An important goal of teacher education programs is to encourage preservice teachers to develop their understanding of student learning and instruction. One way to determine the possible impact of teacher education programs on preservice teachers' understanding of student learning and instruction is to pay attention to preservice teachers' mental models of learning and instruction. Here, I define preservice teachers' mental models as a set of conceptions about learning that represent how students' minds work and how teachers can influence students' minds through instruction (Strauss, 2001). In this study, I qualitatively examined how the mental models of eight preservice, social science teachers developed over a period of one year in a teacher education program. To investigate change over time, I collected three types of data: lesson plans, written rationales, and interviews. I qualitatively analyzed the data and derived a set of relationships among the major categories of instruction, student learning, and learning outcomes. Results were graphically displayed in tables and as consensus models for the group of eight preservice teachers. Findings from this study inform teacher education in several ways. First preservice teachers' mental model development was both incremental and sustained over time. Secondly, preservice teachers' mental models became more complex over time, but many of their earlier conceptions did not disappear as a result of learning. Thirdly, preservice teachers added domain-specific pedagogical knowledge to their mental models as they progressed through their teacher education program. Finally, the development of their mental models occurred in conjunction with their acquisition of a professional vocabulary as well as an increased understanding of lesson planning and assessment. I discuss the findings of this study with respect to teacher education. Specifically, I address mental models and their relationship to instructional planning, role development, and the nature of conceptual change over time. Limitations of this study and directions for future research are described. / A Dissertation Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Summer Semester, 2008. / May 16, 2008. / Teacher Education, Mental Models / Includes bibliographical references. / Susan Carol Losh, Professor Directing Dissertation; Sande D. Milton, Outside Committee Member; Alysia Roehrig, Committee Member; Jeannine E. Turner, Committee Member.

The purpose of this study was to examine effects of enhanced student ratings feedback on instructional practices, teaching effectiveness and student motivation. Sixty high school teachers from three high schools in Turkey participated in this study. Three high schools consisting of twenty teachers each were randomly assigned to no-feedback, feedback-only and enhanced feedback group. Teachers in the enhanced feedback group were provided with their mean scores on each item of the Instructional Activities Feedback Form (IAFF) along with detailed description of each activity and suggestions as to how they might improve their instructional activities. Teachers in the feedback-only group just received numerical average scores on each item of the IAFF. Final student rating results revealed significant differences (p < .05) in favor of the enhanced feedback group on teaching practices, teaching effectiveness and student motivation. Teachers also valued the enhanced feedback process and their perceptions on receiving feedback from their students were very positive. These findings suggest that enhanced student ratings feedback can be used to improve teachers' teaching practices. Suggestions for future research and implications for using enhanced feedback with IAFF are also discussed. / A Dissertation Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Spring Semester, 2006. / May 26, 2005. / Teaching Effectiveness, Student Motivation, Instructional Practices, Teacher Evaluation, Student Ratings / Includes bibliographical references. / John M. Keller, Professor Directing Dissertation; Janice L. Flake, Outside Committee Member; Amy Baylor, Committee Member; Akihito Kamata, Committee Member.

