There is a method called "Functional Resonance Analysis Method" (FRAM) that might interest you. It is in a book called "Barriers and Accident Prevention" by Eric Hollnagel, who has worked extensively in accident analysis and resilience engineering. 

A link to the publisher's website is here: http://www.ashgate.com/isbn/9780754643012

Hope it helps.

I am attaching a couple of articles by Robins and Hernan - they are top shelf epidemiologists and methodologists. I'd encourage you to look up their other papers as well.

Judea Pearl has many resources on his webpage (http://bayes.cs.ucla.edu/jp_home.html). Depending on what you want, the first place I often point people to is this intro to causality lecture (http://singapore.cs.ucla.edu/LECTURE/lecture_sec1.htm) which is also in the appendix of the 2nd edition of his book, Causality.

ps. I was just re-looking at his webpage, also worth watching is his Turing Lecture (http://amturing.acm.org/award_winners/pearl_2658896.cfm).

On causal diagarams to analyse using computers, prefer XML in specific OWL.

SDG serves a generalized tool. 

On data based causality analysis refer to "Capturing Connectivity and Causality in Complex Industrial Processes" by F.Yang et al

http://www.dagitty.net/ is a good resource to build your own DAGs and have a basic understanding of the terminology.

I found this book very useful as it synthesizes across a number of different traditions -it has many causal diagrams 

Stephen L. Morgan and Christopher Winship (2007) Counterfactuals and Causal Inference Methods and Principles for Social Research

http://www.cambridge.org/gb/academic/subjects/sociology/sociology-general-interest/counterfactuals-and-causal-inference-methods-and-principles-social-research

This site sets out a number of papers from across the social sciences and beyond

http://www.wjh.harvard.edu/~cwinship/cfa.html

I also found this on Graphical Causal Models to be insightful

http://www.ssc.wisc.edu/soc/faculty/pages/docs/elwert/Elwert%202013.pdf

This book is particularly clear

Bill Shipley. Correlation in Biology: A User’s Guide to Path Analysis, Structural Equations, and Causal Inference. 

http://www.billshipley.recherche.usherbrooke.ca/my%20book.htm

Imbens and Rubin's book on causal inference just arrived on my doorstep (it's new). I'll start reading so and will report back, but based on other work by the authors it will likely be excellent.

Been looking at Imbens and Rubin. It is a different approach from the causal diagram approach, and they state:

``Pearl's work is interesting, and many researchers find his arguments that path diagrams are a natural and convenient way to express assumptions about causal structures appealling. In our own work, perhaps influenced by the type of examples arising in social and medical sciences,  we have not found this approach to aid drawing of causal inferences, and we do not discuss it further in this text.''  Imbens and Rubin (2015, p.\ 22)

So the book covers a lot on causal inference (particularly for different designs), but not on the diagrams. Rubin and Pearl have debated their merit.

Daniel, you and I are always on the same page! I just got my copy of the book as well! This is a marvelous book, but it will take the better part of a year to get through. Nonetheless, this book will serve as the encyclopedia for causal inference (in the binary treatment case, as the authors point out).

Good to see you on these pages. Yea, at 600+ pages it will be a lot of coffee breaks, and it really isn't a coffee break book. They use it for a one-term course, so I guess that is a guide to how much work is needed (though hopefully much is review).

Nisa, we've written a brief tutorial on casual identification and estimation, which covers some of the key concepts (see attachment). We also have a video recording of a seminar on the same topic (see attached link).

https://bayesia.wistia.com/medias/71n20ufyv9

Article Causality for Policy Assessment and Impact Analysis - Direct...

Hi Nisa, Hi All,

Thanks for the interesting links and comments. We may also be interested in the upcoming book by Miguel Hernan and Jamie Robins on Causal Inference.

! Drafts of selected sections are available on Miguel Hernan´s website at Harvard:

http://www.hsph.harvard.edu/miguel-hernan/causal-inference-book/

Thanks very much everyone for your valuable input and suggestions! These are very interesting reads. Just to share with everyone, I'm also reading Elwart's paper which gives examples in sociology.

Best wishes,

Nisa

Daniel Wright Hi Daniel, why Rubin said the diagram doesn't aid drawing of causal inferences? Is it the causal diagram is problematic or just the diagram doesn't help much in the problems he's interested.

I'm reading Judea's Causality: models, reasoning and inference, in the book he proved the equivalence of his definition and Neyman's theory, which is adopted in Rubin's book.

Senhui Guo , short answer is I don't know; I'm not him. Maybe he just just like equation-format more than diagram format (he is much better in math than most!). Slightly longer answer is that Rubin and Pearl have each been addressing how to estimate causation in non-experimental settings, but have come about it from different perspectives. Both likely think their perspective is more intuitive. Both are brilliant. I see people argue for one approach "as better" than the other. I think that is probably the wrong approach. In my experience I tend to think, "oh wow, X says this" and then read the related paper by Y and see different (long and confusing sounding words ... by both) and see they are related. For good insight into the discussion look for when Pearl is commentating on Gelman's blog.

Daniel Wright Thanks! didn't know there's such a blog. Also, do you know if there're any other blogs or communities, groups that people are actively discussing about these fundamental causality problems? I've seen different perspectives, Judea's, Bernhard's (his principle of independent mechanisms), it's like they're all talking about totally different aspects of causality, and don't know the other's theory, which is quite confusing for people new to this area, like me, about what we should focus on.

Senhui Guo , Gelman's would be in the statistics arena, and there are ones in graphs. Other than Gelman's and my niece's on books she reviews I don't look much at blogs. Most seem just tips based on how the author teaches topic X or outside of academia people writing, well, I dunno why. The amount people have to say has increased with the channels that allow them to say thing.