







































 

 

 

 

 

 

Research on the Influence of Teacher Variables 

on Students’ Mathematical Achievements

Shengnan Bai, Jiwei Han, Canhui Li

Northeast Normal University, Changchun, China 

 

Abstract. Based on the TIMSS2015 data, this study applied a hierar-

chical linear model to explore the influence of teacher variables on stu-
dents’ mathematics scores. Teacher variables were composed of teacher 

characteristic variables, teacher teaching variables, and teacher profes-

sional development variables. The teacher’s characteristic variables 
were teaching age, gender, education, mathematics major, and mathe-

matics education. Teachers’ teaching variables were teaching expecta-

tions, teaching cooperation, teaching enthusiasm, classroom discussion, 
multimedia use, attention to homework, and emphasis on exams. Teacher 

professional development variables had mathematics knowledge train-
ing, mathematics education training, and mathematics curriculum train-

ing. Multi-layer linear analysis found that in the teacher’s characteristic 

variables, the teacher’s teaching age, gender, education, and mathemat-
ics major have a significant effect on the students’ mathematics scores; 

In the teacher’s teaching variables, teachers’ teaching expectations, 
teaching enthusiasm, class discussion, and multimedia use have a signif-

icant impact on students’ mathematics scores. In the teacher profession-

al development variables, participation in mathematics knowledge train-
ing and mathematics education training had a significant positive impact 

on students’ mathematics scores. 
Best Evid Chin Edu 2019; 3(2):347-360. 

Doi: 10.15354/bece.19.ar1266. 

Keywords: Teacher Characteristic Variables; Teacher Teaching Variables; 

Teacher Professional Development Variables; Mathematics Scores

 

 

 
 

Correspondence to: Jiwei Han, Ph.D., Professor, School of Mathematics and Statistics, Northeast Normal Univer-

sity, Changchun, China. Email: hanjw617@nenu.edu.cn. 

About the Author: Shengnan Bai, Ph.D., School of Mathematics and Statistics, Northeast Normal University, 

Changchun, China. Email: baisn012@nenu.edu.cn. 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 348 

Problem 

N the evaluation of education quality, the learning effect of students has always 

been regarded as an important dimension to examine the effect of education. 

Among them, student achievement is one of the main indicators reflecting the effect 

of learning. Teachers are often considered to be the main factor affecting students’ aca-

demic performance. Therefore, exploring the influence of teacher variables on students’ 

academic performance through research has always been an essential research issue in 

educational research. The research of this problem can provide theoretical basis for im-

proving students’ learning, promoting the fair distribution of educational resources and 

the effective training of teacher education. It is a critical problem that has been continu-

ously explored in the field of education. 

Throughout the study of the influence of teacher variables on students’ aca-

demic performance, it can be roughly divided into three categories. The first category 

was to study the influence of teacher characteristic variables on students’ academic per-

formance. The conclusions from this problem study are not the same. Huang and Xin 

found that teachers’ gender, teaching age, education, and majors had no significant in-

fluence on students’ academic performance (Huang & Xin, 2007); whereas Zhang 

found that teachers’ gender, teaching age, and majors had a significant impact on stu-

dent achievement (Zhang, 2012). The second category was to study the influence of 

teachers’ teaching variables on students’ academic performance. Teachers’ teaching 

variables such as teacher’s teaching plan, teaching behavior ( Peterson, et al., 1978) 

(class introduction, classroom language, classroom questions, classroom feedback, class 

waiting, class summary (Huang, et al., 2009; Zhou & Bao, 2012; Huang, 2016; Gao & 

Zhang, 2016; Cao & Yu, 2017), teaching strategy (Walsh-Cavazos, 1994), teaching 

attitude (Bülent & Erden, 2006) were important factors affecting students’ academic 

performance, and some teachers’ teaching variables such as teacher’s teaching coopera-

tion and classroom discussion had no significant influence on students’ academic per-

formance (Zhang, 2010). The third category was to study the influence of teacher pro-

fessional development variables on students’ academic performance. Zhang and Xin 

found that teachers’ preparation methods, peer listening courses, and teaching and re-

search participation had no significant effect on student achievement (Zhang, 2010); but 

