Search Results

A Quantitative Modeling Approach to Examining High School, Pre-Admission, Program, Certification and Career Choice Variables in Undergraduate Teacher Preparation Programs

Description: The purpose of this study was to examine if there is an association between effective supervision and communication competence in divisions of student affairs at Christian higher education institutions. The investigation examined chief student affairs officers (CSAOs) and their direct reports at 45 institutions across the United States using the Synergistic Supervision Scale and the Communication Competence Questionnaire. A positive significant association was found between the direct report's evaluation of the CSAO's level of synergistic supervision and the direct report's evaluation of the CSAO's level of communication competence. The findings of this study will advance the supervision and communication competence literature while informing practice for student affairs professionals. This study provides a foundation of research in the context specific field of student affairs where there has been a dearth of literature regarding effective supervision. This study can be used as a platform for future research to further the understanding of characteristics that define effective supervision.
Date: December 2007
Creator: Williams, Cynthia Savage
Partner: UNT Libraries

The Relationship of the Learning Styles of High School Teachers and Computer Use in the Classroom

Description: This study sought to determine if the dominant learning styles of high school teachers is related to the amount of time computers are used in the classroom by students. It also examined the types of software used by those teachers, and their levels of technology adoption. Subjects (N=177) were from high schools in a large urban school district. Instrumentation included the Gregorc Style Delineator, a modified version of the Snapshot Survey and the Stages of Adoption of Technology. An ANOVA showed no statistical significance between teachers with different dominant learning styles in the numbers of minutes per week that computers were utilized in their classrooms with students. A chi square test showed no statistical significance in the types of software used in the classrooms of teachers with different dominant learning styles. A chi square test showed no statistical significance in the Stages of Technology Adoption of teachers with different dominant learning styles.
Date: August 2005
Creator: Hunnicutt, Robert Lane
Partner: UNT Libraries

Establishing the utility of a classroom effectiveness index as a teacher accountability system.

Description: How to identify effective teachers who improve student achievement despite diverse student populations and school contexts is an ongoing discussion in public education. The need to show communities and parents how well teachers and schools improve student learning has led districts and states to seek a fair, equitable and valid measure of student growth using student achievement. This study investigated a two stage hierarchical model for estimating teacher effect on student achievement. This measure was entitled a Classroom Effectiveness Index (CEI). Consistency of this model over time, outlier influences in individual CEIs, variance among CEIs across four years, and correlations of second stage student residuals with first stage student residuals were analyzed. The statistical analysis used four years of student residual data from a state-mandated mathematics assessment (n=7086) and a state-mandated reading assessment (n=7572) aggregated by teacher. The study identified the following results. Four years of district grand slopes and grand intercepts were analyzed to show consistent results over time. Repeated measures analyses of grand slopes and intercepts in mathematics were statistically significant at the .01 level. Repeated measures analyses of grand slopes and intercepts in reading were not statistically significant. The analyses indicated consistent results over time for reading but not for mathematics. Data were analyzed to assess outlier effects. Nineteen statistically significant outliers in 15,378 student residuals were identified. However, the impact on individual teachers was extreme in eight of the 19 cases. Further study is indicated. Subsets of teachers in the same assignment at the same school for four consecutive years and for three consecutive years indicated CEIs were stable over time. There were no statistically significant differences in either mathematics or reading. Correlations between Level One student residuals and HLM residuals were statistically significant in reading and in mathematics. This implied that the second stage of ...
Date: May 2002
Creator: Bembry, Karen L.
Partner: UNT Libraries

