觀看大圖 相關連結 Click here


FastTEST Highlights






FastTEST 2 is a 32-bit Windows item banker and test assembly system for creating printed tests, surveys, and questionnaires.


FastTEST 2 was designed for easy and efficient item banking and test assembly!


FastTEST 2 is the world's most advanced item banking and test development system!


FastTEST 2 is the item banker and test development system for the FastTEST Professional Testing System, which delivers electronic tests, surveys, and questionnaires including computerized adaptive tests (CATs) using items response theory (IRT)


FastTEST 2.0 and FastTEST Pro are multilingual - they will work with any installed 32-bit Windows font, including right-to-left languages, such as Arabic and Hebrew!


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


The completely redesigned Version 2.0 of the FastTEST Professional Testing System is a complete testing environment.




FastTEST Pro 2 uses the powerful FastTEST 2 Item Banking and Test Development System, which allows you to create and deliver:


Computer-administered tests, including computerized adaptive tests (CAT) based on item response theory (IRT)


Computer-administered questionnaires and surveys, including data collection questionnaires that use extensive branching between and within questions, or testlets


Test questions that can use audio, video, graphics, and popups.


Test questions using multi-windowed screens.


With its capability of delivering adaptive tests (CATs) based on item response theory, combined with a host of other advanced features, FastTEST Pro 2 is the most sophisticated Windows-based testing system available. If you are planning to use FastTEST Pro for CAT, see our companion program POSTSIM 2.0 for Windows. POSTSIM implements post-hoc simulations that allow you to pretest various combinations of CAT parameters to help you design an effective and efficient CAT with any item bank for which you have estimated item parameters using dichotomous IRT methods.



Assessment Systems Corporation 台灣區代理證明


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here












The Laboratory of Educational Research Test Analysis Package, "LERTAP", is a classical item and test analysis system. Lertap also analyzes surveys and mastery tests. Lertap uses a special control language that makes analyses simple and efficient. In Version 5, Lertap's reports have been packaged in a new style, and numerous new features have been added.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


MicroFACT for Windows- Factor Analysis for Dichotomous and Ordered Polytomous Response Data


Niels G. Waller, University of Minnesota, Minneapolis, Minnesota


MicroFACT is an easy-to-use factor analysis program that can tackle mainframe size problems on your PC — quickly! Version 2.0 includes WinMFact, a Windows interface that allows you to build control files by point-and-click and run your analyses! Also included in version 2.0 is an enhanced DOS version of the program that runs the command files created by the Windows interface, or command files that you create independently with a text editor. Version 2.0 also has many other new features.


MicroFACT provides a methodologically defensible and computationally efficient method for factor analysis of both dichotomous and polytomous data sets. Using virtual memory, it can handle analyses of data sets of any size. For dichotomous (i.e., correct/incorrect or keyed/non-keyed) data, MicroFACT can assist you in verifying the dimensionality of your test (which is one of the most basic assumptions of the application of item response theory models). Assessing the dimensionality of a test can be accomplished using a factor analytic approach if appropriate computations are employed. The linear factor analysis model as typically implemented in most statistical packages, however, often fails to use the appropriate measures of correlation for such a factor analysis (tetrachoric correlations). MicroFACT makes it easy to analyze the interrelationships among your test items to determine the number of underlying dimensions. MicroFACT is simple to use and provides detailed (but understandable) results.


For the factor analysis of other types of data, such as polytomous data (i.e., ordered categorical data such as that found in surveys in the social sciences, political science, education, and behavioral sciences, and clinical measures used in medical, psychological and personality research), MicroFACT allows the uses of polychoric correlations, which represents a dramatic step forward in overcoming the biased results produced by the linear factor analysis model. MicroFACT also handles continuously scored and ipsatized item response data.


New Features in version 2


1. MicroFACT can be called from another program to perform Monte Carlo simulations.


2. Data may be written to external files to make simulations easier to conduct.


3. New rotation options: weighted varimax, Harris-Kaiser ortho-oblique, and Oblimin.


4. The ability to rotate from random orientations to avoid local minimum problems.


5. Can import factor pattern, structure, and factor intercorrelation matrices from previous analyses or published studies.


6. New model evaluation features include residual plots, goodness-of-fit indices, and hyperplane counts for rotated factors -- structures and patterns.


7. New factor loading plots.


8. Greater user control of output format.


9. New factor score plots.


10. Improved error handling routines.


11.A complete electronic manual provided as a PDF file, plus on-line Windows Help files.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


LOGIMO - Loglinear and Loglinear IRT Model Analysis


LOGIMO is a program for the estimation and testing of ordinary loglinear models and loglinear IRT models. In loglinear models, the log-probabilities of the various combinations of categorical variables are explained by a linear model. In LOGIMO this linear model may be an ANOVA type model — that is, a model with main and interaction effects of the parameters.


A distinctive feature of LOGIMO is that maximum likelihood estimates of the loglinear model parameters are computed from the sufficient statistic of the parameters rather than the contingency table of the full cross-classification of all variables. Memory requirements for LOGIMO, therefore, depend on the number of model parameters rather than the number of cells in the table.


In loglinear IRT models, the original set of variables is extended with a set of sum score variables. A sum score is computed by summing the scored categories of two or more variables. The category scores are defined a priori by the user. In this way, various types of IRT models can be specified, such as partial credit models, multidimensional partial credit models and multidimensional polytomous latent trait models.


Observed variables can be background variables (e.g., gender, age) rather than items. By testing interactions between background variables and items, invariance of item parameters over subgroups (e.g., item bias) may be studied.


LOGIMO computes maximum likelihood estimates and standard errors of the model parameters, observed and expected sufficient statistics, the kernel of the likelihood, likelihood ratio goodness-of-fit statistics, and Pearson goodness-of-fit statistics for contingency tables specified by the user.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


MSP, Mokken scale analysis for polytomous items, scales test item responses using nonparametric cumulative item response theory. It can test scalability of a given scale or construct one or more unidimensional scales from an item pool. Persons are ordered by scale score and measurement quality is assessed.


MSP can analyze either dichotomous or polytomous data, and can be used for:


Stepwise construction of one or more unidimensional scales from a pool of items;


Evaluation of model fit of a given scale, including the assessment of its reliability and suggestions for removal of misfitting items; and Detection of the extent to which the answer patterns follow the cumulative Guttman scalogram pattern, using Loevinger's H-values and associated significance tests.


