Easy Question Research Articles

Analysis of mathematics exam practice questions for 6th grade elementary students

Students possess varying abilities in problem-solving, ranging from adeptness to struggles or even inability. This study employs a descriptive content analysis design to scrutinize official documents containing practice exam questions. Data collection involved documentation study, with subsequent analysis utilizing validated assessment sheets and rubrics endorsed by experts. The descriptive method facilitated the analysis and interpretation of data meaning, while quantitative analysis processed research data. The study focused on mathematics exam practice scores, revealing several problematic questions. Out of 25 questions, 11 were deemed acceptable, 8 required correction, and 6 were rejected, necessitating replacement. Questions requiring correction or replacement exhibited a low question discriminating power index, failing to effectively gauge students' proficiency in practice exams. Moreover, the majority of practice test questions posed considerable difficulty for students. Out of the 25 questions, students encountered 1 very difficult, 10 difficult, 10 moderates, and only 4 easy questions, which they struggled to solve adequately. Consequently, the findings underscore the importance of revising practice exam questions to ensure they accurately assess students' abilities and cater to a diverse range of skill levels.

Analisis Hasil Ujian Sekolah Berstandar Nasional di Sekolah Dasar Kecamatan Bulu pada Muatan Matematika

<p>This research aims to analyze the level of difficulty of the questions and mastery of basic competencies according to the level of difficulty of the questions that have been found in the Mathematics subject in the national standard elementary school exam results in Bulu sub-district, Temanggung. This research uses a quantitative approach using descriptive methods. The population in this study is data from the USBN results in the Bulu sub-district with a sample of 35 questions answered by 613 grade 6 students, in terms of the level of difficulty the questions were classified as sufficient or medium. This can be seen from the calculation results which show that of the 30 multiple-choice questions that are classified as difficult questions, there are 4 questions (13%), 17 medium questions (57%), and 9 easy questions (30%). Meanwhile, for the 5 essay questions in the difficult category, 1 question (20%), 4 medium questions (80%), and 0 easy questions (0%).</p><p><strong><br /></strong></p><p><strong>Keyword: </strong>item analysis, USBN, level of difficulty, mathematics</p>

Analisis Butir Soal Penilaian Akhir Tahun Pelajaran Bahasa Arab Kelas VII MTs Al-Ma’arif Rakit Banjarnegara Jawa Tengah

Analysis of question items is very important in learning evaluation because it determines the quality of the evaluation tool used. The aim of this research is to analyze the results of the final semester Arabic evaluation at the Al-Ma’arif Junior High School. The researchers conducted a series of tests consisting of measurements of validity, reliability, differential strength, difficulty level, and debilitating power. The method used in this research is quantitative descriptive method. The population of the study was 25 students of the 7th grade in the IT High School Building Amal School Year 2022/2023. As for the sample used, the final evaluation of the Arabic semester of the teaching year 2022/2023 was documentation and observation. The result of this study is; in the validity test found 30% of questions can be said to be valid. Then the remaining 70% is declared invalid because it has a significance value <0,05 and the value of R count is smaller than R Table. Then in the reliability test found using Cronbach's Alpha method, and the element of this question is worth 0.625 with the number of questions (N of items) amounting to 40. Difference shows 6 questions in the category "Bad once in distinction, 25 questions in category "badly once in differentiation", 6 issues in the categories "Sufficient enough in distinguishment", 3 questions with category "Good once in distinctness" whereas with the question of category "good once in discrimination" was not found. Difficulty level there are 25 questions with very easy questions category, 14 questions with easy categories, 1 questions with difficult categories. The fractional strength of the answer option A indicates that the question is not working well with the student as the data indicates more fractionals of the question are rejected than accepted. The same applies to options B, C, and D. Thus, a question that does not meet the revised criteria can be concluded or a new question created and tested again. This research can be used as evaluation material for teachers to create quality evaluation tools. Further research can focus on discussing the development of superior evaluation instruments.

