Assignment: Chapter 17 Exercise: Bivariate Table
Discuss how you would interpret the following bivariate table for an imaginary study comparing the effectiveness of two interventions aimed at preventing future child abuse among high-risk parents.
Table X: Number of Parents Who Were and Were Not Found to Have Committed Future Abuse, by Type of Intervention.
TYPE OF INTERVENTION
OUTCOME INTERVENTION A INTERVENTION B
Abuse 10 10
No Abuse 30 20
TOTAL 40 30
Explain why you would want to employ inferential statistics to further analyze the data.
Present an illustration to understand effect size and its representations.
Assignment: Chapter 17 Case Study: Effect of Stress on Performance
Case Study:
A company conducted a survey to identify the effect of stress on performance. Three hundred candidates were selected. Participants were of the age group 30–45, and of both genders. The nature of work varied from managerial to administrative to operative to sales. Work timings were from 9.00 a.m. to 5.00 p.m.; however, some roles would require duties outside of the office. Based on the designation, salaries varied as well from a range of USD 15,000 to 50,000 per annum.
Identify from the above narrative the data that can be converted to numerals and codes for analysis.
What would be the most ideal form of analysis—Univariate, Bivariate, or Multivariate? Justify.
Identify variables that can be assessed using the levels of measurement.
Assignment: Chapter 17 What does this mean to me? Reflection Activity
Create a hypothesis for a basic causal study between two variables associated with counseling or social work in an area of interest to you. Outline the steps and methods you would implement to interpret your results.
Discuss the importance of obtaining substantive significance in research studies. How would you make sure the causal study you created has substantive significance?
Why is it important to be able to refute chance in examining relationships among variables? Describe a situation you might encounter as a social work practitioner in which you would need to be able to rule out the possibility of a chance of relationship between variables.
