critical review of a research paper

System 89 (2020) 102199

Contents lists available at ScienceDirect

System

journal homepage: www.elsevier.com/locate/system

Exploring the relationship between foreign language motivation and achievement among primary school students learning English in China

Xiaoyi Hu a, *, Sarah McGeown b, 1

a Faculty of English Language and Culture, Guangdong University of Foreign Studies, 2 North Baiyun Street, Guangzhou, 510420, China b Moray House School of Education and Sport, University of Edinburgh, Holyrood Road, Edinburgh, EH8 8AQ, UK

a r t i c l e i n f o

Article history: Received 10 May 2019 Received in revised form 3 January 2020 Accepted 3 January 2020 Available online 10 January 2020

Keywords: China Foreign language motivation Foreign language achievement Expectancy-value theory Gender differences

* Corresponding author. Faculty of English Langua E-mail addresses: 201910114@oamail.gdufs.edu.c

1 Sarah McGeown is a senior lecturer in developm

https://doi.org/10.1016/j.system.2020.102199 0346-251X/© 2020 Elsevier Ltd. All rights reserved.

a b s t r a c t

This study explored primary school students’ foreign language (FL) motivation (expec- tancy-value) and its relationship with their FL achievement. In total, 631 Chinese primary school pupils (324 boys, 307 girls) aged between 9 and 12 years completed a questionnaire to report on their FL motivation (expectancy-value), and their teachers provided measures of their FL achievement in school. Students differentiated between expectancy and value components of motivation, and overall higher reports of value were found. While girls reported higher levels of FL motivation (particularly value), declines in students’ FL motivation (expectancy and value) were found with increasing age. Results from regres- sion analyses illustrated that both the expectancy and value components of motivation were significant independent predictors of students’ FL achievement, accounting for approximately 40% of the variance. However, expectancy emerged as the stronger pre- dictor across all ages and for both boys and girls. In light of these findings, pedagogical strategies such as shifting the emphasis from competitive to cooperative FL learning ac- tivities or providing students with more choice and autonomy over their FL activities can be adopted to enhance primary school students’ FL motivation.

© 2020 Elsevier Ltd. All rights reserved.

1. Introduction

Foreign language (FL) motivation reflects an individual’s drive to engage in and persevere with FL learning, and a considerable body of research has illustrated a positive relationship between FL motivation and FL achievement. However, research to date has predominately been carried out with adult learners (e.g., Hsieh & Schallert, 2008; Khodadady & Khajavy, 2013; Mills, Pajares, & Herron, 2007); very little research has been conducted among children of primary school age. Each year approximately 65e70 million primary school students in China learn English as a FL, since it is one of three core subjects (Ministry of Education, 2018). In this context, children are engaging in compulsory FL leaning, which may be different from the situation with adult learners who often choose to learn a foreign language, or at least have more experience of learning/ education. Despite significant numbers of primary school children engaging in FL learning in China, we know very little about their FL motivation and how it relates to their FL achievement. Given the considerable lack of research investigating young

ge and Culture, Guangdong University of Foreign Studies, Guangzhou, 510420, Guangdong, China. n (X. Hu), S.McGeown@ed.ac.uk (S. McGeown). ental psychology at University of Edinburgh

X. Hu, S. McGeown / System 89 (2020) 1021992

learners, this study was driven by a need to contribute to academic understanding in this area, but also to raise awareness of the importance of FL motivation and the extent to which it is related to children’s FL achievement.

2. Foreign language motivation

Studies of foreign language (FL) motivation have drawn upon different theoretical frameworks in order to understand students’ motivation to learn a FL across different contexts. Much of the work in the FL learning context was initiated and inspired by Gardner’s (1985) social-educational L2 motivational framework. This theory posits that motivation to learn a L2 arises from integrative (desire to learn L2 in order to interact with members of L2 community) and/or instrumental (desire to learn L2 for pragmatic gains) reasons. Subsequent research has drawn upon this or other theoretical frameworks and identified significant positive relationships between FL motivation and achievement (e.g., Hsieh & Schallert, 2008; Liu, 2007; McEown, Noels, & Saumure, 2014).

However, to date the majority of FL motivation research has been conducted among adult/university learners across different cultural contexts, with examples of studies in Iran (Khodadady & Khajavy, 2013), Korea (Joe, Hiver, & Al-Hoorie, 2017), Canada (McEown et al., 2014), and the US (Hern�andez, 2008; Hsieh & Schallert, 2008; Mills et al., 2007), among many others. There are also examples of studies in China, but this research is mostly restricted to work with adults (Huang, 2008; Li & Pan, 2009; Liu, 2007; Liu & Huang, 2011; Liu & Zhang, 2013; Wang, 2008). To date there has been one study of primary school children’s FL motivation in China (Lin, Wong, & McBride-Chang, 2012), focusing specifically on reading skill among a group of children from affluent backgrounds. In studies of FL motivation, a range of theoretical frameworks have been drawn upon, with decisions regarding the choice of framework reflecting its suitability for the sample, context, research question(s), methodology and/or preferences of the research team (see Oga-Baldwin, Fryer, & Larson-Hall, 2019 for a recent review of FL motivation theories and research).

2.1. Studies of motivation among children

Learning as a child during primary school is very different to learning as an adult. Given the compulsory nature of primary school education, it is important to use theoretical frameworks of motivation which are suited to this developmental stage and the controlled and directed environment in which children are learning.

Several theories of motivation typically used among primary school children, although not originally developed in a FL context, have relevance to FL learning (Lee & Bong, 2019; Loh, 2019; McEown & Oga-Baldwin, 2019). For example, in its simplest form, achievement goal theory distinguishes between mastery (desire to increase competence) and performance (desire to demonstrate competence) goal orientations (Lee & Bong, 2019). Self-determination theory, on the other hand, differentiates between autonomous (i.e., intrinsic or internal drivers) and controlled (i.e., extrinsic or external drivers) di- mensions of motivation (Lee & Bong, 2019; McEown et al., 2014; McEown & Oga-Baldwin, 2019). Meanwhile, expectancy- value theory (Chen & Sheu, 2005; Loh, 2019; Mori & Gobel, 2006) distinguishes between students’ expectations of success (expectancy) and the extent to which they value the subject (value) as key dimensions of motivation. These three theories are the most widely used in studies of motivation with children, and have clear relevance to FL learning (Lee & Bong, 2019; Loh, 2019; McEown & Oga-Baldwin, 2019). Importantly, the theories highlight that it is not only the quantity (i.e., level) but also the quality (i.e., type) of motivation that should be considered.

