Perceived teacher support profiles and mathematics engagement, mathematics anxiety, mathematics attitude: A latent profile analysis approach

doi:10.21203/rs.3.rs-5293099/v1

Download PDF

Research Article

Perceived teacher support profiles and mathematics engagement, mathematics anxiety, mathematics attitude: A latent profile analysis approach

https://doi.org/10.21203/rs.3.rs-5293099/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The critical role of perceived teacher support in mathematics learning has been widely recognized, but individual student differences have often been overlooked in previous studies.

Methods

This study adopts a person-centered approach to study the perceived teacher support in the process of mathematics learning, uses latent profile analysis to classify 1314 students, and uses one-way analysis of variance to explore the differences between students with different profiles. Further, a mediation model is established to explore the relationship between different perceived teacher support profiles and and mathematics engagement, mathematics anxiety, and mathematics attitude.

Results

Latent profile analysis identified three profiles: low (5.78%), medium (44.29%), and high perceived teacher support (49.93%). There were no significant differences in gender or grade among students with different profiles. But they show significant differences in mathematics engagement, mathematics anxiety, and mathematics attitude. Further analysis revealed that there are differences in behavioral engagement, cognitive engagement, emotional engagement, classroom anxiety, learning motivation, and learning strategies. The mediation effect analysis showed that students with higher perceived teacher support exhibited stronger mathematics attitudes, which in turn increased their mathematics engagement. However, mathematics anxiety did not serve as a mediating factor.

Conclusion

The study showed that individual differences in perceived teacher support affect students’ mathematics engagement mathematics anxiety, and mathematics attitudes. Students who perceive a higher level of teacher support show a more positive attitude toward mathematics, which promotes mathematics engagement. This study provides empirical evidence for teachers to implement personalized support.

Perceived teacher support

Mathematics engagement

Mathematics anxiety

Mathematics attitudes

Latent profile analysis

In the field of mathematics education, perceived teacher support has been widely recognized and deeply studied as a key factor affecting student learning outcomes [1–3]. Previous studies have found that students’ perceived teacher support is affected by gender and grade. Additionally, they have provided insight into how perceived teacher support can effectively alleviate mathematics anxiety, foster positive learning attitudes, and enhance student engagement in mathematics learning [3–5].

However, it is worth noting that most existing studies on perceived teacher support are variables-centered and based on pluralism and homogeneity assumptions [6–8]. According to the developmental context theory, each student’s perception of and response to teacher support is unique, lacking a uniform developmental pattern and outcome [9]. This implies potential heterogeneity in students’ perceived teacher support, indicating the need for a more detailed, person-centered analysis of how students perceive teacher support. Furthermore, it is important to explore in depth the correlation between students’ different profiles of perceived teacher support and individual characteristics (such as gender and grade), and how these perceptions affect their mathematics engagement, mathematics anxiety, and mathematics attitude in math learning.

The person-centered approach reveals subtle differences in students’ perceived teacher support by capturing the unique mathematics experience of each student [10]. To dissect this phenomenon, we used latent profile analysis to identify different profiles of perceived teacher support. These profiles not only reflect the diversity and complexity of students’ perceived teacher support but also provide the groundwork for further investigation into the underlying influences [6]. Building upon this foundation, and grounded in self-development theory, we further analyzed the specific relationship and mechanism of influence between different perceived teacher support profiles and these variables to elucidate how different perceived teacher support profiles impact students’ mathematics learning [11]. Therefore, the findings from this study offer person-centered insights into the role of perceived teacher support in mathematics learning and contribute to the development of differentiated strategies and support mechanisms to enhance students’ mathematical learning efficacy.

2.1 Perceived teacher support and mathematics learning

Perceived teacher support refers to the care and assistance that students perceive from their educators, covering three aspects: academic, emotional, and competence support [12, 13]. Within the realm of mathematics education, academic support manifests through mathematics teachers’ attentiveness to students’ learning progress and the provision of timely assistance when students encounter challenges. Emotional support is embodied in mathematics teachers’ positive encouragement, unconditional acceptance, deep respect, and strong trust in their students. Competence support entails the encouraging stance adopted by mathematics teachers in motivating students to participate in extracurricular activities and various mathematics competitions [12–14].

Perceived teacher support serves as a crucial catalyst in enhancing students’ mathematics learning efficacy[15, 16]. Self-determination theory provides a comprehensive theoretical framework for our understanding of its role. The theory emphasizes that students’ needs for ability development, autonomy, and social relevance are the core to promote their learning motivation and effectiveness [11, 15]. Teachers provide the necessary support through active interaction, addressing students’ needs across multiple dimensions to enhance mathematics learning effectiveness [2, 17, 18]. Further research indicates that perceived teacher support exerts a broad spectrum of positive effects. It not only positively influences students’ internal driving forces such as learning motivation and interest but also impacts their cognitive processes, learning behaviors, and emotional experiences [19, 20]. Specifically, the supportive role of teachers during the learning process can significantly enhance students’ emotions with mathematics and improve their attitudes toward learning, thus promoting more active and in-depth learning engagement [21, 22]. This comprehensive positive influence collectively serves as a critical driving force for the improvement of students’ mathematics learning outcomes.

2.2 Gender, grade, and perceived teacher support

The existing literature has different views on the relationship between students’ gender and perceived teacher support. Some studies have indicated that girls’ perceived teacher support was significantly higher than boys [23, 24]. The gender socialization hypothesis posits that girls place a higher value on intimacy compared to boys. Consequently, this heightened emphasis on close relationships renders girls more sensitive to teacher support and more influenced by the degree of teacher support they perceive, relative to boys [25]. However, the vulnerability hypothesis (also known as the academic risk hypothesis) suggests that boys may be more likely to feel supported by teachers due to their greater prevalence of serious academic and behavioral issues [25, 26]. However, some studies have also found that gender does not affect perceived teacher support, with no difference in perceived teacher support between boys and girls [27, 28].

For grade, there is some evidence to suggest that the effect of perceived teacher support may differ according to students’ grade levels. For example, a large cross-sectional study indicated that fifth graders and college students reported higher levels of perceived teacher support compared to students at other grade levels in the southeastern United States [29]. However, the studies of basic education in Finland and Norway found that students’ perceived teacher support may decrease slightly as their grade level increases. Younger students reported feeling more supported by their mathematics teachers than older students [30, 31]. The diverse findings from existing research underscore the necessity for further investigation into the relationship between gender, grade, and perceived teacher support.

2.3 Perceived teacher support and mathematics engagement, mathematics anxiety, and mathematics attitude

2.3.1 Perceived teacher support and mathematics engagement

Mathematics engagement refers to the intensity of behavior, emotional level, and the application of cognitive strategies demonstrated by students during mathematics learning activities. It encompasses behavioral engagement, cognitive engagement, and emotional engagement [32, 33]. Behavioral engagement involves actively listening in class, participating in activities, and engaging in discussions. Cognitive engagement refers to engaging in deep learning strategies (e.g., making connections between old and new learning materials), exploring novel problem-solving methods, and evaluating previously acquired skills. Emotional engagement refers to students’ positive or negative feelings about learning and the learning process, including attitudes toward education, a sense of belonging, interest, and related emotional states [32–34].

