483 Studies in Second Language Learning and Teaching Department of English Studies, Faculty of Pedagogy and Fine Arts, Adam Mickiewicz University, Kalisz SSLLT 12 (3). 2022. 483-507 http://dx.doi.org/10.14746/ssllt.2022.12.3.7 http://pressto.amu.edu.pl/index.php/ssllt Second language psychological speaking and listening needs: Scale development, symbiosis, and demographic differences Jian Xu Sichuan International Studies University, Chongqing, China https://orcid.org/0000-0002-2275-6197 xujian@link.cuhk.edu.hk Xuyan Qiu The Hong Kong Polytechnic University, Hong Kong SAR, China https://orcid.org/0000-0003-4368-8039 christyxyq@hotmail.com Abstract This study aimed to develop and validate two parallel scales to measure the psy- chological L2 speaking and listening needs of 863 English-as-a-foreign language (EFL) learners. The associations between three psychological needs (i.e., auton- omy, competence, and relatedness) of L2 speaking and of L2 listening were ex- amined to develop insights into oracy (i.e., integration of speaking and listening) in L2 communication. Subsequently, the impact of demographic variables was explored. The data, collected via a 5-point Likert-scale questionnaire, were ana- lyzed through descriptive and correlation analysis, factor analysis, and ANOVA. Exploratory factor analysis was conducted to determine the factor structures, followed by confirmatory factor analysis for validation. Results demonstrated that the validity and reliability of the two developed scales were satisfactory. L2 speaking autonomy was significantly related to L2 listening autonomy, as were competence and relatedness. The three psychological needs of both L2 speak- ing and listening revealed varying patterns in terms of gender, major, university geographical context, schooling stage (first year to fourth year), and study- abroad experiences. The research findings reinforce the need for integration of Jian Xu, Xuyan Qiu 484 L2 speaking and L2 listening when satisfying university students’ psychological needs, contribute to the research field with the measurement scales of psycho- logical needs in L2 speaking and listening settings, and yield implications for teaching the two language skills integratedly. Keywords: L2 speaking; L2 listening; psychological needs; demographic infor- mation; oracy 1. Introduction Oral second/foreign (L2) communication is an overarching term to refer to L2 listening and speaking, which occur simultaneously in the classroom and in daily interactions (Murphy, 1991). The ability to speak English logically and intelligibly on a particular topic is a key learning goal for English-as-a-foreign language (EFL) learners. Listening comprehension, likewise, is a significant language skill which enables access to various aural and visual L2 resources (Vandergrift, 2007) and whose development has been demonstrated to improve L2 speaking (Vander- grift & Goh, 2012). Driven by the concurrent and symbiotic nature of L2 listening and speaking in oral communication, Goh (2014) used the term oracy to capture their combined equivalent importance. However, existing studies that address L2 speaking and listening together are relatively scarce in EFL research (Goh, 2014). Research focusing on learners’ motivational constructs of both L2 speak- ing and listening is even less frequent. According to self-determination theory (SDT), the degree of a learner’s motivation depends on three psychological needs: autonomy, competence, and relatedness (Ryan & Deci, 2017). In our study, autonomy refers to L2 learners’ feelings of volition when they internalize their actions as an articulation of their free will in L2 speaking or listening prac- tice. Competence is defined as a feeling of effectiveness when the learners feel capable of achieving learning goals and are on the way to successfully perform the target activities in an L2 speaking or listening environment. Relatedness is ev- ident when individuals perceive a connection or develop a significant sense of be- longing with others while communicating, such as with peers and teachers. Even if previous studies have shed light on L2 speaking or listening motivation in isola- tion (e.g., Vandergrift, 2005), the three basic psychological needs and the associ- ations between psychological L2 speaking and listening needs are largely under- explored. Moreover, little is known regarding how L2 learners’ psychological speaking or listening needs are shaped by demographic variables. This study, therefore, intends to address these research gaps by developing two measure- ment scales to investigate L2 learners’ psychological speaking and listening needs. Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 485 2. Literature review 2.1. Basic psychological needs and L2 learning The concepts of intrinsic (autonomous) and extrinsic (controlled) motivation are included in self-determination theory (Ryan & Deci, 2017). When learning is im- portant and integral to the learner’s sense of him/herself, intrinsic motivation to- wards the task will be formed. Learners are extrinsically motivated when learning is done for the purpose of rewards such as grades or appraisal given by teachers. SDT includes five motivational constructs: (1) intrinsic orientation (e.g., enjoy- ment), (2) identified regulation (e.g., recognized value of the extrinsic behavior), (3) introjected regulation (e.g., fear of disapproval), (4) external regulation (e.g., social approval), (5) integrated regulation (e.g., conversion into one’s beliefs) and (6) amotivation (Ryan & Deci, 2017). SDT contends that when specifiable psycho- logical and social needs are satisfied in the context of an individual’s development, this will promote growth, integrity, and well-being. This contention has been proven effective in L2 research (cf. Noels et al., 2000). Researchers such as Noels, Lascano, and Saumure (2019) and Alamer and Almulhim (2021) have examined these basic psychological needs of L2 learners from the SDT perspective in an EFL learning context. These available empirical studies that are related to these psy- chological needs focus on three main aspects, and their findings not only reveal positive relationships between need satisfaction and the enhancement of learner motivation and academic achievement, but also point to the importance of need satisfaction in the development of specific language skills. The first strand of research has focused on psychological language needs and their relation to motivational orientations. For example, Chow and Chu (2007) found that relatedness occurred to Chinese students when they were motivated to fulfil their filial obligations for their parents through satisfactory academic achievement. Carreira (2012) reported that intrinsic motivation and introjected and identified regulation were significantly correlated with auton- omy, competence, and relatedness in L2 learning. Oga-Baldwin et al. (2017) also found that autonomous motivation was significantly associated with need sat- isfaction, and this need satisfaction mediated the relationship between support- ive teaching and student engagement in a language class. These findings were all consistent with Ryan and Deci’s (2017) assertion that satisfying the three basic psychological needs should generate a positive derivative through the em- bodiment of autonomous motivation. The second strand of research into psychological needs, as conceptualized by SDT, has attempted to investigate the relationship between L2 psychological needs and English language learning and achievement. For example, Alamer and Jian Xu, Xuyan Qiu 486 Lee (2019) identified a motivational process to explain L2 Saudi students’ achieve- ment in English. With the three basic psychological needs as the starting point, they found progressive influence first on goal-setting orientation, then motiva- tional emotional state, and finally language achievement outcome. Similarly, Joe et al. (2017) found that L2 achievement was significantly predicted by perceived competence in a Korean secondary-school context. It can thus be inferred that the three basic psychological needs in the L2 context are situated within the motiva- tional process and play a crucial role in L2 learning and development. The third strand of research suggests the potential positive impact of need satisfaction on the development of particular language skills. Alamer (2021) has investigated the impact of basic psychological needs on L2 vocabulary learning. He found that the three basic psychological needs in L2 learning were directly linked to vocabulary knowledge among Saudi university students. However, apart from L2 vocabulary learning, few studies have attended to the three basic psychological needs in the learning process of other language skills, particularly L2 speaking and listening. Given the critical role of psychological speaking and listening needs in communication (Carreira, 2012), a domain-specific investiga- tion can help reveal human functioning in a specific setting (Ryan & Deci, 2001) and more specific implications can be yielded for teaching L2 speaking and lis- tening skills, especially when it comes to oracy instruction. To achieve these pur- poses, there is a need of empirical studies to measure the basic psychological needs of L2 speaking and listening so that the association between the two L2 skills can be explored in a more in-depth way. 2.2. L2 speaking and listening The current research into the association between L2 listening and speaking has been theoretically manifested in