Students benefit from collaborative learning activities, but they do not automatically reach desired learning outcomes when working together (Fischer, Kollar, Mandl, & Haake, 2007, King, 2007). Learners need instructional support to increase the quality of collaborative processes and individual learning outcomes. The core challenge is to find the right type and amount of instructional guidance to ensure that learners accomplish the intended learning goals. Instructional designers have to make sure students stay on track and do not have superficial and off-topic discussions. At the same time, students need enough freedom for the development of valuable processes that evolve during an ongoing conversation. This means that researchers need to examine the impact of external structure (instructions) on the various elements of the learning process to design learning environments with an optimal extent of instructional support. Examining the learning process is essential because simply looking at students' outcome will not provide sufficient information about the effects of instructional guidance. The purpose of this study was to examine the effects of different levels of instructional guidance on students' discussions and outcome in a simulation-based collaborative learning environment. Instructional guidance took the form of computer-supported collaborative learning (CSCL) scripts. A CSCL script is a set of instructions aimed at structuring the learning process. Scripts specify, sequence, and assign activities to collaborative learners (Weinberger, 2003). The main goal of this dissertation was to contribute to the continuing development of effective instructional strategies in computer-supported collaborative settings. A simulation-based learning environment called school efficiency simulation (SEsim), was employed which required learners to collaboratively develop solutions to a complex, authentic problem. SEsim had been designed and developed as a training and research tool that focused on the quality development of schools and included instructional materials, communication tools, and a simulation. The learners' assignment was to reduce the dropout rate of a school and thus to improve its efficiency. Each round of the game simulated one school year and the simulation provided immediate feedback about the learner's performance. Forty five students from two classes from a German gymnasium (a school that is equivalent to an American high school) specializing in pedagogy participated in the study. To experimentally examine the effects of the two different CSCL scripts, students were randomly assigned to two treatment conditions implemented during the collaborative learning task. Students operating under the precise instructional guidance (PIG) treatment condition were assigned roles throughout the discussion and received precise instructions regarding the number and nature of messages to be posted. The other groups, i.e., the general instructional orientation (GIO) groups, received more general instructions and students were not asked to represent a certain role. A mixed methods research approach was used in this study to explore how the scripts would impact students' discussions and the effects on individual performance. The primary focus was set on analyzing students' discussions to explore the scripts' effects on the learning process. The rich qualitative data allowed for obtaining a detailed understanding about the students' reaction to the given instructions. The secondary purpose was set on examining a different aspect of the research problem, namely the effects of the two scripts on students' performance. First, the students' discussions were analyzed to examine how the two scripts would impact students' conversations. Also, two groups, one from the GIO treatment condition and one from the PIG treatment condition, were selected for further analysis in the form of case studies. The purpose of the data analysis was to explore what was going on in students' discussions and to allow for comparison of the two treatment conditions (PIG and GIO) with regard to the nature of students' interactions. How closely the students discussed according to the given directions in the precise instructional guidance (PIG) script was also examined. Results showed that students from the PIG groups partially followed the precise directions and for the most part represented assigned roles. Structural differences between the treatment groups were discovered. Treatment effects reflected by differences in interaction patterns were found: Students in the GIO groups more frequently asked questions and exchanged superficial information whereas students from the PIG groups made suggestions supported by personal beliefs, experiences, and information from the learning materials. Furthermore, the data analysis revealed that the two groups from the case studies employed different problem solving approaches. Second, the effects of the CSCL scripts on students' performance were examined. An independent samples t-test was used to test the hypothesis that students with precise instructional guidance (PIG treatment condition) perform better on the posttest than students receiving general instructions (GIO treatment condition). Results showed that students in the precise instructional guidance groups significantly outperformed the general instructional orientation treatment groups regarding the first outcome variable, the simulation score. There were no differences regarding the second outcome variable, the causal model similarity score. Possible explanations for the results are discussed in relation to findings from related research. Limitations of the present study are presented along with its implications. Future research will include further development of the learning environment focusing on design of adequate assessment techniques and other factors, such as motivation, which may have an impact on students' performance. / A Dissertation Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Spring Semester, 2011. / March 17, 2011. / Collaboration Scripts, Collaborative Learning, Collaboration, CSCL / Includes bibliographical references. / Vanessa Dennen, Professor Directing Dissertation; Jay Rayburn, University Representative; Valerie Shute, Committee Member; Dirk Ifenthaler, Committee Member; Philip Grisé, Committee Member.

