Avoiding and Correcting Bias in Score-Based Latent Variable Regression With Discrete Manifest Items

Lu, Irene R. R.; Thomas, David R.

doi:10.1080/10705510802154323

Cited by 26 publications

(23 citation statements)

References 32 publications

(52 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multistage factor score regression has been historically been more common (e.g., Bollen & Lennox, 1991;Lu & Thomas, 2008;Skrondal & Laake, 2001) and continues to be recommended as a practical approach (e.g., Hayes & Usami, 2020a;Hoshino & Bentler, 2013). In factor score regression, factor scores from a measurement model are created for each construct separately and saved in one step.…”

Section: Using Scores In Subsequent Analysesmentioning

confidence: 99%

Thinking twice about sum scores

2020

View full text Add to dashboard Cite

A common way to form scores from multiple-item scales is to sum responses of all items. Though sum scoring is often contrasted with factor analysis as a competing method, we review how factor analysis and sum scoring both fall under the larger umbrella of latent variable models, with sum scoring being a constrained version of a factor analysis. Despite similarities, reporting of psychometric properties for sum scored or factor analyzed scales are quite different. Further, if researchers use factor analysis to validate a scale but subsequently sum score the scale, this employs a model that differs from validation model. By framing sum scoring within a latent variable framework, our goal is to raise awareness that (a) sum scoring requires rather strict constraints, (b) imposing these constraints requires the same type of justification as any other latent variable model, and (c) sum scoring corresponds to a statistical model and is not a modelfree arithmetic calculation. We discuss how unjustified sum scoring can have adverse effects on validity, reliability, and qualitative classification from sum score cutoffs. We also discuss considerations for how to use scale scores in subsequent analyses and how these choices can alter conclusions. The general goal is to encourage researchers to more critically evaluate how they obtain, justify, and use multiple-item scale scores.

show abstract

Section: Using Scores In Subsequent Analysesmentioning

confidence: 99%

Thinking twice about sum scores

2020

View full text Add to dashboard Cite

show abstract

“…Specifically, it is known that factor score regression estimators are generally biased (see, e.g. Croon, 2002;Hoshino & Bentler, 2013;Lu & Thomas, 2008;Skrondal & Laake, 2001). In this paper, the recently proposed factor regression method by Hoshino and Bentler (2013) is followed.…”

Section: Discussionmentioning

confidence: 98%

Testing two path models to explore relationships between students’ experiences of the teaching–learning environment, approaches to learning and academic achievement

Milienos

2014

Educational Psychology

View full text Add to dashboard Cite

The study explores the relationships between students' experiences of the teaching-learning environment and their approaches to learning, and the effects of these variables on academic achievement. Two three-stage models were tested with structural equation modelling techniques. The Approaches and Study Skills Inventory for Students (ASSIST) and the Experiences of Teaching and Learning Questionnaire (ETLQ) were used to assess approaches to learning and student's experiences of the teaching-learning environment, respectively. These two constructs were then used as either first-or second-stage variables within the path analysis. The model using approaches to learning as a mediating variable showed the best fit with our data; variations in our students' experiences of the teaching-learning environment appear to give rise to their approaches to studying, which subsequently affect their achievement. The deep approach shows no detectable influence on academic achievement in this sample, neither there are any direct effects of experiences of the teaching-learning environment on it. The indirect effects of these experiences on achievement, acting through the strategic and the surface approaches, are related to two aspects of the teaching-learning environment only, namely congruence and coherence in course organisation, and integrative learning and critical thinking. The finding of a reciprocal relation between approaches to learning and experiences of the teaching-learning environment supports previous conclusions about the association between these constructs. The indirect effect of experiences of the teaching-learning environment on achievement, acting through approaches to learning, shows those approaches as a dynamic construct that varies in line with experiences of the teaching-learning environment, and so influences achievement.

show abstract

“…Indeterminacy becomes an even more important concern when the factor scores are subsequently used in an OLS regression to obtain treatment effect estimates. Factor scores contain a degree of uncertainty that is not accounted for in the second-stage OLS regression and causes the regression coefficient estimate to be biased (Skrondal and Laake, 2001;Croon, 2002;Bolck, Croon, and Hagenaars, 2004;Lu and Thomas, 2008;Devlieger, Mayer, and Rosseel, 2016). Devlieger, Mayer, and Rosseel (2016) show that in a model with latent independent variables and an observed dependent variable, the Regression method of computing factor scores yields an unbiased estimate.…”

Section: *mentioning

confidence: 99%

The Gendered Effects of Career Concerns on Fertility

Park,

Rim

2017

SSRN Journal

View full text Add to dashboard Cite

A growing literature reveals that the adverse effect of children on career advancement falls disproportionately on women. This raises the possibility that women respond to career concerns by delaying family formation more than men. Using a panel dataset on lawyers, we find females are less likely to have their first-child before the promotion decision. This fertility gap is not explained away by gender-based sorting or gender differences in marriage-timing and spousal occupation. Two channels drive our results: women bear child-rearing costs and gender-specific promotion thresholds. This implies the focus on the gender wage gap understates gender inequality in the labor market. JEL Codes: J13, J71 * We are grateful to Dan Black, Kerwin Charles, and Jens Ludwig for their guidance and help throughout this project. We would also like to thank Tom Coleman, Bob Michael, and seminar participants at Chicago Harris,

show abstract

Avoiding and Correcting Bias in Score-Based Latent Variable Regression With Discrete Manifest Items

Cited by 26 publications

References 32 publications

Thinking twice about sum scores

Thinking twice about sum scores

Testing two path models to explore relationships between students’ experiences of the teaching–learning environment, approaches to learning and academic achievement

The Gendered Effects of Career Concerns on Fertility

Contact Info

Product

Resources

About