Considering the prevalence of addiction issues within the U.S., this study focused on the topic of recovery from drug and alcohol dependence in order to add to current literature. Prevention and recovery services are two of the most common ways of combating the addiction issue, and counselors are at the forefront of both movements. The bottoming out experience (BOE) and the turning point (TTP) are two common lay terms of factors within changing addictive behaviors, yet the connection of these constructs to recovery remains unstudied and unknown. The current study tested a model that levels of the BOE and the TTP are predictive of early recovery (ER). The data from this study was obtained from a national dataset previously collected from 230 grant-funded addiction treatment centers that utilize the Global Assessment of Individual Needs (GAIN) assessment instrument. A review of the literature gaps, coupled with available data, influenced decisions on research design and statistical analysis procedures. As clear definitions of the BOE, TTP, and ER have not been discovered through research, a descriptive, correlational research design was chosen in order to understand not only what constitutes a BOE, TTP, and ER, but also to discover the relationships between the BOE, TTP, and ER in their natural state. The purpose of correlational studies is to investigate the relationship between two or more variables without researcher manipulation and such designs are common in the counseling and counselor education research field (Heppner, et al., 2008). Because correlational research is exploratory in nature, structural equation modeling (SEM) was utilized to understand the components of each construct and was used to test the hypothesis of the relationships iv between the BOE, TTP and ER. Although SEM is a confirmatory technique, it is frequently used in an exploratory manner because it combines elements of confirmatory factor analysis and multiple regressions and allows for various possibilities of the relationships between constructs and variables (Schrieber, et al., 2006). The literature on ER, BOE, and TTP provides enough evidence to test a theoretical model, which is the purpose of SEM. The hypothesized model assessed data at intake for the BOE, TTP and ER. Once constructs were delineated through measurement models/CFA, SEM path analysis was used to understand how the constructs related to one another. The first three hypotheses were rejected in the study, and measurement model modifications were conducted, which yielded good fit indices. Results from Hypothesis One indicated that hypothesized factors did not load on the BOE, and instead, the BOE at the intake level was a measurement of mental health severity. Results from Hypothesis Two indicated that hypothesized factors did not load on TTP; however, TTP did resulting factor structure created through model modification contained factors of awareness, motivation, and support. Results from Hypothesis Three also indicated that hypothesized indicators did not load into ER; however the resulting factor structure contained indicators of abstinence and environmental support. Lastly, Hypothesis Four yielded three resulting models, all of which had good fit indices. Therefore, hypothesis four was accepted. It is noteworthy that direct effects were not all significant, and the p value in all final models was significant. There was not a significant relationship between the BOE and ER at the intake level; however, there was a significant relationship between the BOE and TTP, as well as TTP and ER at the intake level. The direct v effects between the BOE and ER may have had a role in the significant p values, as well as the large sample size. Within the three resulting models, the BOE had significant relationships with TTP, spiritual support, and motivation. Both spiritual support and motivation also had significant relationships with ER. Therefore, the results from the current study support that there are existing relationships between the BOE and TTP; however, the relationship between the BOE and ER at the intake level was not significant.