Empirical studies have pointed out the positive impact of perceived teacher support on mathematics engagement [17, 33]. Specifically, perceived teacher support significantly promotes students’ behavioral, cognitive, and emotional engagement in mathematics learning [2, 20]. This support helps to establish a positive teacher-student relationship, optimize the mathematics teaching process, and stimulate their motivation for mathematics learning. Motivated by this support, students display increased enthusiasm and confidence, enabling them to focus more on academic tasks and actively engage in mathematics learning and classroom interactions [2, 17]. A theoretical explanation for this phenomenon can be found in self-determination theory, which emphasizes that when students feel supported by their teachers, their basic psychological needs—such as autonomy, competence, and relatedness—are satisfied. The fulfillment of these needs further encourages students to actively engage in academic tasks, thus achieving a higher level of learning and personal development [15, 17].

2.3.2 Mathematics anxiety as a mediator

Mathematics anxiety refers to the tension, anxiety, and fear that individuals experience when confronted with mathematics. This form of anxiety includes problem-solving anxiety, test anxiety, and classroom anxiety, which describe students’ anxiety levels when confronted with mathematics problems, mathematics exams, and mathematics classes, respectively [35, 36].

In the field of mathematics, studies have conclusively shown that students’ perceived teacher support can enhance their interest in mathematical activities [3, 37], Moreover, this support can alleviate anxiety associated with problem-solving, testing, and classroom performance [3, 21, 22]. Further analysis showed that higher teacher support is associated with greater academic enjoyment and lower levels of academic hopelessness and mathematics anxiety [3, 33, 38]. In addition, mathematics anxiety has also been regarded as a key factor affecting mathematics learning engagement [39, 40]. Compared to students with high anxiety, those with low mathematics anxiety are more likely to engage in mathematics learning and are less likely to avoid difficult problems or delay learning tasks[41, 42]. Therefore, this suggests that mathematics anxiety may mediate the relationship between perceived teacher support and mathematics engagement. In other words, Perceived teacher support may have influenced students’ mathematics engagement in the first place by influencing students’ mathematics anxiety levels.

2.3.3 Mathematics attitude as a mediator

Mathematics attitude refers to an individual’s tendency to think, feel, perceive, and act in response to the learning or application of mathematics knowledge and skills [43, 44]. It includes three dimensions: learning motivation, learning beliefs, and learning strategies. Learning motivation stems mainly from an interest in mathematics, learning beliefs involve students’ understanding of and confidence in mathematical knowledge and learning processes, and learning strategies involve students’ cognitive styles in mathematics learning cognitive styles [44, 45].

The accumulated empirical evidence strongly supports the critical role of teacher support in shaping students’ attitudes toward mathematics. Greater support from mathematics teachers corresponds to increased student confidence in their mathematical abilities and more positive attitudes toward mathematics [46, 47]. This positive attitude functions as a potent internal driving force, encouraging students to actively engage in mathematics learning [22, 48]. As a result, students are more inclined to invest time and energy in mathematics learning and to engage more deeply in mathematical activities and knowledge exploration [49, 50]. So, we analyze that mathematics attitudes may play a role as a mediating variable between teacher support and mathematics engagement.

2.4 The need for a person-centered approach to profile perceived teacher support

Most previous studies have adopted a variable-centered approach to explore the relationships and interactions between perceived teacher support and factors influencing mathematics learning. Such research assumes a premise that the impact of students’ perceived teacher support on their mathematics learning is generally consistent and homogeneous [6, 10]. This perspective simplifies the complex reality and treats perceived teacher support as an undifferentiated overall variable, thereby ignoring the individual differences and uniqueness that are prevalent within student groups. However, the developmental situation theory points out that the interaction between an individual and the environment is dynamic and diverse, and each student will show a unique development trajectory in the interaction with the environment [9]. This theoretical perspective reveals that students’ perceived teacher support is likely to be individualized and do not have uniform developmental outcomes [51]. Therefore, the previous hypothesis may have ignored the significant heterogeneity displayed by students’ perceived teacher support in actual educational settings.

The person-centered approach provides new perspectives for understanding the heterogeneity of students’ perceived teacher support. It can identify profiles of students with similar perceptual patterns, and explore differences in mathematics learning outcomes among these profiles[10]. Latent profiling analysis (LPA), as an effective tool in this method, reveals the potential correlation between dominant variables by identifying individuals with similar characteristics, highlighting individual differences and development stages, and tracking group heterogeneity [6]. Although person-centered approaches have significant advantages in revealing individual differences, this does not mean that variable-centered approaches should be replaced [52]. The two are complementary: variable-centered methods are good at accurately quantifying predictor variables and revealing the relationship between variables through statistical analysis. Person-centered approaches more comprehensively assess the complexity and interaction of relationships between predictors, starting from the uniqueness and complexity of individuals or groups [53, 54]. Thus, person-centered analyses can provide a rich complement to variable-centered studies.

To sum up, research on perceived teacher support that takes into account students’ heterogeneity is still lacking. Especially, it remains unclear what the latent profiles of perceived teacher support are, whether students’ characteristics (such as gender and grade) impact perceived teacher support latent profiles, and what the relationships and underlying mechanisms are between different latent profiles of perceived teacher support and mathematics learning factors. Therefore, it is necessary to adopt a person-centered research perspective combined with the latent profile analysis method to conduct an in-depth investigation of students’ perceived teacher support and explore how these profiles affect the factors related to students’ mathematics learning.

This research examined perceived teacher support from a person-centered perspective, utilizing LPA to identify latent profiles of perceived teacher support. Additionally, the study investigated the relationships between these profiles and various factors such as gender, grade, mathematics engagement, mathematics anxiety, and mathematics attitude. Furthermore, we explored the underlying relationships and mechanisms between different profiles of perceived teacher support and mathematics engagement, mathematics anxiety, and mathematics attitude. To guide the study, we formulated four specific research questions (RQs).

RQ1. What are the characteristics of the latent profiles of perceived teacher support?

RQ2. Do gender and grade predict different latent profile probabilities of students’ perceived teacher support?

RQ3. Are differences in mathematics engagement, mathematics anxiety, and mathematics attitudes among students influenced by different latent profiles of perceived teacher support?

RQ4. What are the relationships and mechanisms between different latent profiles of perceived teacher support and mathematics engagement, mathematics anxiety, and mathematics attitude?

4.1 Participants

In this study, students in grades 10–12 from three public high schools in China were investigated. Under the Chinese education system, mathematics education ensures the coherence and integrity of the knowledge system through unified teaching materials and teaching arrangements. At the same time, the unified training of teachers aims to ensure that every student can enjoy a high level of teaching services, and ensure that the teaching standards are consistent so that students can follow a similar and efficient learning path under a unified curriculum framework. In terms of class organization, we implement the achievement-oriented class classification system, which not only reflects the differences between classes but also ensures the level of students within the class is similar. Each class size is carefully controlled to 55 students to optimize teaching management. In addition, the unified test ranking mechanism provides a level playing field for students and helps teachers and education departments understand students’ learning.

A total of 1,350 online questionnaires were distributed, and 1,314 valid questionnaires were obtained (effectiveness rate 97.33%) in this study. Among them, 563 were completed by boys (42.85%) and 751 were completed by girls (57.15%), with an average age of 17.50 ± 1.18 years. There were 736 students in 10th grade (56.01%), 520 students in 11th grade (39.57%), and 58 students in 12th grade (4.42%).

4.2 Data collection

In the data collection stage, we strictly follow the principle of informed consent, and obtain the consent of school administrators, high school students, and their guardians, students are voluntary participants, and they can withdraw at any time, using an anonymous questionnaire to protect privacy. With the assistance of the school, students complete an online questionnaire through WJX (www.wjx.cn) covering perceived teacher support, mathematics engagement, anxiety, and attitude, which takes approximately 30–45 minutes. Before filling in the questionnaire, the researcher informed the students that the purpose of the survey was only for academic research, that is, to understand their overall experience and feelings in the process of mathematics learning, and emphasized objective evaluation and real feedback, and the teacher would not get the results filled in by the students. After the questionnaire is completed, the data should be cleaned strictly and unqualified answers should be eliminated by certain principles to ensure objective and effective results. The principles include these: (1) More than half of the questions had the same or missing answers. (2) The response time for each question is less than 2 seconds [55, 56]. Protect student privacy throughout the process to ensure objective and credible results.

