scholarly journals A Comparison of Methods for Estimating Relationships in the Change Between Two Time Points for Latent Variables

2016 ◽  
Vol 78 (2) ◽  
pp. 232-252 ◽  
Author(s):  
W. Holmes Finch ◽  
Sungok Serena Shim
Keyword(s):  
Simulation Study ◽  
Latent Variables ◽  
Factor Model ◽  
Latent Trait ◽  
Factor Loading ◽  
Change Score ◽  
Standard Errors ◽  
Time Points ◽  
Complex Relationships ◽  
Latent Traits

Collection and analysis of longitudinal data is an important tool in understanding growth and development over time in a whole range of human endeavors. Ideally, researchers working in the longitudinal framework are able to collect data at more than two points in time, as this will provide them with the potential for a deeper understanding of the development processes under study and a much broader array of statistical modeling options. However, in some circumstances data collection is limited to only two time points, perhaps because of resource limitations, issues with the context in which the data are collected, or the nature of the trait under study. In such instances, researchers may still want to learn about complex relationships in the data, such as the correlation between changes in latent traits that are being measured. However, with only two data points, standard approaches for modeling such relationships, such as growth curve modeling, cannot be used. The current simulation study compares the performance of two methods for estimating the correlations among changes in latent variables between two points in time, the two-wave latent change score model and the latent difference factor model. Results of the simulation study showed that both methods yielded generally accurate estimates of the correlation between changes in a latent trait, with relatively small standard errors. Estimation bias and standard errors were lower with larger samples, larger factor loading magnitudes, and more indicators per factor. Further comparisons between the methods and implications of these results are discussed.

scholarly journals Can Variances of Latent Variables be Scaled in Such a Way That They Correspond to Eigenvalues?

2017 ◽  
Vol 6 (6) ◽  
pp. 35 ◽  
Author(s):  
Karl Schweizer ◽  
Stefan Troche ◽  
Siegbert Reiß
Keyword(s):  
Factor Analysis ◽  
Confirmatory Factor Analysis ◽  
Simulation Study ◽  
Latent Variables ◽  
Latent Variable ◽  
Factor Model ◽  
Estimation Method ◽  
Factor Loadings ◽  
Confirmatory Factor ◽  
Theoretical Foundations

The paper reports an investigation of whether sums of squared factor loadings obtained in confirmatory factor analysis correspond to eigenvalues of exploratory factor analysis. The sum of squared factor loadings reflects the variance of the corresponding latent variable if the variance parameter of the confirmatory factor model is set equal to one. Hence, the computation of the sum implies a specific type of scaling of the variance. While the investigation of the theoretical foundations suggested the expected correspondence between sums of squared factor loadings and eigenvalues, the necessity of procedural specifications in the application, as for example the estimation method, revealed external influences on the outcome. A simulation study was conducted that demonstrated the possibility of exact correspondence if the same estimation method was applied. However, in the majority of realized specifications the estimates showed similar sizes but no correspondence. 

scholarly journals Effect of Dichotomization on the Latent Structure of Data

2021 ◽  
Vol 7 ◽  
Author(s):  
Karl Schweizer ◽  
Andreas Gold ◽  
Dorothea Krampen
Keyword(s):  
Simulation Study ◽  
Latent Variables ◽  
Latent Variable ◽  
Factor Model ◽  
Latent Structure ◽  
Factor Loadings ◽  
Dichotomous Data ◽  
Level Data ◽  
Confirmatory Factor

We investigated whether dichotomous data showed the same latent structure as the interval-level data from which they originated. Given constancy of dimensionality and factor loadings reflecting the latent structure of data, the focus was on the variance of the latent variable of a confirmatory factor model. This variance was shown to summarize the information provided by the factor loadings. The results of a simulation study did not reveal exact correspondence of the variances of the latent variables derived from interval-level and dichotomous data but shrinkage. Since shrinkage occurred systematically, methods for recovering the original variance were fleshed out and evaluated.

scholarly journals Dispositional individual differences in cognitive effort investment: establishing the core construct

2021 ◽  
Vol 9 (1) ◽  
Author(s):  
Corinna Kührt ◽  
Sebastian Pannasch ◽  
Stefan J. Kiebel ◽  
Alexander Strobel
Keyword(s):  
Individual Differences ◽  
Latent Variables ◽  
Effortful Control ◽  
Factor Model ◽  
Cognitive Effort ◽  
Self Control ◽  
Behavioral Measures ◽  
Effort Discounting ◽  
Cognitive Motivation ◽  
Goal Directed Behavior

