This paper is the first to examine the motivational disposition of Nepalese learners of L2 English. Based on an adapted version of the questionnaire in (Kormos, Judit & Kata Csizér. 2008. Age-related differences in motivation of learning English as a foreign language: Attitudes, selves, and motivated behavior. Language Learning 58. 327–355. Doi:10.1111/j.1467-9922.2008.00443.x.), we test the robustness and culture-specific applicability of well-known motivational antecedents to this learner population, and we investigate how the effects of these antecedents are mediated by the learners’ gender, age and regional aspects of the educational setting. In doing so, we offer novel ways of analyzing the data: Firstly, we employ random forests and conditional inference trees for assessing the relative importance of motivational antecedents. Secondly, we complement the traditional ‘scale-based approach’, which focuses on holistic constructs like the ‘Ideal L2 Self’, with an ‘item-based approach’ that highlights more specific components of such scales. The results are interpreted with reference to the L2 Motivational Self System (Dörnyei, Zoltán. 2005. The psychology of the language learner: Individual differences in second language acquisition. Mahwah, NJ: Lawrence Erlbaum) and to previous studies on other Asian populations of L2 learners.
Apart from the acknowledgements made directly in the text, we would like to express our gratitude to numerous people involved in the data-collection process as such. First of all, we thank all of the students and pupils who participated in this study and all of the teachers and principals that allowed us to introduce the project to them. Special thanks go to the staff of Himanchal Education Foundation, who accepted the second author of this paper as a volunteer and gave him the opportunity to conduct this study in Nangi Village. Dr Mahabir Pun and Chitra Bahadur Tilija Pun, HEF’s primary representatives in Nepal, as well as Dr Debra Stoner and Dr Leonard Skov, volunteer advisers of HEF in Nebraska, constantly assisted in getting into contact with Nepali learners and teachers. Furthermore, this project would not have been possible without the help of Prakash Gautam, teacher at Enlightened International Academy and Sagarmatha Higher Secondary School, Pokhara, who translated the questionnaire and who allowed Gregor to visit his lessons and introduce the project to his students. We are also obliged to Alex Clark and Joceline Houdijk, two volunteers working in Pokhara, for taking some questionnaires to their classes and even feeding the data into the online version. Furthermore, we owe thanks to Prof. Dr Debendra Bahadur Lamichane, head of National Inventive Boarding School and lecturer at Thribuvan College, Pokhara, for encouraging over 50 students to participate in the project.
Appendix 1. Survey instrument
In this supplementary material, we provide the full questionnaire in a systematic form, where each scale is presented along with all of its constituent items. It goes without saying the participants of our study received a thoroughly randomized version of the questionnaire.
In the following lists, …
items marked with an *asterisk are those whose scores were inverted for the statistical analysis in order to maintain a consistent directionality in what the questions asked for (and in how they relate to motivated learning behaviour).
items printed in italics were added to the present questionnaire (as compared to Kormos and Csizér 2008).
How much would you like to become similar to the people who speak English?
How much do you like English?
How important do you think learning English is in order to learn more about the culture and art of its speakers?
Attitude towards the L2 community
How much do you like the people who live in the United States?
How much do you like the people who live in the United Kingdom?
How much would you like to travel to the UK?
How much would you like to travel to the USA?
How much do you like meeting foreigners from English-speaking countries?
How much do you think knowing English would help you if you travelled abroad in the future?
How much do you think knowing English would help you in the future?
How important do you think English is in the world these days?
Knowledge of English would make me a better educated person.
How much do you like movies made in the United States?
How much do you like the pop music of the USA?
How much do you like the magazines made in the United States?
Vitality of the L2 community
How important is the United Kingdom in the world?
How rich and developed do you think the United Kingdom is?
How rich and developed do you think the United States is?
How important a role do you think the United States plays in the world?
I am sure I will be able to learn a foreign language well.
Learning a foreign language is a difficult task.*
I think I am the type who would feel anxious if I had to speak to someone in a foreign language.*
I feel uneasy speaking English with a native speaker.