The Development of Assessment Instrument Based on Wordwall to Measure Process Science Skills for Environmental Pollution

This study aims to determine the validity, practicality, and effectiveness of an assessment instruments based on Wordwall in measuring students' science process skills of junior high school science learning in environmental pollution materials. This research is a type of development research (Research and Development) that refers to the procedure of the Borg and Gall development model which has stages including research and information collecting, planning, develop the preliminary form of the product, product design validation, main product revisions, main field testing, operational product revision, operational field testing, final product revision. The subject of the study was a grade VII student of SMPN 2 Ajung, Jember. This study used instruments in the form of validation sheets, practicality sheets, student questionnaire sheets, and science process skills tests. Data analysis techniques used in research consist of validity test analysis, practicality test analysis, and effectiveness test analysis. The results showed that out of 20 multiple-choice questions developed from aspects of material, construction, and language based on the assessment of 3 validators were declared very valid and worthy of use without revision. The twenty multiple-choice questions tested have valid question item validity. The assessment instrument has a reliability value of 0.612. The difficulty level analysis obtained 5 easy criteria questions, 14 medium criteria questions, and 1 difficult criteria question. The effectiveness of the assessment instrument obtained the average results of students' science process skills of 55%, including the sufficient category and student response questionnaires of 81%, so they were declared very good.

Evaluation of ChatGPT pathology knowledge using board-style questions.

ChatGPT is an artificial intelligence chatbot developed by OpenAI. Its extensive knowledge and unique interactive capabilities enable its use in various innovative ways in the medical field, such as writing clinical notes and simplifying radiology reports. Through this study, we aimed to analyze the pathology knowledge of ChatGPT to advocate its role in transforming pathology education. The American Society for Clinical Pathology Resident Question Bank 2022 was used to test ChatGPT, version 4. Practice tests were created in each subcategory and answered based on the input that ChatGPT provided. Questions that required interpretation of images were excluded. We analyzed ChatGPT performance and compared it with average peer performance. The overall performance of ChatGPT was 56.98%, lower than that of the average peer performance of 62.81%. ChatGPT performed better on clinical pathology (60.42%) than on anatomic pathology (54.94%). Furthermore, its performance was better on easy questions (68.47%) than on intermediate (52.88%) and difficult questions (37.21%). ChatGPT has the potential to be a valuable resource in pathology education if trained on a larger, specialized medical data set. Those relying on it (in its current form) solely for the purpose of pathology training should be cautious.

Analisis Kesalahan Siswa dalam Menyelesaikan Soal Integral Bentuk Tak Tentu

In this research, researchers analyzed students' errors in solving questions on the topic of indefinite form integrals. The purpose of this research is to determine the patterns of student errors that often occur when working on indefinite form integral questions, as well as the factors that cause them. The subjects in this research were 25 grade 11 high school students. This research was carried out by providing a test sheet in the form of essay questions on the subject matter of indefinite integrals, totaling 5 questions with 1 category of easy questions, 3 categories of medium questions, and 1 category of difficult questions. Based on the results of data analysis, it was obtained that for question number 1 (an easy category question), 100% of respondents answered correctly. Question number 2 (medium category question), 93.3% of respondents answered correctly. Question number 3 (medium category question) 54% of respondents answered correctly. Question number 4 (difficult category question), 46% of respondents answered correctly. Meanwhile, for question number 5 (difficult category question), only 20% of respondents answered correctly.

