EXPERIMENTS ON GOAL-SETTING, SELF-CONTROL AND PERFORMANCE

Abstract

Goal-setting has long been recognised as a powerful tool for enhancing performance (Locke, 1968; Locke & Latham, 1990, 2002). More recently, theoretical economic models have been used to propose that self-set non-binding goals can serve as reference points, leveraging psychological mechanisms such as loss aversion to motivate individuals (Hsiaw, 2013; Koch & Nafziger, 2011; Suvorov & Van de Ven, 2008). However, empirically, the effectiveness of goal-setting interventions can vary significantly based on individual differences and contextual factors (e.g., Clark et al., 2020; Corgnet et al., 2015, 2018; Smithers, 2015). The current study examines the nuanced dynamics of goal-setting through a series of experiments. It contributes to our understanding of how factors such as self-control, loss aversion, gender and goal meaningfulness influence the impact of goal-setting on performance. The aim is to demonstrate both the potential and limitations of goal-setting as a strategy for improving its associated outcomes. This thesis is structured around three self-contained studies, each focusing on different aspects of goal-setting: Chapter 2 examines the interaction between goal-setting, self-control and gender in a lab experiment. Chapter 3 tailors the goal-setting intervention to better resonate with women by connecting goals to charitable donations. Chapter 4 investigates the effects of goal-setting through a field experimentwith university students, examining both the short- and long-term impacts on academic performance. Chapter 2 utilises a real-effort task, whereby the participants are paid based on their effort and are randomly assigned to either a control or goal-setting treatment. The main findings reveal that goal-setting effectively enhances performance for men with low self-control. Moreover, it is found that the interaction between goal-setting and loss aversion has a negative impact: goal-setting is less effective among more loss averse individuals. Chapter 3 extends the investigation by exploring how manipulating the meaningfulness of work moderates goal effectiveness across genders. Motivated by research suggesting that women often derive more meaning from prosocial and communal aspects of work (e.g., Burbano et al., 2023; Konrad et al., 2000), we introduce a charitable donation element to the benefits of work, so that in addition to generating their own income, additional effort generates benefits for others. This approach provides insight into how aligning goals with gender-specific values may enhance the effectiveness of goal-setting. The key findings from this study reveal a complex interaction between gender and goal type. While setting goals alone remains effective for men, a significant gender difference emerges when goals are combined with charitable donations. In this condition, women significantly outperform men. However, among women, the difference between those who set regular goals and those who set goals with donations was not statistically significant. Chapter 4 investigates the effects of goal-setting on academic performance through a field experiment with first-year undergraduate economics students. Unlike previous studies which instructed students to set goals for the final exam (e.g., Clark et al., 2020; van Lent & Souverijn, 2020), this experiment asked students to set goals for a practice exam. This approach allows students to adjust their study strategies by retaking the exam multiple times if their initial goal is not met. The findings reveal that treated students, particularly male students, were approximately five times more likely to attempt the practice exam compared to the control group. On the practice exam, goal-setting students outperformed the control group, with those exhibiting lower self-control benefiting more from the intervention. However, this positive effect did not persist to the final exam, in which goal-setting students unexpectedly underperformed relative to the control group. Interestingly, the interaction between goal-setting and self-control reversed from the practice to the final exam. In summary, this thesis adds to the existing body of literature which demonstrates that self-set non-binding goals can offer a promising and practical solution for improving performance across various domains. At the same time, the current study demonstrates that goal-setting is not always effective and its efficacy is sensitive to numerous factors. As such, this thesis aims to deepen our understanding of how such a simple yet powerful tool can be optimised, considering factors such as self-control, loss aversion and gender.