VTechWorks Repository :: Browsing by Author "Miyazaki, Yasuo"

Browsing by Author "Miyazaki, Yasuo"

Now showing 1 - 20 of 46

Accuracy of Global Fit Indices as Indictors of Multidimensionality in Multidimensional Rasch Analysis
Harrell, Leigh Michelle (Virginia Tech, 2009-10-28)
Most research on confirmatory factor analysis using global fit indices (AIC, BIC, AICc, and CAIC) has been in the structural equation modeling framework. Little research has been done concerning application of these indices to item response models, especially within the framework of multidimensional Rasch analysis. The results of two simulations studies that investigated how sample size, between-dimension correlation, and test length affect the accuracy of these indices in model recovery using a multidimensional Rasch analysis are described in this dissertation. The first study analyzed dichotomous data, with model-to-data misfit as an additional independent variable. The second study analyzed polytomous data, with rating scale structure as an additional independent variable. The interaction effect between global fit index and between-dimension correlation had very large effect sizes in both studies. At higher values of between-dimension correlation, AIC indicated the correct two-dimension generating structure slightly more often than does the BIC or CAIC. The correlation by test length interaction had an odds ratio indicating practical importance in the polytomous study but not the dichotomous study. The combination of shorter tests and higher correlations resulted in a difficult-to-detect distinction being modeled with less statistical information. The correlation by index interaction in the dichotomous study had an odds ratio indicating practical importance. As expected, the results demonstrated that violations of the Rasch model assumptions are magnified at higher between-dimension correlations. Recommendations for practitioners working with highly correlated multidimensional data include creating moderate length (roughly 40 items) instruments, minimizing data-to-model misfit in the choice of model used for confirmatory factor analysis (MRCMLM or other MIRT models), and making decisions based on multiple global indices instead of depending on one index in particular.
Assessing Factors that Distinguish First-Generation College Students from Non First-Generation College Students at an Urban Comprehensive University
Jenkins, Anthony L. (Virginia Tech, 2007-03-26)
The purpose of the study was to compare a freshman cohort of first and non first-generation college students enrolled in an urban university and to identify characteristics that distinguish the two groups in terms of selected demographics, pre-college behaviors and beliefs (expectations and personal traits). Moreover, the study sought to identify variables whose distribution indicated a significant difference between the two groups and rank those variables by order of the strength of association. Data analysis for this study consisted of a combination of chi-square and descriptive discriminate analysis using logistic regression. Chi-square analysis was the preliminary statistical procedure used in this study. I relied on a sequence of chi-square analyses to help identify a list of statistically significant variables to be used in the subsequent descriptive discriminate logistic regression model. Descriptive discriminate analysis was used because its primary function is designed to reveal projected differences among groups (Huberty, 1994). The results revealed seven important characteristics (Reading for pleasure (Hpw0111), Household income (Income), Asked teacher for advise (Act0114), Rate computer skills (Rate0103), Get a bachelor's degree (Futact11), Change major field of study (Futact01) and Obtain recognition by colleague (Goal0103) were commonly statistically significant student characteristics across all race/ethnicity groups, and three (Gain a general education (Reason05), High school grade point average (HSGPA) and Felt overwhelmed (Act0110) were unique to one or some of the groups. These variables can also be viewed as predictors that help identify the likelihood that a student is first-generation. Results of this study had implications for the practice of high school guidance counselors, student and academic affairs practitioners and specifically support services personnel and financial aid officers.
College-Going Behaviors: Are there School Effects for the Rural Student?
Hamill, Bridget (Virginia Tech, 2018-04-30)
This study considered the school effects of college going behavior for rural students. Of interest were the effects of location and college-going culture within a given school. The research questions asked, included: 1. What are the effects of rural school location and college-going culture on public high school graduation? 2. What are the effects of rural school location and college-going culture on college enrollment? 3. For the public high school graduates who enrolled in college, what are the effects of rural school location and college-going culture on the control structure of the college program enrolled? 4. For the public high school graduates who enrolled in college, what are the effects of rural school location and college-going culture on type of college program enrolled (two-year vs. four-year)?> 5. For the public high school graduates who enrolled in college, what are the effects of rural school location and college-going culture on full-time vs. part-time enrollment? The study used data from the HSLS:09 survey. The data was analyzed using Hierarchical Generalized Linear Modeling. This study found that the odds of attending college decreased 18.7% for rural students. There was also a 4.8% decrease in the odds of college enrollment by students from majority White high schools. School's with high mean GPA's were more likely to have students graduate from high school, enroll in college, and attend 4-year institutions. High rates of school problems negatively affected students and demonstrated decreased odds of high school graduation and college enrollment. The role of counselors had demonstrated effects on students. Schools with counseling offices that focused a high number of hours on college counseling increase the odds their students graduate would from high school and attend a 4-year institution. Students attending high schools with a college counselor dedicated to college applications were 4.30 times more likely to attend a not-for-profit institution than a for-profit institution.
