UNT Libraries - Browse

ABOUT BROWSE FEED

Ability Estimation Under Different Item Parameterization and Scoring Models

Description: A Monte Carlo simulation study investigated the effect of scoring format, item parameterization, threshold configuration, and prior ability distribution on the accuracy of ability estimation given various IRT models. Item response data on 30 items from 1,000 examinees was simulated using known item parameters and ability estimates. The item response data sets were submitted to seven dichotomous or polytomous IRT models with different item parameterization to estimate examinee ability. The accuracy of the ability estimation for a given IRT model was assessed by the recovery rate and the root mean square errors. The results indicated that polytomous models produced more accurate ability estimates than the dichotomous models, under all combinations of research conditions, as indicated by higher recovery rates and lower root mean square errors. For the item parameterization models, the one-parameter model out-performed the two-parameter and three-parameter models under all research conditions. Among the polytomous models, the partial credit model had more accurate ability estimation than the other three polytomous models. The nominal categories model performed better than the general partial credit model and the multiple-choice model with the multiple-choice model the least accurate. The results further indicated that certain prior ability distributions had an effect on the accuracy of ability estimation; however, no clear order of accuracy among the four prior distribution groups was identified due to an interaction between prior ability distribution and threshold configuration. The recovery rate was lower when the test items had categories with unequal threshold distances, were close at one end of the ability/difficulty continuum, and were administered to a sample of examinees whose population ability distribution was skewed to the same end of the ability continuum.
Date: May 2002
Creator: Si, Ching-Fung B.

The Analysis of the Accumulation of Type II Error in Multiple Comparisons for Specified Levels of Power to Violation of Normality with the Dunn-Bonferroni Procedure: a Monte Carlo Study

Description: The study seeks to determine the degree of accumulation of Type II error rates, while violating the assumptions of normality, for different specified levels of power among sample means. The study employs a Monte Carlo simulation procedure with three different specified levels of power, methodologies, and population distributions. On the basis of the comparisons of actual and observed error rates, the following conclusions appear to be appropriate. 1. Under the strict criteria for evaluation of the hypotheses, Type II experimentwise error does accumulate at a rate that the probability of accepting at least one null hypothesis in a family of tests, when in theory all of the alternate hypotheses are true, is high, precluding valid tests at the beginning of the study. 2. The Dunn-Bonferroni procedure of setting the critical value based on the beta value per contrast did not significantly reduce the probability of committing a Type II error in a family of tests. 3. The use of an adequate sample size and orthogonal contrasts, or limiting the number of pairwise comparisons to the number of means, is the best method to control for the accumulation of Type II errors. 4. The accumulation of Type II error is irrespective of distributions.
Date: August 1989
Creator: Powers-Prather, Bonnie Ann

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.

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

The Characteristics and Properties of the Threshold and Squared-Error Criterion-Referenced Agreement Indices

Description: Educators who use criterion-referenced measurement to ascertain the current level of performance of an examinee in order that the examinee may be classified as either a master or a nonmaster need to know the accuracy and consistency of their decisions regarding assignment of mastery states. This study examined the sampling distribution characteristics of two reliability indices that use the squared-error agreement function: Livingston's k^2(X,Tx) and Brennan and Kane's M(C). The sampling distribution characteristics of five indices that use the threshold agreement function were also examined: Subkoviak's Pc. Huynh's p and k. and Swaminathan's p and k. These seven methods of calculating reliability were also compared under varying conditions of sample size, test length, and criterion or cutoff score. Computer-generated data provided randomly parallel test forms for N = 2000 cases. From this, 1000 samples were drawn, with replacement, and each of the seven reliability indices was calculated. Descriptive statistics were collected for each sample set and examined for distribution characteristics. In addition, the mean value for each index was compared to the population parameter value of consistent mastery/nonmastery classifications. The results indicated that the sampling distribution characteristics of all seven reliability indices approach normal characteristics with increased sample size. The results also indicated that Huynh's p was the most accurate estimate of the population parameter, with the smallest degree of negative bias. Swaminathan's p was the next best estimate of the population parameter, but it has the disadvantage of requiring two test administrations, while Huynh's p index only requires one administration.
Date: May 1988
Creator: Dutschke, Cynthia F. (Cynthia Fleming)

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.

