Font Size: a A A

Influencing Factors And Prediction Analysis Of Grade One Mathematics In Harbin

Posted on:2021-06-29Degree:MasterType:Thesis
Country:ChinaCandidate:J H MuFull Text:PDF
GTID:2517306314453954Subject:Applied Statistics
Abstract/Summary:PDF Full Text Request
Compared with junior high school,the difficulty of senior one course is greatly improved.When students study mathematics,they need to establish a new way of thinking.After the long summer vacation after the entrance examination,inertia will inevitably increase,which is not conducive to the learning of senior high school curriculum.In this case,it is particularly difficult to make a good transition between junior high school and senior high school and achieve the learning objectives of the first semester of senior high school with high quality.The main content of mathematics teaching in the first semester of senior high school can be divided into three parts:set,function and trigonometric function.The functions and trigonometric functions occupy a large part of the compulsory course and have high requirements for students' abstract thinking,which makes it difficult for students who are just in high school to adapt immediately.In addition,the miscellaneous and knowledge of other modules is not simple for most senior one students.In order to make the freshmen grow up quickly and complete the transition and connection from junior high school to senior high school,it is very important to cultivate students' good learning habits and correct their learning attitude from the beginning of the first semester of senior high school.This paper collects the information of 196 senior high school students who have been studying in key senior high schools and private senior high schools in Harbin in recent three years about the learning situation of mathematics in the first semester of senior high school.Using these data,this paper analyzes the significant factors that affect the final math scores,and establishes a prediction model for the final exam scores based on these important factors.First of all,this paper analyzes the correlation between 14 factors and different types of mathematics final examination scores from three aspects:Students' individual conditions,family environment and teaching resource quality,including gender,source of students,whether to participate in extracurricular remedial classes,boarding or day reading,time spent studying mathematics every day,mathematics scores in the middle school entrance examination,test scores of sets and functions,the test results of the trigonometric function part,the math results of the mid-term examination,the education level of parents,whether the student is the only child,the title of the teacher,the teaching age of the teacher,whether the head teacher is a math teacher,etc.According to the different types of variables,t-test of independent samples,F-test of variance analysis,Pearson chi square test of contingency table analysis and Mann Whitney test are selected respectively U rank sum test and gamma coefficient test are used to make descriptive statistical analysis on the factors that affect the final exam scores.The factors that have significant correlation with the final exam scores are preliminarily screened out,and these important variables are used as the variables in the prediction model.Secondly,according to the timing of the students' tests,this paper analyzes the three types of final examination results in stages,and calculates the accuracy of the prediction.By using the stepwise regression method of multiple regression,the paper establishes a prediction model for the final examination score in stages,and forecasts the final examination score of continuous mathematics.It is proved that the last three stages are obviously better than the first stage,that is,the prediction will be more accurate after the examination score of entering high school.In addition,based on the C4.5 algorithm of decision tree,this paper establishes the prediction model of mathematics grade,forecasts the results of the final examination of mathematics by types and stages,and calculates the accuracy of the prediction of mathematics grade and classification grade.In this way,the students who may be excellent in the final examination can be predicted after each stage examination,and the students who may fail in the final examination can be warned.In addition,the fault tolerance of the model is improved by improving the pass line.The purpose is to strengthen the supervision of students,try to reduce the failure rate of mathematics,and help teachers better achieve the teaching objectives.This paper studies the prediction model of mathematics final examination results in the first semester of Harbin Senior High School,which is significant for teachers to improve the teaching quality of the subjects they teach and reduce the low score rate of students.
Keywords/Search Tags:grade prediction, correlation analysis, decision tree, stepwise regression
PDF Full Text Request
Related items