
EI*JP 3.763 1 3.763 .958 .331
SN*TF 3.076 1 3.076 .783 .379
SN*JP .889 1 .889 .226 .636
TF*JP 8.039 1 8.039 2.046 .157
EI*SN*TF 1.792 1 1.792 .456 .502
EI*SN*JP .207 1 .207 .053 .819
EI*TF*JP 8.032 1 8.032 2.044 .157
SN*TF*JP 6.983 1 6.983 1.778 .187
EI*SN*TF*JP .266 1 .266 .068 .795
ERROR 275.00 70 3.929
TOTAL 2715.18 86

Next, to discover whether any difference existed between the extroversion/introversion

pair and other pairs, 4 paired t-tests were run. Group means and standard deviations are
presented in Table 4.

Table 4. Paired T-test results for the four dichotomies (males, females and total)
Male (df=40) Female (df= 42) Total(df=84)
Type Freq. Mean Sig Type Freq. Mean Sig Type Freq. Mean Sig
E 13 15.33 .907 E 25 16.15 .099 E 38 15.87 .499
I 29 16.03 I 19 16.76 I 48 16.35
S 24 15.23 .335 S 24 16.34 .146 S 48 15.79 .012
N 18 16.67 N 20 16.5 N 38 16.58
T 26 15.88 .983 T 29 16.34 .399 T 55 16.13 .442
F 16 15.78 F 15 16.55 F 31 16.15
J 31 15.85 .938 J 34 16.30 .079 J 65 16.08 .665
P 11 15.84 P 10 16.8 P 21 16.30
 122

As is clear from table 4, females were more of the E type while most males were Is.
Both groups were mostly of the S, T, J types. Regardless of the gender, most of the
participants were introvert, sensing, thinking, and judging. As far as males are concerned, no
significant difference was observed between the four dichotomies of E/I, S/N, T/F, and J/P.
Among girls some traces of difference could be noticed concerning E/I and J/P dichotomies.
Moreover, a look at the total column makes it clear that in this study extroversion-
introversion, thinking-feeling, and judging-perceiving dichotomies appear to have no
relationship with writing ability. The difference between S/N is meaningful, statistically
speaking. A plausible explanation of such significance is that sensing/intuitive is the scale that
directly addresses how people like taking information in, i.e., to learn. In this study sensing
types did better than intuitive writers. This finding is different from prior research where Ns
showed some learning advantage (Ehrman & Oxford, 1990), were excellent readers (Ehrnam,
1994), and were good candidates for learning languages (Ehrman & Oxford, 1995). The
reported studies, however, never addressed writing ability. One may also think of other
potential reasons for this striking difference. May be this is due to small size of the learners or
it can be attributed to the personality type of the rater. Yet, the researcher copied and pasted
the data twice just to get a feeling of how things would appear in a bigger sample. No
difference was traced!

Overall, the present study yielded no consistent pattern to clearly distinguish Es from Is,
Ts from Fs, and Js from Ps. Between. Only the sensing/intuitive dimension showed a
difference. How you gather information is directly related to the way you learn, generally
speaking, and particularly the way you learn language. Moreover, the study demonstrated that
gender did play a role: No meaningful difference was observed among male participants, but
as for the females, being S or N affected their writing performance.

The second research question was posed to detect any relationship between rater and
learner personality types. In, MBTI terms, the researcher is an ISTP, with all dimensions
being very clearly marked: (E=7, I=14, S=16, N=10, T=22, F=2, J=8, P=14. The reader
remembers from table 2 that this type was the least frequently observed. Also it was pointed
out that the individuals in this study showed a great tendency toward the ISTJ. So the
dimensions the rater and most participants had in common were I, S and T. The Ss care for
details and produce lengthy pieces. In retrospect, I have always attached a great significance
to details and have often encouraged the other party, Ns, to improve their essays by adding
details. And when thinking about the T dimension, I find that in academic writing the
thinking element seems more relevant than feeling. In fact, organizational patterning and the
rhetorical devices do count in writing. There is another side to the coin. The INTJs enjoy the
highest mean in writing ability. Interestingly enough, here with this group there are two
common dimensions, the I and T preferences. That could explain partly why these types
scored highest. Similarly, Alfallahy (2004), interested in discovering the relationship between
student, peer, and teacher assessments on the one hand and some psychological and
personality traits on the other, reported high correlations among the three types of assessment,
with learner trait being a crucial factor at work.

