Denise Elmer*
Wasted Time or Informed Delay?
Academic Procrastination, Learning Style, and Self-Talk
Abstract
This research determined if Kolb’s (1976) learning style preferences could be used to identify the personality makeup of academic procrastinators. In addition, intrapersonal communication, or self-talk, as related to procrastination and learning style preference was identified. The 174 participants, 101 males and 65 females, were students enrolled in public speaking classes. The measures for this study were self-report procrastination and learning style surveys published by Tuckman (1991) and Kolb (1976). Results indicated that procrastination between learning style preferences is distinctly different for males. Academic procrastination can be predicted for students preferring active and concrete learning. It is evident from this study that content analysis of significant survey questions can be interpreted as self-talk, providing a basis for qualitative exploration during focus group/interview sessions. Survey self-talk exhibits evidence of poor self-image and self-blame with an inability to integrate courses of action that avoid procrastination. Female students’ self-talk reveals task avoidance and negative self-esteem, whereas male students describe a tendency to avoid tasks, negative self-esteem, and self-blame.
Procrastination is a behavior in which we all indulge. Those of us with low
procrastination tendencies find procrastinators annoying; however, because we
all do it, this irritating behavior is often tolerated. Although most research
on procrastination is in the area of psychology, procrastination does have a
meaningful relationship to the field of communication; it is a strategy used
after self-talk has occurred.
Self-talk is intrapersonal communication and can be described as a cognitive
control process that involves the ability to think into the future, to place
value on a choice, to like or dislike a choice, to comprehend the relationship(s)
between cause and effect, and to control the choice made (Roberts, Edwards,
& Barker, 1987). Self-talk centers around self-concept or the positive and
negative views one has of one’s physical, psychological, and social self.
Another term for this self-view is self-esteem. Knaus (1995), a pioneer in procrastination
research, counsels clients to “listen for anxiety-creating self-talk”
when starting to procrastinate and encourages clients to not only “listen
for self-downing self-talk,” but to confront “negative self-talk”
(par. 52-53). Thus, procrastination is an outcome of negative self-talk.
Academic procrastination is a form of procrastination peculiar to education.
Research indicates that academic procrastination is associated with communication
apprehension, ineffective learning strategies, lower grade point averages, boring
or difficult assignments, unplanned study habits, fraudulent excuses, plagiarism,
anxiety, fear of failure, depression, irrational thought, less self-efficacy,
less self-control, and less delay of gratification (Behnke & Sawyer, 1999;
Chissom & Iran-Nejad, 1992; Ferrari, Johnson, & McCown, 1995; Ferrari
& Beck, 1999; Haycock, McCarthy, & Skay, 1998; Lay, 1986; Milgram, 1995;
Roig & DeTommaso, 1995; Senecal, 1995; Solomon & Rothblum, 1984). Ferrari,
Parker, and Ware (1992) concluded that behavioral outcomes due to procrastination
“encompass both cognitive and affective personality components requiring
further analysis” (p. 496). Their suggested solution is to identify the
personality “makeup” of procrastinators.
Most procrastination investigations focus on personality attributes such as
self-efficacy. However, a test for self-efficacy is less likely to be used in
college classrooms because the results have little relation to the topic being
learned. On the other hand, learning style surveys are being administered more
often in classrooms. In the instance of this research, the use of learning styles
is taught as a public speaking technique to students, while instructors are
encouraged to use survey results as a curriculum device (Sellnow, 1999a, 1999b).
Kolb (1976) defined learning style as an individual’s preferred way to
learn. Curry (1987) enhanced this description by describing learning style as
a cognitive personality process that adapts and assimilates information. Assimilation
centers on the use of learning strategies or “translation-like mechanisms”
that individuals use to “cope with” the environment (p. 5). Chissom
and Iran-Nejad (1992) determined that procrastination is a strategy that students
use to delay learning. Thus, the literature indicates that procrastination is
both a behavior and a strategy resulting from negative self-talk.
Procrastination and self-efficacy
The earliest use of procrastination had one definition, an “informed delay,”
and was thought to be an intelligent choice (Ferrari et al., 1995, p. 4). With
the onset of the Industrial Revolution a more negative connotation, “sloth,”
became popular (p. 4). Sloth denotes avoiding work, and includes manipulating
others to do it for you. Ferrari et al. wrote that with modern-day economic
growth, the term “sloth” changes to “task avoidance,” and
described procrastination as irrational thought which originates in cognitive
theory (p. 5). Characteristics of irrational thought include illogical thoughts,
self-defeating choices that undermine goals, fear of failure, immediate gratification,
and too high standards or perfectionism.
Haycock et al. (1998) tested if procrastination can be predicted by efficacy
expectations using a modified Form G Procrastination Inventory. Haycock et al.
operationalized efficacy using Bandura’s (1986) theory—[efficacy is
a] “generative capability in which cognitive, social, and behavior subskills
must be organized into integrated courses of action” (p. 391). Bandura
noted that “self-doubters” abandon this generative process sooner
than “persevering” individuals (p. 391). Lay (1986), developer of
Form G, confirmed that irrational thought is related to procrastination by concluding
that some people judge time requirements inaccurately, resulting in procrastination.
Describing this thought process as irrational; Lay related a lack of time judgment
to disorganization and an inability to maintain priorities. This conclusion
is similar to Bandura’s definition of self-efficacy as well.
To measure efficacy, Haycock et al. (1998) created the Self-Efficacy Inventory
(SEI) which assesses “behaviors related to the task of doing an important
and difficult project by a specific deadline” (p. 319). Through regression
analysis, Haycock et al. found students with low self-efficacy self-report more
procrastination on all the behaviors needed to accomplish a major project. Haycock
et al. claimed that their findings support Bandura’s argument that “strong
efficacy expectations lead to greater task initiation and persistence, whereas
weak expectations produce task avoidance and less persistence” (p. 321).
According to Tuckman (1989) each of us has a tendency to procrastinate. He suggested
that this proclivity is due to a preference for “ease and pleasure over
hard work and pain,” a conviction that trying will lead to failure or “self-doubt,”
and a capacity to blame external people and events for our neglect (pp. 48-49).