Comparison of the Item Response Theory with Covariates Model and Explanatory Cognitive Diagnostic Model for Detecting and Explaining Differential Item Functioning
Krost, Kevin Andrew (Virginia Tech, 2023-10-06)
In psychometrics, a concern is that the assessment is fair for all students who take it. The fairness of an assessment can be evaluated in several ways, including the examination of differential item functioning (DIF). An item exhibits DIF if a subgroup has a lower probability of answering an item correctly than another subgroup after matching on academic achievement. Subgroups include race, spoken language, disability status, or sex. Under item response theory (IRT), a single score is given to each student since IRT assumes that an assessment is only measuring one construct. However, under cognitive diagnostic modeling (CDM), an assessment measures multiple specific constructs and classifies students as having mastered the construct or not. There are several methods to detect DIF under both types of models, but most methods cannot conduct explanatory modeling. Explanatory modeling consists of predicting item responses and latent traits using relevant observed or latent covariates. If an item exhibits DIF which disadvantages a subgroup, covariates can be modeled to explain the DIF and indicate either true or spurious differences. If an item exhibited statistically significant DIF which became nonsignificant after modeling explanatory variables, then the DIF would be explained and considered spurious. If the DIF remained significant after modeling explanatory variables, then there was stronger evidence that DIF was present and not spurious. When an item exhibits DIF, the validity of the inferences from the assessment is threatened and group comparisons become inappropriate. This study evaluated the presence of DIF on the Trends in International Math and Science Study (TIMSS) between students who speak English as a first language (EFL) and students who do not speak English as a first language (multilingual learners [ML]) in the USA. The 8th grade science data was analyzed from the year 2011 since science achievement remains understudied, the 8th grade is a critical turning point for K-12 students, and because 2011 was the most recent year that item content is available from this assessment. The item response theory with covariates (IRT-C) model was used as the explanatory IRT model, while the reparameterized deterministic-input, noisy "and" gate (RDINA) model was used as the explanatory CDM (E-CDM). All released items were analyzed for DIF by both models with language status as the key grouping variable. Items that exhibited significant DIF were further analyzed by including relevant covariates. Then, if items still exhibited DIF, their content was evaluated to determine why a group was disadvantaged. Several items exhibited significant DIF under both the IRT-C and E-CDM. Most disadvantaged ML students. Under the IRT-C, two items that exhibited DIF were explained by quantitative covariates. Two items that did not exhibit significant nonuniform DIF became significant after explanation. Whether or not a student repeated elementary school was the strongest explanatory covariate, while confidence in science explained the most items. Under the E-CDM, five items initially exhibited significant uniform DIF with one also exhibiting nonuniform DIF. After scale purification, two items exhibited significant uniform DIF, and one exhibited marginally significant DIF. After explanatory modeling, no items exhibited significant uniform DIF, and only one item exhibited marginally significant nonuniform DIF. Examining covariates, home educational resources explained the most with ten items and the strongest positive covariate. Repeated elementary school had the strongest absolute effect. Examining the item content of 14 items, most items had no causal explanation for the presence of DIF. In four items, a causal mechanism was identified and concluded to exhibit item bias. An item's cognitive domain had a relationship with DIF items, with 79% of items under the Knowing domain. Based on these results, DIF that disadvantaged ML students was present among several items on this science assessment. Both the IRT-C and E-CDM identified several items exhibiting DIF, quantitative covariates explained several items exhibiting DIF, and item bias was discovered in several items. Following up on this empirical study, a simulation study was performed to evaluate DIF detection power and Type I error rates of the Wald test and likelihood ratio (LR) test, and parameter recovery when ignoring subgroups, using the compensatory reparameterized unified model (C-RUM). Factors included sample size, DIF magnitude, DIF type, Q-matrix complexity, their interaction effects, and p-value adjustment. Evaluating DIF under the C-RUM, the DIF detection method had the largest effect on Type I error rates, with the Wald test recovering the nominal p-value much better than the LR test. In terms of power, DIF magnitude was the most important factor, followed by Q-matrix complexity. As DIF magnitude increased and Q-matrix complexity decreased, power rates increased. In terms of parameter recovery, the DIF type had the strongest effect, followed by Q-matrix complexity. Nonuniform DIF recovered the parameter more than uniform DIF, while fewer attributes measured by an item improved parameter recovery. Several factors affected DIF detection power and Type I error, including DIF detection method, DIF magnitude, and Q-matrix complexity. For parameter recovery, DIF type had an impact, along with Q-matrix complexity, and DIF magnitude.
