Ana Paula Gomes Jelihovschi
Look Before You Leap: the effects of
cognitive impulsiveness and reasoning
process on rational decision making
Rio de Janeiro
2016
Ana Paula Gomes Jelihovschi
Look Before You Leap: the effects of
cognitive impulsiveness and reasoning
process on rational decision making
Thesis presented as a requirement for obtaining the Master’s Degree in Administration.
Fundação Getulio Vargas
Advisor: Alexandre Linhares
Rio de Janeiro
2016
“Nothing interesting happens in the
comfort zone”
Unknown author
Acknowledgments
This is an extensive list. I would like to thank my advisor Prof. Alexandre Linhares
for his trust and for making this thesis experience lighter. I also have to thank Prof.
Ricardo Lopes for his patience and time since the beginning of this project, and Prof.
Leandro Malloy for his support and attention since my Bachelor’s.
I would like to express my gratitude for my EBAPE friends and colleagues. You are
awesome! It was incredible to have the opportunity to meet such nice and competent
people. Thanks Ebapeanos 2015 for the important support during the program. It was
not easy. Louis, Gus and Rodrigo, thank you for your tips and patience. Leyla, you rock!
Your availability, competence and willingness to help people are admirable. Jo and G.O.,
thank you for the hard working and fun moments! Ula and Renats, thanks for always
being there for me.
Celene, Kaillen, and Adriana, thank you very much for being so helpful and warm.
This place would not stand without you. For everyone who works at EBAPE and makes
this environment productive.
This experience would not be the same without my friends. Belbs and Re, thank you
very much for bringing me back to reality and showing me what really matters. Unicamp,
who is always nearby despite the distance. Shuva and Fe, for sharing this moment and
cheering me on. Lila and Binch, for the partnership and comprehension in my deepest
issues during the toughest times.
To my family, my parents Angela and Sergio for the love and support in all my
decisions, and my brother Paulo Henrique. To my uncle Anatole and my cousin Paulo for
the generosity and kindness. To uncle Enio for sharing his academic experience, shedding
light into academia world. To my godmother Gracinha for having been rooting for me
since my childhood.
Last but not least, thank you Felipe for believing in me more than myself.
We did it!
Contents
1 Introduction
3
2 Theoretical Background and Hypotheses
9
2.1
Executive Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2
Models of Impulsivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3
Dual Process Theories and Decision Making . . . . . . . . . . . . . . . . . 14
3 Method
9
19
3.1
Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2
Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3
Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.4
Statistical Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4 Results
4.1
23
Hypotheses Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5 Discussion and Conclusion
25
Appendices
39
.1
Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
.2
Appendix B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
.3
Appendix C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
.4
Appendix D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Abstract
Impulsivity may lead to several unfortunate consequences and maladaptive behaviors for clinical and non-clinical people. Although many studies discuss the
negative impact of it, few of them emphasize the relationship between cognitive impulsiveness and decision making in non-clinical subjects. The aim of this study is to
investigate the effects of cognitive impulsiveness on decision making and explore the
strategies used by participants to solve problems. For this purpose, we apply two
measures of impulsivity: the self-report Barratt Impulsiveness Scale (BIS-11) and
the performance based Cognitive Reflection Test (CRT).This is the first study that
compares self-report impulsiveness based on BIS-11 and performance-based reflectivity measured by CRT. Moreover, due to the fact that we apply the instruments
on pen and paper, it is possible to evaluate participants’ reasoning processes employed to answer CRT questions. These reasoning processes are related to the role
of Executive Functions for decision making and its relationship with impulsiveness.
In practical terms, we observed participants’ strategies by analyzing their calculation expressions and data organization to answer CRT questions in the paper sheet.
The sample consists of 191 non-clinical adults, professionals, and undergraduate
students from the fields of business, management, and accounting. Results show
that cognitive impulsiveness may negatively affect performance. Moreover, there is
no difference in strategies used by impulsive and non-impulsive people during a decision making, and who calculate in the paper sheet perform better. Finally, people
who inhibit their immediate answers also perform better during a decision making.
Keywords: Executive Functions, Reasoning Process, Impulsivity, Dual
Process, Decision Making, BIS-11, CRT.
2
1
Introduction
How to explain why people from well-known universities or from big companies make
mistakes in simple reasoning questions? Did they skip math classes in high school? Although it could be true for some of them, the majority of educated people may make
mistakes on simple reasoning tasks because they misuse their own cognitive resources.
Frederick (2005) developed the Cognitive Reflection Test (CRT), a three-item task
with simple reasoning problems to measure cognitive ability ( and impulsivity) presenting
the following questions:
(1) A bat and a ball cost $1.10 in total. The bat costs $1.00 more than the ball. How
much does the ball cost?
(2) If it takes 5 machines 5 minutes to make 5 widgets, how long would it take 100
machine to make 100 widgets?
(3) In a lake, there is a patch of lily pads. Every day, the patch doubles in size. If it
takes 48 days for the patch to cover the entire lake, how long would it take for the patch
to cover half of the lake?
He originally applied it on a sample of students from well-known universities and, surprisingly, he found that students from Harvard University (Princeton University) scored
only 1.43 (1.63) on average, on a score that ranges from 0 to 3.
What may explain these intriguing results is what researchers call as cognitive Dual
Process. Literature suggests that people have two types of cognitive processes: System
1 and System 2 (Kahneman and Frederick, 2002; Stanovich and West, 2000; Shafir and
LeBoeuf, 2002). While System 1 is related to an impulsive way of thinking, System 2 is
a reflective style of decision making. That is, even though people know how to answer
specific questions and how to make good decisions they may misjudge if they use the
impulsive cognitive system.
On the other hand, studies on Naturalistic Decision Making suggest that experienced
professionals present great performance using their System 1 cognitive style when solving
problems (Klein, 1999). Klein performed a series of studies where he and his research team
followed the routine of firefighters, pilots, nurses, chess masters, and others specialists to
3
investigate how experienced professionals make decisions in real-life settings under fast
changing circumstances. One of the stories they witnessed about a firefighter team was
the following:
It is a simple house fire in a one-story house in a residential neighborhood. The
fire is in the back, in the kitchen area. The lieutenant leads his hose crew into the
building, to the back, to spray water on the fire, but the fire just roars back at them.
“Odd”, he thinks. The water should have more of an impact. They try dousing it
again, and get the same results. They retreat a few steps to regroup. Then, the
lieutenant starts to feel as something is not right. He doesn’t have any clues; he
just doesn’t feel right about being in that house, so he orders his men out of the
building- a perfectly standard building with nothing out of the ordinary. As soon as
his men leave the building, the floor where they had been standing collapses. Had
they still been inside, they would have plunged into the fire below (p.32).
As we can see, the decision maker was simply performing an automatically-provided
action (Linhares, 2007). The lieutenant was intuitive and not impulsive like Frederick’s
students. Intuition is also related to non-deliberative thinking. However, it involves
pattern matching and recognition of familiar and typical cases (Klein, 1999). Thus, experience plays a key role in successful intuitive decision making. Finally, although students
from the best universities of the world may be used to perform reasoning tasks, the CRT
was unknown to them. In that way, the use of System 1 cognitive style led to some
unexpected students low scores.
In the present study, we are interested in exploring the mechanisms of cognitive impulsiveness. That is, although it is well known that people may provide right or wrong
answers due to the use of these different cognitive processes, we are not sure about the
strategies that could distinguish when they are using one system or the other. More
specifically we do not know whether people may give wrong answers and make bad decisions due to their impulsiveness by using more frequently System 1 or because they do
not have the knowledge to give right answers and make better decisions.
In this way we aim to investigate the effect of impulsivity on decision making and
4
explore the strategies people use to solve problems. Although Frederick evaluated the
of participants providing significant contribution to the literature of cognition and decision making, he and other authors using the same instrument (Alos-Ferrer et al., 2016;
Cueva Herrero et al., 2015; Primi et al., 2015; Toplak et al., 2014, 2011) did not investigate
whether participants could present impulsive personality traits and neither explained the
process of reasoning they used to answer CRT questions. This study intends to fill these
gaps.
To achieve our goal, we selected a sample of non-clinical adults to whom we applied
a survey with three questionnaires. We used the self-report Barratt Impulsiveness Scale
(BIS-11) to evaluate impulsive traits of personality, the CRT to evaluate for decision
making, and sociodemographic questions (Appendix A). Given that we applied the survey on pen and paper, we could evaluate the participants’ strategies by analyzing their
calculation expressions and data organization to answer CRT questions.
The Barratt’s impulsivity model is one of the most widely applied and recognized in
the literature (Stanford et al., 2009). According to this model, impulsiveness trait of personality is composed of three subtypes: non-planning impulsiveness (orientation toward
present and cognitive complexity), motor impulsiveness (act in the spur of the moment),
attentional impulsiveness (lack of attention and concentration) (Patton et al., 1995). The
Cognitive Reflection Test, as mentioned before, is a performance-based three-item task
which measures through simple reasoning questions (Frederick, 2005). The strategies developed to answer CRT questions are strictly related to the Executive Functions which are
the mental processes required when we need to concentrate and pay attention to achieve
a goal. It allows us to mentally play with ideas, consider responses rather than being impulsive, resist temptations, solve problems, and be creative when meeting unanticipated
challenges (Diamond, 2013). Executive Functions act as a manager of our cognitive resources such as planning, decision making, and flexibility which are used to accomplish
an objective.
