The relationship between mathematics interest and mathematics achievement: mediating roles of self-efficacy and mathematics anxiety
The present study
In this study, we propose a serial mediation model, based on existing literatures on how mathematics interest may associate with self-efficacy and affect studentrsquo;s mathematics anxiety, and relation to mathematics achievement. This study contributed by exploring the mediating roles of self-efficacy and mathematics anxiety in the effects of mathematics interest on mathematics achievement (Fig. 1). The following research hypotheses will be tested.
H1. Mathematics interest, self-efficacy, and mathematics achievement positively correlate with each other, and they all negatively correlate with mathematics anxiety.
H2. Self-efficacy mediates the relationship between mathematics interest and mathematics achievement.
H3. Self-efficacy and then mathematics anxiety serial mediate the relationship between mathematics interest and mathematics achievement.
Fig. 1. Hypothesized serial model depicting the relations of the constructs.
3. Method
3.1. Participants
In the present study, the data we analyzed was from a large-scale investigation in China entitled lsquo;Regional Education Assessment Project (REAP)rsquo;. A total of 158161 Chinese eighth graders from 1097 schools were selected from 145 districts and counties of 4 province in the north, south, east and middle of China with two-stage unequal probability sampling. At the first stage, the schools were sorted and stratified by the nature of the school and the school system of districts and counties of each provinces, and then were selected by probability proportionate to size sampling (PPS) method. At the second stage, the students in the schools were selected based on the systematic sampling method.
All of the eighth-grade students extracted from the schools were invited to participate in this study. After removing the missing data, the final sample included 156661 students. Among them, 53% were boys, and 47% were girls. 3.2. Measures
3.2.1. Mathematics interest
Mathematics interest was measured with 4 items drawn from the PISA 2012 student questionnaires. The scale measured studentsrsquo; views on mathematics, including lsquo;I enjoy reading about mathematicsrsquo;; lsquo;I look forward to my mathematics lessonsrsquo;; lsquo;I do mathematics because I enjoy itrsquo;; lsquo;I am interested in the things I learn in mathematicsrsquo;. Ratings ranged from 1 (strongly disagree) to 5 (strongly agree). Internal consistency reliability (Cronbachrsquo;s alpha) of the mathematics interest scale was reasonably high at 0.945.
3.2.2. Self-efficacy
Mathematics self-efficacy contained 8 items that were modified from the PISA 2012 student questionnaires. This scale measured the degree of studentrsquo;s confidence about doing a specific mathematics task. The items were lsquo;determining the relationship between the cost (such as parking fees, telephone charges, etc.) and the amount of timersquo;; lsquo;estimating the nearest integer to an irrational numberrsquo;; lsquo;comparing and calculating which of the two preferential programs is more cost-effective according to the number and unit price of purchased goodsrsquo;; lsquo;using props (such as poker, turntable, etc.) to design a fair gamersquo;; lsquo;drawing conclusions based on known chartsrsquo;; lsquo;drawing with ruler and compass to find points with the same distance from three non-collinear pointsrsquo;; lsquo;folding the rectangular paper to find out the relationship between the formed anglesrsquo;; lsquo;exploring the laws of angles of pentagram, hexagram, etc.rsquo;. Students were not required to solve these problems, just need to rate the degree of their confidence in being able to do so on a 5-point Likert scale ranging from 1 (very difficult) to 5 (very easy). The reliability (Cronbachrsquo;s alpha) of mathematics self-efficacy scale in this study was 0.940.
3.2.3. Mathematics anxiety
Mathematics anxiety was also measured with 5 items drawn from the PISA 2012 student questionnaires. The students were asked to report whether they agree or strongly agree of their worry, nervous, and even feelings of helpless regarding their mathematics learning. The included items in this scale were lsquo;I often worry that it will be difficult for me in mathematics classesrsquo;; lsquo;I get very tense when I have to do mathematics homeworkrsquo;; lsquo;I get very nervous doing mathematics problemsrsquo;; lsquo;I feel helpless when doing a mathematics problemrsquo;; lsquo;I worry that I will get poor lt; grades gt; in mathematicsrsquo;. Ratings ranged from 1 (strongly disagree) to 5 (strongly agree). Internal consistency reliability (Cronbachrsquo;s alpha) of the mathematics anxiety scale was also very high at 0.915.
