|
This study attempts to investigate the hypothesis that there is a correlation between an individual’s personality, as indicated by the personality traits presented by the Jung’s personality test and their ability to learn computing. Both the qualitative and the quantitative approaches to conducting research were used in order to come up with a conclusive answer to the hypothesis that an individual’s personality may very well have a link with their ability in computing.
A literature review that was conducted as being a part of the
qualitative investigations for this study indicates that there are
other studies that have been conducted which provide an indication that
certain personality traits may be indicative of ability in computing.
Individuals who are introverted, sensing, thinking, feeling and judging
were found to be more likely to be successful in the field of computing
then others. Most of the undergraduate and postgraduate students of
computing at British universities possessed these personality traits. A
quantitative study that was carried out for this study supported this
finding from the literature review. However, despite substantial
efforts being made to elicit responses from a substantial number of
individuals, relatively few good responses were forthcoming for
inclusion in the statistical analysis. Pearson’s correlation
coefficients were used to investigate any correlation between the
variables. The results of the statistical analysis indicated a moderate
correlation between introversion, sensing, thinking and judging
attributes of personality type and ability in computing. These results
supported the results of other studies that are discussed in the
literature review. It is felt that longitudinal studies which can
investigate a correlation between personality and ability in computing
amongst those who apply for admission to computing courses will prove
useful in providing further insights into the hypothesis that has been
investigated.
1. Introduction
Computers have become an accepted part of everyday life throughout the
world and are also being widely used at work, with many businesses
depending on their information systems to conduct routine business
activities. Rapid changes in technology have meant that an average
member of the workforce is now expected to have a certain minimum level
of computer literacy
in order to be able to perform most job functions (Burger, 2003, Pp. 1
– 15). Computers, software and the computing interface for interaction
with humans have progressively emerged as being more friendly for the
human users and convenient as well as easy to use. However, despite
these advances and the emergence of a ubiquitous computing age in which
intelligent devices are to be found interacting with humans in every
sphere of human existence, there is still a certain level of computer
anxiety that is to be found amongst a proportion of the population.
Whereas, merely interacting with computers may not be a major problem
for many, the more skilled tasks of programming, database design, an
understanding of the logic associated with computers, hardware design
and many other areas of in – depth knowledge associated with computers
can present a far greater challenge to some individuals as compared to
others.
A lack of learning ability in mathematics, termed the Dyscalculia
Syndrome, has received a certain level of recognition and attention
(Newman, 1998, Pp. 7 – 12). It has been found that some individuals
suffer from a poor memory for mathematics as well as other symptoms and
characteristics that make it difficult for them to learn mathematics.
It has also been observed that when individuals are imparted the same
computer training, some individuals are able to acquire far superior
computing abilities as compared to the others (Burger, 2003, Pp. 1 –
4). The personality of an individual can have at certain level of
influence on their ability in at least some activities associated with
computing (Devito da Cunha, 2003, Pp. 94 – 97). Some students who
usually end up studying business, accountancy or finance have been
found to be far better at interacting with people then at solving the
highly logical problems associated with computing, especially when such
efforts require great concentration and an attention to detail
(Momberg, 2004, Pp. 120 – 122), (Brown, 2004, Pp. 24 – 26), (Simon,
2006, Section 4) and (Darcy, 2005, Pp. 8). Studies have also indicated
that the attitude of women to computer technology is somewhat more
constructive then that of men (Ray, 1999, Pp. 7). Successful scientists
are said to possess certain dominant patterns of behaviour, ethics and
motivation (Jarrard, 2006, Chapter 10) and (Standler, 1998, Section 3).
It has also been claimed that it is possible to test individuals in
order to determine if these individuals are capable of succeeding at
learning computer programming (Dehnadi, 2006, Pp. 5 – 14). Hence, it
has to be concluded that it is entirely possible that an ability to
excel at learning the more intricate tasks associated with computing
may be lacking in some individuals, while other individuals may have
been gifted with just the right attributes to understand computing far
better. In the light of the previous discussion, it may well be asked
if it is possible that the gift for readily understanding computing and
allied skills can somehow be detected in the personality of an
individual. Although human determination and interest can overcome many
an obstacle, it is likely to be very useful if it was possible to
correlate the observable attributes of an individual’s personality with
their likelihood of doing well in studying computing. In the light of
such findings, individuals can make more informed choices about what
they wanted to do with their future.
The personality of an individual has been considered in terms of the
observable attributes that individuals possess and which distinguish
one individual from the other. Although a number of different views
about personality have been presented in literature, there are four
broad approaches related to the discussion of personality (Burger,
2003, Pp. 35 – 51). Personality has been considered in terms of the
psychodynamic, the dispositional, the behavioural and the
phenomenological (Devito da Cunha, 2003, Pp. 34 – 43). Whereas the
psychodynamic approach presents personality as an evolution of the
conflict between a person’s basic needs and the demands of the real
world, the dispositional school considered personality to be the result
of a person’s stable internal characteristics. On the other hand, the
behavioural school of thought considers personality to be a person’s
unique patterns of learned behaviour, while the phenomenological school
considers personality to be the result of the way in which a person
perceives and interacts with the real world. Some definitions of
personality have placed an emphasis on the external appearance and
attributes of individuals, while other approaches emphasise all the
attributes that an individual posses. Still other approaches to
understanding personality have considered the interactions between
personal attributes. Personality has been considered in terms of a
person adapting to their surroundings and some researchers have totally
rejected the concept of personality, preferring instead to describe
human behaviour in terms of their response to environmental factors.
Nevertheless, it has to be appreciated that regardless of whether an
individual’s personality emerged as a dictate of the genetic make-up of
a particular individual or in response to their environment, some
distinguishing as well as observable differences between individuals do
exist.
Personality traits have been considered to be the most appropriate way
of describing and studying an individual’s personality (Barkhuus, 1999,
Pp. 1 – 10). It is possible to asses an individual’s personality either
by observing, interviewing or assessing the individual through tests.
Various tests, based on different theories of psychology, have been
developed and these are capable of describing the personality of an
individual who has taken the test as a model based on a particular
theory of personality (Riley guide, 2006, Pp.1). These tests attempt to
judge an individual by the response that they provide to a set of
questions. The tests can either be based on answers to multiple choice
questions or written answers to questions. Responses to multiple choice
questions are preferred because these are easy to administer. The
answers to the questions that are administered helps psychologists to
place a person on personality scales such as Extraversion (E) -
Introversion (I), Sensing (S) - Intuition (N), Thinking (T) - Feeling
(F), and Judging (J) - Perceiving (P) (Devito da Cunha, 2003, Pp. 34 –
43). The Meyer Briggs Test Instrument or MBTI is the most widely used
instrument to measure an individual’s personality on the previously
mentioned scales. This test is, however, not free and investigations
based on this test instrument can be expensive. Some tests such as
those based on the sixteen personality factor model tend to be more
exacting, but personality testing based on this model can be very long
and difficult to administer as well as being expensive because there
are very many test questions that take a considerable time to answer or
administer, with the questionnaire not being free. The sixteen
personality factor model attempts to more accurately place an
individual’s personality as being a combination of scales of warmth,
reasoning, emotional stability, dominance, liveliness, rule –
consciousness, social boldness, sensitivity, vigilance, abstractedness,
private ness, apprehension, openness to change, self – reliance,
perfectionism and tension (Fehriinger, 2004, Pp. 1).
However, the Jung’s personality test that is based on the Jung’s Theory
of Personality is a test that is readily and freely available, placing
an individual’s personality on scales of Extraversion (E) -
Introversion (I), Sensing (S) - Intuition (N), Thinking (T) - Feeling
(F), and Judging (J) - Perceiving (P) (Similarminds.com, 2006, Pp.1).
Attempts can, therefore, be made to correlate a person’s personality
based on their placement by the Jung’s personality test and their
ability in computer literacy. A geek test has also been developed to
determine, on a percentage scale, the extent to which an individual can
be considered to be a geek and this too can be correlated to ability in
computing (Both.net. 2006, Pp. 1). However, the geek test is not a
serious instrument for the measurement of personality.
This research study attempts to explore if there is a correlation
between certain personality types and ability in computer literacy. A
literature review which attempts to present the important themes that
are to be found in the recent and most relevant published literature
has been presented A survey was also conducted, requiring respondents
to take the Jung’s personality test and to indicate their ability in
computing. The responses were statistically analysed using the SPSS
13.0 package in an attempt to discover any correlations as presented in
the chapters below. Discussions and conclusions based on the findings
from literature review and the survey were then presented.
The next section presents a literature review for the topic of discussion.
2. Literature Review
It has been widely agreed that the general cognitive ability of an
individual has a bearing on a large number of outcomes in their life,
their behaviour and performances including those that are associated
with academic achievement, social outcomes and creativity etc (Kuncel,
2004, Pp. 148 and Pp. 156 – 158). Because cognitive ability is a
complex mixture of the genetic make – up of an individual human, their
life experiences and their interaction with the environment, it is
possible that personality too has a correlation with an aptitude for
learning computing. After all, an individual’s personality is likely to
be somehow influenced by their cognitive ability because the models to
explain an individual’s personality also describe this human attribute
in terms of genetic make – up, interactions with the real world and
past experiences. It has been observed that although the ability and
productivity of computer programmers differs significantly between
programmers, the ability to program productively has no correlation
with a programmer’s personality as measured on the Jung personality
test or the Myers-Briggs Type Indicator of cognitive style, the MBTI
(Darcy, 2005, Pp. 2 – 8). However, it is possible that learning may be
somewhat different to ability in solving complex problems productively
(Tukiainen, 2002, Pp. 45 – 46). This conclusion has been reached after
considering the correlation between college admissions tests and
student performance. It has been demonstrated in studies that on the
MBTI test for personality, intuitive persons are more likely to succeed
at programming then the sensitive persons (Tukiainen, 2002, Pp. 46 –
47). Faculty and industry members have agreed that there are certain
characteristics associated with successful programmers including
“pride”, “understanding the user”, “attitude”, “willingness to study at
home”, “strong desire for learning”, “creative talents”, “writing
skills”, “possession of an ability to explain and justify”, “modelling
skills”, “calmness” and “liking for music” (Sterling, 2003, Pp. 419).