2.2. Expectancy-value theory of motivation

In this study, Eccles et al.’ (1983) expectancy-value theory of motivation was used. This posits that students’ motivation is influenced by their expectancy (both current and future expectations of success) and value (the extent to which they value the task or activity as interesting/enjoyable, useful, and important). This theoretical framework of motivation was chosen as it is suited to contexts where students are engaging in compulsory learning (i.e., autonomous aspects of motivation inherent within self-determination theory are not studied). Furthermore, the distinction between expectancy (expectations of success) and value (interesting/enjoyable, useful, important) was seen an interesting framework to study FL learning motivation among primary school children in China. For example, the Chinese education system places a high value on academic success and the importance of learning English. In China primary school students participate in regular assessments, receiving feedback (i.e., grades) each time (Hu, 2007; Qi, 2016), which are likely to have motivational consequences, specifically for their expectations of success. With regard to value, teaching shifts from cultivating an early interest in FL learning to mastering complex FL skills (Ministry of Education, 2011). In addition, given the status of English in the curriculum (i.e., a compulsory subject from Grade 3 onwards; MoE, 2011), students have an awareness of the value and importance of English from the beginning of their FL instruction. Finally, expectancy-value theory has also been used widely among children of primary school age, and therefore there are measures available which are developmentally appropriate. For these reasons, expectancy-value theory was used in order to examine the relationship between students’ FL motivation and achievement, in addition to examining grade and gender differences.

Until now, expectancy-value theory has not been used to study primary school students’ FL motivation, but it has been used to study students’ motivation across a range of different academic domains (e.g., German, English, literacy, math, physics, biology, chemistry; Archambault, Eccles, & Vida, 2010; Fan, 2011; Gaspard, H€afner, Parrisius, Trautwein, & Nagengast, 2017;

X. Hu, S. McGeown / System 89 (2020) 102199 3

Wang, Chow, Degol, & Eccles, 2017). Past research exploring the relationship between expectancy-value components of motivation and achievement have reported that both expectancy beliefs and value are correlated with achievement outcomes (Guo, Marsh, Parker, Morin, & Yeung, 2015; Simpkins, Davis-Kean, & Eccles, 2006; Wigfield & Eccles, 2000). Interestingly, however, when both expectancy and value are examined simultaneously, expectancy is typically the stronger correlate and/or predictor of performance or achievement, while value better predicts decisions to participate in and/or time spent partici- pating in the activity (e.g., Durik, Vida, & Eccles, 2006; Loh, 2019; Plante, O’Keefe, & Th�eorêt, 2013; Wang & Liou, 2017).

2.3. Gender and grade differences

In studies of motivation with primary school students, grade and gender differences are frequently explored, in order to understand trends and identify whether there are specific groups who may benefit from support in this area. Studies exploring gender differences in school students’ FL motivation have found that girls generally report higher levels of FL motivation than boys (Carreira, 2011; D€ornyei & Csiz�er, 2002; Henry & Apelgren, 2008; Sung & Padilla,1998). With respect to grade differences, declines in FL motivation with years of schooling are common (Carreira, 2011; Ghenghesh, 2010; Henry & Apelgren, 2008; Williams, Burden, & Lanvers, 2002). Nevertheless, it is important to recognize that a few studies have re- ported no significant gender or age differences (Tachibana, Matsukawa, & Zhong, 1996; Wu, 2003).

In research using the expectancy-value theory of motivation, boys have been found to report higher expectancy beliefs in some areas (i.e., maths, science, sports) and girls in others (i.e., music, language, reading) (Fan, 2011; Fredricks & Eccles, 2002; Jacobs, Lanza, Osgood, Eccles, & Wigfield, 2002). However, it is also important to note that other studies have found no evidence of gender differences in students’ expectancy beliefs in particular areas (e.g., reading: Durik et al., 2006). With regard to value, researchers have similarly reported gender-stereotyped differences in students’ value across different academic domains (Durik et al., 2006; Gaspard et al., 2017; Wigfield et al.,1997; Watt, 2004); however, there have been exceptions (e.g., maths: Eccles, Wigfield, Harold, & Blumenfeld, 1993; Jacobs et al., 2002). Grade-related changes have also been reported among school age pupils, with declines in students’ expectancy and value across numerous academic domains (Archambault et al., 2010; Fredricks & Eccles, 2002; Jacobs et al., 2002; Watt, 2004). To summarise, studies of FL motivation or those using the expectancy-value theory of motivation show similarities in their findings. On average, girls report higher levels of motivation, and motivation declines with increasing grade.

3. Study rationale and description

The present study examined the relationship between primary school students’ FL motivation and FL achievement. China has the world’s largest population of primary school students (Liu & Gong, 2001), yet they are an under-researched group. In China, learning English is highly valued and it is a core subject that all primary school students must learn from the age of 8e9 (Ministry of Education, 2018). Despite this, research exploring FL motivation among school students in China has primarily been conducted in high schools (e.g., D€ornyei & Chan, 2013). Research conducted in primary school contexts is rare, with only one identified study focusing on the acquisition of a specific FL skill (i.e., reading: Lin et al., 2012).

It is crucial, from both a theoretical and an educational perspective, to understand primary school students’ FL motivation in China (including reported levels of FL expectancy and value) and also understand the relationship between their FL motivation and achievement. Furthermore, grade and gender differences in FL motivation are also of interest, as under- standing trends across different groups is important to understand areas of need and target support optimally.

To answer our research questions, a cross-sectional study was designed and primary school students’ FL motivation (measured using a self-report questionnaire) and FL achievement (measured using school assessments) were examined. Based on previous research it was expected that FL expectancy would correlate more closely with FL achievement than FL value. Furthermore, it was expected that girls would report higher FL motivation than boys, and that FL motivation would decline with increasing age.

