1. Help Center
  2. Glossaries
  3. Individual glossary pages

Agent Based Approach (ABA)


an approach to systems understanding, problem solving, and analysis-synthesis of various kinds (e.g., process, policy, systems, etc.) that utilizes a CAS perspective as a frame; a specific stepwise method for approaching a given system or problem developed by Drs. Derek and Laura Cabrera.


First, you must understand DSRP, DSRP Method, and CAS, as well as m=it and the ST Loop

ABA is based on the underlying structure of CAS: independent agents + local simple rules >>> yield massively parallel collective dynamics >>>> which result in emergent properties/behaviors of the system. 

With these prerequisite understandings, you will know what you need to perform an ABA. Here are the Steps to ABA:



1. Understand your system. Develop a DSRP map of the system and iterate it until it passes “reality-test.” (Note: This part of the process is its own process in and of itself, that has to do with understanding how DSRP Mapping works)

Map out all the reasons women are underrepresented in science. 

2. Understand the current and desired [POSIWID] goal state of the system

A representative number(~50%) of female scientists

3. Make a short list (3-10) of the salient agents in system at all levels of scale

girls of all ages, parents, teachers, school administrators, schools, grant administrators, granting institutions, policy makers, cultural/political leaders

4. Come up with 3-10 “CAS principles” your and all future recommendations must meet. This list acts as a backcheck on proposed recommendations to see if they pass this litmus.

[These are examples based on the conclusions of systems structure and dynamics in step 1] 

Does it disincentivize girls from pursuing science? If so, rethink it. 

Does it discriminate against competence in science? If so, rethink it. 

Does the act/artifact communicate implicitly or explicitly a difference between males and females in regard to science ability? If so, rethink it. 

Does it cost more than what is currently being spent? If so, rethink it. 

5. Explicate the simple rules for each agent group. Fill in the “CAS Synthesis table” for all agents in item #3 above checking that no simple rule violates any of your CAS Principles in #4 above.

See completed table

6. Creatively create 3-10 recommendations. Be wildly creative but constrained by the principles. Any recommendation that  does adheres to the principles is fair game. 

Science Barbie

Priscilla Pal the Science Gal

By assembling the components of CAS/ABM and DSRP into a “method” (a stepwise process) we are better able to understand and change complex systems in the real world based on their underlying structure.


The artifact of “policy” and the notion of “policy analysis” comes into play here. We see of course that a “policy” is simply a set of guidelines for understanding how agent action (the following of simple rules) will affect emergent properties. We also see that “policy” in this framing of the term is quite a fractal, General, and abstract concept. That is, “analyzing or establishing policy” isn’t something that only a policy expert or government administrator does, but something a parent, team or organizational leader, community organizer, or numerous other roles must do. it makes little difference If the CAS you are attempting to understand and change is a family, a classroom, a team, an organization, a state or federal system, or a global crisis...the steps you must take to understand this CAS, identify the types of actions that can be taken to alter it, and codify those generalizable actions into specific ones (recommendations) is the same. The degree, scale, resources available, timeline, stakeholders, and relative complexity may change but the basic process does not. In other words, policy analysis is a fractal pattern. The methodology described above and the examples provided below are guideposts for and manifestations of this pattern.