the current literature. Brown (2004) argued that L2 listening functions as a part of L2 speaking, and thus without successful listening comprehension individuals cannot communicate effectively. More re- cently, Goh (2014), in her review article, recommended the use of oracy instruc- tion, involving both speaking and listening, in L2 pedagogy so that learners can understand, increase, and manage their learning processes in a holistic way. Qiu and Xu (2021) found that L2 speaking and listening motivation were associated with each other, highlighting the close relationship between L2 speaking and listening in the communication practices. However, the available empirical stud- ies that have investigated the relationship between basic psychological needs in L2 listening and speaking were not found. Thus, the present study seeks to first develop scales of psychological L2 speaking and listening needs and then exam- ine how close L2 speaking and listening needs are associated with each other. Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 487 2.3. The importance of demographic factors to L2 learning L2 learning has also been shown to be both directly and indirectly influenced by demographic variables (Fromkin et al., 2007), but the role that these demo- graphic variables play in psychological L2 speaking and listening needs has not yet been fully clarified. Specifically, gender differences have been shown to play a role in various aspects of L2 learning, such as the level of language anxiety in the EFL learning context (Jiang & Dewaele, 2019), and attitudes toward learning French (L2) and English (L1) and French learning orientations (Baker & Maclntyre, 2000). Furthermore, university locations have been found relevant to the re- gional inequality of general education (Xiang et al., 2020); yet few studies have looked into the role of regional differences in L2 settings. It is assumed that L2 learning in universities is a process that progresses dynamically from the first year to the fourth. For example, Sung (2019) found that an L2 learner’s sense of language competence progressed over time during their university study. Em- pirical evidence has further shown that basic psychological needs vary from ma- jor to major; Lau and Gardner (2019) found that students with different majors had varying learning styles and preferences. Finally, study-abroad experience is an important aspect when distinguishing the level of L2 listening motivation (Xu & Qiu, 2020). However, the studies overviewed above have not investigated how students’ psychological L2 speaking and listening needs are satisfied across dif- ferent demographic groups (e.g., male vs. female). Thus, the present study en- deavors to explore variations in gender, university geographical context, univer- sity schooling stages, majors, and study-abroad experience in relation to psy- chological L2 speaking and listening needs. 3. The present study The present study aimed to develop and validate scales for psychological L2 speak- ing and listening needs among Chinese university students, to examine how the L2 speaking and L2 listening needs were associated with each other, and to un- derstand how the needs varied in terms of demographic qualities. The demo- graphic characteristics explored include gender, major (humanities and social sci- ence, science, and engineering), university geographical context (Beijing, coastal provinces, inland provinces of China), university schooling stage (freshman, soph- omore, junior, or senior), and the presence or absence of study abroad experi- ences. The following three research questions guided the present study. 1. What is the factor structure model of the psychological needs scales on L2 speaking and listening among L2 Chinese students? (RQ1) Jian Xu, Xuyan Qiu 488 2. How are psychological L2 speaking and listening needs correlated with each other? (RQ2) 3. How do psychological L2 speaking and listening needs vary in terms of gender, major, university geographical context, and schooling stage at university, and study abroad experiences? (RQ3) 4. Method 4.1. Participants This study was conducted in a Chinese tertiary context where students are EFL learners taught in compulsory English courses. In total, 863 Chinese university stu- dents, including 274 male and 589 female, from 16 to 22 years old (M = 18.67, SD = 1.26) were recruited. All the participants had been learning English for more than nine years. The surveyed participants represented a spectrum of different majors including language and literature, communication and journalism, geology, engi- neering, computer science, law, management, translation studies, and education. These were categorized into four main groups for analysis: humanities and social