Peer Groups, Success and Punishment as Correlates and Contributors to Pupils’ Academic Performance in Mathematics in Bamenda II Subdivision, Cameroon

The study set out to determine the relative contributions of peer groups, pupils’ success and punishment as motivating factors in mathematics classrooms on pupils’ academic performance in mathematics in Bamenda II Subdivision, Cameroon. Three specific objectives, three research questions and three hypotheses guided the study. The study adopted the descriptive survey and correlational research designs. One hundred and fifty (150) primary three pupils from two functional public primary schools in Bamenda II Subdivision constituted the population of the study. Using the Krejcie and Morgan table, and employing the stratified proportionate sampling technique, 108 pupils were selected to form the sample of the study. Data was collected using a questionnaire. The questionnaire was validated by experts and its reliability of 0.82 assured using the Cronbach Alpha. The data was analyzed using Pearson product moment correlation (r), the coefficient of determination (r2) to answer the research questions and the two tailed p-value for correlation to test the hypotheses at 0.05 level of significance. The findings of the study revealed that peer groups, pupils’ success and punishment in mathematics classrooms all contribute significantly to their academic performance in mathematics. It was therefore recommended that punishment should be carefully utilized by the classroom teachers in a bid to keep them focused on the subject both in and out of the classroom. Furthermore, teachers should always use very easy questions either during set induction or during the introduction of new mathematical concepts—questions within the reach of all pupils. This is because perceived success in mathematics by pupils has proven to be a good motivating factor to students’ in the subject. Lastly, group counseling should be done by both the school counselors and the classroom teachers, aimed at encouraging pupils to assist each other academically, especially in mathematics.

Development of Minimum Competency Assessment (AKM) Model Test for Learning Biology Excretion System Materials for Measuring Science Literacy in SMA Negeri 1 Bandar

The results of the 2018 PISA study show that the ability of students in Indonesia in scientific literacy shows an average score of 389, so scientific literacy is still low. The Ministry of Education and Culture released the Independent Learning Program to replace the National Examination (UN) assessment system with a Minimum Competency Assessment (AKM) and character surveys. Therefore, it is necessary to develop AKM questions on the excretory system material to measure scientific literacy in students. The purpose of this study is to describe the characteristics of the AKM model questions, analyze differences in student learning outcomes and student responses related to the AKM model questions at SMA Negeri 1 Bandar, Batang Regency. This research refers to the development of questions according to Djemari Mardapi which consists of 7 stages that have been modified, The subjects of this study were students of class XI MIPA 4 (experimental class) and class XI MIPA 5 (control class). Analysis of the items on the test results of the AKM model test using Microsoft Excel software and SPSS v.19 software. That is to calculate validity, difficulty index, discriminating power and reliability estimation. Data on the results of differences in student learning averages were measured using a test instrument for the AKM model and student response data was measured using a questionnaire instrument. The results of the study obtained from 25 questions that were tested on students there were 23 valid questions and 2 invalid questions, the results of the difficulty level on the items obtained 12 very easy questions and 13 easy questions, the results of the differentiating power were in the good category and the results of the reliability estimation of items The questions are included in the very high category, and the results of the difference in the average learning outcomes in the experimental class and control class are 73.61> 61.44. The increase in the average learning outcomes of experimental and control class students in their scientific literacy skills was significantly different with an N-gain value of 58.28 > 25.08. The experimental class gave a very good response and the control class gave a good response to the development of the AKM model questions. Thus, it can be concluded that the development of the AKM model test questions can measure students' scientific literacy skills.

PENGEMBANGAN INSTRUMEN TES DIAGNOSTIK THREE-TIER PADA PEMBELAJARAN SISTEM EKSKRESI BERDIFERENSIASI

Diagnostic tests are strategies used in differentiated learning to adjust the course of learning based on students' existing learning needs. There are several types of diagnostic tests, including the three-tier diagnostic test. Explaining a three-tier diagnostic test instrument that is suitable for use in studying differentiated excretory systems is the aim of this research. The research method is research & development with 4D as a development model. 33 students of class The instrument developed consisted of 20 questions which were then tested for content and empirical validity. Based on the results of validity testing, an average of 92% was obtained, which shows that the instrument is very valid. Empirical validity testing obtained 13 valid questions and 7 invalid questions. In addition, a value of 0.732 was obtained from the reliability testing results, which shows that the instrument developed is reliable. Based on the results of testing the level of difficulty of the questions, it was found that there were 3 easy questions, 10 medium ones, and 7 difficult ones, and from the results of testing the differentiating power of the questions it was found that the questions were not good because they had a negative value of 1, bad had 3, fair and good each had 7, respectively. and very well collected 2. Based on content and empirical validity testing, the results showed that only 13 questions could be used in the differentiated excretion learning system, while the other 7 questions could not be used.