4.3 Measures

The reliability of each questionnaire was analyzed using Cronbach’s α coefficients, and their structural validity was evaluated by confirmatory factor analysis (CFA; see Table 1). For reliability, the recommended α is greater than 0.70 [57]. For CFA, the fit metrics that must be reported in general studies include the ratio of chi-square to degree of freedom (χ²/df), the comparative fit index (CFI), the Tuck Lewis index (TLI), the root-mean-square approximation error (RMSEA), and the standardized root mean residual (SRMR). The recommended fitting indexes in the literature are CFI ≥ 0.90, TLI ≥ 0.90, RMSEA ≤ 0.06, and SRMR ≤ 0.10 [58]. Due to the large sample size, the value of χ²/df is too high, so this value cannot be used as a criterion for validity in this study.

4.3.1 Perceived Teacher Support

The perceived teacher support scale was adopted, which was translated and revised by Ouyang[13] based on the original English scale of Babad [59]. This scale has been validated by Liu et al. [12] in the Chinese educational context and can effectively and accurately assess Chinese students’ perceived teacher support. This scale has 19 items and contains 3 dimensions: academic support, emotional support, and competence support (e.g., “The teacher would use encouraging eyes to signal me to get up and answer questions”, “The teacher’s attitude toward me has always been very gentle”, and “The teacher often puts me in charge of things in the class”). Each item is evaluated using a 5-point Likert scale ranging from 1 (completely disagree) to 5 (completely agree), with higher scores on each dimension indicating a higher level of perceived teacher support. Through CFA, 4 items with factor loadings less than 0.6 were deleted.

4.3.2 Mathematics engagement

The mathematics engagement scale was adopted by Liu et al. [33], which was translated and revised based on the original English scale by Wang et al. [60]. This scale has been validated in the Chinese educational environment and can effectively measure students’ mathematics engagement levels. It has 16 items in total, including 3 dimensions: behavioral engagement, cognitive engagement, and emotional engagement (e.g., “I actively engage in class discussions during mathematics class”, “When I study mathematics, I think about how the knowledge is used in real life”, and “I enjoy learning new knowledge in mathematics class”, respectively). Each item is evaluated using a 5-point Likert scale ranging from 1 (completely disagree) to 5 (completely agree). The higher the score is, the greater the student’s investment in mathematics learning.

4.3.3 Mathematics anxiety

The mathematics anxiety scale translated and revised by Zhang [61] based on the original English scale of Richardson and Suinn [62] was selected. This scale has been validated in the Chinese educational setting and can measure students’ mathematics anxiety well. It has 17 items in total, encompassing 3 dimensions: problem-solving anxiety, test anxiety, and classroom anxiety (e.g., “I have trouble concentrating when solving problems”, “I’m always worried about failing a mathematics test”, and “I often feel nervous and uncomfortable in mathematics class”). Each item is evaluated using a 5-point Likert scale ranging from 1 (not at all anxious) to 5 (very anxious). The higher the score is, the more serious the student’s mathematics anxiety is.

4.3.4 Mathematics attitude

The Mathematics Attitude Scale developed by He and Li [44] was used, which has been validated in the Chinese educational setting and can measure students’ attitudes toward mathematics well. It has a total of 27 items, encompassing 3 dimensions: learning motivation, learning beliefs, and learning strategies (e.g., “I’m learning mathematics because it’s fun”, “I’m sure I can do harder mathematics homework”, and “At different learning stages, I will set corresponding mathematics learning goals for myself”, respectively). Each item is evaluated using a 5-point Likert scale ranging from 1 (very inconsistent) to 5 (very consistent). The higher the score is, the more positive the attitude toward mathematics learning. Through CFA, 7 items with factor loadings less than 0.6 were deleted.

Table 1

Reliability and validity of measurement tools
Scale	Dimension	Cronbach’s α	χ²/df	CFI	TLI	RMSEA	SRMR
Perceived teacher support		0.93	6.57	0.94	0.92	0.07	0.06
	Academic support	0.90
	Behavior support	0.88
	Competence support	0.87
Mathematics engagement		0.96	7.11	0.94	0.92	0.07	0.05
	Behavioral engagement	0.92
	Cognitive engagement	0.94
	Emotional engagement	0.95
Mathematics anxiety		0.95	8.09	0.92	0.91	0.07	0.07
	Problem-solving anxiety	0.92
	Test anxiety	0.92
	Classroom anxiety	0.92
Mathematics attitude		0.80	3.93	0.96	0.95	0.05	0.07
	Learning motivation	0.92
	Learning beliefs	0.86
	Learning strategies	0.92
Note: df = degrees of freedom; CFI = comparative fit index; TLI = tucker–lewis index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual.

4.5 Data analysis

In the research, SPSS 26.0 and Mplus 8.3 were used for data management and analysis. To answer RQ1, we used Mplus 8.3 to conduct potential profile analysis and estimated the best model of different combination modes based on three criteria: (1) the lower the AIC, BIC, and BIC (aBIC) are, the better the model fit. (2) For LRT and BLRT, the p-value is significant, indicating that the k-type fitted model performs better than the k-1 type model. Nylund [63]noted that BLRT performs better than LRT. (3) The closer the entropy score is to 1, the greater the probability of accurate classification of an individual. Simplicity and interpretability must also be considered in model selection. This study was conducted to explore the profile of perceived teacher support by students based on LPA, and 2 to 4 types of models were verified to determine the best fitting model. To answer RQ2, based on the LPA results, SPSS 26.0 was used to conduct multiple logistic regression analysis and examine the effects of gender and grade on latent profiles of perceived teacher support. To answer RQ3, SPSS 26.0 was used to conduct a one-way ANOVA to explore the differences between the dimensions of mathematics engagement, mathematics anxiety, and mathematics attitude among students with different profiles of perceived teacher support. To answer RQ4, we further used the PROCESS macro in SPSS to explore the mediating role of mathematics anxiety and mathematics attitudes between different profiles of perceived teacher support and mathematics engagement.

5.1 Common Method Bias

The data in this study were all derived from students’ self-reports and may exhibit common method bias. Statistical analysis of common method deviation was performed using Harman’s single-factor test [64]. The results showed that the variance explanation rate of the first common factor is 29.32% (less than 40%), indicating that there is no serious common method bias exhibited in this study.

5.2 RQ1: Latent profile analysis of perceived teacher support

The profiles of perceived teacher support in the sample were categorized by latent profile analysis. The model parameters of each profile are shown in Table 2, and the fitting indices of 1 to 4 profiles were extracted for model comparison. The AIC, BIC, and aBIC exhibited monotonically decreasing trends as the number of profiles increased; from 3 profiles to 4 profiles, the AIC, BIC, and aBIC values showed a decreasing tendency to level off, and 2 profiles exhibited their highest entropy values. In addition, the LMR values for 4 profiles did not reach statistical significance, which indicates that the fit of the 4-profile model was not significantly better than that of the 3-profile model when considering both simplicity and interpretability. In summary, the 3-profile model was determined to be the optimal potential profile model.