4. Research questions

1. What levels of FL motivation (expectancy and value) do Chinese primary school students report? 2. Are there grade or gender differences in their FL motivation? 3. To what extent do expectancy and value components of FL motivation predict FL achievement? 4. Does the relationship between FL motivation and FL achievement differ depending on grade or gender?

5. Method

5.1. Participants

In total, 631 students in one primary school took part in this study (324 boys, 307 girls). The primary school is located at the border of a suburban and urban area in a city in Guangdong Province, and pupils were from diverse social class

X. Hu, S. McGeown / System 89 (2020) 1021994

backgrounds. The school was regarded as representative as it follows National Curriculum guidelines (Ministry of Education, 2011) for the introduction and teaching of English to students. Each primary school in China has six grades (Primary 1 to Primary 6), and in this school each grade had four classes. Therefore, in this study, students from 12 classes across 3 grades (all 4 classes from each grade) were invited to participate. In total, 215 Year 4 students (110 boys, 105 girls, mean age ¼ 9.61, SD ¼ 0.67), 209 Year 5 students (107 boys and 102 girls, mean age ¼ 10.62, SD ¼ 0.71) and 207 Year 6 students (107 boys, 100 girls, mean age ¼ 11.56, SD ¼ 0.71) participated. All children had Mandarin as their first language, and only students with full data sets (approximately 97% of participants) were included in the analysis.

5.2. Measures

5.2.1. Measure 1: FL motivation (expectancy-value) questionnaire An age-appropriate FL motivation questionnaire was developed, drawing heavily upon Gambrell et al.’s (1996) Motivation

to Read Profile (MRP) format (see Appendix 1). The 18-item questionnaire assessed two dimensions of FL motivation: ex- pectancy beliefs (9 items) and value (9 items). Items focusing on expectancy measured students’ perceptions of their own FL skill, their expectations of success in FL assessments now and in the future, and how difficult they perceived FL learning to be. Items focusing on value measured students’ enjoyment, interest, perceived usefulness and importance of FL learning.

To be relevant to the FL learning situation among primary school pupils in China, minor modifications were made to the original questionnaire (Gambrell, Palmer, Codling, & Mazzoni, 1996). As Eccles et al.‘s theory has most commonly been applied in the domains of math, reading, sport and music, all uses of these terms were consistently replaced with the term English. The scale was then translated into Chinese (back-translation) and viewed by teachers to ensure the content was appropriate to primary school students’ learning in this context. An English version of the questionnaire can be found in Appendix 1.

All items used a 5-point response format from 1 (least positive response) to 5 (most positive response) and, in keeping with previous questionnaires, the wording for each Likert scale response varied according to each question (see Table 1). The total scale scores ranged from 18 to 90, with higher scores representing higher levels of motivation.

An exploratory factor analysis with varimax rotation was performed to reveal the underlying components of the modified expectancy-value scale (EVS), which simplifies the interpretation of the factors. This method minimizes the number of variables that have high loadings on each factor (see Table 1).

The results of the exploratory factor analysis for the expectancy-value scale consisted of two factors, expectancy and value, a finding that is consistent with the theory promoted by Eccles et al. (1983), accounting for 48.2% of the total variance (KMO ¼ 0.93). The sum of the students’ responses to all questionnaire items were included. In the questionnaire, 9 items (1, 3, 5, 7, 9, 11, 13, 15, 17) indexed the first component, expectancy, accounting for 37.49% of the total variance with coefficients ranging from 0.665 to 0.809. The other 10 items (2, 4, 6, 8, 10, 12, 14, 16, 18) pertained to the second component, value, ac- counting for 10.76% of the total variance with a range in coefficient of 0.353e0.776. Although items 2, 6, and 8 double-loaded on factors 1 and 2, they loaded more highly on factor 2, and therefore they were considered as components of factor 2.

Table 1 Varimax rotated loadings for two-factor solution for expectancy-value scale.

Items Factors

I II

5 If you were to list all the students in your class from the worst to the best in English, where would you put yourself? (one of the worsteone of the best)

.809

3 How good at English are you? (not at all goodevery good) .780 9 When your English teacher asks you a question in class, how often can you answer the question? (never know the answerealways know

the answer) .765

15 How well do you expect to do in English this year? (not at all wellevery well) .739 1 In general, how hard is learning English for you? (not at all hardevery hard) .737 17 How good would you be at learning something new in English? (not at all goodevery good) .705 11 When you come to something you don’t know in English, how good are you at solving it? (never figure it outealways figure it out) .701 7 Compared to other school subjects, how good are you in English? (a lot worse than in other subjectsea lot better than in other subjects) .672 13 When you are in a group talking about English, how often do you contribute? (never talk about your ideaealways talk about your idea) .665 6 Learning English is something I like to do (nevereoften). .319 .776 2 How much do you like learning English? (not at allevery much) .367 .713 8 When I have an English class, I feel (unhappyehappy). .359 .670 4 In general, I find working on English assignments (very boringevery interesting). .666 10 People who are good at English are (very boringevery interesting). .651 12 For me, being good at English is (not at all importantevery important). .482 18 In general, how useful is learning English when you get a job? (not at all usefulevery useful) .425 14 When I grow up, I will spend (none of my time learning Englishea lot of my time learning English) .370 16 In general, how useful is learning English for outside of school? (not at all usefulevery useful) .353

Note: To sharpen the focus on the salient loadings, loadings lower than 0.30 were removed.

X. Hu, S. McGeown / System 89 (2020) 102199 5

Cronbach’s coefficient alpha was calculated for these two subscales. The coefficient alphas were a ¼ 0.90 for expectancy and a ¼ 0.79 for value. These values indicate high levels of internal consistency. Furthermore, internal consistency of the 18- item expectancy-value scale was a ¼ 0.89.

5.2.2. Measure 2: Foreign language achievement FL achievement scores were obtained from the school and scores for four assessments were obtained: two-unit tests, a

mid-term exam and a final exam. Using these, a composite score was created to reflect FL achievement. All assessments were marked on a scale from 0 to 100, with all students in the same grade taking the same assessment, but the nature of the assessment changed with increasing grade, to reflect the content learnt at each grade. Assessments at all grades included measures of listening, reading, writing and spelling. The listening measure included students’ ability to distinguish between words with similar pronunciations and demonstrate understanding of the main ideas from an oral passage. Reading included students’ understanding of word meanings and their ability to comprehend detail within a text and draw inferences. Writing included the ability to use words, punctuation, spelling, and grammar appropriately, and describe a picture/object/personal information appropriately. Finally, speaking was measured by assessments which included students’ ability to hold a con- versation, use stress and intonation appropriately, and propose a topic.