Regarding the psychometric measures of impulsivity, to our knowledge, there is no
research assessing the relationship between impulsivity and decision making related to
5
logical and abstract reasoning, based on self-report and behaviors measures such as reasoning process and performance. Actually, most studies using both types of measurements
are interested in investigating the relationship between impulsivity and mental disorders
such as substance abuse (McGue et al., 2001; Tarter et al., 1999; Petry and Casarella,
1999; Dougherty et al., 2009) or obesity (Fields et al., 2013). Studies that also focus on
impulsivity in healthy adults (Reynolds et al., 2006) are interested in investigating the
reliability of impulsivity measures in wide dimensions related to personality traits and
behaviors. In this light, there is no study using both CRT and BIS-11 as measures of
impulsivity.
Considering the assessment of the impulsivity’s subtypes following Barratt’s model, the
present study focuses on the non-planning impulsiveness. From a neurobiology perspective, this subtype is analogous to a “cognitive impulsiveness”, which in turn is associated
with decision making (Bechara et al., 2000). According to Bechara’s model, cognitive impulsiveness is related to an inability to delay gratification which is line with the orientation
toward present characteristic of the non-planning impulsiveness subtype.
This study could provide significant contribution for the development of a tool to
measure non-planning impulsiveness. This is an important subtype of impulsivity which
is hard to evaluate, at least using self-reported measures (Barratt, 1993). Moreover, it
is not proportional to the variety and availability of tools to measure the non-planning
impulsiveness comparing to the others subtypes. That is, even though there are several
tests to measure motor impulsiveness, for instance, there are few tools to measure nonplanning impulsiveness (Malloy-Diniz et al., 2010).
This work could also be valuable for tools that measure Executive Functions. The
advantage of using pen and paper to collect the data is that it allows us to investigate
how participants use their Executive Functions during the reasoning process. We observed CRT answers that do not present any externalization or calculation expression
or reasoning (“No expression”), answers that show some data organization but with no
persistence to the development of calculation (“Organization”) and answers with high manipulation of data, demonstrating a rationale with some structured sequences of reasoning
6
(“Calculation”)1 as shown in Figure 1.
Figure 1: Illustration of Reasoning Strategies
Note: A- Organization: some organization but with no persistence for calculation; B- Calculation: high
manipulation of data, demonstrating a rationale with some structured sequences of reasoning; C- No
expression: no externalization or expression of calculation or reasoning.
Something curious in the CRT answers is that some people answered the question,
erased and changed the answer, showing calculation or not (Figure 2). It seems that
participants, at first, answered impulsively, but then, reconsidered their answers, and
changed their mind. We called this variable as “Erasure”.
Figure 2: Erasure Variable
Note: Erasure: it seems that participants, at first, answered impulsively, reconsidered their answers,
and changed their mind.
1
Due to the fact that our sample is comprised by Brazilians we translated CRT to Portuguese language,
and the CRT first question was adapted to the local cultural (baseball is unknown for the majority of
Brazilian population). Thus, we replaced bat and ball by candy (bala) and bubble gum (chiclete).
7
These data may be useful to evaluate Executive Functions considering some benefits
of the data collection. Such data emerged from an individual and singular procedure
with no explicit instructions about it, as usually occurs with neuropsychological tasks
(Krikorian et al., 1994; Heaton et al., 1993; Bechara et al., 1994). Thus, it presents more
spontaneous and ecological data. More specifically, it may be useful for the evaluation
of fluency, decision making and planning. Fluency is the ability to emit verbal and nonverbal behaviors, in sequence, following implicit and explicit pre-settled rules. Decision
making is a process that involves the choice of one among several alternatives in situations
where there is some level of uncertainty. Planning consists of the ability to create the
best way to achieve a defined goal, regarding the rank of steps and the necessary tools to
accomplish it (Malloy-Diniz et al., 2014). Planning is a central cognitive component of
many problem-solving activities (Krikorian et al., 1994).
The main results of the present study suggest that people with higher levels of cognitive impulsiveness have lower performance on decision making tasks and that the use of
different strategies plays an important role in obtaining better outcomes during decision
making. These findings are highly relevant for the literature of administration, psychology/neuropsychology, and cognitive sciences. Based on the results, loss-limiting strategies
may be developed, especially for people who present high levels of impulsive personality
traits. New tools for cognitive impulsiveness investigation may be created and applied
in more cautious interventions for personal and clinical treatments. Finally, for scholars
investigating cognition and intelligence in humans, data about reasoning process is scarce
(Hofstadter, 2008). Thus, the present study could provide valuable contribution and a
better understanding about this issue.
In conclusion, some critiques regarding studies on judgment and decision making claim
that they focus on performance leaving aside important dimensions (Bonner, 2008). This
study intends to fill this gap presenting important data about intrinsic characteristic,
performance, and reasoning process giving significant contribution to the literature of
judgment and decision making.
This study is divided in five sections. Beyond this brief introduction, the theoretical
8
background about executive functions, models of impulsivity, and dual process compose
the second section. Then, the third section discusses the method employed, and after
that, the results are presented in the forth part. Finally, the fifth offers a discussion and
conclusion.
2
Theoretical Background and Hypotheses
2.1
Executive Functions
Executive Functions are the superior order mental processes required to pay attention
and employ all the necessary cognitive operations possible to achieve a goal. The term
“Executive Functions” was initially founded by Lezak (1982) but the pioneer model was
proposed by Luria (1966), after his investigations on patients with local brain lesions.
According to him, mental processes are composed by three basic units for any mental
activity: 1) attention unit; 2) unit of coding and process (process and storage information
from the environment); 3) unit of programming, control and verification of mental activities. Each unit is localized in a different area of the brain: the first unit in the brainstem
and occipital lobe. The second unit in the temporal and parietal lobes. The third unit in
the frontal lobe, particularly on the prefrontal cortex.
On principle, Executive Functions were limited for goal setting, action initiation, inhibition, planning, flexibility, and monitoring (Santos et al., 2016). However, recent studies
have incorporated other functions related to these mental processes such as the mental
theory, social cognition, strategic process of episodic memory, insight, and metacognition
(Santos et al., 2015; Godefroy et al., 2010). Bechara’s model is an important theory
construct for the Executive Functions related to decision making since it investigates
the personality and emotional problems resulting from prefrontal cortex lesions, as the
Phineas Gage classic case (Bechara et al., 2000). This model is based on the hot components of Executive Functions, which differs from the model proposed in the present study,
since it is built based on the cold components of Executive Functions.
The existence of hot and cold Executive Functions emerged from the correlations
9
among the cerebral circuits (Malloy-Diniz et al., 2014). Hot Executive Functions are
strictly related to the motivational and emotional process such as decision making, social
cognition and mental theory (Zelazo and Müller, 2002). They are linked with affective
aspects and involve the orbitofrontal prefrontal cortex. On the other hand, cold Executive
Functions are linked with categorization, verbal fluency, logical and abstract reasoning
functions which are predominantly related to cognitive processes. Activities which cold
Executive Functions has a role imply the area of dorsolateral prefrontal cortex.
Recently, Diamond (2013) mentioned that there is a general agreement about the
three cores of Executive Functions. The first core is the inhibition which includes the
inhibitory control, self-control, and interference control such as selective attention and
cognitive inhibition. The second core is the working memory which allows us to hold
information in mind and mentally work with it. Finally, the last core is the cognitive
flexibility which is our ability to be able to change perspective spatially or interpersonal.
“It involves being flexible enough to adjust to changed demands or priority, to admit you
are wrong, and to take advantage of sudden unexpected opportunities” (Diamond, 2013,
p.14).
For the present work, it worth to delve more deeply on the concepts and relationship
between the inhibitory control and working memory to have it clearer. Without inhibitory
control we would be at the mercy of impulses (Diamond, 2013). It allows us to change and
choose how we react and behave rather than being unthinking people. Moreover, it has
a role on controlling attention, cognition, behavior, and emotions. From the perspective
of cognitive inhibition, the inhibitory control is responsible to suppress immediate mental
representations, involved with resisting unwanted and extraneous thoughts. Working
memory, as mentioned before, is the brain system that provides temporary storage and
manipulation of information that is necessary for complex cognitive tasks such as language,
comprehension, learning, and reasoning (Baddeley and Hitch, 1974). It presents three
subcomponents: the central executive which is an attentional–controlling system, the
phonological loop responsible for stores and rehearses speech-based information, and the
visual-spacial system that manipulates visual images. More recently, Baddeley (2000)
10
added the episodic buffer which is a limited-capacity temporary storage system that has
the ability to integrate information from several sources.
Generally, both systems, inhibitory control and working memory, support one another
(Diamond, 2013). We must hold our goals in mind to know what is relevant or appropriate
and what to inhibit, which is the case of working memory supporting inhibitory control.
From the opposite side, inhibitory control supports working memory by helping us to
stay focused exclusively in one thing to relate multiple ideas, and resisting repeating old
thought patterns (Diamond, 2013).
To conclude, both systems are crucial for the process of reasoning to correctly answer
CRT questions. While the working memory is necessary to manipulate and store the information given about the questions for a precise reasoning, the inhibitory control support
it activity ensuring that external and internal disturbing stimuli will not interfere on this
process.
2.2
Models of Impulsivity
In the late sixties and early seventies the psychologist Walter Mischel performed the
classic study of the “Marshmallow Test” (Mischel and Ebbesen, 1970; Mischel et al.,
1989), evaluating whether 4-year-old children would choose one marshmallow immediately
or whether they could wait for two marshmallows twenty minutes later. In addition to
yield a very cute video on Youtube 2 , his study provided interesting findings. His aim was
to investigate the underlie processes of “willpower” and self-control to see whether these
children could delay gratification. However, the most interesting point about this study
was its follow up studies showing that preschool delay ability predicts outcomes in the
adulthood related to higher educational achievement, higher sense of self-worth, better
ability to cope with stress (Ayduk et al., 2000; Moffitt et al., 2011) and better body mass
index (Schlam et al., 2013) showing higher abilities to manage self-control.