Abstract Background Individuals tend to avoid effortful tasks, regardless of whether they are physical or mental in nature. Recent experimental evidence is suggestive of individual differences in the dispositional willingness to invest cognitive effort in goal-directed behavior. The traits need for cognition (NFC) and self-control are related to behavioral measures of cognitive effort discounting and demand avoidance, respectively. Given that these traits are only moderately related, the question arises whether they reflect a common core factor underlying cognitive effort investment. If so, the common core of both traits might be related to behavioral measures of effort discounting in a more systematic fashion. To address this question, we aimed at specifying a core construct of cognitive effort investment that reflects dispositional differences in the willingness and tendency to exert effortful control. Methods We conducted two studies (N = 613 and N = 244) with questionnaires related to cognitive motivation and effort investment including assessment of NFC, intellect, self-control and effortful control. We first calculated Pearson correlations followed by two mediation models regarding intellect and its separate aspects, seek and conquer, as mediators. Next, we performed a confirmatory factor analysis of a hierarchical model of cognitive effort investment as second-order latent variable. First-order latent variables were cognitive motivation reflecting NFC and intellect, and effortful self-control reflecting self-control and effortful control. Finally, we calculated Pearson correlations between factor scores of the latent variables and general self-efficacy as well as traits of the Five Factor Model of Personality for validation purposes. Results Our findings support the hypothesized correlations between the assessed traits, where the relationship of NFC and self-control is specifically mediated via goal-directedness. We established and replicated a hierarchical factor model of cognitive motivation and effortful self-control that explains the shared variance of the first-order factors by a second-order factor of cognitive effort investment. Conclusions Taken together, our results integrate disparate literatures on cognitive motivation and self-control and provide a basis for further experimental research on the role of dispositional individual differences in goal-directed behavior and cost–benefit-models.

scholarly journals A New Perspective on the Multidimensionality of Divergent Thinking Tasks

2018 ◽  
Author(s):  
Boris Forthmann ◽  
Paul - Christian Bürkner ◽  
Mathias Benedek ◽  
Carsten Szardenings ◽  
Heinz Holling
Keyword(s):  
Latent Variables ◽  
Divergent Thinking ◽  
Rating Scale ◽  
Latent Trait ◽  
Subjective Ratings ◽  
Task Instructions ◽  
New Perspective ◽  
Psychometric Quality

In the presented work, a shift of perspective with respect to the dimensionality of divergent thinking tasks is introduced moving from the question of multidimensionality across divergent thinking scores to the question of multidimensionality across the scale of divergent thinking scores. We apply IRTree models to test if the same latent trait can be assumed can be assumed across the whole scale in snapshot scoring of divergent thinking tests and holds for different task instructions and varying levels of fluency. This way, multidimensionality can be explored across scale points of a Likert-type rating scale and also multidimensionality due to differences in number of responses of ideational pools can be assessed. It was found that evidence for unidimensionality across scale points was stronger with be-creative instructions as compared to be-fluent instructions which suggests better psychometric quality of ratings when be-creative instructions are used. In addition, latent variables pertaining to low-fluency and high-fluency ideational pools shared around 50% of variance which suggests both strong overlap and evidence for differentiation. The presented approach allows to further examine the psychometric quality of subjective ratings and to examine new questions with respect to within-item multidimensionality in divergent thinking.

Item Response Theory Analysis to Assess Dimensionality of Substance Use Disorder Abuse and Dependence Symptoms

2017 ◽  
Vol 6 (4) ◽  
Author(s):  
Levent Kirisci ◽  
Ralph Tarter ◽  
Maureen Reynolds ◽  
Michael Vanyukov
Keyword(s):  
Substance Use ◽  
Item Response Theory ◽  
Item Response ◽  
Substance Use Disorder ◽  
Factor Model ◽  
Common Factor ◽  
Latent Trait ◽  
Dominant Factor ◽  
Response Theory ◽  
Men And Women

Background. Item response theory (IRT) based studies conducted on diverse samples showed a single dominant factor for DSM-III-R and DSM-IV substance use disorder (SUD) abuse and dependence symptoms of alcohol, cannabis, sedative, cocaine, stimulants, and opiates use disorders. IRT provides the opportunity, within a person-centered framework, to accurately gauge each person’s severity of disorder that, in turn, informs required intensiveness of treatment. Objectives. The aim of this study was to determine whether the SUD symptoms indicate a unidimensional trait or instead need to be conceptualized and quantified as a multidimensional scale. Methods. The sample was composed of families of adult SUD+ men (n=349), and SUD+ women (n=173), who qualified for DSM-III-R diagnosis of substance use disorder (abuse or dependence) and families of adult men and women who did not qualify for a SUD diagnosis (SUD- men: n=190, SUD- women: n=133). An expanded version of the Structured Clinical Interview for DSM-III-R (SCID) was administered to characterize lifetime and current substance use disorders. Item response theory methodology was used to assess the dimensionality of DSM-III-R SUD abuse and dependence symptoms.Results. A bi-factor model provided the optimal representation of the factor structure of SUD symptoms in males and females. SUD symptoms are scalable as indicators of a single common factor, corresponding to general (non-drug-specific, common) liability to addiction, combined with drug-specific liabilities. Conclusions. IRT methodology used to quantify the continuous general liability to addiction (GLA) latent trait in individuals having SUD symptoms was found effective for accurately measuring SUD severity in men and women. This may be helpful for person-centered medicine approaches to effectively address intensity of treatment.