However, even though the previously mentioned characteristics of a
successful programmer may have a correlation with personality, it is
difficult to correlate such characteristics with the MBTI personality
types or the personality types presented by other tests for personality
such as the Jung’s personality test.
Studies that have been conducted have certainly found that there is a
correlation between cognitive ability as indicated by performance on a
paper folding task, behavioural attributes, attitude to learning and
success in programming (Simon, 2006, Pp. 1 – 7). Behavioural attributes
indicated by ability in map sketching, (indicative of spatial
knowledge) and phone book searching, (indicative of strategic /
algorithmic style of articulation), have been positively correlated
with an ability in computer programming. In the previously mentioned
study, a paper folding test was designed to test skills in
visualization and spatial reasoning. A significant but weak correlation
between visualization and spatial reasoning was discovered as an
indicator for ability in programming. In the previously mentioned
study, attitude to learning were also found to have a positive
correlation with success in programming. Those who took a “deep
learning approach” to learning computing, involving deeply engaging
with study material were more successful as compared to those who had a
“surface learning approach” that involved a refusal to get deeply
involved with the study material. It is possible that personality may
have some correlation with behavioural attributes, cognitive ability
and attitudes to learning, with such a correlation pointing to a
correlation between personality and success with computer programming.
However, such a link is likely to be somewhat complex.
Despite various attempts to investigate all sorts of correlations
between an ability to learn computing and individual attributes, it has
been widely accepted that some individuals are just not cut out for
success in fields that require a detailed and in – depth study of
computing (Dehnadi, 2006, Pp. 1 – 2). It has also been mentioned in
literature that the “artistic types” are far more likely to fail in
computing then the “nerds”. This is encouraging for those who are
trying to investigate links between personality and ability to learn
computing because these previously mentioned “types” certainly have
distinct personality attributes. The same study, (Dehnadi, 2006, Pp. 1
– 2), indicated that ability in formal logic, mathematical reasoning as
well as an ability to decipher complex patterns in sequences of
numbers, which are also indicators of cognitive ability, are practical
indications of an ability to learn computing. It is claimed that a test
based on the measurement of the previously mentioned abilities, i.e.
ability in formal logic, mathematical reasoning and identification of
complex sequence patterns has been a reliable indicator of an ability
to learn computing. However, the previously mentioned abilities are
cognitive abilities that may be associated with distinct personality
types, i.e. “nerds” rather then the “artistic types”. Therefore, it can
be concluded that cognitive ability and hence an ability to learn
computing does indeed has a correlation with an individual’s
personality. However, such a correlation may be required to be
carefully evaluated and identified through studies.
In a study conducted by Mancy, (Mancy, 2004, Pp. 1 – 7), field
dependency, which is a cognitive characteristic associated with an
individual’s ability to separate an item from an organized perceptual
field has been found to have a link with an ability to excel in
computer programming (Mancy, 2004, Pp. 1 – 7). However, this cognitive
characteristic has also been linked with an individual’s personality
and hence, it must be concluded that there is a link between an
individual’s personality and their ability to learn computing (Brown,
2004, Pp. 24 – 26). Working memory space of an individual’s mind is
another cognitive ability that has a bearing on ability to learn
science. However, a lack of working memory space does not seem to limit
an individual’s ability to learn computing.
An ability to learn mathematics has been strongly correlated with an
ability to learn computing (White, 2002, Pp. 409 – 412). Cognitive
problem solving ability and ability in Piaget's cognitive formal
operations have been indicators of an aptitude for mathematics which is
also an indication of an aptitude for learning computing. However, it
is not certain if these cognitive attributes have a bearing on an
individual’s personality. Computer self – efficacy, or a perception of
one’s own abilities in computing does have a bearing on decisions to
learn computing and these decisions may best be further assisted with
aptitude testing for computing (Smith, 2002, Pp. 1 – 7).In the light of
the previous discussion, a valid question that may be asked is whether
it is possible for an individual who has certain knowledge of the self
to be in a position to modify their personality through a conscious use
of their cognitive ability? Actors are able to present different
personalities depending on the requirements of the role that they are
acting. Hence, it may be concluded that it is cognitive ability that is
really the determining factor in ability to learn computing and an
individual’s personality is somehow related to the unconscious display
of an individual’s cognitive ability. The mind and hence its cognitive
ability develops with time and in an individual who has achieved a
certain level of mental development that results in the knowledge of
the self, it may be possible for the mind to alter personality and
learning ability, at least to a certain degree. Hence, it has to be
asked if an attraction or liking for computers and computing as opposed
to indicators of computer anxiety may also be indicators of ability to
learn computing (Khorrami, 2001, Pp. 17 – 24)? Any links between a
liking for computers or computer anxiety and personality may also be
worth investigating. It is highly likely that such links do exist
because as has been previously mentioned, the “artistic types”
generally do not like computers and are more likely to suffer from
computer anxiety as compared to the “nerds” who find computers to be
pleasing. Artistic types have also exhibited an inferior ability for
self control as compared to the nerds who do not like taking risks and
this too may well have something to do with cognitive ability.
Despite the previous discussion on personality and cognitive ability
having a bearing on aptitude for learning computing, it has to be
remembered that there are a large number of environmental factors which
may also have a bearing on the success or failure of a student in an
academic endeavour. Financial hardship when studying is an example of a
factor that can have a significant impact on the ability of an
individual to perform well despite ability or previous learning
(Roddan, 2002, Pp. 4 – 10), (Bruer, 2005, Pp. 1) and (Jenkins, 2001,
Pp. 53 – 57). However, certain traits have been mentioned as being
associated with successful scientists and these traits may be
considered to be loosely associated with personality (Jarrard, 2006,
Pp. 197 – 213).
At this juncture, it is appropriate to consider just exactly what
personality is and how this rather unique set of human attributes is
viewed by the psychologists. Personality refers to the permanent or
very slowly changing patterns of thought, emotion and behaviour that
are associated with an individual (Momberg, 2004, Pp. 16 – 38) and
(Devito da Cunha, 2003, Chapter 3). Apart from the previously mentioned
theories of personality which have been briefly discussed in the
introduction, the Five Factor Model is mentioned as a descriptive model
of personality that is widely used and which uses the five dimensions
of extraversion, agreeableness, conscientiousness, neuroticism, and
openness to associate an individual with a personality type. The theory
of personality types is often used to describe personality. The most
influential of the theories of personality types is the Jung’s theory
of personality that is the work of Carl Gustav Jung, a Swiss
psychologist whose ideas have profoundly shaped the field of
psychology. However, it was Allport who had first successfully used the
concept of traits to describe personality (Barkhuus, 1999, Pp. 2 – 12).
The theory of personality types as it stands today places an
individual’s personality on scales of extroverted or introverted,
sensing or intuitive, thinking or feeling and judging or perceiving.
Based on these scales, an individual may have an ENTJ personality, as
an example, which stands for a predominantly extroverted, intuitive,
judging and feeling personality. It has been found that most of the
students who make it to a postgraduate course in computing in the
United Kingdom are either of the SIFJ or the SITJ type personality
(Chandler, 2003, Pp. 5). However, there were marked differences in the
student population with regard to extrovert / introvert and judging /
perceiving categories. Women can be more positive about computers then
men (Ray, 1999, Pp. 7).
Personality has certainly influenced vocational interests of
individuals (Momberg, 2004, Pp. 69).Personality has also been observed
to have an influence on aptitude for learning computing amongst certain
racial groups in South Africa (Burger, 2003, Pp. 112). Personality has
been found to have an influence on the ability for certain activities
in computing such as the task of code – review in programming (Devito
da Cunha, 2003, Chapter 8). Intuitive type personalities or N - types
on the MBTI seem to have a positive correlation with ability to program
as well as ability for code – review. Information – seeking behaviour
has a relationship with computing because the applications of computing
are somewhat related to the management of information and this has been
found to have a certain correlation with an individual’s personality
(Heinström, 2003, “Conclusion”). Personality traits have been found to
have an influence on the learning behaviour of individuals (Heinström,
2003, “Discussion”). There is a significant link between a strategic
approach to learning and a conscientious personality. A significant
link also exists between surface approach to learning, which is
indicative of a lack of ability to learn computing and the personality
trait of neuroticism. However, the “comfort level” of an individual
with computer science and computers has been mentioned as being the
best indicator for success in a computing course (Wilson, 2002, Pp. 17
– 21).
Hence, there is an overwhelming body of evidence which indicates that
there is some sort of a link between personality and the ability to
learn computing. However, there is also a link between cognitive
ability and ability to learn computing and this logically indicates
that there is a link between cognitive ability and personality. Hence,
a study to try and find a correlation between personality and learning
ability in computing is likely to be interesting.
The next chapter presents the methodology for a study to examine any
correlation between personality and an ability to learn computing.
3. Research Methodology
This study, which attempts to examine if there is any correlation
between an individual’s personality and their ability to learn
computing has to rely on a literature review as well as an empirical
investigation in order to gather evidence that may support or oppose
the hypothesis being examined. It is always necessary to learn from any
conclusions which other researchers and scholars may have reached and
the literature review assists in gathering any evidence which may be
available as a result of scholarly investigations that may have been
carried out. The empirical investigation involves gathering of data and
its statistical analysis on order to try and obtain evidence that can
shed some light on the topic of investigation. Any investigations that
are conducted may well provide new insights into the research topic or
challenge some of the existing views that may have been held about the
correlation of personality and an ability to learn computing (Collins,
1999, “Chapter 1) and (Marshall, 1999, Chapters 1 and 2).
There are two traditions of research which have developed and which
have evolved over time along with their terminology, methods and
techniques. These research techniques have been referred to by
different researchers with different names such as qualitative or
quantitative traditions in research, humanistic and scientific
traditions or the positivist and phenomenological traditions of
research. Although the names may be different, they refer to the same
distinctions in the processes of conducting research. The qualitative
tradition consists of case study methods, ethnography and historical as
well as action research and the quantitative tradition consists of
methodologies such as survey research, experimental and
quasi-experimental research as well as research after the occurrence of
a factual event (Collins, 1999, Chapters 1 – 5) and (Marshall, 1999,
Chapters 1 – 4).