science, sciences, engineering, and medicine. Their universities were located in nine Chinese provinces representing major regions (northeast, southeast, northwest, southwest, and the capital city) of China: Beijing, Anhui, Hubei, Jilin, Shandong, Zhejiang, Xinjiang, Shanghai, and Chongqing. These were organized into three sub- categories for analysis: Beijing, coastal provinces, and inland provinces. There were 549 freshmen, 188 sophomores, 87 juniors, and 39 seniors. They came from Tier 1 universities (N = 711), Tier 2 universities (N = 137), and Tier 3 universities (N = 15). The three tiers of universities in China mainland all provided undergraduate courses and bachelor’s degrees. Tier 1 universities are key public universities under the di- rect leadership of the Ministry of Education, whereas Tier 2 universities refer to or- dinary public universities under the leadership of the provincial government and Tier 3 universities are independent and private ones with relatively low admission criteria. Of the 863 participants, 114 had experienced studying abroad, which was defined as any short-term study, tour, or residence abroad for more than one week. 4.2. Scale development The scales were developed through a two-phase process: item-generation, and item piloting. Two recommendations informed the item-generation process. In- itially, we followed Haynes, Richard, and Kubany’s (1995) advice that the content of developed items should originate from multiple sources: 1) prior literature and theoretical frameworks, 2) existing or related scales, and 3) the researchers’ own Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 489 deductive reasoning. Next, we adopted DeVellis’s (2016) recommendation that scale developers develop the items three times and use those items that performed best in the final iteration. Hence, the initial item pool was created based on the basic con- cepts of SDT and psychological needs (Ryan & Deci, 2017) as well as the empirical studies related to psychological needs in L2 contexts (e.g., Alamer, 2021; Carreira, 2012). Additional items were created by modifying existing SDT instruments that were developed to measure the psychological needs in other contexts, such as edu- cation (Deci et al., 2001; Tian et al., 2014). Keeping in mind the requirement that the designed item pool should be broad enough to retain only the best-performing items, the developed English scales were pretested with five students to further identify pos- sible confusions or problems. The English scales were sent to two applied linguistics experts who were non-native speakers of English to check the content, wordings, and readability. Subsequently, the revised questionnaire was piloted with 135 university students and achieved high reliability for the overall scale (Cronbach α = .94) and each of the three dimensions (autonomy, α = .91; competence, α = .96; relatedness α = .90). 4.3. Instruments 4.3.1. Scale of psychological L2 speaking/listening needs In total, 13 items were designed to assess the psychological L2 speaking/listening needs scales: 4 items for autonomy, 5 for competence, and 4 for relatedness. For ex- ample, the samples item for autonomy were “I feel like I can pretty much be myself when speaking in/listening to English,” for the speaking/listening scale. The sample items for competence were “I can communicate the main point(s) of what I want to say/I can understand the gist of what I hear,” for the speaking/listening scale. The sample item for competence was “I experience a warm feeling with the teachers I spend time with,” for the speaking and listening scale. All items were designed using a 5-point Likert scale from (1) strongly disagree to (5) strongly agree. A commonly used rule of thumb for evaluating internal consistency was below: α < 0.5 = unac- ceptable; 0.5 < = α < 0.6 = poor; 0.6 < = α < 0.7 = acceptable; 0.7 < = α < 0.9 = good; α > = 0.9 = excellent (Hair et al., 2010). The overall reliability of the scale of psycholog- ical L2 speaking/listening needs was satisfactory (Cronbach’s α = .91/.91). The relia- bility estimates of autonomy (α = .86/.90), competence (α = .87/.86), and relatedness (α = .96/.92) were all above the cut-off value 0.7, demonstrating good reliability. 