The latest version incorporates many new features and improvements:


An improved definition of the scalability coefficients;


Computation of the frequency distribution and corresponding statistics of the scale score, and the number of Guttman errors per person;


An option to analyze and compare several subgroups in one run;


Checks the assumptions underlying the models for monotone homogeneity and double monotonicity; and Estimates the reliability of a test score.


The scaling procedures of the program can handle up to 100 polytomous items with a maximum of 10 ordered categories within a given range (900 item steps). Persons with values outside this range are deleted. Data for up to 32,000 persons can be analyzed.No missing data is allowed.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


PARELLA - Parametric Item Response Model for Measuring Attitudes and Preferences


H. Hoijtink and W. Molenaar, University of Groningen, Netherlands


W. Post, NIDI, Den Haag, Netherlands


To measure latent person characteristics, such as attitudes and preferences, it is not unusual to construct a set of items indicative of the characteristic in question and collect the responses of a sample of persons to these items. On the basis of these responses a unidimensional representation of both persons and items is constructed so that each person is located at a small distance from the items he/she agrees with or prefers.


The PARELLA model is a parametric item response model that can be used for such measurements of attitudes and preferences. The program can construct a unidimensional representation of both persons and items if:


The responses to the items are dichotomous; and


Proximity relations are expected to be the main determinant of a person's response, i.e., the smaller the (psychological) distance between person and item, the larger the probability of a positive response.


PARELLA (parallelogram analysis) uses marginal maximum likelihood to estimate the locations of the items, and the (nonparametric) density function of the persons’ locations. The expected a posteriori (EAP) estimator is available for the estimation of the location of each person. The PARELLA program offers the possibility to distinguish subsamples of persons and estimates the density function of each subsample. It also distinguishes subsets of items and estimates the location of these items per subsample.


PARELLA offers several goodness-of-fit tests and diagnostics, such as:


Tests for differential item functioning (DIF) at the item level and the item-set level;


Diagnostics for the adequacy of the item characteristic curve specified by the PARELLA model; and


Tests associated with the correlation matrix, with the conditional adjacency matrix, and with some other matrices under a more general, completely nonparametric parallelogram model.


The maximum input for the PARELLA program is 60 items, 10 subsamples, and 300 persons.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


RSP - Rasch Scaling Program


C. A. W. Glas, CITO, Arnhem, Netherlands


J. L. Ellis, Catholic University Nijmegen, Netherlands


RSP analyzes the responses to dichotomous items using the Rasch model, including data collected in structured or incomplete designs. RSP uses conditional maximum likelihood (CML) estimation of item parameters, but also can compute marginal maximum likelihood (MML) estimates of item parameters and the parameters of one or more normal trait distributions.


The incomplete designs analyzed by RSP consist of a number of tests that each consist of a set of items administered to a sample of persons. The tests must be linked by common items or common trait distributions. RSP can analyze up to 32 tests simultaneously, with a maximum number of 96 items in a test and a limit of 600 unique items in the design. There is no restriction on the number of persons taking a test. Subsets of items and tests can also be analyzed.


RSP uses a menu-oriented user interface to implement its computational module for estimating and testing the Rasch model; its data file pre-processor module for data description, recoding, and sorting; and its file viewer, for viewing the output files created by the two other modules.


In addition to computing CML or MML estimates of item parameters, RSP:


Computes estimates of person parameters by maximum likelihood, weighted maximum likelihood, and expected aposteriori (EAP) estimation.


Tests model fit using Andersen's likelihood ratio test, Molenaar's U-test, van den Wollenberg's Q-statistics, and Glas' R-statistics, and provides information for identifying causes of item misfit.


Computes person-fit indices.


Because data analysis is usually an iterative process, for each computational run the user can select specific stages of the analysis and specific computational and output parameters can be set.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


SCA - Simultaneous Principal Components Analysis


H. A. L. Kiers / University of Groningen, Netherlands


Simultaneous Component Analysis (SCA) provides principal components in two or more groups that have the same meaning in all groups. In SCA the "meaning" of a component is defined by the component weights (i.e., the weights used to compute the components as weighted sums of variables). The meaning of the components is the same in all groups because the same component weights are used in all groups. The component weights are determined so that the resulting components optimally summarize (in terms of explained variance) the variables in all groups simultaneously. SCA provides a measure for the quality of the solution. This method is a generalization of PErfect CONgruence analysis (PECON), which is used to determine which components from one group describe the observations in another group.


SCA permits rotation of the components to simplify their interpretation. In addition, the components can be defined as user defined unweighted sum scores of subsets of variables (as in the Multiple Group Method). The program also provides a measure of the quality of the resulting components.


Program Specifications


SCA can analyze data for up to 20 groups. The number of variables ranges from 70 (in 2 groups) to 30 (in 20 groups). The maximum number of people is 9999. Input is by either a variables by persons matrix or a correlation matrix.


SCA is partly interactive. The program determines the order of the different procedures in the program, but the user can choose from several options at every step of the program. During one procedure, the user can adjust the solution and study the loss of quality due to the adjustments.


SCA contains several output options which help interpret the simultaneous components, and which reveal possible differences between the behavior of these components in the different groups. As special features, it contains the Multiple Group Method and PECON.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


UNFOLD - Unidimensional Unfolding


A.W. Vogelesang and P. van Blokland / SWOV, Den Haag, Netherlands


Unidimensional unfolding is a technique for finding an underlying dimension for preferential choice data. If there is a common reference frame for judging stimuli, individual preferences unfold into a "J" (joint) scale for both stimuli and persons. Moreover, the median ranking (a consensus ranking) is expected to be a folded J scale, since a unimodal probability model is assumed to underlie the rankings.


UNFOLD is designed for the analysis of all types of preference data. The input data may be complete or incomplete rankings of preference, ratings, or dichotomous choices. The criterion for the best J scale is that it minimizes the total number of inversions from individual rankings.


