TPC Journal V8, Issue 2 - FULL ISSUE

164 The Professional Counselor | Volume 8, Issue 2 goals and achieving them. These circumstances led to the following customized goals: (a) enhancing her time-management skills, (b) engaging in short- and long-term goal setting, (c) exploring potential academic majors, (d) learning more about how to pay for college, and (e) understanding how college education influences one’s future income and lifestyle. Data collection. The CCRSI (Baker & Parikh Foxx, 2012) was distributed electronically via Qualtrics survey software to the participants upon their submission of the informed consent forms. The pre- treatment CCRSI data served as the baseline (Phase A1) for the study. A common self-monitoring schedule was distributed with instructions for each participant to complete the CCRSI four times during the 2 weeks prior to the beginning of the intervention. The intervention (Phase B) lasted 8 weeks for each participant. Participants completed the CCRSI at the end of each weekly session. During the 2-week withdrawal phase (A2) following the last intervention session, participants were again instructed to follow a common self-monitoring schedule for completing the CCRSI four times. The three participants received a dinner, a gift card, and a certificate of completion from the counselor- investigator at the end of the study. Data analysis. Visual and non-parametric analyses were used to assess the outcomes for each experiment, and non-parametric analyses provided information about the effects of the treatments (Hott et al., 2015). Temporal analysis. The time series data were plotted graphically on x (temporal independent variable) and y (dependent variable) axes for each participant and CCRSI factor. Autocorrelation and regression analyses were used to determine the appropriate statistical analysis procedure. Autocorrelation analysis was used to determine whether each observation within each phase and factor of the study was truly independent. Observations that were not correlated to each other could not be predicted (Bloom, Fischer, & Orme, 2006). Regression analysis was used to determine whether significant trends were present for each phase of each CCRSI factor for each participant (alpha <.05). In cases where there was significant trend and autocorrelation, as well as outliers within each phase, the Robust Conservative Dual-Criteria (RCDC; Borckardt, 2008) method was used as the primary statistical analysis tool. RCDC was used to compare differences between phases for each participant as opposed to traditional parametric methods like the student’s t -test and analysis of variance (ANOVA). Intervention effects. Providing effect sizes in addition to visual analyses enhances the credibility, reliability, and defensibility of single-case research findings (Vannest & Ninci, 2015). Vannest and Ninci (2015) reported that there are several strategies available to estimate effect sizes for SCRD studies. In cases where there is a significant trend and autocorrelation, the G-index (Cohen, 1988) is used to estimate effect sizes. The G-index results were determined by using the regression line and the mean or median from the baseline. The effect size was calculated by using the proportion of participants’ scores in the desired zone above the regression line, which was an expected increase in scores from the baseline to treatment phases. The baseline average was then subtracted from the intervention average, with a positive value indicating improved effects and a negative value indicating decreased effects. Metrics for interpreting G-index effect sizes are: small (< 0.3), medium (0.31 to 0.50), and large (> 0.51). Assessing social validity and unforeseen changes in participants. Each participant completed the ATT measure following the final session of their respective interventions. The counselor-investigator kept field notes for each participant throughout the study.