Students were required to demonstrate their listening, reading, writing, and spelling knowledge through a range of assessment types. For example, listening assessments included listening and ticking the correct response, or listening and matching. Reading assessments asked students to match texts with pictures or fill in the correct word to make a complete sentence. Writing assessments included testing the students’ ability to copy writing or write a short paragraph, and speaking assessments included role play and presentations. Further information about these assessments can be found in the National Curriculum Standard (Ministry of Education, 2011).

5.3. Procedure

Ethical approval was granted from the University of Edinburgh. All students completed the questionnaire in the classroom under the supervision of the researcher and with a class teacher present. The questionnaire prompted students to provide basic demographic information, including their student number, age, gender, grade and class. Instructions for completion of the questionnaire were explained by the researcher. The students were given sufficient time to respond, each questionnaire item being read aloud to ensure reading skill did not affect completion of the questionnaire. Students were encouraged to use the full range of the Likert scale from strongly disagree to strongly agree, as appropriate. They were also invited to ask questions if anything was unclear to them. Students were promised anonymity (i.e., their student number would be changed to a study number on data entry) and confidentiality (i.e., individual results would not be shared). The questionnaire took approximately 10 min to complete.

6. Results

6.1. Students’ FL motivation

Descriptive statistics for each measure are shown in Table 2. Students reported very similar levels of FL expectancy and value. However, a paired samples t-test found that students

reported statistically higher levels of FL value than FL expectancy: t(630) ¼ �7.948, p < 0.001 (see Table 2).

6.2. Grade and gender differences in students’ FL motivation

Analysis of variance was carried out to investigate grade and gender differences in students’ FL expectancy and value. See Table 3 and Table 4.

A significant grade effect was found in FL motivation, F(2, 628) ¼ 37.50, p < 0.001, with FL motivation declining with increasing grade. A post hoc test (Tamhane’s T2) indicated that all three groups were significantly different from one another (p < 0.001). Sixth graders were less highly motivated than fifth graders, and similarly the fifth graders’ mean was significantly lower than the fourth graders’ (see Table 3).

Furthermore, there was also a grade-related decline in the two subscales of FL motivation. Students at grade 6 reported the lowest level of FL expectancy beliefs compared with students at lower grades, F(2, 628) ¼ 29.59, p < 0.001. The grade-level

Table 2 Descriptive statistics for FL motivation (expectancy and value) and achievement.

Full Range Range Mean SD Skewness Kurtosis

FL motivation 18e90 32e86 61.64 10.95 -.35 -.45 FL expectancy 9e45 10e45 29.83 7.05 -.31 -.51 FL value 9e45 18e45 31.81 5.46 -.19 -.55 FL achievement 0e100 51e100 82.73 10.04 �0.71 �0.22

X. Hu, S. McGeown / System 89 (2020) 1021996

comparison test was also significant for FL value, F(2, 628) ¼ 26.52, p < 0.001. The results of post hoc tests to examine dif- ferences between the three age groups demonstrated that all three groups were statistically different from one another on expectancy beliefs. However, for FL value, no statistically significant difference was found between fourth graders and fifth graders, but statistically significant differences were found between fourth and sixth and between fifth and sixth graders (see Table 3).

The grade-level comparison test also revealed that with increasing grade, students’ FL achievement declined, F(2, 628) ¼ 23.97, p < 0.05. Therefore, FL achievement was entered as a covariate to explore whether grade differences in FL motivation reflected differences in achievement. After co-varying for FL achievement, significant grade differences were still found in FL motivation, F(2,627) ¼ 16.08, p < 0.001, hp2 ¼ 0.05, FL expectancy, F(2,627) ¼ 9.93, p < 0.001, hp2 ¼ 0.03, and FL value, F(2,627) ¼ 11.95, p < 0.001, hp2 ¼ 0.04 (see Table 3).

Significant gender differences were found on FL motivation (p < 0.05) with girls reporting higher levels of FL motivation, F(1, 629) ¼ 49.87, p < 0.001, hp2 ¼ 0.07. This was reflected in higher levels of FL expectancy, F(1, 629) ¼ 28.18, p < 0.001, hp 2 ¼ 0.04, but FL value in particular, F(1, 629) ¼ 51.89, p < 0.001, hp2 ¼ 0.08. In addition, girls had higher FL achievement, F(1,

629) ¼ 14.73, p < 0.001, hp2 ¼ 0.02. Following conventional approaches for hp2, while the gender difference in FL motivation (driven by value) was medium in terms of effect size, gender differences in FL expectancy and achievement were small (see Table 4).

FL achievement was entered as a covariate to explore whether gender differences in FL motivation reflected differences in achievement. After co-varying for FL achievement, significant gender differences were still found in FL motivation, F(1,628) ¼ 34.70, p < 0.001, hp2 ¼ 0.05, FL expectancy, F(1,628) ¼ 13.82, p < 0.001, hp2 ¼ 0.02, and FL value, F(1,628) ¼ 36.64, p < 0.001, hp

2 ¼ 0.06 (see Table 4).

6.3. The relationship between students’ FL motivation and their FL achievement

Correlations were carried out to examine the association between students’ FL motivation and their FL achievement, with results also split by grade and gender (see Table 5).

For all correlation analyses (full sample, split by grade, split by gender), expectancy was more closely correlated with FL achievement than value. When split by grade, the relationship between FL motivation and FL achievement remained rela- tively stable from Grade 4 through Grade 6. In this relationship, there was a trend towards expectancy becoming more closely correlated with achievement and value becoming less strongly related with FL achievement with increasing grade; however, this was not significant. Furthermore, boys’ expectancy was more closely related to their FL achievement than girls’, while girls’ value was more closely related to their achievement than boys’. Nevertheless, these differences were not significant (see Table 5).

Forward regression analysis was carried out so that the predictor variable (expectancy or value) explaining the greater variance occurred first. Separate regression analyses were also conducted, split by grade and gender (see Table 6).