It is possible to express facts as objective reality which can be
expressed as quantities and this forms the basis of the positivist
tradition of research which is quantitative research. Such research
relies on numbers, measurements and experiments to derive numerical
relationships under conditions of controlled behaviour that can be
manipulated. On the other hand, the phenomenological tradition attempts
to describe and understand reality that is perceived with narratives
and observations being used to focus on understanding and meaning in
order to yield knowledge and understanding (Collins, 1999, Chapters 1 –
5) and (Marshall, 1999, Chapters 1 – 4).
Situations and settings may be generalised in order to understand
events and why they occur, resulting in predictions that can be made as
a consequence of the research process. Explanations are possible and
understanding develops as a result of seeing things happen. Attempts
can be made to fit observed truths into patterns and deductions can be
made from known truths. Elements being investigated need to be related
to other elements so that the overall picture forms into a unified
model with the unification forming the explanation. An explanation is
possible for something when it can be understood. Rich descriptions are
required for understanding so that relationships may be formed between
different parts. Pattern models result from attempts to fit together
things and the understanding of patterns is the result of research of a
qualitative tradition. Literature review assists in this form of
research. On the other hand, the quantitative method of research uses
the more basic facts or laws in order to determine that which is
required to be explained, so that a deductive model may be constructed
(Collins, 1999, Chapters 1 – 5) and (Marshall, 1999, Chapters 1 – 4).
A distinction exists between prediction and generalisation. Unknown
parts of a pattern cannot be deduced from a known part and this
requires that the deductive model be used in order to maintain the
symmetry of prediction and deduction. Generalisations can become
complex and there may be a great deal of conflict and scrutiny
(Collins, 1999, Chapters 1 – 5) and (Marshall, 1999, Chapters 1 – 4).
Qualitative research is regarded as being more important in situations
when there is a requirement to investigate complex interrelationships
in the more natural or real life situations with a possibility of using
this methodology of research to test theories that have already been
developed. The qualitative and quantitative methods of research usually
work together, with quantitative research being used to further test
theories that have been developed using the qualitative research. On
the other hand, qualitative research is often used to further explain
the results of quantitative investigations. Qualitative methods are,
therefore, useful for rich descriptions of issues being studied with
hopes of achieving better understandings with predictions not being the
main aim and generalisations taking the form of natural generalisation.
Hence, the qualitative research methodology coupled with a quantitative
investigation that was adopted to study the ingredients of a
correlation of personality with ability to learn computing is in fact
the appropriate one (Collins, 1999, Chapters 1 – 5) and (Marshall,
1999, Chapters 1 – 4).
Suitable literature was selected for the literature review as a result
of literature searches that were conducted using internet search
engines and through a process of querying the database of reputed
libraries. Abstracts and keywords were used to select promising
material and this was further narrowed down after considering the
content.
The review of literature indicated that there is a complex correlation
between personality and ability to learn computing. A stronger
correlation between cognitive ability of an individual and ability to
learn computing was available from the literature that was examined for
the literature review. However, because cognitive ability and the
complex brain process that are a characteristic of the human brain also
have a bearing on an individual’s personality, therefore, it was
decided to try and attempt to carry out another empirical study or a
survey that had the capability to shed some light on the research
topic.
There are many variables that are involved in the topic of the
investigation. An ability to learn computing can also be influenced by
environmental factors, the learning environment, teaching methods, any
psychological stresses which an individual may have been subjected to,
age, gender or distractions etc. However, such influences can be
minimised by considering a sufficiently large number of respondents who
have been randomly selected. It was, however, decided to concentrate on
the younger segment of the population whose cognitive ability had not
sufficiently developed for them to be in a position to control their
personality. It was also decided that it was most appropriate to try
and select a totally random set of respondents for the survey who are
geographically spread and are living in different environments so that
the impact of culture and conditions that may exist within a region may
be randomised. Yahoo groups, a collection of mailing lists, presented
an interesting possibility for sending a survey questionnaire to a
large number of individuals from very diverse backgrounds. A survey
email account was created and this email was linked to suitable mailing
lists on which a single email containing the survey questionnaire along
with instructions was emailed. It was felt that a single email asking
individuals to participate in a scientific research survey did not
constitute SPAM. A total of 50,000 possible respondents were contacted
in this way. The questionnaire that was designed along with the
instructions has been presented in (Appendix A).
The questionnaire that was designed consisted of a section that
attempted to measure attributes of an individual’s personality using
the Jung personality test (Similarminds, 2006, Pp. 1). This test is
widely used and can be administered for free. The Myer Briggs
personality test or the MBTI is very widely used to assess personality,
but unfortunately this test required payments to be made for testing
and it is also lengthy (Riley guide, 2006, Pp.1). The sixteen
personality factor test is also available, but this too is a lengthy
set of questions that are not free. A “geek test” was included in the
questionnaire to add some humour and also to add another rather
unconventional dimension of personality for testing (Both.net. 2006,
Pp. 1). A further set of questions was added in order to measure an
interest or ability in computing for individuals who are young and non
– professionals (Burger, 2003, Pp. 126). It was felt that “interest” in
computing rather then “proficiency” in programming languages or other
computing skills was a more appropriate measure of ability in learning
computing. Even though the shortest tests for personality had been
selected, the number of questions that an individual had to answer in
order to complete the whole questionnaire amounted to about 80
questions. It was feared that such a questionnaire, which was shorter
then most questionnaires for testing personality, may deter
participants who cannot be forced to participate in surveys and that
any response was also likely to take substantial time in coming. Hence,
a smaller manual survey with some monitory inducement for responding
was also prepared as a back – up. At least twenty to thirty responses
are required for any kind of a decent statistical analysis to be
possible.
The data which was gathered was analysed using the SPSS 13 statistical
package. Depending on the answers that are provided by a respondent for
the personality test, a personality type is generated which indicates
the percentage values for personality traits on different personality
scales as shown below.
Extroverted (E) 50% Introverted (I) 50%
Intuitive (N) 58.33% Sensing (S) 41.67%
Thinking (T) 51.61% Feeling (F) 48.39%
Perceiving (P) 70.83% Judging (J) 29.17%
Personality type: ENTP
The personality type and the percentage values of different traits are
variables that are related to the personality of an individual. The
answers that are provided for the geek test will provide a percentage
value that is an indicator of the “Geek Character” of a respondent. The
answers that are provided for the questions that are asked in relation
to a likeness or comfort for computers and a level of desire / comfort
for working with computers are used to generate an indicator for an
ability to learn computing as a percentage value. Any correlation
between the variables that have been mentioned was then investigated
using correlation analysis, regression analysis and the analysis of
variance or ANOVA technique as discussed in the next section.
The next section takes a look at the project management and quality aspects of the statistical survey which was carried out.
4. Survey Project Management
Surveys are a means of gathering information and information is used to
make business decisions or decisions about the future. Hence, it is
important that the information that is gathered should be as reliable
and accurate as possible so that the decisions that are made on the
basis of the information collected are the right decisions. Time, money
and effort are spent on designing as well as conducting surveys. Thus,
it is important that surveys should be carefully designed and managed
in order to provide value for money and to provide quality of
information for the effort that has been expanded. Some of the ways of
gathering information include literature search, interviewing people,
the use of focus groups to explore ideas and attitudes of individuals,
personal interviews, telephone surveys, postal mail surveys and the use
of email or the internet to gauge individual opinion in a sample
population (Walonick, 2004, Pp. 1 – 10).
Email or internet surveys are relatively new techniques for gathering
information and these methods for conducting statistical surveys
substantially reduce the cost as well as effort involved in gathering
information. However, survey professionals have advised that email or
internet surveys should be kept as short and simple as possible, being
targeted to specific groups such as in-house employees, individuals
within an organisation, professionals or pre-sampled populations who
are interested in taking a survey. Fears have also been expressed that
because email or internet surveys are relatively easy to administer,
they may loose their effectiveness as a result of being too common with
many individuals receiving a number of invitations for participating in
such surveys (Shannon, 2002, Pp. 1). It is possible to conduct email
and web – based surveys electronically and the web – based approach
offers greater appeal because data can be automatically stored in a
database with validation being possible. However, web – based
approaches to conducting electronic surveys are prone to multiple
entries, server malfunctions and can suffer as a result of a lack of an
ability to offer incentives for participation which may be a crucial
for persuading individuals to participate in the survey. There may be
concerns associated with privacy and confidentiality of individuals
when conducting email or internet surveys (Andrews, 2003, Pp. 5 – 15).
All surveys which are conducted can be thought of as projects
consisting of associated tasks which require time and money to complete
and there are issues of quality associated with these tasks that can
influence the overall product quality, i.e. the quality and reliability
of the information or knowledge gleaned from the survey results. It is,
therefore, important that issues of project scheduling, risk management
and quality be well thought out for all surveys (Walonick, 2004, Pp. 1
– 20) and (National Centre for Education Statistics, 2006, Pp. 1).
This section presents a discussion of the project management and
quality issues associated with the administration of surveys in general
and this project in particular.
4.1 Project Schedule
Some of the tasks that are associated with a project that involves the
administration of a survey include goal clarification, overall study
design, selecting a sample population, designing a questionnaire and
cover letter, conducting a pilot test survey, revising the
questionnaire if required and the mailing, emailing as well as awaiting
or reminding for participation in the survey. Once feedback from
respondents has started to arrive, tasks associated with the editing of
data and coding of open – ended questions, data entry and verification,
data analysis report preparation as well as report distribution must
also be accomplished (Walonick, 2004, Pp. 5). If responses are
inadequate, as was the case for this project, then a back – up survey
using face –to- face interviews with incentives is appropriate. Each of
the previously mentioned tasks or activities has costs associated with
a task as well as the minimum and maximum time in which a task can be
completed. Thus, the conduct of a survey lends itself to the usual
methodology for project management including PERT, CP/M and Gantt
Charts etc (Mind Tools, 2006, Pp. 1).
