Approach to analysing correlated contextual factors: an application for studies on violence

2021 ◽  
pp. injuryprev-2020-043967
Author(s):  
Marizen R Ramirez ◽  
Javier E Flores ◽  
Gang Cheng ◽  
Corinne Peek-Asa ◽  
Joseph E Cavanaugh
Keyword(s):  
Academic Performance ◽  
School Violence ◽  
Principal Components ◽  
Contextual Factors ◽  
Epidemiological Studies ◽  
Standard Errors ◽  
Parameter Estimates ◽  
School Crime ◽  
School Factors ◽  
Health Studies

BackgroundNumerous public health studies, especially in the area of violence, examine the effects of contextual or group-level factors on health outcomes. Often, these contextual factors exhibit strong pairwise correlations, which pose a challenge when these factors are included as covariates in a statistical model. Such models may be characterised by inflated standard errors and unstable parameter estimates that may fluctuate drastically from sample to sample, where the excessive estimation variability is reflected by inflated standard errors.MethodsWe propose a three-stage approach for analysing correlated contextual factors that proceeds as follows: (1) a principal components analysis (PCA) is performed on the original set of correlated variables, (2) the primary generated principal components are included in a multilevel multivariable model and (3) the estimated parameters for these components are transformed into estimates for each of the original contextual factors. Using school violence data, we examined the associations between school crime and correlated contextual school factors (ie, English proficiency, academic performance, pupil to teacher ratio, average class size and children on free and reduced meals).ResultsFrom models ignoring correlations, school crime was not reliably associated with any of the contextual school factors. When models were fit with principal components, school crime was found to be positively associated with a school’s student to teacher ratio, average classroom size and academic performance but negatively associated with the proportion of children who were on free and reduced meals.ConclusionOur multistep approach is one way to address multicollinearity encountered in social epidemiological studies of violence.

scholarly journals The Impact of Work Hours on Depressive Symptoms among Koreans Aged 45 and Over

2021 ◽  
Vol 18 (3) ◽  
pp. 853
Author(s):  
Juyeong Kim ◽  
Eun-Cheol Park
Keyword(s):  
Older Adults ◽  
Depressive Symptoms ◽  
Social Services ◽  
Depression Scale ◽  
Epidemiological Studies ◽  
Work Hours ◽  
Middle Aged ◽  
Health Studies ◽  
Males And Females ◽  
The Impact

Background: Given the documented importance of employment for middle-aged and older adults’ mental health, studies of the association between their number of work hours and depressive symptoms are needed. Objectives: To examine the association between the number of work hours and depressive symptoms in Korean aged 45 and over. Methods: We used data from the first wave to fourth wave of the Korea Longitudinal Study of Aging. Using the first wave at baseline, data included 9845 individuals. Depressive symptoms were measured using the 10-item Center for Epidemiological Studies Depression scale. We performed a longitudinal analysis to estimate the prevalence of depressive symptoms by work hours. Results: Both unemployed males and females aged 45–65 years were associated with higher depressive symptoms (β = 0.59, p < 0.001; β = 0.32, p < 0.001). Females working ≥ 69 h were associated with higher depressive symptoms compared to those working 41–68 h (β = 0.25, p = 0.013). Among those both middle-aged and older adults, both males and females unemployed were associated with higher depressive symptoms. Those middle-aged female working ≥69 h were associated with higher depressive symptoms. Conclusions: An increase in depressive symptoms was associated with unemployed males and females working ≥69 h compared to those working 41–68 h. Although this association was found among middle-aged individuals, a decrease in depressive symptoms in both sexes was associated with working 1–40 h. Depressive symptoms should decrease by implementing employment policies and social services to encourage employers to support middle-aged and older adults in the workforce considering their sex and age differences.