A Comparison of IRT and Rasch Procedures in a Mixed-Item Format Test

Description: This study investigated the effects of test length (10, 20 and 30 items), scoring schema (proportion of dichotomous ad polytomous scoring) and item analysis model (IRT and Rasch) on the ability estimates, test information levels and optimization criteria of mixed item format tests. Polytomous item responses to 30 items for 1000 examinees were simulated using the generalized partial-credit model and SAS software. Portions of the data were re-coded dichotomously over 11 structured proportions to create 33 sets of test responses including mixed item format tests. MULTILOG software was used to calculate the examinee ability estimates, standard errors, item and test information, reliability and fit indices. A comparison of IRT and Rasch item analysis procedures was made using SPSS software across ability estimates and standard errors of ability estimates using a 3 x 11 x 2 fixed factorial ANOVA. Effect sizes and power were reported for each procedure. Scheffe post hoc procedures were conducted on significant factos. Test information was analyzed and compared across the range of ability levels for all 66-design combinations. The results indicated that both test length and the proportion of items scored polytomously had a significant impact on the amount of test information produced by mixed item format tests. Generally, tests with 100% of the items scored polytomously produced the highest overall information. This seemed to be especially true for examinees with lower ability estimates. Optimality comparisons were made between IRT and Rasch procedures based on standard error rates for the ability estimates, marginal reliabilities and fit indices (-2LL). The only significant differences reported involved the standard error rates for both the IRT and Rasch procedures. This result must be viewed in light of the fact that the effect size reported was negligible. Optimality was found to be highest when longer tests and higher proportions of polytomous ...
Date: August 2003
Creator: Kinsey, Tari L.
Partner: UNT Libraries

Structural Validity and Item Functioning of the LoTi Digital-Age Survey.

Description: The present study examined the structural construct validity of the LoTi Digital-Age Survey, a measure of teacher instructional practices with technology in the classroom. Teacher responses (N = 2840) from across the United States were used to assess factor structure of the instrument using both exploratory and confirmatory analyses. Parallel analysis suggests retaining a five-factor solution compared to the MAP test that suggests retaining a three-factor solution. Both analyses (EFA and CFA) indicate that changes need to be made to the current factor structure of the survey. The last two factors were composed of items that did not cover or accurately measure the content of the latent trait. Problematic items, such as items with crossloadings, were discussed. Suggestions were provided to improve the factor structure, items, and scale of the survey.
Date: May 2011
Creator: Mehta, Vandhana
Partner: UNT Libraries

Participation in a study-abroad program and persistence at a liberal arts university.

Description: This study used a quasi-experimental design with 1,237 students to investigate the association between participation in a study-abroad program and persistence at a liberal arts university. The theoretical basis for the study was Tinto's Theory of Individual Departure. The independent variable of interest, also known as the treatment, was participation in the University of Dallas Rome Program during the sophomore year. The control group consisted of students who were qualified to participate in the Rome Program, but chose not to do so. The dependent variable was the number of fall and spring semesters enrolled as an undergraduate at the University of Dallas post-treatment through spring 2003. Nine variables that measured background characteristics, academic integration, and social integration explained 3.8% of the variation in number of semesters enrolled post-treatment. Participation in the Rome Program explained an additional 4.2%. In all of the statistical measures examined in this study (incremental increase in R2, b weights, adjusted β weights, and structure coefficients), there was evidence of an important positive association between participation in the Rome Program and persistence. Based on the b weight in the regression equation, holding all other variables constant, students who participated in the Rome Program persisted on average .83 semesters longer post-treatment at the University of Dallas than those who did not go to Rome. Of the 1,007 students in this study who went to Rome, 96% were enrolled at the University of Dallas one semester after Rome participation and 91% were still enrolled after two semesters. This compared to 80% and 72%, respectively, for the 230 students in the control group. Of the 674 students in the study who went to Rome and had the opportunity to graduate within 4 years, 79% graduated within 4 years. This compared to 51% for the 123 students in the control group. ...
Date: December 2003
Creator: Young, Denise York
Partner: UNT Libraries

Heard but not seen: Instructor-led video and its effect on learning.

