Background: Up until now research investigating alcohol and risk-taking has largely overlooked influences from the social settings in which drinking usually occurs. The thesis therefore examines systematically, risk-taking as a determinant and consequence of alcohol consumption, whilst addressing the independent and combined influences of social contexts. Method: Study 1 –Participants completed online surveys measuring trait impulsivity, risk-taking propensity, and alcohol use behaviours. Study 2 – General risk-taking and computer simulated risky driving were measured before and following 0.6g/kg of alcohol or placebo administration in isolation or in natural friendship groups. Study 3 – Risk-taking was assessed in isolation or in natural friendship groups, following 0.8g/kg of alcohol or placebo consumption. Risk-taking behaviour was measured via The Shuffleboard Game, developed to examine physical risk-taking more akin to real world drinking games. Affective state was further measured both before and after beverage consumption. Study 4 –Intoxication levels, experienced alcohol-related consequences, relative injunctive norms, and risky gambling, were measured in real world alcohol and non-alcohol-related environments. Group size data were also collected. Meta-analysis – A systematic search of Web of Science, PsycINFO and PsycARTICLES, revealed 22 (k = 35) alcohol administration studies measuring risky behaviour. Results: Study 1 found both impulsivity and risk-taking predicted 8-11% of variance in hazardous and harmful alcohol use, and dependence symptoms, and 10-14% when combined. Results suggested some overlap between impulsive and risk-taking traits, yet still supported them as distinct constructs. In Study 2, those who were tested in group contexts were riskier on both general and driving-related tasks, than those in isolation. However, no effect of alcohol or interaction of intoxication and group was found on risky behaviour. Conversely, in Study 3, both alcohol and group contexts were found to independently increase risky behaviour on The Shuffleboard Game, although no interaction of beverage and context was revealed. Further, a more positive mood predicted increased risk-taking behaviour. Study 4 revealed no influence of environment (alcohol versus non-alcohol), intoxication levels or injunctive norms on risky gambling, whereas larger group size was associated with riskier lottery choice in non-alcohol-related environments only. Furthermore, injunctive norms predicted experience of risky alcohol consequences, and were riskier in alcohol-related settings. Finally, the meta-analysis found a small, yet significant effect of acute alcohol consumption on risky behaviours, and more specifically on risky driving and gambling. However, alcohol was not found to influence risk-taking on general (non-specific) risk-taking tasks. Overall conclusions: Overall it was found that social contexts consistently increase individual risky behaviour, whereas alcohol effects on risk-taking are contingent on the risk domain measured. The lack of a combined influence of intoxication and groups highlights the importance of targeting social influences and perceived injunctive norms alongside alcohol consumption to reduce risky behaviour in drinking settings. Moreover, the varied effects of alcohol across risk domains outlines important implications for future research assessing risk-taking. Finally, the thesis finds risk-taking to be a predictor of alcohol consumption behaviours therefore, identifying potential risk-factors to address when attempting to reduce problematic alcohol consumption. Original contribution: The experimental research is the first of its kind to experimentally measure both the influence of alcohol and group contexts on individual risk-taking, as opposed to a collective group decision. Further, the thesis offers new insights into the effect of alcohol consumption on risk-taking as findings suggest variations of intoxication influences across risk-domains. Finally, the thesis contributes a newly developed measure of risky behaviour which potentially demonstrates risk-taking more akin to real-world drinking.