The Four-States Model of Memory Retrieval Experiences

2007 ◽  
Vol 215 (1) ◽  
pp. 61-71 ◽  
Author(s):  
Edgar Erdfelder ◽  
Lutz Cüpper ◽  
Tina-Sarah Auer ◽  
Monika Undorf
Keyword(s):  
Signal Detection ◽  
Memory Retrieval ◽  
Measurement Model ◽  
Standard Errors ◽  
Parameter Estimates ◽  
Unequal Variances ◽  
One Dimensional ◽  
Detection Model ◽  
Memory States ◽  
The One

Abstract. A memory measurement model is presented that accounts for judgments of remembering, knowing, and guessing in old-new recognition tasks by assuming four disjoint latent memory states: recollection, familiarity, uncertainty, and rejection. This four-states model can be applied to both Tulving's (1985) remember-know procedure (RK version) and Gardiner and coworker's ( Gardiner, Java, & Richardson-Klavehn, 1996 ; Gardiner, Richardson-Klavehn, & Ramponi, 1997 ) remember-know-guess procedure (RKG version). It is shown that the RK version of the model fits remember-know data approximately as well as the one-dimensional signal detection model does. In contrast, the RKG version of the four-states model outperforms the corresponding detection model even if unequal variances for old and new items are allowed for.We show empirically that the two versions of the four-statesmodelmeasure the same state probabilities. However, the RKG version, requiring remember-know-guess judgments, provides parameter estimates with smaller standard errors and is therefore recommended for routine use.

scholarly journals Incidencia de factores personales y contextuales sobre el uso de los recursos tecnológicos por el alumnado en América Latina

2021 ◽  
Vol 29 ◽  
pp. 6 ◽  
Author(s):  
Sara Cebrián-Cifuentes ◽  
Gonzalo Almerich ◽  
Jesús Suárez-Rodríguez ◽  
Francesc Pedró
Keyword(s):  
Principal Components Analysis ◽  
Information And Communication Technologies ◽  
Principal Components ◽  
Contextual Factors ◽  
Sixth Grade ◽  
Descriptive Statistics ◽  
Ict Use ◽  
Components Analysis ◽  
Categorical Principal Components Analysis ◽  
America Latina

The use of information and communication technologies (ICT) by students reflects the appropriateness of the ICT integration process. However, the typology of ICT use by students has not been established empirically based on their use at home and at the school. Thus, the purpose of the article is to determine the structure of ICT use by students, together with the influence of personal and contextual factors. A correlational design has been used, with the sample being the sixth-grade students in Latin America who answered the questionnaire on ICT use in the Third Regional Comparative and Explanatory Study (TERCE). The data analysis performed is descriptive statistics, Multivariate Analysis of Variance (MANOVA) and Categorical Principal Components analysis (CATPCA). The results obtained through the descriptive statistics show how the students make a greater use of technological resources in the area of free time than in the academic area. It has been found, by means of a categorical principal components analysis (CATPCA), that student use is structured in three planes: personal, non-school academic and school academic. In addition, the results obtained from the MANOVA indicate that the personal and contextual factors influence the use of ICT, essentially the availability of technological devices and the Internet connection. In the light of the results, it is recommended to implement programs that favour non-school academic use of ICT by students.

scholarly journals Bias and Precision of Parameter Estimates from Models Using Polygenic Scores to Estimate Environmental and Genetic Parental Influences

2020 ◽  
Author(s):  
Yongkang Kim ◽  
Jared V. Balbona ◽  
Matthew C. Keller
Keyword(s):  
Assortative Mating ◽  
Vertical Transmission ◽  
Companion Paper ◽  
Genetic Effects ◽  
Causal Models ◽  
Parental Influences ◽  
Standard Errors ◽  
Parameter Estimates ◽  
Polygenic Score ◽  
Polygenic Scores