A Cross-National Study of Civic Knowledge Test Scores
Gregory, Christopher Ryan (Virginia Tech, 2015-10-23)
The purpose of this study is to examine the relationship among student civic knowledge scores and several different variables each at the student, classroom/school, and national levels using the IEA CIVED study international data set collected in 1999 from 27 countries. The student level predictors included two elements of socioeconomic status (a student's parental education, their home literacy level measured by the number of books at home), student's perception of an open classroom climate, student aspiration of obtaining higher education, and other variables that were identified as relevant to the dependent variable in the literature. The classroom/school level predictors included teacher's degree in civics, in-service training, teaching confidence, and school safety in addition to the compositional variable created as the classroom/school averages by aggregating the student level variables. Then I investigated whether instructional methods focusing on the student activities the teacher employed in the classroom and an open classroom climate were associated after accounting for the above student and school level background variables. National level variables such as GNP, GINI index, democratic system, public education expenditure, and etc. as well as compositional variables obtained by aggregating the classroom/school variables were also added to the model to investigate if they were associated with students' civic knowledge scores and whether they could explain between nations variability. The study used a three-level hierarchical linear model to analyze the data, with number of students, N=56,579, number of classrooms/schools, J=3443, and number of countries, K=27. Some of the key findings was that there were significant variations of civics knowledge among nations, and significant variations of civic knowledge scores between school and within nations, no statistically significant association between teacher's practice and civics knowledge scores, however the student perception of an open classroom climate was significant at all 3 levels. These findings were interpreted in terms of policies and practices that could be implemented to improve students' civic knowledge.
A Cross-national Study of Mathematics Achievement Via Three-level Multilevel Models
Lee, Youjin (Virginia Tech, 2023-01-18)
The present study explored the effects of the national and cultural contexts on students' mathematics achievement. The study also investigated the nature and magnitude of student-level (level 1), school-level (level 2), and country-level (level 3) factors that are associated with math achievement. The Program for International Student Assessment (PISA) 2018 datasets were used. The main predictors focusing on this study included university admission procedure and the country's culture of mindsets about intelligence at level 3, indicating extra-curricular activities at level 2, growth mindset, and resilience self-efficacy at level 1. Other than main predictors, various predictors including country's characteristics, school characteristics, school climate factors, students' demographic characteristics, and non-cognitive abilities were added in the analysis to examine the main predictors are statistically significant after controlling for other predictors. The findings of HLM analysis showed that mathematics achievement is associated with national and cultural contexts since the study found 31.30% of the total variation was accounted for level 3 in math achievement. Also, the significant findings of the study indicated that university admission procedure was significantly associated with country-mean math achievement while the country's culture of mindsets about intelligence was not at level 3. At level 2, providing extra-curricular activities in school was a significant predictor for math achievement. At level 1, a growth mindset and information and Communication Technology (ICT) usage were positively associated with math achievement. The other significant predictors for math achievement were found in the model. In addition, the study found that the compositional effect of ICT usage explained a significant amount of between schools and countries variance even after controlling for other predictors in the analysis. Moreover, the study found several counterintuitive association phenomena due to shift of meaning. These findings were explained in terms of practical and theoretical implications for policymakers, educators, and researchers to improve students' mathematics achievement.
A Cross-National Study of School Violence
Agnich, Laura Elizabeth (Virginia Tech, 2011-05-11)
This study examines the predictors of school violence cross-nationally, testing the applicability of criminological theories of adult violence to violence in the school setting. Using hierarchical linear modeling (HLM), a method of multi-level linear analysis, of the 2007 Trends in International Math and Science Studies (TIMSS) data augmented with data from UN Human Development Reports, UN Demographic Yearbook, CIA World Factbook and the World Health Organization Mortality Database, I determine the predictors of school violence at the school and national levels to determine what variables account for cross-national variation in the level of school violence. Hierarchical linear modeling (HLM) takes into account the structure of nested data, and this study examines schools nested within nations. The relationships between school and national level inequalities, social disorganization, institutional anomie, social support, resource deprivation theories and school violence are tested. Violence is operationally defined as a continuum of aggression ranging from non-physical to physical (see Yu 2003), incorporating low-level as well as more serious forms of interpersonal violence. I find that measures of social disorganization, institutional anomie and resource deprivation at both the school and national levels predict higher levels of violence within schools. Surprisingly, homogeneity rather than heterogeneity is a significant predictor of physical bullying. In addition, math achievement and achievement score variation significantly predict the level of school violence cross-nationally. At the national level, placing too much emphasis on students' achievement on standardized tests may inadvertently create a culture conducive to school violence. Emphasizing a diverse range of ways to measure students' achievement other than standardized testing may reduce the likelihood that students experience strain and engage in violent behavior at school. This research is the first to use multi-level linear analysis to discern the school and national level predictors of school violence.