Tuckman’s theory originates in Bandura’s (1986) social cognitive theory.
This theory assumes that human behavior is due to the interaction of personal
behavior, personal cognition, and external events. This interaction occurs because
of people’s ability to translate external symbols into knowledge that regulates
future behavior. However, as humans make mistakes, this ability does not preclude
flawed assessments from being formed and acted on. Behavior is both motivated
and self-regulated by comparing the desired behavior outcome to internal standards
and regulations. Based on this construct of self-efficacy, academic procrastination
may be the result of self-talk that includes a self-capability judgment to do
a particular educational task.
Tuckman (1991) devised a self-report measure of academic procrastination tendencies
for the purpose of providing college students a self-report scale that may help
“regulate their own learning” and obtain a “behavioral measure
of self-regulated performance with self-efficacy or the degree to which one
believes in one’s own capability to perform” (p. 474). This survey
is a 35-item (TPS35), 4-point, Likert-type scale with a reliability coefficient
of .90. Further testing reduced it to a 16-item scale (TPS16), with a reliability
coefficient of .86. The Tuckman Procrastination Scale (TPS) has a higher reliability
score than either the two-part PASS, .26, .81; or Aitken Procrastination Inventory
(API), .82, two other common measures of student procrastination (Ferrari et
al., 1995).
The TPS is designed to measure (1) a generalized self-description of the inclination
to delay, (2) a penchant to experience difficulty when doing “unpleasant
things,” and (3) a tendency to blame external events or people for any
ensuing ramifications from procrastination choices (Tuckman, 1991, p. 475).
Survey questions Tuckman used to illustrate these dimensions are, respectively,
(1) “When I have a deadline, I wait until the last minute,” (2) “I
look for a loophole or a shortcut to get through a tough task,” and (3)
“I believe that other people do not have the right to give me deadlines”
(p. 475).
To measure self-doubt or self-efficacy, Tuckman (1991) correlated TPS responses
to the General Self-Efficacy Scale (GSE). The purpose of the GSE is to measure
people’s belief in “their ability to overcome obstacles and succeed”
(p. 476). Examples of GSE questions are
(1) “Failure just makes me try harder,” and (2) “I feel insecure
about my ability to do things”
(p. 476). Tuckman concluded that the TPS is a highly reliable measure of academic
procrastination as related to self-efficacy. He described characteristics of
procrastinators as
(1) “time wasters, deadline avoiders, pleasure seekers, and blamers and
resenters of others who make demands on their time,” and as (2) self-doubters
who worry about their capability to perform and rarely expend extra effort on
tasks that would provide additional benefits such as graded bonus points (p.
479).
Learning style theory and self efficacy
Bandura’s (1986) social cognitive theory is the “conception-matching
process whereby symbolic representations are translated into appropriate courses
of actions” (p. 390). However, it does not explain how this practical learning
is acquired. Examination of Kolb’s (1976) experiential learning theory
indicates that learning is a four-stage cycle that moves from observations and
reflections, to formation of abstract concepts and generalizations, to testing
implication of concepts in new situations, to concrete experience, which again
leads to observations and reflections—or as Kolb summarized, “the
central process of human adaptation to the social and physical environment”
(p. 6). These four phases can be explained by what Kolb described as the “core
of the model,” which is how “experience is translated into concepts
which in turn are used as guides in the choice of new experiences” (p.
2). In other words, the learning cycle is a marriage of learning and problem
solving (Kolb, Rubin, & McIntyre, 1971).
Note the similarity of Bandura and Kolb’s theories. According to Bandura,
self-efficacy is dependent on the ability to recognize external experiences,
whereas, Kolb decided learning begins with the ability to recognize external
symbols. Both abilities integrate these external symbols into concepts based
on past values and experiences, thus producing guides. These guides self-regulate
behavior and suggest new experiences, resulting in action or, as in the case
of procrastination, avoiding action.
Kolb (1976) based his theory on the Jungian concept that “fulfillment in
adult development is accomplished by higher level integration and expression
of nondominant modes of dealing with the world” (p. 2). This learning model
consists of two dimensions with polar opposites: (1) concrete experience/abstract
conceptualization (CE/AC) and (2) active experimentation/reflective observation
(AE/RO). The CE/AC dimension stimulates cognitive growth and learning, including
the ability “to plan ahead ideationally and think and perform symbolically”
(p. 3). The AE/RO dimension represents the ability to apply learning actively
after having tested (reflected) it against internal standards. On the CE/AC
dimension are two distinct learning types, Divergers (CE/feel) and Convergers
(AC/think). The AE/RO dimension represents the two distinct learning types of
Accommodators (AE/do) and Assimilators (RO/watch). To determine learning style,
Kolb devised a Learning Style Inventory (LSI) that ranks an individual’s
preferred way to learn. Hence, student learning can begin at any point within
the four-stage cycle; it depends on the individual’s learning style personality
characteristics.
Each learning style personality has assigned distinctive descriptors. Divergers’
(CE/feel) strength is action after reflection; they are imaginative and emotional
(Kolb, 1976). Convergers’ (AC/think) strength is practical application
of ideas; they are unemotional and focused on single solutions. Accommodators’
(AE/do) strength is to do. They are intuitive, preferring to adapt to the immediate
situation and using trial and error to solve problems. Assimilators (RO/watch)
prefer the creation of theoretical concepts with a lack of concern for the actual
application of theory.
