Risk taking behaviours can enable one to be explorative and creative leading to benefits that were previously inaccessible, but they can also be detrimental to the safety, mental and physical health of the individual if the risk is not assessed properly. It is in this light that examining the mechanisms involved in making decisions involving risk should be further researched. It is proposed that these risk taking behaviours are guided by mental processes, specifically intelligence level, that lead to informed and effective risk taking decisions. The current study examines general GPA attainment with the degree of risk taking attitudes and perceptions in 70 undergraduate students at York University. The methodology being a correlation design using a modified version of The DOSPERT scale by Columbia University which measures individual’s willingness or aversion to taking risks and their perceptions of how risky a situation is. In addition a general questionnaire including basic demographic data along with education oriented questions, in order to gather information about academic standing and so forth.
Risk taking can be defined as undertaking a task or action that involves a challenge a person takes in order to obtain some sort of benefit, when there is an element of uncertainty involved in the outcome.(citation) These behaviours can be threatening to the individual, others or aspects of their life such as their job or relationships. In this sense, risk taking behaviours can either be a characteristic expressed by the individual, situational factors, or a unique interaction between both. Previous research indicates that commonly accepted interpretations of risk attitudes are often considered to be a personality trait (Weber, 1998) which is often referred to as the “sensation seeking” trait and is indeed a part of risk taking (Rossi & Cereatti, 1993), but not as big of a role as once perceived. Sensation seeking, according to Zukerman (who has done extensive research on the subject), is specifically defined as “the need for varied, novel, and complex sensations and experiences and the willingness to take physical and social risks for the sake of such experiences” (1979, p. 10).However, this concept fails to generalize to populations of individuals who do not possess this trait, thus not recognizing other mechanisms that lead to engaging in risk taking. More specifically, those who are deemed generally risk aversive according to this model may still engage in behaviour categorized as risky and cannot be explained by this view of a sole characteristic responsible for this behaviour. Individual differences pertaining to the assessment of the situation must play a vital role. Research has supported this notion as individuals have not been consistently risk taking or aversive across various situations in laboratory settings (Schoemaker, 1990).
The concept of risk taking as a component of an individual’s attitude has gone through a dynamic change over the years much like other components of overall attitude or personality traits. These characteristics that compromised a person’s general attitude or personality were always believed to be rigid, stable and strongly correlated with conditioning in early childhood or genetic/biological factors (Eysenck & Eysenck, 1985). Research has indicated low correlations between these fixed behaviours and predictable responses to situations, shedding light on the actual flexibility of these traits depending on the situational determinants (Mishel & Shoda, 1995). In fact many aspects of psychology acknowledge the influence of external influences which has shifted the direction of thinking about and explaining behaviours in theoretical frameworks, including decision making. The earliest research on decision making occurred in the 1950’s and relied heavily on gambles as stimuli creating hypothesis about human judgements and decisions (Smith & Kosslyn, 2007). These theories, which include the expected utility model, prospect theory and more recently the numerous risk return models have reflected the shift. The Expected Utility model, provided the framework for most models of decision making, assuming rational behaviour on the part of the decision maker in assessing the consequences, assigning utilities (values) to the object, action or person then multiplying the utilities by the likelihood of an outcome and then choosing the option with highest utility (Smith & Kosslyn, 2007). In order for this complex processing to occur, to assess values and minimize possible losses, a significant cognitive function must be activated. Evidence supporting the cognitive aspect of this process comes from research using single cell recording techniques with non human primates identifying small clusters of neurons that become activated when assessing gamble type situations involving the prospects of winning juice (Glimcher, 2003). This highlights the role of cognitive processing when analyzing a situation with benefits and losses over quick, impulsive decisions; especially when the outcomes are seen as desirable or beneficial to the individual. The value of the outcome varies across individuals and may account for differences in risk perception. Furthermore, psychological risk return models perceive risk perception as differing between individuals and influenced by the content and context of the situation (Weber, 1998). Therefore, the notion of risk taking strictly being a personality trait is challenged and can be seen as a function of external influences that vary on the individual’s level of processing the information. The risk return model proposes that a breakdown of behaviour is between perceived benefits and perceived risks, with an individual willingness and ability to trade off units that are valuable for units of risk or loss, including analysis of perceived risk involved which leads the person to the expected outcome or a decision. For example, if an individual was deciding to drive over to a friend’s holiday party when there was a forecast of bad weather- one might weigh the benefits of seeing friends and enjoying themselves as more valuable then sitting as home anticipating a possible snow storm. Additionally, they just put winter tires on which tip the scale in favour of making the decision to go to the event because in the unlikely event of the snow storm, they will be prepared. This provided multiple ways in which an individual’s characteristics and/ or situation can affect choices involving risk, however these models constantly overlook cognitive factors and even more specifically characteristics like intelligence. Frederick reflected on the inadequate amount of research arguing that, “Despite the diversity of phenomena related to IQ, few have attempted to understand-or even describe-its influences on judgment and decision making” (2005, p. 25). Unfortunately, overlooking such a distinct characteristic of an individual’s personality fails to observe all the factors involved in decision making.