Description: Educators and instructional designers are seeking ways to increase levels of learning. One of the ways this is being done is through cognitive load theory which attempts to reduce cognitive load through a better understanding of working memory and the factors that impact its function. Past studies have found that working memory processes visual and auditory information using separate and non-sharable resources (dual coding theory) and that by properly utilizing multimedia elements, information processing in working memory is more efficient (multimedia learning). What is not known is the effect that instructor-led video, which uses the visual channel but delivers no information, has on the cognitive load of the learner. Further, will the introduction of multimedia elements make the information processing of the learner more efficient? This study examined three ways in which instructional designers may create a more efficient learning environment through a better understanding of multimedia learning. First, by using the theories of multimedia learning, I examined a more efficient use of sensory memory. By minimizing extraneous load, which communication theory calls noise, on working memory through increased utilization of the visual and auditory channels, the effectiveness of instruction was increased. Secondly, the multimedia effect, defined as using visual helps and guides with spoken and written text, was shown to assist working memory in processing new information into existing schema. Last, by using the personalization principle set forth by Clark and Mayer (2008), I used both the video feed and multimedia together to foster a more social or conversational presentation to the learner.
Date: August 2008
Creator: Holder, David E.
Partner: UNT Libraries

Placement in the prekindergarten bilingual and English as a second language programs as a predictor of reading achievement of 3rd grade students.

Description: At the beginning of the 21st century, few challenges for educators compared to that of meeting the academic needs of the growing number of limited English proficient (LEP) students. Divergent views on whether those needs were best met through instruction in the student's first language and English, known as bilingual education, or instruction solely in English, compounded the challenge and led to varied language support programs. The present study looked at the prekindergarten (preK) language support program as a predictor of 3rd grade reading achievement of students with the intention of helping educators understand how best to serve LEP students. The study included an analysis of 3rd grade reading achievement for four groups of students with a primary home language of Spanish who attended bilingual or ESL prekindergarten. Multivariate analysis of variance (MANOVA) followed by descriptive discriminant analysis (DDA) was used to analyze scores from the Iowa Tests of Basic Skills (ITBS) reading test and the Texas Assessment of Knowledge and Skills (TAKS) reading test. No statistically significant difference in 3rd grade reading achievement was found among the four groups at the .05 level. There was, however, a small-to-medium effect size. The MANOVA indicated that the group to which the students belonged accounted for 5.5% of the variance in their scores. The DDA revealed the ITBS explained most of the difference in the group performance. The findings suggest that ESL instruction is a viable option to bilingual instruction for LEP preK students.
Date: December 2004
Creator: Vannoy, Martha
Partner: UNT Libraries

Spatial Ability, Motivation, and Attitude of Students as Related to Science Achievement

Description: Understanding student achievement in science is important as there is an increasing reliance of the U.S. economy on math, science, and technology-related fields despite the declining number of youth seeking college degrees and careers in math and science. A series of structural equation models were tested using the scores from a statewide science exam for 276 students from a suburban north Texas public school district at the end of their 5th grade year and the latent variables of spatial ability, motivation to learn science and science-related attitude. Spatial ability was tested as a mediating variable on motivation and attitude; however, while spatial ability had statistically significant regression coefficients with motivation and attitude, spatial ability was found to be the sole statistically significant predictor of science achievement for these students explaining 23.1% of the variance in science scores.
Date: May 2011
Creator: Bolen, Judy Ann
Partner: UNT Libraries

Bias and Precision of the Squared Canonical Correlation Coefficient under Nonnormal Data Conditions