Curriculum Track And Its Influences On Predicting High School Dropout Likelihood
Mohd Kamalludeen, Rosemaliza (Virginia Tech, 2012-07-06)
Dropping out of school is a major concern as high school graduation credentials have been used as an important measurement tool to define post-secondary success. Numerous researchers presented a multitude of factors that predict dropouts at individual and school levels. Curriculum track choice, or high school course-taking sequence, defines students' schooling career and ultimately the post-secondary path that they choose (Plank, DeLuca, & Estacion, 2008). Scholars have debated on various outcomes related to dropouts influenced by various curriculum choices, namely academic, career and technical education (CTE), dual enrollment, and general curriculum. Several argued students following academic tracks are more likely to graduate. Others claim that CTE benefits students who are at-risk and suppresses dropout likelihood (Rumberger & Sun, 2008). New vocationalism or dual enrollment has proven successful at reducing dropout rates. This study attempted to investigate the influence of curriculum track and CTE program areas on dropout likelihood while controlling for possible individual differences. Analysis was conducted via Hierarchical Generalized Linear Modeling (HGLM) due to the nested data structure of Education Longitudinal Study of 2002 (ELS). Variables included were academic background, academic and career aspiration, school-sponsored activity participation, school minority composition, school average student socio-economic status (SES), school type (private or public), school urbanicity, CTE courses offered at the school, and demographic indicators (gender, race, and SES). Findings reflect higher dropout likelihood among general curriculum participants than academic and occupational concentrators after controlling for all possible individual differences. Dual concentrators had 0% dropout rate, and therefore comparison with other curriculum tracks was not possible via HGLM analysis. Results suggest substantial importance of academic background, post-secondary education plans, and school-sponsored activity participation in predicting dropout likelihood. Comparing CTE program areas, Family and Consumer Sciences, Human Services, Public Services, Health and Education (Human Services area) participants were more likely to drop out than other program areas while Technology Education participants were less likely to drop out than Human Services and 2 or more CTE program area participants. Results suggest 9th grade overall GPA and school-sponsored activity participation as substantial predictors of dropout likelihood among occupational concentrators. Variability across schools was insignificant.
Detecting Rater Centrality Effect Using Simulation Methods and Rasch Measurement Analysis
Yue, Xiaohui (Virginia Tech, 2011-07-14)
This dissertation illustrates how to detect the rater centrality effect in a simulation study that approximates data collected in large scale performance assessment settings. It addresses three research questions that: (1) which of several centrality-detection indices are most sensitive to the difference between effect raters and non-effect raters; (2) how accurate (and inaccurate), in terms of Type I error rate and statistical power, each centrality-detection index is in flagging effect raters; and (3) how the features of the data collection design (i.e., the independent variables including the level of centrality strength, the double-scoring rate, and the number of raters and ratees) influence the accuracy of rater classifications by these centrality-detection indices. The results reveal that the measure-residual correlation, the expected-residual correlation, and the standardized deviation of assigned scores perform better than the point-measure correlation. The mean-square fit statistics, traditionally viewed as potential indicators of rater centrality, perform poorly in terms of differentiating central raters from normal raters. Along with the rater slope index, the mean-square fit statistics did not appear to be sensitive to the rater centrality effect. All of these indices provided reasonable protection against Type I errors when all responses were double scored, and that higher statistical power was achieved when responses were 100% double scored in comparison to only 10% being double scored. With a consideration on balancing both Type I error and statistical power, I recommend the measure-residual correlation and the expected-residual correlation for detecting the centrality effect. I suggest using the point-measure correlation only when responses are 100% double scored. The four parameters evaluated in the experimental simulations had different impact on the accuracy of rater classification. The results show that improving the classification accuracy for non-effect raters may come at a cost of reducing the classification accuracy for effect raters. Some simple guidelines for the expected impact of classification accuracy when a higher-order interaction exists summarized from the analyses offer a glimpse of the "pros" and "cons" in adjusting the magnitude of the parameters when we evaluate the impact of the four experimental parameters on the outcomes of rater classification.