Comparison of Methods for Computation and Cumulation of Effect Sizes in Meta-Analysis

Description: This study examined the statistical consequences of employing various methods of computing and cumulating effect sizes in meta-analysis. Six methods of computing effect size, and three techniques for combining study outcomes, were compared. Effect size metrics were calculated with one-group and pooled standardizing denominators, corrected for bias and for unreliability of measurement, and weighted by sample size and by sample variance. Cumulating techniques employed as units of analysis the effect size, the study, and an average study effect. In order to determine whether outcomes might vary with the size of the meta-analysis, mean effect sizes were also compared for two smaller subsets of studies. An existing meta-analysis of 60 studies examining the effectiveness of computer-based instruction was used as a data base for this investigation. Recomputation of the original study data under the six different effect size formulas showed no significant difference among the metrics. Maintaining the independence of the data by using only one effect size per study, whether a single or averaged effect, produced a higher mean effect size than averaging all effect sizes together, although the difference did not reach statistical significance. The sampling distribution of effect size means approached that of the population of 60 studies for subsets consisting of 40 studies, but not for subsets of 20 studies. Results of this study indicated that the researcher may choose any of the methods for effect size calculation or cumulation without fear of biasing the outcome of the metaanalysis. If weighted effect sizes are to be used, care must be taken to avoid giving undue influence to studies which may have large sample sizes, but not necessarily be the most meaningful, theoretically representative, or elegantly designed. It is important for the researcher to locate all relevant studies on the topic under investigation, since selective or even random ...
Date: December 1987
Creator: Ronco, Sharron L. (Sharron Lee)

A Comparison of Some Continuity Corrections for the Chi-Squared Test in 3 x 3, 3 x 4, and 3 x 5 Tables

Description: This study was designed to determine whether chis-quared based tests for independence give reliable estimates (as compared to the exact values provided by Fisher's exact probabilities test) of the probability of a relationship between the variables in 3 X 3, 3 X 4 , and 3 X 5 contingency tables when the sample size is 10, 20, or 30. In addition to the classical (uncorrected) chi-squared test, four methods for continuity correction were compared to Fisher's exact probabilities test. The four methods were Yates' correction, two corrections attributed to Cochran, and Mantel's correction. The study was modeled after a similar comparison conducted on 2 X 2 contingency tables and published by Michael Haber.
Date: May 1987
Creator: Mullen, Jerry D. (Jerry Davis)

A comparison of the Effects of Different Sizes of Ceiling Rules on the Estimates of Reliability of a Mathematics Achievement Test

Description: This study compared the estimates of reliability made using one, two, three, four, five, and unlimited consecutive failures as ceiling rules in scoring a mathematics achievement test which is part of the Iowa Tests of Basic Skill (ITBS), Form 8. There were 700 students randomly selected from a population (N=2640) of students enrolled in the eight grades in a large urban school district in the southwestern United States. These 700 students were randomly divided into seven subgroups so that each subgroup had 100 students. The responses of all those students to three subtests of the mathematics achievement battery, which included mathematical concepts (44 items), problem solving (32 items), and computation (45 items), were analyzed to obtain the item difficulties and a total score for each student. The items in each subtest then were rearranged based on the item difficulties from the highest to the lowest value. In each subgroup, the method using one, two, three, four, five, and unlimited consecutive failures as the ceiling rules were applied to score the individual responses. The total score for each individual was the sum of the correct responses prior to the point described by the ceiling rule. The correct responses after the ceiling rule were not part of the total score. The estimate of reliability in each method was computed by alpha coefficient of the SPSS-X. The results of this study indicated that the estimate of reliability using two, three, four, and five consecutive failures as the ceiling rules were an improvement over the methods using one and unlimited consecutive failures.
Date: May 1987
Creator: Somboon Suriyawongse

A Comparison of Three Criteria Employed in the Selection of Regression Models Using Simulated and Real Data

