A review of process data studies in educational assessment

Batch process performance monitoring has been achieved primarily using process measurements with the extracted information being associated with the physical parameters of the process. With increasing attention now being paid to the application of on‐line real‐time process analytics through spectrometry, together with the FDA Process Analytical Technologies (PAT) initiative, the use of spectroscopic information for enhanced monitoring of reactions is gaining impetus. The harmonious integration of process data and spectroscopic data then becomes a major challenge. By integrating the process and spectroscopic measurements for multivariate statistical data modelling and analysis, it is conjectured that improved process understanding and fault diagnosis can be achieved. An investigation into combining process and spectral data using multiblock and multiresolution analysis is proposed and the results from the analysis of experimental data from two industrial application studies are presented to demonstrate the improvements achievable in terms of process performance monitoring and fault diagnosis.

Students’ problem-solving strategies and the differences among strategy groups were explored by analyzing the process data collected during student interactions with computer-based science items. Data were gathered from 1516 eleventh-grade students from 4 schools in China. Analyses of the sequences of students’ response actions revealed that the students were divided into four strategy groups when designing experiments to solve scientific problems: the scientific and rigorous strategy (18.5%), scientific and less rigorous strategy (25.4%), incomplete strategy (31.5%), and chaotic strategy (24.6%). The heatmaps of response actions for each strategy and the frequencies of the most representative response sequences were further explored to understand the students’ detailed trajectories. The results showed that successful problem solvers were generally inclined to explore all possibilities of experimental combinations and design experiments scientifically and rigorously based on the relevant scientific principles. Moreover, the timestamps of response actions were explored to show that the students who adopted the scientific and rigorous strategy spent more time seeking solutions, suggesting that students may need sufficient time to solve complex and authentic scientific problems. The findings enrich the literature on using process data to address theoretical issues in educational assessment and provide students with individualized instructional needs for teachers to improve students’ scientific problem-solving competency. Keywords: process data, scientific problem-solving, computer-based assessment, China

In recent years the educational community has increasingly embraced digital technologies for the purposes of developing alternative learning environments, providing diagnostic feedback, and fostering the development of socalled 21st -century skills. This special issue is dedicated to bridging recent work from the disciplines of educational and psychological assessment and educational data mining (EDM) via the assessment design and implementation framework of evidence-centered design (ECD). It consists of a series of five papers: one conceptual paper on ECD, three applied case studies that use ECD and EDM tools, and one simulation study that relies on ECD for its design and EDM for its implementation. In this introduction, we discuss the underlying rationales for the special issue in more detail, provide a short introduction to ECD, and describe the focus of the five selected papers.

Experimental design and statistical data analyses are fundamental components of animal science research. Proper design of experiments and adequate sampling permits testing hypotheses raised by researchers and sets the stage for collecting required data and subsequent statistical analysis. When designing experiments, researchers should respect rules of randomization of treatments to avoid statistical bias and permit proper inference to be drawn. Use of sample sizes that result in adequate statistical power to identify the hypothesized differences among factor levels of interest is key and should be driven by formal processes determining such. Best practices for data collection should be performed to obtain high quality data by reducing collection (e.g., mislabeling, improper technique) and measurement errors. With sound data, appropriate and optimal statistical methods should be used to generate valid results. The statistical method deployed should be chosen based on assumptions about residuals (e.g., normality, correlation, and homogeneity) and on the type of data (e.g., quantitative continuous or categorical). The appropriate statistical model used should also be consistent with the experimental design to validate the respective test statistics. The science of statistics is changing rapidly. With the development of high-throughput technologies, the generation of large datasets, high performance and sophisticated models and the interest in Big Data, the training of animal science graduate students in data management and rigorous statistical analyses is more important than ever. In order to meet the demands of current trends, animal science graduate students must be trained in several complex statistical and computational skills to meet the challenges imposed by these complicated, sophisticated and nuanced analytical methods. The livestock production sector will benefit from improved training, use of advanced and appropriate experimental designs, and collection and analysis of quality data in research.