Zhao and Yan found that teachers’ preparation ways, training participation, peer-to-peer 

assessment, and teaching and research participation have a significant impact on student 

achievement (Zhao, et al., 2013). From these studies we can find that although the 

teacher variables were closely related to the students’ academic performance, the con-

clusions had not yet reached a consistent agreement. There are two reasons for this: first, 

the measurement tools developed by the researchers were different, and the quality of 

the different measurement tools was directly related to the research results. Herlihy et al 

 
 

About the Author: Canhui Li, School of Mathematics and Statistics, Northeast Normal University, Changchu, 

China. Email: lich380@nenu.edu.cn. 

Conflict of Interests: None. 

I 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 349 

indicated that the validity and reliability of the teacher evaluation system affect the con-

clusion of the role of teacher variables on student achievement (Herlihy, et al., 2005); 

second, the statistical analysis methods of choice are different, previous studies have 

not considered the Hierarchicaled nesting structure of teacher and student data, and tra-

ditional regression analysis of teacher variables and student variables as independent 

variables may cause bias in the results. 

The research on the influence of teacher variables on students’ academic per-

formance has made two new developments in recent years. On the one hand, the quality 

of the measuring tools has improved. With the international emphasis on basic educa-

tion, the emergence of international evaluation projects such as TIMSS and PISA has 

largely contributed to the improvement of the quality of measurement tools, which has 

led researchers to pay attention to the development of measurement tools, thereby pro-

moting the quality of measurement tools. On the other hand, the methods of education 

statistics have been improved and improved. Samples of social science research often 

come from different levels and units. The resulting data is usually multi-level nested 

data, and the general statistical methods have problems for the processing of nested data. 

Therefore, with the continuous development of social sciences, Hierarchical linear 

models have begun to enter the eyes of educational researchers. This model solves the 

limitations of traditional regression analysis in dealing with multi-level nested data and 

becomes a new method of educational statistics research. In this context, we try to di-

rectly use the eighth-grade measurement data in TIMSS2015, and use the Hierarchical 

Linear Model (HLM) to analyze the influence of teacher variables on students’ mathe-

matics scores. In order to find teacher variables that can promote student achievement, 

this study proposes the following three questions: 

 Question 1: How does the teacher characteristic variable affect student math-

ematics scores? 

 Question 2: How does the teacher’s teaching variable affect student mathemat-

ics scores? 

 Question 3: What effect does the teacher’s professional development variable 

have on student mathematics? 

Methodology 

Data Selection 

The TIMSS project focuses on the development of students’ mathematical and scientific 

academic achievements. TIMSS2015 collects information from students, teachers, and 

schools and encodes data by standardizing tests, questionnaires, and observing video 

lessons. 

In TIMSS2015, a total of 45 countries participated in the eighth-grade stand-

ardized test. This study selected data from all countries participating in the eighth grade 

test and referred to it as TIMSS2015 data. All data is from the official TIMSS website 

(http://timssandpirls.bc.edu/timss2015/). 

http://timssandpirls.bc.edu/timss2015/


Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 350 

The subjects of this study were selected from the data of TIMSS2015. After 

eliminating the missing values of the relevant variables and satisfying at least 20 stu-

dents per teacher, there are 45,321 student samples, the ratio of male to female is 1:1.4; 

the number of teachers is 1,806, and the ratio of male to female is 1:1.2. 

Variable Selection and Description 

Based on the data of TIMSS2015, this study explored the influence of teacher variables 

on students’ mathematics scores based on the control of student variables in the nested 

relationship between students and teachers. Based on previous researches, the recent 

exploration of student family factors, teacher teaching factors and teacher professional 

development factors were introduced. At the same time, the TIMSS2015 standardized 

test questions and questionnaires were also adjusted and introduced into the study. 