The purpose of this study was to examine the impact of an intervention – self-generated analogy instruction – on the situational interest (interest generated primarily by certain conditions, stimuli, and/or concrete objects in the environment that focus attention), individual interest (a personal form of interest that is specific to the individual, relatively stable and develops over time) and intrinsic motivation to learn (a drive to engage in the learning process or school-related activities for their own sake) of students who are at-risk of dropping out of high school. Survey research on high school dropouts identified dropouts' "dislike of school" as the most common reason attributed to their leaving school. In addition, qualitative studies examining reasons dropouts give for leaving traditional school settings, showed that students' dislike of school often translated into feelings of classroom boredom. In combating classroom boredom, dropout prevention researchers suggest that educators and researchers working with at-risk students try current, cognitively based methods of motivating students. In this study, self-generated analogy instruction based on Generative Learning Theory, was presumed to be a potentially effective classroom teaching tool that exhibits recommended characteristics for meeting the instructional needs of at-risk students. In this study self-generated analogy instruction was introduced within two intact high-school biology classrooms, and its effectiveness was investigated through a mixed method research design, comprised of both qualitative and quantitative techniques for collecting data. Results showed that the level of situational interest, individual interest and motivation to learn increased for most students and provide evidence to support interest and motivation theories which suggest that a positive relationship exists between the three variables. In addition, the study suggests that pre-service and veteran teachers receive training / professional development with respect to effective cognitive strategies that may promote at-risk students' situational interest, individual and motivation to learn in the classroom with the prospect of a decrease in potential dropout behavior. / A Dissertation Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Fall Semester, 2005. / July 29, 2005. / Interest Research, Self-Generated Analogies, School Motivation to Learn, Dropout Prevention / Includes bibliographical references. / Marcy Driscoll, Professor Co-Directing Dissertation; Alysia Roehrig, Professor Co-Directing Dissertation; Sherry Southerland, Outside Committee Member; Akihito Kamata, Committee Member.

Multivariate meta-analysis refers to the statistical analyses of a collection of studies where at least some studies provide multiple effect-size estimates that may or may not represent multiple constructs. Multiple endpoint studies typically involve research designs where individuals in one treatment group and one control group produce measures on multiple variables. These types of studies will likely lead to statistically dependent effect sizes. The dependence that arises from multiple endpoint studies in the meta-analytic framework has not been thoroughly studied. The main purpose of this thesis was to investigate the impact of dependence from multiple endpoint studies utilizing homogeneity measures commonly found in current meta-analyses. This thesis is comprised of two sections: simulation and model estimation. Both sections replicated 3,000 meta-analyses for varying conditions. The simulation section varied study sample size, number of studies, between-outcomes correlation, and dependency structure. The model estimation section used generalized least squares estimation to analyze many of the same conditions. The standardized mean difference was the utilized effect-size estimator and Type I error rates of Q statistics were the primary unit of analysis. Results showed that increased dependence among effect sizes is associated with increased Type I error rates of Q statistics. More specifically, under very dependent conditions, Type I error rates were significantly greater than their nominal levels regardless of within-study sample size and number of studies, sometimes with more than a twofold inflation. The model estimation section demonstrated that using generalized least squares estimation to account for multiple endpoint dependency maintains Type I error rates within nominal levels and is preferable to incorrectly assuming independence. / A Thesis Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Summer Semester, 2011. / June 13, 2011. / Homogeneity, Meta-analysis, Multiple endpoints / Includes bibliographical references. / Betsy Becker, Professor Directing Thesis; Yanyun Yang, Committee Member; Daniel McGee, Committee Member.

Many college athletes continue to demonstrate low levels of motivation and achievement in the classroom (Simons, Van Rheenen, & Covington, 1999). Institutions of higher learning do not seem to be able to devise effective strategies that would resolve this situation. Collegiate athletes present a "motivational contradiction," in that many are motivated to succeed in the athletic domain yet, seem to lack the motivation to learn and achieve at high levels in the classroom. Perhaps this is due to the arrival of many athletes at college campuses in a state of low readiness for academic challenge. Mitchell and Piatrowka's (1974) research indicates that under-prepared students are likely to have deficits in motivation and are more likely to be distracted in their academic tasks than their successful peers. Although all students should be beneficiaries of academic support, campus athletes may comprise a group of clients that are in particular need of such assistance. The role of a collegiate student-athlete entails a variety of expectations as well as special challenges (Ervin, Saunders, Gillis, & Hoprabo, 1985; Parham, 1993). As the term suggests, student-athletes are expected to fulfill a dual role, that of student as well as athlete (Adler & Adler, 1987; Parham, 1993). Many athletes entering collegiate sport are academically and socially unprepared (Bruno, Holland, & Ward, 1988; Staurt 1985). If proper services are not provided to guide them, many succumb to a multitude of powerful stressors on the playing field, in the classroom, and in their social lives (Felston & Wilcox, 1993). Results of this thesis research should be helpful to those who counsel and advise student-athletes. Therefore one purpose of this study was is to provide information, which should help clarify the perceived intensity, frequency and nature of stressors encountered by university student-athletes. A second purpose of equal importance was to determine perceptions held by the study's athlete participants about the university provided support services designed to help the manage their stressors. Results of this research indicate that athletes at Florida State University view support service as comprehensive and helpful. The staff members who provide services are perceived as being well prepared. In response to the first hypothesis GPA was not found to correlate with student-athlete's overall perception of effectiveness about the academic services. The study also discovered that there were no statistically significant correlation between high school GPA and perceptions about effectiveness of academic support services and only a slight low and negative correlation between high school GPA and amount of study hall hours attended per week. Contrary to what hypothesis four predicted there was no difference in perceptions between race, sport, gender, and revenue and non-revenue sports. The result of the last hypothesis stated that regardless of gender, stress appraisal was not significantly correlated with amount of study hall hours or perceptions about the services. Studies such as this one should be conducted every few years to make sure that the service effectiveness is being maintained. / A Thesis Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Fall Semester, 2003. / October 10, 2003. / Student-Athletes, Tutorial Services, Life Stresors / Includes bibliographical references. / David Pargman, Professor Directing Thesis; Gershon Tenenbaum, Committee Member; Linda Schrader, Committee Member.