4.4. Data collection and analysis The Chinese version of the questionnaires was distributed online. The transla- tion was completed by the first author. The translated questionnaires were then Jian Xu, Xuyan Qiu 490 proofread by a professor in applied linguistics. An English language learner was invited to pick up the language errors. Then, the back translation was conducted by the second author to ensure language accuracy. Accordingly, the first author addressed the issues arising from the translation (e.g., clarity of questionnaire items). Concerning the sample, we first identified our population as Chinese uni- versity students and then considered the demographic variables when listing some potential participating institutions. Next, the purposive sampling ap- proach was used and English course instructors were contacted to request that data be collected from their students. Those course instructors who agreed to help send the electronic questionnaire link to their students during their class periods. The participants’ consent was obtained before data collection. The eth- ical approval was also obtained from the first author’s university. The questionnaire data were initially screened before data analysis. To achieve the first goal of this study, exploratory factor analysis (EFA) and confirm- atory factor analysis (CFA) were conducted. EFA was conducted with SPSS 25.0 in the pilot study (N = 135), and CFA was subsequently performed in the main study (N = 863), using Mplus 8.0 to test both the theoretical model and an ac- curate specification of multiple hierarchies to confirm whether each indicator (item) loaded significantly on the expected factors. Oblimin with Kaiser Normal- ization was the rotation method. The number of factors to be retained was as- sessed by judging whether eigenvalues were greater than 1, and if factor load- ings were greater than 0.3, and by evaluating the scree plot of eigenvalues, and interpretability of factor structure (Hair et al., 2010). Maximum likelihood esti- mation with robust standard errors was employed when performing CFA to as- sess the hypothesized models. When evaluating the appropriateness of model fit, we followed Kline’s (2016) suggestion that we establish whether the speci- fied model fit the empirical data. The size of standardized factor loadings was interpreted according to Cohen’s (1992) guidelines: trivial < 0.1 ≤ small < 0.3 ≤ moderate < 0.5 ≤ large with the cut-off values being 0.1, 0.3, and 0.5, respec- tively. Alternative models were tested to select better model and were also cal- culated to further provide evidence for construct validity. To achieve the second goal, CFA models were computed with the 863 partic- ipants to see how closely the three psychological needs of L2 speaking were associ- ated with those of L2 listening. For the third goal, t-tests and ANOVA tests were performed independently to pinpoint the differences in autonomy, competence, and relatedness among 863 participants. Independent t-tests were conducted to explore the roles of gender differences and study abroad experience, while ANOVA tests were used to investigate regional and disciplinary differences, and also the dif- ferences of school year. When the ANOVA was significant, post hoc tests (Tukey Test) were conducted to see differences between specific groups. Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 491 5. Results 5.1. Psychometric properties of L2 psychological speaking and listening needs scales 5.1.1. Exploratory factor analysis We first performed an EFA with the pilot data (N = 135). The Kaiser-Mayer-Olkin of sampling adequacy was 0.91, which was above the cut-off threshold of 0.5, thus suggesting that EFA was able to yield discrete and reliable factors. Bartlett’s test of sphericity (χ² = 3124.06, df = 325, p < 0.01) indicated that the correlations between items were large enough to perform the EFA. As presented in Table 1, the pilot data generated six factors, as expected, which explained 77.36% of the variance. All of these factor loadings were above 0.3. A scree plot was not included because the factor dimensions in this study were set theoretically. Only LA4 showed cross-load- ings on both speaking and listening autonomy, but its item loading was heavier on its corresponding factor, and thus it was retained. Consequently, all the validated items in the pilot study were subject to CFA in a larger sample. Table 1 EFA results of psychological L2 speaking and listening needs Item 1 2 3 4 5 6 SA1 .56 SA2 .51 SA3 .32 SA4 .37 SC1 .61 SC2 .64 SC3 .40 SC4 .75 SC5 .86 SR1 .79 SR2 .93 SR3 .97 SR4 .98 LA1 .80 LA2 .85 LA3 .74 LA4 .79 LC1 .58 LC2 .81 LC3 .51 LC4 .64 LC5 .81 LR1 .56 LR2 .69 LR3 .56 LR4 .59 Note. SA(1) = speaking autonomy, SC(2) = speaking competence, SR(3) = speaking relatedness, LA(4) = listening autonomy, LC(5) = listening competence, LR(6) = listening relatedness Jian Xu, Xuyan Qiu 492 5.1.2. Confirmatory factor analysis Based on our theoretical underpinnings, the hypothesized first-order correlated factor structure of psychological L2 speaking needs was tested with a larger sam- ple (N = 863). There were three factors: autonomy (4 items), competence (5 items), and relatedness (4 items). The factor loadings can be seen in Figure 1, and all the factor loadings were statistically