Browsing by Author "Wu, Jiun-Yu"

Now showing 1 - 20 of 22

A Solution to Modeling Multilevel Confirmatory Factor Analysis with Data Obtained from Complex Survey Sampling to AvoidConflated Parameter Estimates
(2017-09-22)
Wu, Jiun-Yu
;
Lin, John J. H.
;
Nian, Mei-Wen
;
Hsiao, Yi-Cheng
;
教育研究所
The issue of equality in the between-and within-level structures in Multilevel Confirmatory Factor Analysis (MCFA) models has been influential for obtaining unbiased parameter estimates and statistical inferences. A commonly seen condition is the inequality of factor loadings under equal level-varying structures. With mathematical investigation and Monte Carlo simulation, this study compared the robustness of five statistical models including two model-based (a true and a mis-specified models), one design-based, and two maximum models (two models where the full rank of variance-covariance matrix is estimated in between level and within level, respectively) in analyzing complex survey measurement data with level-varying factor loadings. The empirical data of 120 3rd graders' (from 40 classrooms) perceived Harter competence scale were modeled using MCFA and the parameter estimates were used as true parameters to perform the Monte Carlo simulation study. Results showed maximum models was robust to unequal factor loadings while the design-based and the miss-specified model-based approaches produced conflated results and spurious statistical inferences. We recommend the use of maximum models if researchers have limited information about the pattern of factor loadings and measurement structures. Measurement models are key components of Structural Equation Modeling (SEM); therefore, the findings can be generalized to multilevel SEM and CFA models. Mplus codes are provided for maximum models and other analytical models.
Dynamics of Teacher-Student Relationships: Stability and Change Across Elementary School and the Influence on Children's Academic Success
(2012-07-01)
Spilt, Jantine L.
;
Hughes, Jan N.
;
Wu, Jiun-Yu
;
Kwok, Oi-Man
;
交大名義發表
"This study modeled teacher-student relationship trajectories throughout elementary school to predict gains in achievement in an ethnic-diverse sample of 657 academically at-risk students (mean age = 6.57 years, SD = .39). Teacher reports of warmth and conflict were collected in Grades 1-5. Achievement was tested in Grades 1 and 6. For conflict, low-stable (normative), low-increasing, high-declining, and high-stable trajectories were found. For warmth, high-declining (normative) and low-increasing patterns were found. Children with early behavioral, academic, or social risks were underrepresented in the normative trajectory groups. Chronic conflict was most strongly associated with underachievement. Rising conflict but not declining Conflict coincided with underachievement. The probability of school failure increased as a function of the timing and length of time children were exposed to relational adversity."
Editorial: Recent Advancements in Structural Equation Modeling (SEM): From Both Methodological and Application Perspectives
(2018-10-09)
Kwok, Oi-Man
;
Cheung, Mike W. L.
;
Jak, Suzanne
;
Ryu, Ehri
;
Wu, Jiun-Yu
Exploring factors influencing residents' health outcomes in long-term care facilities: 1-year follow-up using latent growth curve model
(2014-11-01)
Liu, Li-Fan
;
Weng, Rhay-Hung
;
Wu, Jiun-Yu
;
教育研究所
;
Institute of Education
Purpose This study explored the residents' health outcomes of long-term care (LTC) facilities and examined the risk factors in individual and institutional levels during 1 year of admission. Methods The study included four stages of interviews with residents in 31 nursing homes and 64 residential care homes. Three hundred and twenty-five residents at baseline were interviewed, and 206 completed the interviews at follow-up. Five outcomes including residents' physical/mental functional status and subjective health status in Short Form-36 were analyzed using latent growth curve models (LGCMs). Results Only the physical component summary (PCS) had increased significantly. The most influential risk factors to outcomes were the intra-individual-level time-varying variables, including self-rated health and with/without tubing care. Some predictive inter-individual-level factors were also found. For institutional characteristics, small-sized homes (<49 beds) with low occupancy rates showed a lower growth rate in residents' mental component summary (MCS) and PCS over 1 year and private sector homes showed the most significant growth rates in MCS. Conclusions The methodological strength using LGCMs provides a framework for systematically assessing the influence of risk factors from various levels on residents' outcomes and follow-up change. It is evident that factors in various levels all influenced residents' outcomes which support critical information for case mix and quality management in LTC facilities. Under the scenario of a surplus of institutional care in Taiwan, we suggest that institutions must focus more on residents' psychological well-being and care quality, especially in small-sized homes in relation to the outcomes of its residents.