I would get tense if a foreigner asked me for the way in English.
If there was an opportunity to meet an English speaker, I would feel nervous.
I am worried that native speakers of English would find my English strange.
It embarrasses me to volunteer answers in our English class.
I get nervous when I am speaking in my English class.
I am afraid the other students will laugh at me when I speak English.
I always feel that the other students speak English better than I do.
People around me tend to think that it is a good thing to know foreign languages.
Nobody really cares whether I learn English or not.*
For people where I live learning English is not really necessary.*
My parents consider foreign languages important school subjects.
My parents really encourage me to study English.
My parents encourage me to practise my English as much as possible.
My parents have stressed the importance English will have for me in my future.
My parents feel that I should really try to learn English.
Language-learning attitudes and experience
Learning English is really great.
I really enjoy learning English.
I find learning English really interesting.
I think that foreign languages are important school subjects.
Studying English will help me to understand people from all over the world (not just English-speaking countries).
If I could speak English well, I could get to know more people from other countries (not just English-speaking countries).
I would like to be able to use English to communicate with people from other countries.
In the future, I imagine myself working with people from other countries (not just English speaking countries).
In the future, I really would like to communicate with foreigners.
Ideal L2 Self
I like to think of myself as someone who will be able to speak English.
If my dreams come true, I will use English effectively in the future.
Whenever I think of my future career, I imagine myself being able to use English.
When I think about my future, it is important that I use English.
I can imagine speaking English with international friends.
The things I want to do in the future require me to speak English.
The job I imagine having in the future requires that I speak English well.
Ought-to L2 Self
If I fail to learn English I'll be letting other people down.
Learning foreign languages makes me fear that I will feel less Nepali because of it.*
Learning English is necessary because it is an international language.
For me to be an educated person I should be able to speak English.
I study hard in order not to disappoint my teacher.
I am responsible for the quality of my English.
I am not responsible for the quality of my English.*
The quality of my English does not depend on factors that I cannot control (such as the quality of my teachers, lessons and learning material).
The quality of my English depends on factors that I cannot control (such as the quality of my teachers, lessons and learning material).*
The quality of my English depends how much effort I put into my studies.
No matter how much I study, it does not improve my English.*
If I want to speak English better, I have to work harder in the future.
If I continue studying English as I do now, my English will be very good.
Perceived L2 difficulty
I think English is more difficult than Nepali.
I like English but it is so hard to learn.
I think English is a very difficult language.
I don't like English because it is so difficult.
Motivated learning behaviour
I am willing to work hard at learning English.
I am determined to push myself to learn English.
I can honestly say that I am really doing my best to learn English.
It is very important for me to learn English.
Learning English is one of the most important aspects in my life.
If an English course was offered in the future, I would like to take it.
If I could have access to English-speaking radio, I would try to listen to it often.
When I hear an English song on the radio, I listen carefully and try to understand all the words.
I always look forward to our English classes.
Sometimes I study more than is expected of me.
I like to improve my English outside of the classroom.
Appendix 2. Statistical analysis
In this supplementary material, we detail the analytical procedures of our study. All analyses were performed in the open-software R, version 3.2.3 (R Development Core Team 2015).
As explained in the main text, we used principal component analysis (PCA) in conjunction with factor analysis (FA) for testing the dimensionality of the purported scales. For each scale in question, we began the analysis by checking three crucial prerequisites for both PCA and FA: First, we applied the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy in order to test whether, given a particular set of items to be factorized, our sample of observations (nmax=149) was sufficient to apply factor analysis in the first place.  Following recommendations by Kaiser (1974), we took KMO values>0.5 to be acceptable; lower values indicated that no PCA/FA could be performed (without, for example, removing any of the items in question, cf. also Backhaus et al. 2016: 398). Secondly, the correlational structure of the respective items was tested, in two complementary ways: Bartlett’s test of sphericity (Bartlett 1951), implemented in the R function cortest.bartlett() in the ‘psych’ package, determines whether there is a minimal correlational structure in the data (by comparing the data to a so-called identity matrix without any correlations). The opposite scenario, i. e. that the items correlate too highly, is as problematic for FA as such extreme multicollinearity is for multiple regression models, because it becomes impossible to gauge the unique contributions of highly correlated items to a given factor. This scenario was tested by calculating the so-called determinant, using the det() function from the R base distribution; the determinant value should be substantially greater than 0.00001 (since perfect correlation, so-called singularity, is indicated by a determinant of zero, cf. Field et al. 2012: 771).
