For this assignment, you will answer the evaluation questions that you developed (information provided) You will use the appropriate statistical procedure to determine if there was a significant difference in the outcomes. There should be a minimum of two analyses. One should answer the question related to the proximal outcome for your intervention. The second question is to answer the question related to the distal outcome for the intervention. Please state each question that was articulated and provide a statement of the null and alternate hypothesis for each question. Present the results and interpretation of the results in terms of the null hypothesis. There should be a minimum of one chart or tables in the document. The narrative should refer to the charts or tables. The tables are to be incorporated in the text and not placed in an appendix. Please read and apply to assignment Parametric vs Non-Parametric Tests and Assumptions Hopefully, by now you have noticed that when discussing option for analysis, the text refers to “parametric” and “non-parametric” statistical tests. Parametric tests are used “with dependent variables that are at the interval or ratio level of measurement.” (Rubin, p. 157, 2013). Many of the statistical procedures we encounter are parametric, including correlation, regression, t -tests and ANOVA. An additional consideration for the use of these statistics for analyzing data is establishing whether the characteristics of the data to be analyzed meet the assumptions necessary for valid interpretation of the results. Two assumptions common to parametric tests is the assumption of normality of the means for variables of groups that are compared and equality of variances for those groups. With sample sizes of 30 or more, the violation of the normality assumption is usually not a significant issue. However, to be sure various univariate analyses such as Shapiro-Wilk’s significance test can be used. The measure of skewness and kurtosis are also used to assess whether a distribution of values conform to the assumption of normality. Graphical methods such as the histogram could give a rough indication of normality. Box plots and Normal Probability Plots also give a bit more detailed information on the distribution of data and its conformance with normality. Tests for the equality of variances could include the Bartlett’s or Levene’s tests. Many other procedures are available for testing these assumptions. There are other assumptions, depending on the statistical procedures requiring various tests. Non-parametric tests are used when the variables for analysis are not at the interval or ratio level. These tests may also be used when the data does not meet the assumptions for parametric statistical procedures. Post Hoc Tests Although post hoc tests won’t be part of the statistics that will be used in this class, it is still good to have a knowledge of what they are and the purpose of using them. These tests are designed for situations in which the researcher has already obtained a significant result with an ANOVA test. The ANOVA test only tells you that there is a signficant difference between the means of groups—but if there are three or more means being compared, you don’t know which mean or means are significantly different from the others. Post hoc tests provide this information. Among the more popular tests are Scheffe test, the Least Significant Difference (LSD) test, and Tukey’s HSD. Effect Size Statistics The effect size is a measure of the strength of a phenomenon or a sample -based estimate of that quantity. An effect size calculated from data is a descriptive statistic that conveys the estimated magnitude of a relationship. It does not provide any information regarding whether the data is generalizable to the population- there is no measure of statistical significance. There are several effect size statistics, three will be highlighted here. Although there is a numeric indication of the strength of the relationship, the qualitative distinction of “small”, “medium” and “large” effect size is often used. The Pearson r correlation is an effect size measure. As you are aware, the values for correlation range from -1 to 1. According to Cohen (1988, 1992), the effect is is small if the value of r is around .1, medium if it is around .3 and large if r is around .5. Cohen’s d effect size is the difference of two population means and is divided by the standard deviation from the data. The formula: For nominal data, Chi-square statistic is used. If there are two categories to the data, Cramer’s phi can be used. If there are more than two, Cramer’s V statistics can be used. References Bettany-Saltkov, J. & Whittaker, . (2013). Selecting the most appropriate inferential statistical test for your quantitative research study. Journal of Clinical Nursing, 23, 1520-1531. Ham, ., Huggins-Hoyt, Y. & Pettuss, J. (2016). Assessing statistical change indices in selected social work intervention research studies. Research on Social Work Practice, 26(1), 44-52. Rubin, A. (2010). Statistics for evidence-based practice and evaluation, (2nd ed.). Brooks Cole. please provide Permalinks for all citations used The Templeton Family Services & Support Program seeks to support households with children who have special needs. It is characterized by three main programs namely After School Program- Templeton Middle School, Anger Management Skills Training, and Parenting Education Program. The project also captures an important aspect highlighted by Doucet, et al,. (2010). It agrees to the fact that parents undergo mental health problems and need programs that will assist them manage their anger which is usually the cause of child abuse and neglect. These programs have proximal (goals to be achieved over a short period of time) and distal outcomes (long-term objectives to be attained over a long period of time). Rubin (2013) highlights the research process and the significance of the research questions. Therefore, the study captures two important questions that will become the basis for the evaluation. Evaluation Question linked to Proximal Outcome Question 1. How has the participant been able to manage frustration in their daily life after the 10-week anger management program? To determine if the parent or adult has been in a position to manage frustration, the Inventory for Assessment of Stress Management Skills is going to be used. The 17-item scale evaluates the respondent’s point of view on their ability to use stress management approaches (Templeton Family Services & Support Program, .). The measure will request the participant to rate 5-point response scale on the effectiveness of managing stress. Evaluation Question linked to Distal Outcome Question: Has the participant developed a set of strategies to use in effectively managing anger-related events five to ten years to come? To ascertain if the participant has gained strategies to manage anger-related events five to ten years to come, the Inventory for Assessment of Stress Management Skills (ISBF) will be used. ISBF has the ability of determining the kind of strategies the participants have picked and the ability to implement it. It will show if the participant undertakes the 17 items in the scale extremely well or not (Templeton Family Services & Support Program, ). The higher the value and total score the more the ability of the individual to use stress management skills five to ten years to come. Reference Doucet, ., Letourneau, ., & Stoppard, . (2010). Contemporary paradigms for research related to women’s mental health. Health Care for Women International, 3, 296 – 312. Rubin, A. (2013). Statistics for Evidence-Based practice and evaluation (3rd ed). Brooks Cole. Templeton Family Services & Support Program. (.). About.