Data files must be in ASCII (text-only) format. The best qualitative and quantitative J scales can be determined for from 4 to 9 stimuli out of a maximum of 24. A separate analysis is performed for each number of stimuli. Therefore, results for any subset of stimuli are never dependent on previous steps in the analysis. A nested set of J scales will be found if a stable continuum underlies the data: the stimuli on a smaller scale will be included in the larger scale in the same order. There are options for analyzing specific qualitative or quantitative J scales. The test for goodness-of-fit of the unfolding model to the data is based on a non-parametric error model for ranking data.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


WINMIRA 2001 - Latent class analysis, dichotomous and polytomous Rasch models


Matthias von Davier / University of Kiel, Germany


WINMIRA is standalone software for estimating and testing a large number of discrete mixture models for categorical variables. Models with a nominal as well as continuous latent variables, and combinations of both, can be estimated with the software. WINMIRA 2001 can be used for analyses with the Latent Class Analysis (LCA), with the Rasch model (RM), with the Mixed Rasch model (MRM) and with Hybrid models (HYBRID) for dichotomous and polytomous data.


New features of WINMIRA 2001 include:


Improved estimation algorithms - Estimation of the polytomous Rasch models is speeded up and improved. A completely rewritten algorithm ensures increased convergence speed.


User friendly help - The online help system was completely rewritten and includes a detailed description of all new features.


Full SPSS support - Reads and writes data directly in SPSS file format. Person parameter estimates and standard errors, posterior probabilities of class-membership (in the case of mixture models) as well as person-fit statistics can now be appended directly to the SPSS data-file. In addition, EXCEL and other spreadsheet data files can be imported and exported by means of using the 'save as' option to generate tab-delimited files.


Improved and extended parameter constraints - Logistic parameter constraints (i.e., item parameters (for dichotomous models), or item locations as well as threshold distances (for polytomous models) can be imposed both within and between classes. Alternatively, category probabilities can also be constrained.


More documentation - All features of the software are outlined in a 140- page manual that is included in the registered version of WINMIRA 2001. Depending on the license purchased, the latest version of the manual is delivered in high quality printable electronic form (PDF and PostScript) and is optionally accompanied by a printed manual.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


RUMMFOLD SS and PP - Rasch Unidimensional Models for Measurement for Unfolding Response Models


David Andrich, Murdoch University


RUMMFOLDss and RUMMFOLDpp are easy-to-use mouse-driven Windows programs for scaling attitude and preference data. Both estimate the person trait levels and item location parameters of the one-parameter logistic Rasch unfolding measurement model (RUMM). This model assumes a symmetric single-peaked item response function for the items, in which the probability of a correct response decreases as the distance between the person's trait level and the item's location increases in either direction. Unfolding models can arise from two data collection designs - the direct-response single-stimuls (SS) design and the pair-comparison or pairwise preference (PP) design.


RUMMFOLDss


In the direct-response design, respondents respond directly to a single item at a time by replying "agree" or "disagree" to an attitude or preference item on a questionnaire (i.e., each item on the multi-item questionnaire is answered independently of the other items). Direct responses may also be obtained from the dichotomization of a multipoint Likert-type attitude item, such as an item administered in a five-point rating scale format (e.g., strongly disagree, disagree, neutral, agree, strongly agree). These dichotomized responses are then the input to RUMMFOLDss, which analyzes attitude and preference data collected by single-stimulus (direct-response) designs and provides estimates of item location parameters.


RUMMFOLDpp


In the pair-comparison data collection design, the attitude or preference statements are presented in pairs. The respondent then chooses the statement in each pair that he/she prefers, resulting in "pairwise preference" data. In pairwise preference data, the person's location on the trait continuum is assumed to govern their choices as expressed in their selection of one statement over another in each of the statement pairs of the questionnaire. These pairwise preference data are then analyzed according to the Rasch unfolding model for attitude and preference data by RUMMFOLDpp, which provides estimates of item location parameters.


Both RUMMFOLDss and RUMMFOLDpp use as input an ASCII file of person IDs and coded item responses. Output of both programs includes item location parameters, associated item location standard errors, and the results of a chi-square test of fit. For each person, the programs provide trait level estimates and their associated standard deviations. Output options also allow for graphical display of the results for both person and item parameter estimates and related statistics.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


Quest - Interactive Test Analysis System


Raymond J. Adams, Siek-Toon Khoo


Quest offers a comprehensive test and questionnaire analysis environment by providing a data analyst with access to the most recent developments in Rasch measurement theory, as well as a range of traditional analysis procedures. It includes an easy-to-use control language with flexible and informative output. Quest can be used to construct and validate variables based on both dichotomous and polytomous observations. It scores and analyzes such instruments as multiple choice tests, Likert-type rating scales, short answer items, and partial credit items.


The Rasch analysis provides item estimates, case estimates, and fit statistics; and the results from this analysis can be accessed through a variety of informative tables and maps. Additional analyses report counts, percentages, and point-biserals for each possible response to each item. A variety of reliability indices are also available.


Quest runs in batch mode, interactive mode, or a combination of the two. The batch mode conveniently allows a one-step submission of more routine analysis jobs, while the interactive environment facilitates exploration of the properties of test data. In interactive mode, the Quest Display Manager allows analysis results to be viewed on the screen.


Special Features


Several distinct features in Quest set it apart from other test analysis software.


Subgroup and Subscale Analyses. Quest allows you to define subgroups and subscales so that analyses can be performed for any combination of subgroup and subscale.


User-defined Variables. Quest lets you specify variables that define subgroups, and correlate variables with case estimates on any subscale.


Anchoring Parameter Estimates. Quest allows you to anchor (i.e. fix) any item or case estimate to known values, typically obtained from previous analyses. This facilitates equating of tests and item banking.


Dealing with Missing Data. Quest provides flexible procedures for dealing with missing data. You can make Quest ignore certain codes or "holes" in that data. This means you can calibrate several test forms together as long as links exist between forms with common cases or common items.


Exporting Files. Quest lets you export analysis results to text files with a choice of tabs, spaces, or commas as field delimiters to facilitate importation into database, spreadsheet, or other programs.


Scoring and Recording of Data. Quest provides flexible and easy-to-use methods for test scoring and data recoding. Easy recoding of data facilitates regrouping of items and redefinition of scores.


Differential Item Functioning (New in Version 2). The new Quest compare command provides Mantel-Haenszel and Rasch tests of differential item functioning.