For the PCAs, we used the PCA()function from the ‘FactoMineR’ package. Missing values in our data were imputed by the function imputePCA() from the ‘missMDA’ package (cf. Josse and Husson 2012 for technical details). For determining the optimal number of principal components, we inspected the eigenvalues of each component, using scree plots (cf. Catell 1966) in conjunction with Kaiser’s (1960) criterion of choosing components with eigenvalues greater than 1. The ideal scenario in our case is, of course, one in which a single principal component emerges and where, following Stevens’ (2002) guidelines for our sample size of 100<n<200, each item has a loading of>0.4 onto the component in question.
For the factor analysis, we used the function factanal() from the R base distribution. In each case, we took the smallest number of components suggested as adequate by the PCA (i. e. ideally just one) and tested whether this yields a satisfactory factor solution in FA. In order to improve interpretability with multi-component solutions, we employed ‘promax’ as an oblique method for factor rotation.  To inspect the factor loadings, we examined the pattern matrix and again took Stevens’>0.4 criterion as the benchmark for the contribution of an item to a factor. Finally, we also used the maximum-likelihood residual analysis implemented in factanal(), which indicates whether the number of factors specified is sufficient in order to account for the variation in the participating items. 
The last step of the analysis was to assess the reliability of the resulting scales, applying Cronbach’s alpha (Cronbach 1951). Following standard recommendations in the literature (e. g. also Dörnyei 2010b: 95 on questionnaire data), we used α ≥ 0.7 as the critical threshold for the reliability of a scale. In Appendix 1, where we provide the full questionnaire, we also indicate which items were reversed for the reliability analysis.
In the following overview, we turn to each scale from the paper individually again and provide the central results from the PCA, FA and reliability analysis. Unless specified otherwise, it can be taken for granted that both the Bartlett criterion and the determinant yielded acceptable results.
Integrativeness (items 1–3): KMO=0.57. PCA suggests that a single-component solution may be acceptable, but this is not corroborated by FA (item 3 showed a loading<0.4 and a single-factor solution is not judged sufficient). α(1–3)=0.48, α(1–2)=0.42. We thus obtain a heterogeneous and unreliable scale, even if shorted to the more homogeneous items 1 and 2.
Attitudes towards the L2 speakers and the community (items 4–8): KMO=0.7. PCA suggests that a single-component solution is perfectly adequate. In FA, a single-factor solution is not judged sufficient by the significance test, but all items show loadings>0.4 on the factor. The variance accounted for (and the solution as a whole) can be improved if item 8 is removed, and the final scale reported in the paper is reliable at α(4–7)=0.71.
Instrumentality (items 9–12): An initial analysis showed that item 12, which (in contrast to Kormos and Csizér 2008) we had reallocated to this scale because of its instrumental flavour, introduced severe heterogeneity, as it did not correlate with the other items on the scale. Consequently, we abandoned it from the scale and operated with items 9–11 only. KMO=0.64. PCA then suggested a satisfactory one-component solution. FA’s significance test did not consider a single factor to be sufficient, but again all items had acceptable loadings of>0.4 on the factor. The scale is borderline reliable (α(9–11)=0.68) and we hence retained it in further analyses.
Cultural interest (items 13–15): KMO=0.57. PCA suggests that a single-component solution may be acceptable, but this is not corroborated by FA (item 3 showed a loading<0.4 and a single-factor solution is not judged sufficient). α(13–15)=0.57. The scale is thus both heterogeneous and unreliable.