Evaluating the Artificial Intelligence Performance Growth in Ophthalmic Knowledge.

We aim to compare the capabilities of Chat Generative Pre-Trained Transformer (ChatGPT)-3.5 and ChatGPT-4.0 (OpenAI, San Francisco, CA, USA) in addressing multiple-choice ophthalmic case challenges. Both models' accuracy was compared across different ophthalmology subspecialties using multiple-choice ophthalmic clinical cases provided by the American Academy of Ophthalmology (AAO) "Diagnosis This" questions. Additional analysis was based on image content, question difficulty, character length of models' responses, and model's alignment with responses from human respondents. χ2 test, Fisher's exact test, Student's t-test, and one-way analysis of variance (ANOVA) were conducted where appropriate, with p<0.05 considered significant. GPT-4.0 significantly outperformed GPT-3.5 (75% versus 46%, p<0.01), with the most noticeable improvement in neuro-ophthalmology (100% versus 38%, p=0.03). While both models struggled with uveitis and refractive questions, GPT-4.0 excelled in other areas, such as pediatric questions (82%). In image-related questions, GPT-4.0 also displayed superior accuracy that trended toward significance (73% versus 46%, p=0.07). GPT-4.0 performed better with easier questions (93.8% (least difficult) versus 76.2% (middle) versus 53.3% (most), p=0.03) and generated more concise answers than GPT-3.5 (651.7±342.9 versus 1,112.9±328.8 characters, p<0.01). Moreover, GPT-4.0's answers were more in line with those of AAO respondents (57.3% versus 41.4%, p<0.01), showing a strong correlation between its accuracy and the proportion of AAO respondents who selected GPT-4.0's answer (ρ=0.713, p<0.01). Our study demonstrated that GPT-4.0 significantly outperforms GPT-3.5 in addressing ophthalmic case challenges, especially in neuro-ophthalmology, with improved accuracy even in image-related questions. These findings underscore the potential of advancing artificial intelligence (AI) models in enhancing ophthalmic diagnostics and medical education.

ChatGPT Conquers the Saudi Medical Licensing Exam: Exploring the Accuracy of Artificial Intelligence in Medical Knowledge Assessment and Implications for Modern Medical Education.

Background The application of artificial intelligence (AI) in education is undergoing rapid advancements, with models such as ChatGPT-4 showing potential in medical education. This study aims to evaluate the proficiency of ChatGPT-4 in answering Saudi Medical Licensing Exam (SMLE) questions. Methodology A dataset of 220 questions across four medical disciplines was used. The model was trained using a specific code to answer the questions accurately, and its performance was assessed using key performance indicators, difficulty level, and exam sections. Results ChatGPT-4 demonstrated an overall accuracy of 88.6%. It showed high proficiency with Easy and Average questions, but accuracy decreased for Hard questions. Performance was consistent across all disciplines, indicating a broad knowledge base. However, an error analysis revealed areas for further refinement, particularly with category (Option) A questions across all sections. Conclusions This study underscores the potential of ChatGPT-4 as an AI-assisted tool in medical education, demonstrating high proficiency in answering SMLE questions. Future research is recommended to expand the scope of training and evaluation as well as to enhance the model's performance on complex clinical questions.

Renewable Portfolio Standards

State-level renewable portfolio standards (RPSs) aim to encourage renewable energy and discourage greenhouse gas (GHG) emissions from the electric power sector in the United States. Do they work? Some prominent government agencies and advocacy groups assert that U.S. renewables growth has been largely due to RPSs. That seems unlikely, given that in most regions, renewables exceed RPS requirements. But it is not an easy question to answer, thanks to interstate trading and the possibility that states with abundant renewable resources might set the most ambitious RPS goals. We combine the best features of four recent academic studies, using ordinary least-squares and instrumental variables approaches. In some specifications, RPSs do appear to reduce the use of natural gas to generate electricity and decrease GHG emissions, while boosting the use of wind and solar power. But the effects are small—consistent with the academic findings and in contrast to the public claims and policy goals.