significant (p < .01). The hypothesized L2 listening needs first-order correlated factor structure results suggested that the size of all standardized factor loadings of this test were also large and that all the factor loadings were statistically significant (p < .01) (see Figure 1). Figure 1 and Figure 2 show the standardized results for the first-order and bifactor correlated model. Table 2 Model fit indices of four models (N = 863) Fit indices x2/df p CFI TLI RESEA SRMR Cutoff criteria ＜5 ＜.00 > .90 >.90 < .08 < .08 Model 1 (first-order-speaking) 6.19 .00 .94 .92 .07 .03 Model 2 (first-order-listening) 8.48 .00 .91 .90 .08 .04 Model 3 (bifactor-speaking) 5.63 .00 .95 .93 .07 .12 Model 4 (bifactor-listening) 6.62 .00 .94 .92 .08 .12 Autonomy A1/AU1 e1.38/.37 .79/.79 A2/AU2 e2.31/.31.83/.83 A3/AU3 e3.23/.23 .88/.88 A4/AU4 e4.26/.26 .86/.86 Competence C1/CO1 e1.31/.27 .83/.85 C2/CO2 e2.53/.57 .69/.66 C3/CO3 e3.25/.26.86/.86 C4/CO4 e4.22/.24 .89/.87 Relatedness R1/RE1 e1.35/.42 .81/.76 R2/RE2 e2.30/.38.84/.79 R3/RE3 e3.43/.21 .76/.89 R4/RE4 e4.21/.30 .89/.84 C5/CO5 e5.21/.37 .89/.79 Figure 1 CFA of first-order correlated factor structure of psychological L2 speak- ing/listening needs (Model 1/Model 2) Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 493 To ensure that the above hypothesized model was the one with the closest model fit to the data, we tested the bifactor models of both of the developed scales. As shown in Model 3 (Figure 2), the size of factor loadings on the right side was large and all the factor loadings were statistically significant (p < .01). In Model 4, also shown in Figure 2, the size of factor loadings on the right side was large as well and all the factor loadings were statistically significant (p < .01). Models 1 and 3 were equivalent, as were Models 2 and 4. The CFA model fit indices of the four models are summarized in Table 2, and all the results show that the data fit the model well. The value of x2/df was greater than 5 possibly because of the large sample size, which was a limitation. Models 3 and 4 had better model fit than Model 1 and Model 2 because of the lower value of x2/df and higher value of CFI and TLI, so we determined that the bifactor CFA models of psychological L2 speak- ing and listening needs better matched our theoretical underpinnings. Autonomy A1/AU1 .81/.82 .81/.82 A2/AU2 .85/.84 .05/.05 A3/AU3 .89/.88 .04/.04 A4/AU4 .86/.86 .09/.08 Competence C1/CO1 .89/.89.88/.88 C2/CO2 .73/.72.10/.10 C3/CO3 .89/.90.07/.06 C4/CO4 .90/.90 .06/.01 Relatedness R1/RE1 .78/.77 .91/.98 R2/RE2 .83/.80 .11/.20 R3/RE3 .76/.90 .10/.01 R4/RE4 .89/.86 .04/.03 C5/CO5 .90/.82 .01/.05 Psychological speaking/listening needs Figure 2 CFA of bifactor models of psychological L2 speaking/listening needs (Model 3/Model 4) 5.1.3. Descriptive and correlation analysis After presenting the results of EFA and CFA, results of descriptive and correlation analysis were presented in this larger sample (N = 863). Mean, standard deviations, Jian Xu, Xuyan Qiu 494 skewness, kurtosis, and item-total correlation for the two scales on the item level were reported. Drawing on the Oxford’s (1990) classification of the level of self-reported questionnaire responses, a mean in the range of 3.5-5.0 was cate- gorized as a high level, 2.5-3.4 medium level and 1.0-2.4 low level. The descrip- tive analysis of psychological L2 speaking needs at the item level showed means for each of the 13 items ranging from 2.66 to 4.09, with standard deviations ranging from 0.92 to 1.09 (see Table 3). The value of skewness and kurtosis in the speaking study fell within the cut-off value range of between ± 3.0 and be- tween ± 8.0 respectively, indicating the univariate normality of the responses (Kline, 2016). The item-total correlation ranged from .56 to .73, which indicates that each item was closely correlated with the sum of items. In addition, descriptive analysis of psychological L2 listening needs at item level showed means for each of the 13 items ranging from 2.24 to 4.13, with stand- ard deviations ranging from 0.87 to 1.10 (see Table 4). Similarly, the value of skew- ness and kurtosis for the listening study was less than the cut-off value of ± 3.0 and ± 8.0, signifying the univariate normality of the responses. Violations of normality, multicollinearity and outliers were not detected in either of the psychological L2 speaking or listening needs scales. The item-total correlation ranged from .46 to .74, which shows that each item was well associated with the sum of items. Table 3 Descriptive analysis of psychological L2 speaking needs at the item level (N = 863) When speaking in English, I feel that… M SD Skewness Kurtosis Autonomy (4 items) 1. I have a sense of choice about how to speak in English (e.g., the choice of language expression, content, speaking strategies). 