Description: Researchers who make predictions from educational data are interested in choosing the best regression model possible. Many criteria have been devised for choosing a full or restricted model, and also for selecting the best subset from an all-possible-subsets regression. The relative practical usefulness of three of the criteria used in selecting a regression model was compared in this study: (a) Mallows' C_p, (b) Amemiya's prediction criterion, and (c) Hagerty and Srinivasan's method involving predictive power. Target correlation matrices with 10,000 cases were simulated so that the matrices had varying degrees of effect sizes. The amount of power for each matrix was calculated after one or two predictors was dropped from the full regression model, for sample sizes ranging from n = 25 to n = 150. Also, the null case, when one predictor was uncorrelated with the other predictors, was considered. In addition, comparisons for regression models selected using C_p and prediction criterion were performed using data from the National Educational Longitudinal Study of 1988.
Date: December 1994
Creator: Graham, D. Scott

A Comparison of Three Item Selection Methods in Criterion-Referenced Tests

Description: This study compared three methods of selecting the best discriminating test items and the resultant test reliability of mastery/nonmastery classifications. These three methods were (a) the agreement approach, (b) the phi coefficient approach, and (c) the random selection approach. Test responses from 1,836 students on a 50-item physical science test were used, from which 90 distinct data sets were generated for analysis. These 90 data sets contained 10 replications of the combination of three different sample sizes (75, 150, and 300) and three different numbers of test items (15, 25, and 35). The results of this study indicated that the agreement approach was an appropriate method to be used for selecting criterion-referenced test items at the classroom level, while the phi coefficient approach was an appropriate method to be used at the district and/or state levels. The random selection method did not have similar characteristics in selecting test items and produced the lowest reliabilities, when compared with the agreement and the phi coefficient approaches.
Date: August 1988
Creator: Lin, Hui-Fen

A comparison of traditional and IRT factor analysis.

Description: This study investigated the item parameter recovery of two methods of factor analysis. The methods researched were a traditional factor analysis of tetrachoric correlation coefficients and an IRT approach to factor analysis which utilizes marginal maximum likelihood estimation using an EM algorithm (MMLE-EM). Dichotomous item response data was generated under the 2-parameter normal ogive model (2PNOM) using PARDSIM software. Examinee abilities were sampled from both the standard normal and uniform distributions. True item discrimination, a, was normal with a mean of .75 and a standard deviation of .10. True b, item difficulty, was specified as uniform [-2, 2]. The two distributions of abilities were completely crossed with three test lengths (n= 30, 60, and 100) and three sample sizes (N = 50, 500, and 1000). Each of the 18 conditions was replicated 5 times, resulting in 90 datasets. PRELIS software was used to conduct a traditional factor analysis on the tetrachoric correlations. The IRT approach to factor analysis was conducted using BILOG 3 software. Parameter recovery was evaluated in terms of root mean square error, average signed bias, and Pearson correlations between estimated and true item parameters. ANOVAs were conducted to identify systematic differences in error indices. Based on many of the indices, it appears the IRT approach to factor analysis recovers item parameters better than the traditional approach studied. Future research should compare other methods of factor analysis to MMLE-EM under various non-normal distributions of abilities.
Date: December 2004
Creator: Kay, Cheryl Ann

A Comparison of Traditional Norming and Rasch Quick Norming Methods

Description: The simplicity and ease of use of the Rasch procedure is a decided advantage. The test user needs only two numbers: the frequency of persons who answered each item correctly and the Rasch-calibrated item difficulty, usually a part of an existing item bank. Norms can be computed quickly for any specific group of interest. In addition, once the selected items from the calibrated bank are normed, any test, built from the item bank, is automatically norm-referenced. Thus, it was concluded that the Rasch quick norm procedure is a meaningful alternative to traditional classical true score norming for test users who desire normative data.
Date: August 1993
Creator: Bush, Joan Spooner

A Comparison of Two Criterion-Referenced Item-Selection Techniques Utilizing Simulated Data with Item Pools that Vary in Degrees of Item Difficulty

Description: The problem of this study was to examine the equivalency of two different types of criterion-referenced item-selection techniques on simulated data as item pools varied in degrees of item difficulty. A pretest-posttest design was employed in which pass-fail scores were randomly generated for item pools of twenty-five items. From the item pools, the two techniques determined which items were to be used to make up twelve-item criterion-referenced tests. The twenty-five items also were rank ordered according to the discrimination power of the two techniques.
Date: May 1974
Creator: Davis, Robbie G.