AbstractIn a companion paper Balbona et al. (Behav Genet, in press), we introduced a series of causal models that use polygenic scores from transmitted and nontransmitted alleles, the offspring trait, and parental traits to estimate the variation due to the environmental influences the parental trait has on the offspring trait (vertical transmission) as well as additive genetic effects. These models also estimate and account for the gene-gene and gene-environment covariation that arises from assortative mating and vertical transmission respectively. In the current study, we simulated polygenic scores and phenotypes of parents and offspring under genetic and vertical transmission scenarios, assuming two types of assortative mating. We instantiated the models from our companion paper in the OpenMx software, and compared the true values of parameters to maximum likelihood estimates from models fitted on the simulated data to quantify the bias and precision of estimates. We show that parameter estimates from these models are unbiased when assumptions are met, but as expected, they are biased to the degree that assumptions are unmet. Standard errors of the estimated variances due to vertical transmission and to genetic effects decrease with increasing sample sizes and with increasing $$r^2$$ r 2 values of the polygenic score. Even when the polygenic score explains a modest amount of trait variation ($$r^2=.05$$ r 2 = . 05 ), standard errors of these standardized estimates are reasonable ($$< .05$$ < . 05 ) for $$n=16K$$ n = 16 K trios, and can even be reasonable for smaller sample sizes (e.g., down to 4K) when the polygenic score is more predictive. These causal models offer a novel approach for understanding how parents influence their offspring, but their use requires polygenic scores on relevant traits that are modestly predictive (e.g., $$r^2>.025)$$ r 2 > . 025 ) as well as datasets with genomic and phenotypic information on parents and offspring. The utility of polygenic scores for elucidating parental influences should thus serve as additional motivation for large genomic biobanks to perform GWAS’s on traits that may be relevant to parenting and to oversample close relatives, particularly parents and offspring.

Some Results on the Behavior of Alternate Covariance Structure Estimation Procedures in the Presence of Non-Normal Data

1989 ◽  
Vol 26 (2) ◽  
pp. 214-221 ◽  
Author(s):  
Subhash Sharma ◽  
Srinivas Durvasula ◽  
William R. Dillon
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Covariance Structure ◽  
Standard Errors ◽  
Parameter Estimates ◽  
Simulation Experiments ◽  
Estimation Techniques ◽  
Structure Estimation ◽  
Estimation Procedures ◽  
Simulation Results

The authors report some results on the behavior of alternative covariance structure estimation procedures in the presence of non-normal data. They conducted Monté Carlo simulation experiments with a factorial design involving three levels of skewness, three level of kurtosis, and three different sample sizes. For normal data, among all the elliptical estimation techniques, elliptical reweighted least squares (ERLS) was equivalent in performance to ML. However, as expected, for non-normal data parameter estimates were unbiased for ML and the elliptical estimation techniques, whereas the bias in standard errors was substantial for GLS and ML. Among elliptical estimation techniques, ERLS was superior in performance. On the basis of the simulation results, the authors recommend that researchers use ERLS for both normal and non-normal data.

Who Succeeds at Dental School? Factors Predicting Students’ Academic Performance in a Dental School in Republic of Korea

2013 ◽  
Vol 77 (12) ◽  
pp. 1616-1623 ◽  
Author(s):  
Jung-Joon Ihm ◽  
Gene Lee ◽  
Kack-Kyun Kim ◽  
Ki-Taeg Jang ◽  
Bo-Hyoung Jin
Keyword(s):  
Academic Performance ◽  
Dental School ◽  
Republic Of Korea ◽  
School Factors

scholarly journals Analysis of algebraic weighted least-squares estimators for enzyme parameters

1992 ◽  
Vol 288 (2) ◽  
pp. 533-538 ◽  
Author(s):  
M E Jones
Keyword(s):  
Least Squares ◽  
Coefficient Of Variation ◽  
Constant Coefficient ◽  
Weighted Least Squares ◽  
Simulated Data ◽  
Standard Errors ◽  
Parameter Estimates ◽  
Spreadsheet Program ◽  
Least Squares Estimators ◽  
Constant Coefficient Of Variation