In this way, having a good self-control is a great advantage in life. According to Stanford et al. (2009), the question about whether a person could modulate his/her cognition
2
https://www.youtube.com/watch?v=QX_oy9614HQ
11
and behavior to fit the demands of a given environment is crucial in almost any situation.
Thus, the role of impulsivity among healthy populations has great attention of scientists
interested in investigating their characteristics and consequences on several contexts such
as employment behaviors (Everton et al., 2005), health behavior (De Wit et al., 2007)
and educational performance (Diamantopoulou et al., 2007). Impulsive behaviors may
lead to unfortunate consequences such as substance abuse (Swann et al., 2004) and eating
disorders (Dawe and Loxton, 2004). Moreover, it has an important role in psychiatric
diagnostics (Moeller et al., 2001). Actually, impulsivity may be characterized as an executive inhibitory dyscontrol which means that it is a maladaptive behavior that presents
bad results in some Executive Functions activities (Enticott et al., 2006). Thus, impulsivity is an important concern for research and that is the reason for the several models
about it.
Despite the usual descriptions about the bad consequences of impulsivity, Dickman
(1990) suggests that we have a functional and a dysfunctional impulsivity. The functional
impulsivity may be the tendency to act with little forethought in an optimal situation.
It may be very useful in sports, and for experienced people in tough circumstances as
discussed in the introduction about Klein’ studies. The dysfunctional impulsivity, on the
other hand, is the tendency to act with less forethought than most of people, leading a
person to difficult situations.
Impulsivity may also be a range of behaviors regarding preference for immediate reward
(temporal-impulsivity), the tendency to make premature decisions (reflection impulsivity),
and responses (motor-impulsivity) (Caswell et al., 2015). Moeller et al. (2001) suggest
that impulsivity is a quick unplanned behavior that leads to thoughtless actions. Two
models of impulsivity are suggested by Swann et al. (2002). The first one is the rewarddiscounting model where impulsivity may be the inability to wait for a larger reward in
a long term. The second model suggests that impulsivity is a response without adequate
assessment of the context. For Whiteside and Lynam (2001) impulsivity is a multi-faceted
construct which is composed by urgency, (lack of) premeditation, (lack of) perseverance,
and sensation seeking.
12
Finally, the base for our study is the model of Patton et al. (1995). They suggest
that impulsivity has three subcomponents: motor impulsivity, attentional impulsivity
and impulsivity for not planning. Here, the motor and attentional subcomponents are
related to the lack of ability in the inhibitory control.
This model was initially developed to separate impulsiveness from anxiety and include
information from different perspectives such as psychological model, social model, medical
model and behavioral model (Barratt, 1993). That is, it integrates cognitive, behavioral,
biological and environmental data. Some studies (Barratt, 1985; Patton et al., 1995;
Barratt, 1993) raised an inquisitive discussion about the cognitive aspect of impulsivity.
Cognition is a process that underlie impulsiveness as a whole (Patton et al., 1995). This
assumption is also raised in a systematic review about the psychometric properties of
the scale (Vasconcelos et al., 2012). The same systematic review detected that the nonplanning impulsiveness tends to be identified in most of the studies while the instability
of attentional items on different studies may be related to the hypothesis about the role
of cognition on impulsivity. Anyway, a fast cognition and the relationship between lack
of planning ahead and characteristics of high-impulsive subjects was found by the scales’s
author (Barratt, 1993).
Kagan (1966) was also interested in investigating cognitive process, such as impulsivity
and which he called “reflection-impulsivity”. The author explored the difference on the
problem solving quality among children with similar level of intelligence by observing
their performance and predisposition to be impulsive or reflective on cognitive tasks.
According to him, people who take more time to evaluate their cognitive process are
called “reflective” while people who give the first answers and ideas that emerge on their
minds are called “impulsive”. Previously, this author created the Matching Familiar
Figure (MFF) which is a task to measure reflection-impulsivity in children (Kagan et al.,
1964). Although we have similar interests, our sample and methods are very different.
We are bringing a divergent perspective about reflectivity and impulsivity, and besides
the literature contributions, we are also considering the practical implications on the daily
lives of healthy adults.
13
Finally, the interest in investigating this paradigm of and impulsivity has been studied
by several authors who are absorbed in evaluating the negative impact for clinical population to not reflect (Djamshidian et al., 2013; Solowij et al., 2012; Townshend et al., 2014).
Furthermore, in several fields of knowledge such as Cognition, Social Psychology, Economics, and Management, researchers are also concerned with discovering how cognitive
processes may influence reasoning (Osman, 2004), learning (Sun et al., 2005) and decision
making (Kahneman and Frederick, 2002) in non-clinical populations. In the present study
we are interested in investigating the relationship between reasoning process and higher
cognitive process such as decision making.
2.3
Dual Process Theories and Decision Making
Dual process theories are interested in understanding the mechanisms that underlie the
process of thinking. The first studies about it started in the 1970s and 1980s, concerned
in investigating performances on reasoning tasks (Wason and Evans, 1975; Evans et al.,
1983). These reasoning processes have been characterized and named in several ways. In
the literature of rationality and decision making, for instance, researchers suggest that we
have two different ways of thinking which they call System 1 and System 2 (Sloman, 1996;
Stanovich and West, 2000; Shafir and LeBoeuf, 2002; Kahneman and Frederick, 2002;
Strack and Deutsch, 2004)3 . System 1 is characterized as automatic, rapid and requires
low cognitive effort. For this reason, some authors call System 1 as the impulsive (Strack
and Deutsch, 2004), automatic (Schneider and Shiffrin, 1977) and intuitive (Hammond,
1996) system. On the other hand, System 2 refers to the analytic and reflective thinking,
which requires higher cognitive effort and time. When System 2 is activated, people are
usually more concentrated and using their abstract reasoning. This system is also called
systematic (Chaiken, 1980; Chen and Chaiken, 1999) controlled (Schneider and Shiffrin,
1977) and rule based (Sloman, 1996; Smith and DeCoster, 2000).
Modern theory about dual process raised a question about the reasoning processes
nomenclatures. Although the names System 1 and System 2 are very common, we agree
3
These theorists suggest that the usual common name “System 1” and “System 2” (Kahneman and
Frederick, 2002; Stanovich, 1999) is more neutral than others forms.
14
with the suggestion that this is not the more appropriate way to refer to these processes
(Evans and Stanovich, 2013). The authors claim that the “Dual System” term is ambiguous because it may also refers to a two minds hypothesis. Thus, we will call these
processes as Type 1 and Type 2 as proposed by the authors which may be more strictly
related to and impulsivity ways of thinking as suggested by Frederick (2005) for the CRT
measures.
From a neuroscience perspective,there are actually three neural systems related to
decision making (Noël et al., 2013). The third neural system suggested by the authors
is responsible for craving sensations and, consequently for addictions, such as gambling
and drug addiction. The first system is the amygdala-striatum dependent system which
mediates automatic and habitual behaviors; the second system is the prefrontal cortex
dependent system related to self-regulation and forecasting the future consequences of a
behavior; and the third system is the insula dependent system which is the responsible
for the reception of interoceptive signals and their translation into felling state presenting
significant influence in decision making and impulse control related to risk, reward and
uncertainty (Noël et al., 2013). The first and second systems are analogous to the dual
process theories of cognition. That is, the first one is fast and automatic while the second
one is slow and deliberative.
As mentioned in the section 2.1, Type 2 way of thinking is strictly related with its
operation. To review, working memory is one of the core Executive Functions (Diamond,
2013), responsible to provide temporary storage and manipulation of information necessary for complex cognitive tasks such as language, comprehension, learning, and reasoning.
It has an attentional limited control system (episodic buffer) which limits its cognitive
capacity (Baddeley, 2000; Cowan, 2012). Thus, while working memory is necessary for
a successful functioning of the Type 2 process during a decision making, it may be true
that it does not play a role whether Type 1 is in charge (Table 1).
In this way, the role of each process during a decision making is very singular. We
usually live our daily lives in a default mode, making decisions when automatic thinking
is appropriate. This is noticeable in situations such as brushing teeth, making the way
15
from home to work, and expressing when something is disgusting. However, when we
face unusual situations, and attention is necessary, the automatic thinking may not be
appropriate and the use of Type 2 process could present better results (Kahneman, 2011).
Considering the relationship between both processes, modern theories about dual process
suggest that Type 1 is responsible for biasing people while Type 2 is responsible for
inhibiting and suppressing the default responses that emerge from Type 1 (Evans, 2003).
The necessary effort to suppress Type 1 is high and requires high inhibitory control
(Pennycook et al., 2015). Thus, Type 1 and Type 2 are usually competing with each
other.
Table 1: Cluster Of Attributes Usually Associated To The Models Of Dual System And
Dual Process Theories Related to Decision Making (adapted from Evans and Stanovich
(2013)).
Type 1 process (impulsive)
Type 2 process (reflective)
Defining features
Does not require working memory
Requires working memory
Autonomous
Cognitive decoupling: mental simulation
Typical correlates
Fast
Slow
High capacity
Capacity limited
Parallel
Serial
Nonconscious
Conscious
Biased response
Normative responses
Contextualized
Abstract
Automatic
Controlled
Associative
Rule based
Experience-based decision making
Consequential decision making
Independent of cognitive ability
Correlated with cognitive ability
The conflict and competition of Type 1 and Type 2 responses is an important topic in
16
the context of judgment and decision making. According to Tversky (1974), both types of
cognitive processes may lead to totally different answers for the same questions whether
people make use of heuristics to make decisions. Regarding the characteristics of each
process, the likelihood of a person to give wrong answers and make bad decisions using
Type 1 may be much higher than using Type 2.