scholarly journals Longitudinal Validation of the Levenson Self-Report Psychopathy (LSRP) Scale in a High-Risk Dutch Community Sample

Assessment ◽  
2020 ◽  
pp. 107319112097513
Author(s):  
Sophie A. Wissenburg ◽  
Carlo Garofalo ◽  
Arjan A. J. Blokland ◽  
H. Palmen ◽  
Martin Sellbom
Keyword(s):  
Construct Validity ◽  
Factor Structure ◽  
Factor Model ◽  
Psychopathic Traits ◽  
Self Report ◽  
Longitudinal Measurement Invariance ◽  
Time Points ◽  
Longitudinal Measurement ◽  
Three Factor ◽  
Over Time

The Levenson Self-Report Psychopathy (LSRP) scale is a self-report measure that can be used to assess psychopathic traits in community samples, and recent research suggested that its three-factor model (Egocentricity, Callousness, and Antisocial) has promising psychometric properties. However, no study to date has validated the LSRP in a longitudinal framework. The present study sought to validate the LSRP scale in a longitudinal design using a sample of Dutch emerging adults ( ns = 970 and 693 at time points 1 and 2, respectively). We assessed longitudinal measurement invariance and the stability of psychopathic traits over an 18-month time period, from age 20 to age 21.6. Furthermore, we replicated and extended findings on the factor structure, reliability, and construct validity of the Dutch LSRP scale. Confirmatory factor analysis revealed that the three-factor model fit the data well. Evidence of partial longitudinal measurement invariance was observed, which means that the Dutch translation of the LSRP scale is measuring an equivalent construct (and overall latent factor structure) over time. Psychopathic traits were relatively stable over time. The three LSRP subscales showed largely acceptable levels of internal consistency at both time points and showed conceptually expected patterns of construct validity and predictive validity, with a few notable exceptions.

scholarly journals Precision and Sample Size Requirements for Regression-Based Norming Methods for Change Scores

Assessment ◽  
2020 ◽  
pp. 107319112091360
Author(s):  
Zhengguo Gu ◽  
Wilco H. M. Emons ◽  
Klaas Sijtsma
Keyword(s):  
Sample Size ◽  
Simulation Study ◽  
Change Score ◽  
Minimum Sample Size ◽  
Change Scores ◽  
Satisfactory Precision ◽  
Similarities And Differences ◽  
Minimum Sample

To interpret a person’s change score, one typically transforms the change score into, for example, a percentile, so that one knows a person’s location in a distribution of change scores. Transformed scores are referred to as norms and the construction of norms is referred to as norming. Two often-used norming methods for change scores are the regression-based change approach and the T Scores for Change method. In this article, we discuss the similarities and differences between these norming methods, and use a simulation study to systematically examine the precision of the two methods and to establish the minimum sample size requirements for satisfactory precision.

Evaluating a Brief Measure of Social Self-Efficacy among U.S. Adolescents

2011 ◽  
Vol 109 (3) ◽  
pp. 907-920 ◽  
Author(s):  
Keith J. Zullig ◽  
Daniel A. Teoli ◽  
Robert F. Valois
Keyword(s):  
Factor Analysis ◽  
Self Efficacy ◽  
Factor Model ◽  
Factor Loading ◽  
Asian Students ◽  
Public High School ◽  
Social Self ◽  
Efficacy Measure ◽  
Principal Axis Factor Analysis ◽  
Being Bullied

Preliminary data were collected to evaluate the performance of a social self-efficacy measure among 4,061 public high school adolescents. Principal-axis factor analysis was followed by a 4-way between-groups analysis of variance (ANOVA) to test for differences in the Total score means on selected demographic estimates and their interactions. Relations between the Total score and selected risk behaviors were examined through a series of one-way ANOVAs and the Tukey HSD test. Factor analysis results suggested a one-factor model best explained the factor structure of the scale items (factor loading range = .64–.77, eigenvalue = 4.05, h2 = .51). Females reported a significantly higher mean Total social self-efficacy rating than males, while White students reported a significantly higher mean Total social self-efficacy rating than Black and Asian students. Statistically significant lower mean Total social self-efficacy ratings were also noted for those who reported physical fighting, avoiding school, and being bullied.

Sign in / Sign up

Export Citation Format

Share Document