Microsoft Project can be readily used as a project management tool for
managing surveys and Gantt Charts as well as many other types of
analysis related to project management for surveys may be conducted by
using this project management software. Although statistical surveys
that are required to be conducted are of different types involving
different resources and complexities, a simple Gantt chart and work
breakdown structure for the simple survey that was carried out for this
project are presented below:
Gantt chart for the Correlation between Personality and Computer Literacy Project
Task Time Involved in Task (Days)
Minimum Expected Maximum
1. Goal
clarification
0.25 0.50 1.00
2. Study design 1.00 2.00 3.00
3. Selecting the sample 0.25 0.50 1.00
4. Designing the questionnaire
and cover letter 0.50 1.00 2.00
5. Conduct pilot test 2.00 3.00 4.00
6. Examine and Revise questionnaire (if necessary) 0.50 1.00 2.00
7. Locating the sample (if necessary) 0.50 1.00 1.50
8. Wait for response
time 3.00
4.00 5.00
9. Send reminders to get non-respondents 2.00 3.00 4.00
and conduct back – up face –to- face
survey.
10. Editing the data and
coding open-ended questions 2.00 3.00 4.00
11. Data entry and verification 3.00 4.00 5.00
12. Analyzing the data 2.00 3.00 4.00
13. Preparing the report 3.00 4.00 5.00
14. Printing & distribution of the report 2.00 3.00 4.00
Expected duration of survey related activities: 22 days 33 days 45.5 days
Resources allocated to the project: 1 trained professional and a
computer with statistics software + internet connection. Hence, the
above figures can also be considered in terms of man days.
Work Breakdown Structure for the Correlation between Personality and Computer Literacy Project
It should be noted that the previous Gantt chart and work breakdown
which has been presented for this simple survey project may be
considered to be trivial. However, very lengthy and complex statistical
surveys are also possible in real life. Examples of more complex
surveys include a survey related to the health of a national population
or a survey for the educational performance of children in schools in a
region. Such complex surveys need appreciable resources to be allocated
to them and in certain situations, the project has to be managed
carefully to deliver reliable results in time and within the allocated
resources. The project management techniques which have been mentioned
previously can, therefore, become important for large and complex
surveys. Sometimes, a fixed set of resources such as man power or funds
may be required to be allocated to a project. In such a situation, the
survey managers need to be able to determine with a reasonable degree
of accuracy the funds, resources and the time duration of a survey.
Experienced survey managers will be able to assign probabilities
associated with the minimum, expected and maximum times for the
completion of tasks associated with a survey so that the probability of
completing a survey within an estimated time can be determined. This is
illustrated below (Walonick, 2004, Pp. 5) and (Palisade, 2005,
Downloaded Demo Software and Tutorial):
Task Uncertainty Presented as a Normal Distribution (Palisade Demo Software)
If probabilities can be assigned to the completion times associated
with various tasks in a survey, then it is possible to estimate the
completion time for a survey with a certain level of probability being
assigned to this occurrence using project risk estimation software such
as the @RISK software by Palisade (Palisade, 2005, Various Sections).
Probabilistic times for the completion of a complex survey can then be
estimated and this often leads to a probabilistic estimation for the
completion costs associated with a survey. Such estimates can be of
great assistance when professional organisations are bidding to carry
out surveys for a client and need to estimate an amount to charge for
their services. The graphs below illustrate the probabilistic
completion times for a complex survey and these estimates can provide
probabilistic costs associated with a survey (Palisade, 2005,
Downloaded Demo Software and Tutorial). It is, therefore, possible to
estimate, manage and control surveys in a professional manner.
Typical Results of the @RISK Project Simulation Showing Probabilistic
Time for Completion of a Project with Task Uncertainties (Palisade Demo
Software)
Costs are important, especially for large surveys. The cost associated
with a survey depends on the size of the sample from which a survey is
required to be conducted and the resources that are allocated to a
survey as well as how quickly it is desired to have the results. For
larger and complex surveys, the resources that are used can be
described on a software management tool such as Microsoft Project and
this software tool when used with a project risk analysis tool such as
@RISK by Palisade which will provide probabilistic estimates of the
costs associated with conducting a survey. However, for a simple survey
such as the one which was carried out for this project, the following
cost estimation will adequately describe the associated costs
(Walonick, 2004, Pp. 5):
Activity Cost (GBP)
Proposal typing and survey design. 50
Cover letter and questionnaire design. 50
Emailing of survey questionnaires. 10
Following up on non-respondents. 10
Mailing list cost 20
Payments for survey web pages. 150
Payment for backup survey in a cyber café 100
Incentives. 50
Data entry and verification. 50
Statistical analysis programmer. 50
Preparation of the final report. 100
_______
APPROXIMATE TOTAL: 640
_________
Approximate Survey Costs for the Survey conducted to Correlate Personality and Computer Literacy
It should be noted that the previously mentioned costs are only
approximate costs and that incentives were needed for the back – up
survey for this project because adequate responses were not received
for the email survey.
The next two sections take a look at the management of risks and quality in statistical surveys.
4.2 Risk Management
The risks that are associated with conducting a survey can be broadly
categorised as follows (Andrews, 2003, Pp. 1 – 20), (Walonick, 2004,
Pp. 1 – 20) and (Office of Management and Budget. 2006, Sections 1 –
6):
- Risks associated with gathering inaccurate data that is of poor
quality and which does not reflect the reality of a situation.
- Risks which are associated with the management of a survey and which
can result in cost overruns, failure to complete a survey within the
projected time frames or inability to get a sufficient number of valid
responses.
- Risks associated with privacy and data protection concerns of those who have been invited to participate in a survey.
- Risks associated with presenting a poor image of those who are
conducting the survey as a result of poorly designed questionnaires,
email campaigns, survey forms, web pages or communications.
The previously mentioned risks are somewhat interlinked. A well
designed survey form on nicely presented web pages with a
professionally presented email campaign is likely to get people
interested in the survey. If individuals are interested then it is more
likely that they will respond and data will be available for analysis.
Spamming into the emails of others with poorly designed email campaigns
for a survey can also result in reports to law enforcement agencies,
blocking of any further communications or abusive responses. Greater
accuracy of the data that is being gathered is possible with well
designed survey questionnaires which keep people who are participating
interested. It is also important to ensure that multiple responses from
kids mucking around with on-line forms are avoided by sending out
password coded email invitations to those who are likely to be
responsible and will respond because they have an interest in the
survey. The key to conducting a successful survey is to concentrate on
quality at all stages of the survey. Quality is important when defining
the goals of a survey, designing the survey, designing the
questionnaire, designing web pages and forms for the survey as well as
when analysing the results or processing the data. Longer survey
questionnaires get fewer responses then shorter ones and incentives can
be very effective in getting individuals to respond. As an example, if
a lucky draw to win a computer or even a car is an incentive for those
who are participating in a survey, then the response rate is likely to
be terrific. Such an offer can be included in internet surveys, but it
is generally less likely that individuals will respond to a lengthy
email survey because small incentives such as a chocolate or a box of
tissues cannot be readily given over the internet and the body language
is missing in email or web communications. Hence, it is always better
to have at least some of the sampling for a survey being conducted
through face –to- face encounters or interviews so that at least some
reliable data is available (Walonick, 2004, Pp. 1 – 20) and (Office of
Management and Budget. 2006, Sections 1 – 6).
The title for a questionnaire should be kept short and simple, lengthy
questions should be avoided, the language in all communications as well
as the questionnaire should be simple, considerations for convenience
as well as presentation and asking questions which are likely to evoke
the truth are all important for getting reliable responses. It is also
important to give a survey adequate time to run so that a reasonable
number of responses can be collected. The progress of a survey should
be carefully monitored so that additional efforts can be made to
acquire further responses if responses from efforts that have been made
are not forthcoming. Unfortunately eliciting responses is a bit like
selling and those who may want to part with their opinions cannot be
forced to do so (Walonick, 2004, Pp. 1 – 20) and (Office of Management
and Budget. 2006, Sections 1 – 6). In this project, the responses that
were received for the internet survey were miniscule and when it was
obvious that this was the case, a face –to- face survey for a sample
with incentives for the respondents was carried out as a back - up.
Every effort was made to minimise the length of the questionnaire, but
the information which was required to be collected required that the
questionnaire was to be of an above average length.
The next section takes a look at issues related to quality management in a survey.
4.3 Quality Management
The concept of quality that is associated with a statistical survey may
be considered in terms of the following seven dimensions (Lynn, 2001,
Pp. 1 – 10), (Biemer, 2003, Chapters 1 – 5) and (Centre for Health
Promotion, 1999, Chapters 1 – 5):
- Relevance of the data gathered to the information that is required.
This attribute of quality is related to the purpose for which a survey
was conducted. A statistical survey is required to provide information
which answers the questions that were required to be answered and for
which a survey was carried out. This means that the right questions
have to be asked from the right people in the right way.
- Accuracy of the information collected. Every survey will present a
result that is slightly different from the reality in a population.
This error arises because “errors of non-observation” in which a
sufficiently large portion of the total population could not be sampled
and also because there were errors in measurements or the data which
was collected. Errors arise because the questionnaire which was
designed failed to ask the right questions or because an insufficient
number of responses were able to be collected. Interviewers can fail to
ask the right questions or they can corrupt the data that is being
gathered. The respondents may not want to answer questions honestly and
there may be processing errors or analysis errors when the data which
has been gathered is processed or analysed. Various sources of error
can add up to increase the overall error in the survey results.
- The timeliness of the information gathering refers to a survey being
able to provide results in a timely manner so that inputs can be
included for the decisions that are required to be made. A planned
survey will be designed to be completed in a time frame which will make
it possible for the information gathered to be available for decisions.
However, if a survey slips from its planned schedule then it may not be
possible to provide reliable information from a survey for decision
making.
- Accessibility of results and information refers to the availability
of information that has been gathered for users. The cost involved in
providing survey results, the availability of processed or raw data as
well as the medium involved in providing access to information have an
influence on accessibility.