Vitality of the L2 community (items 16–19): KMO=0.72. PCA suggests a clearly one-dimensional structure with high loadings of all items. FA with one factor confirms loadings>0.4, but fails the significance test. In view of the strong signals from all other criteria, however, the scale was included in its original composition. This is confirmed by reliability analysis (α(16–19)=0.78).
Linguistic self-confidence (items 20–22): In this case, the Bartlett test yields a non-significant result (p=0.08), and KMO=0.44. Therefore, the correlational structure of these items is so weak that not even the basic prerequisites for PCA/FA are fulfilled.
Language use anxiety (items 23–26): KMO=0.73. Both PCA and FA suggest a one-dimensional structure with high loadings of all items onto the single factor. This monofactorial solution also passes FA’s significance test and is reliable at α(23–26)=0.74.
Classroom anxiety (items 27–30): KMO=0.65. An initial analysis showed that item 30 makes the scale inconsistent and leads to an unacceptable alpha value. Removing this item returns a single-component structure in PCA and a monofactorial solution in FA with all loadings>0.4. This scale is reliable at α(27–30)=0.73.
Milieu (items 31–34): This scale violates some crucial prerequisites for PCA/FA, i. e. Bartlett’s test is non-significant (p=0.08) and KMO=0.49. Even if we remove item 32 because of its large amount of NAs, the results do not improve significantly. Not surprisingly, then, the scale is also completely unreliable (α(31–34)=0.09).
Parental encouragement (items 35–38): KMO=0.77. Although both PCA and FA suggest a satisfactory one-dimensional solution, some improvement in the factor loadings and in the variance explained by a single factor can be achieved if item 36 is removed. The resulting scale is reliable at α(35,37,38)=0.74.
Language-learning attitudes (items 39–42): KMO=0.51. Both PCA and FA suggest a two-component solution with complex cross-loadings and an overall unreliable composition (α(39–42)=0.51). As explained in the paper, a subscale of items 39 and 42 achieves a much higher degree of reliability (α(39,42)=0.7) and was hence retained in further analyses.
International posture (items 43–47): Item 45 had to be removed due to a large number of NAs. For the remaining items, KMO=0.53. PCA suggests that a single-component solution might be possible (though certainly not optimal), and all signals of a FA clearly speak against this. Removing individual items does not improve this situation, and the scale is not reliable (α(43–47)=0.55).
Ideal L2 Self (items 48–54): KMO=0.76. Both PCA and FA suggest a two-dimensional construct, but since an outstanding first component emerges in PCA, with high loadings of all items, and since a monofactorial solution is still reliable (α(48–54)=0.71), we had sufficient reason to believe that this construct is valid enough (and certainly more homogeneous than truly diverse scales). Note that removing items with the lowest loadings does not improve the residual diagnostics in FA, nor does it lead to a more reliable alpha value.
Ought-to L2 Self (items 55–59): KMO=0.59. Both PCA and FA suggest a complex (at least two-factor) structure with many cross-loadings, and an insufficient α(55–59) of −0.52.
Self-efficacy (items 60–67): KMO=0.54. Both PCA and FA suggest a very heterogeneous composite of three factors. As far as reliability is concerned, neither the scale in its original composition (α(60–67)=0.06) nor in a version reduced to the most strongly correlating items (α(60, 64, 67)=0.52) reaches the criterion level.
Perceived L2 difficulty (items 68–71): KMO=0.69. Both PCA and FA suggest a satisfactory one-component solution, as all items show high loadings on the first dimension and even the residual analysis shows that a monofactorial analysis is sufficient. The scale is reliable at α(68–71)=0.7.
Motivated learning behaviour (items 72–82): A first round of analysis showed that the two items we added to the original scale (i. e. 81 and 82) introduced heterogeneity. Without them, KMO=0.83. PCA suggests that a one-dimensional solution may be warranted, with high loadings of all items on the first dimension (similarly to the Ideal L2 Self above). A similarly good loading structure emerges in a monofactorial FA (even though this does not pass the residual test). The scale is reliable at α(72–80)=0.83.