Exploring factors influencing residents\' health outcomes in long-term care facilities: 1-year follow-up using latent growth curve model
(2014-11-01)
Liu, Li-Fan
;
Weng, Rhay-Hung
;
Wu, Jiun-Yu
;
師資培育中心
;
Center of Teacher Education
Purpose This study explored the residents\' health outcomes of long-term care (LTC) facilities and examined the risk factors in individual and institutional levels during 1 year of admission. Methods The study included four stages of interviews with residents in 31 nursing homes and 64 residential care homes. Three hundred and twenty-five residents at baseline were interviewed, and 206 completed the interviews at follow-up. Five outcomes including residents\' physical/mental functional status and subjective health status in Short Form-36 were analyzed using latent growth curve models (LGCMs). Results Only the physical component summary (PCS) had increased significantly. The most influential risk factors to outcomes were the intra-individual-level time-varying variables, including self-rated health and with/without tubing care. Some predictive inter-individual-level factors were also found. For institutional characteristics, small-sized homes (<49 beds) with low occupancy rates showed a lower growth rate in residents\' mental component summary (MCS) and PCS over 1 year and private sector homes showed the most significant growth rates in MCS. Conclusions The methodological strength using LGCMs provides a framework for systematically assessing the influence of risk factors from various levels on residents\' outcomes and follow-up change. It is evident that factors in various levels all influenced residents\' outcomes which support critical information for case mix and quality management in LTC facilities. Under the scenario of a surplus of institutional care in Taiwan, we suggest that institutions must focus more on residents\' psychological well-being and care quality, especially in small-sized homes in relation to the outcomes of its residents.
Eye movements predict students' computer-based assessment performance of physics concepts in different presentation modalities
(2014-05-01)
Chen, Sheng-Chang
;
She, Hsiao-Ching
;
Chuang, Ming-Hua
;
Wu, Jiun-Yu
;
Tsai, Jie-Li
Despite decades of studies on the link between eye movements and human cognitive processes, the exact nature of the link between eye movements and computer-based assessment performance still remains unknown. To bridge this gap, the present study investigates whether human eye movement dynamics can predict computer-based assessment performance (accuracy of response) in different presentation modalities (picture vs. text). Eye-tracking system was employed to collect 63 college students' eye movement behaviors while they are engaging in the computer-based physics concept questions presented as either pictures or text. Students' responses were collected immediately after the picture or text presentations in order to determine the accuracy of responses. The results demonstrated that students' eye movement behavior can successfully predict their computer-based assessment performance. Remarkably, the mean fixation duration has the greatest power to predict the likelihood of responding the correct physics concepts successfully, followed by re-reading time in proportion. Additionally, the mean saccade distance has the least and negative power to predict the likelihood of responding the physics concepts correctly in the picture presentation. Interestingly, pictorial presentations appear to convey physics concepts more quickly and efficiently than do textual presentations. This study adds empirical evidence of a prediction model between eye movement behaviors and successful cognitive performance. Moreover, it provides insight into the modality effects on students' computer-based assessment performance through the use of eye movement behavior evidence. (C) 2014 Elsevier Ltd. All rights reserved.
Indirect Effects of Child Reports of Teacher-Student Relationship on Achievement
(2012-05-01)
Hughes, Jan N.
;
Wu, Jiun-Yu
;
Kwok, Oi-man
;
Villarreal, Victor
;
Johnson, Audrea Y.