Description: This dissertation: (a) investigated the degree to which the squared canonical correlation coefficient is biased in multivariate nonnormal distributions and (b) identified formulae that adjust the squared canonical correlation coefficient (Rc2) such that it most closely approximates the true population effect under normal and nonnormal data conditions. Five conditions were manipulated in a fully-crossed design to determine the degree of bias associated with Rc2: distribution shape, variable sets, sample size to variable ratios, and within- and between-set correlations. Very few of the condition combinations produced acceptable amounts of bias in Rc2, but those that did were all found with first function results. The sample size to variable ratio (n:v)was determined to have the greatest impact on the bias associated with the Rc2 for the first, second, and third functions. The variable set condition also affected the accuracy of Rc2, but for the second and third functions only. The kurtosis levels of the marginal distributions (b2), and the between- and within-set correlations demonstrated little or no impact on the bias associated with Rc2. Therefore, it is recommended that researchers use n:v ratios of at least 10:1 in canonical analyses, although greater n:v ratios have the potential to produce even less bias. Furthermore,because it was determined that b2 did not impact the accuracy of Rc2, one can be somewhat confident that, with marginal distributions possessing homogenous kurtosis levels ranging anywhere from -1 to 8, Rc2 will likely be as accurate as that resulting from a normal distribution. Because the majority of Rc2 estimates were extremely biased, it is recommended that all Rc2 effects, regardless of which function from which they result, be adjusted using an appropriate adjustment formula. If no rationale exists for the use of another formula, the Rozeboom-2 would likely be a safe choice given that it produced the greatest ...
Date: August 2006
Creator: Leach, Lesley Ann Freeny
Partner: UNT Libraries

Missing Data Treatments at the Second Level of Hierarchical Linear Models

Description: The current study evaluated the performance of traditional versus modern MDTs in the estimation of fixed-effects and variance components for data missing at the second level of an hierarchical linear model (HLM) model across 24 different study conditions. Variables manipulated in the analysis included, (a) number of Level-2 variables with missing data, (b) percentage of missing data, and (c) Level-2 sample size. Listwise deletion outperformed all other methods across all study conditions in the estimation of both fixed-effects and variance components. The model-based procedures evaluated, EM and MI, outperformed the other traditional MDTs, mean and group mean substitution, in the estimation of the variance components, outperforming mean substitution in the estimation of the fixed-effects as well. Group mean substitution performed well in the estimation of the fixed-effects, but poorly in the estimation of the variance components. Data in the current study were modeled as missing completely at random (MCAR). Further research is suggested to compare the performance of model-based versus traditional MDTs, specifically listwise deletion, when data are missing at random (MAR), a condition that is more likely to occur in practical research settings.
Date: August 2011
Creator: St. Clair, Suzanne W.
Partner: UNT Libraries

Predictive Relationships among Learner Characteristics, Academic Involvement, and Doctoral Education Outcomes

Description: The literature identifies multiple factors pertinent to learner characteristics and learning experiences that may promote doctoral education outcomes, and yet little quantitative research has examined relationships between those factors deemed important in the effectiveness of doctoral education. This study sought to examine predictive relationships among doctoral students’ learner characteristics, their involvement in mentorship and intellectual community, and doctoral education outcomes. Using Astin’s theory of involvement and the literature on signature pedagogies in doctoral education as conceptual guides, a survey instrument was constructed for the purpose of measuring variables identified as relevant to the effective formation of scholars. Central to the conceptualization of this study was academic involvement as represented by mentorship and intellectual community. The instrument was validated in a two-stage pilot testing process and administered to doctoral candidates at three public Texas higher education institutions. Of the 217 participants, the majority were female, White (Non-Hispanic), US citizens, and were pursuing education doctorates. Data were analyzed using multivariate statistical analyses. Reliability and validity estimates indicated psychometric integrity of the 20 observed variables measured to represent the constructs of mentorship and intellectual community. Results indicated that doctoral students’ learner characteristics were not notably predictive of doctoral students’ degree of involvement in mentorship and intellectual community (p < .05, R2 = .23). Doctoral students’ degree of academic involvement was strongly predictive of outcomes (p < .001, R2 = .58), particularly student satisfaction with the doctoral education experience and self-efficacy in conducting various forms of scholarly work. Of this effect, more tangible outcomes such as scholarly productivity and degree progress were not meaningfully related to academic involvement. Regardless of the frequency of academic involvement, students perceived faculty mentorship and intellectual community as very important. The predictive value and perceived importance of faculty mentorship and intellectual community highlight the critical role faculty and peer ...
Date: December 2011
Creator: Anderson, Baaska
Partner: UNT Libraries

Parental decision-making regarding their child's participation in a middle-school talent search.