Statistical parameters of reliable scales
This part of the appendix relates to Section 4 of the paper. The relevant data can be found in Table 3 on the next page, which has to be read as follows: For each of the reliable scales listed in the first column, we provide comparative data on three different subsamples, i. e. female/male, remote/urban and pupil/student. We first list the mean and standard deviation for each part of the respective subsample (e. g. female versus male) and then compare the difference in their average scores by means of non-parametric Wilcoxon rank-sum tests (excluding missing values on a pairwise basis).
|Scale||Items||Sample contrasts||Statistics of sample contrasts||Correlation with motivated learning behaviour|
|Attitudes to the L2 community||4–7||female:male||3.91||3.90||0.6||0.77||2474||0.636||0.2||0.64||0.08||<0.001||3.20||0.001**|
|Vitality of the L2 community||16–19||female:male||4.07||4.11||0.65||0.70||2448.5||0.565||0.017||0.267||0.445||0.011||1.51||0.131|
|Language use anxiety||23–26||female:male||2.98||2.91||1.01||0.98||2737||0.563||−0.366||−0.20||<0.001||0.044||0.15||0.294|
|Perceived L2 difficulty||68–71||female:male||2.70||2.67||0.82||0.93||2651.5||0.813||−0.458||−0.171||<0.001||0.075||1.89||0.059*|
|Motivated learning behaviour||72–80||female:male||4.29||4.14||0.47||0.59||2932||0.174|
In the right-hand part of the table, we then display the (non-parametric) correlation coefficient Spearman’s ρ (rho) for the correlation of the scale in question with the scale of motivated learning behaviour, i. e. the mean value of items 72–80. In testing these correlations for significance, we generally employed one-tailed tests because our hypotheses were clearly directional (e. g. we supposed that an Ideal L2 Self would correlate positively with motivated learning behaviour). However, for testing such correlations in the subsamples (e. g. differences between male and female participants), two-tailed tests were used.
The last two columns of the table feature a test statistic z and its associated p-value. z was obtained by applying Fisher’s r-to-z transformation, which allows comparing whether two correlation coefficients (e. g. that of the rural and the urban subsample) differ significantly from one another. 
As is explained in the main text, our specific data set (with many predictors relative to the number of observations, heavy deviations from normality, potentially correlated predictors, etc.) suggests employing a specific non-parametric regression procedure known as conditional inference trees (CIT) and random forests (RF). The gist of these techniques is described in the main text, and we elaborate on them here and add some specifications of the software and algorithms we drew on.
Individual CITs were produced by using the ctree() function in the ‘party’ package (cf. Strobl et al. 2009a), which has an inbuilt ‘stop criterion’ when no significant association between a predictor and the response variable can be detected anymore, thus preventing further splits of the tree. The function works on the basis of an unbiased algorithm, which (in contrast to other implementations like rpart()) does not favour variables with many distinct levels (such as continuous or multilevel categorical predictors over binary ones). The function is thus particularly suitable to our specific mix of predictors.
RFs are so-called ensemble methods in which many individual trees are grown from random subsets of the data. For each of the trees involved, the whole sample is randomly divided into a learning set (the so-called ‘in-bag’ observations) and a test set (the so-called ‘out-of-bag’ observations), and only a random subset of the predictor variables is used to generate the tree. In this way, a truly diverse set of trees is grown, and by averaging over these trees, the relative importance of each predictor can be assessed more reliably than in a single tree model for the data (since a whole ensemble of trees is less sensitive to idiosyncratic properties of the specific sample at hand). The relative weight of a predictor is gauged by a rather complex procedure of permutation (cf. Breiman 2001). The basic idea is that the values of the predictor variable in question are shuffled so that there is no association with the response variable anymore; if this permuted version of the predictor weakens the performance of the model as compared to using the original, unpermuted predictor, one can conclude that the predictor in question is essential for modelling the response variable (cf. also Tagliamonte and Baayen 2012: 160–161 for a digestible summary of the technicalities).