Table 2

Fit Indices for the Four Models using Latent Profile Analysis (N = 1314)
Profile	AIC	BIC	aBIC	Entropy	LMR	BLRT	Group size for each profile
Profile	AIC	BIC	aBIC	Entropy	LMR	BLRT	1	2	3	4
1-profile	62210.94	62366.37	62271.07	-	-	-	1314
2-profile	55806.19	56044.51	55898.39	0.933	0.000	0.000	545	769
3-profile	53671.35	53992.56	53795.62	0.931	0.043	0.000	76	582	656
4-profile	52029.17	52433.27	52185.50	0.932	0.161	0.000	440	41	571	262

According to Fig. 1, the three dimensions of perceived teacher support exhibit a balanced distribution in terms of scores. This indicates that students generally thought that the three dimensions of perceived teacher support were interconnected. In other words, when the level of support in one dimension is high, the level of support in the other dimensions is also high. 5.78%, 44.29%, and 49.92% of students were categorized into profile 1, profile 2, and profile 3, respectively.

A one-way ANOVA was performed using the three dimensions of perceived teacher support as the dependent variable and the three latent profiles as the independent variable. The results showed (Table 3) that the main effect of the latent profile was significant for the three dimensions of perceived teacher support. Post hoc (LSD) test results revealed that students with the first profile had lower scores in the three dimensions than students with other profiles did, which indicated that students in this profile perceived less support from teachers for their mathematics learning and could be referred to as having low perceived teacher support. The students in the second profile score at a medium level in the three dimensions and can be referred to as having medium perceived teacher support. The high school students in the third profile exhibited significantly greater scores for perceived teacher support than those in the other profiles of high school students. The students in this profile perceived teachers to have strong support for their mathematics learning which can be referred to as high perceived teacher support.

Table 3

Descriptive statistics, ANOVA, and Post Hoc (LSD) Tests for the three latent profiles of perceived teacher support on three-dimensional scores
Class	Academic Support	Emotional Support	Competence Support
Class 1(76)	2.67 ± 0.80	2.45 ± 0.80	2.25 ± 0.89
Class 2(582)	4.17 ± 0.44	4.09 ± 0.53	3.06 ± 0.89
Class 3(656)	5.30 ± 0.47	5.34 ± 0.50	4.15 ± 1.08
F(2,1314)	1518.00***	1502.96***	260.19***
Post hoc(LSD)	1 < 2 < 3	1 < 2 < 3	1 < 2 < 3
Effect	0.698	0.696	0.284
Note: ^***p < 0.001

5.3 RQ2: Gender, grade, and latent profiles of perceived teacher support

To further explore the characteristics of students’ perceived teacher support, multiple logistic regression analysis was conducted using the latent profile of perceived teacher support as the dependent variable and gender (girls served as the reference group) and grade (10th grade served as the reference group) as the independent variables. Taking low teacher support as the reference profile, the analysis yielded the odds ratio coefficient (Table 4), which reflects the ratio of different genders and grades in each latent profile of perceived teacher support. The results indicated that gender and grade were not valid predictors of the latent profile grouping of students’ perceived teacher support. Specifically, there were no significant gender and grade level differences between the medium perceived teacher support, high-perceived teacher support, and low-perceived teacher support profiles.

Table 4

Multiple logistic regression of the gender and grade variables on latent profiles of perceived teacher support
		Profile 2			Profile 3
		OR	95%CI		OR	95%CI
Gender	Male	0.99	0.60	1.61	1.27	0.78	2.06
	Female	1			1
Grade	10	1.06	0.24	4.80	0.52	0.12	2.24
	11	0.74	0.17	3.35	0.32	0.08	1.39
	12	1			1
Note: ^p < 0.05, ^**p < 0.001
5.4 RQ3: The three latent profiles of perceived teacher support and mathematics engagement, mathematics attitude, and mathematics anxiety

ANOVA was used to explore the differences between students with different potential profiles of perceived teacher support and mathematics engagement, mathematics anxiety, and mathematics attitude. According to the analysis results (Table 5), students with different potential profiles of perceived teacher support exhibited significant differences in mathematics engagement, mathematics anxiety, and mathematics attitude.

Those in profile 1 (low perceived teacher support) obtained scores significantly lower than those in the other two profiles on mathematics engagement and mathematics attitude. Among these profiles, behavioral engagement, cognitive engagement, emotional engagement, learning motivation, and learning strategies yielded the same results. However, the scores for mathematics anxiety were significantly greater for profile 1 than those for the other two types of students, with the same results for classroom anxiety. Among the three profiles of students, profile 2 (medium perceived teacher support) had scores for mathematics engagement, mathematics anxiety, and mathematics attitude in the middle. Furthermore, behavioral engagement, cognitive engagement, emotional engagement, classroom anxiety, learning motivation, and learning strategies exhibited the same results. Students in profile 3 (high perceived teacher support) yielded scores that were significantly greater than did students in terms of mathematics engagement and mathematics attitude, among which behavioral engagement, cognitive engagement, emotional engagement, learning motivation, and learning strategies had the same results. The opposite conclusion is drawn for mathematics anxiety, and the same is true for classroom anxiety.

Table 5

ANOVA and post hoc (*LSD*) tests for the three latent profiles of perceived teacher support on mathematics engagement, mathematics anxiety, and mathematics attitude
	Profile M(SD)			F(2,1314)	Effect	Post hoc (LSD)
	Profile 1	Profile 2	Profile 3	F(2,1314)	Effect	Post hoc (LSD)
ME	2.95 ± 0.88	3.33 ± 0.64	3.77 ± 0.69	92.80***	0.124	1 < 2 < 3
ME_BE	3.19 ± 0.94	3.49 ± 0.68	3.86 ± 0.70	58.99***	0.083	1 < 2 < 3
ME_CE	2.80 ± 1.02	3.15 ± 0.74	3.62 ± 0.81	73.44***	0.101	1 < 2 < 3
ME_EE	2.84 ± 0.97	3.35 ± 0.76	3.82 ± 0.81	87.83***	0.118	1 < 2 < 3
MA1	3.20 ± 0.10	3.10 ± 0.04	2.95 ± 0.03	11.83***	0.009	3 < 2 < 1
MA_PSA	3.44 ± 0.96	3.42 ± 0.78	3.35 ± 0.94	1.06	0.002	3 < 2 < 1
MA_TA	3.19 ± 1.11	3.13 ± 0.90	3.02 ± 1.08	2.19	0.003	3 < 2 < 1
MA_CA	2.96 ± 1.03	2.73 ± 0.89	2.48 ± 1.07	15.05***	0.022	3 < 2 < 1
MA2	3.02 ± 0.54	3.20 ± 0.52	3.52 ± 0.55	69.04***	0.095	1 < 2 < 3
MA_LM	2.85 ± 1.02	3.10 ± 0.88	3.62 ± 0.89	64.07***	0.089	1 < 2 < 3
MA_LB	3.34 ± 0.90	3.30 ± 0.74	3.31 ± 0.97	0.11	0.000	2 < 3 < 1
MA_LS	2.86 ± 0.83	3.21 ± 0.67	3.63 ± 0.72	77.14***	0.105	1 < 2 < 3
Note: ME means Mathematics engagement; ME_BE means Mathematics engagement_ Behavioral Engagement; ME_CE means Mathematics engagement_ Cognitive Engagement; ME_EE means Mathematics engagement_ Emotional Engagement; MA₁ means Mathematics anxiety; MA_PSA means Mathematics anxiety_ Problem Solving Anxiety; MA_TA means Mathematics anxiety_ Test Anxiety; MA_CA means Mathematics anxiety_ Classroom Anxiety; MA₂ means Mathematics attitude; MA_LM means Mathematics attitude_ Learning Motivation; MA_LB means Mathematics attitude_ Learning Beliefs; MA_LS means Mathematics attitude_ Learning Strategies. ^p < 0.05, ^**p < 0.001