The Development of a Conceptual Framework for Identifying Functional, Expressive, Aesthetic, and Regulatory Needs for Snowboarding Helmets
Chae, Myung-Hee (Virginia Tech, 2006-11-10)
The purpose of this research was to identify the design characteristics and attitudes that impact the use of snowboarding helmets and to test statistically a proposed conceptual framework for identifying perceived importance of functional, expressive, aesthetic, and regulatory (FEAR) needs of snowboarding helmets for current snowboarders. Data for this study was collected online. The final sample was composed of 391 participants, which represented a 13.67% response rate. Multiple comparisons were used to examine mean differences among the FEAR variables, as well as attitudes toward helmet use. A multiple linear regression was used to test four proposed hypotheses. The results of hypotheses revealed that there was an impact between attitudes toward helmet use and perceived importance of functional needs, but this relationship depended on the level of expressive needs, aesthetic needs, and helmet usage. The typical impact of functional needs on attitudes toward helmet use was positive (slope = .013) when all variables were at their respective means (Hypothesis 1). Hypothesis 2 tested to see if there was an impact between attitudes toward helmet use and perceived importance of expressive needs, but again, this relationship depended on the level of functional needs and helmet usage. The typical impact of expressive needs on attitudes toward helmet use was positive (slope = .014) when all variables were at their respective means. Similarly, the impact between attitudes toward helmet use and the perceived importance of aesthetic needs was dependent on the level of functional needs and helmet usage. The impact of aesthetic needs on attitudes toward helmet use was typically negative (slope = -.012) when all variables were at their respective means (Hypothesis 3). Finally, Hypothesis 4 looked at the impact between attitudes toward helmet use and the perceived importance of regulatory needs. Unlike the other three hypotheses, this relationship did not depend on any other variables. The impact of regulatory needs on attitudes toward helmet use was positive, and the strength of association was .010. Although hypothesis 1-3 were substantially supported, and hypothesis 4 was fully supported, from a statistical point of view, the interaction effects between the independent variables (i.e., FEA needs) and the covariate (i.e., helmet usage) limit the findings, so we can not really state that the hypotheses were supported. However, based on information obtained from the respondents in this study, the application of a FEAR needs assessment of snowboarding helmets could help to enhance the overall performance of snowboarders. In other words, the improvement of helmet functionality, expressive qualities, aesthetic attributes and regulatory needs would provide a more enjoyable snow activity to participants. Thus, the conceptual framework of the perceived importance of FEAR needs would be acceptable to understand the attitudes toward helmet use among snowboarders.
Diagnostic Modeling of Intra-Organizational Mechanisms for Supporting Policy Implementation
Mutcheson, Brock (Virginia Tech, 2016-06-28)
The Virginia Guidelines for Uniform Performance Standards and Evaluation Criteria for Teachers represented a significant overhaul of conventional teacher evaluation criteria in Virginia. The policy outlined seven performance standards by which all Virginia teachers would be evaluated. This study explored the application of cognitive diagnostic modeling to measure teachers' perceptions of intra-organizational mechanisms available to support educational professionals in implementing this policy. It was found that a coarse-grained, four-attribute compensatory, re-parameterized unified model (C-RUM) fit teacher perception data better and had lower standard errors than the competing finer-grained models. The Q-matrix accounted for the complex loadings of items to the four theoretically and empirically driven mechanisms of implementation support including characteristics of the policy, teachers, leadership, and the organization. The mechanisms were positively, significantly, and moderately correlated which suggested that each mechanism captured a different, yet related, component of policy implementation support. The diagnostic profile estimates indicated that the majority of teachers perceived support on items relating to "characteristics of teachers." Moreover, almost 60% of teachers were estimated to belong to profiles with perceived support on "characteristics of the policy." Finally, multiple group multinomial log-linear models (Xu and Von Davier, 2008) were used to analyze the data across subjects, grade levels, and career status. There was lower perceived support by STEM teachers than non-STEM teachers who have the same profile, suggesting that STEM teachers required differential support than non-STEM teachers. The precise diagnostic feedback on the implementation process provided by this application of diagnostic models will be beneficial to policy makers and educational leaders. Specifically, they will be better prepared to identify strengths and weaknesses and target resources for a more efficient, and potentially more effective, policy implementation process. It is assumed that when equipped with more precise diagnostic feedback, policy makers and school leaders may be able to more confidently engage in empirical decision making, especially in regards to targeting resources for short-term and long-term organizational goals subsumed within the policy implementation initiative.
