2 question on second draft fix

Does the arrest of partners to victims of domestic violence in African-American populations lead to high chances of premature death of the victims?

Are the domestic violence victims more likely to die if their abusive partners are jailed or left to rehabilitate at homes?

2. State the main independent variables? Were any of the independent variables manipulated? Were there any unbalanced covariates and what are they? (Hint: see table 1, look for p-values < 0.10)

The researcher changes independent variables in research to measure the impact of that manipulation on the variable, which is dependent. A researcher can apply it at various levels to indicate any differences in the outcomes. The independent variables in the article were identified as the victim's race, gender, and employment status.

Several unbalanced covariates were also identified in the article as follows. As a way to raise the levels of precision of the approximation of the main effect (suspect arrest), a Cox hazard model was run by the authors taking into consideration the predictive influence of three imbalanced covariates discovered at the baseline

1. Victim has a prior arrest

2. Suspect employed

3. Suspect has a prior arrest

According to Sherman and Harris (2014), the group arrested possessed more prior victim arrests that led to lowered mortality prediction, a higher number of the working suspects, and therefore higher victim mortality; and a large number of suspects with priors that led to reducing of the predicted victim mortality.

3. What are the main dependent variables? Discuss how they were measured.

Dependent variable refers to the variable that is being subjected to testing or measurement in a given experiment. These variables change due to the manipulation of the independent variables of the investigation. The primary dependent variable was victim mortality.

The dependent variable was measured through the number of partner arrests that led to the victims' death compared to the suspects who were just issued with a warning.

• The hazard (risk) ratio as used by Sherman and Harris (2014) was a measure of the likelihood of death for domestic violence whose partners were arrested compared to those who were just warned. Sherman and Harris (2014) allude that 30 % of the recorded mortality would have been voided when there are no arrests, provided all other factors are held unchanged. Again, 40 % of deaths would have been avoided within the black victims, while for whites, it would have been 6 %.

4. Discuss in detail the data corpus that Sherman and Harris (2014) employed in their secondary analysis.

The primary role of data corpus is verification of a hypothesis concerning a language, such as the usage of given sounds and words. Mortality data is critical data that affects a population, for instance, the statistics for the number of deaths and their causes.

The data corpus of this mortality study comprised of two important sources of data: 

1. Purchased mortality data from the Wisconsin Office of Vital Statistics

2. Death Master File (DMF), also commercially known as the Social Security Death Index (SSDI)  

3. MILDVE (1992) experiment data

As per the National Information Technical Service website, the Death Master File (DMF) derived from the Social Security Administration (SSA) has more than 83 million death records as reported to SSA. Further, the file contains data such as social security number, name, date of birth, and death.

5. What is the role of random assignment in their study? Please discuss. Define Intention-to-Treatment (ITT) analysis. Try your best to explain why, Sherman and Harris employed ITT analysis in their study.

Random assignment describes the utilization of chance processes in experiments to ensure each participant possesses a similar space to be allocated to any particular group. Sherman and Harris (2014) used random assignment to assist in the reduction of any potential systematic differences for groups at the time an experiment starts and, therefore, mitigation of any threats to the variables. However, this process didn't always equalize the confounding variables.

Intention-to-Treatment analysis refers to a method of doing results analysis in randomized studies in which all the participating subjects are made part of the analysis statistically and analyzed as per categories they were initially assigned, regardless of any form of treatment they received.

According to Sherman and Harris (2014), they used the intention-to-treat analysis to assess the long-term differences between the victim mortality.

6. Discuss in detail the findings of the Sherman and Harris (2014) study.

The results, according to Sherman and Harris (2014), are robust, as shown by the several sensitivity tests. However, the imbalances measured in covariate units show that even though the random assignment was utilized, results still could contain imbalances in the factors not measured but that are also predisposing `to the victim mortality.

Sherman and Harris (2014) find that arrests related to domestic violence had a compromising effect on the status of women as far as respectability is concerned, especially in the African-American women population.

Also, the African-American victims of domestic violence are more likely to pass away after an arrest of a partner as compared to the whites. Furthermore, the study found out that there could be more stress from abrupt demotion in the social aspects and which could be attributed to the deaths.