Therefore, the dependent variable of this study is the student’s mathematics score. The 

independent variable is divided into two layers: 

 The first level is the student variables, which are divided into student back-

ground variables and learning motivation variables. Among them, student 

background variables: gender, age, family book collection, home computer, 

home learning desk, mobile phone, parental education (mother education, fa-

ther education); learning motivation variables: mathematics learning interest, 

mathematics self-confidence. According to the Pearson correlation analysis, the 

mother’s education and father’s education are highly correlated, so the princi-

pal component analysis is used to form a new variable of parental education; 

other variables are not highly correlated. Table 1 shows the definition and 

scoring of student variables. 

 The second level is the teacher variable, which is divided into teacher charac-

teristic variables, teacher teaching variables and teacher professional develop-

ment. Among them, teacher characteristic variables: teaching age, gender, edu-

cation, mathematics major, mathematics education major; teacher teaching var-

iables: teaching expectation, teaching cooperation, teaching enthusiasm, class-

room discussion, multimedia use, attention to homework, emphasis on exams; 

teacher profession development variables: mathematics knowledge training, 

mathematics education training, mathematics curriculum training. After Pear-

son correlation analysis, the teacher variables did not reach a high level of cor-

relation. Table 2 shows the definition and scoring method of teacher variables. 

Data Analysis 

Based on the data of TIMSS2015, this paper uses SPSS22.0 to organize data and use 

HLM6.08 for Hierarchical linear analysis. Taking the student’s mathematics score as 

the dependent variable, without adding the independent variable, a zero model is estab-

lished to test whether the sample data is suitable for Hierarchical linear analysis. On this 

basis, students and teacher variables are added to the previous model one by one to es- 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 351 

Table 1. Definition and Scoring of Student Variables. 

Variable Variable Description Scoring Method 

Student Variable 

Age 
Calculated from the year 
of birth and month 

Continuous type 

Gender Student gender 
Discrete 2- point scoring, 1 for women, 2 
for men 

Family Collection Family book collection 
Discrete 5- point scoring, the higher the 
score, the more books 

Home Computer 
Is there a computer in the 
house? 

Discrete 2- point scoring, 1 is yes, 2 is no 

Home Desk 
Does the student have a 
desk? 

Discrete 2- point scoring, 1 is yes, 2 is no 

Mobile Phone 
Does the student have a 
mobile phone? 

Discrete 2- point scoring, 1 is yes, 2 is no 

Parental Education   
(new variables formed by mother's edu-
cation and father's education) 

Math Interest 
Do students like to learn 
mathematics? 

Discrete 4- point scoring, the higher the 
score, the lower the mathematical inter-
est 

Mathematics Self-
Confidence 

Learning mathematics 
confidence 

Discrete 4- point scoring, the higher the 
score, the lower the confidence 

 

 

tablish five models. For example, model 2 is based on model 1 to add student learning 

motivation variables, and finally form a complete Hierarchical linear model. On the 

basis of controlling student variables, the influence of teacher variables on students’ 

mathematics scores is investigated. 

The specific model is expressed as follows: 

First level: student level 

Dependent variable: 

            

 

Second level: teacher level 

                                          

 

subscripts "j" and "i" in the expression of the Hierarchical linear model represent the 

teacher’s number and the student’s individual number, respectively, and "ij" is the i
th

 

student in the j
th

 class. β0j represents the mathematical average of the students taught by 

the j
th

 teacher. If the difference between teachers reaches a significant level, it is neces-

sary to further analyze which factors of the teacher make the students’ mathematics 

scores show significant differences. 

The establishment of a Hierarchical linear model is presented in Table 3. 

 

 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 352 

Table 2. Teacher Variables of the Definition and Scoring. 

Variable Variable Description Scoring Method 

Teacher Variable     

Teacher Characteristic Variable 

Teaching Age Years of teaching Continuous type 

Gender   Consistent with student gender 

Education   
Discrete 7- point scoring, the higher 
the score, the higher the education 

Mathematic Major 
Whether the field of study is 
mathematics 

Discrete 2- point scoring, 1 is yes, 2 
is not 

Mathematics Educa-
tion 

Is the field of study a math-
ematics education? 