This report summarizes the current statistical analysis capability of OVIS and how it works in conjunction with the OVIS data readers and interpolators. It also documents how to extend these capabilities. OVIS is a tool for parallel statistical analysis of sensor data to improve system reliability. Parallelism is achieved using a distributed data model: many sensors on similar components (metaphorically sheep) insert measurements into a series of databases on computers reserved for analyzing the measurements (metaphorically shepherds). Each shepherd node then processes the sheep data stored locally and the results are aggregated across all shepherds. OVIS uses the Visualization Tool Kit (VTK) statistics algorithm class hierarchy to perform analysis of each process's data but avoids VTK's model aggregation stage which uses the Message Passing Interface (MPI); this is because if a single process in an MPI job fails, the entire job will fail. Instead, OVIS uses asynchronous database replication to aggregate statistical models. OVIS has several additional features beyond those present in VTK that, first, accommodate its particular data format and, second, improve the memory and speed of the statistical analyses. First, because many statistical algorithms are multivariate in nature and sensor data is typically univariate, interpolation of data is required to provide simultaneous observations of metrics. Note that in this report, we will refer to a single value obtained from a sensor as a measurement while a collection of multiple sensor values simultaneously present in the system is an observation. A base class for interpolation is provided that abstracts the operation of converting multiple sensor measurements into simultaneous observations. A concrete implementation is provided that performs piecewise constant temporal interpolation of multiple metrics across a single component. Secondly, because calculations may summarize data too large to fit in memory OVIS analyses batches of observations at a time and aggregates these intermediate intra-process models as it goes before storing the final model for inter-process aggregation via database replication. This reduces the memory footprint of the analysis, interpolation, and the database client and server query processing. This also interleaves processing with the disk I/O required to fetch data from the database - also improving speed. This report documents how OVIS performs analyses and how to create additional analysis components that fetch measurements from the database, perform interpolation, or perform operations on streamed observations (such as model updates or assessments). The rest of this section outlines the OVIS analysis algorithm and is followed by sections specific to each subtask. Note that we are limiting our discussion for now to the creation of a model from a set of measurements, and not including the assessment of observations using a model. The same framework can be used for assessment but that use case is not detailed in this report.

Satellite image data generated from remote sensors around the world have different resolutions and are processed at varying levels from Level 0 to Level 3, with each level containing vast amounts of information. Due to the problem of data size, many ocean science researchers use L3 images, which have a spatial resolution of 4 km or 9 km. However, in order to overcome problems such as red tides or to analyze the marine ecosystem based on ocean color satellite research, researchers must generate data by changing the parameters of images at various levels. There is also a need for immediate access to satellite image information using analytical and visualization tools. Considering those requirements, TuPiX-OC (Turning PiXels into knowledge and science-OceanColor) provides an environment to design and execute algorithms for data processing and analysis of satellite image data by data type. TuPix-OC, which has a distributed architecture, is an analytical platform that supports data import, level conversion, DB integration, analysis and processing, and visualization. TuPiX-OC stores satellite data in a massive storage device, and provides an online platform for satellite data conversion/analysis/ visualization. For satellite data processing, TuPiX-OC converts NASA-provided binary files into files that can be analyzed by users. Moreover, TuPiX-OC includes various algorithms for satellite data selection and utilization of satellite images. Preliminary Experiments of TuPiX-OC's satellite image data processing capability showed that it was able to process 35 times as many images as the open source software SeaDAS.

This article presents a review of the combined analysis of digital terrain models (DTMs) and remotely sensed data in landscape investigations. The utilization of remotely sensed data with DTMs has become an important trend in geomatics in the past two decades. Models of more than ten quantitative topographic variables are employed as ancillary data in the treatment of images. The article reviews the methods for DTM derivation and the basic problems of DTM operation that are important for handling DTMs with imagery, namely: 1) the choice of a DTM network type; 2) DTM resolution; 3) DTM accuracy; and 4) the precise superimposition of DTMs and images. The processing of remotely sensed data and DTMs in combination is used in the following procedures: 1) the image correction of the topographic effect; 2) the correction of geometric image distortion; 3) image classification; 4) statistical and comparative analyses of landscape data; and 5) three-dimensional landscape modelling. These procedures are applied to solve a wide range of problems in geobotany, geochemistry, soil science, geology, glaciology and other sciences. The joint use of imagery and DTMs can increase the total amount of information extracted from both types of data. The trend has been towards the incorporation of the combined analysis of remotely sensed data and DTMs into mixed environmental models. The following potential applications of the treatment of imagery in association with DTMs are identified: 1) the prediction of the migration and accumulation zones of water, mineral and organic substances moved by gravity along the land surface and in the soil; 2) the investigation of the relationships between topographically expressed geological structures and landscape properties; 3) the improvement of geological engineering in industrial planning (e.g., the construction of nuclear power stations, oil and gas pipelines and canals); and 4) the monitoring of existing industries. Digital models of plan, profile, mean and total accumulation curvatures, and nonlocal and combined topographic attributes should be included in data processing both to solve the problems indicated and to improve the outcome of some regular tasks (for example, the prediction of soil moisture distribution and fault recognition).