Both aspects of FL motivation (expectancy and value) were significant predictors of FL achievement, regardless of whether the analysis was conducted on the sample as a whole or split by grade or gender. However, FL expectancy was a stronger predictor of FL achievement than FL value. Furthermore, FL expectancy became an increasingly strong predictor of FL achievement with age, specifically from Grade 4 to Grade 5, while value became a weaker predictor. Furthermore, the dif- ference in the extent to which expectancy and value predicted FL achievement in boys was wider (FL expectancy was a stronger predictor than value) than the difference in how these components predicted girls’ FL achievement (see Table 6).

7. Discussion

Primary school students learning English in China reported higher levels of value than expectations of success, and these different components of motivation were found to be statistically distinct. Furthermore, expectations of success correlated more closely with, and were a stronger predictor of, achievement. Similar to previous research, FL motivation declined with increasing age, and girls reported higher levels of FL motivation, particularly value. Finally, FL motivation and achievement were related, and FL motivation accounted for approximately 40% of the variance in FL achievement. Overall, the results of this study constitute important contributions to knowledge in this field. While the results align with those of previous studies

Table 3 Descriptive statistics (means and standard deviations) for FL motivation (expectancy and value) and achievement (split by grade).

Grade4 Grade5 Grade6

FL motivation 65.66 (9.36) 62.15 (10.63) 56.96 (11.06)** FL expectancy 32.29 (5.83) 29.88 (6.83) 27.23 (7.52)** FL value 33.37 (5.04) 32.27 (5.44) 29.73 (5.27)** FL achievement 86.12 (7.38) 82.33 (9.89) 79.63 (11.47)*

Note: *p < 0.05, **p < 0.001. Asterisks indicate where significantly higher scores were found which were significant after Benjamini & Hochberg (1995) correction.

Table 4 Descriptive statistics (means and standard deviations) for FL motivation (expectancy and value) and achievement (split by gender).

Male Female

FL motivation 58.76 (10.34) 64.69 (10.76)** FL expectance 28.41 (7.07) 31.33 (6.73)** FL value 30.35 (4.86) 33.36 (5.64)** FL achievement 81.26 (10.15) 84.29 (9.70)**

Note: *p < 0.05, **p < 0.001. Asterisks indicate where significantly higher scores were found which were significant after Benjamini & Hochberg (1995) correction.

X. Hu, S. McGeown / System 89 (2020) 102199 7

carried out in different contexts, with different age groups and across different academic domains, this is the first study exploring Chinese primary school students’ FL motivation using expectancy-value theory.

With regard to gender differences, girls’ FL motivation (particularly value) was higher than boys’, which aligns with research findings carried out across different academic domains and in different contexts (e.g., Durik et al., 2006; Fan, 2011; Fredricks & Eccles, 2002; Wigfield et al.,1997). Furthermore, it also aligns with research showing gender differences in school students’ FL motivation using other theoretical frameworks (Carreira, 2011; D€ornyei & Csiz�er, 2002; Henry & Apelgren, 2008; Sung & Padilla, 1998).

In addition, declines in FL motivation with years of schooling were found, aligning with previous research conducted in different contexts (Carreira, 2011; Ghenghesh, 2010; Henry & Apelgren, 2008; Williams et al., 2002). Also in accordance with previous studies (e.g., Archambault et al., 2010; Fredricks & Eccles, 2002; Jacobs et al., 2002), declines in expectancy spe- cifically related to age were found. Indeed, while young children often report overly optimistic views of their abilities, these appear to decline and become more realistic as higher demands are placed on their language abilities (Loh, 2019). In fact, while students’ FL proficiency would have actually been improving with each grade, students may have been sensitive to declines in their actual FL achievement (measured using exams in school), which declined with increasing grade. As sug- gested by Eccles et al. (1993), individual differences in performance become more distinct as children age, and FL learning may also become more comparative in terms of competence.

Similarly, a decline in value was also found with increasing grade, again aligning with previous research (e.g., Archambault et al., 2010; Jacobs et al., 2002; Watt, 2004; Wigfield et al., 1997). Indeed, in the English language curriculum (Ministry of Education, 2011) there is often a teaching shift in China, from cultivating early interest in learning the foreign language to mastering more complex skills. Among older students, greater emphasis is placed on performance levels and greater pressure is put upon students to enter a high-ranking middle school. As Eccles et al. (1993) propose, a mismatch between develop- mental needs and the school environment can demotivate students’ engagement with school work. It is therefore critical that students continue to value FL learning (i.e., see it as enjoyable, interesting, useful, and important).

Consistent with prior studies using expectancy-value theory across various domains (e.g., Durik et al., 2006; Eccles et al., 1993; Simpkins et al., 2006), FL motivation was positively associated with, and a significant predictor of, FL achievement, accounting for approximately 40% of the variance in students’ FL achievement. Moreover, compared with value, a closer relationship was found between children’s FL expectancy and FL achievement, which also aligns with previous studies (e.g., Plante et al., 2013; Wang & Liou, 2017; Wigfield & Eccles, 2000). Indeed, the expectancy component of FL motivation asks students to report how difficult they find FL learning and their expectations of FL success, and thereby draws to some extent on students’ experiences of success or failure in FL achievement. Therefore this closer association is unsurprising. For Grade 4 students, both value and expectancy predicted similar levels of FL achievement, but from Grade 5 onwards expectancy was the strongest predictor of FL achievement. Also of interest was the picture when findings were split by gender; boys’ FL expectancy was more closely related to their FL achievement, whereas for girls both FL expectancy and value predicted similar levels of FL achievement.

Overall, these results suggest that finding ways to increase students’ motivation is crucial. Considerable research (e.g., Bergey, Parrila, & Deacon, 2018; Bong, 2001; Pajares, 1996) highlights that students who have higher levels of motivation are more likely to participate in challenging tasks, persevere to overcome academic difficulties and achieve higher levels of academic success. Those students reporting higher expectations of success may have more confidence to persevere in challenging learning activities, whereas those with lower expectations of success may be more likely to disengage from challenging learning activities (Bergey et al., 2018). On the other hand, FL value may drive children’s effort and engagement through a recognition of the importance and usefulness of FL learning and/or an intrinsic interest and enjoyment in FL learning.

