Structured Analysis: Methods and Practice

In the beginning of ‘Structured Analytic Techniques for Intelligence Analysis’, Richards J. Heuer Jr. and Randolph H Pherson identify four broad categories of analytic methods. These identified approaches are distinguished by the name of the analytic methods used, the type of quantification if any, the type of data that are available, and the type of training that is expected or required. Each of the four categories is distinct, however at times, there is considerable overlap amongst particular categories. These four broad categories are Expert Judgment, Structure Analysis, Quantitative Methods Using Expert-Generated Data and Quantitative Methods Using Empirical Data.
In the beginning of ‘Structured Analytic Techniques for Intelligence Analysis’, Richards J. Heuer Jr. and Randolph H Pherson identify four broad categories of analytic methods. These identified approaches are distinguished by the name of the analytic methods used, the type of quantification if any, the type of data that are available, and the type of training that is expected or required. Each of the four categories is distinct, however at times, there is considerable overlap amongst particular categories. These four broad categories are Expert Judgment, Structure Analysis, Quantitative Methods Using Expert-Generated Data and Quantitative Methods Using Empirical Data.

Expert Judgement is considered to be the most traditional form of intelligence analysis. Properly executed expert judgement should combine subject matter expertise and critical thinking. A key characteristic of expert judgement, from structure analysis, is that it is usually an individual analyst’s effort, instead of an effort by a team of analysts. Until the information is written down and conveyed to others, an analyst applying expert judgement has exclusive control and access to their information. If information is withheld or there is a failure to convey it, expert judgement can create an informational bottleneck effect. This technique usually requires a postgraduate education in social sciences or liberal arts fields and is often coupled with country or language expertise.

Structure Analysis requires a step-by-step process that externalizes analysts thinking, making it apparent to other analysts. Structure analysis usually becomes a collaborative effort with different and potentially conflicting perspectives. This allows for great access of information amongst analysts and prevents any bottlenecking of information (which could arise during the application of expert judgement). Techniques commonly applied during structure analysis include Structure Brainstorming, Scenarios, Indicators, Analysis of Competing Hypotheses and Key Assumptions Check. Typically this knowledge and its application was learned within the Intelligence Community, however now some colleges and universities are beginning to offer related programs of study.

Quantitative Methods Using Expert-Generated Data is applied when there is an absence of empirical data. The absence of empirical data, can lead to the generation of quantitative data by expert opinion and subject probability judgements. This category also includes the application of the following techniques, Bayesian Inference, Dynamic Modeling and Simulation. This technique usually requires a graduate education in the fields of mathematics, information science, operations research, business, or other related sciences.

Quantitative Methods Using Empirical Data varying considerably from expert-generated data. Econometric modeling is one common example of this method. Here empirical data are collected through various types of sensors and are utilized. This technique typically requires graduate level education in statistics, economics or hard sciences.

While each of these four broad categories are important and deserve further study and observation, Heur and Pherson’s focus lies exclusively with Structured Analysis. Structured Analysis requires a step-by-step process amongst analysts that externalizes their thinking which allows greater access to information and multiple perspective testing and scrutiny.
blog comments powered by Disqus