Of course, generalizing the findings of this research to the population has its limitations
and any attempt to do so should be with caution. One of the limitations was the small number
of students involved, 86. Researchers have reported samples twice as big, and even more.
Additionally, one may refer to the homogeneity of the participants. The performance range
observed was 7.25 (19.25-12). Probably more heterogeneous participants would reveal a
 123

different pattern. Next, having just one rater in this study makes the generalization of the
outcomes rather risky. Still one more factor could be added to the list: the essay topic which
was never controlled. It might just be the case that different topics opting for different
rhetorical organizations could introduce still different impacts.

This study was, however, its own reward. The teacher learned about the types issue. As
a teacher, now I understand how unfair my previous evaluations were. In retrospect, I think I
unjustly pushed them to the direction I favored. This familiarity was quite awakening to me.
Interpreting student writing this time around, I’ll be more careful. As a result of my conscious
application of the MBTI, I believe I’ll be better able to read and respond effectively to student
writing in future.

The participants, too, came to know their types. Discovering that other types also exist,
they find it easier to try others preferences as well (Callahan, 2000). Leaving his own
“comfort zones”, to borrow the term from Ehrman and Oxford (1995), would benefit the
learner. Offering some practice and training, it may be argued, would pave the way for the
learners. Indeed, materials developers can best facilitate this task by offering “a variety of
carefully constructed writing prompts” (Callahan, 2000, p. 74). According to Ehrman and
Oxford (1995), studies aiming at investigating psychological types are promising in that they
offer “an accessible conceptual framework for language trainers and learners ….greater self-
regulation and better learning performance” (p. 324).
 124

References
Alfallahy, I. (2004). The role of some selected psychological and personality traits of the rater
in the accuracy of self- and peer- assessment. System, 32 (3), 407-26.

Bailey, P., Onwuegbusie, A. J., & Daley, C. E. (2000). Using learning style to predict foreign
language achievement at the college level. System, 28, 115-33.

Beauvois, M. H., & Eledge, J. (1996). Personality types and megabytes: student attitudes
toward computer mediated communication (CMC) in the language classroom. CALICO
Journal, 13, 19-45.

Callahan, S. (2000). Responding to the invisible student. Assessing Writing, 7 (1), 57-77.

Ehrman, M. E. (1994). The type differentiation indicator and adult foreign language learning
success. Journal of Psychological Type, 30 (1), 10-29.

Ehrman, M. E., & Oxford, R. (1990). Adult language learning styles and strategies in an
intensive training setting. Modern Language Journal, 74 (3), 311-27.

Ehrman, M.E. & Oxford, R. (1995). Cognition plus: correlates of language learning success.
Modern Language Journal, 79 (1), 67- 89.

Hismanoglu, M. (2000). Language learning strategies in foreign language learning and

teaching. The Internet TESL Journal, 6 (8), n.p.

Hoseini, K. (2003). On the relationship between personality type and performance in the
IELTS Test”. Unpublished MA thesis, Tarbiat Modaress University, Tehran, Iran.

Myers, I. B., & Briggs, K. C. (1998). Myers- Briggs Type Indicator. STEP I/Self-Scorable,
Form M. Palo Alto, CA: Consulting Psychologists Press.

Wilz, B. (2000). Relationship between personality type and Grade Point Average of technical
college students. Unpublished MA thesis, University of Wisconsin-Stout.

Weigle, S. C., Boldt, H. and Valsecchi, M.I. (2003). Effects of task and rater background on
the evaluation of ESL studet writing: a pilot study. TESOL Quarterly. 37(2) 345-54.

Zhenhui, R. (2001). Matching teaching styles with learning styles in East Asian contexts. The
Internet TESL Journal, 2 (7), n.p.

The researcher is assistant professor, her area of interest being writing and the
related issues.
Fahimeh Marefat
English Department, Faculty of Persian Literature and Foreign Languages of
Allameh Tabatabaii University
Allameh Tabatabaii Street
Pol-e-Modirat
Chamran High Way
Tehran, Iran.