The effect of student-reported teacher student relationship quality (TSRQ) on academic motivation and achievement was investigated among a sample of 690 academically at-risk elementary students (52.8% male). Measures of TSRQ, achievement, and motivation were collected annually for 3 consecutive years, beginning when participants were in Grade 2 (24.8%) or Grade 3 (74.6%). Child-reported conflict was stable across the 3 years, whereas warmth declined. Boys and African American students reported greater conflict than did girls and Caucasian and Hispanic students. Girls and African American students reported higher warmth than boys and non-African American students. Using path analysis, the authors tested the hypothesis that measures of student motivation in Year 2 mediated the effects of conflict and warmth in Year 1 on reading and math achievement in Year 3. Child-perceived conflict predicted cross-year changes in teacher-rated behavioral engagement, which, in turn, predicted cross-year changes in reading and math achievement. Math competence beliefs also mediated the effect of child-perceived warmth on math achievement. Effects controlled for stability of measures across time, the within-wave association between measures, and baseline measures of IQ and economic adversity. Implications of findings for improving the academic achievement of students at-risk for school failure are discussed.
Interrelations of Growth in Letter Naming and Sound Fluency in Kindergarten and Implications for Subsequent Reading Fluency
(2017-09-01)
Clemens, Nathan H.
;
Lai, Mark H. C.
;
Burke, Mack
;
Wu, Jiun-Yu
;
交大名義發表
Although letter naming fluency (LNF) and letter sound fluency (LSF) measures are widely available to educators for assessing early literacy skills of kindergarten children, better understanding of the contributions of these skills to reading development can help improve the interpretation of LNF and LSF data for instructional decisions. This study investigated the interrelations of growth in LNF and LSF across the kindergarten year and their unique roles in predicting later reading fluency. Piecewise parallel-process growth models indicated that although LNF and LSF were highly correlated at kindergarten entry, fall LNF status was positively predictive of LSF growth across the fall. Bidirectional effects were present, as initial LSF was also a positive predictor of LNF growth across the fall; however, its effects were not as strong as those of initial LNF on LSF growth. More importantly, both initial status and growth in LNF and LSF were uniquely predictive of first-grade reading fluency, indicating the independent effects of each on subsequent text reading skills. Indirect effects were also observed for kindergarten LNF and LSF growth on reading fluency in second and third grades. Implications for kindergarten assessment and instruction are discussed.
Monitoring Early First-Grade Reading Progress A Comparison of Two Measures
(2014-05-01)
Clemens, Nathan H.
;
Shapiro, Edward S.
;
Wu, Jiun-Yu
;
Taylor, Aaron B.
;
Caskie, Grace L.
This study compared the validity of progress monitoring slope of nonsense word fluency (NWF) and word identification fluency (WIF) with early first-grade readers. Students (N = 80) considered to be at risk for reading difficulty were monitored with NWF and WIF on a 1-2 week basis across 11 weeks. Reading skills at the end of first grade were assessed using measures of passage reading fluency, real and pseudoword reading efficiency, and basic comprehension. Latent growth models indicated that although slope on both measures significantly predicted year-end reading skills, models including WIF accounted for more variance in spring reading skills than NWF, and WIF slope was more strongly associated with reading outcomes than NWF slope. Analyses of student growth plots suggested that WIF slope was more positively associated with later reading skills and discriminated more clearly between students according to successful or unsuccessful year-end reading outcomes. Although both measures may be used to monitor reading growth of at-risk students in early first grade, WIF may provide a clearer index of reading growth. Implications for data-based decision-making are discussed.
Revisit the Effect of Teaching and Learning with Technology
(2013-01-01)
Lee, Yuan-Hsuan
;
Waxman, Hersh
;
Wu, Jiun-Yu
;
Michko, Georgette
;
Lin, Grace
We re-examined the effect of teaching and learning with technology on student cognitive and affective outcomes using the meta-analytic technique. Screening studies obtained from an electric search of databases such as PsyInfo and ERIC resulted in 58 studies (1997-2011). Overall, effect sizes were small to moderate across the cognitive and affective outcome measures. Specific teaching/learning components such as context/making sense, challenging activity, instructional conversation, and joint productivity were associated with effect sizes. Instructional features such as objectives, pattern of student computer use, and type of learning task also moderated effect sizes. Suggestions are made for pre-service teachers and in-service teachers to include these instructional features and teaching strategies in teaching and learning with technology.