Discrete 2- point scoring, 1 is yes, 2 
is not 

Teacher Teaching Variable 

Teaching Expecta-
tion 

Teachers' expectations for 
classroom teaching 

Discrete 5- point scoring, the higher 
the score, the lower the expectation 

Teaching Coopera-
tion 

Degree of cooperation be-
tween teachers 

Discrete 4- point scoring, the higher 
the score, the lower the cooperation 
frequency 

Teaching Passion Degree of love for teaching 
Discrete 4- point scoring, the higher 
the score, the lower the level of en-
thusiasm 

Class Discussion 
Frequency of discussion 
among students in the class-
room 

Discrete 4- point scoring, the higher 
the score, the lower the discussion 
frequency 

Multimedia Use 
Whether to use multimedia 
in classroom teaching 

Discrete 2- point scoring, 1 is yes, 2 
is not 

Pay Attention To 
Homework 

Frequency of arranging math 
assignments 

Discrete 5- point scoring, the higher 
the score, the more important 

Pay Attention To 
The Exam 

Emphasis on classroom 
tests 

Discrete 3- point scoring, the higher 
the score, and the less important it is. 

Teacher Professional Development Variable 

Mathematical 
Knowledge Training 

Whether to participate in 
mathematics knowledge 
training 

Discrete 2- point scoring, 1 for partic-
ipation, 2 for not participating 

Mathematics Educa-
tion Training 

Whether to participate in 
mathematics education train-
ing 

Discrete 2- point scoring, 1 for partic-
ipation, 2 for not participating 

Mathematics Course 
Training 

Whether to participate in 
mathematics training 

Discrete 2- point scoring, 1 for partic-
ipation, 2 for not participating 

 

 

Results 

According to the statistical principle of the Hierarchical linear model, in order to ex-

plore the influence of teacher variables on students’ mathematics scores, the first is to 

establish a zero model to estimate the contribution of teacher-to-teacher differences and 

intra-teacher differences (student differences) to the total difference, if the difference 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 353 

Table 3. Establishment of Hierarchical Linear Model. 

 
Student Variable Teacher Variable 

Zero 
Model 

    

Model 
1 

Student background variable: 
Gender, age, family book collec-
tion, home computer, home desk, 
mobile phone, parental education 

  

Model 
2 

Students learn motivation varia-
bles: 
Mathematical learning interest, 
mathematics self-confidence 

  

Model 
3 

  
Teacher feature variables: 
Teaching age, gender, education, mathemat-
ics major, mathematics education 

Model 
4 

  

Teacher teaching variables: 
Teaching expectations, teaching cooperation, 
teaching enthusiasm, class discussion, multi-
media use, attention to homework, and em-
phasis on exams 

Model 
5 

  

Teacher professional development variables: 
Mathematical knowledge training, mathemat-
ics education training, mathematics course 
training 

 

 

between teachers is significant, it is necessary to further analyze which teacher variables 

make the students’ mathematics scores show significant differences. The zero model 

results are presented in Table 4. 

It can be seen from Table 4 that the intra-class correlation coefficient ICC of 

the student’s math score is 0.68, which means that about 68.42% of the total difference 

in the student’s math score is caused by the teacher variable. At the same time, when 

the ICC is greater than 0.138, the difference between groups is significant (Dedrick, et 

al., 2009). This indicates that there is a significant difference between teachers, that is, 

the difference in mathematics scores of different teachers has reached a significant level 

(p < 0.001), which is consistent with previous empirical findings (Li & Ni, 2006). 

Because of the significant differences among teachers, this study adds the three 

aspects of teacher variables to the second level of independent variables, and examines 

their influence on students’ mathematics scores, thus providing some theoretical refer-

ences for the improvement of teachers’ teaching process. Due to space limitations, the 

full results of each model are listed in Table 5. 