An algorithm for the least-squares estimation of enzyme parameters Km and Vmax. is proposed and its performance analysed. The problem is non-linear, but the algorithm is algebraic and does not require initial parameter estimates. On a spreadsheet program such as MINITAB, it may be coded in as few as ten instructions. The algorithm derives an intermediate estimate of Km and Vmax. appropriate to data with a constant coefficient of variation and then applies a single reweighting. Its performance using simulated data with a variety of error structures is compared with that of the classical reciprocal transforms and to both appropriately and inappropriately weighted direct least-squares estimators. Three approaches to estimating the standard errors of the parameter estimates are discussed, and one suitable for spreadsheet implementation is illustrated.

Relating Latent Class Assignments to External Variables: Standard Errors for Correct Inference

2014 ◽  
Vol 22 (4) ◽  
pp. 520-540 ◽  
Author(s):  
Zsuzsa Bakk ◽  
Daniel L. Oberski ◽  
Jeroen K. Vermunt
Keyword(s):  
Latent Class ◽  
Asymptotic Variance ◽  
Latent Class Model ◽  
Monte Carlo Study ◽  
Correction Method ◽  
Standard Errors ◽  
Classification Error ◽  
Parameter Estimates ◽  
Science Literature ◽  
External Variables

Latent class analysis is used in the political science literature in both substantive applications and as a tool to estimate measurement error. Many studies in the social and political sciences relate estimated class assignments from a latent class model to external variables. Although common, such a “three-step” procedure effectively ignores classification error in the class assignments; Vermunt (2010, “Latent class modeling with covariates: Two improved three-step approaches,” Political Analysis 18:450–69) showed that this leads to inconsistent parameter estimates and proposed a correction. Although this correction for bias is now implemented in standard software, inconsistency is not the only consequence of classification error. We demonstrate that the correction method introduces an additional source of variance in the estimates, so that standard errors and confidence intervals are overly optimistic when not taking this into account. We derive the asymptotic variance of the third-step estimates of interest, as well as several candidate-corrected sample estimators of the standard errors. These corrected standard error estimators are evaluated using a Monte Carlo study, and we provide practical advice to researchers as to which should be used so that valid inferences can be obtained when relating estimated class membership to external variables.

scholarly journals The Role of Contextual Factors in Cyberspace Harm, Anxiety, Aggression, and Students' Academic Performance

2021 ◽  
Author(s):  
Zahra Sadat Aghaei ◽  
Maryam Kian
Keyword(s):  
Academic Performance ◽  
High School Students ◽  
Public Schools ◽  
Contextual Factors ◽  
Validity And Reliability ◽  
School Students ◽  
Academic Field ◽  
Research Population ◽  
Post Hoc

Background: Nowadays, cyberspace has presented both concerns and interests for the educational researchers. This study aimed to investigate the role of contextual factors in cyberspace harm, anxiety, aggression, and academic performance. Methods: A descriptive correlation research method was applied. The research population covered all high school students in Yazd (21,328 students). The number of 377 students were chosen as the sample through a stratified sampling based on the Cochran's formula. The research tools included the questionnaires of contextual factors, and the standard scales of cyberspace harm, anxiety, aggression, and academic performance. The validity and reliability of the scales were determined. Data were analyzed by variance, t-test, and Tukey post hoc test. Results: The findings showed that there are significant differences in cyberspace harm according to some contextual variables such as gender, educational and academic levels, educational district, type of school, field of study, and parents' occupation. More precisely, the harm of cyberspace was more high among the schools of District 2, non-public schools, and the academic field of humanities. Moreover, there was no relationship between the parents' education and the other variables. However, there was a significant relationship between the parents' occupation, regarding the employed mothers, as well as anxiety. Conclusion: In general, it can be concluded that cyberspace could provide students some challenges as it is affected by various contextual factors. According to the findings, several practical suggestions were presented in the study.

Sign in / Sign up

Export Citation Format

Share Document