Procrastination, learning style, self-efficacy and self-talk
Perhaps the relationship between procrastination, learning style, self-efficacy
and self-talk has its best explanation within the typology established by Curry’s
(1987) research, which examined the reliability and validity of major North
American, Australian, and European learning style instruments. Using psychometric
evidence, Curry reorganized twenty-one learning style instruments into a system
with three layers. Curry likened this system to an onion with the inner layer,
or core, representing the “central personality dimension” (Hickcox,
1995, p. 28). The middle layer or center focused on a “theme of information-processing,”
and the outer layer is that which has immediate interaction and is influenced
by the environment (pp. 28-29). Curry associated the Myers-Briggs Type Indicator
test (MBTI) with the inner core and the LSI with the middle layer. The middle
layer, or information-processing dimension, represents the cognitive ability
of an individual to “intersect personality levels, individual differences
and environmentally based learning format choices” (p. 32). This characterization
is similar to Bandura’s (1986) interpretation of self-efficacy, “the
processes governing the interrelationship between knowledge and action”
(p. 390), and Roberts et al. (1987) construct of self-talk, a cognitive control
process that involves thinking into the future, placing value on a choice, liking
or disliking a choice, comprehending the relationship(s) between cause and effect,
and controlling the choice made.
Figure One illustrates how the information-processing
dimension encompasses the interrelationship between self-efficacy, experiential
learning, procrastination and self-talk. Students are exposed to external symbols
in the way of textbooks and classroom board notes, and experiences through assignments
and interaction with instructors and classmates. How this information is acquired
depends on their individual learning style type or strength. Once external knowledge
is acquired, it is processed internally by a cognitive comparison of observations,
concepts, and values, resulting in self-regulated behavior and new experience
choices. If self-efficacy is high, the academic objective is accomplished; if
self-efficacy is low, the academic objective is avoided, i.e., procrastination
occurs.
Purpose
The purpose of this research was to test if Kolb’s (1976) learning style
preferences could be used to identify the personality makeup of procrastinators.
As there has been limited re-testing of the TPS, another purpose was to re-test
the reliability of the scale (Ferrari et al., 1995). Lastly, this study examines
how a self-completed survey is an articulation of self-talk.
Hypotheses
A comparison of MBTI and self-regulated performance and procrastination characteristics
to learning style characteristics suggests that a student preferring one learning
style dimension is unlikely to procrastinate more than another, as both dimensions
have procrastination characteristics (Briggs Myers, 1993; Ferrari et al. 1992;
Kolb, 1976, 1981; Tuckman, 1991, 1998). However, prior personality makeup research
did not consider negative self-efficacy as a factor (Ferrari et al. 1992). Tuckman
(1991) defined procrastination as “the lack or absence of self-regulated
performance; the tendency to avoid or put off an activity under one’s control”
and established that procrastination is a behavior related to low self-efficacy
(p. 474). Self-efficacy and learning style preference are interactions of self-talk
or personal behavior, personal cognition, and external events; therefore, it
is probable that learning style type can predict procrastination.
Within the AE/RO (do/watch) dimension, reflection inhibits action, and inversely,
action inhibits reflection (Kolb, 1976); therefore, an Assimilator (RO/watch)
learning style type is more likely to procrastinate than an Accommodator (AE/do).
Within the CE/AC (feel/think) dimension, a Converger (AC/think) is more likely
to “perform symbolically” and a Diverger (CE/feel) to experience (Kolb,
1976); therefore, a Converger learning style type is more likely to procrastinate.
Alternatively, a combined preference for the two styles most inclined to procrastinate,
Assimilator and Converger, may be more indicative of procrastination. Lastly,
it appears that the Assimilator and Converger preferences may be prone to negative
self-efficacy. Rank order then, from most likely to procrastinate, should be
Assimilator due to preference for inhibition of action, Converger due to preference
for symbolic performance, Diverger due to preference to experience, and Accommodator
due to preference to carry out plans.
Therefore, because it is probable that learning style preference is related
to procrastination, there are two hypotheses for this research.
Hypothesis One: students with the learning styles of Assimilator (RO/watch)
and Converger (AC/think) are more likely to academically procrastinate than
students with the styles of Accommodator (AE/do) and Diverger (CE/feel). Within
the combined types, RO/AC (watch/think) is more likely to procrastinate CE/AE
(feel/do).
Hypothesis Two: the rank order, from highest to lowest, of learning style
academic procrastination tendency, is Assimilator, Converger, Diverger, and
Accommodator.
Questions asked
Individual procrastination questions were tested to assess what variety of self-talk
occurred within each learning style. It should be noted that these questions
have not been interpreted as self-talk in other studies; however, the items
measured in these questions fit the definition of intrapersonal communication
(Roberts et al., 1987). There was no prediction as to which questions would
be significant for each type.
Method
Participants
The 174 participants in this study were students enrolled in public speaking
classes at a Midwest, land-grant, state university. Classes are capped at twenty-two
students each, with approximately 800 enrolled each semester. Public speaking
is a required class for all students planning to graduate from this university
of 9,000-plus students. Participants were invited to take part by their instructors
and received extra credit points for completing a procrastination and learning
style inventory. Confidentiality of students’ identities was maintained
by use of control identification numbers. The LSI is a regular class assignment,
usually completed outside of the classroom, although some instructors had students
complete the surveys in class. In either instance, the purpose of the survey
and instructions were covered orally with students.
Research instruments
The measures for this study were two survey instruments: the first a self-report
procrastination survey (TPS) published by Tuckman (1991), and the second a self-report
learning style survey (LSI) published and copyrighted by Kolb (1976). If a respondent
left a reply blank or scores did not fall within defined ranges, the response,
as well as the respondent’s corresponding TPS or LSI, was eliminated from
the study. Ultimately, eight responses were rejected.
Procrastination instrument
The TPS is a 35-item, 4-point, Likert-type scale with a reliability of .90 (Tuckman,
1991). Using factor analysis, Tuckman reduced this 35-item scale to a 16-item
scale consisting of questions that loaded .40 or higher with an alpha reliability
of .86. The 16-item scale takes less time to complete than the 35-item scale
and can be more quickly used by students for self-assessment of procrastination
tendencies. The 16-point scale is contained within the 35-item scale, so both
were administered at the same time. Self-efficacy and self-regulated behaviors
are measured by default (Tuckman, 1991).
Methods used to re-test the reliability of the scale were alpha reliability
and factor analysis. As in Tuckman’s study, factors loading .40 and higher
were examined for reliability and interpretation of procrastination as related
to self-efficacy and self-regulated behavior. A minimum threshold of .80 was
predetermined to be an acceptable alpha reliability.