In order to properly manage the depletion of a petroleum reservoir one must understand its physical characteristics and their relationship to production performance. Often the challenge is to interpret an extensive amount of available information. This paper presents an approach to this common petroleum engineering problem using statistical analysis. Statistical analysis can provide an improved understanding of any reservoir. Beginning with the compilation of a comprehensive database, intrinsic relationships among physical parameters and production are characterized. Mapping of statistically processed data allows field-wide interpretation and visual comparison with the geological model. Probabilities of production success can be mathematically determined and used to focus optimization efforts on areas of the reservoir with favourable characteristics. The techniques provide an unbiased appraisal of the available data that can lead to re-examination of assumptions about the underlying mechanisms governing production behaviour. Technical hypotheses can be tested for consistency by determining if expected correlations are present in the data. A recent evaluation of the performance of a heavy oil waterflood at Golden Lake, Saskatchewan, is discussed to illustrate the application of the method. Modification of well completion practices and the reservoir depletion strategy have resulted from the study, and favourable areas for infill drilling have been identified. Introduction The initial challenge when analysing the performance of a petroleum reservoir is often to marshal and interpret an overwhelming amount of raw data. Technological advances in reservoir characterization and data processing have provided access to an abundant supply of numerical data. Coincidentally, the importance of understanding reservoir heterogeneity to maximize depletion efficiency is increasingly recognized. Complex variation of fluid and rock properties within reservoirs is the norm, demanding more detailed description and sophisticated technical analysis. Techniques for processing large databases and distilling critical knowledge are therefore of growing interest to the petroleum engineer. The data overload problem may be addressed by arbitrarily discarding information or by working with averaged values. A compromise between these shortcuts and utilization of all of the data available is usually necessary; however, the quality of technical analysis may be sacrificed by over simplification. A method of extracting the important information concealed in an extensive dataset is required. Determination of the significant factors controlling production success will usually suggest the proper course to optimize the field. Statistical analysis techniques are well suited to achieving these objectives. Statistical analysis of data is best regarded as a supplement to, not a replacement for, the engineering analysis typically undertaken to solve technical problems. As an integrated component of a study, it can generate ideas, provide direction, and confirm that the data actually supports the theoretical conclusions. The value added by statistical analysis comes from ensuring no important trends hidden in the data are overlooked, and from the improved level of confidence in the results. Petro-Canada initiated a comprehensive technical review of the Golden Lake heavy oil field in 1994. The scope of the work included standard waterflood analysis, laboratory experiments, field tests, and numerical simulation in conjunction with the statistical analysis discussed in this paper.

During this digital era, France, like many other countries, is undergoing a transition from paper-based assessments to digital assessments in education. There is a rising interest in technology-enhanced items which offer innovative ways to assess traditional competencies, as well as addressing problem solving skills, specifically in mathematics. The rich log data captured by these items allows insight into how students approach the problem and their process strategies. Educational data mining is an emerging discipline developing methods suited for exploring the unique and increasingly large-scale data that come from such settings. Data-driven methods can be helpful when trying to make sense of process data. However, studies have shown that didactically meaningful findings are most likely generated when data mining techniques are guided by theoretical principles on subjects’ skills. In this study, theoretical didactical grounding has been essential for developing and describing interactive mathematical tasks as well as defining and identifying strategic behaviors from the log data. Interactive instruments from France’s national large-scale assessment in mathematics have been pilot tested in May 2017. Feature engineering and classical machine learning analysis were then applied to the process data of one specific technology-enhanced item. Supervised learning was implemented to determine the model’s predictive power of students’ achievement and estimate the weight of the variables in the prediction. Unsupervised learning aimed at clustering the samples. The obtained clusters are interpreted by the mean values of the important features. Both the analytical model and the clusters enable us to identify among students two conceptual approaches that can be interpreted in theoretically meaningful ways. If there are limitations to relying on log data analysis in order to determine learning profiles, one of them is the fact that this information remains partial when it comes to describing the complete cognitive activity at play, the potential of technology-enriched problem solving situations in large-scale assessments is nevertheless obvious. The type of findings this study produced is actionable from teachers’ perspective in order to address students’ specific needs.