Interface


The Quest user interface is designed primarily to encourage you to explore test data. The interactive environment and the Display Manager included in the program allow you to browse through the results of commands. On the basis of these results, you can decide upon the next sequence of commands stored in an external submit file. Quest also allows you to stack jobs, which means that any number of jobs can be lined up and run without attendance.


Quest can read information about anchoring, item names, and so forth from external files and can redirect output to external files. This capability helps you to easily organize and store Quest input and output.


The Quest emphasis for output is graphical. Visual displays are used to summarize key information such as item difficulty and fit.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


POLY-DIMTEST - Latent Unidimensionality Assessment for Polytomously Scored Data


POLY-DIMTEST is a hypothesis testing procedure that assesses lack of latent unidimensionality for a polytomously scored educational or psychological test. It does so by assessing the statistical significance of the possible dimensional distinctiveness between two user-specified subtests: the Assessment Subtest (AT) and the Partitioning Subtest (PT).


POLY-DIMTEST is a modified version of DIMTEST that can be used to assess lack of latent unidimensionality for polytomously scored item response data. As with DIMTEST, it can be used in a confirmatory mode, or in an exploratory mode provided the user provides a procedure to select an AT item set for analysis.


In confirmatory mode, POLY-DIMTEST assesses the user-selected AT that a priori has been judged to be possibly dimensionally distinct from the PT; in exploratory mode, using cross-validation, POLY-DIMTEST assesses unidimensionality by using a statistically selected AT item set that is potentially maximally dimensionally distinct from the rest of the test (i.e., the PT). POLY-DIMTEST is completely nonparametric and requires neither parametric IRT modeling nor estimation of item response functions.


Specifications


POLY-DIMTEST can analyze data files containing up to 12,000 examinees and 151 items.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


CONCOV - Covariance Curve Estimation


Whereas the dimensionality assessment procedures—DIMTEST, HCA/CCPROX, and DETECT—examine dimensionality by averaging over all examinee ability levels, CONCOV nonparametrically estimates the conditional covariance curve for each item pair as a function of the underlying composite latent ability best measured by the total test score. The resulting curves as well as their confidence intervals are graphically displayed for easy evaluation by the user. CONCOV presumes dichotomous item scoring. Such nonparametric estimated conditional covariance curves and nonparametric estimates of the item response functions can together be used to estimate a non-unidimensional IRT model for test data when unidimensionality fails.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


HCA/CCPROX(DOS) - Hierarchical Cluster Analysis


(DOS version for dichotomously or polytomously scored data; see DIMPACK for a Windows version for use with dichotomously scored data.)


HCA/CCPROX performs a latent multidimensionality-sensitive hierarchical cluster analysis on either dichotomous or polytomous scored items. This nonparametric procedure is able to quickly cluster the items into progressively larger and larger relatively dimensionally homogeneous groups. It allows the user to examine the test’s dimensionality at a variety of agglomoration levels, ranging from which pairs of items are most closely dimensionally related, to which two item clusters best dimensionally summarize the entire test. If the items of a test exist in approximately dimensionally homogeneous item clusters, then there should exist a level in the hierarchy at which the clusters found by HCA/CCPROX will maximally agree with this approximate simple structure. Even if approximate simple structure does not hold, HCA/CCPROX will tend to find dimensionally disparate clusters, and hence its uses are not limited to tests demonstrating approximate simple structure. HCA/CCPROX is useful in tandem with a DIMTEST assessment of dimensionality, because it can test hypotheses concerning HCA/CCPROX selected clusters.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


Includes SIBTEST, POLY-SIBTEST, Crossing SIBTEST, DIFCOMP, and DIFSIM


The William Stout Institute for Measurement


These products were originally announced as separate DOS programs. Due to popular demand they are now only available in a package for 32-bit Windows.


Differential Item Functioning


DIFCOMP computes the theoretical amount of DIF/DBF for an item or item bundle for a given two-dimensional IRT model known to produce DIF. It computes the theoretical b index, which measures the amount of DIF/DBF present, for suspect multidimensional items or item bundle given a particular pair of multidimensional ability distributions for the focal and reference groups and the two-dimensional item response funtions of the studied item(s). The user specifies parameters of the bivariate normal distribution of the target (intended) ability and the secondary ability for two populations of examinees, multidimensional logistic parameters for the DIF-displaying item(s) with respect to each ability, whether the amount of unidirectional or crossing DIF/DBF is to be computed, and the desired precision of the integration. Differences in expected item or item bundle scores are computed at multiple locations on the target ability continuum and averaged over ability to compute an approximate theoretical b index. DIFCOMP also permits assessment of DIF amplification (i.e., amount of DBF) by allowing the user to enter parameters for multiple items and compute the combined theoretical b index for the resulting bundle of items.


DIFSIM - Multidimensional Differential Item Functioning


DIFSIM generates item responses for focal and reference populations of examinees based on Shealy-Stout’s multidimensional model for DIF. The user specifies the marginal bivariate distributions of the target ability and each of up to 10 secondary abilities, logistic parameters for all items with respect to each dimension with item parameters supplied by the user or randomly generated, number of examinees in each population, and number of DIF and "valid" items (assumed strictly unidimensional). An output file for each population containing the examinees’ item responses is made. The generated data can then be used, e.g., to estimate the SIBTEST theoretical b index of DIF/DBF magnitude for an item or items. DIFCOMP is useful with DIFSIM: it computes the theoretical value of b for a given multidimensional DIF model like that used in DIFCOMP to simulate DIF/DBF data.


SIBTEST - Differential Item/Bundle Functioning


SIBTEST implements a nonparametric statistical method of assessing DIF/DBF (differential item functioning/differential bundle functioning) in an item or bundle of items based on Shealy-Stout’s (1993) multidimensional model for DIF. Matching on target ability is done by matching on total test score or other score believed to validly measure the target ability (i.e., the dimension best measured by reported test score), such as total score with certain items removed that are believed to be DIF-producing. With a flexible front-end program, the user specifies the particular DIF/DBF hypothesis tests to be performed—including which item or items will be tested, which population (reference, focal, or either) DIF/DBF will be tested to occur against, and which items will be used to construct the matching score. SIBTEST uses a recently developed nonlinear regression correction procedure to match examinees, which has demonstrated particularly good effectiveness in controlling the false flagging of non-DIF items.