Description: The present study sought to identify variables that predicted parental decision-making regarding their child's participation in a national gifted and talented identification program for middle school students and subsequent participation in recommended educational options. One hundred sixty-nine parents of students who qualified for either the 2001-2002 or 2002-2003 Duke Talent Identification Program participated in the study. The students were drawn from two large public school districts and six small private schools in a large metropolitan area in the southwestern United States. Both quantitative and qualitative methods were used to identify the variables predictive of parental decision-making regarding talent search participation. Each parent completed a questionnaire consisting of both multiple-choice and open-ended questions. Selected parents participated in structured follow-up interviews. The results of the study indicated that parental perception of the helpfulness of school personnel in explaining the purpose and process of the talent search was most predictive of participation in the talent search. The educational level of the father, parent's prior awareness of the purpose and process of talent search, and the number of enrichment activities in which the child had previously participated were also predictive of talent search participation. Qualitative data indicated that parents of both participants and nonparticipants had a limited understanding of the purpose, diagnostic power, and potential benefits of the talent search. Very few parents chose to seek extracurricular or curricular/instructional options following the talent search testing. Qualitative data indicated that parents did not choose these options due to cost, logistical concerns regarding the special programs, and reservations about the developmental appropriateness of such options for middle school students. Although talent searches are sponsored and administered by organizations outside the local school, this study suggests that parents mostly rely on their local school for notification of their child's nomination, information on the purpose and benefits of ...
Date: May 2005
Creator: Ray, Janet
Partner: UNT Libraries

The Role of Brand Equity in Reputational Rankings of Specialty Graduate Programs in Colleges of Education: Variables Considered by College of Education Deans and Associate Deans Ranking the Programs

Description: Seeking to identify and further understand the variables considered when ranking specialty programs in colleges of education, this research study surveyed all deans, and associate deans responsible for graduate education, at United States institutions that offer the terminal degree in at least one of the ten education specialty areas. The study utilized a three-dimension model of brand equity from the marketing literature, which included the elaboration likelihood model of persuasion. Descriptive statistics determined that research by the faculty of the specialty program is the variable most widely considered by deans and associate deans when determining reputation. In order to determine what predicts a person's motivation to correctly rank programs, a principal components analysis was utilized as a data reduction technique, with parallel analysis determining component retention. The model identified five components which explained 66.224% of total variance. A multiple regression analysis determined that characteristics of a specialty program was the only statistically significant predictor component of motivation to correctly rank programs (&#946; = .317, p = .008, rs2 = .865); however, a large squared structure coefficient was observed on perceived quality (rs2 = .623). Using descriptive discriminant analyses, the study found there is little evidence that marketing efforts have differing effects on groups. Further, a canonical correlation analysis that examined the overall picture of advertising on different groups was not statistically significant at F (15, 271) = .907, p = .557, and had a relatively small effect size (Rc2 = .099).
Date: May 2010
Creator: Lamb, Keith Whitaker
Partner: UNT Libraries

Measuring the Perceived Transfer of Learning and Training for a Customer Service Training Program Delivered by Line Managers to Call Center Employees in a Fortune 200 Financial Services Company