3.2.4. Mathematics achievement
The items in the mathematics performance test was developed based on Mathematics Curriculum Standard for Compulsory Education (2011 version) (Ministry of Education, Peoplersquo;s Republic of China, 2012). The test examined eighth grade studentsrsquo; mathematics performance from both the content dimension (number and algebra, graphics and geometry, statistics and probability) and cognitive dimension (knowing, understanding and application). To ensure the quality of the test instrument, the item developing team experienced several procedures: (1) interviewing 6 representative students to decide whether they could understand each item of the test instrument; (2) revising those problematic items based on 30 studentsrsquo; responses to each item; (3) further revising the problematic items after reviewing 300 studentsrsquo; responses; (4) independent review of experts from external institutions to help us revise the items.
There were 2 parallel test booklets (A and B). Students were randomly assigned to do one of the booklets. Our analyses were based o
剩余内容已隐藏,支付完成后下载完整资料
数学兴趣与数学成就之间的关系:自我效能与数学焦虑的中介作用
目前的研究
在这项研究中,我们基于现有的文献提出了一个串行调解模型,其中涉及数学兴趣如何与自我效能感相关联并影响学生的数学焦虑以及与数学成绩的关系。该研究通过探索自我效能感和数学焦虑在数学兴趣对数学成就的影响中的中介作用而做出了贡献(图1 )。以下研究假设将得到检验。
1. 数学兴趣,自我效能感和数学成就相互之间呈正相关,而它们与数学焦虑呈负相关。
2. 自我效能感介导了数学兴趣与数学成就之间的关系。
3. 自我效能感以及随后的数学焦虑系列介导了数学兴趣与数学成就之间的关系。
图1.假设序列模型描述了构建体之间的关系
方法
参加者
在本研究中,我们分析的数据来自中国一项名为“区域教育评估项目(REAP)”的大规模调查。从华北,华南,华东和华中4个省的145个区县中抽取了来自1097所学校的158161名中国八年级学生,进行了两阶段不等概率抽样。在第一阶段,根据学校的性质和各省区县的学校系统对学校进行分类和分层,然后通过与大小抽样法(PPS)成比例的概率进行选择。在第二阶段,根据系统的抽样方法选择学校的学生。
从学校提取的所有八年级学生都应邀参加了这项研究。删除缺失的数据后,最终样本包括156661名学生。其中,男孩占53%,女孩占47%。
措施
数学兴趣
对数学的兴趣通过从PISA 2012学生调查表中抽取的4个项目进行衡量。该量表衡量了学生对数学的看法,包括“我喜欢阅读数学”;“我期待着我的数学课”;“我之所以做数学是因为我喜欢它”;“我对我在数学中学到的东西很感兴趣”。评分范围从1(强烈不同意)到5(强烈同意)。数学兴趣量表的内部一致性可靠性(Cronbachs alpha)相当高,为0.945。
自我效能