However, recent literature about the dynamic between both cognitive processes on
decision making shows that the effort of Type 2 to avoid cognitive bias may be insufficient
regarding that humans are seen as cognitive misers (Toplak et al., 2014). That is, people
minimize their cognitive effort, having a strong tendency to rely more in an intuitive
process than on a deliberative process (Kahneman, 2011). Although Type 2 processing
enables us to solve novel problems with accuracy, it comes with a cost of cognitive load
and cognitive effort. In this way, we have a tendency to act according to Type 1 process
which has low computational expenses (Toplak et al., 2014). Thus, the stronger reliability
on intuitive process may explain the wrong answers of decision makers on the CRT first
question about the bat and the ball (De Neys et al., 2013).
The propensity of minimizing the cognitive effort has drawn attention of psychologists
who are interested in studying individual differences in the tendency of being cognitive
misers (Stanovich, 1999; Stanovich and West, 2000). They aim to understand how these
differences may affect the propensity of engagement on reasoning and reflection during
rational tasks to lead to correct answers. In this line, this work may be very useful and
may also give great contribution to the psychology literature interested in investigating
individual differences and the propensity to engage in a more effortful reasoning, since we
compare impulsiveness traits of personality with reasoning processes.
Finally, the rationality behind these theories is that impulsivity affects decision making
and that strategies used for it would also play an important role on performance. That is,
impulsiveness, specially non-planning impulsiveness according to Barratt’s model, leads
people to focus on the present reward, not evaluating the consequences of their decisions
in a medium and long term, which leads them to unfavorable solutions to a problem.
Moreover, non-planning impulsiveness may lead to less calculation and manipulation of
17
data. Specifically, people with higher levels of this subtype of impulsivity present planning deficits by not checking whether the impulsive answer could be wrong. Hence the
manipulation of data may be reduced or null which leads to the following hypotheses:
H1 : High levels of non-planning impulsiveness trait negatively affects performance on
CRT.
H2 : High levels of non-planning impulsiveness trait negatively affects the development
of calculation to answer CRT questions.
Regarding the strategies used by participants, for a successful accomplishment of several daily activities, people should clearly identify their final objective, building a plan of
goals and use a hierarchical organization that makes its execution feasible (Malloy-Diniz
et al., 2014). Nevertheless, this subject should perform the planned steps constantly evaluating the success of each one correcting the unsuccessful ones, adopting new strategies if
necessary. Great ability on inhibitory control, working memory and cognitive flexibility
are important to plan, pay attention and persist until the end of a task. Thus, regarding
the reasoning processes in tasks that involve algebra knowledge application, we consider
that calculation represent the closest to the necessary steps for a successful achievement
of a goal using the Executive Functions properly, as stated in the hypothesis below:
H3 : The more participants manipulate data to answer CRT questions, the better will
be their performance on CRT. That is, calculation positively affects CRT performances.
As mentioned before, some participants presented erasures. It seems that some of them
answered the question impulsively but reconsidered the impulsive answer changing it. It
may be true that they initially succumbed for the impulsive answer, but they rethought
about it and recalculated the question, leading us for our fourth hypothesis:
18
H4 : Erasure positively affects CRT performances.
Finally, the present study could contribute to the literatures of decision making, cognitive science and psychology. As Frederick (2005) mentions, decision researchers are more
interested in investigating the average effect of some experimental manipulation leaving
individual differences aside. Our findings could contribute to unfold this important issue.
3
Method
3.1
Participants
The sample was comprised of 191 non-clinical participants which were professionals
(74.8%) and undergraduate students (25.2%) from the fields of business, accounting, and
management. Women composed 44.3% of the sample and participants mean age was 33.9
years (SD=10.23). In total, 191 participants answered the surveys, but 7 participants did
not inform their monthly income, 11 did not answered some CRT question and, 1 of them
left one BIS-11 question unanswered. Thus, due to missing values, the analyses sample
number varies between 183 and 179 in the hypotheses tests.4 Participants were volunteers recruited from a well-known entertainment company in Brasil, from public sector
Administration program from Getulio Vargas Foundation (FGV), and from Accountancy
undergraduate program from State University of Rio de Janeiro (UERJ) which are prestigious universities in Brasil. Thus, we assume that all participants were able to read,
interpret questions and perform the four basic math operations (add, subtract, multiply, and divide). The inclusion criteria were ages between 18 to 61 years old and have
completed higher education or underway. Participants were excluded from the study
otherwise.
4
Appendix B presents the hypotheses tests using the variables with no missing values showing that
results do not change.
19
3.2
Instruments
• Cognitive Reflection Test (CRT) (Frederick, 2005): it is a performance-based threeitem task. Once it is comprised by three reasoning questions which respondents may
answer right or wrongly, the CRT score ranges from 0 (no right answer) to 3 (all
answers right). It measures one type of cognitive ability: the tendency to override
a premature response that is usually incorrect and to engage in reflective reasoning
which usually leads to right answers. The translation was carried out by researchers
themselves. The CRT first question was adapted to the local culture reality since
baseball is an unknown sport to the majority of Brazilians. Thus we replaced “bat”
and “ball” by candy (bala) and bubble gum (chiclete).
• Barratt Impulsiveness Scale (BIS-11, translated version (Malloy-Diniz et al., 2010):
this is a self-report Likert scale from 1 to 4 (1- rarely/never; 4-almost always/always)
constituted by 30 items that evaluate the behavior construct and personality of
impulsivity. This scale measures the three subtypes of impulsivity (non–planning
impulsiveness, attentional impulsiveness, motor impulsiveness) and the total impulsiveness which is the sum of the subtypes. Example of a question measuring
non-planning impulsiveness is “I like to think about complex problems”, while for
attentional impulsiveness it is “I am a steady thinker”, and for motor impulsiveness
“I act on impulse”.
3.3
Procedures
Procedures were very simple. Participants answered the survey on paper in a one session of 30 minutes maximum. As mentioned before, one part of the respondents answered
the survey during an event of a well-known big-sized entertainment company in Brasil,
while the other part, composed by undergraduate students and public tender students,
answered it at their classroom after they return from a break in their classes.
Individually, they answered the CRT, BIS-11 and demographics questions after signing
the consent form. Researchers gave directions to answer the questions by themselves
20
without consulting external sources. When in doubt, participants were told to ask help
of the researchers who remained present in the room during procedures.
Each CRT questionnaire had a blank space for their use in case they need to do some
calculation. However, in order to not influencing participants on their decisions of using
or not such blank spaces, nothing was said about the possibility of making calculation in
those spaces. At the end, participants returned the protocol with their answers and their
consent form. We stored both documents separately, and gave them a code to ensure
anonymity.
3.4
Statistical Analyses
Descriptive analysis was used in order to characterize the sample as depicted in Table 2.
Age, gender, occupation, and income are the demographics characteristics of the sample.
The annual income variable is a categorical variable ranging from up to US$ 7,536.23
(income0) to above US$ 48,985.50 (income5)5 .
Aiming to define which criteria would be used to differentiate the cognitive processes
variables (“No expression”, “Organization”, “Calculation”, and “Erasure”), three researchers analyzed and coded these variables of each questionnaire to reach a consensus
about it. Moreover, BIS-11 is a 30 questions Likert scale that ranges from 1 to 4. Thus,
it is important to emphasize that there are 11 questions aiming to assess non-planning
impulsiveness, 11 questions to evaluate motor impulsiveness, 8 questions to investigate
attentional impulsiveness resulting in a total of 30 questions that evaluate a total impulsiveness measurement in the BIS-11 scale. The minimum score for non-planning and
motor impulsiveness is 11 with a maximum of 44, for attentional impulsiveness is a minimum score of 8 with a maximum of 32, and for total impulsiveness the score is a minimum
of 30 with a maximum of 120. Data were analyzed using Stata version 14.1.
5
Annual income was asked to participants in BRL (the Brazilian currency), which were converted into
U.S. dollars at the average rate for the period of data collection, i.e., USD 1 = BRL 3.45.
21
Table 2: Summary Statistics
Variable
Mean (SD)
Min
Max
N
income0 (up to US$7.536,23)
0.21 (0.41)
0
1
184
income1 (US$7,536.23 to US$13,188.40)
0.07 (0.25)
0
1
184
income2 (US$13,188.00 to US$18,840.57)
0.13 (0.34)
0
1
184
income3 (US$18,840.57 to US$30,144.92)
0.18 (0.39)
0
1
184
income4 (US$30,144.92 to US$48,985.50)
0.21 (0.41)
0
1
184
income5 (above US$48,985.50)
0.2 (0.4)
0
1
184
Gender (1=female)
0.44 (0.5)
0
1
185
Occupation (1=professional)
0.74 (0.43)
0
1
191
Age
33.9 (10.24)
18
61
185
Iat
16.73 (3.33)
10
27
191
Imt
19.05 (4.14)
12
41
191
Inpt
23.38 (4.42)
13
35
191
It
59.16 (9.23)
37
87
191
NExps
2.37 (1.01)
0
3
191
Orgs
0.14 (0.41)
0
2
191
Calcs
0.49 (0.89)
0
3
191
CRTs
1.21 (1.12)
0
3
187
Erass
0.18 (0.45)
0
3
191
Iat = Total Attentional Impulsiveness; Imt = Total Motor Impulsiveness;
Inpt= Total Non-Planning Impulsiveness; IT = Total Impulsiveness;
NExps = Sum of No expression; Org = Sum of Organization;
Calcs = Sum of Calculation; CRTs = Sum of Right Answers on CRT; Erass = Sum of Erasures
Table 3 presents the correlation analysis applying Pearson Correlation Coefficients6 .