Description: The purpose of this study was to explore what effect manager involvement in the delivery of training has on employee learning (transfer of learning) and on student behavior after training (transfer of training). Study participants were randomly assigned to the experimental and control groups and a customer service training program was delivered with and without manager involvement. Quantitative and qualitative data were collected immediately after training using a retrospective pretest-then/posttest-now instrument developed to measure the participants' perceived transfer of learning. Quantitative and qualitative data were collected approximately 4 weeks after training also using a retrospective pretest-then/posttest-now instrument developed to measure the participants' perceived transfer of training. Quality assurance data generated by the organization for the first full month after the training program was completed were collected to measure the actual transfer of training. A 13-item version of the Marlowe-Crowne Social Desirability Scale (MC-C) was included with the perceived transfer of training survey to measure the potential for self-perception bias with the perceived transfer of learning and the perceived transfer of training data. ANOVA results for the perceived transfer of learning and perceived transfer of training data indicated that there were no statistically significant differences between the experimental and control groups. ANOVA results for the actual transfer of training data mirrored the results found for the perceived transfer of training. The possibility of self-perception bias in using the retrospective pretest-then/posttest-now instruments was recognized as a study concern with MC-C data indicating a much higher level of social desirability with the sample population than with reported non-forensic norms. A slight positive influence on the transfer of learning and on the transfer of training was found when a participant's direct manager was involved in the delivery of training.
Date: December 2006
Creator: Perez, Gustavo A.
Partner: UNT Libraries

Well-Being of Gifted Students Following Participation in an Early-College-Entrance Program

Description: The concepts of well-being and life satisfaction are explored in this study of the experiences and psychological traits of highly-gifted students who have been radically accelerated into an early-college-entrance program. The study was conducted after participation in the early-college-entrance program. The primary focus of the study is on personal well-being and life satisfaction including the variables of subjective well-being, efficacy, and the dispositional traits of cheerfulness, seriousness, and bad mood. These variables are gathered as the initial phase of a longitudinal study of the early-college entrants' personal and professional experiences, their life satisfaction, and dispositions. The subjects for this study were participants in the Texas Academy of Math and Science (TAMS). TAMS is a state run early-college-entrance program at the University of North Texas in Denton.
Date: December 2006
Creator: Boazman, Janette Kay
Partner: UNT Libraries

Stratified item selection and exposure control in unidimensional adaptive testing in the presence of two-dimensional data.

Description: It is not uncommon to use unidimensional item response theory (IRT) models to estimate ability in multidimensional data. Therefore it is important to understand the implications of summarizing multiple dimensions of ability into a single parameter estimate, especially if effects are confounded when applied to computerized adaptive testing (CAT). Previous studies have investigated the effects of different IRT models and ability estimators by manipulating the relationships between item and person parameters. However, in all cases, the maximum information criterion was used as the item selection method. Because maximum information is heavily influenced by the item discrimination parameter, investigating a-stratified item selection methods is tenable. The current Monte Carlo study compared maximum information, a-stratification, and a-stratification with b blocking item selection methods, alone, as well as in combination with the Sympson-Hetter exposure control strategy. The six testing conditions were conditioned on three levels of interdimensional item difficulty correlations and four levels of interdimensional examinee ability correlations. Measures of fidelity, estimation bias, error, and item usage were used to evaluate the effectiveness of the methods. Results showed either stratified item selection strategy is warranted if the goal is to obtain precise estimates of ability when using unidimensional CAT in the presence of two-dimensional data. If the goal also includes limiting bias of the estimate, Sympson-Hetter exposure control should be included. Results also confirmed that Sympson-Hetter is effective in optimizing item pool usage. Given these results, existing unidimensional CAT implementations might consider employing a stratified item selection routine plus Sympson-Hetter exposure control, rather than recalibrate the item pool under a multidimensional model.
Date: August 2009
Creator: Kalinowski, Kevin E.
Partner: UNT Libraries

Attenuation of the Squared Canonical Correlation Coefficient Under Varying Estimates of Score Reliability