A Comparison of Two Differential Item Functioning Detection Methods: Logistic Regression and an Analysis of Variance Approach Using Rasch Estimation

Description: Differential item functioning (DIF) detection rates were examined for the logistic regression and analysis of variance (ANOVA) DIF detection methods. The methods were applied to simulated data sets of varying test length (20, 40, and 60 items) and sample size (200, 400, and 600 examinees) for both equal and unequal underlying ability between groups as well as for both fixed and varying item discrimination parameters. Each test contained 5% uniform DIF items, 5% non-uniform DIF items, and 5% combination DIF (simultaneous uniform and non-uniform DIF) items. The factors were completely crossed, and each experiment was replicated 100 times. For both methods and all DIF types, a test length of 20 was sufficient for satisfactory DIF detection. The detection rate increased significantly with sample size for each method. With the ANOVA DIF method and uniform DIF, there was a difference in detection rates between discrimination parameter types, which favored varying discrimination and decreased with increased sample size. The detection rate of non-uniform DIF using the ANOVA DIF method was higher with fixed discrimination parameters than with varying discrimination parameters when relative underlying ability was unequal. In the combination DIF case, there was a three-way interaction among the experimental factors discrimination type, relative ability, and sample size for both detection methods. The error rate for the ANOVA DIF detection method decreased as test length increased and increased as sample size increased. For both methods, the error rate was slightly higher with varying discrimination parameters than with fixed. For logistic regression, the error rate increased with sample size when relative underlying ability was unequal between groups. The logistic regression method detected uniform and non-uniform DIF at a higher rate than the ANOVA DIF method. Because the type of DIF present in real data is rarely known, the logistic regression method is recommended for ...
Date: August 1995
Creator: Whitmore, Marjorie Lee Threet

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.

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

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

The Effect of Psychometric Parallelism among Predictors on the Efficiency of Equal Weights and Least Squares Weights in Multiple Regression

Description: There are several conditions for applying equal weights as an alternative to least squares weights. Psychometric parallelism, one of the conditions, has been suggested as a necessary and sufficient condition for equal-weights aggregation. The purpose of this study is to investigate the effect of psychometric parallelism among predictors on the efficiency of equal weights and least squares weights. Target correlation matrices with 10,000 cases were simulated so that the matrices had varying degrees of psychometric parallelism. Five hundred samples with six ratios of observation to predictor = 5/1, 10/1, 20/1, 30/1, 40/1, and 50/1 were drawn from each population. The efficiency is interpreted as the accuracy and the predictive power estimated by the weighting methods. The accuracy is defined by the deviation between the population R² and the sample R² . The predictive power is referred to as the population cross-validated R² and the population mean square error of prediction. The findings indicate there is no statistically significant relationship between the level of psychometric parallelism and the accuracy of least squares weights. In contrast, the correlation between the level of psychometric parallelism and the accuracy of equal weights is significantly negative. Under different conditions, the minimum p value of χ² for testing psychometric parallelism among predictors is also different in order to prove equal weights more powerful than least squares weights. The higher the number of predictors is, the higher the minimum p value. The higher the ratio of observation to predictor is, the higher the minimum p value. The higher the magnitude of intercorrelations among predictors is, the lower the minimum p value. This study demonstrates that the most frequently used levels of significance, 0.05 and 0.01, are no longer the only p values for testing the null hypotheses of psychometric parallelism among predictors when replacing least squares weights ...
Date: May 1996
Creator: Zhang, Desheng

Effect of Rater Training and Scale Type on Leniency and Halo Error in Student Ratings of Faculty

Description: The purpose of this study was to determine if leniency and halo error in student ratings could be reduced by training the student raters and by using a Behaviorally Anchored Rating Scale (BARS) rather than a Likert scale. Two hypotheses were proposed. First, the ratings collected from the trained raters would contain less halo and leniency error than those collected from the untrained raters. Second, within the group of trained raters the BARS would contain less halo and leniency error than the Likert instrument.
Date: May 1987
Creator: Cook, Stuart S. (Stuart Sheldon)

