This study has three primary objectives. First to provide insight into possibilities for moderation/mediation analysis in education research by discussing analytical methods based on the counterfactual and developed in other disciplines. Second is to highlight specific issues related to mediation and the estimation of effects when the outcome is dichotomous. The final objective is to apply counterfactual-based analytical methods to two questions pertaining to student success in community college mathematics, using a dataset spanning twenty years of student transcript information. This study will estimate moderated and mediated effects, use a generalized linear model approach to calculate odds and risk ratios, and incorporate a measure of possible confounding known as the E-value in the investigation of two research questions pertaining to student success in community college mathematics. Prior grades were the most important predictor of success, both in a second sequential mathematics course and for credential completion. While a small association was found between timing of first math class and credential completion, there is no evidence that timing has a causal effect. Policies hoping to enhance student success should focus on effective advising rather than specifying either the timing between sequential math courses or on the timing of the first math class. Researchers using logistic regression should estimate risk ratios instead when an outcome is not rare. Research using regression models should include sensitivity analyses as a standard practice. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com.bibliotheek.ehb.be/en-US/products/dissertations/individuals.shtml.]