The investigated variables are the total of the three subtypes of impulsivity (motor impulsiveness, attentional impulsiveness and non-planning Impulsiveness), the total impulsiveness, the sum of the No expression, Erasure, Organization, and Calculation variables
to answer CRT test, and the sum of right answers in CRT for each respondent.
An Ordinal Least Square (OLS) and an Ordered Logistic Regression (Ologit) were
performed to test the hypotheses. The dependent variables are interval and ordinal. Thus,
6
Considering that the variables in the cross-correlation table are interval and ordinal variables, we
also performed a correlation analysis employing Spearman Correlation as shown in Appendix C.
22
Table 3: Cross-Correlation Table
Variables
1
2
3
4
5
6
1-Iat
1.000
2-Imt
0.337
1.000
3-Inpt
0.308
0.517
1.000
4-It
0.660
0.819
0.822
1.000
5-NExps
-0.044
-0.025
0.068
0.005
1.000
6-Orgs
0.055
0.118
0.008
0.077
-0.487
1.000
7-Calcs
0.025
-0.025
-0.081
-0.041
-0.917
0.098
8-CRTs
0.002
-0.060
-0.214
-0.129
-0.288
0.061
9-Erass
0.067
0.101
0.038
0.088
0.030
-0.082
Iat = Total Attentional Impulsiveness; Imt = Total Motor Impulsiveness;
Inpt= Total Non-Planning Impulsiveness; IT = Total Impulsiveness;
NExps = Sum of No expression; Orgs = Sum of Organization;
Calcs = Sum of Calculation; CRTs = Sum of Right Answers on CRT; Erass
7
8
9
1.000
0.300
0.004
1.000
0.221
1.000
= Sum of Erasures
the results of both methods may be useful to evaluate the findings’ robustness. The OLS
presents a simple result, and its coefficient allows a direct interpretation. On the other
hand, the Ologit is also appropriate for the analyses since our dependent variables may
be ordered from 0 (poor performance), which are people who got all three answers wrong
or people who did not calculate to answer them, to 3 (excellent performance), people who
scored the three CRT questions or people who calculated in the paper sheet to answer
all of them. Finally, both methods admit the control of important variables which could
influence the outcomes of the dependent variables such as income (Dohmen et al., 2010).
Attentional impulsiveness and motor impulsiveness were added to the model as control
variables for the non-planning impulsiveness effects. Moreover, Organization variable is a
control variable for the effect of Calculation 7 .
4
Results
4.1
Hypotheses Tests
Table 4 presents the results. Results of both regression methods were the same regarding p-value. In this way, OLS coefficients are reported because they allow a direct
interpretation. The following results do not present standardized coefficients because
7
Appendix D presents two tables. The first one is the analyzes of Hypotheses Tests with no demographic variables in all models and the second table demonstrate the effects of all control variables in all
models.
23
variables are on the same scale.
Table 4: Hypotheses Tests
Ordinal Least Square (OLS)
Inpt
Imt
Iat
Orgs
Calcs
Erass
N
R
2
Adjusted R
Pseudo R
2
F/χ
2
Ordered Logistic Regression (Ologit)
CRTs
Calcs
CRTs
Calcs
-0.04*
-0.01
-0.10*
-0.03
(0.02)
(0.02)
(0.04)
(0.05)
-0.00
0.02
-0.02
0.07
(0.02)
(0.02)
(0.05)
(0.05)
0.02
0.01
0.06
-0.00
(0.02)
(0.02)
(0.05)
(0.06)
0.16
0.38
(0.15)
(0.33)
0.32***
0.65***
(0.08)
(0.17)
0.43**
0.91**
(0.16)
(0.35)
179
183
0.295
0.107
0.235
0.049
2
9.81
2.12
179
183
0.131
0.069
62.89
22.39
Standard parentheses: * p < 0.05, ** p < 0.01, *** p < 0.001
Demographic variables (income, age, gender, occupation) were added in all models as controls.
Iat and Imt are control variables for the effect of Inpt.
Orgs is a control variable for the effect of Calculation.
Iat = Total Attentional Impulsiveness; Imt = Total Motor Impulsiveness;
Inpt= Total Non-planning impulsiveness; IT = Total Impulsiveness;
NExps = Sum of No expression; Orgs = Sum of Organization;
Calcs = Sum of Calculation; CRTs = Sum of right answers on CRT; Erass = Sum of Erasures
Hypothesis 1 predicted that high levels of non-planning impulsiveness (Inpt) would
affect performance on CRT (CRTs) negatively. After entering the demographic variables
and controlling for the other subtypes of impulsivity, the regressions showed that hypothesis 1 was supported (β = −.04, p < .05). Despite the small coefficient value, this result
suggests that people who are usually present-oriented and do not think carefully may
make worse choices during a decision making process than people with lower levels of
non-planning impulsiveness who are more future-oriented and careful to make decisions.
Hypothesis 2 suggested that higher levels of non-planning impulsiveness would lead
people to less frequently perform calculations to answer CRT questions. Results show
24
that this hypothesis was not supported. People with higher levels of non-planning impulsiveness would not necessarily execute less calculation to answers CRT questions. That is,
there is no difference between people with high levels of cognitive impulsiveness and people
with low levels of cognitive impulsiveness in their strategies to answer CRT questions.
Hypothesis 3 stated that higher levels of manipulation on data, which means a deeper
development of a structured reasoning and calculation to answer CRT questions, would
lead participants to better performance on CRT than those who do not perform calculation. The hypothesis was supported (β = .32, p < .001). It suggests that the development
of a complete reasoning with the achievement of a goal may lead to positive outcomes.
Finally, the last hypothesis suggested that people who presented erasures to answer
CRT questions, that is, people who gave an answer at first but changed their mind presenting another answer, would have better performances on CRT. The hypothesis was
supported (β = .43, p < .01) proposing that those who are able to inhibit and reconsiders
immediate responses may make better decisions.
5
Discussion and Conclusion
The aim of this study is to investigate the effect of impulsiveness on decision mak-
ing and explore the strategies people use to solve problems. For this purpose, we apply
two measures of impulsivity: Barrat Impulsiveness Scale (BIS-11) (Patton et al., 1995)
and Cognitive Reflection Test (CRT) (Frederick, 2005) in a sample of 191 professionals and undergraduate students from the field of business, accounting, and management.
BIS-11 is a self-report scale based on a model where impulsivity is composed by three
subtypes: motor impulsiveness, attentional impulsiveness, and non-planning impulsiveness. The CRT is a performance-based three-item task which aims to measure cognitive
ability. The questionnaires are applied on paper, therefore it is possible to evaluate which
strategies participants used to answer the questions, which we named as “No expression”,
“Erasure”, “Organization”, and “Calculation”. Calculation was considered the strategy
with the highest data manipulation compared to the other strategies, and the closest one
related to the best employment of Executive Functions. Organization was treated as the
25
beginning of a logical reasoning organization of information and data. Erasure was the
act of rethink about an answer, while No expression was the act of presenting an answer
with no externalization of reasoning processes. Table 5 summarizes the evidence collected
in this study from all tested hypotheses.
Table 5: Summary of Results
Hypotheses
H1: High levels of non-planning impulsiveness trait negatively
Supported?
P-value
Coefficient
Yes
.03
-.04
No
.49
-.01
Yes
< .01
.32
Yes
< .01
.43
affects performance on CRT;
H2: High levels of non-planning impulsiveness trait negatively
affects the development of calculation to answer CRT questions;
H3: The more participants manipulate data to answer CRT questions,
the better will be their performance on CRT.
That is, calculation positively affects CRT performances;
H4: Erasure positively affects CRT performances.
Results show some interesting findings. The first hypothesis outcome suggests that
higher levels of non-planning impulsiveness lead to worse performance on CRT. On daily
life it may appear in the form of frustration and stressful situations. Non-planning impulsiveness may lead people to disadvantageous choices since there may not be a logical
reasoning for the choices performed in contexts where it is required. In unusual situations
planning has a key role to satisfactory solutions for a problem (Krikorian et al., 1994).
Thus, whether people do not have much previous experience in a given setting, it takes
them more time and more cognitive effort to develop a hierarchical plan and follow it
successfully. When they act without these concerns, they make mistakes that lead them
to frustration and to incapable thoughts that limit their abilities to reach a positive performance. The problem may be the way they are dealing with the situation and not who
they think they are.
There are common situations in our lives where planning and structured reasoning
are required for successful choices. Some examples are a school exam, when buying a
new cellphone, a computer, an apartment, and a car. In a company context, the lack of
planning and logical reasoning may lead to unfavorable choices when strategic decisions
need to be performed and better strategy design is not implemented due to impulsive
26
decision making.
Another finding is that maybe there is no difference between the strategies used by
people with high levels of non-planning impulsiveness and non-impulsive people to answer
CRT questions. This may be explained due to compensatory behaviors (Anderson and
Bulik, 2004), that is, impulsive people that are aware about their impulsive features may
calculate aiming to mitigate their impulsiveness. When making decisions it may be true
that impulsive people put extra effort on it because they know they are impulsive and
that their impulsiveness could lead to impairments on their decision making processes.