Disaggregating Within-Person and Between-Person Effects in the Presence of Linear Time Trends in Time-Varying Predictors: Structural Equation Modeling Approach
Hori, Kazuki (Virginia Tech, 2021-06-01)
Educational researchers are often interested in phenomena that unfold over time within a person and at the same time, relationships between their characteristics that are stable over time. Since variables in a longitudinal study reflect both within- and between-person effects, researchers need to disaggregate them to understand the phenomenon of interest correctly. Although the person-mean centering technique has been believed as the gold standard of the disaggregation method, recent studies found that the centering did not work when there was a trend in the predictor. Hence, they proposed some detrending techniques to remove the systematic change; however, they were only applicable to multilevel models. Therefore, this dissertation develops novel detrending methods based on structural equation modeling (SEM). It also establishes the links between centering and detrending by reviewing a broad range of literature. The proposed SEM-based detrending methods are compared to the existing centering and detrending methods through a series of Monte Carlo simulations. The results indicate that (a) model misspecification for the time-varying predictors or outcomes leads to large bias of and standard error, (b) statistical properties of estimates of the within- and between-person effects are mostly determined by the type of between-person predictors (i.e., observed or latent), and (c) for unbiased estimation of the effects, models with latent between-person predictors require nonzero growth factor variances, while those with observed predictors at the between level need either nonzero or zero variance, depending on the parameter. As concluding remarks, some practical recommendations are provided based on the findings of the present study.
The Effect of Organizational Characteristics on School Effectiveness: A Multilevel Analysis of the Gulf Cooperation Council States
Alenezi, Abdulaziz Sh (Virginia Tech, 2023-01-10)
According to the findings of the 2019 Trends in International Mathematics and Science Study (TIMSS), fourth-grade students from Gulf Cooperation Council (GCC) countries—consisting of Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates—performed below average compared to other countries on mathematics and science assessments. Despite this, little organizational research has examined potential factors that might have contributed to these results or sought to quantify the variability in school effectiveness in GCC countries. Hence, the present study sought to address this gap by quantifying the variability in school effectiveness in these countries. Using TIMSS 2019 data and multilevel analysis within each GCC member state, the study found school effectiveness varied significantly, ranging between 17% and 60%, considerably more than the variation typically seen in Western countries. In addition, several school-level organizational factors showed a significant impact on school effectiveness. Schools with more adequate resources, higher-quality teachers, greater parental involvement at the school level, and a safer and more orderly environment tended to display higher effectiveness as measured by average mathematics achievement. This finding should encourage researchers and policymakers to have more informed discussions about school effectiveness in the region.
Effect of Unequal Sample Sizes on the Power of DIF Detection: An IRT-Based Monte Carlo Study with SIBTEST and Mantel-Haenszel Procedures
Awuor, Risper Akelo (Virginia Tech, 2008-06-19)
This simulation study focused on determining the effect of unequal sample sizes on statistical power of SIBTEST and Mantel-Haenszel procedures for detection of DIF of moderate and large magnitudes. Item parameters were estimated by, and generated with the 2PLM using WinGen2 (Han, 2006). MULTISIM was used to simulate ability estimates and to generate response data that were analyzed by SIBTEST. The SIBTEST procedure with regression correction was used to calculate the DIF statistics, namely the DIF effect size and the statistical significance of the bias. The older SIBTEST was used to calculate the DIF statistics for the M-H procedure. SAS provided the environment in which the ability parameters were simulated; response data generated and DIF analyses conducted. Test items were observed to determine if a priori manipulated items demonstrated DIF. The study results indicated that with unequal samples in any ratio, M-H had better Type I error rate control than SIBTEST. The results also indicated that not only the ratios, but also the sample size and the magnitude of DIF influenced the behavior of SIBTEST and M-H with regard to their error rate behavior. With small samples and moderate DIF magnitude, Type II errors were committed by both M-H and SIBTEST when the reference to focal group sample size ratio was 1:.10 due to low observed statistical power and inflated Type I error rates.