On the TPS instrument, students indicated their self-rated level procrastination
tendency by selecting “1, That’s me,” “2, That’s my
tendency,” “3, Not my tendency,” or “4, Not me.” The
minimum score possible on the TPS 35-item scale is 35, and describes a high
degree of self-reported procrastination. The maximum score possible is 140,
and describes a non-procrastinator. Minimum and maximum scores for the TPS16
item scale are 16 and 64.
Each procrastination question with its corresponding numerical response was
entered as a variable, as well as the total procrastination score, i.e., the
sum of the 35 questions and the sum of the 16 questions. Factor analysis resulted
in a new scale, PS14, for which a procrastination score variable was entered,
i.e., the sum of the 14 questions. The minimum score possible on the PS14 item
scale is 14 and the maximum score is 56.
Learning style instrument
Kolb’s (1981) measure is a twelve-sentence completion list. Scores are
self-ranked on a 4-point, Likert-type scale that has a test, re-test reliability
of .92 to .97 (Veres, Sims, & Locklear, 1991). The learning style survey
is an assignment in the text used in the public speaking class (Sellnow, 1999b).
Students rank their learning style preferences of Assimilator (RO/watch), Converger
(AC/think), Accommodator (AE/do), or Diverger (CE/feel) by selecting “1,
Least like me,” “2, Third most like me,” “3, Second most
like me,” or “4, Most like me.” The four individual learning
style scores range from 12 (“Least like me”) to 48 (“Most like
me”). The four respective scores always total 120.
Each self-ranked total was entered as a variable: feel (CE), watch (AO), think
(AC), and do (AE), as well as each student’s primary learning style type,
which was determined by subtracting the CE score from the AC score and the RO
score from the AE score. These two differences were plotted on Kolb’s (1981)
“Learning Style-Type Grid” and determined the participants’ primary
learning type. This procedure was repeated with the difference between the total
means to determine the participants-as-a-group primary type. To determine the
group’s learning style profile, learning style score means were plotted
on Kolb’s “Cycle of Learning Graph.” Profiles by sex were not
plotted.
To facilitate the comparison of procrastination tendencies between learning
style preferences, scores were summed to create various combination variables.
These variables were (1) the CE/AC (feel/think) dimension, (2) the AE/RO (do/watch)
dimension, (3) the CE/AE (feel/do) types, and (4) the RO/AC (watch/think) types.
Lastly, dichotomous coding of the highest score of each feel/do and watch/think
types created a variable that encompassed all learning style preferences. In
total, there were nine learning style variables for testing against two procrastination
variables.
Distribution analysis
Each learning style and procrastination variable was tested for normal distribution
by examining the ratio of skew to standard error, the 95 percent confidence
interval ranges, and the Kolmogorov-Simirnov with Lilliefors Significance Correction
tests of normality. It was determined to use log (base 10) transformation if
any variables were a probability of more than .05 on the tests of normality;
this transformation established a more normal distribution of scores with z
scores. However, two variables exceeded .05, and all variables had outliers
after log transformation; therefore, it was decided to eliminate all outliers.
Elimination of outliers in one variable inevitably created an outlier in another
variable that did not have one before. These occurrences supported Kolb’s
(1984) theory that learning style scores are dialectically opposed rather than
unitary. Knowing that outliers would most likely not be eliminated, it was decided
to use the first log transformation variables for hypothesis testing and measures
of association.
Hypothesis testing
To determine if procrastination by learning style preference would be attributed
to a normal distribution in the population, one-way ANOVA procedures were used
to the test TPS16 point and PS14 point procrastination means between and within
each learning style variable. Significant results, p < .05, were hypothesized
for watch (AO), think (AC), summed RO/AC (watch/think) types, and coded CE/AE
RO/AC (feel/do, watch/think) types. Individual procrastination questions were
also tested with one-way ANOVA procedures to check what types of self-talk occurred
with each learning style. There was no prediction as to which questions would
be significant for each type. All significant results were tested further with
Bonferroni and Tukey a methods, to identify which specific means differ.
Measures of association
To determine if procrastination and learning style preference were related,
Pearson correlation coefficients were used to measure the strength of the association
between the variables. Positive values for a variable indicated that as procrastination
increases, so does the strength of the learning style preference. A negative
value indicated that, as procrastination increases, the strength of the learning
style preference decreases. For example, if, as hypothesized, the CE/AE (feel/do)
preference procrastinates less than the RO/AC (watch/think), the correlation
coefficients would be negative CE/AE (feel/do) and positive RO/AC (watch/think).
Results
Distribution of participants was reflective of the usual enrollment in this
university’s public speaking classes. Specifically, there were 92 first-year
students, 53 sophomores, 14 juniors, and 7 seniors; 101 were male and 65 were
female students. The highest percentage of students was first-year, 55.4 percent,
and sophomore, 31.9 percent; years with lower representation were juniors, 8.4
percent, and seniors, 4.2 percent, N=166. Dstributiion between sexes was 39.2
percent female, n =65, and 60.8 percent male, n=101.
Factor analysis of the 35-item procrastination scale failed to duplicate Tuckman’s
original 16-item scale (Table One). Instead, using
principal component analysis, factor analysis of the 35-item procrastination
scale extracted 11 components. Component 1 had an eigenvalue of 7.9 and explained
22 percent of the variance. Components 2 through 11 had eigenvalues ranging
from 6.6 to 2.9, with all 11 components explaining 63% of the variance. The
remaining components (12-35) had eigenvalues ranging from .98 to .21. Using
Tuckman’s (1991) original selection criteria of a .40 loading, Component
1 yielded a 1-factor rotated component matrix solution of 14 items with an alpha
reliability of .86. Cronbach alpha reliability of the 35-item scale was .79
and of the 16-item scale (TPS16) was .88. The 35-item scale was not used in
any further procedures due to reliability less than .80; instead, the TPS16
and the new scale from this study’s factor analysis, PS14, were used.