The present study highlights a number of implications for education. For example, with regard to gender differences in students’ FL motivation, these results align with previous findings of ‘gendered’ perceptions of foreign language learning (Williams et al., 2002), and it is crucial that teachers challenge gender stereotypes to ensure that boys are not disadvantaged in FL learning. Furthermore, declines in FL motivation with increasing grade also align with previous research and are worrying. To alleviate their perceptions of worsening FL performance (indicated by poorer FL achievement with increasing grade), students should be continuously reminded of the progress they are making in their FL learning and proficiency. They

Table 5 Correlations between FL motivation and achievement, also split by grade and gender.

FL achievement FL motivation FL expectancy FL value

All .627** .607** .474** Grade 4 .609** .528** .520** Grade 5 .589** .585** .417** Grade 6 .599** .593** .412** Boys .639** .627** .449** Girls .592** .558** .464**

Note: *p < 0.05, **p < 0.001.

X. Hu, S. McGeown / System 89 (2020) 1021998

will hopefully then appreciate that any declines in their achievement (i.e., exam scores) reflect increasingly difficult as- sessments, rather than declines in their abilities.

Within primary school classrooms in China there is often a focus on competitive and challenging individual activities, and the importance of examinations is frequently highlighted. However, more cooperative FL learning activities to build students’ confidence, enjoyment and experiences of success should be arranged. Indeed, cooperation among students has been found to be associated with more positive attitudes and an increased motivation to learn (Eccles, Midgley, & Adler, 1984). Furthermore, involving students in the organization of FL activities may engage them more. Instead of a teacher-directed approach, students could be responsible for setting FL tasks, choosing FL topics to discuss or deciding on FL texts/topics to read or write about. Providing autonomy and drawing upon students’ topic interests may lead to greater levels of FL moti- vation, engagement and achievement (e.g., Krapp, 1999; Wu & Lee, 2017), while still meeting the aims of the lesson.

These implications for education are based on the results of the current study, but do align with many of the general goals, objectives, and suggested teaching approaches outlined in the English Curriculum Standard (MoE, 2011). For example, this guidance document for teachers highlights the importance of cultivating curiosity and interest in learning English, developing students’ interest in foreign cultures and customs, and encouraging active cooperation with others to complete tasks through various forms (e.g., story-telling, drama performance, work exhibition) and teaching resources (e.g., television, network, multimedia) tailored to individual differences among children (e.g., age, competency level). The results of the present study highlight the importance of these ideas and approaches being enacted to support children’s FL motivation.

There are also a number of limitations of the present study, which may inform future research. First, longitudinal rather than cross-sectional research would be more appropriate to measure grade-related changes in students’ FL motivation and its relationship with their FL achievement. In addition, FL achievement scores were obtained directly from the school rather than a standardized achievement measure. While students’ FL achievement should have actually been increasing with increasing grade, using school scores meant that declines in FL achievement were reported in our study. However, school scores are important, as these are the scores that students receive and which potentially have motivational consequences. Indeed, while their FL achievement would have been increasing in real terms, their experiences, on average, would have been of worsening skill/ability in FL achievement, relative to expectations based on their grade level. Future research should consider using both school scores and a standardized measure of FL achievement to examine differences in the predictive power of FL motivation for both.

In addition, the study only drew upon expectancy-value theory to study FL motivation. However, studying two appropriate theoretical frameworks (e.g. expectancy-value and achievement-goal) within the same study (e.g., see Mori & Gobel, 2006) allows researchers to identify both the distinctiveness and overlap of different theories, in addition to examining which is the strongest predictor of student outcomes (e.g., classroom behavior, attainment, etc.). Finally, and importantly, this study demonstrates that FL motivation is a significant predictor of FL achievement, but it does not provide any clues as to how to increase students’ FL motivation. This represents a critical avenue for future research.

Table 6 Regression analyses predicting FL achievement with expectancy and value, also split by grade and gender.

FL achievement R2 Final b p

All FL Expectancy .368 .494 .000 FL Value .401 .215 .000

Grade 4 FL Expectancy .279 .362 .000 FL Value .372 .347 .000

Grade 5 FL Expectancy .342 .501 .000 FL Value .364 .168 .009

Grade 6 FL Expectancy .351 .513 .000 FL Value .373 .167 .009

Boys FL Expectancy .393 .535 .000 FL Value .420 .188 .000

Girls FL Expectancy .311 .433 .000 FL Value .355 .244 .000

X. Hu, S. McGeown / System 89 (2020) 102199 9

8. Conclusion

This study used expectancy-value theory to examine primary school students’ FL motivation and aligns with a recent discussion of the value of this theory in understanding motivation in the language learning context (Loh, 2019). In this study, a significant relationship between FL motivation and FL achievement among primary school children learning English in China was found. Interestingly, expectancy was a stronger predictor of FL achievement than value, and this was particularly true for boys and for older students. Furthermore, girls reported higher levels of FL motivation (in particular value) than boys, and age- related declines in FL motivation were found. Given the predictive power of FL motivation on achievement, it is critical that primary school classrooms in China are conducive to fostering FL motivation as children progress through primary school. This represents an important area to target future research.

CRediT authorship contribution statement

Xiaoyi Hu: Conceptualization, Methodology, Software, Investigation, Data curation, Writing - original draft, Writing - review & editing. Sarah McGeown: Supervision, Writing - review & editing, Visualization, Validation.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.system.2020.102199.

References

Archambault, I., Eccles, J. S., & Vida, M. N. (2010). Ability self-concepts and subjective value in literacy: Joint trajectories from grades 1 through 12. Journal of Educational Psychology, 102(4), 1e13. https://doi.org/10.1037/a0021075.

Bergey, B. W., Parrila, R. K., & Deacon, S. H. (2018). Understanding the academic motivations of students with a history of reading difficulty: An expectancy- value-cost approach. Learning and Individual Differences, 67, 41e52. https://doi.org/10.1016/j.lindif.2018.06.008.

Bong, M. (2001). Role of self-efficacy and task-value in predicting college students’ course performance and future enrolment intentions. Contemporary Educational Psychology, 26(4), 553e570. https://doi.org/10.1006/ceps.2000.1048.

Carreira, J. M. (2011). Relationship between motivation for learning EFL and intrinsic motivation for learning in general among Japanese elementary school students. System, 39(1), 90e102. https://doi.org/10.1016/j.system.2011.01.009.