However, regarding that the first hypothesis is supported, this second finding suggests
that some people presented wrong logical sequence of reasoning and gave wrong answers,
which is possible to confirm by checking the questionnaires. Thus, participants with high
and low levels of non-planning impulsiveness might have had presented a logical reasoning
sequence while manipulating data to answer CRT questions, but this logical sequence was
wrong. This finding is interesting considering that one of the sample inclusion criteria
was higher education completed or underway, which implies knowledge of basic math
calculations.
The third hypothesis is supported. Calculation affects performance in CRT positively.
It suggests that when people persist in what they planned their outcomes are better
assuming their plan was effective. This result seems to be intuitive but it is important to
highlight the relevance of an adequate employment of Executive Functions. Calculating
requires a plan for the necessary procedures to manipulate the given information. After
a plan is structured, it is necessary the following of this plan making an evaluation of the
choices, and changing the strategy when necessary for a successful achievement of a goal.
At the same time, people need to keep important information in mind, and manipulate
this information presenting satisfactory use of their working memory. Furthermore, paying
attention to what matters to achieve the goal is also fundamental for successful results. In
this line, the positive relationship between calculation and performance on CRT makes the
importance of Executive Functions improvement more evident for satisfactory solutions
on tasks where logical reasoning is required.
27
The last hypothesis suggests that people who reevaluate their answers may also have
better performances on decision making. Even though some participants presented wrong
answers due to impulsive reasoning, if they changed their mind they could have had
better results. Thus, the inhibitory control may play an important role to achieve the
best results possible on rational decision making. In the present study, participants could
think again about their choice and change their mind with no bad consequences. In
daily lives, sometimes it may not be possible. Therefore, specially in new situations, it is
important to inhibit prompt thoughts evaluating options carefully and, after this, making
the choice. As the follow-up of the Marshmallow Study discovered that children who did
not eat the marshmallow and waited for the other one became successful adults in several
dimensions (Mischel et al., 1989), this study suggests that inhibiting a prompt thought
and careful thinking lead to better outcomes during a decision making. Moreover, this
finding could contribute to the literature on reasoning process conflicts which investigates
the dynamic between Type 1 and Type 2 and the factors that lead to Type 2 engagement
(Pennycook et al., 2015). The action of reevaluating the given answer may represent the
process of Type 2 monitoring the Type 1 output.
The main contributions of this study are related to a methodological advancement
for decision making and impulsiveness literature. Differently from studies on decision
making that usually give emphasis only on people’s performance, we complement such
evidences adding data about intrinsic characteristics, reasoning process and performance.
We also contribute to the literature of impulsivity by presenting evidence using BIS11, based on the second order sub-scale outcomes which are different from the usual
studies that present evidence based only on the total score. Moreover, we implemented
the contribution on impulsiveness literature with non-clinical population using BIS-11.
Finally, to our knowledge, this is the first study that uses BIS-11 and CRT as measures
of cognitive impulsiveness.
Some limitations of this study are that some people may not present calculation or
other reasoning expressions by hand on paper but they may have done it mentally or
using other resources such as the table surface or their own palm, for instance. Another
28
limitation is that the sample is restricted to specific fields of study and professionals with
similar specialities.
Moreover, we cannot observe the relationship between Calculation and Erasure when
participants present both on their CRT answers. That is, we do not know the sequence
performed by participants; if they first answered with no calculation, then rethought,
performed calculation and changed answer (Erasure); or if they first performed some calculation and answered them, rethought, performed calculation again and changed answer
(Erasure) as shown in Figure 3. Finally, the data collection was not performed in a standardized way since one part of the data was collected during an event of a company and
the other part was collected in a classroom.
Figure 3: Erasure and Calculation
Future research could investigate further performance on reasoning tasks related to
neuropsychological tools that evaluate inhibitory control, decision making, attention, and
non-verbal fluency. Furthermore, it could be fruitful to evaluate one’s cognitive effort and
the awareness of self-impulsiveness more deeply. Thus, it would be possible to investigate the issue of compensatory behavior and whether participants are presenting higher
cognitive effort compared with the presence or absence of calculation on their answers.
Regarding sample analyses, the replication of this study with different participants
could be valuable. That is, it would be interesting to investigate the performance on
CRT, cognitive processes and impulsiveness traits with students from different fields and
professionals with others specialities. Another suggestion would be to measure the time
a participant takes to answer each CRT question to investigate the relationship between
time and impulsive decision making, and to collect information about the processes of
reasoning for more detailed analyses. Osman (2013) has a point when she suggests that
29
there are few studies measuring time of responses in tasks that measures dual processes.
Time measurement could contribute for the reliability of the differences on Type 1 and
Type 2 processes of reasoning. Moreover, the assessment of emotions and somatic markers
could bring important insights about the reasoning processes and the investigation of
the cognitive overload effects on CRT performances could also be interesting. Finally,
physiological measures as brain activation using Functional Magnetic Response Image
(FMRI) or electroencephalogram (EEG) could be used to assess the coherence between
neurophysiological activation, behavior, and feelings.
30
References
Alos-Ferrer, C., Garagnani, M., and Hügelschäfer, S. (2016). Cognitive reflection, decision
biases, and response times. Frontiers in Psychology, 7:1402.
Anderson, C. B. and Bulik, C. M. (2004). Gender differences in compensatory behaviors,
weight and shape salience, and drive for thinness. Eating behaviors, 5(1):1–11.
Ayduk, O., Mendoza-Denton, R., Mischel, W., Downey, G., Peake, P. K., and Rodriguez,
M. (2000). Regulating the interpersonal self: strategic self-regulation for coping with
rejection sensitivity. Journal of personality and social psychology, 79(5):776.
Baddeley, A. (2000). The episodic buffer: a new component of working memory? Trends
in cognitive sciences, 4(11):417–423.
Baddeley, A. D. and Hitch, G. (1974). Working memory. Psychology of learning and
motivation, 8:47–89.
Barratt, E. (1993). Impulsivity: integrating cognitive, behavioral, biological, and environmental data. American Psychological Association Washington, DC.
Barratt, E. S. (1985). Impulsiveness subtraits: Arousal and information processing. Motivation, emotion, and personality, 5:137–146.
Bechara, A., Damasio, A. R., Damasio, H., and Anderson, S. W. (1994). Insensitivity to
future consequences following damage to human prefrontal cortex. Cognition, 50(1):7–
15.
Bechara, A., Damasio, H., and Damasio, A. R. (2000). Emotion, decision making and the
orbitofrontal cortex. Cerebral cortex, 10(3):295–307.
Bonner, S. E. (2008). Judgment and decision making in accounting. Prentice Hall.
Caswell, A. J., Bond, R., Duka, T., and Morgan, M. J. (2015). Further evidence of the
heterogeneous nature of impulsivity. Personality and individual differences, 76:68–74.
31
Chaiken, S. (1980). Heuristic versus systematic information processing and the use of
source versus message cues in persuasion. Journal of personality and social psychology,
39(5):752.
Chen, S. and Chaiken, S. (1999). The heuristic-systematic model in its broader context.
Dual-process theories in social psychology, 15:73–96.
Cowan, N. (2012). Working memory capacity. Psychology press.
Cueva Herrero, C., Iturbe-Ormaetxe, I., Mata-Pérez, E., Ponti, G., Sartarelli, M., and
Yu, H. (2015). Cognitive (ir) reflection: New experimental evidence. Technical report,
Instituto Valenciano de Investigaciones Económicas, SA (Ivie).
Dawe, S. and Loxton, N. J. (2004). The role of impulsivity in the development of substance
use and eating disorders. Neuroscience & Biobehavioral Reviews, 28(3):343–351.
De Neys, W., Rossi, S., and Houdé, O. (2013). Bats, balls, and substitution sensitivity:
Cognitive misers are no happy fools. Psychonomic Bulletin & Review, 20(2):269–273.
De Wit, H., Flory, J. D., Acheson, A., McCloskey, M., and Manuck, S. B. (2007). Iq
and nonplanning impulsivity are independently associated with delay discounting in
middle-aged adults. Personality and Individual Differences, 42(1):111–121.
Diamantopoulou, S., Rydell, A.-M., Thorell, L. B., and Bohlin, G. (2007). Impact of executive functioning and symptoms of attention deficit hyperactivity disorder on children’s
peer relations and school performance. Developmental neuropsychology, 32(1):521–542.
Diamond, A. (2013). Executive functions. Annual review of psychology, 64:135.
Dickman, S. J. (1990). Functional and dysfunctional impulsivity: personality and cognitive correlates. Journal of personality and social psychology, 58(1):95.
Djamshidian, A., Sanotsky, Y., Matviyenko, Y., O’Sullivan, S. S., Sharman, S., Selikhova,
M., Fedoryshyn, L., Filts, Y., Bearn, J., Lees, A. J., et al. (2013). Increased reflection
impulsivity in patients with ephedrone-induced parkinsonism. Addiction, 108(4):771–
779.
32
Dohmen, T., Falk, A., Huffman, D., and Sunde, U. (2010). Are risk aversion and impatience related to cognitive ability? The American Economic Review, 100(3):1238–1260.
Dougherty, D. M., Mathias, C. W., Marsh-Richard, D. M., Furr, R. M., Nouvion, S. O.,
and Dawes, M. A. (2009). Distinctions in behavioral impulsivity: Implications for
substance abuse research. Addictive disorders & their treatment, 8(2):61.
Enticott, P. G., Ogloff, J. R., and Bradshaw, J. L. (2006). Associations between laboratory
measures of executive inhibitory control and self-reported impulsivity. Personality and
Individual Differences, 41(2):285–294.