3.52 1.05 -.17 -.60 2. I feel free to express my thoughts about the speaking topics. 3.48 1.07 -.11 -.69 3. I feel like I can pretty much be myself when speaking in English. 3.58 1.09 -.21 -.74 4. My decisions during the English-speaking process (e.g., decisions on lan- guage expression, content, speaking strategies) stem from what I person- ally want to do. 3.65 1.04 -.28 -.58 Competence (5 items) 1. I can communicate the main point(s) of what I want to say. 3.46 .92 -.06 -.33 2. I can solve communication problems when I do not know how to say something. 3.18 .96 .06 -.23 3. I can accomplish a task in real life (e.g., asking the price of an item in a store). 3.57 .99 -.28 -.40 4. I can produce fluent English speech. 2.93 1.02 .22 -.21 5. I can produce grammatically accurate English speech. 2.66 .99 .36 -.08 Relatedness (4 items) 1. I experience a warm feeling with the teachers I spend time with. 3.96 .95 -.51 -.52 2. The teachers are generally friendly towards me. 4.09 .92 -.67 -.22 3. I get along with my teachers. 4.03 .93 -.55 -.44 4. I get along with my peers. 3.98 .96 -.59 -.24 Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 495 Table 4 Descriptive analysis of psychological L2 listening needs at item level (N = 863) When listening to English, I feel that… M SD Skewness Kurtosis Autonomy (4 items) 1. I have a sense of choice about how to listen to English texts (e.g., listen- ing strategies). 3.77 1.02 -.48 -.30 2. I feel free to select the listening topics or text types that I am interested in. 3.70 1.08 -.43 -.57 3. I feel like I can pretty much be myself when listening to English. 3.62 1.10 -.33 -.68 4. My decisions during the English listening process (e.g., decisions on lis- tening strategies) stem from what I personally want to do. 3.66 1.05 -.31 -.61 Competence (5 items) 1. I can understand the gist of what I hear. 3.39 .90 -.34 .08 2. I can figure out the meanings of words or phrases I do not know. 2.63 .91 .30 -.03 3. I can use the information heard in English to accomplish a task in real life (e.g., follow directions to an unknown address). 2.92 .95 .03 -.33 4. I can understand speech of different English dialects (e.g., Japanese Eng- lish, Indian English, etc.). 2.24 .87 .30 -.43 5. I can understand speakers’ underlying meaning. 2.38 .90 .28 -.14 Relatedness (4 items) 1. I experience a warm feeling with the teachers I spend time with. 3.83 1.01 -.41 -.60 2. The teachers are generally friendly towards me. 4.13 .91 -.80 .11 3. I get along with my teachers. 4.06 .91 -.61 -.36 4. I get along with my peers. 4.05 .90 -.58 -.31 As seen in Table 5, the mean of relatedness of L2 speaking at factor level was above 4 (4.01), showing a high level. The mean of competence was just above 3 (3.16), demonstrating a medium level, and the mean of autonomy was between 3 and 4 (3.56), suggesting a high level. Likewise, the mean of relatedness of L2 listening at factor level was high (4.02), the mean of competence was medium (2.71), and the mean of autonomy was high (3.69). Results of correlation analysis in Table 5 showed that the three psychological L2 speaking needs correlated significantly with each other at p < .001 level. According to Cohen’s (1988) criteria, competence and auton- omy were related strongly (r = .61) as were autonomy and relatedness (r = .66), while relatedness and competence were associated moderately (r = .41). Results of corre- lation analysis in Table 5 showed that the three psychological L2 listening needs cor- related with each other significantly at p < .001 level. Autonomy and relatedness cor- related strongly (r = .70). The associations between autonomy and competence (r = .37), and between competence and relatedness (r = .29) were moderate. Table 5 Descriptive analysis, correlation analysis, and reliability estimate of L2 speaking/listening needs at factor level (N = 863) M SD 1 2 3 α 1 Autonomy 3.56/3.69 .93/.93 1 .86/.90 2 Competence 3.16/2.71 .79/.72 .61**/.37** 1 .87/.86 3 Relatedness 4.01/4.02 .89/.83 .66**/.70** .41**/.29** 1 .96/.92 Note. **p < .001 Jian Xu, Xuyan Qiu 496 5.2. Associations between psychological L2 speaking and listening needs To pinpoint the associations between L2 speaking and L2 listening, we first calcu- lated a measurement model which entailed both their psychological needs (two second-order models; see Figure 3). This yielded a satisfactory model fit (x²/df = 4.95; p < .001; CFI = .91; TLI = .90; RMSEA = .07, SRMR = .05). In the interest of conciseness, only the latent variables are displayed in Figure 3, which shows that the two con- structs were highly associated (β = .81). Then, the association between L2 speaking and listening autonomy, between L2 speaking and listening competence, and