20250220 - directed acyclic graphs (DAGs)

_site.yml

@@ -30,6 +30,8 @@ navbar:
         href: dataManagement.html
       - text: Developmental Scaling
         href: developmentalScaling.html
+      - text: Directed Acyclic Graphs
+        href: dags.html
       - text: Electroencaphalography/Event-Related Potentials
         href: eegERP.html
       - text: Exploratory Data Analysis

dags.Rmd

@@ -0,0 +1,28 @@
+---
+title: "Directed Acyclic Graphs (DAGs)"
+---
+
+```{r setup, include = FALSE}
+knitr::opts_chunk$set(
+  echo = TRUE,
+  error = TRUE,
+  warning = FALSE,
+  comment = "")
+```
+
+When conceptualizing and designing a study, or when developing plans to test a research question, it is important to draw a directed acyclic graph (DAG).
+DAGs, like [path diagrams](figures.html#pathDiagrams), are causal diagrams.
+Causal diagrams depict the hypoothesized causal processes that link two or more variables.
+Path diagrams are typically used after analysis to describe and report the findings in analysis (when using path analysis, [factor analysis](factorAnalysis.html), or [structural equation modeling](sem.html)).
+By contrast, DAGs are particularly useful when designing a study or before analysis, because they can help specify which variables it is important to control for and—just as importantly—which variables it is important not to control for.
+
+When drawing a DAG for your study, draw all the variables that link the hypothesized cause to the hypothesized effect, including confounders, mediators, and colliders.
+In your study, it is important to control for confounders.
+Moreover, it is important not to control to control for mediators when you are interested in the total effect of the predictor on the outcome.
+In addition, it is important not to control for descendants of the outcome variable.
+When there is a collision, it is important not to control for the collider when examining the association between the two causes of the collider.
+The only time when one should control for a collider is when the collider is also a cause (i.e., confound) of both the predictor and outcome variable rather than a common effect of both.
+
+For more information on DAGs, including ancestors, descendants, confounders, and colliders, see here: https://isaactpetersen.github.io/Fantasy-Football-Analytics-Textbook/causal-inference.html#sec-causalDiagrams.
+
+After determining what variables are confounders and what are important to control for, there are various ways one can control for variables, as described here: https://isaactpetersen.github.io/Fantasy-Football-Analytics-Textbook/causal-inference.html#sec-causalInferenceControlVariables.

index.Rmd

@@ -12,6 +12,7 @@ knitr::opts_chunk$set(
 - [Bayesian Analysis](bayesian.html)
 - [Data Management](dataManagement.html)
 - [Developmental Scaling](developmentalScaling.html)
+- [Directed Acyclic Graphs](dags.html)
 - [Electroencaphalography/Event-Related Potentials](eegERP.html)
 - [Exploratory Data Analysis](eda.html)
 - [Factor Analysis](factorAnalysis.html)