5.5 RQ4: Mechanisms between the three latent profiles of perceived teacher support and mathematics engagement, mathematics anxiety, and mathematics attitude

Considering the mediating effects of mathematics anxiety and mathematics attitude, we explored the relationships between different profiles of perceived teacher support, mathematics engagement, mathematics anxiety, and mathematics attitude. The SPSS macro program compiled by Hayes (2017) was used to test the mediating effect based on bootstrapping. Specifically, Model 4 of the model test was applied, and the sampling was performed 5,000 times. The perceived teacher support profile was used as the independent variable, and the perceived teacher support profile was dummy-coded (D1: 0 = low perceived teacher support, 1 = medium perceived teacher support, D2: 0 = low perceived teacher support, 1 = high perceived teacher support). After controlling for gender and grade, with mathematics anxiety and mathematics attitude as mediating variables, a dual mediation model was established using mathematics learning investment as the dependent variable.

The results of Fig. 2 show that, when low perceived teacher support is used as a reference, the predictive effect of medium perceived teacher support on mathematics anxiety is not significant (β = -0.12, p > 0.05). However, it has a significant positive impact on mathematics attitude and mathematics engagement (0 < β_s < 0.5, p < 0.05). High perceived teacher support has a significant negative impact on mathematics anxiety (β = -0.34, p < 0.05) and a significant positive impact on mathematics attitude and mathematics engagement (0.3 < β_s < 0.9, p < 0.05). The predictive effect of mathematics anxiety on mathematics engagement is not significant (β = -0.02, p > 0.05), and mathematics attitude positively affects mathematics engagement (β = 0.63, p < 0.05).

The results of the mediation effect are shown in Table 6. We found that the confidence intervals for the mediation effect test of mathematics anxiety between medium and high perceived teacher support and mathematics engagement were [-0.01, 0.02] and [-0.01, 0.04]. The confidence intervals for the mediation effect test of mathematics attitudes between medium and high perceived teacher support and mathematics engagement were [0.06, 0.32] and [0.38, 0.65]. Therefore, mathematics attitudes have a mediating role between perceived teacher support profiles and mathematics engagement, with mediation effect sizes of 0.19 and 0.51; This implanted high perceived teacher support, compared to medium perceived teacher support, has a stronger effect on students’ mathematics attitude.

Table 6

The Effect Sizes of the Mediation Effects
	95% confidence interval
	Effect Size (SE)	LLCI	ULCI
Total effects
M_PTS\(\:\to\:\)Mathematics engagement	0.51(0.11)	0.29	0.73
H_PTS\(\:\to\:\)Mathematics engagement	1.10(0.11)	0.88	1.32
Indirect effects
Indirect effect 1: M_PTS\(\:\to\:\text{M}\text{a}\text{t}\text{h}\text{e}\text{m}\text{a}\text{t}\text{i}\text{c}\text{s}\:\text{a}\text{n}\text{x}\text{i}\text{e}\text{t}\text{y}\to\:\)Mathematics engagement	0.003(0.01)	-0.01	0.02
Indirect effect 2: H_PTS\(\:\to\:\text{M}\text{a}\text{t}\text{h}\text{e}\text{m}\text{a}\text{t}\text{i}\text{c}\text{s}\:\text{a}\text{n}\text{x}\text{i}\text{e}\text{t}\text{y}\to\:\)Mathematics engagement	0.01(0.01)	-0.01	0.04
Indirect effect 3: M_PTS\(\:\to\:\text{M}\text{a}\text{t}\text{h}\text{e}\text{m}\text{a}\text{t}\text{i}\text{c}\text{s}\:\text{a}\text{t}\text{t}\text{i}\text{t}\text{u}\text{d}\text{e}\to\:\)Mathematics engagement	0.19(0.07)	0.06	0.32
Indirect effect 4: H_PT\(\:\to\:\text{M}\text{a}\text{t}\text{h}\text{e}\text{m}\text{a}\text{t}\text{i}\text{c}\text{s}\:\text{a}\text{t}\text{t}\text{i}\text{t}\text{u}\text{d}\text{e}\to\:\)Mathematics engagement	0.51(0.08)	0.38	0.65
Direct effects
M_PTS\(\:\to\:\)Mathematics engagement	0.32(0.09)	0.14	0.49
H_PTS\(\:\to\:\)Mathematics engagement	0.57(0.09)	0.40	0.75
Note: Effect Size, standardized effect; SE, standard error; LLCI, lower level confidence interval; ULCI, upper level confidence interval.

Previous research on perceived teacher support has mostly used a variable-centered approach, ignoring the existence of different profiles of perceived teacher support and possible differences in their associations with other factors. Focusing on a person-centered approach, this study used LPA to examine students’ different perceived teacher support profiles and explored whether gender and grade predict the probability of student profiles. In addition, the study examined the relationships and underlying mechanisms between different perceived teacher support profiles and mathematics learning factors.

6.1 RQ1: Latent profile analysis of perceived teacher support

This study revealed three different potential profiles of students’ perceived teacher support: low perceived teacher support (5.78%), medium perceived teacher support (44.29%), and high perceived teacher support (49.92%). Significant differences were observed in the scores across the three dimensions of perceived teacher support. Students in the high perceived teacher support profile scored higher on all dimensions of perceived teacher support, followed by those in the medium perceived teacher support profile, and those in the low perceived teacher support profile scored the lowest on all dimensions. These findings reflect the heterogeneous nature of the characteristics of perceived teacher support and indicate that a person-centered study can reveal individuals’ differences and enable potential profile analysis to determine the underlying characteristics of the variables [7, 8].

6.2 RQ2: Gender, grade, and latent profiles of perceived teacher support

The results show that there are no significant differences in perceived teacher support across different profiles based on gender and grade. Gender and grade did not predict the different potential profile probabilities of perceived teacher support. Although this result is inconsistent with the expectations of the gender socialization hypothesis and the vulnerability hypothesis [25, 26], it aligns with research suggesting the consistency of perceived teacher support across genders [27, 65]. One possible explanation for this result is the increasing emphasis on gender equality and the evolution of educational practices, which encourage teachers to treat all students equitably and reduce gender bias. This may lead to students perceiving teacher support as more gender-neutral [27, 65]. At the same time, students become more gender-conscious and focus more on internal needs than external stereotypes, further weakening the impact of gender on perceived teacher support[66].

The influence of grade level is inconsistent with the results of previous studies, showing no significant difference in perceived teacher support profiles. The reason may lie in the current educational practice and cultural background. First, the modern education system emphasizes continuity, so that different grades maintain stable educational environments, goals, content, and methods, which may allow students to maintain relatively consistent learning experiences and support from mathematics teachers [67]. Second, education reforms promote the professional development of teachers and ensure that mathematics teachers at all grades provide stable support [68]. In addition, students adapt to school life as they grow older, forming stable learning attitudes and emotions, which further stabilizes their perceived teacher support [22].