Teacher Network of Relationships Inventory: Measurement Invariance of Academically At-Risk Students Across Ages 6 to 15
(2015-03-01)
Wu, Jiun-Yu
;
Hughes, Jan N.
;
教育研究所
;
Institute of Education
We tested the longitudinal measurement invariance of the Teacher Network of Relationships Inventory (TNRI), a teacher-report measure of teacher-student relationship quality (TSRQ), on a sample of 784 academically at-risk students across ages 6 to 15 years by comparing the model for each subsequent year with that of the previous year(s). The TNRI was constructed with 22 items to form 3 correlated factors: Warmth, Conflict, and Intimacy. Cronbach\'s alphas ranged from .87 to .96 across 9 years. Both metric and scalar measurement invariance held for 9 years, indicating that scores on the TNRI have similar meaning across these ages. The TNRI also demonstrated measurement invariance across gender and race/ethnicity. Findings support that the TNRI is an appropriate measure for investigating substantive issues related to developmental changes in TSRQ from early childhood through adolescence, including gender and ethnic/racial differences in TSRQ across these ages. Based on repeated-measures ANOVAs, each scale decreased across the 9 years, although the growth patterns for scales differed somewhat: Conflict had a linearly decreasing pattern, Warmth declined most notably as students make the transition to adolescence, whereas Intimacy scores dropped off noticeably at the transition from early to late childhood. Research limitations and implications for practice are discussed.
The effect of individual differences in the inner and outer states of ICT on engagement in online reading activities and PISA 2009 reading literacy: Exploring the relationship between the old and new reading literacy
(2012-06-01)
Lee, Yuan-Hsuan
;
Wu, Jiun-Yu
;
教育研究所
;
Institute of Education
With the prevalence of ICT, the concept of reading literacy has evolved to encompass both online reading and printed texts. This study clarifies the relationship between reading printed texts and online electronic texts from the perspective of individual differences in the inner and outer phases of ICT in a partial mediation model. We used the PISA 2009 data with 297,295 fifteen-year-old students (49.6% males) across 42 regions. The inner state of ICT represents students' attitude toward computers and confidence in high-level ICT tasks, whereas the outer state of ICT represents students' access to IDT facilities at home or school. The indirect results showed students' reading literacy improved with better attitude toward computers, confidence, and ICT availability at home, as long as the effect was mediated through engagement in online reading activities, even though availability of ICT at home had a direct and negative impact on PISA 2009 reading literacy. (C) 2012 Elsevier Inc. All rights reserved.
The indirect effects of online social entertainment and information seeking activities on reading literacy
(2013-09-01)
Lee, Yuan-Hsuan
;
Wu, Jiun-Yu
;
師資培育中心
;
Center of Teacher Education
Research showed distinct effects of different online activities on reading literacy or learning outcomes; however, no explanation about this link was provided. The current study investigated the effects of two genres of online reading activities on reading literacy based on knowledge of metacognitive strategies in a mediation analysis. Participants were 87,735 fifteen-year-old students (49.8% girls) across 15 regions in the PISA 2009 dataset. We divided online reading activities into social entertainment and information-seeking activities and controlled for gender, socioeconomic status, and the availability of Information and Communication Technologies (ICT) at home and at school. The indirect effects of knowledge of metacognitive strategies helped to explain why social entertainment and information-seeking activities would predict reading literacy differently. More frequent information-seeking activities predicted better knowledge of metacognitive strategies, which in turn predicted better reading literacy, while more frequent social entertainment activities predicted poorer knowledge of metacognitive strategies, which in turn led to poorer reading literacy. Suggestions were made to guide students in engaging in more online information-seeking reading activities, and incorporate instruction of metacognitive strategies for both online and offline reading, thereby improving students' reading literacy in both printed and digital formats. (C) 2013 Elsevier Ltd. All rights reserved.