The Effects of Cumulative Social Capital on Job Outcomes of College Graduates
Wang, Yadan (Virginia Tech, 2008-10-31)
The current study drew on a large and diverse body of literature on social capital and aimed to understand its role in the process of transition from college to work. In particular, this research studied the cumulative effects of social capital formed in high school years and college years and examined its relationship with job outcomes. The study used the National Education Longitudinal Study (NELS) to examine whether early investment in the social capital of young adolescents produced better job outcomes in their adulthood. Families and schools were two primary sources of social capital considered in the current study. Parental involvement in a young person's life, extra-curricular activities and participation in volunteer organizations were some of the forms of social capital hypothesized to influence job outcomes after college. Structural equations modeling was used to trace the effects of the presence of social capital as early as the 8th grade in shaping student's later career status. The longitudinal data measured social capital beginning in the 8th grade and every 2 years thereafter, so that the cumulative effects of the social capital resources were investigated. Overall, the hypothesized model was found to fit the data and the findings have suggested a set of positive and direct effects of social capital on job outcomes.
The Effects of Social Support from Parent, Teacher, and Peers on High School Students' Math Achievement: The Mediational Role of Motivational Beliefs
Duan, Xuejing (Virginia Tech, 2018-07-02)
The present study explored the direct influences of contextual social support, including parental involvement, perceived teacher support, and peer influence, on 11th-grade students' math achievement. The study also examined the indirect influences of these contextual social support factors on students' achievement through their math motivation in math courses. The first follow-up year data of High School Longitudinal Study of 2012 (HSLS: 09) was used for this study. Structural equation modeling (SEM) served as the main statistical technique to examine the relationships among variables. The results of this study showed three sets of important findings. The first set showed that students' perception of teacher support and peer influence were significantly and directly related to students' math achievement, with the relationship between peer influence and math achievement being positive and the relationship between perceived teacher support and math achievement being negative. Controlling for other variables in the model, parental involvement was not significantly related to student math achievement. The second set of findings demonstrated that math motivation indeed plays a significant role in mediating the relationships of social support (from teachers and peers, but not from parental involvement) and student math achievement in high school. The third set of findings indicated that both family SES and prior math achievement influenced student social support and math achievement. Furthermore, two main deviations were found between White/Asian and African-American/Hispanic student models. Perceived teacher support negatively and significantly influenced White/Asian students' math achievement, but it had no significant influence on African-American/Hispanic students. In addition, math motivation had a stronger influence on the math achievement for White/Asian students than African-American/Hispanic students. The present study makes significant theoretical and practical contributions to the body of knowledge on the role of parental involvement, perceived teacher support, and peer influence on math achievement at the high school level using nationally representative data.
Examining Social Capital as a Predictor of Enrollment in Postsecondary Education for Low SES Students: A Multilevel Analysis
Stimpson, Matthew (Virginia Tech, 2009-03-23)
This study examined whether measures of social capital were significant predictors of enrollment in postsecondary education for students from a low SES background. Results take the form of two articles. The first article addresses enrollment in four-year institutions of postsecondary education, and the second article addresses enrollment in two-year institutions of postsecondary education. The research questions for this study were: 1. Does probability of enrollment in a four-year postsecondary institution or a two-year postsecondary institution for low SES students differ by mean school SES? 2. Does probability of enrollment in a four-year postsecondary institution or a two-year postsecondary institution for low SES students differ by school locale? 3. When controlling for contextual or environmental variables and student background characteristics, are low SES students with higher levels of social capital more likely to enroll in a four-year postsecondary institution or a two-year postsecondary institution than low SES students with lower levels of social capital? 4. When controlling for contextual or environmental variables, background characteristics, and level of social capital does probability of enrollment in a four-year institution of postsecondary education or a two-year postsecondary institution vary by race for low SES students? When controlling for school level variables, academic achievement and preparation, and select background characteristics, low SES students with higher levels of social capital are more likely to enroll in a four-year college. Students whose parents expected them to obtain more education and those students who obtained more information about attending college were more likely to enroll in a four-year university. In the analysis of enrollment in four-year institutions of postsecondary education, African American low SES students were three times more likely to enroll in a four-year college or university than low SES Caucasian students. Only one measure of social capital, information acquisition, was significantly related to enrollment in a two-year institution of postsecondary education. No significant variability in probability of enrollment in a two-year institution of postsecondary education was observed by either of the school level variables used. Race was not a significant factor when controlling for background characteristics and the measures of social capital used in this study.