Introducing standard patient-reported measures (PRMs) into routine maternity care: A pre-implementation qualitative study on women’s perspectives in Finland

BackgroundSystematically using standard patient-reported measures (PRMs) in clinical routines is trending. The International Consortium for Health Outcomes Measurement (ICHOM) has developed condition-specific standard sets of patient-centred measures, one of which is the Pregnancy and Childbirth Standard (PCB) set, where standard PRMs are included. There is limited knowledge on the use of ICHOM PCB set-included PRMs (ICHOM-PCB-PRMs) in routine care. This study investigates women’s perspectives on the future implementation of standard ICHOM-PCB-PRMs in routine maternity care in Finland.MethodsSemi-structured interviews were conducted. Pregnant and postpartum women were asked to evaluate each ICHOM-PCB-PRM in several dimensions, e.g., importance and quality of questions, and to provide their views on future implementation in terms of benefits, difficulties, and practices. With the predefined topics and themes, deductive analysis was applied. Ethical committee approval (HUS 220/880/2015) and research permissions were obtained.Results22 women participated. Participants felt that most of the ICHOM-PCB-PRMs were important, relevant, understandable, and appropriately designed, and agreed that some changes in ICHOM-PCB-PRMs were needed, e.g., adding other important measures, changing the wording, and adding open-ended questions. Women would be hesitant to answer questions honestly if follow-up actions were unclear. Most “outcome” measures could be asked repeatedly as maternal health status changes over time, and “experience” measures could be asked separately for different service providers. Disagreements regarding data collection at birth were observed. PRMs were regarded as a way for women to express their thoughts and feelings. Our participants were concerned about the possible consequences of negatively answering the PREMs questions and the availability of follow-up care. Participants expected that they could answer short and easy questions digitally before appointments, and that instructions and follow-up actions based on their answers should be available.ConclusionICHOM-PCB-PRMs could be applicable in Finnish maternity care, but some modifications may be required. Careful consideration is needed regarding how and when PRMs questions are asked for eliciting more accurate and honest answers and minimizing women feeling judged, embarrassed, or offended. Follow-ups should be available according to women’s responses and needs. This study provides insights on the adoption and implementation of standard PRMs in routine maternity care.

Student Mistakes in Solving Math Story Problems (Class V Study at SDN Sedatigede I Sidoarjo)

In working on word problems in mathematics, students often find answers that are not quite right. The purpose of this research is to describe the pattern of errors when working on math story problems, the factors that cause work errors, and alternative learning solutions to minimize student errors in solving word problems. Research data collection techniques used math tests in the form of story questions and student interviews. Factors that cause students' mistakes in solving math word problems, are the lack of mastery of mathematical concepts and understanding of problems in story problems, students' learning methods are not organized (not every day), they are lacking in doing exercises especially story problems, working on easy questions Of course, if you have difficulty don't ask directly to the teacher or friends who can (students choose to be silent), are not careful in carrying out calculations (addition, subtraction, multiplication and division operations), don't pay close attention to question orders, don't carry out the procedures used to solve problems. story problems and lack of mastery of material concepts related to story problems.