Scales’ 16 and 14 procrastination score means were midway between “that’s
my tendency” and “that’s not my tendency, ” with respective
means of 39.64 and 34.34 (SDs = 7.25 and 6.32, respectively; N=16). Tests of
normality pre-log (base 10) transformation significances were .20 each. Post-log
transformation significances were .01 and .05, respectively. In other words,
conversion of procrastination scores resulted in a more normal procrastination
distribution, thus increasing test results reliability.
Learning style primary types were evenly distributed for all participants (Table
2). While the individual learning style preference standard deviations were
larger than for a normal distribution, the ratio of skew to standard error of
the mean indicated that distribution is symmetric (Table
2). Learning style score ranges for Divergers (CE/feel), Convergers (AC/think),
and Accommodators (AE/do) met the 95 percent confidence interval. Scores for
Assimilators (RO/watch) had the most variance, with 33 percent of the scores
below the 95 percent confidence interval range of 27 to 45. Tests of normality
were df 166, sig. = .04, .05, .06, and .04, respectively. A log (base 10) transformation
yielded normality tests with significance of .00, .05, .00, and .01, respectively.
Plotting learning style score means on Kolb’s (1981) “Cycle of Learning
Graph” reflected a group learning style profile of Assimilator (RO/watch),
Converger (AC/think), Accommodator (AE/do), and Diverger (CE/feel) with a primary
learning type of Assimilator. The data points were very close to the grid center,
indicating that, while participants prefer abstract learning through reflective
observation, participants had a balanced approach in their learning style. Repeating
the same procedure for each sex resulted in a Diverger (CE/feel) type preference
for female students and an Assimilator (RO/watch) type preference for male students.
While female students preferred concrete learning through action after reflection,
male students preferred abstract learning through reflective observation. Again,
data points’ placements on the grid reflect a balanced approach in learning.
Combined learning style dimensions and types’ standard deviations were
also larger than a normal distribution (Table 3); however,
the ratio of skew to standard error of the mean indicated that the distribution
was symmetric. Ranges for the CE/AC (feel/think) dimension, the AE/RO (do/watch)
dimension, and the CE/AE (feel/do) combined types met the 95% confidence interval.
Scores for RO/AC (watch/think) combined types had the most variance, with 5%
of the scores placing below and 1% above the 95% confidence interval range of
45 to 77. Tests of normality were df 166, sig. = .20. Log (base 10) transformation
of the four variables, in the above order, yielded normality test significances
of .07, .01, .20, and .02. Normality test significances for the Feel/Do Watch/Think
factor were .00 and .20 pre and post log transformation.
Results: Hypothesis One
There was limited support of Hypothesis One. Findings indicate that levels of
procrastination between learning style preferences are distinctly different
for males and support Kolb’s (1984) learning model that consists of two
dimensions with polar opposites (Table 4). For male
students only, one-way ANOVA procedures by learning style dimensions were significant,
p = .03, and explain 4 percent of the procrastination variance. A Type I error,
the null hypothesis may be true or that procrastination may not be predicted
by learning style, cannot be ruled out based on homogeneous subsets of Tukey
a post hoc tests. All other one-way procedures were not statistically significant.
Results: Hypothesis Two
Procrastination order on both scales was Assimilator (RO/watch), Converger (AC/think),
Accommodator (AE/do), and Diverger (CE/feel). This order failed to support Hypothesis
Two, that procrastination rank order, from highest to lowest, is Assimilator,
Converger, Diverger, and Accommodator. In order, the means for learning style
preference procrastination were TPS16: 37.96, 39.56, 40.80, and 41.11 (SDs =
7.23, 7.16, 7.84, and 6.47, respectively) and PS14: 32.93, 34.39, 35.14, and
35.63 (SDs = 6.34, 5.88, 35.14, and 6.01, respectively). The combined types
procrastination order is the reverse of what was hypothesized; rank procrastination
order was CE/AE (feel/do) rather then RO/AC (watch/think). Respective means
and standard deviations (in parentheses) were TPS16: 38.7 (6.7) and 41.9 (7.7),
and PS14: 33.5 (5.7) and 36.6 (6.8).
Results: self-talk questions
The individual questions of the procrastination scale were examined to determine
what type of self-talk is employed when procrastinating. Tests of normality
were significant for all procrastination questions (df 166, sig. = .00 for each).
Significant, p < .05, one-way ANOVA procedures for questions on the procrastination
scales are shown in Table 5. Questions beginning with
a letter are found on the TPS16 scale; questions beginning with a number are
found on the PS14 scale. Self-talk for all participants were self-descriptions
of low self-efficacy, self-blame, and a tendency to avoid tasks, explaining
2 percent of the procrastination variance.
Female student self-talk appeared to be self-descriptions of low self-efficacy,
self and external blame, and a tendency to avoid tasks, explaining 4 percent
to 6 percent of the procrastination variance. Self-talk for male student selections
was self-descriptions of low self-efficacy, self and external blame, and a tendency
to avoid tasks, explaining 3 percent to 4 percent of the procrastination variance.
Results: measures of association
Contrary to Hypothesis One, the summed CE/AE (feel/do) types had a positive
correlation to procrastination, and the RO/AC (watch/think) types had a negative
correlation (Table 6). However, when learning styles
are looked at individually, none had a correlation to procrastination. This
association indicates that students preferring the concrete/active (CE/AE) approach
to learning are most likely to procrastinate. Male students preferring the Converger
(AC/think) learning style had a negative correlation to procrastination. This
correlation indicates that male students preferring an abstract (AC) approach
to learning are least likely to procrastinate.
Individual questions of the procrastination scale were tested to determine what
type of intrapersonal communication or self-talk is related to procrastination.
Positive correlations (Table 5) were considered only
for questions with a significant ANOVA. Note, all but one significant question
is found on both scales. Question (V & 7) “I am a time waster now,
and I can’t seem to do anything about it” has a positive correlation
for all students favoring Accommodator (AE/do) and the summed CE/AE (feel/do)
types. There was a negative correlation for all students favoring the Assimilator
(RO/watch) and Converger (AC/think) learning styles, and summed RO/AC (watch/think)
types. This question is a self-description of low self-efficacy and self-blame.
