Category: Suicide and Self-Injury

Symposium

Exploratory Data Mining With the Application of Exploratory Mediation for Nonsuicidal Self-Injury

Friday, November 17
12:00 PM - 1:30 PM
Location: Cobalt 502, Level 5, Cobalt Level

Keywords: Statistics | Self-Injury | Suicide
Presentation Type: Symposium

Given the increase of exploratory data mining applications in psychology, most research has focused on the use of univariate approaches. In contrast, multivariate extensions of various exploratory data mining methods combine both confirmatory and exploratory aspects of statistical modeling. The first part of the talk focuses on multivariate extensions of decision trees. One method includes structural equation model trees (SEM Trees; Brandmaier, von Oertzen, McArdle, & Lindenberger, 2013), which has been used to create clinical cutoffs for non-suicidal self-injury (Ammerman, Jacobucci, Kleiman, Muehlenkamp, & McCloskey, 2016). In a similar vein, mixed effects decision trees can be applied to ecological momentary assessment data through the longRPart2 package (Jacobucci, Serang, & Grimm, 2017). The second part of the talk focuses on how regularization has been extended to multivariate models, in the form of both network psychometrics (Epskamp, Rhemtulla, & Borsboom, 2017) and regularized structural equation modeling (RegSEM; Jacobucci, Grimm, & McArdle, 2016). RegSEM, seen as a family of tools for regularization with psychological and behavioral data, offers a flexible approach to model generation and modification, with an emphasis on generalizability. Detail is provided on RegSEM as implemented in exploratory mediation (Serang, Jacobucci, Brimhall, & Grimm, 2017), covering how this can be used as an exploratory method for identifying mediators. This is applied with an example from the National Longitudinal Study of Adolescent to Adult Health dataset. Each method is introduced briefly, next providing a simplistic example for pedagogical purposes, and ending with a reference to programming code and further resources. Throughout the presentation, topics unique to clinical research, such as missing data, measurement error (and the need for latent variables), and finally complex interactions and models, are discussed and how data mining methods can be tailored to these issues.

Ross Jacobucci

University of Notre Dame

Presentation(s):

Send Email for Ross Jacobucci


Assets

Exploratory Data Mining With the Application of Exploratory Mediation for Nonsuicidal Self-Injury



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Exploratory Data Mining With the Application of Exploratory Mediation for Nonsuicidal Self-Injury