{"id":643,"date":"2018-06-18T07:30:00","date_gmt":"2018-06-18T11:30:00","guid":{"rendered":"http:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/?p=643"},"modified":"2018-06-17T18:13:33","modified_gmt":"2018-06-17T22:13:33","slug":"causal-inference-in-ecology-concluding-thoughts","status":"publish","type":"post","link":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/blog\/2018\/06\/18\/causal-inference-in-ecology-concluding-thoughts\/","title":{"rendered":"Causal inference in ecology – Concluding thoughts"},"content":{"rendered":"

Causal inference in ecology – links to the series<\/a><\/p>\n

Last week<\/a> I concluded that the Rubin causal model isn\u2019t likely to help me make causal inferences with the kinds of observational data I collect. I also argued that <\/p>\n

\nIt does, however, illuminate the ways in which additional data from different systems could be combined (informally) with the data I collect1<\/a><\/sup> to make plausible causal inferences.\n<\/p><\/blockquote>\n

From the one data set I analyzed last week, I concluded that we could see an association<\/strong><\/em> between rainfall and stomata density in Protea<\/em> sect. Exsertae<\/em> but that we couldn\u2019t claim (on the basis of this evidence alone) that the differences in rainfall caused<\/strong><\/em> differences in stomata density. Why do I claim that \u201cadditional data from different systems [can] be combined (informally) with [these] data to make plausible causal inferences\u201d? Here\u2019s why.<\/p>\n

Think back to when we discussed controlled experiments<\/a>. I pointed out that by randomizing individuals across treatments we statistically control for the chance that there\u2019s some unmeasured factor that influences the results. It\u2019s not as good as a perfectly controlled experiment in which the individuals are identical in every way except for the one factor whose causal influence we are trying to estimate, but it\u2019s pretty good. Well, if we have a lot of observations from different systems – different taxa, different ecosystems, different climates – and we get higher stomata densities in areas with more annual rainfall, as we did in Protea<\/em> sect. Exsertae<\/em>, we also know that these other systems differ from Protea<\/em> sect. Exsertae<\/em> in many different ways in addition to those having to do with annual rainfall. That\u2019s not as good as randomization, but it suggests that the association we saw in that small group of plants in the Cape Floristic Region is similar to associations elsewhere. That means the association is stable across a broader range of taxa or ecosystems or climates, or all three than our limited data showed, suggesting that there is a causal relationship.<\/p>\n

Now it still doesn\u2019t show that it\u2019s mean annual rainfall, per se<\/em>, that matters. It could still be something that\u2019s associated with mean annual rainfall not only in the CFR but also in the other systems we studied. If we happened to find that the association always held, that it was never violated in any system we still couldn\u2019t exclude the possibility that the \u201ctrue\u201d causal factor was this other thing we aren\u2019t measuring, but it begins to become a bit implausible – rather like claiming that it\u2019s not smoking that causes cancer, it\u2019s something else that\u2019s associated with smoking that causes cancer.2<\/a><\/sup><\/p>\n

This kind of argument doesn\u2019t produce logical certainty, but re-read the post on falsification<\/a> and you\u2019ll see that even if a well-controlled experiment fails to give the results predicted by a hypothesis, it is very difficult to be sure that it\u2019s the hypothesis that\u2019s wrong. It may be that the experimental conditions don\u2019t match those presumed by the hypothesis, in which case we can\u2019t say anything about the truth or falsity of the hypothesis. In other words, even the classical hypothesis test can\u2019t reject a hypothesis with certainty. There\u2019s always judgment involved. It can\u2019t be escaped.<\/p>\n

Bottom line: If you\u2019re willing to reject a hypothesis based on a failed experiment because you\u2019re willing to examine all of the factors influencing the experimental conditions and conclude that none of them are the problem,3<\/a><\/sup> you should be as willing to use evidence from a range of associational studies combined with some theory (whether a formal mathematical model or verbal description of the mechanics of a system) to build a case for a causal relationship from observational data. In neither case will you be certain of your conclusions. Your conclusions will merely be more or less plausible depending on how much and how strong your evidence is.<\/p>\n

As scientists,4<\/a><\/sup> we are more like detectives than logicians. We build cases. We don\u2019t build syllogisms.<\/p>\n

    \n
  1. Remember what I wrote in that last footnote. ↩<\/a><\/li>\n
  2. You could argue that if the two factors, the \u201ctrue\u201d causal factor and the one we measure, are invariably connected that there is really only one factor. That\u2019s a longer philosophical discussion that I don\u2019t have the energy to get into – at least not now. ↩<\/a><\/li>\n
  3. Notice that reaching this conclusion depends on your background knowledge about the system and its components, i.e., prior knowledge, not observations from the experiment itself. ↩<\/a><\/li>\n
  4. Or at least as ecologists and evolutionists. ↩<\/a><\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"

    Causal inference in ecology – links to the series Last week I concluded that the Rubin causal model isn\u2019t likely to help me make causal inferences with the kinds of… Read more ><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[],"class_list":["post-643","post","type-post","status-publish","format-standard","hentry","category-statistics"],"_links":{"self":[{"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/posts\/643","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/comments?post=643"}],"version-history":[{"count":2,"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/posts\/643\/revisions"}],"predecessor-version":[{"id":645,"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/posts\/643\/revisions\/645"}],"wp:attachment":[{"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/media?parent=643"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/categories?post=643"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/darwin.eeb.uconn.edu\/uncommon-ground\/wp-json\/wp\/v2\/tags?post=643"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}