The Influence of Teacher Characteristic Variables on Students’ 

Mathematical Achievements 

 

 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 354 

Table 4. Estimation of Differences between Teachers and within 
Teachers in Mathematics Scores. 

 
Variance Standard Error Contribution Rate 

Inter-Teacher 8,900.70 94.34 68.42% 

Intra-Teacher 4,107.92 64.09 31.58% 

 

 

Model 3 is to examine the impact of teacher characteristic variables on students’ math-

ematical performance. The survey results showed: teacher’s teaching age (b = 1.09, p < 

0.001), gender (b = -8.98, p = 0.025), education (b = 48.84, p < 0.001), mathematics 

major (b = 12.20, p = 0.005). The impact on students’ mathematics scores has reached a 

significant level. Compared with teachers with shorter teaching ages, teachers with 

longer teaching ages will have better mathematics scores for teachers who teach older 

classes. Female teachers have better math scores than those taught by male teachers. 

The higher the teacher’s academic qualifications, the better the mathematics scores of 

the students in the taught class will be. Students in classes taught by non-mathematics 

teachers have higher math scores than those who were taught by teachers who graduat-

ed from mathematic major. In addition, the teacher characteristic variable explained 

15.12% of the student’s average math score difference. 

The Influence of Teachers’ Teaching Variables on Students’ 

Mathematical Achievements 

Model 4 includes the teacher’s teaching variables based on Model 3, aiming to examine 

the influence of teacher’s teaching variables on students’ mathematics scores. The re-

sults of Model 4 indicate: teacher’s teaching expectations (b = -23.53, p < 0.001), teach-

ing enthusiasm (b = 9.72, p = 0.002), class discussion (b = 18.01, p < 0.001), multime-

dia use (b = -11.90, p = 0.005) has a significant impact on students’ math scores. The 

higher the teacher’s teaching expectation and the higher the frequency of multimedia 

use, the more helpful the mathematics scores of the students in this class. The higher the 

number of class discussions, the less effective the teaching. The teacher’s enthusiasm 

for teaching has a negative predictive ability for students’ mathematics scores. This may 

be due to the occurrence of the "Simpson’s Paradox" or the poor original grades of the 

corresponding students of the teachers. The specific reasons still need further investiga-

tion and research. In addition, the teacher’s teaching variables reduce the variance of 

students’ mathematics scores to 6556.75, explaining 8.85% of the students’ average 

math scores. 

The Influence of Teachers’ Professional Development Varia-

bles on Students’ Mathematical Achievements 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 355 

Model 5 is based on Model 4 and incorporates three variables of teacher professional 

development. The main examination is the influence of teacher professional develop-

ment (mathematics knowledge training, mathematics education and mathematics train-

ing) on students’ mathematics scores. The results of Model 5 analysis show that teach-

ers’ mathematics knowledge training (b= -9.98, p = 0.041) and mathematics education 

training (b = -14.15, p = 0.005) have significant influence on students’ mathematics 

scores. The mathematics course training (b = 8.11, p = 0.078) had no significant effect 

on the student’s mathematics scores. This shows that teachers’ participation in mathe-

matics knowledge training and mathematics education training will have a positive ef-

fect on classroom teaching, so that the average mathematics score of the students taught 

is higher. These three variables explain the student’s average math score difference of 

1.13%. 

Conclusion and Discussion 

Teachers are a pivotal factor influencing students’ learning outcomes. The study found 

that, in overall, the addition of teacher variables has led to a significant decline in teach-

er variability. The teacher characteristic variable explained 15.12% of the difference in 

the average mathematics score of the students; the teacher teaching variable explained 

8.85% of the difference in the average mathematics score of the students; the teacher 

professional development variable explained 1.13% of the difference in the average 

mathematics score of the students. Specifically, the results of each variable’s impact on 

student mathematics are also presented in Table 5. These findings provide a basis for 

improving teachers’ classroom teaching, strengthening teachers’ professional develop-

ment, and formulating relevant educational policies. 