The experimental design utilized in this study tested the decision-making and reported cue utilization among poker players. Forty-five participants, 15 in each group, comprised expert, intermediate, and novice poker players. Subjects completed the Computer Poker Simulation Task (CPST), comprised of 60 hands of the No-Limit Texas Hold 'Em. During the CPST, participants engaged in a think-aloud protocol in which they literally "thought out loud" throughout the hand. The 60 hands were broken into six separate conditions, two time conditions nested within three style conditions. All hole cards for the participant and computer players, and all community cards, i.e. the flop, turn, and river, were identical across players. The DM performance measures of Expected Value (EV) and profit were subjected to Repeated Measures ANOVAs (RM-ANOVA). The think-aloud protocol was coded to determine if there were skill-level differences in reported cue utilization, using chi square analysis. RM-ANOVAs revealed several significant (p Experts displayed higher EV scores than novices at all stages of play. In addition, expert players outperformed intermediate players at the pre-flop and flop stages. Also, intermediate player's EV scores were higher than novice player's EV scores on the turn. Intermediates scores were higher than novices on the river, but this difference was not significant. All skill-level groups displayed higher EV and profit scores in the timed condition. Novices displayed a large difference in EV scores between the non-timed and timed conditions, while expert and intermediate players displayed only small differences. Experts and novices exhibited large differences in profit scores between the timed and non-timed conditions, with novices displaying the greatest difference. Intermediates displayed moderate differences in profit scores, but these differences were modest in comparison to novices and experts. In the non-timed condition, expert players displayed higher EV scores than novice players at all stages of play, other than the flop. Intermediate players' EV scores were higher than novice players' EV scores on the turn and river. In the timed condition, expert players outperformed novice players in EV scores on the pre-flop, flop, and turn stages. Expert players also displayed higher EV scores than intermediate players on the pre-flop and flop stages. To examine differences in EV scores between the time and non-timed conditions for each skill group by stage of play, effect sizes were calculated. Experts displayed larger EV scores in the timed condition in the flop and turn stages but performed better on the river stage in the non-timed condition. Intermediates displayed higher EV scores in the flop and turn stages in the timed condition. In the turn and river stages, novices displayed a large difference in EV scores between the timed and non-timed conditions, in favor of the timed condition. Considering the think-aloud protocol, expert players reported processing more thoughts than intermediate players, who reported processing more thoughts than novice players. The majority of reported thoughts processed by expert players were of the "Opponent Behavior" and "Advanced Poker Considerations" nature. Intermediate players reported focusing the majority of their attention on three categories: "Opponent Behaviors", "Basic Poker Considerations", and "Advanced Poker Considerations". The majority of reported thoughts processed by novice players were from the "Basic Poker Considerations" and "Other" categories. Overall, expert and intermediate poker players outperformed novice players in DM performance. This difference was largest at later stages of the hand. This can be attributed to greater attendance to the most relevant stimuli, which has the best chance of triggering a correct response (Alain, 1991), and enables prediction of later events allowing subsequent behavior to be planned (Eccles et al., 2002). Experienced players outperformed novices in all styles conditions, which infers that the ability of experts to make better decisions is useful against any opponent style. In the timed condition, contrary to research in other areas, the novices displayed the greatest increase in DM performance. Experts have been found to activate higher-level complex strategies when they had to plan several actions (Poplu et al., 2003). In the non-timed condition, experts and intermediates were given the time required to engage in more complex DM strategies and evaluative DM processes. However, in the timed condition, the initial heuristically-based decisions may have been distorted, and due to time constraints evaluative processes could not occur. / A Dissertation Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Fall Semester, 2009. / November 4, 2009. / Decision-making, Think-aloud Protocol, Gambling, Thought Processes, Poker / Includes bibliographical references. / Gershon Tenenbaum, Professor Directing Dissertation; Michael Mondello, University Representative; Robert Eklund, Committee Member; David Eccles, Committee Member.