Factors Influencing Undergraduate Women's Educational Aspirations
Davis, Sharrika D. (Virginia Tech, 2009-04-02)
Education is one key to economic prosperity and a predictor of overall life satisfaction. The further one progresses through the educational pipeline, the more likely it is that she may prosper. However, in a society bolstered by patriarchal systems, economic and educational inequalities exist among the genders. Educational aspirations are influenced by students' socialization experiences. Faculty teach students about their discipline. Families influence educational pursuits. Peers serve as reinforcements or challenges to academic progress. All three groups are socialization agents to students pursuing higher education. Research indicates that various socialization agents influence whether students pursue an undergraduate degree. However, there is little literature specifically focused on women and less on the relationship between women's undergraduate socialization experiences and their decision to enroll in graduate studies. The purpose of this study was to determine whether certain collegiate experiences (with family, faculty and peers) predict undergraduate women's expectation to enroll in graduate study and to determine if the experiences influence expectation to enroll by race. The sample included women who completed the College Student Experiences Questionnaire (CSEQ) Fourth edition. The study employed logistic regression to explore the relationship between undergraduate women's educational aspirations and family, faculty and peer influences. In addition, I examined whether the associations between family, faculty and peers differed by race/ethnicity. The results of the logistic regression revealed that academic ability (GPA) and peer experiences influenced advanced degree aspirations. In addition, race/ethnicity does matter, i.e., being of African-American or Latina decent is associated with a higher level of advanced degree aspiration. Also, as frequency of interactions between faculty and African-American women increase — aspiration decreases. These findings suggest that it is important to consider the various factors that influence advanced degree aspiration. This is especially important since advanced degrees can be elemental to economic prosperity.
Family and Clinician Effects on Costs of Psychiatric Emergency Services Dispositions
Nielson, L. Reece (Virginia Tech, 2009-03-26)
Families play a key role in psychiatric emergency services (PES). Given the cost of PES in terms of dollars and restrictiveness, clients, families, providers, payers, and policymakers involved in these services need more understanding of how families affect these key PES outcomes. Marriage and family therapy theories offer frameworks for understanding family and provider system dynamics in PES. This study explores how family presence and family quality influence restrictiveness and cost of PES dispositions, and how they moderate the effect of suicide risk, homicide risk, and inability to care for self on those outcomes. The sample of 306 clients and 33 clinicians was drawn from the records of a mobile PES unit serving a rural area. A regression-based, quantitative methodology, Hierarchical Linear Modeling (HLM), was employed to explore associations between restrictiveness and client risk and family factors, as well as differences in dispositions between PES clinicians. In order to extend practical implications, the same questions were also examined in monetary terms by translating restrictiveness into cost of dispositions. Results show inability to care for self and suicide risk to be the strongest predictors of increased restrictiveness and cost. Family quality appeared to reduce restrictiveness but not cost and only when not considering interactions with individual risk factors. When interactions were considered, family quality exhibited a statistically significant disordinal interaction with inability to care for self. That is, when clients were unable to care for self, positive family quality worked toward increasing restrictiveness and cost, perhaps due to families seeking help for the client. However, when clients were able to care for self, positive family quality worked in the opposite direction (i.e., toward reducing restrictiveness and cost). Theoretical and practical implications of this interaction were considered. There was found no significant variability in dispositions and associated costs between clinicians, which may be evidence of standardized clinician training and procedures. Non-standardized instrumentation, lack of comparison with other programs or sites, and limited cell sample size are limitations of the study. This study shows the complexity of family systems in PES and provides basis for recommendations for future research and clinical practice.
Going Beyond the Outcome Assessment Minimum: Toward a Framework to Assess Students' Integrative Learning in a University General Education Program
Li, Mengyun (Virginia Tech, 2023-01-17)
Prior research has demonstrated the efficacy of general education coursework among American college students (Ball, 2012; Rosenzweig, 2009). Traditional models of general education programs are predicated on the understanding that exposure to a broad set of educational experiences creates well-rounded graduates (Roche, 2010). However, emerging research shows the importance of integrative learning experiences including general education programs (Lowenstein, 2015). These programs are just now at the initial stages of development and implementation at colleges and universities making it possible to study direct effects on student learning. What remains, however, is limited ways to measure such learning in emerging programs. One large, research university in a mid-Atlantic state provides opportunity to construct a measure of integrated learning. This study addressed the salient literature on general education in higher education today and then used quantitative methods and qualitative methods to investigate an empirically based measure of integrative learning. Findings revealed the continuous process of integrative learning from disciplinary knowledge to application to real world and established an initial framework for assessing students learning outcomes of integration. Finally, the research provided implications for researchers and practitioners to utilize the instrument and extend it to a wider range of students and academic programs.

Browsing by Author "Miyazaki, Yasuo"

Results Per Page

Sort Options