The indirect relationship of media multitasking self-efficacy on learning performance within the personal learning environment: Implications from the mechanism of perceived attention problems and self-regulation strategies
(2017-03)
Wu, Jiun-Yu
;
教育研究所
;
Institute of Education
Media multitasking, characterized by simultaneous engagement in multiple media forms, is prevalent among university students within the personal learning environment. However, those who think they are capable of multitasking usually overestimate their ability to perform the actual tasks. This study examined university students\' learning performance from the perspectives of their media multitasking self-efficacy, perceived attention problems, and self-regulation strategies using the revised Online Learning Motivated Attention and Regulatory Strategies scale. Participants were 696 university students (275 males, 39.51%) in Taiwan. The author developed the media multitasking self-efficacy scale through open-ended interviews and pilot tested the measures using an exploratory factor analysis. The confirmatory factor analysis verified the uni-factor structure of the instrument. Second-order confirmatory factor analysis validated the two orthogonal higher order constructs of perceived attention problems and self-regulation strategies as well as their subscales. Results from the multilevel structural equation model revealed significant negative indirect relationship between media multitasking self-efficacy and learning performance via both students\' perceived attention problems and self-regulation strategies. Study findings have implications for prevention and intervention of university students\' media-related attention problems and poor regulation strategy use within the personal learning environment. (C) 2016 The Author. Published by Elsevier Ltd.
The predictive validities of individual working-memory capacity profiles and note-taking strategies on online search performance
(1970-01-01)
Wu, Jiun-Yu
;
教育研究所
;
Institute of Education
Online search involves multitasking and may demand better working-memory capacities (WMC) and additional cognitive aids. Given the constraints of human cognition, we tested the effectiveness of note-taking strategies on university students' online search performance. Also examined were the profile configurations of WMC tests in silence and in irrelevant speech as well as its interaction effects with note-taking strategies on online search performance. Among the participants of 60 university students, we found a four-group WMC profile composition that differed significantly across WMC scores in both silence and irrelevant speech. Results showed that WMC profiles, note-taking groups and their interaction effect were significant factors on online search performance. Both Free Note and Matrix Note were beneficial for the students in low silent WMC profile groups. Study results have both theoretical and practical implications, advancing our understanding of the interrelationship of WMC under different conditions and informing instructional practice of online search performance.
University students\' Motivated Attention and use of regulation strategies on social media
(2015-11-01)
Wu, Jiun-Yu
;
教育研究所
;
Institute of Education
The Internet has become a major platform for learning in higher education. Besides rich informational resources, however, the Internet offers an abundance of distractors that challenge students\' attention. This study investigated university students\' perceived attention state and use of regulatory strategies using the Online Learning Motivated Attention and Regulatory Strategies scale (OL-MARS). Participants were 230 undergraduate and graduate students recruited from two universities located in central and northern Taiwan. The exploratory factor analysis revealed four subscales in the OL-MARS, including perceived attention discontinuity, social media notification, behavioral strategies, and mental strategies. Results showed that mental and behavioral strategies were positively associated with criterion variables, including Internet self-efficacy, online search strategies, and final course grades, but negatively correlated with time spent on the Internet and social media. Whereas, perceived attention discontinuity and social media notification mostly had a modest correlational relationship with these validating variables in an opposite direction. Cluster analysis identified five types of profiles: the Motivated Strategic, the Unaware, the Hanging On, the Non-Responsive and the Self-Disciplined. Group membership exhibited mean differences in Internet self-efficacy, online search strategies, final course grades, and time spent on the Internet and social media. The study results validated the constructs in meta-attention for theory development, provided the OL-MARS scale as an effective meta-attention measurement tool to assess university students\' knowledge of attention and regulation of attention, and proposed the specific intervention and attention regulation training for each profile group. (C) 2015 Elsevier Ltd. All rights reserved.