Previous research has revealed that team experience was strongly associated with a greater portion of SK (MacMillan, Entin, & Serfaty, 2004) and a highly developed shared mental model enhanced team performance (Mathieu, Goodwin, Heffner, Salas, & Cannon-Bowers, 2000; Carley, 1997; Stout, Cannon-Bowers, Salas, & Milanovich, 1999; Wittenbaum, Vaughan, & Stasser, 1998). However, the SK-performance relationship has not yet been established in the team sport context. This study examined the effects of playing position and experience on shared knowledge (SK). An additional purpose was to examine SK's effect on performance in a sample of high school basketball players. At the end of their season, participants completed the Demographic Information, Basketball Experience, and Structured Training Questionnaire, the General Shared Knowledge Test, the Diagram Based Shared Knowledge Test, and the Video Based Shared Knowledge Test over the course of four assessment sessions. The present study anticipated that teammates that interact more often will share more knowledge (i.e. guards would share more knowledge with guards and posts would share more knowledge with posts), teammates (dyads and triads) and teams that have played together for longer periods of time will display more SK, and teams that display more SK will be more coordinated, and subsequently, will display higher performance levels than teams with less SK. A MANOVA followed by an ANOVA was employed to elicit the effect of experiential variables on SK. An RM-ANOVA followed by an ANOVA was performed to elicit the effects of SK on team performance scores. The first hypothesis, where teammates interacting more often were expected to display more SK was not supported, while the second and third hypotheses, which asserted greater experience (dyads/triads and team) was associated with greater SK, were supported. Also, these findings supported the fourth hypothesis, where teams with higher SK perform at a higher level, shown by a greater statistical performance. As teams gain experience working together (i.e., practicing and competing), their SK increases, and subsequently their operations are more coordinated (McIntyre & Salas, 1995). If teammates share knowledge, they can anticipate their teammates' behaviors and act accordingly, thus aiding to team coordination (Stout, Cannon-Bowers, Salas, & Milanovich, 1999; Wittenbaum, Vaughan, & Stasser, 1998). The results concurred with previous research that, in basketball, as teams practice and play together team familiarity increases, offensive and defensive strategy becomes more ingrained, and each member gains knowledge of each other's strengths, styles, habits, and preferences. The increased team SK allows teammates to better coordinate each other's actions and adapt to the changing basketball environment. Subsequently, increased coordination results in greater team performance. / A Thesis Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Fall Semester, 2005. / November 2, 2005. / Team coordination, Team communication / Includes bibliographical references. / Gershon Tenenbaum, Professor Directing Thesis; David Eccles, Committee Member; Tristan Johnson, Committee Member.