Description: Research pertaining to the distortion of the squared canonical correlation coefficient has traditionally been limited to the effects of sampling error and associated correction formulas. The purpose of this study was to compare the degree of attenuation of the squared canonical correlation coefficient under varying conditions of score reliability. Monte Carlo simulation methodology was used to fulfill the purpose of this study. Initially, data populations with various manipulated conditions were generated (N = 100,000). Subsequently, 500 random samples were drawn with replacement from each population, and data was subjected to canonical correlation analyses. The canonical correlation results were then analyzed using descriptive statistics and an ANOVA design to determine under which condition(s) the squared canonical correlation coefficient was most attenuated when compared to population Rc2 values. This information was analyzed and used to determine what effect, if any, the different conditions considered in this study had on Rc2. The results from this Monte Carlo investigation clearly illustrated the importance of score reliability when interpreting study results. As evidenced by the outcomes presented, the more measurement error (lower reliability) present in the variables included in an analysis, the more attenuation experienced by the effect size(s) produced in the analysis, in this case Rc2. These results also demonstrated the role between and within set correlation, variable set size, and sample size played in the attenuation levels of the squared canonical correlation coefficient.
Date: August 2010
Creator: Wilson, Celia M.
Partner: UNT Libraries

Comparisons of Improvement-Over-Chance Effect Sizes for Two Groups Under Variance Heterogeneity and Prior Probabilities

Description: The distributional properties of improvement-over-chance, I, effect sizes derived from linear and quadratic predictive discriminant analysis (PDA) and from logistic regression analysis (LRA) for the two-group univariate classification were examined. Data were generated under varying levels of four data conditions: population separation, variance pattern, sample size, and prior probabilities. None of the indices provided acceptable estimates of effect for all the conditions examined. There were only a small number of conditions under which both accuracy and precision were acceptable. The results indicate that the decision of which method to choose is primarily determined by variance pattern and prior probabilities. Under variance homogeneity, any of the methods may be recommended. However, LRA is recommended when priors are equal or extreme and linear PDA is recommended when priors are moderate. Under variance heterogeneity, selecting a recommended method is more complex. In many cases, more than one method could be used appropriately.
Date: May 2003
Creator: Alexander, Erika D.
Partner: UNT Libraries

Investigating the hypothesized factor structure of the Noel-Levitz Student Satisfaction Inventory: A study of the student satisfaction construct.

Description: College student satisfaction is a concept that has become more prevalent in higher education research journals. Little attention has been given to the psychometric properties of previous instrumentation, and few studies have investigated the structure of current satisfaction instrumentation. This dissertation: (a) investigated the tenability of the theoretical dimensional structure of the Noel-Levitz Student Satisfaction Inventory™ (SSI), (b) investigated an alternative factor structure using explanatory factor analyses (EFA), and (c) used multiple-group CFA procedures to determine whether an alternative SSI factor structure would be invariant for three demographic variables: gender (men/women), race/ethnicity (Caucasian/Other), and undergraduate classification level (lower level/upper level). For this study, there was little evidence for the multidimensional structure of the SSI. A single factor, termed General Satisfaction with College, was the lone unidimensional construct that emerged from the iterative CFA and EFA procedures. A revised 20-item model was developed, and a series of multigroup CFAs were used to detect measurement invariance for three variables: student gender, race/ethnicity, and class level. No measurement invariance was noted for the revised 20-item model. Results for the invariance tests indicated equivalence across the comparison groups for (a) the number of factors, (b) the pattern of indicator-factor loadings, (c) the factor loadings, and (d) the item error variances. Because little attention has been given to the psychometric properties of the satisfaction instrumentation, it is recommended that further research continue on the SSI and any additional instrumentation developed to measure student satisfaction. It is possible that invariance issues may explain a portion of the inconsistent findings noted in the review of literature. Although measurement analyses are a time-consuming process, they are essential for understanding the psychometrics characterized by a set of scores obtained from a survey, or any other form of assessment instrument.
Date: December 2008
Creator: Odom, Leslie R.
Partner: UNT Libraries

Determination of the Optimal Number of Strata for Bias Reduction in Propensity Score Matching.