How should the non-Indigenous speak? A discussion of decolonizing academia, positioning, and freedom of speech

ABSTRACT How can we engage in Indigenous research that allows multiple perspectives and knowledge production that is open to epistemic diversity? Answering calls for decolonization of the academy and the need for researchers to do their homework, I use my position as an ‘inbetweener’ in a Sámi, Indigenous context, experiences with peer reviewers as gatekeepers, and theoretical and methodological discussions about Indigenous research, to reflect upon this question. The review processes are seldomly discussed, although reviewers have the power to decide whether research gets published or not. Questions remain on how research should be carried out and who can speak, especially in a Sámi-Norwegian context where Indigenous identity and being Sámi or not by no means are easy questions, due to colonialism and Norwegianization processes. There is confusion regarding how to do research, and what positioning means. I argue that Indigenous methodology raises a dilemma because it one the one hand offers criticism of previous western, ‘dirty’ research, urging non-Indigenous researchers to involve and ‘do their homework’, while simultaneously emphasizing that research should be done by and for Indigenous peoples, with their worldviews as a starting point. Discussions on positionality show that the outsider/insider dichotomy is problematic and that for some scholars, being Indigenous is a precondition for doing valid and important research. Non-Indigenous researchers may be associated with a colonial sin, or shame. Being morally inferior, I argue, makes it difficult for the researcher both to involve, and to have critical approaches. Doing homework should involve being able, and allowed, to engage with Indigenous peoples and societies and go beyond the colonial gaze coloured by the us/them, and victim/sinner dichotomies. Secondly, I also call upon universities, that are now institutionalizing Indigenous perspectives, to take responsibility and offer support to the individual researchers in their struggle to ‘do things right’.

Analysis of science subject questions in class IX junior high school using Anates 4.0 computer program

One of the activities to assess test quality is item analysis, which considers the general quality of the test and the quality of each item. The study was completed after tests were collected and run on all samples, and the results served as a guide for future quality improvement of the test. This study aims to analyze the validity, reliability, level of difficulty, discriminating power, and effectiveness of the distractors on Class IX Science questions at junior high school (SMP) Negeri 9 Batanghari. The data analysis technique used is a quantitative descriptive technique. This research was conducted at SMP Negeri 9 Batanghari. Class IX students of SMP Negeri 9 Batanghari were used as research subjects. The population is all class IX students at SMP Negeri 9 Batanghari, totaling 50 students. The results of this study indicate that, in terms of (1) validity, 25 items (83%) are valid, and five items (17%) are invalid. (2) Reliability: The questions have excellent reliability, with a reliability coefficient of 0.87. (3) Difficulty level: There are seven easy questions (23%), 18 moderate questions (60%), and five difficult questions. (4) The ability to discriminate the items is considered good (moderate, good, and very good) for up to 86% of the questions. (5) The quality of the distractor is considered very good for up to 100% of the questions. Therefore, this science question can be used to assess student learning outcomes at SMP Negeri 9 Batanghari because it fulfills the requirements of a learning evaluation instrument.

Development of PeduliLindungi Application Features Using Design Thinking

The PeduliLindungi app has supported the government in controlling the spread of COVID-19. The purpose of this research is to explore new user needs and create design prototypes for the latest features of the PeduliLindungi application. Design thinking is used to know empathizing, define, ideate, and prototype. Figma Tools was chosen to develop a prototype that visualizes the latest features of the PeduliLindungi application. The results of this research are information about pain points from application users, problem statements, problem solving ideas, and user interface features for the new PeduliLindungi application. Based on the results of user pain points, the features developed are covid-19 event advertisements, Online Registration, Night Mode, Covid-19 educational videos, Tourist Visits, and Vaccine History. Based on the evaluation results using the Single Easy Question (SEQ), it is found that advertising, registration, night mode, and vaccine history are worth 7, meaning that the level of use is basic to use. Tourist visits and educational videos are worth 5.75, which means that improvements need to be made by taking into account notes from respondents during the testing process.

