• Department
  • Undergraduate Program
  • Graduate Program
  • People
  • Research
  • GIVE
  • UC Davis Psychology Quantitative Program

    Quantitative Brown Bag Series

    Location:  Young Hall 166 (Unless specified otherwise)
    Thursdays from 1:35 PM - 2:35 PM (Unless specified otherwise)
    Edit BB Events

    Calendar Administrator:   


    ACADEMIC YEAR:      2014 - 2015 Print Page
    Fall 2014
    Bayesian versus Frequentist Estimation of Multitrat-Multimethod Confirmatory Factor Models 
    SPEAKER: Dr. Jonathan Helm

    Campbell and Fiske's (1959) separation of trait, method, and unique variance across a set of multitrait-multimethod (MTMM) manifest variables directly translates to a confirmatory factor model, and several reports support this approach for partitioning variance (Cole, 1987; Widaman, 1985; Schmitt & Stults, 1986). However, researchers selecting this approach often encounter estimation problems (i.e., failed convergence or solutions with out-of-bounds estimates; Widaman, 1985). Mathematical investigations have identified several potential sources of these problems (Kenny & Kashy, 1992; Grayson & Marsh, 1994), forcing applied researchers to face an analytic conundrum when performing MTMM data analysis. The advent of Bayesian estimation for structural models offers many new opportunities, including the ability to fit models that would fail to converge when estimated within a frequentist framework (Scheines, Hoijtink, & Boomsma, 1999; Asparouhov & Muthén, 2010). Based on the non-identification problems that typically arise when fitting the CTCM model to MTMM data (Kenny & Kashy, 1992; Grayson & Marsh, 1994), and extra modeling flexibility provided by Bayesian estimation, the current paper examines the differences between maximum-likelihood (ML) and Bayesian estimation of the CTCM model. *Prior to Dr. Helm's presentation Dr. Emilio Ferrer will be introducing our area's new graduate students. He'll also be giving updates on the area search for a new faculty member, with emphasis on opportunities for us to attend prospective candidates' job talks later this quarter and/or early winter. The meeting will conclude with asking for volunteers to present at fall brown bag meetings.
    Modeling Time-varying Interdependence 
    SPEAKER: Dr. Jonathan Helm

    Inter-dependence between two individuals may be estimated using a variety of statistical techniques, but these models typically assume that dependence remains constant across repeated measures. To the extent that theory predicts different patterns of dependence as a function of time, current analytic approaches may not adequately test relevant hypotheses. This dissertation focuses on the development of a novel method for analyzing changes in dependence, as it unfolds over time for a sample of dyads. The first chapter gives detail of a specific theory that may benefit from the new method, explains why several common methods cannot investigate change in dependence, and outlines criteria for an approach to summarize change in dependence appropriately. The second chapter introduces an approach that satisfies the criteria, describes the technique analytically, and tests its mathematical properties via simulation. The third chapter describes an extension of the method that accounts for measurement error, and, via simulation, summarizes situations when the extension out-performs the simpler version. The fourth chapter provides an application of the method to empirical data to inform the theory outlined in the first chapter. The fifth chapter summarizes the benefits gained from the method, describes the limitations and assumptions, and suggests future steps for further innovation to the proposed method.
    Fit Index Sensitivity to Restricted Factor Analysis Model Misspecification 
    SPEAKER: Joseph E. Gonzales

    While there is contention over the use of fit indices (Barret, 2007; Bentler, 2007), they are often used to assess whether models are tenable representations of data, often utilizing rules of thumb (Hu & Bentler, 1999) that, despite being over-generalized (Marsh, Hau, & Wen, 2004), have been widely adopted. Previous work has shown that interindividual models of individual data are relatively insensitive to heterogeneity in intraindividual factor structures with respect to factor loadings (Molenaar, 2004). In the present study I expand on this finding using simulated data to evaluate if fit indices are sensitive to heterogeneity in factor structure when cases generated using either a one or two-factor model are randomly mixed and fit with a one or two-factor CFA model. Results suggest that AIC and BIC were able to discriminate between one and two-factor models with as little as ~10% heterogeneity (9:1 one-factor to two-factor cases), but CFI, TLI, and RMSEA all failed to reject the one-factor model until heterogeneity reached ~20%. SRMR performed the worse; failing to reject the one-factor model until ~55% heterogeneity was present. The most efficient measure of fit was the chi-square, which rejected the one-factor model with the smallest proportion of heterogeneity considered (5%). Note that the two-factor model was not closely evaluated for rejection, as its fit to data was consistently excellent. In the case of the two-factor model, inspection of the correlation between the two factors could be used to justify appropriateness of a one-factor model. However, with greater heterogeneity of one to two-factor model generated data, estimated correlations will be reduced depending on the strength of the correlation in the generating two-factor model. Consequently, inspection of the estimated factor correlations, an obvious indicator for the one-factor model, would likely be less informative, or possibly masked depending on the amount of heterogeneity in the sample.
    Brown Bag Meeting Cancelled 

    No Quantitative Brown Bag Today
    The mad-genius paradox: Can creative people be more mentally healthy but highly creative people more mentally ill? 
    SPEAKER: Dr. Dean K. Simonton