These self-descriptions indicate that students who prefer to learn by the concrete
and active experimentation approaches (CE/AE) are more inclined to a poor self-image
and blame themselves for an inability to integrate courses of action that avoid
procrastination. Those students who learn by abstract concepts and reflective
observation (RO/AC) have a more positive self-image are less likely to blame
themselves for their procrastination (Table 6).
For female students, question (5) “I get right to work, even on life’s
unpleasant chores” was positively correlated to the summed CE/AC (feel/think)
dimension and negatively correlated to the summed AE/RO (do/watch) dimension.
The reverse was true for question (R & 6) “I am an incurable time waster,”
which was positively correlated to AE/RO (do/watch) and negatively correlated
to CE/AC (feel/think). Question (5) is a self-description of a tendency to avoid
tasks, indicating that female students’ preference for learning by concrete
experience and abstract concepts (CE/AC) results in task avoidance, whereas
female students who prefer active experimentation and reflective observation
(AE/RO) learning are less likely to avoid tasks. Question (R & 6) is a self-description
of low self-efficacy, indicating that female students who favor concrete experience
and abstract learning concepts tend towards negative self-efficacy, whereas
those who favor active experimentation/reflective observation learning tend
towards positive self-efficacy (Table 6).
For male students, question (CC & 12) “I always finish important jobs
with time to spare” and question (HH & 14) “Putting off until
tomorrow is not the way I do it” were negatively correlated to Convergers
(AC/think). Question (HH & 14) was also negatively correlated to the CE/AC
(feel/think) and positively correlated to AE/RO (do/watch). Both questions are
self-descriptions of a tendency to avoid tasks. These results indicate that
male learning style preference for learning by active experimentation/reflective
observation (AE/RO) results in task avoidance, whereas male students who prefer
abstract thought (AC/think) and concrete experience/abstract concepts (CE/AC)
learning are less likely to avoid tasks.
Questions (V & 7) “I am a time waster now, and I can’t seem to
do anything about it” and (FF & 13) “I get stuck in neutral even
though I know how important it is to get started” self-described a tendency
toward low self-efficacy and self-blame. Correlations were positive for Diverger
(CE/feel) and summed CE/AE (feel/do) types, indicating that male students preferring
to learn by the concrete and active experimentation approaches (CE/AE) are more
inclined to blame themselves for procrastinating behavior with a corresponding
poor self-image. Correlations were negative for Assimilator (RO/watch), Convergers
(AC/think), and summed RO/AC (watch/think) types, indicating that male students
who learn by abstract concepts and reflective observation (RO/AC) are less prone
to low self-blame and self-esteem; that is, they are better able to determine
a course of action that avoids procrastination with a corresponding positive
self-image (Table 6).
Discussion
Personality makeup
This analysis is the second study to relate similar personality characteristics
to academic procrastination. Ferrari et al. (1992) found a positive correlation
to Perceiving, one of the four preferences on the MBTI psychological type scale.
Characteristics of Perceiving include being spontaneous, open-ended, flexible,
adaptable, loose and open to change, and feeling energized by last-minute pressures
(Briggs Myers, 1993). Personality characteristics of the summed CE/AE (feel/do)
types include empathy, imagination, extroversion, and risk-taking; Accommodator
(AE/do) characteristics include adaptation, risk-taking, use of trial and error,
and high reliance on others for information (Kolb, 1978, 1984). Notice the similarities
between Perceiving and Accommodator; both are impulsive and carefree with Diverger
(CE/feel) adding the aspect of sociability. An impulsive and carefree approach
to academic tasks can result in time frittered away and deadlines shirked.
On the other hand, Convergers (AC/think) tend to be less sociable and focus
on developing practical solutions. These characteristics are similar to Judging,
also on the MBTI scale; Judging, as does Converger, has a negative correlation
to procrastination (Ferrari et al., 1992). Convergers tend to pre-plan and complete
tasks before deadlines; consequently, it appears that the major difference between
Converger and Accommodator is the ability to self-regulate performance.
An explanation for this occurrence, may be found in Kolb’s (1976) research
on learning style principles: certain careers are associated with certain learning
styles; students tend to select a major that best fits with their learning style;
and students tend to change majors or learning style preference if there is
a mismatch (Kolb, 1978). The university from which data were gathered specializes
in engineering and agriculture. Engineering is primarily associated with the
Converger (AC/think) learning style and agriculture with Converger and Assimilator
(RO/watch) types. Smaller specialties offered are secondary education, business,
and arts such as theater and journalism, all of which are associated with either
Accommodator (AE/do) or Diverger (CE/feel) types (Kolb, 1981, 1984). Over one-third
(41 percent of the learning style types are Accommodators and Divergers in the
first and second year grades. Distribution of the Accommodator and Diverger
learning styles suggests that students specializing in engineering and agriculture
have either not selected a major that best fits with their learning style, or
have not changed their major or learning style preference to relieve the inconsistency,
whereas students preferring a Converger style have selected a major compatible
with their learning style preference. Findings indicate that failure to resolve
the mismatch of learning style to major seems to result in procrastination tendencies
for Accommodators and Divergers at this university.
Intrapersonal communication
The premise that students favoring Assimilator (RO/watch) and Converger (AC/think)
styles are more likely to procrastinate presumes that both styles are prone
to negative self-efficacy, which is based on the personality descriptors for
Judging, i.e., self-doubter, worrier, feelings of guilt, and doubting one’s
own capability to perform (Briggs Myers, 1993; Tuckman, 1991). The assumption
that Assimilator and Converger styles have an inability to self-regulate performance
is founded on personality preference that these styles would rather generate
ideas and defer action. Correlations suggest the opposite; that is, Accommodator
(AE/do) and Diverger (CE/feel) styles have negative self-efficacy and less ability
to self-regulate performance; Converger styles have positive self-efficacy and
self-regulate performance. Evidence of negative self-efficacy is found for all
students favoring Accommodator (AE/do) and the summed CE/AC (feel/do) dimension
in the negative self-talk and negative self-concept, “I am a time waster
now” (Table 7). Furthermore, there is evidence
of an inability to integrate a course of action or regulate self-performance
in the statement, “and I can’t seem to do anything about it.”