The Effectiveness of a Mediating Structure for Writing Analysis Level Test Items From Text Based Instruction

Description: This study is concerned with the effect of placing text into a mediated structure form upon the generation of test items for analysis level domain referenced test construction. The item writing methodology used is the linguistic (operationally defined) item writing technology developed by Bormuth, Finn, Roid, Haladyna and others. This item writing methodology is compared to 1) the intuitive method based on Bloom's definition of analysis level test questions and 2) the intuitive with keywords identified method of item writing. A mediated structure was developed by coordinating or subordinating sentences in an essay by following five simple grammatical rules. Three test writers each composed a ten-item test using each of the three methodologies based on a common essay. Tests were administered to 102 Composition 1 community college students. Students were asked to read the essay and complete one test form. Test forms by writer and method were randomly delivered. Analysis of variance showed no significant differences among either methods or writers. Item analysis showed no method of item writing resulting in items of consistent difficulty among test item writers. While the results of this study show no significant difference from the intuitive, traditional methods of item writing, analysis level test item generation using a mediating structure may yet prove useful to the classroom teacher with access to a computer. All three test writers agree that test items were easier to write using the generative rules and mediated structure. Also, some relief was felt by the writers in that the method theoretically assured that an analysis level item was written.
Date: August 1989
Creator: Brasel, Michael D. (Michael David)

An Empirical Comparison of Random Number Generators: Period, Structure, Correlation, Density, and Efficiency

Description: Random number generators (RNGs) are widely used in conducting Monte Carlo simulation studies, which are important in the field of statistics for comparing power, mean differences, or distribution shapes between statistical approaches. Statistical results, however, may differ when different random number generators are used. Often older methods have been blindly used with no understanding of their limitations. Many random functions supplied with computers today have been found to be comparatively unsatisfactory. In this study, five multiplicative linear congruential generators (MLCGs) were chosen which are provided in the following statistical packages: RANDU (IBM), RNUN (IMSL), RANUNI (SAS), UNIFORM(SPSS), and RANDOM (BMDP). Using a personal computer (PC), an empirical investigation was performed using five criteria: period length before repeating random numbers, distribution shape, correlation between adjacent numbers, density of distributions and normal approach of random number generator (RNG) in a normal function. All RNG FORTRAN programs were rewritten into Pascal which is more efficient language for the PC. Sets of random numbers were generated using different starting values. A good RNG should have the following properties: a long enough period; a well-structured pattern in distribution; independence between random number sequences; random and uniform distribution; and a good normal approach in the normal distribution. Findings in this study suggested that the above five criteria need to be examined when conducting a simulation study with large enough sample sizes and various starting values because the RNG selected can affect the statistical results. Furthermore, a study for purposes of indicating reproducibility and validity should indicate the source of the RNG, the type of RNG used, evaluation results of the RNG, and any pertinent information related to the computer used in the study. Recommendations for future research are suggested in the area of other RNGs and methods not used in this study, such as additive, combined, ...
Date: August 1995
Creator: Bang, Jung Woong

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.

The Generalization of the Logistic Discriminant Function Analysis and Mantel Score Test Procedures to Detection of Differential Testlet Functioning

Description: Two procedures for detection of differential item functioning (DIF) for polytomous items were generalized to detection of differential testlet functioning (DTLF). The methods compared were the logistic discriminant function analysis procedure for uniform and non-uniform DTLF (LDFA-U and LDFA-N), and the Mantel score test procedure. Further analysis included comparison of results of DTLF analysis using the Mantel procedure with DIF analysis of individual testlet items using the Mantel-Haenszel (MH) procedure. Over 600 chi-squares were analyzed and compared for rejection of null hypotheses. Samples of 500, 1,000, and 2,000 were drawn by gender subgroups from the NELS:88 data set, which contains demographic and test data from over 25,000 eighth graders. Three types of testlets (totalling 29) from the NELS:88 test were analyzed for DTLF. The first type, the common passage testlet, followed the conventional testlet definition: items grouped together by a common reading passage, figure, or graph. The other two types were based upon common content and common process. as outlined in the NELS test specification.
Date: August 1994
Creator: Kinard, Mary E.