    The persistent mad-genius controversy concerns whether creativity and psychopathology are positively or negatively correlated. Remarkably, the answer can be "both"! The debate has unfortunately overlooked the fact that the creativity-psychopathology correlation can be expressed as two independent propositions: (a) among all creative individuals, the most creative are at higher risk for mental illness than are the less creative and (b) among all people, creative individuals exhibit better mental health than do non-creative individuals. In both propositions, creativity is defined by the production of one or more creative products that contribute to an established domain of achievement. Yet when the typical cross-sectional distribution of creative productivity is taken into account, these two statements can both be true. This potential compatibility is here christened the mad-genius paradox. This paradox can follow logically from the assumption that the distribution of creative productivity is approximated by an inverse power function called Lotka's Law. Even if psychopathology is specified to correlate positively with creative productivity, creators as a whole can still display appreciably less psychopathology than in the general population because the more at risk creative geniuses represent an extremely tiny proportion of those contributing to the domain. The hypothesized paradox has important scientific ramifications.
    "Issues, Questions, and Tentative Answers in Studying Physiological Synchrony" and "The Autonomic Regulation of Parenting: Parasympathetic Control Supports Positive Parenting" 
    SPEAKER: Dr. Paul Hastings & Jonas Miller

    Human biological systems vary, at least in part, according to external stimuli. Accordingly, many research enterprises aim to understand the interwoven nature of environment and physiology. Social interaction represents one specific stimulus of great interest, and many investigations report correspondence between one individual’s behavior and another’s physiology. Yet, given that social interaction requires at least two people, an extended perspective examines the association between individuals’ physiology, often called physiological synchrony. In this case physiological responses are repeatedly measured for both individuals and are modeled as a dependent system (i.e. each individual’s current response predicts the other’s concurrent, or future, response). This perspective offers several testable hypotheses, including the presence of physiological covariation and between-dyad differences in that covariation. Although these tests may be appealing, proper application requires a statistical model that aligns with those hypotheses. The current presentation outlines several issues that arise from the assumptions inherent in the statistical models most often used to test physiological synchrony, questions these raise about our understanding of physiological synchrony, and suggestions for modifications of analytic approaches that may advance this area of work.

    Raising children can be challenging at times, but reacting to one's child as a threatening stimulus - that is, responding to an interaction as a fight-or-flight situation - would undermine a parent's ability to manage challenging interactions effectively. Emotion regulation promotes positive parenting, and physiological activity is a critical component of emotion regulation. In this talk, I will present findings from two data sets on the links between autonomic control of cardiac arousal and mothers' warm-supportive versus negative-punitive approaches to socialization.


    Thanksgiving Holiday 
    No brown bag this week. Happy Thanksgiving and safe travels.


    Finals Week 
    No brown bag this week. Good luck with finals, whether taking or grading them.
    #BlackLivesMatter when Explaining Disparities in Police Contact 
    SPEAKER: Melissa McTernan

    In honor of Martin Luther King, Jr. Day this week, and of the national conversations regarding race and police contact since the events of Ferguson, MO, this presentation will explore racial disparities in the context of community-police interactions. While most researchers agree that these disparities can not entirely be explained by criminal behavior, much research still needs to be done to explore the complexities of police interaction with the public and what role race plays during these interactions. I will review existing literature on racial disparities in police contact, and present results from my own analysis of the Police-Public Contact Survey (PPCS) of 2008.
    Setting the record straight: Models for nonlinearity and for nonnormality 
    SPEAKER: Melissa McTernan

    Standard OLS regression is a familiar and straightforward method for studying relations between constructs. However, in psychology research we are often interested in data for which OLS regression is not appropriate. In this talk I focus on the problems of nonlinearity of the parameters and nonnormality of the residuals. OLS regression is not sufficient for data with either of these characteristics, and therefore other methods must be employed. Moving away from a familiar method and making a decision about what alternative method to use can be difficult in practice because there are many methods to choose from and the literature on those methods can be daunting for an unfamiliar reader. A goal of this talk is to describe a few alternative methods to OLS and when they are appropriate to apply in practice. Specifically, I will be reviewing linear regression with OLS, nonlinear regression, generalized linear modeling, generalized nonlinear modeling, and direct likelihood optimization approaches.
    Absence Makes the Heart Grow Fonder: An Introduction to Planned Missing Data Designs 
    SPEAKER: Nathan Smith

    Missing data is ubiquitous in psychological research. Powerful methods and software have been developed to address this problem and are successful at recovering parameter estimates without bias when modeling from incomplete data sets. These methods are most successful when the data are missing at random (MAR) or missing completely and random (MCAR). In planned missing data designs (PPMD) participants are intentionally assigned to conditions where they will not respond to all items/measures or measurement occasions. This allows experimenters to leverage resources, reduce respondent fatigue, and potentially improve data quality by decreasing unplanned missingness. This talk will provide a general overview of applications of planned missing data designs and address two well researched PMDD’s, the 3-Form Design and 2 Method measurement design.
    Small Sample Inference in Linear Mixed Models 
    SPEAKER: Timothy Banh

    Small samples are frequently encountered in psychological research. In this talk, I will discuss the implications of small sample size in regards to linear mixed models. When using SAS PROC MIXED, there are different denominator degrees of freedom options available. I will discuss the motivation for including these degrees of freedom options and talk about one particular method, the Kenward-Roger method. This method is a small sample correction that leads to an F distribution for the null distribution of the fixed effects. The motivation behind the Kenward-Roger method and simulations exploring its behavior will be discussed.
    No Meeting 

    Organizational Meeting 
    Fundamentals of Item Response Theory and its Applications 
    SPEAKER: Dr. Tim Gaffney
    California State University, Sacramento
    SPEAKER: Matthew Miller
    SPEAKER: Dr. Joel Steele Portland State University