- The interpretability of the information collected. This refers to the
survey results including sufficient additional information that makes
it possible to interpret the information which has been gathered. As an
example, if there are insufficient numbers of responses then it should
be possible to determine why a poor response has been presented.
Interpretability is enhanced by providing metadata or additional
information about the survey, samples of population to which the survey
was presented and the time as well as the manner in which the survey
was conducted. Any additional observations which can explain the survey
results will help enhance interpretability.
- The coherence of the information that is provided as a result of the
survey refers to the correlation that is possible between results from
different samples or even different studies which may have been carried
out. Coherence is a measure of the reliability of survey results.
- The costs associated with conducting a survey is also sometimes
considered to be a measure of survey quality. Efforts should be made to
gather the requisite information economically so that reliable results
can be provided to others in a cost effective manner. This often
requires that the correct decisions are required to be made by the
survey managers to minimise costs without unduly compromising
reliability of information that has been gathered.
For the survey which was conducted, efforts were made to properly
design the questionnaire, sample a sufficiently large number of
individuals, minimise costs, conduct a small back – up survey from a
sample which was provided with incentives and to correlate results with
other studies in order to enhance the quality of the survey report.
However, much more could have been done if more funds and time was
available.
The next section of this report takes a look at the statistical analysis for the survey.
5. Statistical Analysis
Although a total of 50,000 possible respondents were contacted, the
response rate was extremely low. Only two individuals had responded by
February 11, 2006. It was only the backup paid survey that was
conducted which produced any results whatsoever. A total of 20
individuals responded to this paid survey. Out of the two responses
that were received for the email survey, one was invalid because the
individual responding to the questionnaire failed to respond to all the
questions asked and had filled in the responses to two questions only.
Hence, after trying to elicit responses from very many individuals,
there were a total of only twenty responses that could be included in
the study.
Every complete response that was received was assigned a personality
type that was measured using the Jung’s personality test and the
percentage value that was assigned to a personality type on the
Extraversion (E) - Introversion (I), Sensing (S) - Intuition (N),
Thinking (T) - Feeling (F), and Judging (J) - Perceiving (P) scales was
also included in the study. The geek test is not a serious test of
personality and this test was included in the questionnaire only to add
a bit of humour to the questionnaire by asking about the observed
personality traits of those individuals who have a lifestyle which
makes them very engrossed in computers as well as computing. Each
response for the geek test has four possible answers and the answer C
in the test indicates a maximum correlation with the Geek personality.
The number of C answers that an individual has provided are counted and
divided by the number of questions to determine the percentage value of
geek in an individual’s personality.
The questions for the measurement of an ability to learn computing can
elicit a set of responses that range from “Strongly Disagree” to
“Strongly Agree”. Depending on the question that is asked, the answer
will either strongly support an aptitude in computing or will indicate
a strong aversion to studying computing. A strong aptitude for learning
computing indicated by a given response can be assigned 4 points while
a strong aversion can be assigned 1 point. Agreement with an aptitude
to learn computing can be assigned 3 points and disagreement with an
ability to learn computing can be assigned 2 points. An individual with
the strongest ability to learn computing can possibly gain a total of
120 points in the questionnaire by providing answers which indicate a
strong aptitude for learning computing on all 30 questions. The marking
scheme to assign a value for ability for learning computing for a
respondent is shown below for the set of questions asked:
Strongly Disagree Disagree Agree Strongly Agree.
1. Computers do not scare me 1 2 3 4
2. I am no good with computers 4 3 2 1
3. I like working with computers 1 2 3 4
4. Working with computers makes 4 3 2 1
me nervous
5. I like trying new problems with 1 2 3 4
computers
.
.
.
10. I feel aggressive and hostile 4 3 2 1
About computers
.
.
.
15. I like sticking with a 1 2 3 4
computing problem
.
.
.
.
30. I do not enjoy talking about 4 3 2 1
computers.
After examining the responses provided by an individual who has
completed the computer ability questionnaire, their total marks for the
questionnaire can be assigned and this divided by 120 and then
multiplied by 100 will provide their percentage score for the ability
to learn computing. Hence, a percentage value can be assigned to all
variables for the survey, except the personality type variable which is
indicated by personality types that are the strongest for an
individual, such as ISFJ or ENFJ etc.
The results for the twenty responses that are included in the questionnaire are shown in the table below:
No. Age Personality Extroverted % Introverted % Sensing % Intuitive %
1 45 ISFJ 29.63 70.37 56.76 43.24
2 28 ENFJ 60 40 51.61 48.39
3 18 ISFP 41.67 58.33 51.52 48.48
4 32 ESTJ 55.56 44.44 55.56 44.44
5 32 ISTJ 40.74 59.26 51.52 48.48
6 24 ESFP 62.96 37.04 66.67 33.33
7 19 ESTJ 58.62 41.38 55.56 44.44
8 25 ISFJ 48.78 51.22 54.05 45.95
9 23 ISTJ 46.67 53.33 54.84 45.16
10 31 ENFP 54.55 45.45 42.86 57.14
11 23 INTJ 47.62 52.38 50 50
12 20 ESFP 53.13 46.88 50 50
13 42 ENTJ 54.55 45.45 48.57 51.43
14 28 ESTP 54.84 45.16 64.52 35.48
15 16 ENFP 52.94 47.06 45.16 54.84
16 35 ENTJ 51.35 48.65 44 56
17 19 INTP 47.5 52.5 43.59 56.41
18 19 INTP 47.5 52.5 43.59 56.41
19 26 INTJ 42.86 57.14 41.94 58.06
20 31 ISTP 56.67 43.33 53.33 46.67
Feeling % Thinking % Judging % Perceiving % Geek Test % Learning Ability %
51.28 48.72 52.78 47.22 36 80.83
54.05 45.95 51.72 48.28 14.29 65
56.76 43.24 45 55 14.29 62.5
42.5 57.5 59.38 40.63 28.57 66.67
42.86 57.14 59.26 40.74 21.43 75
58.33 41.67 45 55 28.57 63.33
50 50 50 50 35.71 58.33
53.49 46.51 55.56 44.44 35.71 55.83
42.86 57.14 53.85 46.15 35.71 58.33
60 40 46.15 53.85 28.57 64.17
47.73 52.27 52.17 47.83 28.57 67.5
57.58 42.42 43.33 56.67 28.57 66
39.29 60.71 51.61 48.39 14.29 60
48 52 44.12 55.88 14.29 64
52.5 47.5 48.15 51.85 35.71 63.25
46.88 53.13 54.05 45.95 35.71 54
41.67 58.33 39.39 60.61 28.57 64
41.67 58.33 39.39 60.61 42.86 65.83
45.95 54.05 51.61 48.39 35.71 62.5
44.44 55.56 44.12 55.88 21.43 68.33
Data Collected from Survey
The data that was gathered was processed using SPSS 13 and the following descriptive statistics were obtained:
Descriptive Statistics
N Minimum Maximum Mean Std. Deviation
Age 20 16.00 45.00 26.8000 7.89803
Valid N (list wise) 20
Descriptive Statistics
N Minimum Maximum Mean Std. Deviation
Age 20 16.00 45.00 26.8000 7.89803
Extroverted % 20 29.63 62.96 50.4070 7.78407
Introverted % 20 37.04 70.37 49.5935 7.78389
Sensing % 20 41.94 66.67 51.2825 6.80970
Intuitive % 20 33.33 58.06 48.7175 6.80970
Feeling % 20 39.29 60.00 48.8920 6.33435
Thinking % 20 40.00 60.71 51.1085 6.33451
Judging % 20 39.39 59.38 49.3320 5.86943
Perceiving % 20 40.63 60.61 50.6685 5.86852
Geek % 20 14.29 42.86 28.2280 8.82753
Ability % 20 54.00 80.83 64.2700 6.07781
Valid N (list wise) 20
Descriptive Statistics for the Survey Data
The hypothesis that has to be tested from the research data that has been collected is as follows:
H0: There is a correlation between personality trait variables such as
extroversion, introversion, sensing, intuitiveness, feeling, thinking,
judging, perceiving, geek nature and ability in computing.
Pearson’s correlation coefficient is appropriate to check for
correlation and the following ranges of this coefficient are considered
(Momberg, 2004, Section 4.3):
- r = 0.01 to 0.20 indicates weak relationship
- r = 0.21 to 0.50 indicates moderate relationship
- r= 0.51 to 0.80 indicates a strong relationship
- r= 0.81 to 1 indicates a very strong relationship
The correlations table below presents correlations between personality trait variables and ability to learn computing:
Correlations between Personality Trait Variables and Ability to Learn Computing
Complete Correlation Table for all Survey Variables
A regression analysis for the personality variables extroversion,
introversion, sensing, intuitiveness, feeling, thinking, judging,
perceiving and ability to learn computing produced the following
results:
Regression Analysis for Personality Trait Variables and Ability to Learn Computing
ANOVA’s for personality variables and ability to learn computing are presented below:
ANOVA
Sensing %
Sum of Squares df Mean Square F Sig.
Between Groups 615.888 16 38.493 .435 .883
Within Groups 265.180 3 88.393
Total 881.068 19
ANOVA
Feeling %
Sum of Squares df Mean Square F Sig.
Between Groups 658.403 16 41.150 1.188 .509
Within Groups 103.952 3 34.651
Total 762.355 19
ANOVA
Thinking %
Sum of Squares df Mean Square F Sig.
Between Groups 658.443 16 41.153 1.188 .509
Within Groups 103.952 3 34.651
Total 762.395 19
ANOVA
Judging %
Sum of Squares df Mean Square F Sig.
Between Groups 614.109 16 38.382 2.847 .211
Within Groups 40.444 3 13.481
Total 654.553 19
ANOVA
Perceiving %
Sum of Squares df Mean Square F Sig.
Between Groups 613.908 16 38.369 2.846 .211
Within Groups 40.444 3 13.481
Total 654.352 19
ANOVA
Geek %
Sum of Squares df Mean Square F Sig.
