# Conclusion

From the current study, it appears that personality traits as measured by the Big Five Inventory, and learning styles measured by the Index of Learning Styles are poor predictors of future academic success for students in an introductory computer programming course.

Surprisingly, time management practices and self-assessment do not seem to be good predictors of academic success either. Is it possible that computer science students either “get it” or they don’t without a considerable amount of time spent studying the topic?

There are a number of possible explanations for our results. The small sample size of this study may be a factor, and a more broadly based study with a larger sample size may find significance in the future. This study also operationally defined academic performance as the students’ results on midterm and final exams. In reality, instructional methods in computer science programs almost invariably involve completing large and complex programming assignments and projects that are not trivial in their time requirements. It is possible for some students to do well on an exam without much study or preparation but if they do not have the time management ability and skill to apply themselves to their programming projects, their final marks and GPA will suffer. Furthermore, using the time management survey with students in other programs of study might provide useful insights into whether time management ability among IT students is somehow fundamentally different from the ability of students in other programs, or whether this survey is fundamentally flawed in its attempt to measure this variable. However, there is some indication that the passion that students feel for computers in general—whether that passion is realized in playing games or developing software—may be a predictor of success in computer software courses.

A better class of predictor involves understanding how students solve logical problems. However, not all logical problems seem to be equally useful in this regard.

Many logical problems ask participants to evaluate the best of available options. These “gaming” paradigm questions are a class of disjunctive reasoning problems that require participants to explicitly or implicitly compute the probability of various outcomes. Participants are expected to make the choice the results in the most reward (usually money) with the least amount of risk. Participants are then given some latitude on how much risk they are prepared to accept for an expected amount of reward. The goal of these problems is to determine if the respondents are consistent in their assessment of risk and reward. These problems show little to no relevance to the ability of students to acquire computer programming skills.

The decision tree class of disjunctive reasoning problems is more interesting to analyze. Although only one of the three problems of this class shows statistical relevance, this might be explainable using Toplak and Stanovich’s (2002) concept of “cognitive miserliness” in which thinkers will often take a pass on a complicated problem if such an option is presented to them. Only 5 subjects in the current study did not chose option “c”, “cannot be determined”, for question 2 and only 12 subjects did not chose this option for question 10. Toplak and Stanovich would consider the remaining subjects to be “cognitive misers” who did not want to spend the time to work through this problem and instead took the pass provided. Among those students who were not cognitive misers, all students correctly solved problem 2 and 75% correctly solved problem 10. If option c was not available to students, students would be forced to work through the decision tree of these problems. This in turn might increase the predictive power of these problems.

Rule based problems showed much more promise in their ability to predict academic success. Problem 12 strongly correlated with success yet could easily be solved by simply applying the rules to the scenarios and then rejecting those scenarios which violated any of the supplied rules. As such, the correlation of this problem may be an indicator of the students’ ability to semantically parse and understand the problem rather than the student’s reasonability ability in itself. Similarly, question 4 can be solved using basic algebra provided the student can adequately parse the problem text and understand what he or she is to solve. Problem 6 requires an understanding of logical inference, however many IT and CS students have not been exposed to logical deduction by the time they enter their first semester programming class. This may explain why this problem was a poor predictor of academic success: many students did not adequately understand the rules upon which this problem was based and thus even an adequate parsing of the problem text was not sufficient to allow them to apply the rules to solve the problem as they understood it.

The best predictor of success seems to be those problems which require students to both reason disjunctively and envision the problem in an iterative or sequential way. Students with this ability seem to be able to understand how to apply loops and other control structures within software programming such that each iteration through a loop or each decision through a problem will bring the problem closer to its resolution.

Future work in this area may be best served by creating an inventory of problem-modeling, decision tree, and rule-based problems of various complexities and then testing this inventory against new intake students in programs of computer science and information technology. Rule-based problems must be constructed with the expectation that once students can correctly parse the problem text, they will be competent in applying the rules required to solve the problem.

Evaluation of curricula and teaching methodologies at the secondary educational level are also warranted. Curricula modifications in primary and secondary educational systems to support students in develop these logical and critical thinking skills at an earlier age could better prepare students for a future career in the Information and Communication Technology industry.