Female students’ Accommodator (AE/do) and Diverger (CE/feel) types express
a negative self-concept when they describe themselves as “incurable time
wasters.” Male students’ Accommodator and Diverger types express negative
self-esteem by describing themselves as “time wasters” and being “stuck
in neutral.” In addition, male students differ from female students by
describing a tendency to avoid tasks or regulate self-performance. Males depict
themselves as “putting off until tomorrow” as the way to “do
it.” On the other hand, male Converger (AC/think) types assert, “I
always finish important jobs with time to spare” (Table
7).
What may be underlying this negative self-talk, negative self-efficacy and unregulated
self-performance is a resentment of structure and rules, a procrastination characteristic
of Perceiving (Briggs Myers, 1993). Evidence of structure and rules can be found
in questions that positively correlated to Accommodation (AE/do) and Divergers
(CE/feel) summed types. Academic structure and rules are apparent in the implied
expectations of deadlines, studying, and ability to prioritize. Procrastinators
“wait until the last minute,” find studying “boring,” and
report inertia even though they “know how important it is to get started.”
Perceived negative societal attitudes towards procrastination are apparent in
the implication that procrastination is a disease or malady as procrastinators
self-describe themselves as “incurable.” Negative self-talk and general
self-efficacy are strong predictors of procrastination, thus an area for future
research is to determine if there is a direct relationship between learning
style preference and self-efficacy (Ferrari et al., 1992; Knaus, 1995).
Whereas examination of individual procrastination scale questions provides us
with a sense of the kind of self-communication occurring, there has been little
research on procrastination as a communication strategy. Behnke and Sawyer (1999)
have correlated procrastination to communication apprehension while Chissom
and Iran-Nejad (1992) have specifically identified procrastination as a strategy
students use to avoid learning. An area of promising research identifies procrastination
as a compliance-gaining strategy (Schneider & Beaubien, 1996). Typically,
procrastination is viewed as avoidance of an interpersonal interaction; however,
Schneider and Beaubien describe it as an intrapersonal communication device
employed by physicians to avoid a difficult (i.e., no-win) situation. In a compliance-gaining
situation, an agent selects a strategy (intrapersonal communication), which
results in a persuasive interaction with a target (interpersonal communication)
and (hopefully) the desired outcome for the agent. If procrastination is selected
as the compliance-gaining strategy (intrapersonal communication), interaction
avoidance and (hopefully) the desired outcome for the agent result.
Applying this framework suggests that persuasive interaction (interpersonal
communication) and procrastination (intrapersonal communication) are subject
to situational factors and the level of power the agent and target (deTurck,
1995). For example, students often complete major papers the night before they
are due. A student’s decision (intrapersonal communication) to complete
a major assignment the night before it is due (procrastination strategy) may
be predicated on the current relationship (referent power) with the target or
experiences (situational factors) that resulted in past high grades. The target
may accept this interaction avoidance because an instructor prefers not to use
his/her authority to discipline or otherwise control (coercive, legitimate,
and expert power) the agent’s behavior. Eventually, this strategy may become
habit on the part of the agent and culturally accepted on the part of the target
(Hill, Hill, & Barrall, 1978; Knaus, 1995). Findings from this study indicate
that the agent is most likely a risk-taking Accommodator (AE/do) that feels
energized by last-minute pressures. Further research on procrastination as a
compliance-gaining technique, as opposed to a personality characteristic, is
warranted.
Limitations
Tuckman’s (1991) procrastination survey was devised to help college students
“regulate their own learning.” (p. 474). Due to its high reliability,
.88, the 16-point scale is an excellent tool for student self-administration,
self-evaluation, and self-use; making use of it in college student class advising
and planning divisions may be useful. However, a limitation of the scale is
that some students found the phrasing puzzling, and as with all self-report
tools, results are contingent on the honesty of participants. In this region,
another limitation is a “desire to please” which may have influenced
some responses. In addition, two methods were employed to collect these data:
completion in the class and take home. Not all the take-home surveys were returned,
leading to the conclusion that some high procrastination scores are missing.
The sample itself, which may have been underrepresented with juniors and seniors,
was primarily white, Midwestern students whose procrastination tendencies may
not be representative of other regions in the United States.
Other findings that may be attributed to regional differences are a factor analysis
of 14 items instead of 16 on the 35-item procrastination scale. The 14-item
scale suggested self-blame, whereas the 16-item scale includes external blame
for procrastination tendencies. The above findings suggest that researchers
may want to include factor analysis and alpha reliability when using the 35-item
scale in order to assure a scale reliability of at least .80 and to ascertain
regional differences. Lastly, while the transformed raw data from individual
learning types and procrastination indicated a normal population distribution,
there is less confidence in the transformed summed dimensions and types. This
limitation should be kept in mind when applying results.
Conclusions
The general purpose of this study was to determine if procrastination could
be predicted by known personality characteristics such as learning style preference.
Although there is some evidence to support this association, limitations indicate
that caution is needed with using this information. A more specific purpose
was to introduce the investigation of procrastination to the field of communication.
To date, research has been limited, yet it is evident from this study that content
analysis of significant survey questions can be interpreted as self-talk, providing
a basis for qualitative exploration during focus group and interview sessions.
For example, society may view procrastination as a disease, and students agree
with this view by describing themselves as “incurable” when answering
a survey, but will they support this theme during a qualitative process in which
they describe their internal perspective of procrastination?
Another area of investigation is within intrapersonal communication, specifically,
as a compliance-gaining strategy used to convert no-win situations to ones of
“informed delay” (Ferrari et al., 1995, p. 4). The earliest use of
procrastination had one definition, an “informed delay,” and was thought
to be a wise or prudent choice (p. 4). However, delay may be an unavoidable
choice for some students as today’s college students are busy people. Within
a semester’s 15-week time span, students carrying full academic loads are
attending classes 12 or more hours a week, with a corresponding homework load
of 36 hours or more. Many are also involved in extracurricular activities, such
as sports and other campus organizations, which can vary from an occasional
hour to 10 hours or more a week in active participation. Many more must work
a part-time job, usually 20 hours a week, to help finance their college opportunity.