6.3 RQ3: The three latent profiles of perceived teacher support and mathematics engagement, mathematics attitude, and mathematics anxiety

Current research shows that students with different perceived teacher support profiles have significant differences in their mathematics engagement. Specifically, mathematics engagement increases progressively across students with low, medium, and high perceived teacher support. Further analysis revealed students with high perceived teacher support showed significantly higher overall mathematics engagement, encompassing elevated levels of behavioral, cognitive, and emotional engagement in mathematics learning. This finding aligns with self-determination theory which suggests that supportive teacher relationships enhance student engagement in learning. When teachers provide clear guidance, timely feedback, and emotional support, students’ interest in mathematics can be enhanced, and their needs for learning autonomy and ability can be satisfied, thus comprehensively promoting engagement in mathematics learning [15, 17]. This finding is also consistent with previous variable-centered studies, that is, students’ high perceived teacher support is positively correlated with their mathematics engagement, which highlights the core role of teacher support in students’ mathematics engagement [2, 17, 33]. Our research further confirms the person-centered perspective that there is a positive association between different perceived teacher support profiles and mathematics engagement. This finding highlights the importance of perceived teacher support for student mathematics engagement and provides a new perspective for understanding the increase in student mathematics engagement.

The results of this study reveal significant differences in mathematics anxiety among students with different perceived teacher support profiles, in line with previous variable-centered research. Specifically, students with high perceived teacher support exhibit lower levels of mathematics anxiety [3, 38]. Perceived teacher support can significantly enhance students’ confidence in mathematics, thus effectively reducing mathematics anxiety [3, 33, 38]. Further analysis showed that different teacher support profiles had significant differences in influencing students’ classroom anxiety. Students with high perceived teacher support had the lowest levels of anxiety in the classroom [3, 21, 69]. However, this study did not find significant differences between problem-solving anxiety and test anxiety. This may be due to the high difficulty and complexity of high school mathematics [1]. Even if students feel supported by their teachers, they still face great pressure when solving problems and taking exams. Therefore, this study suggests that while improving teacher support to alleviate students’ classroom anxiety, we should also pay attention to two more complex and multi-dimensional areas, namely problem-solving anxiety and test anxiety. Therefore, this study suggests that while improving teacher support to alleviate students’ classroom anxiety, we should also pay attention to two more complex and multi-dimensional areas, namely problem-solving anxiety and test anxiety.

The study also found significant differences in mathematics attitudes among students with different perceived teacher support profiles. Further analysis that there were significant differences in learning motivation and learning strategies. Consistent with previous variable-centered studies, students with higher levels of perceived teacher support demonstrated more positive attitudes toward mathematics. Additionally, these students demonstrated significantly enhanced learning motivation, proactively exploring learning opportunities and showing a willingness to overcome challenges [50]. They show higher self-regulation ability in learning strategies and can flexibly adjust methods to optimize learning effects [20, 70]. However, no significant differences were found in learning beliefs, which may be because students’ inherent beliefs are stable and not susceptible to fluctuations by external factors such as teacher support [71].

6.4 RQ4: Mechanisms between the three latent profiles of perceived teacher support and mathematics engagement, mathematics anxiety, and mathematics attitude

Regarding mathematics attitudes, the research revealed a close relationship between teacher support and students’ attitudes toward mathematics. Specifically, moderate to high levels of perceived teacher support had a significant positive impact on students’ mathematics attitudes. Students who feel supported by their teachers show greater confidence, strong interest, sustained enthusiasm, and more positive attitudes in mathematics learning [46, 72]. Further analysis revealed the key mediating role of mathematics attitude between teacher support and mathematics engagement. This means that by providing support, teachers can improve students’ mathematical attitudes, which in turn promotes their more active engagement in mathematical learning activities [22, 49, 50]. This finding highlights the role of mathematical attitudes as a bridge between teacher support and mathematics engagement.

When examining the mediating effect of mathematics anxiety, the findings showed an interesting pattern: moderate perceived teacher support had no significant effect on mathematics anxiety, while high perceived teacher support significantly reduced it. This suggests that mathematics anxiety can only be effectively alleviated when students feel adequately supported by their teachers [3, 73]. It is worth noting that mathematics anxiety has been theoretically and empirically explained to have an impact on mathematics engagement [18, 34]. However, in this study, it did not show this predicted effect. This is mainly attributed to the following two reasons: on the one hand, mathematics attitude plays a dominant role in the mediation model, and its positive influence may cover up the potential interference of mathematics anxiety; on the other hand, there may be potential interaction effects between teacher support and mathematics anxiety. In this study, even though higher anxiety levels were prevalent among the student population, this phenomenon did not significantly affect mathematics engagement, possibly because teacher support became an effective moderator. Specifically, teacher support may effectively alleviate students’ anxiety about mathematics through frequent support and assistance to learners, thereby weakening the negative impact of anxiety on mathematics engagement.

This study provides an innovative person-centered perspective on mathematics education, focusing on students’ perceived teacher support. However, it also has several limitations, which provide clear directions for improvement in future research.

First, this study is a cross-sectional study. Although it can explore the relationship between perceived teacher support and relevant factors in the process of mathematics learning to a certain extent, these variables may have mutual effects in the time dimension, so longitudinal research is needed to further clarify their relationship. Second, although the person-centered approach reveals different characteristics of students’ perceived teacher support because this approach focuses on identifying groups of students with similar perceived characteristics (i.e., “profiles”), it may have overlooked the possible qualitative differences among students within the same profile to a certain extent. To have a more comprehensive understanding of the impact of perceived teacher support on students’ learning, future research can combine person-centered and variable-centered approaches to reveal the complexity of students’ perceived teacher support. Third, this study relies on students’ self-report scales. There may be discrepancies between perceived and actual teacher support, influenced by direct factors such as “episodic memory”, thus affecting the reliability of the data. To improve the accuracy and comprehensiveness of the data, future research should optimize the questionnaire design, such as introducing time points and situational guidance to assess the impact of episodic memory. Additionally, researchers could consider integrating various data collection methods, such as interviews and observations to capture students’ real behaviors and attitudes in math learning from multiple perspectives and dimensions.

Previous variable-centered studies on perceived teacher support have ignored the heterogeneity among students, while this study adopted a person-centered research approach to provide a new method and perspective for students’ perceived mathematics teacher support. LPA revealed three different characteristics of perceived teacher support: low, medium, and high perceived mathematics teacher support. These profiles did not differ significantly in gender and grade but had significant differences in mathematics engagement, anxiety, attitudes and learning behaviors, motivation, and strategies. In addition, mathematics attitudes mediate the relationship between perceived teacher support and mathematics engagement. These enrich previous findings on variable-centered perceived teacher support.

The results of this study have some implications for mathematics education. First, there are individual differences in students’ perceived teacher support [6–8]. Therefore, mathematics teachers should focus on providing fair support to build a positive teacher-student relationship and ensure that students have a good mathematics learning experience [67]. Second, the higher the perceived teacher support, the more positive mathematics engagement and attitude, and the lower their anxiety [3, 4]. This suggests that teachers should provide a high level of support, such as timely help, positive encouragement, and a positive learning environment to promote mathematics learning [2, 17]. In addition, mathematics attitudes served as a key mediating variable. This finding suggests that by increasing the level of support, teachers can actively shape students’ positive attitudes towards mathematics, stimulate their interest and enthusiasm, and thus enhance their mathematics engagement [22, 34].

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Shaanxi Normal University, the institution where the first author works. All participants were acquainted with the research and provided written informed consent.

Consent for publication

Not Applicable.

Conflicts of interest

There is no conflict of interest; we conducted this study only as part of our research program.

Funding

This work supported by the National Social Science Fund of China [BCA240056].

Author Contribution

HL.M and Y.Z contributed to the study's conception and design. YF. H, J. Z, B. J, and Y. Z were responsible for material preparation and data collection. B. J and Y. Z carried out data analysis. Y.Z wrote the initial draft of the manuscript. ZL. P and HL. M revised the manuscript. All authors have reviewed and approved the final manuscript.