Between Groups 1149.213 16 71.826 .650 .757
Within Groups 331.367 3 110.456
Total 1480.580 19
Analysis of Variance for Personality Trait Variables and Ability to Learn Computing
The previously mentioned statistical analyses are all that are likely
to be required for an analysis of this survey data. Out of all the
statistical analysis that was performed, the correlation analysis is
the most important and this is necessary to test the hypothesis under
consideration, i.e. that personality traits do have a correlation with
the ability to learn computing. From the correlation table that was
produced as a result of the correlation analysis in which Pearson’s
correlation coefficients have been presented, the following conclusions
may be drawn after considering the significant value ranges of the
Pearson’s correlation coefficients that have already been presented:
- The personality trait of extroversion had a significant correlation
with an ability to learn computing, with a Pearson’s correlation
coefficient of -0.488. This will tend to indicate that the more
extroverted that an individual is, the less likely they are to have an
ability to learn computing. Extroversion is negatively correlated with
introversion and this is to be expected because they are the extremes
of the same Extroversion – Introversion personality scale. Hence, an
introverted individual is more likely to have the ability to learn
computing.
- Age is positively correlated with an ability to learn computing and
this indicates that as an individual gets older, they are more likely
to become better at learning computing. Age is also positively
correlated with introversion, sensing, thinking and judging. This means
that according to the results of the survey, as an individual gets
older they are more likely to become introverted, have a better sensing
ability, are likely to be more thinking personality and are also likely
to be good at judging. An older person is also less likely to have a
“geek” element to their personality. These results are in agreement
with what is widely believed as a result of observations that many
individuals have made about the personality of older individuals.
- Sensing is positively correlated with ability to learn computing,
while intuitiveness has been found to be negatively correlated with the
previously mentioned variable. This may be regarded as being something
sensible because sensing is about reacting to what is observable in the
real environment, while intuitiveness is about what can be “guessed”.
Hence, those individuals who are more in tune with reality are the ones
who are more likely to be successful at learning computing.
- Feeling is negatively correlated with ability to learn computing,
while the thinking personality trait is positively correlated with the
ability to learn computing. However, the level of correlation with a
Pearson’s coefficient of 0.03 indicates that the correlation between
thinking and ability to learn computing is not very strong. A thinking
person is likely to be a “geek” and is also likely to have the
personality traits of introversion, intuitiveness and perception.
- A judging personality is more likely to have the ability to learn
computing rather then a perceiving personality. Judging involves
observing and then making mental judgements based on such observations.
It has been found from practical observations this is what most
successful computer professionals do. A perceiving person does observe
the environment, but may lack the ability of transforming these
observations into mental models that are so important for developing a
cognitive understanding of the world and the environment in which an
individual lives.
- Surprisingly, contrary to popular beliefs, geeks do not have a
positive correlation with an ability to learn computing. This does not
mean that they do not try very hard or are not interested in computing,
but that such a negative correlation can mean a large number of things
including the observation that geeks may not be very bright. Hence,
despite spending very long hours in their lonely world, geeks may still
not be able to make much of a sense out of computing. However, it has
to be remembered that the geek test that was included in the
questionnaire is not a serious test of personality and was only
included in the survey questionnaire in order to add some humour, that
could have possibly persuaded a larger number of individuals to
respond to the survey.
Because the values of the Pearson’s correlation coefficients in the
correlation table or matrix are within significant ranges, achieving
values of 0.488 for correlation of ability and introversion and 0.349
for correlation of age with ability to learn computing, therefore, it
can be said that the hypothesis under investigation is an acceptable
hypothesis. A Pearson’s correlation coefficient of 0.488 indicates a
moderately strong correlation between introversion and the ability to
learn computing. Other personality traits such as sensing, thinking and
judging do indicate a weak correlation with Person’s coefficient values
of 0.149 and 0.034.
Unfortunately, the number of individuals who had responded to the
survey questionnaire was so small that the statistical analysis of the
results cannot be considered to be highly conclusive. Despite the best
possible attempts to request participation from individuals, only
twenty two individuals could be persuaded to respond and out of all
those who responded, two had submitted incomplete questionnaires. A
total of about 50,000 individuals were informed about the survey
through email. Hence, it can only be concluded that surveys of this
type are best conducted by personal contact on groups such as students
when they are applying for college or university entrance, or on
existing students in computing courses at selected times. It is not
possible to simplify the questionnaire any further, because the Jung’s
personality test is the least lengthy of all of the most widely used
personality tests such as the MBTI and 16 factors test. Longitudinal
studies may be carried out with students who can be persuaded to submit
themselves for personality testing at the time of their entrance to a
college, school or university. Alternatively, it is quite possible that
if the survey were to be required for a research project being
conducted by a reputable institution, that a greater number of
responses may be obtained, resulting in a far more reliable study.
The survey form can be made more attractive with check boxes and radio
buttons or drop down menus that do not require individuals to copy and
paste responses. However, this will require coming to an arrangement
with a web survey company that will charge to provide data for viewing.
The survey form that was developed was the best possible using free
features. Using radio - buttons, drop down menus or check boxes will
have meant that the data will not have been available for analysis.
However, it has to be wondered if the email approach to asking for
individual opinion will ever get a reasonable response.
The regression analysis and the analysis of variance that have been
presented are not required or necessary to test the hypothesis under
investigation (Momberg, 2004, Section 4.3). However, results have been
presented for the sake of completeness and ease in case of any other
questions or investigations being required. The regression analysis
attempts to develop a model for linear relationships between ability
and personality trait variables. The results of this analysis indicate
that a model with introversion as a variable may be possible. Analysis
of variance testing or ANOVA between ability and personality trait
variables indicates a level of correlation between variables that has
already been indicated by the correlation analysis using Pearson’s
coefficients (Brightman, 1994, Pp. 466 – 734). .
The next section discusses the findings that have come to light as a
result of the statistical analysis and the literature review, which are
the quantitative and the qualitative research methodologies that have
been employed in this investigation.
6. Discussion and Findings
The literature review that was presented for this study indicates that
there is a correlation between personality and ability to learn
computing. Although, the literature review, which is the qualitative
study that was carried out, indicates a stronger correlation between
cognitive ability and ability to learn computing, cognitive ability,
which is a measure of the efficiency of the brain processes in an
individual, also has an influence on personality. Hence, the literature
review also indicates that there is a correlation between personality
and ability to learn computing. The study by Chandler indicates that
most of the postgraduate students of computing in the United Kingdom
were of either the SIFJ or the SITJ type personality (Chandler, 2003,
Pp. 5). Entry into a postgraduate course in computing at a United
Kingdom university is a definite indication of a demonstrated ability
in learning computing. Thus it may be concluded that, generally
speaking, SIFJ or the sensing, introverted, feeling, judging
personality as well as the sensing, introverted, thinking and judging
personalities are likely to do very well in computing. Admission to a
postgraduate degree course in computing is not merely an indication of
having the ability of being able to learn computing, but it is more of
an indication of having the potential to excel in learning computing.
The quantitative part of the investigation that was conducted for this
study, or the survey that was conducted, indicates that there is a
significant correlation between introversion and ability to learn
computing. Weak correlations were found between judging, thinking and
sensing personality traits and ability to learn computing. Hence,
according to the quantitative study that was conducted, the personality
that is most likely to do well in computing is the SITJ type
personality. Feeling was, however, found to be weakly and negatively
correlated with ability to learn computing and hence, the quantitative
study does not indicate that the SIFJ type personality is likely to do
well by learning computing. Thus, there is a certain level of agreement
between the qualitative and quantitative investigations that were
carried out for the investigation on personality having a correlation
with ability in learning computing. The quantitative study, however,
indicated that the strongest correlation existed between the
personality trait of introversion and ability to learn computing
because the Pearson’s correlation coefficient for this personality
trait had the strongest correlation with the ability variable that was
being investigated.
In the Chandler study, (Chandler, 2003, Figure 2), the postgraduate
students at Portsmouth were found to be mostly introverted (87 %) and
judging (93 %) as well as thinking (86 %) and sensing (80 %). The
strongest correlation that existed between ability to learn computing
and a personality trait that was to be found in all the undergraduate
as well as postgraduate students of the United Kingdom computing
courses was a judging personality (an average of 82.33 %). This was
followed by 79 % for sensing, 75.67 % for introversion and 62.67 % for
the thinking personality trait.
The personality trait of sensing indicates that individuals acquire
information by sensing the environment rather then by looking for
patterns and connections. A sensing personality is likely to have a
more of an in – depth perception as compared to the surface observation
attitude of the intuitive personality. Other studies that were
presented in the literature review also indicated that surface learners
are not likely to do well in computing as compared to those who have a
more in – depth approach. A judging personality trait is indicative of
a love for an ordered approach to life rather then the spontaneous
approach associated with the perceiving types, who are also the
artistic types. The judging personality trait is, therefore, indicative
of those who are likely to be logical and systematic with a preference
for having a certain order in their mental models as well as life.
Thinking is also associated with logical decision making rather then
people – centred decision making and thus, thinking is indicative of
ability in logic which is a cognitive ability that has been stated to
be correlated with ability in learning computing. Introversion refers
to people who focus internally, as compared to those who focus
externally or on the environment. The marketing and sales types are
those who are more likely to be extroverted, while the computing types
are likely to be more interested in focusing on solving their mental
problems and models. Hence, the quantitative as well as the qualitative
studies have both indicated that there are certain personality types
that are more likely to do well be learning computing as compared to
others.
It can also be concluded that cognitive ability or ability in
mathematics as well as logic does show up in personality. Individuals
who are ordered in their mental approach are likely to be ordered in
all aspects of their life, with extroverted individuals being better
for marketing or sales rather then computing.
Despite the fact that the qualitative and quantitative studies both
present a correlated result, it is felt that the quantitative study or
the survey that was conducted for this investigation is not reliable.