Given this academic lifestyle, it is surprising that academic procrastination
rates are not higher. Perhaps it is time to include informed delay within the
definition of academic procrastination.
Lastly, expanding the understanding of procrastination by way of a communication
lens may augment the multitude of self-help resources advocating the use of
self-talk as a “cure” for procrastination. As a behavior, procrastination
is obvious, however, as an intrapersonal communication device, procrastination
is less clear; this study offers new insight into procrastination as a strategy
one employs after self-talk has occurred, thus inviting further research through
a communication standpoint.
References
Bandura, A. (1986). Social foundations of thought and action:
A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.
Behnke, R., & Sawyer, C. (1999, Fall). Public speaking procrastination as
a correlate of public speaking communication apprehension and self-perceived
public speaking competence. Communication Research Reports, 16 (1), 40-47.
Briggs Myers, I. (1993). Introduction to type (5th ed.). (Rev. by L. K. Kirby
& K. D. Myers). Palo Alto, CA: Consulting Psychologists Press.
Chissom, B., & Iran-Nejad, A. (1992). Development of an instrument to assess
learning strategies. Psychological Reports, 71, 1001-1002.
Curry, L. (1987). Integrating concepts of cognitive or learning style: A review
with attention to psychometric standards. Ottawa, Ontario: Canadian College
of Health Service Executives.
DeTurck, M. (1995). Persuasion and compliance-gaining. In R. Hartman & L.
Texter (Eds.), Advanced interpersonal communication (pp. 271-284). Dubuque,
IA: Kendall/Hunt.
Ferrari, J., & Beck, B. (1999, Summer). Affective responses before and after
fraudulent excuses by academic procrastinators. Education, 118 (4), 529-538.
Ferrari, J., Johnson, J., & McCown, W. (1995). Procrastination and task
avoidance theory, research and treatment. New York: Plenum Press.
Ferrari J., Parker, J., & Ware, C. (1992). Academic procrastination: Personality
correlates with Myers-Briggs types, self-efficacy, and academic locus of control.
Journal of Social Behavior and Personality, 7 (3), 495-502.
Haycock, L., McCarthy, P., & Skay, C. (1998, Summer). Procrastination in
college students: The role of self-efficacy and anxiety. Journal of Counseling
and Development, 76, 317-324.
Hickcox, L. (1995). Learning styles: A survey of adult learning style inventory
models. In R. Sims & S. Sims (Eds.), The importance of learning styles (pp.
25-47). Westport, CT: Greenwood Press.
Hill, M., Hill, D., Chabot, A., & Barrall, J. (1978). A survey of college
faculty and student procrastination. College Student Journal, 12, 256-262.
Knaus, W. (1995). Overcoming procrastination: A new look [On-line]. Retrieved
from Albert Ellis Institute Home Page World Wide Web: http://www.irebt.org/essaays/procrst1.html.
Accessed January 14, 2000.
Kolb, D. A. (1976). Learning style inventory technical manual. Boston: McBer
and Company.
Kolb, D. A. (1978). Learning style inventory technical manual (Rev. ed.). Boston:
McBer and Company.
Kolb, D. A. (1981). Learning style inventory: Self-scoring inventory and interpretation
booklet. Boston: McBer and Company.
Kolb, D. A. (1984). Experiential learning experience as the source of learning
and development. Englewood Cliffs, NJ: Prentice-Hall.
Kolb, D. A., Rubin I., & McIntyre, J. (1971). Organizational psychology
an experiential approach. Englewood Cliffs, NJ: Prentice-Hall.
Lay, C. (1986). At last, my research article on procrastination. Journal of
Research in Personality, 20, 474-495.
Milgram, N. (1995, March). Correlates of academic procrastination: Discomfort,
task aversiveness and task capability. The Journal of Psychology, 129 (2), 145-156.
Roberts, C., Edwards, R., & Barker, L. (1987). Intrapersonal communication
processes. Scottsdale, AZ: Gorsuch Scarisbrick.
Roig, M., & DeTommaso, L. (1995, October). Are college cheating and plagiarism
related to academic procrastination? Psychological Reports, 77 (2), 690-698.
Abstract from http://www.pals.msus.edu/cgi-bin/pa…1/te%20%20procrastination/di%200034
Accessed January 14, 2000.
Schneider, D., & Beaubien, R. (1996, Summer). A naturalistic investigation
of compliance-gaining strategies employed by doctors in medical interviews.
The Southern Communication Journal, 61 (4), 332-341.
Sellnow, D. (1999a). The process of public speaking (Rev. ed.). Orlando, FL:
Harcourt College Custom Publishers.
Sellnow, D. (1999b). Public speaking applications (2nd ed.). New York: Forbes
Custom Publishing.
Senecal, C. (1995, October). Self-regulation and academic procrastination. The
Journal of Social Psychology, 135 (5), 607-620.
Solomon, L., & Rothblum, E. (1984). Academic procrastination: Frequency
and cognitive-behavioral correlates. Journal of Counseling Psychology, 31 (4),
503-509.
Tuckman, B. (1989, March). Procrastination busting. Educational Technology,
29 (3), 48-49.
Tuckman, B. (1991). The development and concurrent validity of the procrastination
scale. Educational and Psychological Measurement, 51, 473-480.
Tuckman, B. (1998, Winter). Using tests as an incentive to motivate procrastinators.
The Journal of Experimental Education, 66 (2), 141-148.
Veres III, J., Sims, R., & Locklear, T. (1991). Improving the reliability
of Kolb’s revised learning style inventory. Educational and Psychological
Measurement, 51, 143-150.
* Denise Elmer is an instructor in the Department of Communication, Drama and Journalism at Angelo State University, San Angelo, TX, 76909. She may be contacted at her personal e-mail address <GraciewKW@aol.com>. Portions of this research were presented at Central States Communication Association convention April 2000, Detroit, MI. The author thanks the two anonymous reviewers and Dr. R. Collins for their assistance in the development of this manuscript.