数学自我效能感包含了从PISA 2012学生调查表中修改的8个项目。该量表衡量了学生对完成特定数学任务的信心程度。这些项目是“确定成本(例如停车费,电话费等)和时间之间的关系”;“估计最接近无理数的整数”;“根据所购商品的数量和单价比较和计算两个优惠方案中哪一个更具成本效益”;“使用道具(例如扑克,转盘等)设计公平竞赛”;“根据已知图表得出结论”;“用尺子和罗盘绘图以找到与三个非共线点具有相同距离的点”;“折叠矩形纸以找出形成的角度之间的关系”;“探索五角星,六边形等的角度定律”。不需要学生解决这些问题,只需要在5分李克特量表(从1(非常困难)到5(非常容易))上评估他们对做到这一点的信心程度。在这项研究中,数学自我效能感量表的信度(Cronbachs alpha)为0.940。
数学焦虑
还从PISA 2012学生调查表中抽取了5个项目来测量数学焦虑。要求学生报告他们是否同意或强烈同意他们对数学学习的担心,紧张甚至无助的感觉。这个量表中包含的内容是“我经常担心我在数学课上会很困难”;“当我必须做数学作业时,我会变得非常紧张”;“我在解决数学问题时非常紧张”;“在做数学问题时,我感到束手无策”;“我担心我的数学成绩会很差”。评分范围从1(强烈不同意)到5(强烈同意)。数学焦虑量表的内部一致性可靠性(Cronbachs alpha)也很高,为0.915。
数学成就
数学性能测试中的项目是根据《义务教育数学课程标准》(2011年版)制定的(中华人民共和国教育部,2012年))。该测试从内容维度(数字和代数,图形和几何,统计和概率)和认知维度(知识,理解和应用)两个方面检查了八年级学生的数学表现。为了确保测试仪器的质量,项目开发团队经历了几个程序:(1)采访6名有代表性的学生,以确定他们是否可以理解测试仪器的每个项目;(2)根据30位学生对每个问题的回答来修改这些有问题的项目;(3)在复习300名学生的回答后,进一步修改有问题的项目;(4)独立审查来自外部机构的专家,以帮助我们修改项目。
有两本平行的测试手册(A和B)。学生被随机分配做一本小册子。我们的分析是基于项目反应理论,通过使用Rasch模型获得学生的数学能力价值。然后,我们将其转换为平均得分为500且标准偏差为100的量表得分。A和B测试手册的可靠性分别为.926和.885。
程序
要求学生完成数学成绩测试,然后完成自我报告纸质问卷,包括数学兴趣,自我效能感和数学焦虑量表。所有参与者都被告知,我们收集的数据将被保密。然后,指导学生如何填写天平。要求学生在大约90分钟内完成数学成绩测试和问卷调查。
仪器验证和数据分析
由于评估变量中存在缺失值,因此我们计算了缺失数据的程度。所有缺少每个变量数据的案例比例均小于4%。在所有变量上不遗漏的案例比例为99%。另外,我们使用小的MCAR(小&鲁宾,2002 )来测试丢失数据的机制,结果表明,缺失值没有被完全随机缺失(chi; 2 = 7656.543,DF = 2639,p lt;0.001) 。然后,在数据分析之前将所有丢失的数据删除。为了验证问卷的结构,我们进行了验证性因素分析(CFA),以建立量表中各项的阶乘结构(Schreiber,Nora,Stage,Barlow和King,2006年)。数据分析的下一步是计算数学兴趣,自我效能感,数学焦虑和数学成就之间的描述性和相关性统计量。然后,由于数据的嵌套性质,计算类内相关系数。为了测试理论模型(见图1 ),使用Mplus 7程序执行SEM。此外,还进行了自举分析,对数据进行了重新采样并替换了10,000次,以检查调解效果的重要性。偏差校正区间和置信区间均为95%(Taylor,MacKinnon,&Tein,2008)。作为一种统计方法,SEM可用于对变量之间的关系的假设模型进行统计检验,以确定其与数据一致的程度或与数据的拟合程度。如果拟合优度足够,则表明变量之间的假定关系是合理的(Byrne,2001年)。为了评估模型的拟合度,本研究采用了公认的指标(例如CFI,TLI,RMSEA,SRMR)。对于这些指数,通常认为CFI,TLI值大于.90可以接受,而RMSEA和SRMR值小于0.08则表示是合理的拟合(Bollen,1989 ; Browne&Cudeck,1993 ; Byrne,2001 ; Hu&Bentler, 1999 )。
结果
初步分析
为了验证问卷的结构,在3个相互关联的量表(数学兴趣,自我效能感,数学焦虑)上进行了验证性因素分析(CFA)。三因素结构的结果提供了一个很好的适合的数据与chi; 2 = 53484.819,DF = 116; p lt;0.001;CFI = .896;TLI = .879;RMSEA = .054;SRMR = .034。3个量表的所有因子负荷都在0.60以上。
表1显示了所有研究变量之间的描述性统计量和相关性。结果表明,数学兴趣,自我效能感和数学成就相互之间呈显着正相关,而与数学焦虑呈显着负相关。