Using Design-Based Latent Growth Curve Modeling With Cluster-Level Predictor to Address Dependency
(2014)
Wu, Jiun-Yu
;
Kwok, Oi-Man
;
Willson, Victor L.
;
教育研究所
;
Institute of Education
The authors compared the effects of using the true Multilevel Latent Growth Curve Model (MLGCM) with single-level regular and design-based Latent Growth Curve Models (LGCM) with or without the higher-level predictor on various criterion variables for multilevel longitudinal data. They found that random effect estimates were biased when the higher-level predictor was not included and that standard errors of the regression coefficients from the higher-level were underestimated when a regular LGCM was used. Nevertheless, random effect estimates, regression coefficients, and standard error estimates were consistent with those from the true MLGCM when the design-based LGCM included the higher-level predictor. They discussed implication for the study with empirical data illustration.
Using SEM to Analyze Complex Survey Data: A Comparison between Design-Based Single-Level and Model-Based Multilevel Approaches
(2012)
Wu, Jiun-Yu
;
Kwok, Oi-man
;
教育研究所
;
Institute of Education
Both ad-hoc robust sandwich standard error estimators (design-based approach) and multilevel analysis (model-based approach) are commonly used for analyzing complex survey data with nonindependent observations. Although these 2 approaches perform equally well on analyzing complex survey data with equal between-and within-level model structures (B. O. Muthen & Satorra, 1995), the performances of these 2 approaches for analyzing multilevel data with unequal between-and within-level structures have not yet been systematically examined. In this study, we extended B. O. Muthen and Satorra's (1995) study by comparing these 2 approaches and an additional model-based maximum model for analyzing multilevel data considering number of clusters, cluster size, intraclass correlation, and the equality of different level structures. The simulation results showed the model-based maximum model generally performed well across conditions. This model is also recommended as an alternative for analyzing nonindependent survey data, especially when the information of the higher level model structure is not known.
Using supervised machine learning on large-scale online forums to classify course-related Facebook messages in predicting learning achievement within the personal learning environment
(2020-01-02)
Wu, Jiun-Yu
;
Hsiao, Yi-Cheng
;
Nian, Mei-Wen
;
教育研究所
;
Institute of Education
This paper demonstrated the use of the supervised Machine Learning (ML) for text classification to predict students' final course grades in a hybrid Advanced Statistics course and exhibited the potential of using ML classified messages to identify students at risk of course failure. We built three classification models with training data of 76,936 posts from two large online forums and applied the models to classify messages into statistics-related and non-statistics-related posts in a private Facebook group. Three ML algorithms were compared in terms of classification effectiveness and congruency with human coding. Students with more messages endorsed by two or more ML algorithms as statistics-related had higher final course grades. Students who failed the course also had significantly fewer messages endorsed by all three ML algorithms than those who passed. Results suggest that ML can be used for identifying students in need of support within the personal learning environment and for quality control of the large-scale educational data.
Using time pressure and note-taking to prevent digital distraction behavior and enhance online search performance: Perspectives from the load theory of attention and cognitive control
(2018-11-01)
Wu, Jiun-Yu
;
Xie, Chen
;
教育研究所
;
Institute of Education
Digital distraction has become a chronic behavior problem for university students that interferes with their attention and learning within personalized learning environment. Applying a randomized control block trial with a counterbalanced experimental design, we investigated the effect of note-taking strategies and time pressure on students' task-irrelevant browsing behaviors and online search performance. Students were randomly assigned to a matrix note, conventional note, or note-free condition and then to one of the four experimental blocks, counterbalancing the order of online search topics and time pressure. Results of the study showed that 44 out of 60 participants (73.33%) conducted at least an irrelevant browsing. Students in the matrix note condition and those under high time pressure exhibited a lower task distraction rate. Moreover, students taking matrix notes demonstrated better online-search performance. Note-taking strategies and task distraction rate were significant predictors of participants' online search performance controlling for their prior knowledge. Study results are consistent with the load theory of attention and cognitive control and have implications to reduce student distraction and improve learning effectiveness.