{"id":337,"date":"2011-09-04T15:00:04","date_gmt":"2011-09-04T22:00:04","guid":{"rendered":"http:\/\/www.mii.ucla.edu\/causality\/?p=337"},"modified":"2011-09-04T15:00:04","modified_gmt":"2011-09-04T22:00:04","slug":"comments-on-an-article-by-grice-shlimgen-and-barrett-gsb-regarding-causation-and-judea-pearls-mediation-formula","status":"publish","type":"post","link":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/2011\/09\/04\/comments-on-an-article-by-grice-shlimgen-and-barrett-gsb-regarding-causation-and-judea-pearls-mediation-formula\/","title":{"rendered":"Comments on an article by Grice, Shlimgen and Barrett (GSB): “Regarding Causation and Judea Pearl’s Mediation Formula”"},"content":{"rendered":"

Stan Mulaik called my attention to a recent\u00a0article by Grice, Shlimgen and Barrett (GSB)\u00a0(linked here\u00a0http:\/\/psychology.okstate.edu\/faculty\/jgrice\/personalitylab\/OOMMedForm_2011A.pdf<\/a> )\u00a0which is highly critical of structural equation modeling\u00a0(SEM) in general, and of the philosophy and tools that I\u00a0presented in “The Causal Foundation of SEM” (Pearl 2011)\u00a0(\u00a0http:\/\/ftp.cs.ucla.edu\/pub\/stat_ser\/r370.pdf<\/a>.) \u00a0In particular, GSB disagree\u00a0with the conclusions of the Mediation Formula —\u00a0a tool for assessing what portion of a given effect is mediated through a specific pathway.<\/p>\n

I responded with a detailed account of the disagreements\u00a0between us (copied below), which can be summarized as follows:<\/p>\n

Summary<\/span><\/p>\n

1. The “OOM” analysis used by GSB is based strictly\u00a0on frequency tables (or “multi-grams”)\u00a0and, as such, cannot assess cause-effect relations\u00a0without committing to some causal assumptions.\u00a0Those assumptions are missing from GSB account,\u00a0possibly due to their rejection of SEM.<\/p>\n

2. I define precisely what is meant\u00a0by “the extent to which the effect of X on Y\u00a0is mediated by a third variable, say Z,” and\u00a0demonstrate both, why such questions are important in decision making\u00a0and model building and why they cannot be captured by\u00a0observation-oriented methods such as OOM.<\/p>\n

3. Using the same data and a slightly different design,\u00a0I challenge GSB to answer a simple\u00a0cause-effect question with their method (OOM), or with any\u00a0method that dismisses SEM or causal algebra as unnecessary.<\/p>\n

4. I further challenge GSB to present us with ONE RESEARCH QUESTION\u00a0that they can answer and that is not answered\u00a0swiftly, formally and transparently by the\u00a0SEM methodology presented in Pearl (2011).\u00a0(starting of course with the same assumptions and same data.)<\/p>\n

5. I explain what gives me the assurance that no\u00a0such research question will ever be found, and why even the late\u00a0David Friedman, whom GSB lionize for his staunch critics of SEM, has\u00a0converted to SEM thinking at the end of his life.<\/p>\n

6. I alert GSB to two systematic omissions from\u00a0their writings and posted arguments, without which\u00a0no comparison can be made to other methodologies:
\n(a) A clear statement of the research question that the\u00a0investigator attempts to answer, and
\n(b) A clear statement of the assumptions that the investigator\u00a0is willing to make about reality.<\/p>\n

Click here for the full response.<\/a> <\/p>\n

=======Judea<\/p>\n","protected":false},"excerpt":{"rendered":"

Stan Mulaik called my attention to a recent\u00a0article by Grice, Shlimgen and Barrett (GSB)\u00a0(linked here\u00a0http:\/\/psychology.okstate.edu\/faculty\/jgrice\/personalitylab\/OOMMedForm_2011A.pdf )\u00a0which is highly critical of structural equation modeling\u00a0(SEM) in general, and of the philosophy and tools that I\u00a0presented in “The Causal Foundation of SEM” (Pearl 2011)\u00a0(\u00a0http:\/\/ftp.cs.ucla.edu\/pub\/stat_ser\/r370.pdf.) \u00a0In particular, GSB disagree\u00a0with the conclusions of the Mediation Formula —\u00a0a tool for assessing […]<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11,26,29],"tags":[],"class_list":["post-337","post","type-post","status-publish","format-standard","hentry","category-discussion","category-mediated-effects","category-opinion"],"_links":{"self":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/posts\/337","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/comments?post=337"}],"version-history":[{"count":0,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/posts\/337\/revisions"}],"wp:attachment":[{"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/media?parent=337"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/categories?post=337"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/causality.cs.ucla.edu\/blog\/index.php\/wp-json\/wp\/v2\/tags?post=337"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}