Screening for Potential, Assessing for Achievement: A Study of Instrument Validity for Early Identification of High Academic Potential in Norway

Campbell, Janine Anne; Idsoe, Ella Cosmovici; Størksen, Ingunn

doi:10.1080/00313831.2022.2042735

Cited by 2 publications

(3 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many different ways to assess student achievement [22][23][24]. Numerous studies assessed students' performance and assessment in general [25][26][27], whereas other research assessed students for a particular goal, including academic accomplishment [5]. For each of the activities listed above, some cutting-edge research studies that assessed student performance in various ways are discussed below.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Similarly, in [11], the authors provided an overview of EDM by going through its core concepts. Both studies included summaries and analyses of the EDM industry and its processes, as well as inclusive learning analytics, but they did not adhere to the requirements for a systematic literature analysis [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28].…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Student Knowledge Assessment Using Machine Learning Techniques

Alruwais

Zakariah

2023

Sustainability

View full text Add to dashboard Cite

The process of learning about a student’s knowledge and comprehension of a particular subject is referred to as student knowledge assessment. It helps to identify areas where students need additional support or challenge and can be used to evaluate the effectiveness of instruction, make important decisions such as on student placement and curriculum development, and monitor the quality of education. Evaluating student knowledge assessment is essential to measuring student progress, informing instruction, and providing feedback to improve student performance and enhance the overall teaching and learning experience. This research paper is designed to create a machine learning (ML)-based system that assesses student performance and knowledge throughout the course of their studies and pinpoints the key variables that have the most significant effects on that performance and expertise. Additionally, it describes the impact of running models with data that only contains key features on their performance. To classify the students, the paper employs seven different classifiers, including support vector machines (SVM), logistic regression (LR), random forest (RF), decision tree (DT), gradient boosting machine (GBM), Gaussian Naive Bayes (GNB), and multi-layer perceptron (MLP). This paper carries out two experiments to see how best to replicate the automatic classification of student knowledge. In the first experiment, the dataset (Dataset 1) was used in its original state, including all five properties listed in the dataset, to evaluate the performance indicators. In the second experiment, the least correlated variable was removed from the dataset to create a smaller dataset (Dataset 2), and the same set of performance indicators was evaluated. Then, the performance indicators using Dataset 1 and Dataset 2 were compared. The GBM exhibited the highest prediction accuracy of 98%, according to Dataset 1. In terms of prediction error, the GBM also performed well. The accuracy of optimistic forecasts on student performance, denoted as the performance indicator ‘precision’, was highest in GBM at 99%, while DT, RF, and SVM were 98% accurate in their optimistic forecasts for Dataset 1. The second experiment’s findings demonstrated that practically no classifiers showed appreciable improvements in prediction accuracy with a reduced feature set in Dataset 2. It showed that the time required for related learning objects and the knowledge level corresponding to a goal learning object have less impact.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Evaluating Student Knowledge Assessment Using Machine Learning Techniques

Alruwais

Zakariah

2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…To ensure the construction of a valid and reliable test to measure an item and a person ability, each of these stages must be properly carried out. In line with this Janine et al (2022) the metrics of reliability and validity pertain to how well a method or test measures an outcome. All assessments that strive to produce a genuine estimate of a test taker's competence and support legitimate inferences like…”

Section: Introductionmentioning

confidence: 96%

Determining the Validity and Reliability of Business Studies Achievement Test for Senior Secondary Schools in Nigeria

2023

FeD

View full text Add to dashboard Cite

The study aimed at ascertaining the validity and reliability of business studies achievement test for senior secondary schools in Nigeria. The study adopted an instrumentation design. The area of the study was Anambra State. The population of the study consisted of all the 23,077 students and 586 business studies teachers in two hundred and sixty-one senior secondary schools in Anambra State. Multi-stage sampling techniques involving two stages were adopted for this study. Thus proportionate stratified and purposive sampling techniques were employed at the first and second stages respectively. This was used to determine the sample size of the study, which was 630 made up of 393 students in senior secondary school one and 238 business studies teachers in Anambra State. The instrument Business Studies Acheivement Test (BSAT) was face validated by five experts, while the content validation was carried out by using a table of specifications which consisted of content areas and behavioral objectives structured after Bloom's taxonomy of cognitive domain. To generate items for Business Studies Achievement Test, the instruments were distributed to 238 business studies teachers by the researcher and his assistants in their respective schools during the 2021/2022 session. Business Studies Acheivement Test (BSAT) was used for data collection, the BSAT was a single multiple-choice objective test with an expected response of either pass (1) or fail. This helped in determining the construct validity of the instrument as well generating items for the business studies achievement test. The generated instrument was pilot tested in six secondary schools in the six educational zones in Anambra state. The developed instrument was found to be valid and reliable with an estimate of 0.96 obtained for placing students in business studies in senior secondary schools in Nigeria. Based on the findings of the study it was recommended that teachers, researchers, and relevant agencies should Relevant educational agencies, teachers, researchers should endeavor to establish the validity of any instrument they develop for any educational purpose to ensure the items are valid and relevant to the subject objective also instrument for educational purposes should be reliable, this will facilitate the better decision for students’ placement and achievement.

show abstract

Screening for Potential, Assessing for Achievement: A Study of Instrument Validity for Early Identification of High Academic Potential in Norway

Cited by 2 publications

References 43 publications

Evaluating Student Knowledge Assessment Using Machine Learning Techniques

Evaluating Student Knowledge Assessment Using Machine Learning Techniques

Determining the Validity and Reliability of Business Studies Achievement Test for Senior Secondary Schools in Nigeria

Contact Info

Product

Resources

About