Evans, J. S. B. (2003). In two minds: dual-process accounts of reasoning. Trends in
cognitive sciences, 7(10):454–459.
Evans, J. S. B., Barston, J. L., and Pollard, P. (1983). On the conflict between logic and
belief in syllogistic reasoning. Memory & cognition, 11(3):295–306.
Evans, J. S. B. and Stanovich, K. E. (2013). Dual-process theories of higher cognition
advancing the debate. Perspectives on psychological science, 8(3):223–241.
Everton, W. J., Mastrangelo, P. M., and Jolton, J. A. (2005). Personality correlates of
employees’ personal use of work computers. CyberPsychology & Behavior, 8(2):143–153.
Fields, S., Sabet, M., and Reynolds, B. (2013). Dimensions of impulsive behavior in obese,
overweight, and healthy-weight adolescents. Appetite, 70:60–66.
Frederick, S. (2005). Cognitive reflection and decision making. The Journal of Economic
Perspectives, 19(4):25–42.
Godefroy, O., Azouvi, P., Robert, P., Roussel, M., LeGall, D., and Meulemans, T. (2010).
Dysexecutive syndrome: diagnostic criteria and validation study. Annals of neurology,
68(6):855–864.
Hammond, K. R. (1996). Human judgment and social policy: Irreducible uncertainty,
inevitable error, unavoidable injustice. Oxford University Press.
33
Heaton, R. K. et al. (1993). Wisconsin card sorting test manual. Odessa: Psychological
Assessment Resources.
Hofstadter, D. R. (2008). Fluid concepts and creative analogies: Computer models of the
fundamental mechanisms of thought. Basic books.
Kagan, J. (1966). Reflection-impulsivity: The generality and dynamics of conceptual
tempo. Journal of abnormal psychology, 71(1):17.
Kagan, J., Rosman, B. L., Day, D., Albert, J., and Phillips, W. (1964). Information
processing in the child: Significance of analytic and reflective attitudes. Psychological
Monographs: General and Applied, 78(1):1.
Kahneman, D. (2011). Thinking, fast and slow. Macmillan.
Kahneman, D. and Frederick, S. (2002). Representativeness revisited: Attribute substitution in intuitive judgment. Heuristics and biases: The psychology of intuitive judgment,
49.
Klein, G. (1999). Sources of power: How people make decisions. MIT press.
Krikorian, R., Bartok, J., and Gay, N. (1994). Tower of london procedure: a standard
method and developmental data. Journal of clinical and Experimental Neuropsychology,
16(6):840–850.
Lezak, M. D. (1982). The problem of assessing executive functions. International journal
of Psychology, 17(1-4):281–297.
Linhares, A. (2007). Free will and the power of veto: convergent evidence from decisionmaking. European Cognitive Science Conference, 2007, Delphi, Greece, pages 01–04.
Luria, A. R. (1966). Higher cortical functions in man. New York, Basic books.
Malloy-Diniz, L., Paula, J. d., Loschiavo-Alvares, F., Fuentes, D., and Leite, W. (2010).
Exame das funções executivas. Avaliação Neuropsicológica. Porto Alegre: Artmed,
pages 94–113.
34
Malloy-Diniz, L. F., Paula, J. J. d., Sedó, M., Fuentes, D., Leite, W. B., et al. (2014).
Neuropsicologia das funções executivas e da atenção. Neuropsicologia-Teoria e Prática
[2ed.], pages 115–138.
McGue, M., Iacono, W. G., Legrand, L. N., and Elkins, I. (2001). Origins and consequences of age at first drink. ii. familial risk and heritability. Alcoholism: clinical and
experimental research, 25(8):1166–1173.
Mischel, W. and Ebbesen, E. B. (1970). Attention in delay of gratification. Journal of
Personality and Social Psychology, 16(2):329.
Mischel, W., Shoda, Y., and Rodriguez, M. L. (1989). Delay of gratification in children.
Science, 244(4907):933–938.
Moeller, F. G., Barratt, E. S., Dougherty, D. M., Schmitz, J. M., and Swann, A. C. (2001).
Psychiatric aspects of impulsivity. American journal of psychiatry, 158(11):1783–1793.
Moffitt, T. E., Arseneault, L., Belsky, D., Dickson, N., Hancox, R. J., Harrington, H.,
Houts, R., Poulton, R., Roberts, B. W., Ross, S., et al. (2011). A gradient of childhood
self-control predicts health, wealth, and public safety. Proceedings of the National
Academy of Sciences, 108(7):2693–2698.
Noël, X., Brevers, D., and Bechara, A. (2013). A triadic neurocognitive approach to
addiction for clinical interventions. Frontiers in psychiatry, 4:179.
Osman, M. (2004). An evaluation of dual-process theories of reasoning. Psychonomic
bulletin & review, 11(6):988–1010.
Osman, M. (2013). A case study dual-process theories of higher cognition—commentary
on evans & stanovich (2013). Perspectives on Psychological Science, 8(3):248–252.
Patton, J. H., Stanford, M. S., et al. (1995). Factor structure of the barratt impulsiveness
scale. Journal of clinical psychology, 51(6):768–774.
Pennycook, G., Fugelsang, J. A., and Koehler, D. J. (2015). Everyday consequences of
analytic thinking. Current Directions in Psychological Science, 24(6):425–432.
35
Petry, N. M. and Casarella, T. (1999). Excessive discounting of delayed rewards in substance abusers with gambling problems. Drug and alcohol dependence, 56(1):25–32.
Primi, C., Morsanyi, K., Chiesi, F., Donati, M. A., and Hamilton, J. (2015). The development and testing of a new version of the cognitive reflection test applying item
response theory (irt). Journal of Behavioral Decision Making.
Reynolds, B., Ortengren, A., Richards, J. B., and de Wit, H. (2006). Dimensions of
impulsive behavior: Personality and behavioral measures. Personality and individual
differences, 40(2):305–315.
Santos, F. H., Andrade, V. M., and Bueno, O. F. (2015). Neuropsicologia Hoje. Artmed
Editora.
Schlam, T. R., Wilson, N. L., Shoda, Y., Mischel, W., and Ayduk, O. (2013). Preschoolers’
delay of gratification predicts their body mass 30 years later. The Journal of pediatrics,
162(1):90–93.
Schneider, W. and Shiffrin, R. M. (1977). Controlled and automatic human information
processing: I. detection, search, and attention. Psychological review, 84(1):1.
Shafir, E. and LeBoeuf, R. A. (2002). Rationality. Annual review of psychology, 53(1):491–
517.
Sloman, S. A. (1996). The empirical case for two systems of reasoning. Psychological
bulletin, 119(1):3.
Smith, E. R. and DeCoster, J. (2000). Dual-process models in social and cognitive psychology: Conceptual integration and links to underlying memory systems. Personality
and social psychology review, 4(2):108–131.
Solowij, N., Jones, K. A., Rozman, M. E., Davis, S. M., Ciarrochi, J., Heaven, P. C.,
Pesa, N., Lubman, D. I., and Yücel, M. (2012). Reflection impulsivity in adolescent
cannabis users: a comparison with alcohol-using and non-substance-using adolescents.
Psychopharmacology, 219(2):575–586.
36
Stanford, M. S., Mathias, C. W., Dougherty, D. M., Lake, S. L., Anderson, N. E., and
Patton, J. H. (2009). Fifty years of the barratt impulsiveness scale: An update and
review. Personality and Individual Differences, 47(5):385–395.
Stanovich, K. E. (1999). Who is rational?: Studies of individual differences in reasoning.
Psychology Press.
Stanovich, K. E. and West, R. F. (2000). Advancing the rationality debate. Behavioral
and brain sciences, 23(05):701–717.
Strack, F. and Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and social psychology review, 8(3):220–247.
Sun, R., Slusarz, P., and Terry, C. (2005). The interaction of the explicit and the implicit
in skill learning: a dual-process approach. Psychological review, 112(1):159.
Swann, A. C., Bjork, J. M., Moeller, F. G., and Dougherty, D. M. (2002). Two models of impulsivity: relationship to personality traits and psychopathology. Biological
psychiatry, 51(12):988–994.
Swann, A. C., Dougherty, D. M., Pazzaglia, P. J., Pham, M., and Moeller, F. G. (2004).
Impulsivity: a link between bipolar disorder and substance abuse. Bipolar disorders,
6(3):204–212.
Tarter, R., Vanyukov, M., Giancola, P., Dawes, M., Blackson, T., Mezzich, A., and
Clark, D. B. (1999). Etiology of early age onset substance use disorder: a maturational
perspective. Development and Psychopathology, 11(04):657–683.
Toplak, M. E., West, R. F., and Stanovich, K. E. (2011). The cognitive reflection test
as a predictor of performance on heuristics-and-biases tasks. Memory & Cognition,
39(7):1275–1289.
Toplak, M. E., West, R. F., and Stanovich, K. E. (2014). Assessing miserly information processing: An expansion of the cognitive reflection test. Thinking & Reasoning,
20(2):147–168.
37
Townshend, J. M., Kambouropoulos, N., Griffin, A., Hunt, F. J., and Milani, R. M.
(2014). Binge drinking, reflection impulsivity, and unplanned sexual behavior: impaired decision-making in young social drinkers. Alcoholism: Clinical and Experimental
Research, 38(4):1143–1150.
Tversky, K. D. (1974). Judgment under uncertainty: Heuristics and biases. Science,
185:1124–1131.
Vasconcelos, A. G., Malloy-Diniz, L., and Correa, H. (2012). Systematic review of psychometric proprieties of barratt impulsiveness scale version 11 (bis-11). Clinical Neuropsychiatry: Journal of Treatment Evaluation, 9(2):61–74.