Acknowledgements

Not applicable

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Lei W, Zhang H, Deng W, Wang H, Shao F, Hu W. Academic self-efficacy and test anxiety in high school students: A conditional process model of academic buoyancy and peer support. School Psychol Int. 2021;42(6):616–37.
Tao Y, Meng Y, Gao Z, Yang X. Perceived teacher support, student engagement, and academic achievement: A meta-analysis. Educational Psychol. 2022;42(4):401–20.
Wang Z, Borriello GA, Oh W, Lukowski S, Malanchini M. Co-development of math anxiety, math self-concept, and math value in adolescence: The roles of parents and math teachers. Contemp Educ Psychol. 2021;67:102016.
Ober TM, Brodersen AS, Rebouças-Ju D, Hong MR, Carter MF, Liu C, et al. Math attitudes, engagement, and performance of high school students on high and low-stakes tests of statistics knowledge. J STEM Educ Res. 2022;5(3):402–38.
Liu R-D, Zhen R, Liu YDY, Wang J, Jiang R, Xu L. Teacher support and math engagement: roles of academic self-efficacy and positive emotions. Educational Psychol. 2018;38(1):3–16.
Hickendorff M, Edelsbrunner PA, McMullen J, Schneider M, Trezise K. Informative tools for characterizing individual differences in learning: Latent class, latent profile, and latent transition analysis. Learn Individual Differences. 2018;66:4–15.
Marsh HW, Lüdtke O, Trautwein U, Morin AJ. Classical latent profile analysis of academic self-concept dimensions: Synergy of person-and variable-centered approaches to theoretical models of self-concept. Struct Equation Modeling: Multidisciplinary J. 2009;16(2):191–225.
Zhou J, Hawrot A. Latent profile analysis of students’ perception of German classroom climate: outcomes and covariates. Learn Environ Res. 2023;1–22.
Lerner RM. Concepts and theories of human development. Routledge; 2018.
Laursen B, Hoff E. Person-centered and variable-centered approaches to longitudinal data. Merrill-Palmer Q (1982-). 2006;377–89.
Deci EL, Ryan RM. The what and why of goal pursuits: Human needs and the self-determination of behavior. Psychol Inq. 2000;11(4):227–68.
Liu Xxian, Gong SY, Zhang H, po, Yu Qlei, Zhou Z. jin. Perceived teacher support and creative self-efficacy: The mediating roles of autonomous motivation and achievement emotions in Chinese junior high school students. Thinking Skills and Creativity. 2021;39:100752.
Ouyang D, Song F. A research on the relation among teachers’ expectation, self-conception of students’ academic achievement, students’ perception of teacher’s behavioral supporting and the study achievement. Unpublished master thesis) Guangxi Normal University, Guilin, China. 2005.
Federici RA, Skaalvik EM. Students’ perceptions of emotional and instrumental teacher support: Relations with motivational and emotional responses. Int Educ Stud. 2014;7(1):21–36.
Ryan RM, Deci EL. Self-determination theory: Basic psychological needs in motivation, development, and wellness. Guilford; 2017.
Shin H, Chang Y. Relational support from teachers and peers matters: Links with different profiles of relational support and academic engagement. J Sch Psychol. 2022;92:209–26.
Gan Y, Peng J. Effects of teacher support on math engagement among Chinese college students: A mediated moderation model of math self-efficacy and intrinsic value. Child Youth Serv Rev. 2024;156:107369.
Li H, Zhang M, Hou S, Huang B, Xu C, Li Z, et al. Examining the dynamic links among perceived teacher support, mathematics learning engagement, and dimensions of mathematics anxiety in elementary school students: A Four-wave longitudinal study. Contemp Educ Psychol. 2023;75:102211.
Wilkie KJ, Sullivan P. Exploring intrinsic and extrinsic motivational aspects of middle school students’ aspirations for their mathematics learning. Educational Stud Math. 2018;97:235–54.
Yang Y, Li G, Su Z, Yuan Y. Teacher’s emotional support and math performance: The chain mediating effect of academic self-efficacy and math behavioral engagement. Front Psychol. 2021;12:651608.
Chen J, Lin CH, Chen G. Adolescents’ self-regulated and affective learning, teacher support and digital reading literacy: A multilevel latent profile approach. Comput Educ. 2023;205:104883.
Luo R, Zhang A, Wang Y, Li H, Xu Y, Guo K et al. Math attitudes and math anxiety predict students’ perception of teacher support in primary school, but not vice versa. Br J Educ Psychol. 2023;e12628.
Lei H, Cui Y, Chiu MM. The relationship between teacher support and students’ academic emotions: A meta-analysis. Front Psychol. 2018;8:2288.
Wentzel KR, Muenks K, McNeish D, Russell S. Peer and teacher supports in relation to motivation and effort: A multi-level study. Contemp Educ Psychol. 2017;49:32–45.
Ewing AR, Taylor AR. The role of child gender and ethnicity in teacher–child relationship quality and children’s behavioral adjustment in preschool. Early Child Res Q. 2009;24(1):92–105.
Hamre BK, Pianta RC. Early teacher–child relationships and the trajectory of children’s school outcomes through eighth grade. Child Dev. 2001;72(2):625–38.
Bayram Özdemir S, Özdemir M. How do adolescents’ perceptions of relationships with teachers change during upper-secondary school years? J Youth Adolesc. 2020;49(4):921–35.
Hughes JN, Luo W, Kwok OM, Loyd LK. Teacher-student support, effortful engagement, and achievement: A 3-year longitudinal study. J Educ Psychol. 2008;100(1):1–14.
Rice L, Barth JM, Guadagno RE, Smith GP, McCallum DM. The role of social support in students’ perceived abilities and attitudes toward math and science. J Youth Adolesc. 2013;42:1028–40.
Havik T, Westergård E. Do teachers matter? Students’ perceptions of classroom interactions and student engagement. Scandinavian J educational Res. 2020;64(4):488–507.
Rautanen P, Soini T, Pietarinen J, Pyhältö K. Dynamics between perceived social support and study engagement among primary school students: A three-year longitudinal survey. Soc Psychol Educ. 2022;25(6):1481–505.
Fredricks JA, Blumenfeld PC, Paris AH. School engagement: Potential of the concept, state of the evidence. Rev Educ Res. 2004;74(1):59–109.
Liu RD, Zhen R, Ding Y, Liu Y, Wang J, Jiang R, et al. Teacher support and math engagement: roles of academic self-efficacy and positive emotions. Educational Psychol. 2018;38(1):3–16.
Quintero M, Hasty L, Li T, Song S, Wang Z. A multidimensional examination of math anxiety and engagement on math achievement. Br J Educ Psychol. 2022;92(3):955–73.
Beilock SL, Maloney EA. Math anxiety: A factor in math achievement not to be ignored. Policy Insights Behav Brain Sci. 2015;2(1):4–12.
Passolunghi MC, De Vita C, Pellizzoni S. Math anxiety and math achievement: The effects of emotional and math strategy training. Dev Sci. 2020;23(6):e12964.
Maloney EA, Ramirez G, Gunderson EA, Levine SC, Beilock SL. Intergenerational effects of parents’ math anxiety on children’s math achievement and anxiety. Psychol Sci. 2015;26(9):1480–8.
Li H, Zhang A, Zhang M, Huang B, Zhao X, Gao J, et al. Concurrent and longitudinal associations between parental educational involvement, teacher support, and math anxiety: The role of math learning involvement in elementary school children. Contemp Educ Psychol. 2021;66:101984.
Luo Z, Luo W. Discrete achievement emotions as mediators between achievement goals and academic engagement of Singapore students. Educational Psychol. 2022;42(6):749–66.
Li H, Zhang M, Hou S, Huang B, Xu C, Li Z, et al. Examining the dynamic links among perceived teacher support, mathematics learning engagement, and dimensions of mathematics anxiety in elementary school students: A Four-wave longitudinal study. Contemp Educ Psychol. 2023;75:102211.
Aldrup K, Klusmann U, Lüdtke O. Reciprocal associations between students’ mathematics anxiety and achievement: Can teacher sensitivity make a difference? J Educ Psychol. 2020;112(4):735.
Choe KW, Jenifer JB, Rozek CS, Berman MG, Beilock SL. Calculated avoidance: Math anxiety predicts math avoidance in effort-based decision-making. Sci Adv. 2019;5(11):eaay1062.
Neale DC. The role of attitudes in learning mathematics. The Arithmetic Teacher [Internet]. 1969; https://api.semanticscholar.org/CorpusID:172815910
He X, Li Y. Compilation of Mathematics Attitude Scale for Junior High School Student and Its Test for Reliability and Validity. J Math Educ. 2013;22(2):37–42.
Lin SH, Huang YC. Development and application of a Chinese version of the short attitudes toward mathematics inventory. Int J Sci Math Educ. 2016;14:193–216.
Lee G, Simpkins SD. Ability self-concepts and parental support may protect adolescents when they experience low support from their math teachers. J Adolesc. 2021;88:48–57.
Yang X, Li J, Guo X, Li X. Group interactive network and behavioral patterns in online English-to-Chinese cooperative translation activity. Internet High Educ. 2015;25:28–36.
Tapia M, Marsh IIGE. Attitudes toward mathematics inventory redux. Acad Exch Q. 2005;9(3):272–6.
Quintero M, Hasty L, Li T, Song S, Wang Z. A multidimensional examination of math anxiety and engagement on math achievement. Br J Educ Psychol. 2022;92(3):955–73.
Vidić T, Đuranović M. Students’ attitudes towards mathematics and their perceptions of teacher support, enthusiasm, classroom management and their own behavior. J Educational Sci Psychol. 2020;10(2).
Howard MC, Hoffman ME. Variable-centered, person-centered, and person-specific approaches: Where theory meets the method. Organizational Res Methods. 2018;21(4):846–76.
Wormington SV, Linnenbrink-Garcia L. A new look at multiple goal pursuit: The promise of a person-centered approach. Educational Psychol Rev. 2017;29:407–45.
Lanza ST, Cooper BR. Latent class analysis for developmental research. Child Dev Perspect. 2016;10(1):59–64.
Xu J. Student-perceived teacher and parent homework involvement: Exploring latent profiles and links to homework behavior and achievement. Learn Individual Differences. 2024;109:102403.
DeSimone JA, Harms PD, DeSimone AJ. Best practice recommendations for data screening. J Organizational Behav. 2015;36(2):171–81.
Huang JL, Curran PG, Keeney J, Poposki EM, DeShon RP. Detecting and deterring insufficient effort responding to surveys. J Bus Psychol. 2012;27:99–114.
Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika. 1951;16(3):297–334.
Meyers LS, Gamst G, Guarino AJ. Applied multivariate research: Design and interpretation. Sage; 2016.
Babad E. Measuring and changing teachers’ differential behavior as perceived by students and teachers. J Educ Psychol. 1990;82(4):683.
Wang MT, Fredricks JA, Ye F, Hofkens TL, Linn JS. The math and science engagement scales: Scale development, validation, and psychometric properties. Learn Instruction. 2016;43:16–26.
Zhang P. Research on the Relationship between Mathematics Learning Anxiety and Mathematics Achievement of Senior High School Students -The mediating role of mathematics learningself-efficacy and mathematics learning strategies [phd]. Central China Normal University; 2021.
Richardson FC, Suinn RM. The mathematics anxiety rating scale: psychometric data. J Couns Psychol. 1972;19(6):551.
Nylund KL. Latent transition analysis: Modeling extensions and an application to peer victimization [phd]. University of California, Los Angeles Doctoral dissertation; 2007.
Zhou H, Long L. Statistical remedies for common method biases. Adv Psychol Sci. 2004;12(06):942.
Hughes JN, Cao Q. Trajectories of teacher-student warmth and conflict at the transition to middle school: Effects on academic engagement and achievement. J Sch Psychol. 2018;67:148–62.
Burke PJ, Stets JE. Identity theory: Revised and expanded. Oxford University Press; 2022.
Jeong-Mi Moon EMC, Sebastian J. Streamlining your school: understanding the relationship between instructional program coherence and school performance. School Eff School Improv. 2022;33(2):260–79.
Thurm D, Barzel B. Effects of a professional development program for teaching mathematics with technology on teachers’ beliefs, self-efficacy and practices. ZDM Mathematics Education. 2020;52(7):1411–22.
Maloney T, Matthews JS. Teacher care and students’ sense of connectedness in the urban mathematics classroom. J Res Math Educ. 2020;51(4):399–432.
León J, Núñez JL, Liew J. Self-determination and STEM education: Effects of autonomy, motivation, and self-regulated learning on high school math achievement. Learn Individual Differences. 2015;43:156–63.
Demedts F, Reynvoet B, Sasanguie D, Depaepe F. Unraveling the role of math anxiety in students’ math performance. Front Psychol. 2022;13:979113.
Ober TM, Coggins MR, Rebouças-Ju D, Suzuki H, Cheng Y. Effect of teacher support on students’ math attitudes: Measurement invariance and moderation of students’ background characteristics. Contemp Educ Psychol. 2021;66:101988.
Li H, Zhang A, Zhang M, Huang B, Zhao X, Gao J, et al. Concurrent and longitudinal associations between parental educational involvement, teacher support, and math anxiety: The role of math learning involvement in elementary school children. Contemp Educ Psychol. 2021;66:101984.