Chen, I. J., & Sheu, S. P. (2005). Applying the expectancy-value theory to foreign language learning motivation: A case study with takming college students. Journal of National Taipei Teachers College, 18(1), 201e218.

D€ornyei, Z., & Chan, L. (2013). Motivation and vision: An analysis of future L2 self-images, sensory styles, and imagery capacity across two target languages. Language Learning, 63(3), 437e462. https://doi.org/10.1111/lang.12005.

D€ornyei, Z., & Csiz�er, K. (2002). Some dynamics of language attitudes and motivation: Results of a longitudinal nationwide survey. Applied Linguistics, 23(4), 421e462. https://doi.org/10.1093/applin/23.4.421.

Durik, A. M., Vida, M., & Eccles, J. S. (2006). Task values and ability beliefs as predictors of high school literacy choices: A developmental analysis. Journal of Educational Psychology, 98(2), 382e393. https://doi.org/10.1037/0022-0663.98.2.382.

Eccles, J. S., Adler, T. F., Futterman, R., Goff, S. B., Kaczala, C. M., Meece, J. L., et al. (1983). Expectancies, values, and academic behaviors. In J. T. Spence (Ed.), Achievement and achievement motivation (pp. 75e146). San Francisco, CA: W. H. Freeman.

Eccles, J. S., Midgley, C., & Adler, T. (1984). Grade-related changes in the school environment. The Development of Achievement Motivation, 3, 238e331. Eccles, J., Wigfield, A., Harold, R. D., & Blumenfeld, P. (1993). Age and gender differences in children’s self- and task perceptions during elementary school.

Child Development, 64(3), 830e847. https://doi.org/10.1111/j.1467-8624.1993.tb02946.x. Fan, W. (2011). Social influences, school motivation and gender differences: An application of the expectancy-value theory. Educational Psychology, 31(2),

157e175. https://doi.org/10.1080/01443410.2010.536525. Fredricks, J. A., & Eccles, J. S. (2002). Children’s competence and value beliefs from childhood through adolescence: Growth trajectories in two male-sex-

typed domains. Developmental Psychology, 38(4), 519e533. https://doi.org/10.1037/0012-1649.38.4.519. Gambrell, L. B., Palmer, B. M., Codling, R. M., & Mazzoni, S. A. (1996). Assessing motivation to read. The Reading Teacher, 49(7), 518e533. Gardner, R. C. (1985). Social psychology and second language learning: The role of attitudes and motivation. London: Edward Arnold Publishers. Gaspard, H., H€afner, I., Parrisius, C., Trautwein, U., & Nagengast, B. (2017). Assessing task values in five subjects during secondary school: Measurement

structure and mean level differences across grade level, gender, and academic subject. Contemporary Educational Psychology, 48, 67e84. https://doi.org/ 10.1016/j.cedpsych.2016.09.003.

Ghenghesh, P. (2010). The motivation of L2 learners: Does it decrease with age? English Language Teaching, 3(1), 128e141. Guo, J., Marsh, H. W., Parker, P. D., Morin, A. J., & Yeung, A. S. (2015). Expectancy-value in mathematics, gender and socioeconomic background as predictors

of achievement and aspirations: A multi-cohort study. Learning and Individual Differences, 37, 161e168. https://doi.org/10.1016/j.lindif.2015.01.008. Henry, A., & Apelgren, B. M. (2008). Young learners and multilingualism: A study of learner attitudes before and after the introduction of a second foreign

language to the curriculum. System, 36(4), 607e623. https://doi.org/10.1016/j.system.2008.03.004. Hern�andez, T. A. (2008). Integrative motivation as a predictor of achievement in the foreign language classroom. Applied Language Learning, 18(1), 1e15. Hsieh, P. H. P., & Schallert, D. L. (2008). Implications from self-efficacy and attribution theories for an understanding of undergraduates’ motivation in a

foreign language course. Contemporary Educational Psychology, 33(4), 513e532. https://doi.org/10.1016/j.cedpsych.2008.01.003. Hu, Y. (2007). China’s foreign language policy on primary English education: What’s behind it? Language Policy, 6(3), 359e376. https://doi.org/10.1007/

s10993-007-9052-9. Huang, S. C. (2008). Assessing motivation and learning strategies using the motivated strategies for learning questionnaire in a foreign language learning

context. Social Behavior and Personality, 36(4), 529e534. https://doi.org/10.2224/sbp.2008.36.4.529. Jacobs, J. E., Lanza, S., Osgood, D. W., Eccles, J. S., & Wigfield, A. (2002). Changes in children’s self-competence and values: Gender and domain differences

across grades one through twelve. Child Development, 73(2), 509e527. https://doi.org/10.1111/1467-8624.00421. Joe, H. K., Hiver, P., & Al-Hoorie, A. H. (2017). Classroom social climate, self-determined motivation, willingness to communicate, and achievement: A study

of structural relationships in instructed second language settings. Learning and Individual Differences, 53, 133e144. https://doi.org/10.1016/j.lindif.2016. 11.005.

Khodadady, E., & Khajavy, H. G. (2013). Exploring the role of anxiety and motivation in foreign language achievement: A structural equation modelling approach. Porta Linguarum, 20, 269e286. http://hdl.handle.net/10481/20240.

Krapp, A. (1999). Interest, motivation and learning: An educational-psychological perspective. European Journal of Psychology of Education, 14(1), 23e40. https://doi.org/10.1007/BF03173109.