Crossing SIBTEST


Crossing SIBTEST tests for crossing DIF, the condition in which reference and focal group differences in expected score at each fixed target ability value on an item or item bundle between two populations change sign at some point on the target ability scale. Analogous to SIBTEST, this program uses a regression correction technique to appropriately match examinees from the two populations, and computes the expected score differences for a suspect item or item bundle at varying ability levels. Fundamental to the procedure is its capacity to identify the location of the crossing point (e.g. the target ability level at which the expected score difference changes sign). Output includes an estimate of the theoretical b crossing-DIF index, the point at which crossing occurs, the relative amount of DIF/DBF occurring against each population (necessary because crossing DIF results in DIF against both populations), and the results of the DIF/DBF significance test. The procedure is also sensitive to noncrossing DIF/DBF, although it differs some in algorithmic detail from regular SIBTEST.


Polytomous SIBTEST


Polytomous SIBTEST handles polytomously scored item responses. This extension of SIBTEST is straightforward, except for some changes to account for the additional score categories in the matching criterion. It uses revised reliability estimates in performing the regression correction, with the regression correction being a generalization of the original linear regression correction.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


MULTISIM - Multidimensional Item Response Theory Analysis


Simulates dichotomous latent multidimensional test responses using a multidimensional compensatory logistic IRT model. MULTISIM can generate test response data having up to four latent dimensions for up to 120 items. It uses as an underlying latent ability distribution a user-specified multivariate normal distribution. The underlying item parameters are specified in a user input file. The number of examinees simulated is essentially unlimited.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


T-Rasch : Non-Parametric Rasch Analysis


I. Ponocny and E. Ponocny-Seliger


T-Rasch provides test construction and detailed item analysis using the Rasch model. T-Rasch is especially useful with small samples of examinees since it uses no asymptotics (which require large samples). T-Rasch is limited to tests with a maximum of 30 items.


T-Rasch implements non-parametric goodness-of-fit tests for the Rasch model. These tests are essentially equivalent to the "naïve" null hypothesis that only one psychological dimension is measured by a test or questionnaire, and that the (unweighted) raw score is fair and provides complete information about the ability of a person. The items must be dichotomously scored (two-category) or at least dichotomized. It is possible to use a priori item parameter values if they are considered to be known sufficiently well from calibration samples, in particular for person-fit applications.


For the user, the main characteristic of T-Rasch is the possibility of generating detailed directed alternative hypotheses easily by means of the menu.


Local stochastic independence: Sets of items suspected to constitute separate scales (i.e., a set is too homogenous in comparison to the others to be consistent with the Rasch model) may be examined by highlighting them in a scroll bar. The example below shows how the second subscale (out of three) is specified to consist of Items 6, 8, and 11; clicking on Select, the highlighted Items 13, 15, and 16 will be added to them.


T-Rasch calculates non-asymptotic p-values referring to most powerful tests for each of the scales separately, as well as a combined p-value for all scales together. A similar procedure is implemented for suspicious item pairs.


Item Bias: If it is suspected that some items are too difficult or easy with respect to a certain subgroup (either defined by an external criterion like gender, origin, or education, or by the internal criterion low/high raw score), item numbers can be selected by means of the menu for too easy ("+")/too difficult ("-") items.


There are also two screening procedures available: one tests for differential item functioning for every item, the other for too high or too low pairwise item-intercorrelations (both with a global fit statistic as well as individual p-values with and without alpha adjustment; the program "searches" for model deviations).


For known item parameters, exact tests are computed, yielding (intended or as by-products) fit statistics for each person. For unknown item parameters, a Monte Carlo algorithm is implemented, the precision of which depends only on the number of simulated cases. All tests are exact (or at least Monte Carlo approximations of exact) Neyman-Pearson tests with the Rasch model as the null hypothesis.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


MULTILOG 7 - Analysis of Multiple-Category Response Data


David Thissen, Wen-Hung Chen and Darrell Bock


A research-oriented IRT program that handles all the major models: 1, 2, and 3 parameter logistic, multiple nominal categories, graded rating scale model, partial credit model, multiple-choice model, and constrained parameter models.


Previously only available as DOS software, MULTILOG has been updated to the current Windows operating systems: 98, NT, 2000, ME, and XP. Extensive online Help has been added, covering almost completely the newly edited book that combines all four programs in one volume.


Annotated program input and output can be found throughout the online Help system and the new manual. A graphic module adds item characteristic curves, item information curves, and test information curves to MULTILOG. The plots may be saved in the Windows Metafile format (WMF) and are then easily included in applications such as MS Word.


Although the programs are command driven (an analysis is specified by creating a command file with the included editor), MULTILOG features a dialog-box user interface to assist first-time users or occasional users with writing such a command file.


Some Features of MULTILOG 7


Easy to use graphical user interface


One, two and three-parameter logistic models


Samejima's model for graded responses


Bock's model for nominal (non-ordered) responses


Steinberg's model for multiple-choice items


Handling of multiple-alternative items, such as multiple-choice tests or Likert-type attitude questionnaires


Scoring of items with multiple alternatives


Differential item functioning (DIF)


Handling of data from several populations simultaneously


Analysis of mixtures of items types


Testing of item parameters across groups


Handling of equality constraints and fixed parameters


Presentation quality IRT graphics, can be imported in Word, Access, etc.


Detailed online HELP documentation include description of interface, syntax, and examples.


Includes a PDF file of a complete user’s manual. Printed copies of the manual are available for $40.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖

BOJA is a DOS program which is particularly useful when classical parametric methods of statistical analysis are inappropriate.


For the bootstrap, the resampling procedures include the standard bootstrap, the balanced bootstrap, and nested bootstrap resampling schemes.


The jackknife resampling procedures include the standard, the random, and the grouped jackknife. The internal statistical applications include the analysis of means, medians, (co)variances, correlations, linear regression, and analysis of variance (randomized designs).


External statistical applications can be dealt with easily, given that the proper parameter estimates are available on file. The results of the analyses can also be inspected graphically.


Input Requirements


No missing values. Maximum 45 variables, 9999 observations. Full matrix input.