Previous research on risk taking has focused on individuals already engaging in high risk behaviours such as ecstasy users (see: Dunn et all, 2009) and those populations in jeopardy to engage in risky behaviours such as adolescents (see: Berten & Rossem, 2009). But most do not examine cognitive factors, opting for strictly social ones in order to plan and implement appropriate intervention techniques. With the exception of a study conducted by Jaccard, Dodge and Ramos (2005), which was a longitudinal design that looked at the role of perceived intelligence and follow up pregnancy a year later. Out of the 8,500 teenage girls assessed those who had higher perceived intelligence initially had lower rates of pregnancy compared to girls who had lower perceived intelligence. It has also been shown that there is a moderate correlation between perceived intelligence and actual IQ scores (Bailey & Mettetal, 1977; Gabriel, Critelli, & Ee, 1994). In addition, IQ scores have been found to be negatively related to delinquency, skipping classes, abusing alcohol and drugs (Moffit, Gabrielli, Mednick & Schulsinger, 1981). Therefore decision making could have a cognitive component related to intelligence levels that negate bad decisions. In the domain of risk preferences however, there is no collective agreement on the impact of cognitive ability or intellectual ability and insufficient amount of research on the topic (Donkers, Melenberg and van Soest, 2001). Even though a lack of empirical evidence out there to support this notion, Lubinski and Humphreys (1997) argue that even though general intelligence and other cognitive abilities have been overlooked as factors involved in decision making, they very well may play an important role and should be examined further. However, the little evidence out there to support this concept is fairly thought provoking.
Many researchers have acknowledged the existence of two types of cognitive processes that either allow us to process quickly and automatically or require deeper and slower cognitive processing (as cited in Sloman, 1996; Chaiken and Trope, 1999; Kahneman and Frederick, 2002). These two processes labelled by Stanovich and West (2000) are better known as “system one” and “system two” and require varying degrees of processes and attention. An individual recognizing the person entering the room as their boss is using system one, where as someone asked to divide a large number into decimals requires activation of system two- deeper and more complex forms of processing requiring conscious awareness, motivation and access to previously learned rules (Frederick, 2005). This theory may ascertain why more intelligent people may not use simple heuristics when making decisions, but analyze them further than average choosing the more beneficial or logical one. Frederick (2005) demonstrated this when studying individuals while doing a cognitive reflective task (a three-item math-puzzle test) designed to elicit an incorrect “intuitive” answer (generated by System one) that needs to be overridden by System two intervention exercising deeper processing. These individuals were more likely to use system two on a variety of tasks. This suggests that traditional choice models may turn out to be valid for at least a subset of the general population, those who have a greater ability to use rational/analytic processing in their decisions (Weber & Johnson, 2008) and use system two. In fact, Frederick (2005) did find a moderate correlation between CRT scores and conventional IQ measures. Therefore, intelligence levels do play a role in decision making but not necessarily in risk taking. However, Benjamin and Shapiro (2005) found that students who scored higher on the SAT’s in the math section were more likely to take a risk when making decisions in realistic situations with a higher expected outcome when it involved low potential loss over a lesser, guaranteed outcome. This counters the notion that more intelligent people, take less risks- perhaps individuals feel more confident in their respective domains only. In this example, students who scored high in math knew the implications and probabilities involved in their decisions so therefore might have been willing to take the risk. Research also indicates that there are individual differences concerning the amount of risk taking or risk perceptions in different modalities, such as gambling, finance, and personal decisions (Mac-Crimmon & Wehrung, 1986, 1990). This only furthers the conundrum of the variability in individuals processing during decision making
Previous research has somewhat described a relationship existing between intelligence and decision making, as one needs to analyze the risks and benefits involved. Duckworth and Seligman (2005) investigated relations between IQ and final GPA and found a moderately significant correlation (r =.32), therefore a relationship between GPA scores and decision making behaviours is feasible to explore. Especially, with decisions that do not strictly involve ones that are specific to individual’s domains of knowledge, but ones that cover a wide sphere of issues encountered in everyday life which previous research has failed to explore.