The Teacher’s Teaching Age Has a Significant Impact on the 

Student’s Mathematical Performance 

The results in Table 5 show that the teacher’s teaching age can have a significant pre-

dictive effect on students’ mathematics scores. At the same time, in the literature that 

has explored the relationship between teachers’ teaching age and students’ mathematics 

scores, the conclusions are not completely consistent. For example, the relationship be-

tween teachers’ teaching age and the student’s learning is positively related, or only 

affects the student’s learning to a certain extent, or the two do not matter (Huang & Xin, 

2007; Zhang, 2012; Zhang, 2010; Xie, et al., 2008; Xin, et al., 2004). This shows that it 

is not the teacher’s teaching age that has a direct impact on the students’ mathematics 

scores. However, These factors affect teachers’ understanding of the knowledge and 

grasp of the teaching process, making teachers with a longer teaching age pay more 

attention to the combination of mathematical knowledge and students’ cognitive level 

and laws, thus showing that the students' mathematical results will be better. This shows 

that relevant education departments and schools should not exclude new teachers. They 

should pay more attention to the development trend of new teachers and provide more 

 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 356 

Table 5. The Influence of Teacher Factors on Students' Mathematics 
Scores. 

 
  Fixed Effect Random Effect 

Hierarchical 
Linear 
Model 

Variable 
Regression 
Coefficient 
(B) 

SEM 
Inter-
teachers 

Intra-
teachers 

Interpretation 
Rate 

Zero Model       8,900.70 4,107.92   

 Model 1 

      7,196.61 3,822.46 19.15%/6.94% 

Student Background Variable 

Gender 4.04*** 0.70       

Age -6.92*** 0.61       

Family Book 
Collection 

7.87*** 0.31       

Home Com-
puter 

7.31*** 0.67       

Home Desk 0.19 0.75       

Mobile 
Phone 

1.38 0.83       

Parental 
Education 

8.66*** 0.50       

Model 2 

  
 

  8,474.77 3,204.58 /16.16% 

Student Learning Motivation Variable  

Mathematical 
Learning 
Interest 

-7.31*** 
0.39 
  

      

Mathematics 
Self-
Confidence 

-25.23*** 0.52       

Model 3 

      7,193.02 3,204.52 15.12 % / 

Teacher Characteristic Variable 

Teaching 
Age 

1.09*** 
0. 
18 

      

Gender -8.98* 4.01       

Education 48.84*** 3.36       

Mathematic 
Major 

12.20** 4.26       

Mathematics 
Education 

-5.25 3.82       

Model 4 

      6,556.75 3,204.09 8.85 % / 

Teacher Teaching Variable 

Teaching 
Expectation 

-23.53*** 2.45       

Teaching 
Cooperation 

3.94 2.29       

Teaching 
Passion 

9.72** 3.11       

Class Dis-
cussion 

18.01*** 2.33       



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 357 

(Continued) 

 

Multimedia Use -11.90** 4.20       

Pay Attention To 
Homework 

-1.13 1.84       

Pay Attention To The 
Exam 

-5.21 3.90       

Model 5 

      6,482.58 3,204.15 1.13 % / 

Teacher Professional Development Variable 

Mathematical 
Knowledge Training 

-9.98* 4.89       

Mathematics Educa-
tion Training 

-14.15** 4.94       

Mathematics Course 
Training 

8.11 4.61       

Note: *: P<0.05, **: P<0.01, ***: P<0.001. Interpretation rate refers to the percentage of decrease in 
teacher-to-teacher variation between teachers and teachers compared to the previous model after adding 
a new factor. 

 

 

channels for the communication between new teachers and expert teachers. This is more 

conducive to narrowing the differences in teaching mathematics among students. 