System Requirements


This software is a DOS based program, which will run in the Windows environment, unless specific exceptions are noted. If you experience difficulty opening this software, please refer to the following support page on the Microsoft website:


Microsoft Support Article ID# 324767 - "Error message when you install or start an MS-DOS or 16-bit Windows-based program"


If you are unable to get this software to run on your machine, or if you have questions before purchasing the program, please contact our technical support department.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖

GRADAP can be used to define, manipulate, and analyze graphs and networks of various kinds.


It provides facilities for simple graphs, digraphs, and valued graphs, including the detection of cliques and components; all major types of point and network centrality measures; spatial autocorrelation; and variance degree.


New graphs can be generated from the original data, with the help of selection, aggregation and induction.


Facilities to group points and lines in sets make the analysis of subgraphs and partial graphs very easy.


Uses a data structure especially designed for network data, facilitating data interfacing with relational databases such as dBase, INGRES or Oracle.


Graphs can be extended with extra graphical elements (i.e., extra lines, points, and information). All parallel lines in a graph can be combined. Points can be condensed or induced.


System Requirements


This software is a DOS based program, which will run in the Windows environment, unless specific exceptions are noted. If you experience difficulty opening this software, please refer to the following support page on the Microsoft website:


Microsoft Support Article ID# 324767 - \"Error message when you install or start an MS-DOS or 16-bit Windows-based program\"


If you are unable to get this software to run on your machine, or if you have questions before purchasing the program, please contact our technical support department.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖

LISREL/PRELIS - Easy Structural Equation Modeling for Windows and Mac


New Statistical Applications


LISREL for Windows is no longer limited to structural equation modeling (SEM). The latest LISREL for Windows includes the following statistical applications.


LISREL for structural equation modeling.


PRELIS for data manipulations and basic statistical analyses.


MULTILEV for hierarchical linear and non-linear modeling.


SURVEYGLIM for generalized linear modeling.


CATFIRM for formative inference-based recursive modeling for categorical response variables.


CONFIRM for formative inference-based recursive modeling for continuous response variables.


MAPGLIM for generalized linear modeling for multilevel data.


New in LISREL 8.8


Structured latent curve models


The LISREL CO command has been extended to include the exponential (EXP) and natural logarithm (LOG) operators as well as parentheses. This allows LISREL users to fit, for example, the structured latent curve models outlined in Browne (1993).


Factor analysis of ordinal variables


Classical exploratory factor analysis assumes that the observed variables are continuous. The PRELIS OFA command implements exploratory factor analysis of ordinal variables as described in Jöreskog & Moustaki (2006).


Generalized linear models (GLIMs) for multilevel data


The new statistical application MAPGLIM fits generalized linear models to multilevel data. Users can select from the multinomial, Bernoulli, Poisson, binomial, negative binomial, Normal, Gamma and inverse Gaussian sampling distributions. The corresponding link functions include the log, cumulative logit, cumulative probit, complementary log-log and logit link functions.


Observational Residuals


Bollen and Arminger (1991) introduced observational residuals for structural equation models. LISREL 8.8 for Windows allows users to compute observational residuals along with latent variable scores for the latent variables of the model. This implementation is described and illustrated in Jöreskog, Sörbom & Wallentin (2006)


Writing parameter estimates, standard error estimates and measures of fit to a PSF


The PV, SV and GF keywords on the LISREL OU command or the SIMPLIS LISREL output command have been extended to allow users to save the parameter estimates, standard error estimates and measures of fit to a PSF. This is especially useful for Monte Carlo studies.


Changes to the graphical user interface (GUI)


The main window of LISREL 8.8 for Windows is now entitled LISREL for Windows. The revised Export Data option on the File menu of the main window allows users to export data to various data formats such as SPSS, SAS, SYSTAT, Statistica, etc.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖

POSTSIM - Post-Hoc (Real-Data) Simulation of Computerized Adaptive Testing


POSTSIM 2.0 implements post-hoc or \"real-data\" simulations of computerized adaptive testing.


In implementing a computerized adaptive test (CAT) testing program, post-hoc simulation is an important step prior to live implementation of a CAT.


Post-hoc simulation allows you to evaluate various CAT testing parameters prior to live testing, including entry points into the CAT, item selection options, scoring methods, and termination criteria, so that your live CAT will function optimally with the item bank that you have calibrated using dichotomous item response theory (IRT) models.


POSTSIM includes all of the CAT options in the FastTEST Professional Testing System, as well as some other CAT options not yet available in FastTEST Pro.


Whereas monte carlo simulations are typically useful in the early stages of investigating the performance characteristics of CAT procedures, post-hoc simulation is particularly useful when a calibrated item bank is available, prior to the implementation of a live CAT. In contrast to monte carlo simulations, post-hoc simulations use the responses of real examinees to real items.


To implement a post-hoc simulation, you will need item response data from a group of examinees who have answered a set of multiple-choice items administered as a conventional test. In the case of CAT, this test would typically consist of the IRT-calibrated items that will comprise the CAT item bank. The post-hoc simulation then \"re-administers\" those items to the examinees (using the responses they have already provided) \"as if\" the item bank had been administered using various CAT procedures.


POSTSIM will implement simulations for item banks of up to 999 items, with no limit on the number of examinees.


Input requirements include a data file in the same format used in our Item and Test Analysis Package with ITEMAN, RASCAL, and/or XCALIBRE, and a file of IRT item parameter estimates from a 1-, 2-, or 3-parameter logistic model (.g., an XCALIBRE output .PAR file).


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖

PARSCALE 4 - IRT Scaling, Item Analysis, and Scoring or Rating Scale Data


This most versatile of all IRT rating-scale programs now includes adjustments for differences in rater severity. PARSCALE 4 now includes:


DIF of rating scale items


Handles up to 15 categories


User options for Samejima’s graded response model generalized to rating scales or Master’s partial credit model with or without discriminating power parameters.


Allows mixtures of rating scale items and multiple-choice items with or without guessing.


Handles multiple subtests and weighted combinations of subtest scores.


Previously only available as DOS software, the programs have been updated to the current Windows operating systems: 98, NT, 2000, ME, and XP. Extensive online Help has been added, covering almost completely the newly edited book that combines all four programs in one volume.