Computational advancements in the last couple of decades have brought forth a new era of international large-scale assessments (ILSAs) in which the administration of computer-based tests is becoming the norm. Beyond collecting the correct/incorrect answers for each item, computer-based assessments are also able to collect a range of actions performed by the respondents in the computer testing application during the course of the test administration. Both respondents’ actions – starting a unit, clicking a button, spending time until inputting or submitting an answer, and so forth – and their overall behavior – what they do with their keyboards, mice, and even their own eyes – are recorded as a new set of data called “process data,” which are normally time-stamped and can be stored in the so-called log files. There may be plenty of useful insight into the respondent’s cognitive process, and process data can potentially become a relevant element in the scoring process of an assessment, validate test score interpretations, to name a few possibilities for the analysis of such data. This chapter aims to contribute to the body of knowledge in this area by offering (1) an introduction to process data, what kind of data it contains, its relation to the cognitive process, and how it can be organized into a proposed six-layered ecological framework that facilitates its analysis; (2) a literature review of 37 seminal and state-of-the-art publications produced in the last decade on the topic, which are then analyzed both in their chronological perspective as well as in how they fit into the ecological framework; and (3) a discussion of the potential and limitations of using process data in the assessment framework, including a view of what could be the next steps in the analysis of process data from ILSAs.KeywordsComputer-based assessmentProcess dataLog filesEcological frameworkILSA

Background: Online learning became a popular mode of education in the recent Corona Virus Disease-2019 (COVID-19) pandemic. For medical field in India, online mode of learning and assessment is novel. Aims and Objectives: This study aims at determining the attitude of students, identifying unacknowledged problems and their possible solutions associated with online learning, and to explore the future scope of online learning and assessment for medical education in India. Materials and Methods: The study population consisted of 199 medical students who had taken continuous six months of online learning and assessment in the 2020 COVID-19 pandemic period over varzious platforms such as Google Classroom, Zoom Meet, Cisco Webex etc. A questionnaire was made on Google form and distributed through online platforms, and students’ responses with informed consent were collected and analyzed. Responses were in form of Likert scale that ranged from “Strongly agree” to “Strongly disagree.” Results: About 41.4% of participants agreed that online learning and assessments save time and resources. About 78.3% of students agreed that technical errors make it difficult to continue online classes. About 78.3% of students agree that responsibility of learning is on students. About 77.7% of students agreed that academic integrity and honesty are vital in online assessment. Conclusion: If online classes can be improved by solving technical errors, better infrastructure, and faculty training, it may be continued as a supplement to classroom education for lectures, but not for practicals.

The purpose of this research is to describe the planning, implementation, and obstacles in implementing CBT. It is a qualitative research using phenomenological approach. The subjects of this research are Accounting teachers and students in Vocational High School 1 Turen. The method for collecting the data are interview and documentation. The validation of data was conducted by the triangulation of sources. Data analysis using Miles and Huberman technic. The research shows that (1) Planning Computer Based Test (CBT) for midterm and final test includes the formulation of objectives, selecting officers, preparing the system, facilities, and infrastructures needed for the test, socialization for teachers and students, preparing the test items and CBT simulation. (2) The implementation of the computer-based test (CBT) for midterm and final test is the responsibility of Proctor and technician. Teachers' role during the CBT only as the supervisor of the test and assess the result. (3) Problems in the implementation of CBT are the trouble of internet connection, electricity, and the chance for the student to browse easily. There are some strategies to overcome those problems such as giving a clear instruction as the supervisor of CBT, preparing enough genset, empowering wifi connection and server for the stability of internet connection.

Undergraduates’ acceptability and attitude towards CBT as the general mode of testing in undergraduate programs were considered in this study. The survey research design was adopted; the population consists of the students in the Faculty of Education at the University of Benin. A sample of 200 was drawn from the total number of students by random sampling. The instrument – Students Attitude and Acceptability of the CBT Scale- was used for data collection, it was developed by the researchers, and the instrument had a reliability coefficient of .076 and .082 for sections B and C respectively. Descriptive statistics (Mean and Standard Deviation) were used in answering the research questions while inferential statistics (t-test and ANOVA) were used for hypotheses testing. All statistical data analysis were done using the IBM-SPSS statistical package version 20. Findings revealed that the students have positive Attitude towards the CBT and the level of acceptability of the CBT was found to be high. However, there was no significant difference in the attitude and acceptability by sex, but there was a difference in acceptability by area of study whereas there was none in attitude. It was recommended among others that CBT be introduced as the general mode of testing in the undergraduate programmes except otherwise, in areas where it is not practicable.

