(Shiny version of Test Analysis Module)
This is an interactive online platform for test paper analysis and formative assessment using Rasch models. It can provide feedback to teachers by examining the reliability and validity of test papers and the quality of test items. It also provides information regarding students’ Zone of Proximal Development (ZPD).
You may perform your analysis by following the below steps:
1. Import and Analyze:Download the data template and data input manual, and then import the data. Then check your data to see if any errors occurred while inputting. After that, you may perform the analyses by clicking 'Analyze'.
2. Once the analyses are finished, you may view various outputs in:
- Wright Map: pictures comparing distributions of person ability against item difficulty.You may download the outputs and a whole report for further usage.
Note: ShinyTAM utilizes 'TAM' package in R to perform psychometric analysis based on Rasch model.Robitzsch, A., Kiefer, T., & Wu, M. (2020). TAM: Test analysis modules. R package version, 3, 5-19.
License:This program is a free online application and you can use or redistribute it by using the following citation:
Kwan, K., Xu, K., & Zhu, J. (2022-2024). ShinyTAM [Online application; Beta version 1.3]. Retrieved from https://shinytam.eduhk.hk
You may first download the data template and data input manual from the following buttons:
You may preview your data file to check if there is any problem.
EAP reliability: a reliability measure of the test by expected a posteriori method.
Difference between Mean Person Ability and Mean Item Difficulty: a measure to evaluate the discrepancy between the general population ability and the whole test difficulty.
Mean Point-Measure Correlation: the mean association between each item and the whole test.
EAP reliability: a reliability measure of the test by expected a posteriori method.
Difference between Mean Person Ability and Mean Item Difficulty: a measure to evaluate the discrepancy between the general population ability and the whole test difficulty.
Mean Point-Measure Correlation: the mean association between each item and the whole test.
Note: Classes with only 1 person were omitted in the plot.
EAP reliability: a reliability measure of the test by expected a posteriori method.
Difference between Mean Person Ability and Mean Item Difficulty: a measure to evaluate the discrepancy between the general population ability and the whole test difficulty.
Mean Point-Measure Correlation: the mean association between each item and the whole test.
Note: Classes with only 1 person were omitted in the plot.
EAP reliability: a reliability measure of the test estimated by expected a posteriori method.
Mean Point-Measure Correlation: the mean association between each item and the whole test.
A summary of item statistics, including mean percentage score, item difficulty, item fit, point-measure correlation, mean percentage score of each class.
Note. For item fit, '*' represents a fit value > 2, and '#' represents a fit value <0.5.
The report of multiple choice (MC) items: Option Frequency Percentage (Mean Percentage Score / Mean Person Ablity)
The report of non-Multiple-Choice (non-MC) items: Category (Marks) Frequency Percentage (Mean Percentage Score / Mean Person Ablity)
A scatterplot of person percentage score against person fit. Note that a value around 1 indicates the best person fit.
A summary of person statistics, including percentage score, person ability, person fit, performance-fit classification, items that a person is expected to get 1 more/less mark, and ZPD.
A summary of person scores, including expected score, observed score, score gap, and item classification.
You may download the following outputs:
Table Outputs