Some Features of PARSCALE 4


One, two, and three-parameter logistic models


Samejima's model for graded responses


Master's partial credit model


Generalized partial credit model


Analysis of rating scale items such as open-ended essay questions


Analysis of multiple-choice items


Differential item functioning (DIF)


Analysis of mixtures of item types


Rater's-effect analysis


Multiple-group polytomous item response models


Presentation quality IRT graphics, can be imported in to Word, Access, etc.


Detailed online HELP documentation including syntax and examples.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


"LERTAP"為”Laboratory of Educational Research Test Analysis Package”的縮寫，是一各以Excel為基礎的應用程式。


LERTAP可以提供各種分析圖形及報表，同時有完整的程式來檢查測驗的作弊功能。


LERTAP的強項是提供圖形分析結果：包括分位數圖、分組測驗分數盒圖、特徵值陡坡圖(Eigenvalue scree plot)、分數長條圖(score histograms)及散佈圖(scatterplots)。


LERTAP同時也可計算分類測量之條件標準誤。因為不斷的研究發展，本軟體已成功的應用於認證的考試。


Lertap 5 is a classical item, test, and survey analysis system which has continually been refined and updated for the last decade.


An Excel-based application, Lertap 5 produces a variety of tabular reports and related graphics.


Lertap extends the features commonly found in classical item and test analysis programs by providing support for users of mastery, licensing, and certification tests.


It also has comprehensive routines for detecting test cheating.


A particular strength of Lertap is its incorporation of graphical portrayals of results, including quintile plots of distractor functioning, boxplots of group test scores, eigenvalue scree plot displays, score histograms, and scatterplots.


Lertap also computes conditional standard errors of measurement and indices of classification consistency.


Backed by extensive documentation, Lertap runs on both Windows and Macintosh systems.


A new version for use with Excel 2007 is ready for users to try.


(Excel 2007 is the latest version of Excel for Windows.


It runs under Windows XP as well as Windows Vista.)


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


包含DIMTEST, DETECT及HCA/CCPROX for dichotomously scored data.




DIMTEST是一假設檢定的程序，用於評估潛在維度變數是否存在。


常用於教育及心理測驗上，操作的模式包含肯證(Confirmatory mode)和探索模式(Exploratory mode)。DETECT主要功能是做維度分析(Dimensionality Analysis)；而HCA/CCPROX 是做階層集群分析(Hierarchical Cluster Analysis)。


使用的維度敏感近似矩陣(Dimensionally-Sensitive Proximity Matrices)的技巧。


DIMPACK - Nonparametric Dimensionality Analysis Package


Includes DIMTEST, DETECT, and HCA/CCPROX for Dichotomously Scored Data.


For polytomously scored data, see POLY-DIMTEST and HCA/CCPROX.


These latent dimensionality structure programs previously have been distributed as separate DOS programs.


Due to ongoing demand for them, they are now being made available in one package for 32-bit Windows.


Essentially no changes have been made in the three main programs, but the Windows interface facilitates selection of program options and provides ready access to help documentation.


A printed manual that includes relevant research literature on the methods accompanies the CD.


This new Windows package includes DIMTEST, DETECT, and HCA/CCPROX for dichotomously scored data.


DIMTEST 2 is a hypothesis testing procedure that assesses lack of latent unidimensionality for an educational or psychological test, operating in a confirmatory mode or in an exploratory mode.


DETECT nonparametrically provides a detailed dimensional description of a test in which the items are grouped around several distinct reading passages or underlying psychological or cognitive constructs; DETECT also operates in either a confirmatory or an exploratory mode. HCA/CCPROX performs a latent multidimensionality-sensitive hierarchical cluster analysis on dichotomously scored items, allowing the user to examine the test’s dimensionality at a variety of agglomeration levels.


The new Windows interface in DimPack facilitates selection of program options and provides ready access to help documentation.


New Features in DIMTEST 2


DIMTEST 2 is a major upgrade to the earlier DIMTEST, allowing longer tests and larger samples to be analyzed:


The upper limit on the number of items is 151 (increased from 50)


The upper limit on the number of examinees is 12,000 (increased from 2,000). 12,000 will almost always provide the statistical power the user needs.


The earlier version of DIMTEST required that the test be divided into three subtests: AT, AT2, and PT, where in addition to the user selected subtest AT discussed above, a bias-correcting AT2 was also needed.


Thus, the AT2 requirement resulted in "sacrificing" a portion of the PT subtest to correct for statistical bias.


DIMTEST 2 eliminates AT2. Instead of using AT2, AT-based simulated data are used to correct for statistical bias.


Research has shown this to result in a more powerful hypothesis-testing statistic that still adheres well to the nominal Type 1 error rate.


Furthermore, the elimination of AT2 enables DIMTEST to be applied to shorter tests and to be used with AT's that have larger numbers of items.


One important exploratory capability provided by the old DIMTEST was an automatic procedure, called FAC, by which a linear factor analysis was applied to a training sample to find a single maximally dimensionally distinct AT, which was then tested by DIMTEST on a cross-validation sample.


DIMTEST 2 replaces FAC with a new procedure called ATFIND that is more effective at finding a dimensionally distinct AT when multidimensionality holds. Importantly, ATFIND is based on the same basic statistical building blocks as DIMTEST, namely estimated item pair conditional covariances. In particular, ATFIND is based on a combination of DETECT and HCA/CCPROX. Research has shown ATFIND to result in increased DIMTEST statistical power (sometimes dramatically so) while still adhering well to the nominal Type 1 error rates.


DETECT – Dimensionality Analysis


One of the most common types of test formats is one in which the items are grouped around several distinct reading passages or underlying psychological or cognitive constructs—for example, algebra, geometry, and trigonometry in a mathematics test.


Using estimated item pair conditional covariances, the DETECT procedure nonparametrically provides a detailed dimensional description of this type of test, which is said to exhibit approximate simple structure.


DETECT begins by using clusters obtained from the dimensionality-sensitive cluster analysis procedure HCA/CCPROX and then uses a customized genetic algorithm to efficiently search through all of the possible item cluster partitions to find the one that maximizes the DETECT statistic.