Teacher’s Academic Qualifications, Mathematics Knowledge 

Training, and Mathematics Majors Will Significantly Affect 

Students’ Mathematics Scores 

The results in Table 5 show that the higher the teacher’s academic qualifications, the 

better the mathematics scores of the students in the taught class will be. Students who 

have participated in mathematics knowledge training have better math scores. Students 

in classes taught by non-mathematics teachers have higher math scores than those who 

were taught by teachers who graduated from mathematic major. Generally speaking, 

teachers with high academic qualifications have more knowledge of mathematics, and 

teachers who participate in mathematics knowledge training will have a better under-

standing of the methods of teaching mathematics. Comparing the influence of these 

three variables on students’ mathematics scores, it is found that the mathematics 

knowledge and methods of teaching that teachers have had positive effects on students’ 

mathematics scores. However, students in classes taught by non-mathematics teachers 

will have higher math scores. There may be two reasons for this. First, compared to 

non-mathematics teachers, those graduated from mathematic major have more 

knowledge, but their ability to teach mathematical knowledge to students in the corre-

sponding grade is not as good as non-mathematics teachers. Second, the focus of the 

two types of teachers is different from that of the current basic education. 

Teachers without background of mathematics major will pay more attention to 

the training of students’ mathematics skills. Teachers of mathematics majors pay more 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 358 

attention to the cultivation of students’ mathematical thinking and mathematics thinking. 

However, for the improvement of students’ test scores, it is better to train students’ 

math skills. In addition, mathematics education is designed to help students master the 

mathematics, skills, ideas, and methods necessary for modern life and further learning. 

Therefore, in the process of teaching, teachers can not choose the sea tactics solely for 

the test results, nor can they focus only on the improvement of mathematical thinking 

and neglect the students’ ability to accept at this stage. This shows that the cultivation 

of mathematical knowledge and skills, mathematical ideas and methods complement 

each other. Teachers should pay attention to this and organically combine them, so that 

students can understand the ideas and methods of mathematics in the training of math-

ematics knowledge and skills. At the same time, it is recommended that the education 

department or school regularly organize mathematics knowledge training to provide 

more ways for teachers to master the methods of teaching mathematics. 

The Influence of Teachers’ Mathematics Education Major and 

Mathematics Education Training on Students’ Mathematical 

Achievement 

The results in Table 5 show that there is no significant difference in the mathematics 

scores of the students taught by the mathematics education majors and the non-

mathematics education teachers. Students who have participated in mathematics educa-

tion training have better math scores. Comparing the influence of two variables on stu-

dents’ mathematics scores, it is found that teachers master more mathematics education 

theories have positive effects on students’ mathematics scores. However, the teaching 

effects of teachers in mathematics education and non-mathematics education teachers 

are not significantly different. The reason may be that the teaching of traditional math-

ematics education theory does not use actual cases as a carrier. Therefore, teachers can-

not truly apply the learned education theory to classroom teaching, thus making the ef-

fect of educational theory learning greatly compromised. In view of this, it is recom-

mended to combine educational theory and practice when teaching educational theory, 

to provide teachers with more practical teaching cases, or to answer the problems en-

countered by teachers in real teaching. In addition, the relevant education departments 

are also encouraged to regularly organize mathematics education training, and strive to 

implement education theory into classroom teaching. 

Overall, the study has made some meaningful conclusions. At the same time, 

there are problems in this study, through the interpretation of inter-teacher variation. 

We find that the addition of teacher variables does cause a large decline in teacher-to-

teacher variation, but there are still large inter-teacher differences. There are two rea-

sons for this: First, the TIMSS study is conducted once every four years, so the stu-

dent’s math scores are not continuous, and the characteristics of the student’s current 

math scores cannot be linked to the teacher factors at this time. Secondly, the student’s 

academic performance itself is affected by many factors, and is also affected by the ini-

tial mathematics scores, and the students’ mathematics scores themselves may also dif-



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 359 

fer. Therefore, the influence of teacher variables on students’ mathematics scores is still 

somewhat exploratory. 

 

 

 

 

 

 

References

Bülent, A., & Erden, M. (2006) The effects of 

primary school teachers’ attitudes towards 

the mathematics achievement fourth grade 

students by gender. J Edu Faculty, 8(1): 13-

21. 