For growing random forests, we used the cforest() function of the ‘party’ package.  For the reasons mentioned above, we used the default option controls=cforest_unbiased to ensure unbiased treatment of different types of variables in each tree. Furthermore, in assessing the relative impact of the predictors, we extracted the conditional permutation variable importance, using the varimp(obj, conditional=T) routine. This procedure, proposed in Strobl et al. (2008), specifically ensures that the importance of correlated predictors is not overestimated and thus reduces the potential problem of collinearity. In growing the random forest, we used the default mtry=5 setting (i. e. with five randomly preselected predictor variables at each split of each tree) but with a considerably larger number of trees than specified in the default settings (3,000 instead of 500), in order to do justice to the rather high number of predictor variables and to make the results more reliable (cf. Strobl et al. 2009b: 343). Following recommended practices, we tried several different parameter settings (including varying random seeds), but obtained stable results as far as the relative variable importance of the strongest predictors was concerned (while the relative ranking of less influential predictors changed across the various runs).
For determining the statistical significance of a predictor in the random forest solution, we followed Strobl, Malley and Tutz (2009: 343), who suggest that “all variables with importance that is negative, zero, or positive but with a value that lies in the same range as the negative values can be excluded from further exploration. The rationale for this rule of thumb is that the importance of irrelevant variables varies randomly around zero. Therefore, positive variation of an amplitude comparable to that of negative variation does not indicate an informative predictor variable, whereas positive values that exceed this range may indicate that a predictor variable is informative.” More specifically, we took the absolute value of the most negative predictor as a threshold for the significance of the positive predictors.
The goodness-of-fit of CITs and RFs can be obtained in a similar way as for multiple linear regression models, i. e. by using the regular formula for the coefficient of determination: R2=1 – SSres/SStot, where SSres denotes the residual sum of squares and SStot the total sum of squares (cf. Field et al. 2012: 250). This procedure capitalizes on the fact that, in both CITs and RFs, one can compare the predicted and the observed values of the response variable (based on a single model in CITs and the average of many models in RFs). What we observe in our data with regard to R2 (compare Figures 2 and 3, or 4 and 5) is rather typical: The goodness-of-fit of a single CIT is usually slightly worse than the predictive accuracy of a whole ensemble of trees, i. e. of a RF, especially if the latter is grown from a very large number of trees.
Backhaus, Klaus, Bernd Erichson, Wulff Plinke & Rolf Weiber. 2016. Multivariate Analysemethoden: Eine anwendungsorientierte Einführung, 14th edn. Berlin & Heidelberg: Springer.10.1007/978-3-662-46076-4Search in Google Scholar
Bong, Mimi & Richard E. Clark. 1999. Comparison between self-concept and self-efficacy in academic motivation research. Educational Psychologist 34(3). 139–153. Doi:10.1207/s15326985ep3403_1.Search in Google Scholar
Boo, Zann, Zoltán Dörnyei & Stephen Ryan. 2015. L2 motivation research 2005–2014: Understanding a publication surge and a changing landscape. System 55. 145–157. Doi:10.1016/j.system.2015.10.006.Search in Google Scholar
Csizér, Kata & Zoltán Dörnyei. 2005. The internal structure of language learning motivation and its relationship with language choice and learning effort. The Modern Language Journal 89(1). 19–34. Doi:10.1111/j.0026-7902.2005.00263.x.Search in Google Scholar
