The Logic of Inference: The Science of Uncertainty

Describing and explaining social phenomena is a complex task. Box’s quote “All models are wrong. Some models are useful” speaks to the point that it is a near impossible undertaking to fully explain such systems—physical or social—using a set of models. Yet even though these models contain some error, the models nevertheless assist with illuminating how the world works and advancing social change.

The competent quantitative researcher understands the balance between making statements related to theoretical understanding of relationships and recognizing that our social systems are of such complexity that we will always have some error. The key, for the rigorous researcher, is recognizing and mitigating the error as much as possible.

As a consumer of research, you must recognize the error that might be present within your research and the research of others.

To prepare for this Discussion select one of the quantitative articles with social change implications attached. As you read the article, reflect on George Box’s quote.

Post a brief description (2–3 sentences) of the article you found.

Describe how you think the research in the article is useful (e.g., what population is it helping? What problem is it solving?).

Using Y=f(X) +E notation, identify the independent and dependent variables.

How might the research models presented be wrong? What types of error might be present in the reported research?