This work replicates previous research used to criticize the use of principle component (PCA) and exploratory factor analysis (EFA). Factor analysis is often used when research is being conducted in a new area and there is limited background knowledge about what one might expect to find. It’s useful because of its ability to reduce the number of variables in a dataset by examining the correlations between items and identifying latent dimensions that account for measured item variance and covariation. A criticism of this method is that it detects patterns in random data where there is no underlying structure. We tested this criticism by replicating the methods in a 1981 paper by Robert Vierra and David Carlson. We generated random data in R Studio and looked at the variance accounted for by the factors identified by EFA and PCA. We also examined the Type I error rate in sample correlation matrices tested with Bartlett's test. Additionally, we have begun a literature review looking at the modern methods similar to exploratory factor analysis, including machine learning, in order to understand how these criticisms have been addressed by these modern approaches.

Measurement in Psychology usually assumes continuous dimensions. Correspondingly, theoretical constructs are typically conceptualized as dimensions, meaning that observations differ in degree. However, theoretical constructs may also be categorical, i.e., rather than differing by degree, differing qualitatively. Different construct operationalization’s may lead to different scientific conclusions. Self esteem has long been of interest in Psychology. It has been conceptualized in various different ways, but always dimensional. Some self-esteem measures (for example, the The Rosenberg Self Esteem scale) seek to measure globally how much one values themselves as a person, without focusing on specific domains. On the other hand, other self-esteem measures (for example, the State self esteem scale) focus on perceived capabilities in specific domains, such as performance, social, and appearance. In the National Survey of Youth, participants rated themselves in specific areas, and then again as they would like to be. Using these data, we examined how different operationalization’s of self-esteem are associated with a few psychologically interesting outcomes such as drug use and crime. Specifically, we compared continuous and categorical representations of self-esteem. By examining the assumption of continuous latent variables, we shed light on how scientific conclusions differ based on construct operationalization.
 

Latent class models are frequently used to classify observations into homogeneous albeit qualitatively different sub-groups. Typically, these models are unrestricted, meaning there is no a priori structure or restrictions on model estimates. This approach is similar to many cluster analysis approaches. The study examines modeling conclusions when an unrestricted model is applied to data generated from a highly restrictive population model through a simulation study. We expect that a researcher will conclude fewer classes than the true number specified in the population. Simulated data for this work is based on a substance use example identified in other research. The population was generated based on six classes, but with a within class error rate fixed to 5%. Preliminary analyses indicate that commonly employed model selection criteria indicate that models with fewer than six classes are preferable. This is important because most analysts using latent class models use unrestricted models with little apparent consideration to what realistic error rates may be. In research domains where smaller error rates may be plausible, restrictive models of the sort used to generate the data for this study may be worth considering.

Research has shown that positive outcome expectancy (i.e., anticipated positive consequences as a result of engaging in a behavior) and positive affective responses to exercise play an important role in predicting exercise behavior as well as future exercise intentions. Unlike most research in which the valence (i.e., positive - negative) of outcome expectancy and affective responses are typically assessed through survey items whose valence is predetermined, the current study used a spontaneous word association test (WAT) to examine the subjective meaning of exercise-related outcome expectancy and mood before exercise, and how they are related to exercise intentions and behaviors. One-hundred and eleven undergraduate and graduate students (74 females; 64 undergraduate students) performed an exercise-related WAT in which they were asked to write down the first word or a short phrase that comes to mind when they thought of (a) things that might happen from exercising and (b) their mood right before exercising. Subsequently, they were instructed to rate their subjective valence (i.e., positive-negative) to each of their responses to the WAT. They also completed a series of online self-report surveys that included assessments of exercise intentions and past exercise behavior. Path analyses will be conducted to test our hypotheses that the subjective valence of their mood right before exercising is expected to be directly associated with exercise behavior whereas the subjective valence of outcome expectancy is expected to be indirectly associated with exercise behavior through exercise intentions. This study examines the potential value and usefulness of a spontaneous word association procedure to more accurately measure affective evaluations of an individual’s feelings and thoughts toward exercise behavior.

Alzheimer’s disease (AD) is a neurodegenerative illness affecting millions of individuals in the United States and represents the leading cause of dementia worldwide. AD is pathologically defined by abnormal accumulation of interneuronal plaques composed of amyloid beta protein and intraneuronal tangles composed of tau protein. The focus of our current research involves understanding how tau protein homeostasis is restored using the model organism C. elegans. One process by which protein homeostasis is modulated is through the endoplasmic reticulum unfolded protein response (UPRER), which is composed of three signaling branches. Initially, the UPRER acts to restore normal protein folding, but if restoration is impossible, it triggers apoptosis. Our lab became interested in this signaling pathway after we identified the UPRER master transcription factor XBP-1s as a modifier of tau pathology (tauopathy). Specifically, XBP-1s overexpression protects against tauopathy in our C. elegans model. Previous RNA sequencing (RNAseq) analysis identified several downstream target genes upregulated with XBP-1s overexpression in our pathological tau C. elegans model. To follow up, I am working on understanding the molecular basis of XBP-1s-mediated tauopathy suppression by generating C. elegans models containing target genes of interest. Assessment of the effect on tauopathy phenotypes includes behavioral analysis by performing a liquid thrashing assay and tau protein analysis via the Western blot technique. Using the microinjection technique, I have generated a transgenic C. elegans model overexpressing hsp-4, an upregulated XBP-1s target gene identified by RNAseq that potentially modulates tauopathy. We are currently determining whether overexpression of hsp-4 can ameliorate tauopathy similar to XBP-1s overexpression. Gaining further information about the role of these XBP-1s target genes in tauopathy using the model organism C. elegans is valuable for future translational studies in mammals, along with being integral to research for neuroprotective therapies for neurodegenerative diseases such as AD.

Race is a social construct that was originally developed to differentiate between superior and inferior human beings. Race has been primarily studied from a monoracial perspective, which prohibits multiracial individuals to express themselves because their racial identity does not fit into the neatly, pre-established monoracial categories. Multiracial is used as an umbrella term to describe an individual whose identity is comprised of multiple races or ethnicities (multi-ethnic). Previous research has found that multiracial students experience life in ways that are unique compared to their monoracial peers. Multiracial students are often required to conform to socially constructed monoracial boxes, and often have negative experiences from microaggressions, monoracial family member comments, and lack of demographic categorization. Additional studies have found that the denial of multiracial identities have the potential to cause lower levels of self-esteem along with decreased motivation. The multiracial population in America is steadily growing and, if current trends continue, will only continue to do so. As a result, an increased number of multiracial students on college campuses are expected. This paper explores multiracial student identities within higher education at a public university in the Pacific Northwest and aims to learn about racial identity beyond a monoracial paradigm. The knowledge gained from this research will contribute to higher education administration, faculty, and staff to inform them of the ways in which they can validate multiracial students’ identities on their college campuses.