Cao, Y.M., Yu, G.W. (2017) Research on the 

critical level of the critical pedagogical be-

haviors in middle school mathematics class-

room. J Math Edu, 26(1): 2-6. 

Dedrick, R.F. (2009) Multilevel modeling: A 

review of methodological issues and applica-

tions. Rev Edu Res, 79(1): 69-102. 

Gao, X., & Zhang, B. (2016) Research on the 

evaluation and compilation of question 

strings by high school mathematics teachers. 

J Math Edu, 25(3): 66-70. 

Herlihy, C. (2005) State and local efforts to in-

vestigate the validity and reliability of scores 

from teacher evaluation systems. Teach Coll 

Rec, 116(1): 463-465. 

Huang, H., & Xin, T. (2007) The influence of 

teacher’s classroom teaching behavior on 

students’ academic achievements: A cross-

cultural study. Psychol Develop 

Edu ,23(4):57-62. 

Huang, X.F., Pang, Y.L., & Li, S.Z. (2009) In-

heritance and development of teaching be-

havior of mathematics classroom teachers: 

Comparative study of three video lessons. J 

Math Edu, 18(6): 54-57. 

Huang, Y.C. (2016) Four elements of teacher’s 

classroom teaching behavior. J Math Edu, 

25(1): 72-74. 

Li, Q., Ni, Y.J. (2006) Multi-level analysis of the 

influence of teacher variables on pupils’ 

math learning achievements. J Teach Edu, 

18(3): 74-80. 

Peterson, P.L., Marx, R.W., & Clark, C.M. 

(1978) Teacher planning, teacher behavior, 

and student achievement. Am Edu Res J, 

15(3): 417-432. 

Walsh-Cavazos, S.A. (1994) Study of the effects 

of a mathematics staff development module 

on teachers’ and students’ achievement. 

Texas Tech Univ, Texas. 

Xie, M., Xin, T., & Li, D.W. (2008) The influ-

ence of teacher qualification and career de-

velopment factors on students’ mathematical 

achievements: A cross-cultural comparison. 

Stud Psychol Behav, 6(2): 124-129. 

Xin, T., Xu, Z., & Tatsuoka, K. (2004) Linkage 

between teacher quality, student achieve-

ment and cognitive skills: A rule-space 

model. Stud Edu Eva, 30(3): 205-223. 

Zhang, W.J. (2010) The influence of teacher 

variables on mathematical achievements in 

the fourth grade of primary school: A value-

added study. Chin J Edu, 06(2): 69-76. 

Zhang, Y.M. (2012) An empirical study of the 

influence of teacher factors and student fac-

tors on students’ academic achievement: A 

Hierarchical linear model analysis based on 

large-scale test data. J Teach Edu, 24(4):56 -

62. 

Zhao, J.et al. (2013) The impact of professional 

development practice of Chinese teachers on 

student achievement: an analysis based on 

surveys in five cities. Global Edu Outlook, 

42(2): 22-33. 

Zhou, C., & Bao, J.S. (2012) The strategy of 

forming students’ high level mathematical 

thinking: The viewpoint of frontline teachers. 

J Math Edu, (4): 37-39. 



Bai, et al. Teacher Variables and Students’ Mathematical Achievements. 

Vol.3, No. 2, 2019 360 

Received: 02 August 2019 

Revised: 22 September 2019 

Accepted: 19 October 2019 

 

 

The Chinese version of this article has been published in Teacher Edu Res 2019; 31(3):70-76. The 

English version has been authorized for being publication in BECE by the author(s) and the Chi-

nese journal. 

白胜南, 韩继伟, 李灿辉. 教师变量对学生数学成绩影响的研究. 教师教育研究, 2019, 

31(3):70-76. 

 


	Article-ShennanBai-BECE_title_30Nov2019
	Article-ShennanBai-MainText-BECE_30Nov2019