In February 2007 the German TerraSAR-X satellite will be launched and the TerraSAR-X mission will enter its approximately 5 months commissioning phase. At that time, the challenging developments on both sides, the advanced high- resolution multi-mode SAR instrument on the one hand and the corresponding sophisticated TerraSAR-X ground segment on the other hand will prove correct interaction and functioning. Screening and processing of the SAR data is the task of the DLR developed TerraSAR Multi Mode SAR Processor TMSP. Preceded by data reception, transcription including decryption and followed by archiving, cataloguing and product delivery, processing of the data by the TMSP is the central part of the SAR data workflow implemented in the Payload Ground Segment PGS. Space and ground segment have been subject to intense complete system testing on ground. Here, the compatibility of SAR instrument commanding, SAR instrument operations and subsequent SAR data processing has been successfully proven for the various acquisition modes of the sensor. Compliance of specified and measured product performance has been investigated as far as possible utilizing simulated point target SAR data. However, the real challenge will be the screening and SAR processing of TerraSAR-X data acquired in orbit and linked down to the receiving station. Therefore, the complete reception and processing chain will be properly tuned and adjusted to the properties of the received TerraSAR-X payload data. The TMSP algorithms have to be configured, e.g. thresholds for calibration pulse analysis, estimation window sizes for SAR data analysis, parameterization of estimation algorithms. Also the configuration of product variants with respect to resolution and radiometric quality will be checked and refined. This paper presents the very first experiences in reception, transcription, screening and processing of TerraSAR-X data with respect to performance, throughput and quality. During the TMSP checkout phase the compatibility of instrument commanding and SAR processing have to be verified and the accordance of SAR performance prediction and the obtained product performance and quality have to be investigated. First characteristics of the SAR data with respect to raw data statistics, calibration pulse analysis and Doppler centroid measurements will be shown. As far as available examples of SAR image products featuring the different image modes, Stripmap, ScanSAR and Spotlight at different incidence angles and polarizations will be displayed and a first estimate of product performance parameters will be given.

Introduction. Benz[a]pyrene is one of the most dangerous carcinogens of group I according to the IARC classification; it is metabolically activated by enzymes of the cytochrome system P450 (CYP) with the formation of reactive metabolites capable of damaging cells and DNA, which leads to carcinogenesis. The HepaRG cell line, which exhibits active enzymes of phases I and II of biotransformation, serves as a model for studying polycyclic aromatic hydrocarbons (PAH) metabolism and evaluating the effect of CYP modulators ketoconazole, α-naphthoflavone and 6-formylindolo[3,2-b]carbazole (FICZ) on cell metabolism and the possibility of cytoprotection when exposed to benz[a]pyrene.The aim of the study was to evaluate metabolic changes under the cytoprotective action of ketoconazole, α-naphthoflavone and FICZ against the background of benz[a]pyrene toxicity in the HepaRG cell line model.Material and methods. To identify metabolomic biomarkers of PAH intoxication in HepaRG cells, changes in the metabolomic profile were analyzed using high-resolution chromatographic mass spectrometry. Potential biomarkers were identified through statistical data analysis methods.Results. The possibility of using a metabolomic analysis of conditioned media to identify markers of benz[a]pyrene exposure and evaluate the effects of CYP modulators in vitro using the HepaRG cell line has been shown. Changes in the metabolomic profile of conditioned media after exposure to benz[a]pyrene on HepaRG cells in the presence of α-naphthoflavone indicate a decrease of the PAHs toxic effect on cells.Limitations. The study was performed in vitro on a cell culture without analyzing the dose-dependent changes in the metabolome after exposure to the compounds studied and does not allow extrapolating the results obtained to a living organism.Conclusion. According to the results of the metabolomic profiling of the HepaRG cellular media, a number of metabolomic markers were identified: 5-hydroxytryptophol, lysophosphatides, arginine and glutamic acid, which indicate the cytoprotective effect of α-naphthoflavone against the background of the toxic effect of benz[a]pyrene.Compliance with ethical standards. The study does not require the submission of a biomedical ethics committee opinion or other documents.Authors’ contribution: Ilyushonok S.K. – collection and processing of material, statistical processing, writing a text; Rogovskaya N.Yu. – collection of material, writing a text; Gladchuk A.S. – editing; Sharapov Ya.A. – writing a text; Perevoznikov I.E. – writing a text; Beltyukov P.P. – data processing and interpretation, text editing; Podolskaya E.P. – research concept and design; Babakov V.N. – the concept and design of the study, editing. All co-authors are responsible for approving the final version of the article and ensuring the integrity of all parts of the article.Conflict of interests. The authors declare no conflict of interest.Funding. The work was carried out within the framework of the state contract (2224388100182000000000000/25.456.22.9), cipher “Monitoring-SOZ”.Received: August 08, 2025 / Accepted: October 2, 2025 / Published: November 19, 2025