Assignment task:
In depth respond to the below paragraphs. What were good points? What can they work on? what research question can help them go deeper? Include apa sources and one bible verse
Descriptive inference by KKV Chap 2 is the process of using systematic/specific methods in describing the state of the world (as perceived) or a particular phenomenon based on data (The reliance on objectivity). This methodology relies on the process of summarizing observations, establishing patterns, and making statements about relationships w/o necessarily implying causality. This is a very crucial aspect of social science research due to the fact that it provides the basis or foundation upon which causal inferences are built.
Causal inference is the process of determining or establishing whether a cause-and-effect relationship exists between variables and estimating the magnitude and direction of effect, which is a vector quantity. This often involves careful research designs such as experiments, observational studies, or statistical methods (i.e., Models) for the purpose of isolating the causal effect from the confounding factors associated with the system being studied.
They believe in increasing the number of observations (n) because a larger number of observations typically improves the validity, reliability, and robustness of inferences made from the data collected. With that said the increase in (n) will typically enhance generalizability, improve the statistical power of the observations, reduce inherent biases and random errors( epsilons), provide better control of confounding variables, the inclusion of more variation for the analysis, and improvement in the causal inference.
Lijphart suggests that one should increase the number of cases which means expanding (n), concentrating the numerical order of comparable cases, reducing the number of variables, and the use of statistical methods for the purpose of controlling variables. By addressing the notion of increasing the number of cases or expanding (n), researchers can improve the robustness of their findings, limit errors or epsilons, as well as strengthen the overall reliability of the conclusions. Addressing the notion of focusing on comparable cases, requires selecting cases that are similar, thus limiting or controlling variability within the system. Addressing the notion of reducing the number of variables, requires being very selective when choosing variables critical to the study or analysis, and finally the use of statistical methods for the purpose of controlling the variable.
Causality plays a role by identifying key relationships, developing an explanatory framework, having a predictive power, and the improvement of the theory refinement and development. "KKV understand process tracing as the search for intervening variables that link an independent variable with dependent variable." (Mahoney, 2010, p 123). This overall shows that causality by default moves research beyond correlation, which in turn allows the researcher to explain and make predictions based on the findings.
This requires the formulation of explanations that will articulate how and why cause leads to an effect. This will give a template that will lead to a better understanding of the variables in the context of their relationships to each other. This in turn should allow for the ability to utilize testing and/or refinement. "Complexity begets an occupational hazard for all policy scholars: access to valid, reliable, and useful data." (Kay, 2015, p 2).
It should be first noted that observable implications are testable predictions that should be observed if a causal theory is correct. This is predicated on data patterns that fall in line with causal relationships proposed by a theory. This will for all practical purposes allow for the empirical testing of a theory. Addressing causal mechanisms is the process of explaining how and why a cause leads to an effect. With all that said and done, it provides the logic or reasoning that connects the independent variable to the dependent variable. The fundamental differences lie in the form over function, which deals with having a correct theory as opposed to the process that is limited to the cause and effect. "Indeed, the policy mechanism approach will not rival established and emerging policy process frameworks -- rather, it collectively challenges them and points to a way forward." (Lindquist, p 34).
Social scientists will typically take the approach of process tracing, comparative case studies, experiments, interviews, etc. These are some of the most common methods at hand. In the context of process tracing, this approach is a qualitative method that involves or requires the examination of steps that will have a link of a cause to an effect. This is generally deflected in a limited number of cases that are studied.
Some of the main challenges associated with the identification of causal mechanisms stem from the difficulty of measuring, observing, and establishing a validation process which creates a nexus between cause and effect. Addressing the issue of measurement, there can be difficulties measuring the components of the mechanism, which often derives from the fact that the variables are abstract by nature, complex, and/or subjective by nature. Addressing the issue of observing, unobservable processes could take place which could create a skewing effect on the analysis. This skewing effect will create a distortion in the established conclusion of the study which by default will be misleading. To further note, this will make it very difficult to gather evidence of the mechanism at play, which in turn will bring into question the reliability of the validation process.