The purpose of this study was to compare the effects of associative and dissociative intervention strategies upon perceived exertion in undergraduate and graduate female students (n = 13, 18 - 24 years of age) exercising on stationary bicycles. Participants had some experience with cycling, as they were recruited from physical education spinning classes at a large southeastern university. They were assigned to each treatment condition, and performed the same 10-minute cycling task for four consecutive weeks. The order of treatment was randomized to discourage order effects. Interventions applied were derived from Stevinson and Biddle's (1999) two-dimensional coping strategy model. The first dimension was task relevance, comprised of associative (e.g. bodily sensations, pace) and dissociative (e.g. daydreams, environmental distractions) processes. The second dimension was direction of attention, i.e. internal or external. Therefore, the model yields four coping strategy types: internal association, external association, internal dissociation, and external dissociation. It was hypothesized that (1) participants in the internal and external dissociation conditions would report lower RPE than participants in the internal and external associative conditions, (2) there would no difference between the internal and external associative conditions, and (3) there would be no difference between the internal and external dissociative conditions. The first session was the same for all participants, consisting of a sub maximal multiple-stage test aimed at assessing the participants VO2 max and corresponding heart rates. The following four sessions each introduced a different intervention, but maintained an equivalent physical load. The task required the participants to ride an exercise bicycle for a total of 20-minutes during each session: comprised of a five-minute warm-up, a 10-minute physical task, and a five-minute cool down. The participants were asked to maintain a 75% HRmax range throughout the 10-minute cycling task. In order to examine the three hypotheses, a repeated measures ANOVA with one within subject factor (treatment condition) was used to analyze the data. The dependent variable is RPE as reported by the participants. Two main effects were observed: treatment condition and time. One interaction effect was also found: treatment by time. Results yielded significant differences (p < .01) between the associative and dissociative treatments. The associative treatments provided higher perceived exertion levels than the dissociative treatments for the same physical load. Therefore, task relevance may affect perceived exertion. However, the differences in perceived exertion were not significant between the internal association and external association (p = .22) and the internal dissociation and external dissociation treatments (p = .99). These findings suggest that the directional dimension of the model may not affect the perceived exertion of endurance tasks as much as task relevance. / A Thesis Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Master of Science. / Spring Semester, 2004. / March 25, 2004. / Perceived exertion, Cyclists, Female, Coping strategies, Association, Dissociation / Includes bibliographical references. / David Pargman, Professor Directing Thesis; Gershon Tenenbaum, Committee Member; Akihito Kamata, Committee Member.

Recent developments in multilevel modeling made it possible to model the relationships between item properties and examinee properties within the multilevel and structural equation modeling framework. In this study, the performance of the multilevel two parameter logistic item response model (2PL IRM) was investigated for estimating item difficulty and discrimination parameters and equating a test among different grade levels under the presence of differential item functioning (DIF) by using real and simulation data. A statewide data, designed for vertical scaling, were used with three different adjacent grade levels. The data were collected for the Florida Comprehensive Achievement Test in 2001. In addition, simulated data comparable to large-scale assessment data from two grade levels were analyzed to control for conditions of different numbers of DIF items. The performance of 2PL IRMs with modeling of the DIF and inclusion of an examinee-level variable was compared with traditional IRT for the development of a vertical scale. It was found that 2PL IRM without any DIF parameter produced the same item difficulty and discrimination parameters. Furthermore, 2PL IRM generated the same scale score as traditional IRT. The inclusion of an examinee variable (grade level) in 2PL IRM produced a better vertical scale in comparison to 2PL IRT. The modeling of nonuniform DIF for some of the anchor items, in addition to the examinee-level variable, resulted in the same scale as the previous model; however, the modeling of uniform DIF for some of the anchor items distorted the vertical scale. A small simulation study was designed to investigate the effects of DIF items on vertical equating with respect to presence of uniform, nonuniform, and both nonuniform and uniform DIF exhibiting on some of anchor items. It was found that distortion of the scale increased as the number of nonuniform DIF items increased in the anchor set. The scale distortion got larger than the effects of having one type of DIF when items in the anchor set had both types of DIF at the same time. There was one conflicting result: Increasing the number of uniform DIF items in an anchor set decreased the scale distortion when only uniform DIF items were present. However, this could have been the result of random error due to the limited simulation size. There was one drawback of multilevel IRM in using the large-scale assessment data. The computation time needed to complete the calibration process was far beyond practicality for a comprehensive state testing program. However, multilevel IRM potentially provides more flexibility for investigating the dimensions that affect the success. Directions for future research and limitations are also discussed. / A Dissertation Submitted to the Department of Educational Psychology and Learning Systems in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy. / Spring Semester, 2006. / January 20, 2006. / FCAT, Developmental Scale, Differential Item Functioning, Vertical Equating, Multilevel Item Response Model, GLLAMM / Includes bibliographical references. / Akihito Kamata, Professor Directing Dissertation; Colleen Kelley, Outside Committee Member; Richard Tate, Committee Member; Albert Oosterhof, Committee Member.