Wason, P. C. and Evans, J. S. B. (1975). Dual processes in reasoning?
Cognition,
3(2):141–154.
Whiteside, S. P. and Lynam, D. R. (2001). The five factor model and impulsivity: Using
a structural model of personality to understand impulsivity. Personality and individual
differences, 30(4):669–689.
Zelazo, P. D. and Müller, U. (2002). Executive function in typical and atypical development.
38
.1
Appendix A
39
Figure 4: Barrat Impulsiveness Scale (BIS-11)
40
Figure 5: Cognitive Reflection Test (CRT)
Cognitive Reflection Test
Instruções: Abaixo estão 3 questões que buscam medir processos de raciocínio. Por
favor, leia com atenção e responda nos respectivos campos.
1. Uma bala e um chiclete custam R$1,10. A bala custa R$1,00 a mais do que
o chiclete. Quanto custa o chiclete?
(informe apenas o valor): R$_______.
2. Se 5 máquinas levam 5 minutos para fabricar 5 componentes, quanto
tempo (em minutos) levará 100 máquinas para produzir 100
componentes?
(informe apenas o tempo): _______ minutos.
3. Em um lago, há um conjunto de vitórias-régias. Todo dia o conjunto
dobra de tamanho. Se leva 48 dias para que o conjunto de vitórias-régias
cubra todo o lago, quantos dias levaria para que o conjunto cobrisse a
metade do lago?
(informe apenas o tempo): _______ dias.
41
Figure 6: Demographic Questions
Questionário Sociodemográfico
1. Sexo:
(__) Feminino
(__) Masculino
2. Qual a sua idade? ______ anos.
3. Estado Civil: ___________________.
4. Quantos filhos você tem? ________ .
5. Você concluiu NÍVEL MÉDIO TÉCNICO?
(__) Não
(__) Sim. Há quantos anos: _________
6. Você concluiu NÍVEL SUPERIOR?
(__) Não
(__) Sim, Há _____anos.
Em qual área? ________________________________________________________
7. Você concluiu PÓS-GRADUAÇÃO?
(__) Não
(__) Sim, Especialização/MBA em _____________________________. Há _____anos.
8. Você RESIDE na Capital ou no Interior do Estado do Rio de Janeiro? (Se em ambos,
indicar somente o principal).
(__) Capital
(__) Interior
9. Você é NATURAL de qual Estado? (onde você nasceu?) UF: _____
10. Ainda com relação a seu local de NASCIMENTO, você nasceu na Capital ou no
Interior do Estado?.
(__) Capital
(__) Interior
11. QUANTO À SUA RENDA. Lembrando que suas respostas são absolutamente
confidenciais, pedimos que marque a faixa de renda bruta mensal em que VOCÊ se
encontra atualmente:
(__) Até R$ 2.000,00.
(__) Acima de R$ 5.000,00 até R$ 8.000,00.
(__) Acima de R$ 2.000,00 até R$ 3.500,00.
(__) Acima de R$ 8.000,00 até R$ 13.000,00.
(__) Acima de R$ 3.500,00 até R$ 5.000,00.
(__) Acima de R$ 13.000,00.
12. QUANTO À RENDA FAMILIAR. Favor marcar a faixa de renda bruta mensal em
que se encontra atualmente O CHEFE DA SUA FAMÍLIA:
(__) Até R$ 2.000,00.
(__) Acima de R$ 5.000,00 até R$ 8.000,00.
(__) Acima de R$ 2.000,00 até R$ 3.500,00.
(__) Acima de R$ 8.000,00 até R$ 13.000,00.
(__) Acima de R$ 3.500,00 até R$ 5.000,00.
(__) Acima de R$ 13.000,00.
42
.2
Appendix B
Table 6: Hypotheses Tests With No Missing Value
Ordinal Least Square (OLS)
Inpt
Imt
Iat
Orgs
Calcs
Erass
Ordered Logistic Regression (Ologit)
CRTs
Calcs
CRT
Calcs
-0.04*
-0.01
-0.10*
-0.02
(0.02)
(0.02)
(0.04)
(0.05)
-0.00
0.01
-0.02
0.06
(0.02)
(0.02)
(0.05)
(0.05)
0.02
0.01
0.06
-0.01
(0.02)
(0.02)
(0.05)
(0.06)
0.16
0.38
(0.15)
(0.33)
0.32***
0.65***
(0.08)
(0.17)
0.43**
0.91**
(0.16)
(0.35)
N
179
179
R2
0.295
0.104
Adjusted R2
0.235
0.045
Pseudo R2
F/χ2
9.81
1.98
179
179
0.131
0.065
62.89
21.08
Standard errors in parentheses
* p<0.05, ** p<0.01, *** p<0.001
Demographic variables (income, age, gender, occupation) were added in all models as controls.
Iat and Imt are control variables for the effect of Inpt.
Orgs is a control variable for the effect of Calculation.
Iat = Total Attentional Impulsiveness; Imt = Total Motor Impulsiveness;
Inpt= Total Non-planning impulsiveness; IT = Total Impulsiveness;
NExps = Sum of No expression; Orgs = Sum of Organization;
Calcs = Sum of Calculation; CRTs = Sum of right answers on CRT; Erass = Sum of Erasures
43
.3
Appendix C
Table 7: Spearman Correlation Table
Variables
1
2
3
4
5
6
7
8
1 Iat
1.000
2 Imt
0.265
1.000
3 Inpt
0.288
0.560
1.000
4 It
0.621
0.777
0.837
1.000
5 Nexps
-0.039
0.004
0.057
0.022
1.000
6 Orgs
0.078
0.077
0.010
0.069
-0.513
1.000
7 Calcs
0.026
-0.020
-0.038
-0.026
-0.922
0.208
1.000
8 CRTs
0.009
-0.027
-0.218
-0.139
-0.267
0.050
0.263
1.000
9 Erass
0.069
0.079
0.013
0.044
0.032
-0.078
-0.015
0.212
Iat = Total Attentional Impulsiveness; Imt = Total Motor Impulsiveness;
Inpt= Total Non-planning impulsiveness; IT = Total Impulsiveness;
NExps = Sum of No expression; Orgs = Sum of Organization;
Calcs = Sum of Calculation; CRTs = Sum of right answers on CRT; Erass = Sum of Erasures
44
9
1.000
.4
Appendix D
Table 8: Hypotheses Tests With No Demographic Variables As Controls
Ordinal Least Sqaure (OLS)
Inpt
Imt
Iat
Orgs
Calcs
Erass
CRTs
Calcs
CRTs
Calcs
-0.06**
-0.02
-0.11**
-0.03
(0.02)
(0.02)
(0.04)
(0.04)
0.01
0.00
0.01
0.02
(0.02)
(0.02)
(0.04)
(0.04)
0.01
0.01
0.03
0.02
(0.02)
(0.02)
(0.04)
(0.05)
0.14
0.31
(0.18)
(0.33)
0.35***
0.62***
(0.07)
(0.15)
0.57***
1.08***
(0.15)
(0.32)
N
187
191
R2
0.184
0.009
Adjusted R2
0.157
-0.006
Pseudo R2
F/χ2
Ordered Logistic Regrssion (Ologit)
9.49
0.61
Standard errors in parentheses
* p<0.05, ** p<0.01, *** p<0.001
45
187
191
0.078
0.001
39.31
0.65
Table 9: Hypotheses Tests Demonstrating the Effects of Control Variables
Inpt
Imt
Iat
Orgs
Calcs
Erass
Gender (1=female)
Age
Occupation (1=professional)
income1
income2
income3
income4
income5
Ordinal Least Square(OLS)
CRTs
Calcs
-0.04*
-0.01
(0.02)
(0.02)
-0.00
0.02
(0.02)
(0.02)
0.02
0.01
(0.02)
(0.02)
0.16
(0.15)
0.32***
(0.08)
0.43**
(0.16)
-0.42*
-0.14
(0.17)
(0.14)
-0.03**
-0.03*
(0.01)
(0.01)
0.61
-0.90**
(0.40)
(0.31)
0.46
0.43
(0.38)
(0.27)
0.42
1.03*
(0.50)
(0.40)
0.40
1.24***
(0.51)
(0.36)
0.34
1.25**
(0.50)
(0.38)
0.57
1.38***
(0.53)
(0.40)
179
183
0.295
0.107
0.235
0.049
Ordered Logistic Regression(Ologit)
CRTs
Calcs
-0.10*
-0.03
(0.04)
(0.05)
-0.02
0.07
(0.05)
(0.05)
0.06
-0.00
(0.05)
(0.06)
0.38
(0.33)
0.65***
(0.17)
0.91**
(0.35)
-0.86**
-0.28
(0.32)
(0.35)
-0.07**
-0.07*
(0.02)
(0.03)
1.22
-3.04***
(0.72)
(0.91)
0.98
0.66
(0.76)
(0.81)
0.81
2.80**
(0.85)
(1.07)
0.73
3.65**
(0.91)
(1.13)
0.74
3.73**
(0.89)
(1.13)
0.92
4.04***
(0.93)
(1.15)
179
183
N
R2
Adjusted R2
Pseudo R2
0.131
9.81
2.12
62.89
F/χ2
Standard errors in parentheses:* p<0.05, ** p<0.01, *** p<0.001
income1 = (US$7,536.23 to US$13,188.40)
income2 = (US$13,188.00 to US$ 18.840.57)
income3 = (US$18,840.57 to US$30,144.92)
income4 = (US$30,144.92 to US$48,985.50)
income5 = (above US$48,985.50)
46
0.069
22.39