Csizér, Kata & Judit Kormos. 2009. Learning experiences, selves and motivated learning behavior: A comparative analysis of structural models for Hungarian secondary and university learners of English. In Zoltán Dörnyei & Ema Ushioda (eds.), Motivation, language identity and the L2 Self, 98–119. Clevedon: Multilingual Matters.10.21832/9781847691293-006Search in Google Scholar
Dörnyei, Zoltán. 2005. The psychology of the language learner: Individual differences in second language acquisition. Mahwah, NJ: Lawrence Erlbaum.Search in Google Scholar
Dörnyei, Zoltán. 2009. The L2 motivational self system. In Zoltán Dörnyei & Ema Ushioda (eds.), Motivation, language identity and the L2 Self, 9–42. Clevedon: Multilingual Matters.10.21832/9781847691293-003Search in Google Scholar
Dörnyei, Zoltán. 2010b. Questionnaires in second language research: Construction, administration, and processing, 2nd edn. With contributions from Tatsuya Taguchi. New York & London: Routledge.10.4324/9780203864739Search in Google Scholar
Dörnyei, Zoltán. 2010a. Researching motivation: From integrativeness to the Ideal L2 Self. In Susan Hunston & David Oakey (eds.), Introducing applied linguistics: Concepts and skills, 74–83. London: Routledge.Search in Google Scholar
Dörnyei, Zoltán, Christine Muir & Zana Ibrahim (2014). Directed Motivational Currents: Energising language learning by creating intense motivational pathways. In David Lasagabaster, Aintzane Doiz & Juan Manuel Sierra (eds.), Motivation and foreign language learning: From theory to practice, 9–29. Amsterdam & Philadelphia: John Benjamins.10.1075/lllt.40.01dorSearch in Google Scholar
Field, Andy, Jeremy Miles & Zoe Field. 2012. Discovering statistics using R. Los Angeles etc.: Sage.Search in Google Scholar
Gardner, Robert C. 1985. Social psychology and second language learning: The role of attitudes and motivation. London: Edward Arnold.Search in Google Scholar
Gardner, Robert C. 2001. Integrative motivation and second language acquisition. In Zoltán Dörnyei & Richard Schmidt (eds.), Motivation and second language acquisition, 1–19. Honolulu, HI: University of Hawai’i Press.Search in Google Scholar
Gobel, Peter, Siew Ming Thang & Setsuko Mori. 2016. Revisiting studies on causal attribution in ESL/EFL contexts: Toward an alternative model. In Matthew T. Apple, Dexter Da Silva & Terry Fellner (eds.), L2 selves and motivations in Asian contexts, 11–28. Clevedon: Multilingual Matters.10.21832/9781783096756-003Search in Google Scholar
Henry, Alastair. 2011. Gender differences in L2 motivation: A reassessment. In Samuel A. Davies (ed.), Gender gap: Causes, experiences and effects, 81–101. New York: Nova Science.Search in Google Scholar
Higgins, E. Tory. 1998. Promotion and prevention: Regulatory focus as a motivational principle. Advances in Experimental Social Psychology 30. 1–46. Doi:10.1016/s0065-2601(08)60381-0.Search in Google Scholar
Jonkisz, Ewa, Helfried Moosbrugger & Holger Brandt. 2011. Planung und Entwicklung von Tests und Fragebogen. In Helfried Moosbrugger & Augustin Kelava (eds.), Testtheorie und Fragebogenkonstruktion, 2nd edn., 27–74. Berlin & Heidelberg: Springer.10.1007/978-3-642-20072-4_3Search in Google Scholar
Josse, Julie & François Husson. 2012. Handling missing values in exploratory multivariate data analysis methods. Journal de la Société Française de Statistique 153(2). 79–99.Search in Google Scholar
Kormos, Judit & Kata Csizér. 2008. Age-related differences in motivation of learning English as a foreign language: Attitudes, selves, and motivated behavior. Language Learning 58. 327–355. Doi:10.1111/j.1467-9922.2008.00443.x.Search in Google Scholar
Kormos, Judit, Kata Csizér, Adrienn Menyhárt & Dóra Török. 2008. ‘Great Expectations’: The motivational profile of Hungarian English language students. Arts and Humanities in Higher Education 7(1). 65–82. Doi:10.1177/1474022207084884.Search in Google Scholar