In addition to quickly finding the dimensionally most appropriate grouping of the test’s items and thereby estimating the number of dominant dimensions present, its two resulting statistics provide a summary of the test’s total amount of multidimensionality (equivalently, a measure of the lack of fit by a unidimensional model) and a measure of the degree to which it displays approximate simple structure.


This information is often useful from a statistical robustness perspective for those wishing to use IRT methodologies based on unidimensionality.


DETECT presumes dichotomous item scoring.


DETECT can be run in either confirmatory or exploratory mode.


In confirmatory mode, the user specifies a set of non-overlapping clusters for the test items and DETECT calculates its indices for that clustering.


In exploratory mode, the data are split into a training sample and a cross-validation sample.


The genetic algorithm searches for the best clusters on the training sample and then the cross-validation sample is used to calculate the DETECT indices.



HCA/CCPROX – Hierarchical Cluster Analysis with Dimensionally-Sensitive Proximity Matrices


HCA/CCPROX performs a latent multidimensionality-sensitive hierarchical cluster analysis on dichotomously scored items.


This nonparametric procedure is able to quickly cluster the items into progressively larger and larger relatively dimensionally homogeneous groups.


It allows the user to examine the test’s dimensionality at a variety of agglomeration levels, ranging from which pairs of items are most closely dimensionally related, to which two-cluster solution best dimensionally summarize the entire test.


If the items of a test exist in approximately dimensionally homogeneous item clusters, then there should exist a level in the hierarchy at which the clusters found by HCA/CCPROX will maximally agree with this approximate simple structure.


Even if approximate simple structure does not hold, HCA/CCPROX will tend to find dimensionally disparate clusters, and hence its uses are not limited to tests demonstrating approximate simple structure.


HCA/CCPROX is useful in tandem with a DIMTEST assessment of dimensionality, because DIMTEST can test hypotheses concerning HCA/CCPROX selected clusters.


System Requirements


Windows 98 or Higher.


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


ConQuest - Generalized Item Response Modeling


ConQuest Version 2.0 is a computer program for fitting item response (Rasch) and latent regression models.


It provides a comprehensive and flexible range of item response models (IRM) to analysts, allowing them to examine the properties of performance assessments, traditional assessments and rating scales.


ConQuest 2.0 also offers the wider measurement and research community the most up-to-date psychometric methods of multifaceted item response models, multidimensional item response models, latent regression models and drawing plausible values.


Now with 30 Day Free Trial and Free Manual!


ConQuest Version 2.0 Enhancements


ConQuest Version 2.0 has incorporated many enhancements to the 1998 release. These include:


plots of item characteristic curves


user-defined fit statistics


estimation of population characteristics such as percentages above a cut-point on a scale,


more user-friendly interface


Models ConQuest Can Fit


Rasch’s Simple Logistic Model


Rating Scale Model


Partial Credit Model


Ordered Partition Model


Linear Logistic Test Model


Multifaceted Models


Generalized Unidimensional Models


Multidimensional Item Response Models


Latent Regression Models


How Does ConQuest Fit These Models?


ConQuest produces marginal maximum likelihood estimates for the parameters of the models. The estimation algorithms it uses are adaptations of the quadrature method described by Bock and Allen (1981) and the Monte Carlo method of Volodin and Adams (1995). The fit of the models is ascertained by generalizations of the Wright and Masters (1982) residual-based methods developed by Wu (1997). The manual contains a summary of these procedures.


Some Applications of ConQuest


Performing item analysis


Examining differential item functioning


Exploring rater effects


Estimating latent correlations and testing dimensionality


Drawing plausible values


ConQuest is available with both a graphical user interface (GUI) and a simple command line, or console, interface. The console version of the program is available for all of the ConQuest platforms except Windows 3.1x. The GUI version is available for all Windows platforms.


Program Limits


Item response categories: 65


Item codes: 200


Number of generalised items for each case: 1000


Response model parameters: 3000


Latent dimensions: 15


Elements in each facet: 1000


Characters in a line of data and length of filename path: 3072


Lines specified in a formal statement: 20


Variables in a model statement: 1000


The number of item steps is not limited except by system memory.


Comparing ConQuest and Quest


Quest fits Rasch’s simple logistic model (for sets of dichotomously scored items), Andrich’s Rating Scale Model (for analyzing Likert scales) and Master’s Partial Credit Model (for analyzing polytomous items or mixtures of dichotomous and polytomous items).


Quest also implements Mantel-Haenszel approaches to assess differential item functioning.


ConQuest can fit a wider range of item response models than Quest.


ConQuest can also fit both unidimensional and multidimensional latent regression models.


Quest uses joint maximum likelihood to estimate parameters, whereas ConQuest uses marginal maximum likelihood.


Quest is available for Macintosh OS, DOS, Windows 95 and Windows NT, but it does not have a GUI interface.


ConQuest is available for Windows and Windows NT with a GUI interface and a console interface.



Assessment Systems Corporation 台灣區代理證明


價格根據授權模式而定

立刻索取此產品報價單

觀看大圖 相關連結 Click here


FastTEST 可讓您的測試問題容易地進入階層式並且列印，選項模組可讓您在電腦上管理測驗。


FastTEST Professional 包含所有FastTEST功能，另增影像、合乎項目(matching items)、適應測試樣板(adaptive testing templates)及使用項目反應原理(item response theory)。


FastTEST Pro Version 2.0 is completely new and has been redesigned for maximum flexibility to make it easy and efficient for you to develop and implement a wide range of PC-based electronic tests.


FastTEST Pro's new modular structure works in conjunction with the powerful FastTEST 2 Item Banking and Test Development System to deliver a wide variety of PC-based tests, surveys, and questionnaires.


You can combine multiple instruction sequences with multiple tests, including conventional tests, randomized tests, complex branched tests, or fully adaptive tests based on item response theory. Follow any test with an optional review and/or a detailed or custom designed report that includes scores from one or multiple tests. Each question in your test can be designed according to your needs and can include multiple graphics, audios, and videos.


FastTEST Pro's distributed computing design allows tests to be delivered on a network, on independent testing stations, from CDs or USB memory drives, or sent as email attachments and easily installed and delivered on remote computers. Want to take a closer look? Try FastTEST Pro 2.0 Free for 30 Days!


價格根據授權模式而定

立刻索取此產品報價單