Description: Previous research implementing stratification on the propensity score has generally relied on using five strata, based on prior theoretical groundwork and minimal empirical evidence as to the suitability of quintiles to adequately reduce bias in all cases and across all sample sizes. This study investigates bias reduction across varying number of strata and sample sizes via a large-scale simulation to determine the adequacy of quintiles for bias reduction under all conditions. Sample sizes ranged from 100 to 50,000 and strata from 3 to 20. Both the percentage of bias reduction and the standardized selection bias were examined. The results show that while the particular covariates in the simulation met certain criteria with five strata that greater bias reduction could be achieved by increasing the number of strata, especially with larger sample sizes. Simulation code written in R is included.
Date: May 2010
Creator: Akers, Allen
Partner: UNT Libraries

Dimensions of Social Network Position As Predictors of Employee Performance.

Description: Research of social networks has revealed that certain components of network position can have an impact on organizational effectiveness, yet relatively little research has been conducted on network position and individual performance. This study sought to determine if a relationship exists between an employee's social network position and an individual's job performance. The participant organization was a network of individuals within an Information Technology (IT) department at a major defense company. A social network analysis (SNA) was conducted to determine the employee's network position, measured by centrality and constraint. Centrality refers to the extent to which an individual is connected to others. Constraint refers to how constrained or inhibited an individual is within the network. Performance was measured by annual appraisal ratings provided by the employee's supervisor. Hierarchical regression analysis was performed to determine relationships between the dependent variable (performance) and independent variables of centrality and constraint. Secondary variables also studied in relation to the model included education level, service years (tenure), job grade, and age. The overall model revealed 17% of variance explained. The primary predictors of network position, centrality and constraint, were not statistically significant predictors of performance ratings. Three variables, job grade, tenure and age, were found to be statistically significant predictors of employee performance. Further research is suggested to provide additional insight into the predictive value of these variables.
Date: August 2007
Creator: Burton, Paul
Partner: UNT Libraries

A Comparison of Five Robust Regression Methods with Ordinary Least Squares: Relative Efficiency, Bias and Test of the Null Hypothesis

Description: A Monte Carlo simulation was used to generate data for a comparison of five robust regression estimation methods with ordinary least squares (OLS) under 36 different outlier data configurations. Two of the robust estimators, Least Absolute Value (LAV) estimation and MM estimation, are commercially available. Three authormodified variations on MM were also included (MM1, MM2, and MM3). Design parameters that were varied include sample size (n=60 and n=180), number of independent predictor variables (2, 3 and 6), outlier density (0%, 5% and 15%) and outlier location (2x,2y s, 8x8y s, 4x,8y s and 8x,4y s). Criteria on which the regression methods were measured are relative efficiency, bias and a test of the null hypothesis. Results indicated that MM2 was the best performing robust estimator on relative efficiency. The best performing estimator on bias was MM1. The best performing regression method on the test of the null hypothesis was MM2. Overall, the MM-type robust regression methods outperformed OLS and LAV on relative efficiency, bias, and the test of the null hypothesis.
Access: This item is restricted to UNT Community Members. Login required if off-campus.
Date: August 2001
Creator: Anderson, Cynthia, 1962-
Partner: UNT Libraries

Convergent Validity of Variables Residualized By a Single Covariate: the Role of Correlated Error in Populations and Samples

Description: This study examined the bias and precision of four residualized variable validity estimates (C0, C1, C2, C3) across a number of study conditions. Validity estimates that considered measurement error, correlations among error scores, and correlations between error scores and true scores (C3) performed the best, yielding no estimates that were practically significantly different than their respective population parameters, across study conditions. Validity estimates that considered measurement error and correlations among error scores (C2) did a good job in yielding unbiased, valid, and precise results. Only in a select number of study conditions were C2 estimates unable to be computed or produced results that had sufficient variance to affect interpretation of results. Validity estimates based on observed scores (C0) fared well in producing valid, precise, and unbiased results. Validity estimates based on observed scores that were only corrected for measurement error (C1) performed the worst. Not only did they not reliably produce estimates even when the level of modeled correlated error was low, C1 produced values higher than the theoretical limit of 1.0 across a number of study conditions. Estimates based on C1 also produced the greatest number of conditions that were practically significantly different than their population parameters.
Date: May 2013
Creator: Nimon, Kim
Partner: UNT Libraries