The DOSPERT scale developed by the Center for Decision Sciences at Columbia University is an excellent scale that addresses various aspects of risky decision making that individuals encounter in everyday life like gambling, alcohol use, ethics etc and will be used in the study. This study takes into account previous studies that involve decision making, in specific risk taking attitudes and attempt to explore its connection with GPA attainment. This is important as certain intervention strategies or programs can be diverted to populations that have been previously overlooked. Individuals in university may seem as rational humans capable of protecting themselves from these situations; however it is a time in their lives that expose them to risky situations without guidance from parental influences. It is a time where they can experiment and explore, having severe implications for later on in life under certain circumstances.
This study proposes that individuals with higher GPA, abstain from risky situations, and therefore avoid risk taking behaviours. Because it may affect or interfere with all the hard work they have put into attaining an outstanding academic standing, compared to those individuals who have lower GPA’s.
The 90 participants in this study were randomly selected from York University in Toronto, Canada. The sample included both males and females from various ethnic backgrounds, with ages ranging from 18-28 years old who volunteered their time. It has been proven that risk taking behaviours decrease significantly with age (Jianakoplos & Bernasek, 2006) therefore a limitation has been put on the age criteria. Individuals must be currently enrolled in the university as well.
The Domain-specific risk-taking (DOSPERT) Scale was developed at the Center for Decision Sciences at Columbia University. Elke Weber, Ann-Renee Blais and Nancy Betz, the founding researchers behind the scale, were motivated by the theoretical risk return trade off frame work and a large body of supporting empirical results (Honoch, Johnson & Wilke,2006). This scale allows for the assessment of both general risk attitudes, measured by the level of willingness to engage in a particular activity, and broken down into five domains that address a variety of decisions encountered in everyday life. These five sub categories include, social, recreational, ethical, financial (questions about investment and gambling were both included), and health and safety. The second part of the DOSPERT, addresses respondent’s perceptions about many of the same questions and the score is looked at as an overall perceptual attitude by measuring the level of risk a situation may have according to the individual. Examples of the questions are as follows, “How likely are you to engage in unprotected sex?” (health/safety) or “how likely are you to bet a days income on a sporting event?” (financial). These questions are rated on a 5 point Likert scale (modified recently from 7) in order to increase its reliability.(Visser, Krosnick, & Lavrakas, 2000). The continuum ranging from not at all likely to very likely and not at all risky to extremely risky
The two sections have 30 identical questions and both are rated on the seven point Likert scale. Also being modified in this study is the exclusion of the perceptions of benefits component which does not have much relevance to this study and due to time constraints. Research supports that this scale has been validated, and replicated in a variety of settings and populations. Also the shorter version is more adaptable to different cultures and a wiser proportion of the population (see; https://decisionsciences.columbia.edu/dospert/index.htm). Weber, et al (2002), have found the internal reliability to be in the range of .70 to .84 (mean a = .78) for the risk taking scale and .70 to .81 (mean a = .77) for the risk perception scale from the original 48- item scale. In addition, further empirical support for the DOSPERT Scale’s validity came from a study conducted by Zuniga and Bouzas (2005), where Mexican high school student’s blood alcohol measure was predicted by the scores on the health/safety and recreational risks subscales in the DOSPERT.
In addition to this scale, participants will need to fill out a questionnaire of demographic data such as, age, gender, years enrolled in school and most importantly most recent GPA score.
Individuals at York University will be approached and asked to participate their time in a study. After they have volunteered, they will be given an informed consent and informed of all the implications pertaining to the study and their rights as participants. They will then be given the demographic questionnaire to fill out along with the modified DOSPERT Scale containing 60 questions. They will be told to answer as truthfully as possible as their names will not be kept with their responses, but a coding system will be used to maintain anonymity. Also upon completion they will be debriefed about the study they participated in.
This study is a non experimental correlation design- as the goal is to find the relationship between GPA and risk taking attitudes and perceptions and employing a survey style method. The GPA and the risk taking attitudes and perception scores will be cross examined and will be analyzed using the correlation coefficient also known as Pearsons r. The prediction on the hypothesis remains that the higher the GPA of an individual, the less likely they will engage in risky behaviours, and the more likely they will perceive a situation to be risky, therefore yielding a negative correlation.