类内相关系数(ICC)
考虑到数据的嵌套性质,检查了学校在数学成绩,数学兴趣,自我效能感,数学焦虑方面的差异。方差的随机效应分析结果表明,类内相关系数(ICC)对数学成绩的贡献为33.7%,对数学兴趣的看法为7.4%,对自我效能感的看法为10.7%,对数学焦虑的看法为4.9%。所有学校之间的差异都是显着的。因此,考虑到样本的非独立性,在Mplus 7中使用“类型=复杂”来调整标准误差。
直接效果
我们首先分析了对数学成绩的数学兴趣的直接作用,直接模型的拟合度是可以接受的:chi; 2 = 55345.375,DF = 131; p lt;0.001;CFI = .901;TLI = .884; RMSEA = .052;SRMR = .035。结果表明,数学兴趣对数学成绩有显着影响(beta; = .151,p lt;.001),表明较高的数学兴趣与较高的数学成绩有关。
表格1描述性统计量和研究变量之间的相关性。
多变的 |
1个 |
2个 |
3 |
4 |
1.数学兴趣 |
1个 |
|||
2.自我效能感 |
.550 ** |
1个 |
||
3.数学焦虑 |
-.107 ** |
-.079 ** |
1个 |
|
4.数学成就 |
.310 ** |
.358 ** |
-.239 ** |
1个 |
吝啬的 |
3.33 |
3.74 |
2.90 |
569.10 |
标准偏差 |
1.11 |
.92 |
1.09 |
77.16 |
注意:** p lt;.01。
图2.数学兴趣,自我效能感,数学焦虑和数学成就的结构方程模型。
调解分析
类似地,结果表明,数学兴趣对自我效能感有显着的预测作用(beta; = .581,p lt;.001),自我效能对数学成就具有显着的影响(beta; = .260,p lt; .001)和数学焦虑(beta; = -.109,p lt;.001),以及数学焦虑对数学成绩的显着影响(beta; = -.212,p lt;.001)(见图2 )。此外,还计算了自举分析,对数据进行了重新采样并替换了10,000次,以检查调解效果的重要性(请参见表2)。)。结果表明两条间接路径是重要的。该模型解释了数学成就方差的19.7%。两种间接途径的介导作用占总作用的52.06%。具体而言,间接途径的中介作用是数学兴趣→自我效能感→数学成就,为47.94%,表明自我效能感在中介数学兴趣和数学成就方面起着重要作用。此外,间接途径,数学兴趣→自我效能感→数学焦虑→数学成就的中介效应为4.13%,这表明自我效能和随后的数学焦虑系列之间的中介作用是数学兴趣与数学成就之间的关系。
讨论
本研究通过验证一个理论模型来确认数学兴趣和数学成就之间的关系,从而为文献做出了贡献。这项研究的主要发现是:(1)数学兴趣,自我效能感和数学成就相互之间呈显着正相关,而与数学焦虑呈显着负相关;(2)自我效能在调解数学兴趣和数学成就方面发挥了重要作用;(3)自我效能感继而又是数学焦虑的中介,它直接联系了数学兴趣与数学成就之间的关系。一般而言,本研究结果为数学兴趣与数学成就之间的关联提供了经验证据。
我们的研究数据支持了先前研究的发现,即数学兴趣与学生的数学成就呈正相关(例如,Reeve等,2015 ;Schiefele等,1992 ; Simpkins等,2006 )。对数学更感兴趣的学生通常会付出更大的努力,提高唤醒水平,增加与数学相关的活动所花费的时间,并且表现出更深的认知加工能力和更好的自我调节能力(Fisher等,2012 )。因此,对数学有较高兴趣的学生比对数学没有兴趣的同学表现出更好的数学成绩。一些研究还表明,数学成绩反过来会影响学生的数学兴趣(鲍默特(Baumert),施纳贝尔(Schnabel)和莱尔克(Lehrke),1998 ;Ganley&Lubienski,2016 ;Kouml;ller等,2001 )。此外,那些兴趣浓厚的学生在将来的学习中更有可能选择一门高级数学课程(Kouml;ller等,2001 ),通过该课程他们的数学学习可以形成一个良性循环。然而,
表2对中介效果进行引导测试。
路径 |
标准化 beta; |
东南 |
降低 |
上 |
MInt→S-Eff→机械 |
0.151 *** |
0.002 |
0.147 <!--剩余内容已隐藏,支付完成后下载完整资料
资料编号:[405849],资料为PDF文档或Word文档,PDF文档可免费转换为Word |
以上是毕业论文外文翻译,课题毕业论文、任务书、文献综述、开题报告、程序设计、图纸设计等资料可联系客服协助查找。