No competing interests reported.

Download PDF

Editor invited by journal
24 Oct, 2024
Editor assigned by journal
24 Oct, 2024
Submission checks completed at journal
22 Oct, 2024
First submitted to journal
19 Oct, 2024

You are reading this latest preprint version

Perceived teacher support profiles and mathematics engagement, mathematics anxiety, mathematics attitude: A latent profile analysis approach

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1. Introduction

2. Literature review

2.1 Perceived teacher support and mathematics learning

2.2 Gender, grade, and perceived teacher support

2.3 Perceived teacher support and mathematics engagement, mathematics anxiety, and mathematics attitude

2.3.1 Perceived teacher support and mathematics engagement

2.3.2 Mathematics anxiety as a mediator

2.3.3 Mathematics attitude as a mediator

2.4 The need for a person-centered approach to profile perceived teacher support

3. The Present Study

4. Method

4.1 Participants

4.2 Data collection

4.3 Measures

4.3.1 Perceived Teacher Support

4.3.2 Mathematics engagement

4.3.3 Mathematics anxiety

4.3.4 Mathematics attitude

4.5 Data analysis

5. Results

5.1 Common Method Bias

5.2 RQ1: Latent profile analysis of perceived teacher support

5.3 RQ2: Gender, grade, and latent profiles of perceived teacher support

6. Discussion

6.1 RQ1: Latent profile analysis of perceived teacher support

6.2 RQ2: Gender, grade, and latent profiles of perceived teacher support

7. Limitations and future directions

8. Conclusion and implications

Declarations

Conflicts of interest

Funding

Author Contribution

Acknowledgements

Data Availability

References

Additional Declarations

Status:

Version 1