Kormos, Judit & Thom Kiddle. 2013. The role of socio-economic factors in motivation to learn English as a foreign language: The case of Chile. System 41. 399–412. Doi:10.1016/j.system.2013.03.006.Search in Google Scholar
Kormos, Judit, Thom Kiddle & Kata Csizér. 2011. Systems of goals, attitudes, and self-related beliefs in second-language-learning motivation. Applied Linguistics 32(5). 495–516.10.1093/applin/amr019Search in Google Scholar
Liaw, Andy & Matthew Wiener. 2002. Classification and regression by randomForest. R News 2(3). 18–22.Search in Google Scholar
Lohmann, Arne. 2013. Is tree hugging the way to go? Classification trees and random forests in linguistic study. VIEWS: Vienna Working Papers in Linguistics 22. 1–16. http://anglistik.univie.ac.at/views (accessed 3 March 2016).).Search in Google Scholar
Noels, Kimberley A. 2003. Learning Spanish as a second language: Learners’ orientations and perceptions of their teachers’ communication style. Language Learning 53. 97–136. Doi:10.1111/1467-9922.53225.Search in Google Scholar
R Development Core Team. 2015. R: A language and environment for statistical computing. Version 3.2.3. Vienna, Austria: R Foundation for Statistical Computing.Search in Google Scholar
Ryan, Stephen (2008). The Ideal L2 Selves of Japanese learners of English. University of Nottingham dissertation.Search in Google Scholar
Stevens, James P. 2002. Applied multivariate statistics for the social sciences, 4th edn. Hillsdale, NJ: Lawrence Erlbaum.Search in Google Scholar
Strobl, Carolin, Anne-Laure Boulesteix, Thomas Kneib, Thomas Augustin & Achim Zeileis. 2008. Conditional variable importance for random forests. BMC Bioinformatics 9. 307. Doi:10.1186/1471-2105-8-25.Search in Google Scholar
Strobl, Carolin, James Malley & Gerhard Tutz. 2009b. An introduction to recursive partitioning: Rationale, application, and characteristics of classification and regression trees, bagging, and random forests. Psychological Methods 14(4). 323–348. Doi:10.1037/a0016973.Search in Google Scholar
Tagliamonte, Sali A. & R. Harald Baayen. 2012. Models, forests and trees of York English: Was/were variation as a case study for statistical practice. Language Variation and Change 34. 135–178. Doi:10.1017/s0954394512000129.Search in Google Scholar
Taguchi, Tatsuya, Michael Magid & Mostafa Papi. 2009. The L2 motivational self system amongst Chinese, Japanese and Iranian learners of English: A comparative study. In Zoltán Dörnyei & Ema Ushioda (eds.), Motivation, language identity and the L2 Self, 66–97. Clevedon: Multilingual Matters.10.21832/9781847691293-005Search in Google Scholar
Ushioda, Ema. 2001. Language learning at university: Exploring the role of motivational thinking. In Zoltán Dörnyei & Richard Schmidt (eds.), Motivation and second language acquisition, 91–124. Honolulu, HI: University of Hawai’i Press.Search in Google Scholar
Yashima, Tomoko, Lori Zenuk‐Nishide & Kazuaki Shimizu. 2004. The influence of attitudes and affect on willingness to communicate and second language communication. Language Learning 54(1). 119–152. Doi:10.1111/j.1467-9922.2004.00250.x.Search in Google Scholar
You, Chenjing & Zoltán Dörnyei. 2016. Language learning motivation in China: Results of a large-scale stratified survey. Applied Linguistics 37(4). 495–519. Doi:10.1093/applin/amu046.Search in Google Scholar
Zimmerman, Barry J. 1995. Self-efficacy and educational development. In Albert Bandura (ed.), Self-efficacy in changing societies, 202–231. Cambridge: Cambridge University Press.10.1017/CBO9780511527692.009Search in Google Scholar
Zimmerman, Barry J., Albert Bandura & Manuel Martinez-Pons. 1992. Self-motivation for academic attainment: The role of self-efficacy beliefs and personal goal setting. American Educational Research Journal 29(3). 663–676. Doi:10.2307/1163261.Search in Google Scholar
© 2018 Walter de Gruyter GmbH, Berlin/Boston