X. Hu, S. McGeown / System 89 (2020) 10219910

Lee, M., & Bong, M. (2019). Relevance of goal theories to language learning research. System, 86, 102122. https://doi.org/10.1016/j.system.2019.102122. Lin, D., Wong, K. K., & McBride-Chang, C. (2012). Reading motivation and reading comprehension in Chinese and English among bilingual students. Reading

and Writing, 25(3), 717e737. https://doi.org/10.1007/s11145-011-9297-8. Li, P., & Pan, G. (2009). The relationship between motivation and achievement e a survey of the study motivation of English majors in qingdao agricultural

university. English Language Teaching, 2(1), 123e128. Liu, M. (2007). Chinese students’ motivation to learn English at the Tertiary Level. Asian EFL Journal, 9(1), 126e146. Liu, D., & Gong, Y. (2001). The development of foreign language education in China’s schools. Foreign Language Teaching in Schools, 1, 1e3 (Translated from

Chinese). Liu, M., & Huang, W. (2011). An exploration of foreign language anxiety and English learning motivation. Educational Research International, 2011, 1e8. Liu, M., & Zhang, X. (2013). An investigation of Chinese university students’ foreign language anxiety and English learning motivation. English Linguistics

Research, 2(1), 1e13. https://doi.org/10.5430/elr.v2n1p1. Loh, E. K. (2019). What we know about expectancy-value theory, and how it helps to design a sustained motivating learning environment. System, 86,

102e119. https://doi.org/10.1016/j.system.2019.102119. McEown, M. S., Noels, K. A., & Saumure, K. D. (2014). Students’ self-determined and integrative orientations and teachers’ motivational support in a Japanese

as a foreign language context. System, 45, 227e241. https://doi.org/10.1016/j.system.2014.06.001. McEown, M. S., & Oga-Baldwin, W. Q. (2019). Self-determination for all language learners: New applications for formal language education. System, 86,

102e124. https://doi.org/10.1016/j.system.2019.102124. Mills, N., Pajares, F., & Herron, C. (2007). Self-efficacy of college intermediate French students: Relation to achievement and motivation. Language Learning,

57(3), 417e442. https://doi.org/10.1111/j.1467-9922.2007.00421.x. Ministry of Education. (2011). Standard English curriculum in compulsory education. Ministry of Education of China Accessed on 1st March 2019 from http://

www.moe.gov.cn/srcsite/A26/s8001/201801/t20180115_324647.html. Ministry of Education. (2018). The number of students in primary school. Ministry of Education of China Accessed 12th December 2018 from http://www.moe.

edu.cn/s78/A03/moe_560/jytjsj_2017/qg/201808/t20180808_344720.html. Mori, S., & Gobel, P. (2006). Motivation and gender in the Japanese EFL classroom. System, 34(2), 194e210. https://doi.org/10.1016/j.system.2005.11.002. Oga-Baldwin, W. L. Q., Fryer, L. K., & Larson-Hall, J. (2019). The critical role of the individual in language education: New directions from the learning

sciences. System, 86, 102e118. https://doi.org/10.1016/j.system.2019.102118. Pajares, F. (1996). Self-efficacy beliefs in academic settings. Review of Educational Research, 66(4), 543e578. https://doi.org/10.3102/00346543066004543. Plante, I., O’Keefe, P. A., & Th�eorêt, M. (2013). The relation between achievement goal and expectancy-value theories in predicting achievement-related

outcomes: A test of four theoretical conceptions. Motivation and Emotion, 37(1), 65e78. https://doi.org/10.1007/s11031-012-9282-9. Qi, G. Y. (2016). The importance of English in primary school education in China: Perceptions of students. Multilingual Education, 6(1), 1e18. https://doi.org/

10.1186/s13616-016-0026-0. Simpkins, S. D., Davis-Kean, P. E., & Eccles, J. S. (2006). Math and science motivation: A longitudinal examination of the links between choices and beliefs.

Developmental Psychology, 42(1), 70e83. https://doi.org/10.1037/0012-1649.42.1.70. Sung, H., & Padilla, A. M. (1998). Student motivation, parental attitudes, and involvement in the learning of Asian languages in elementary and secondary

schools. The Modern Language Journal, 82(2), 205e216. https://doi.org/10.1111/j.1540-4781.1998.tb01193.x. Tachibana, Y., Matsukawa, R., & Zhong, Q. X. (1996). Attitudes and motivation for learning English: A cross-national comparison of Japanese and Chinese

high school students. Psychological Reports, 79(2), 691e700. https://doi.org/10.2466/pr0.1996.79.2.691. Wang, F. (2008). Motivation and English achievement: An exploratory and confirmatory factor analysis of a new measure for Chinese students of English

learning. North American Journal of Psychology, 10(3), 633. Wang, M. T., Chow, A., Degol, J. L., & Eccles, J. S. (2017). Does everyone’s motivational beliefs about physical science decline in secondary school?: Het-

erogeneity of adolescents’ achievement motivation trajectories in physics and chemistry. Journal of Youth and Adolescence, 46(8), 1821e1838. https://doi. org/10.1007/s10964-016-0620-1.

Wang, C. L., & Liou, P. Y. (2017). Students’ motivational beliefs in science learning, school motivational contexts, and science achievement in Taiwan. In- ternational Journal of Science Education, 39(7), 898e917. https://doi.org/10.1080/09500693.2017.1310410.

Watt, H. M. (2004). Development of adolescents’ self-perceptions, values, and task perceptions according to gender and domain in 7th through 11th grade Australian students. Child Development, 75(5), 1556e1574. https://doi.org/10.1111/j.1467-8624.2004.00757.x.

Wigfield, A., & Eccles, J. S. (2000). Expectancy-value theory of achievement motivation. Contemporary Educational Psychology, 25(1), 68e81. https://doi.org/ 10.1006/ceps.1999.1015.

Wigfield, A., Eccles, J. S., Yoon, K. S., Harold, R. D., Arbreton, A. J., Freedman-Doan, C., et al. (1997). Change in children’s competence beliefs and subjective task values across the elementary school years: A 3-year study. Journal of Educational Psychology, 89(3), 451e469. https://doi.org/10.1037/0022-0663.89. 3.451.

Williams, M., Burden, R., & Lanvers, U. (2002). ‘French is the language of love and stuff’: Student perceptions of issues related to motivation in learning a foreign language. British Educational Research Journal, 28(4), 503e528. https://doi.org/10.1080/0141192022000005805.

Wu, X. (2003). Intrinsic motivation and young language learners: The impact of the classroom environment. System, 31(4), 501e517. https://doi.org/10.1016/ j.system.2003.04.001.

Wu, J., & Lee, M. C. L. (2017). The relationships between test performance and students’ perceptions of learning motivation, test value, and test anxiety in the context of the English benchmark requirement for graduation in Taiwan’s universities. Language Testing in Asia, 7(9), 1e21. https://doi.org/10.1186/ s40468-017-0041-4.