Twenty sets of responses is not a significantly large figure on which
conclusions may be based with a degree of reliability. However, it is a
pity that despite some very honest and concerted attempts to get
individuals to respond, only personal contacts were able to provide
answers. The email campaign to elicit responses from individuals was a
total flop despite the fact that 50,000 individuals were emailed a
request to participate in the survey. Only two responded and both of
the two did not fill in the survey form correctly, without bothering to
take the Jung’s personality test. Hence, not a single email result was
included in the survey study. Perhaps, a more reliable study can be
carried out if individual contacts were to be made with a number of
researchers approaching individuals who they know, so that they can be
persuaded to fill out the long questionnaire for the measurement of
personality and ability in computing. The Jung’s test is the shortest
of the personality tests that are available, with both the MBTI and the
sixteen personality factor model being far longer. Hence, it is
unlikely that the survey questionnaire can be simplified. However,
perhaps personal contact can persuade individuals to take an
investigation more seriously and submit their responses. Most of the
email recipients had a lack of appreciation for the gravity of the
research that was being conducted and the desire to donate their time
was lacking. Perhaps if an official email was sent with the backing of
a reputable academic institution, the response could have been more
favourable.
The next chapter presents some conclusions and recommendations that
have been derived from the investigations that were carried out.
7. Conclusions and Recommendations
Both the quantitative as well as the qualitative investigations that
were carried out for this study indicate that there is indeed a
correlative link between certain personality traits and ability to
learn computing. Surprisingly, the qualitative study was well supported
by the quantitative study and both these studies indicated that the
personality traits of introversion, thinking, judging and sensing are
associated with ability to learn computing. Many studies have already
been carried out by a number of researchers that have attempted to
address the question of personality having a correlation with ability
in computing and most of these studies that have been mentioned in the
literature review support the hypothesis that was examined. Although
the studies that were discussed in the literature review indicated a
stronger correlation between cognitive ability and ability in
computing, cognitive ability does have an impact on personality because
it is related to the brain processes that also have an impact on
personality. Ability in logical reasoning and mathematics as well as a
tendency to get involved deeply into things is usually associated with
the judging and introversion personality traits and this ability is
indicative of success in computing.
Despite the very considerable attempts that were made to elicit
responses from a large number of individuals for the survey, only
twenty good responses were available and these were included in the
statistical analysis. It is felt that longitudinal studies can be
carried out on students who have a preference for studying computing
right from the time that they appear for admissions tests at a
university or college and until they graduate. Not only will such
attempts make it possible to acquire more substantive and reliable
data, but these attempts are more likely to succeed in uncovering even
deeper mysteries such as whether an individual’s personality changes
significantly over time, especially during the four or so years that
they may be studying for an undergraduate degree in computing. Although
personality refers to some very slowly changing traits in an
individual, it is worth investigating how the percentage values of the
different personality trait variables will change as a student
progresses through the program of studies. If a longitudinal study is
carefully conducted by involving consciences faculty members and
councillors, then it is very likely that the impact of any other
external variables that may have an impact on students can also be
noticed.
Studies that attempt to correlate personality attribute with vocational
or study interests are useful because they provide the ability for
individuals to make informed decisions about their future. Although the
interest that an individual may have for a field of study and the
statistical nature of any studies will mean that should there be a
student who shows a keen interest in computing or any other field, with
a certain level of aptitude demonstrated by performance in scholastic
tests, then they should not be excluded from a course of study or a
vocation because of their personality. However, it has to be recognised
that individuals should lead fulfilling lives by making the best of
what the Creator has given them and thus anything that will add to
humans being able to make better and informed decisions is likely to be
useful. Studying at a college or university can be expensive for a
student and their family as well as those who fund educational
institutions. Hence, it is important that individuals should attempt to
do well in courses that are suited to them.
The quantitative study did not indicate that there was a correlation
between the so called geek traits of personality and ability in
computing. Geeks are certainly introverted, but they are not sensing,
judging or thinking type personalities. The image that a geek conjures
in people’s minds is usually that of individual who is so engrossed in
computing that they have become rather detached from reality. Geeks are
also perceived to have a disorderly personality. Success in computing
requires a certain ability to be in tune with reality and the
environment as well as a logical or ordered approach to life.
References / Bibliography
1. Access and Privacy Branch, Government of Alberta. August, 2003.
Conducting Surveys: A Guide to Privacy Protection. Government of
Alberta. Retrieved: March 15, 2006.
2. Adams, D. K., Ammons, R. B., University of Kentucky. College of Arts
and Sciences. Department of, & Psychology 1954, Learning theory,
personality theory, and clinical research. The Kentucky symposium.
(Eleven lectures given under the auspices of the Department of
Psychology in the College of Arts and Sciences of the University of
Kentucky ... 1953) New York & London.
3. Andrews, D., Nonnecke, B., Preece, J. (2003) Electronic survey
methodology: A case study in reaching hard to involve Internet Users.
International Journal of Human-Computer Interaction. 16, 2, 185-210.
Retrieved: March 15, 2006.
4. Barkhuus, Louise. 1999. Allport’s Theory of Traits – A Critical
Review of the Theory and Two Studies. Concordia University. Retrieved:
February 6, 2006.
5. Barrett, J. 2000, Test yourself! : test your aptitude, personality and motivation and plan your career London: Kogan Page.
6. Benjamin, D. P. 1990, Change of representation and inductive bias Boston; London: Kluwer Academic.
7. Biemer, Paul P., Lyberg, Lars E. March, 2003. Introduction to Survey Quality. John Wiley. Retrieved: March 15, 2006.
8. Birnbaum, M. H. 2000, Psychological experiments on the Internet San Diego, Calif.; London: Academic Press.
9. Boden, M. A. & University of Sussex. 1993, Multiple personality
and computational models University of Sussex, School of Cognitive and
Computing Science.
10. Both.net. 2006. The Compleat Geek Test. Both.net. Retrieved: February 7, 2006.
11. Bridgewater, D., University of Newcastle upon Tyne. Dept. of
Education, & University of Newcastle upon Tyne 2000, Some effects
of computer aided learning as revision aids and for improving graphing
skills in secondary school science Newcastle upon Tyne: University of
Newcastle upon Tyne.
12. Brightman, Harvey and Schneider, Howard. 1994. Statistics for Business Problem Solving. South Western Publishing.
13. Brown, J. S. & Goldstein, I. P. 1977, Computers in a learning
society: testimony for the House Science and Technology Subcommittee on
Domestic and International Planning, Analysis, and Cooperation [S.l.:
s.n.].
14. Brown, Nigel and Graff, Martin. 2004. Student performance in
assessments: an exploration of the relevance of personality traits,
identified using meta programmes. University of Glomorgan. Retrieved:
February 6, 2006.
15. Bruer, John T. 2005. The Mind’s Journey from Novice to Expert. James S. McDonnell Foundation. Retrieved: February 6, 2006.
16. Burger, Andreas Johannes. 2003. SOME PSYCHOLOGICAL AND BIOGRAPHICAL
PREDICTORS OF COMPUTER PROFICIENCY: AN ANALYSIS OF THE POTENTIAL OF A
NOVICE TO BECOME A GOOD COMPUTER USER. University of the Free State.
Retrieved: February 6, 2006.
17. Burgess, Thomas F. May, 2001. A General Introduction to the Design
of Questionnaires for Survey Research. University of Leeds. Retrieved:
March 15, 2006.
18. Carter, P. J. 2004, IQ and psychometric tests : assess your personality, aptitude and intelligence London : Kogan Page.
19. Centre for Health Promotion, University of Toronto. 1999.
Conducting Survey Research. University of Toronto. Retrieved: March 15,
2006.
20. Chandler, Jane et al. 2003. EXTROVERT OR INTROVERT? THE REAL
PERSONALITIES OF COMPUTING STUDENTS. Retrieved: February 6, 2006.
21. Chapman, G. M. 1997, Computer based learning in science Ceske Budejovice : JUCB.
22. Collins, K. 1999. Participatory Research: A Primer. Prentice Hall
South Africa (Pty) Ltd, 1999. Retrieved: February 7, 2006.
23. Corman, L, S. 1986. Cognitive Style, Personality Type and Learning
Ability as Factors in Predicting the Success of the Beginning
Programming Student. ACM SIGCSE Bulletin, Vol. 18, No# 4, December
1986, pages 80-89
24. Cunha, A. D. D., Greathead, D., & University of Newcastle upon
Tyne. Computing Science 2004, Code review and personality : is
performance linked to MBTI type? Newcastle upon Tyne : University of
Newcastle upon Tyne, Computing Science.
25. David P. Darcy, Meng (Jessie) Ma. 2005. Exploring Individual
Characteristics and Programming Performance: Implications for
Programmer Selection. 38th Hawaii International Conference on System
Science. Retrieved: February 6, 2006.
26. Deakin University. 2006. Designing and Conducting Surveys. Deakin University. Retrieved: March 15, 2006.
27. Decker, Adrienne. 2003. I want to be a Computer Scientist When I
Grow Up: Evaluating the Skills Necessary for Computer Science. Radford
University. Retrieved: February 6, 2006.
28. Dehnadi, Saeed and Richard Bornat. 2006. The Camel Has Two Humps
(Wotking Title). University of Middlesex. Retrieved: February 6, 2006.
29. Devito da Cunha, Alessandra. 2003. The Myers Briggs Personality
Type as a Predictor of Success in the Code-Review Task. University of
Newcastle. Retrieved: February 6, 2006.
30. Economist.com. 2006. Research Tools – Style Guides. Economist.com. Retrieved: February 6, 2006.
31. Fehriinger, Heather M. 2004. Contributions and Limitations of
Cattell's Sixteen Personality Factor Model. Rochester Institute of
Technology. Retrieved: February 7, 2006.
32. Giannoutsos, E. & University of Durham 2003, Stress, attitudes
and personality in computing in students (1990-1993) Durham: University
of Durham.
33. Heinström, Jannica. 2003. Five personality dimensions and their
influence on information behaviour. Information Research, Vol. 9 No. 1,
October, 2003. Retrieved: February 6, 2006.
34. Hirsh, H. 1990, Incremental version-space merging: a general framework for concept learning Boston, [Mass.]; London: Kluwer.