Feasibility Analysis of Items for End of Semester Assesment Questions Class VII Indonesian Subjects of SMPN 1 Ngaglik

This research is to determine the feasibility of the test items on the Indonesian language PAS questions for class VII at SMP Negeri 1 Ngaglik for the 2021/2022 academic year usingdocument analysis research with a quantitative descriptive approach. Sources of research data, namely the test questions grid, PAS question sheets in the form of student responses. The research subjects were 32 class VII B students at Ngaglik 1 Public Middle School for the 2021/2022 academic year. The research variable is the feasibility of the Indonesian PAS questions which consist of 40 questions. The research instruments were in the form of documentation guidelines, documentation collection techniques, and data analysis techniques in this study using descriptive statistical techniques. The research results show(1) validity level of 40 questions100% valid, (2) the reliability of the test items was 0.807 and declared reliable, (3) ITK, namely there were 6 easy questions, 30 moderate questions, 3 difficult questions, and 1 very difficult question; IDB there are 27 questions that are appropriate and 13 questions that are not feasible. (4) distractor analysis, there are 30 items (92.5%) that have very good distractor quality and 3 items (7.5%) that have poor distractor quality.

Conversational pragmatics: memory reporting strategies in different social contexts.

Previous studies in conversational pragmatics have showed that the information people share with others heavily depends on the confidence they have in the correctness of a candidate answer. At the same time, different social contexts prompt different incentive structures, which set a higher or lower confidence criterion to determine which potential answer to report. In this study, we investigated how the different incentive structures of several types of social contexts and how different levels of knowledge affect the amount of information we are willing to share. Participants answered easy, intermediate, and difficult general-knowledge questions and decided whether they would report or withhold their selected answer in different social contexts: formal vs. informal, that could be either constrained (a context that promotes providing only responses we are certain about) or loose (with an incentive structure that maximizes providing any type of answer). Overall, our results confirmed that social contexts are associated with different incentive structures which affects memory reporting strategies. We also found that the difficulty of the questions is an important factor in conversational pragmatics. Our results highlight the relevance of studying different incentive structures of social contexts to understand the underlying processes of conversational pragmatics, and stress the importance of considering metamemory theories of memory reporting.

Evaluating the Performance of ChatGPT in Ophthalmology: An Analysis of Its Successes and Shortcomings

Foundation models are a novel type of artificial intelligence algorithms, in which models are pretrained at scale on unannotated data and fine-tuned for a myriad of downstream tasks, such as generating text. This study assessed the accuracy of ChatGPT, a large language model (LLM), in the ophthalmology question-answering space. Evaluation of diagnostic test or technology. ChatGPT is a publicly available LLM. We tested 2 versions of ChatGPT (January 9 "legacy" and ChatGPT Plus) on 2 popular multiple choice question banks commonly used to prepare for the high-stakes Ophthalmic Knowledge Assessment Program (OKAP) examination. We generated two 260-question simulated exams from the Basic and Clinical Science Course (BCSC) Self-Assessment Program and the OphthoQuestions online question bank. We carried out logistic regression to determine the effect of the examination section, cognitive level, and difficulty index on answer accuracy. We also performed a post hoc analysis using Tukey's test to decide if there were meaningful differences between the tested subspecialties. We reported the accuracy of ChatGPT for each examination section in percentage correct by comparing ChatGPT's outputs with the answer key provided by the question banks. We presented logistic regression results with a likelihood ratio (LR) chi-square. We considered differences between examination sections statistically significant at a P value of < 0.05. The legacy model achieved 55.8% accuracy on the BCSC set and 42.7% on the OphthoQuestions set. With ChatGPT Plus, accuracy increased to 59.4% ± 0.6% and 49.2% ± 1.0%, respectively. Accuracy improved with easier questions when controlling for the examination section and cognitive level. Logistic regression analysis of the legacy model showed that the examination section (LR, 27.57; P= 0.006) followed by question difficulty (LR, 24.05; P < 0.001) were most predictive of ChatGPT's answer accuracy. Although the legacy model performed best in general medicine and worst in neuro-ophthalmology (P < 0.001) and ocular pathology (P= 0.029), similar post hoc findings were not seen with ChatGPT Plus, suggesting more consistent results across examination sections. ChatGPT has encouraging performance on a simulated OKAP examination. Specializing LLMs through domain-specific pretraining may be necessary to improve their performance in ophthalmic subspecialties. Proprietary or commercial disclosure may be found after the references.