35. Hoffman, E. & ebrary, I. 2002, Psychological testing at work:
how to use, interpret, and get the most out of the newest tests in
personality, learning style, aptitudes, interests, and more! New York:
McGraw-Hill.
36. J Heinström, Jannica. 2000. The impact of personality and
approaches to learning on information behaviour. Information Research,
Vol. 5 No. 3, April 2000. Retrieved: February 6, 2006.
37. Jarrard. 2006. The Scientist. University of Utah College of Mines and Earth Sciences. Retrieved: February 6, 2006.
38. Jenkins, Andrew. 2001. Companies’ Use of Psychometric Testing and
the Changing Demand for Skills: A Review of the Literature. London
School of Economics and Political Science. Retrieved: February 6, 2006.
39. Jenkins, Tony. 2002. On the Difficulty of Learning to Program. University of Leeds. Retrieved: February 6, 2006.
40. Kaluzniacky, E. & ebrary, I. 2004, Managing psychological
factors in information systems work: an orientation to emotional
intelligence Hershey PA: Information Science.
41. Khorrami-Arani, Olivia. 2001. Researching Computer Self – Efficacy.
International Education Journal Vol 2, No 4, 2001. Educational Research
Conference 2001 Special Issue. Retrieved: February 6, 2006.
42. Kleen, Laura J. 2001. MALICIOUS HACKERS: A FRAMEWORK FOR ANALYSIS
AND CASE STUDY. AIR FORCE INSTITUTE OF TECHNOLOGY, USA. Retrieved:
February 6, 2006.
43. Kuncel, Nathan R. Hezelett, Sarah A and Ones, Deniz S. 2004.
Academic Performance, Career Potential, Creativity, and Job
Performance: Can One Construct Predict Them All? Journal of Personality
and Social Psychology Copyright 2004 by the American Psychological
Association, Inc. 2004, Vol. 86, No. 1, 148–161. Retrieved: February 6,
2006.
44. Laird, P. D. 1988, Learning from good and bad data Boston, Mass.; Lancaster: Kluwer Academic.
45. Learning and Teaching Support Network. Centre for Information and
& Computer, S. Innovations in Teaching and Learning in Information
and Computer Sciences: ITALICS electronic journal [S.l.]: LTSN-ICS.
46. Learning and Teaching Support Network. Centre for Information and,
Computer Sciences. Conference (, University of Ulster. Faculty of
Informatics, Learning and Teaching Support Network. Centre for
Information and, Computer, S., LTSN Centre for Information and Computer
Sciences (Conference: & Aug, E. 2000, 1st annual Conference of the
LTSN Centre for Information and Computer Sciences: "the move from
teaching to learning”: 23-25 Newtown abbey: Faculty of Informatics,
University of Ulster.
47. Linn, M. C. & Hsi, S. 2000, Computers, teachers, peers: science learning partners Mahwah, N.J.: L. Erlbaum Associates.
48. Lyall, R. K. 1995, Designing computer software that takes account
of individual learning styles: submitted for the requirement of a
degree of Master of Science (Science Education), Faculty of Science,
University of Glasgow, Glasgow, October 1995.
49. Lynn, Peter. 2001. A Quality Framework for Longitudinal Studies. University of Essex. Retrieved: March 15, 2006.
50. Mancy, Rebecca and Reid Norman. 2004. Aspects of Cognitive Style
and Programming. University of Glasgow. Retrieved: February 6, 2006.
51. Marshall, C. and Rossman, GB. (1999). Designing qualitative research. Thousand Oaks, California: Sage Publications, 1999
52. Mind Tools. 2006. Project Planning and Management Tools. Mind Tools. Retrieved: March 15, 2006.
53. Molyneux-Hodgson, S., Butterfield, A., & Sutherland, R. 1998,
Mathematical competencies of GNVQ science students: teaching, learning
and computer based activities.
54. Momberg, Christine. 2005. The Relationship between Personality
Traits and Vocational Interests in a South African Context. University
of Pretoria. Retrieved: February 6, 2006.
55. Morris, A. M. & University of Leeds 1984, A study of some
personality factors in relation to computer based learning University
of Leeds.
56. National Centre for Educational Statistics. 2006. Statistical
Standards. National Centre for Educational Statistics. Retrieved: March
15, 2006.
57. National Foundation for Educational Research in England and Wales.
& Publishing Company. Test Division 1973, Catalogue of tests: tests
of aptitude, ability, intelligence, personality, interest, attitudes,
and environment Windsor: The Company.
58. Newman, Renee. M. 1998. THE DYSCALCULIA SYNDROME. RENEE M. NEWMAN. Retrieved: February 6, 2006.
59. Office of Management and Budget. 2006. Proposed Standards and
Guidelines for Statistical Surveys. United States Office of Management
and Budget. Retrieved: March 15, 2006.
60. Office of Naval Research & Northwestern University
Technological Institute Computer and information sciences: 1st
Symposiun on learning adaptation and control in information systems:
Papers.
61. O'Shea, T., Fung, P., & Open University Centre for Information
Technology and Education 1992, Computer science students' perceptions
of learning formal reasoning methods Open University, Centre for
Information Technology and Education.
62. O'Shea, T., Fung, P., & Open University Centre for Information
Technology and Education 1992, Learning to reason formally about
programs : an observational study of computer science students Open
University, Centre for Information Technology and Education.
63. Oster, A. & University of Lincoln 2002, Computers in Israeli
schools: a case study of the effect of their introduction on children's
problem-solving skills and attitudes to learning in science Lincoln:
University of Lincoln.
64. Palisade. 2005. Palisade Software for Risk and Decision Analysis. Palisade. Retrieved: March 16, 2006. From:
65. Palisade. 2005. Risk Analysis Books. Palisade. Retrieved: March 16, 2005.
66. Ray, Charles M et al. 1999. Men’s and Women’s Attitudes toward
Computer Technology: A Comparison. Office Systems Research Journal,
vol. 17, no. 1, Spring 1999. Retrieved: February 6, 2006.
67. Riley Guide.com. 2006. Self-Assessment Resources. Riley Guide.com. Retrieved: February 7, 2007.
68. Roddan, Matt. 2002. The Determinants of Student Failure and
Attrition in First Year Computing Science. University of Glasgow.
Retrieved: February 6, 2006.
69. Salzberg, S. L. 1990, Learning with nested generalized exemplars Boston; London: Kluwer Academic.
70. Shannon, David M et al. 2002. Using Electronic Surveys: Advice from
Survey Professionals. Practical Assessment, Research & Evaluation,
8(1). Retrieved March 14, 2006
71. Shevels, T. 1992, Computer aptitude and personality factors: an empirical study.
72. Similarminds.com. 2006. The Jung Personality Test. Similarminds.com. Retrieved: February 7, 2006.
73. Simon et al. 2006. Predictors of Success in a First Programming
Course. Australian Computer Society. Retrieved: February 6, 2006.
74. Singh, N. 1987, An artificial intelligence approach to test
generation Boston, Mass.; Lancaster: Kluwer Academic Publ., c1987.
75. Smith, Sheila M. 2002. Using the Social Cognitive Model to Explain
Vocational Interest in Information Technology. Information Technology,
Learning, and Performance Journal, Vol. 20, No. 1, Spring 2002.
Retrieved: February 6, 2006.
76. Standler, Ronald B. 1998. Creativity in Science and Engineering. Standler, Ronald B. Retrieved: February 6, 2006.
77. Sterling, Gregory D and Thomas M. Brinthaupt. 2003. Faculty and
Industry Conceptions of Successful Computer Programmers. Journal of
Information Systems Education, Vol. 14(4). Retrieved: February 6,
2006.
78. Sutton, R. S. 1992, Reinforcement learning Boston; London: Kluwer Academic Publishers.
79. Symposium on relationships between learning theory, p. t. & and
clinical research (University of Kentucky : 1954, Learning theory,
personality theory and clinical research : [eleven lectures given under
the auspices of the Department of Psychology in the College of Arts and
Sciences of the University of Kentucky on March 13 and 14, 1953] [New
York] : [Wiley].
80. Tawney, D. A. 1979, Learning through computers: an introduction to
computer assisted learning in engineering, mathematics and the sciences
at tertiary level London [etc.]: Macmillan.
81. Tawney, D. A., Council for Educational Technology for the United
Kingdom, & National Development Program 1976, Simulation and
modelling in science : computer assisted learning London : Council for
Educational Technology.
82. Teitelbaum-Kronish, P. 1984, Relationships of selected cognitive
aptitudes and personality characteristics of the online searcher to the
quality of performance in online bibliographic retrieval.
83. Tomkins, S. S., Messick, S., Educational Testing Service, &
Princeton University 1963, Computer simulation of personality: frontier
of psychological theory new York; London: Wiley.
84. Tou, J. T. & Wilcox, R. H. 1964, Computer and information
sciences: collected papers on learning, adaptation and control in
information systems Washington, D.C.: Spartan Books.
85. Touretzky, D. S. 1991, Connectionist approaches to language learning Boston: Kluwer Academic Publishers.
86. Tukiainen, Markku and Eero Mönkkönen. 2002. Programming Aptitude
Testing as a Prediction of Learning to Program. University of Joensuu,
Finland. Retrieved: February 6, 2006.
87. University of Kent. 2006. Aptitude Tests. University of Kent. Retrieved: February 6, 2006.
88. Walonick, David. S. 2004. Excerpts from Survival Statistics. Stat
Pac, Inc., 8609 Lyndale Ave. S. #209A, Bloomington, MN 55420.
Retrieved: March 15, 2006.
89. White, Garry and Sivitanides Marcos. 2002. An Empirical
Investigation of the Relationship between Success in Mathematics and
Visual Programming Courses. Southwest Texas State University.
Retrieved: February 6, 2006.
90. Wilson, Brenda Cantwell. 2002. A study of factors promoting
Success in Computer Science including gender differences. Murray State
University. Retrieved: February 6, 2006.
91. Wong, W. & Taaffe, G. 1958, Relationships between selected
aptitude and personality tests and lip reading ability Los Angeles:
John Tracy Clinic.
|