be- tween L2 speaking and listening relatedness were calculated in another first-order CFA model with six latent variables, which also generated a good model fit (x²/df = 4.89; p < .001; CFI = .91; TLI = .90; RMSEA = .07, SRMR = .05). Considering publication word and space limits, this model is not illustrated here, but the correlation coeffi- cients obtained from this CFA model are presented in Table 6. The correlation coef- ficients between psychological L2 speaking and listening needs were all above .05, suggesting a large association in L2 oral communication (see Table 6). Autonomy Competence Relatedness Psychological speaking needs Psychological listening needs Autonomy Competence Relatedness .97 .97 .98 0.81 .76 .98 .97 Figure 3 The measurement model of psychological L2 listening and speaking needs Table 6 The correlation between psychological L2 speaking and listening needs L2 speaking-listening needs Correlation coefficient via CFA L2 speaking and listening autonomy .99** L2 speaking and listening competence .73** L2 speaking and listening relatedness .85** Note. **p < .001 Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 497 5.3. Demographic variations This section displays the demographic variations across gender, university geo- graphical context, school years, disciplinary differences, and study abroad expe- rience. Regarding English speaking (Table 7), results of independent t-tests showed that female participants manifested significantly higher degrees of au- tonomy and relatedness than their male counterparts whereas no significant differences were found in competence. Regarding English listening, female learners were more confident about their competence than males, but the de- grees of autonomy and relatedness between the two groups were similar. Table 7 Gender differences (speaking/listening) Gender N Autonomy Competence Relatedness M (SD) M (SD) M (SD) Female 589 3.60 (0.90)/3.72 (0.90) 3.19 (0.76)/2.77 (0.67) 4.09 (0.83)/4.05 (0.79) Male 274 3.46 (1.00)/3.61 (0.98) 3.11 (0.86)/2.59 (0.82) 3.86 (0.97)/3.95 (0.91) t 2.174/1.604 1.394/3.315 3.537/1.705 p .030/.109 .164/.001 <.001/.089 Cohen’s d .15/- -/.24 .25/- Furthermore, ANOVA test results revealed significant differences among EFL learners from the three university geographical contexts of China for all the psychological speaking and listening needs (Table 8). Specifically, according to the post hoc (Tukey test) results in Table 9, students from Beijing and coastal provinces obtained significantly higher degrees of autonomy and relatedness than those from inland provinces in both speaking and listening. Also, Beijing participants were more confident about their speaking and listening than their inland peers, and the mean for listening competence for those from coastal provinces was significantly higher than those from inland provinces. Table 8 University geographical context (speaking/listening) Area N Autonomy Competence Relatedness M (SD) M (SD) M (SD) Beijing 90 3.80 (1.00)/3.93 (0.90) 3.47 (0.81)/3.03 (0.79) 4.23 (0.77)/4.20 (0.75) Coastal 149 3.73 (0.80)/3.89 (0.83) 3.26 (0.72)/2.96 (0.67) 4.22 (0.75)/4.17 (0.73) Inland 624 3.48 (0.94)/3.60 (0.94) 3.09 (0.80)/2.60 (0.70) 3.93 (0.92)/3.95 (0.86) F 7.834/9.143 10.509/26.089 9.782/6.620 p <.001/<.001 <.001/<.001 <.001/<.001 Jian Xu, Xuyan Qiu 498 Table 9 Tukey post hoc results for university geographical context Speaking Listening Autonomy Competence Relatedness Autonomy Competence Relatedness Beijing vs Coastal p = .838 p =.108 p = .997 p = .938 p = .753 p = .970 Beijing vs Inland p = .006 d = .33 p < .001 d = .42 p = .007 d = .35 p = .005 d = .36 p < .001 d = .58 p = .024 d = .31 Coastal vs Inland p = .009 d = .29 p = .054 p = .001 d = .35 p = .002 d = .33 p < .001 d = .53 p = .011 d = .28 When comparing the differences among students in different school years in the program, we again observed significant differences in competence and relatedness for speaking and competence for listening (Table 10). Unsurprisingly, Year 4 students were more confident about their speaking and listening skills than Year 1 students (Table 11), and Year 3 students felt more related to teachers and peers when learning English speaking than their Year 2 counterparts. Table 10 School years (speaking/listening) Year N Autonomy Competence Relatedness M (SD) M (SD) M (SD) Year 1 549 3.59 (0.96)/3.72 (0.96) 3.12 (0.82)/2.67 (0.71) 4.05 (0.90)/4.04 (0.85) Year 2 188 3.45 (0.90)/3.59 (0.89) 3.18 (0.78)/2.71 (0.75) 3.85 (0.87)/3.90 (0.82) Year 3 87 3.66 (0.80)/3.79 (0.76) 3.21 (0.69)/2.86 (0.75) 4.17 (0.77)/4.12 (0.75) Year 4 39 3.37 (0.92)/3.41 (0.99) 3.54 (0.63)/3.02 (0.67) 4.00 (0.87)/4.01 (0.78) F 1.986/2.382 3.756/4.325 3.332/1.848 p .114/.068 .011/.005 .019/.137 Table 11 Tukey post hoc results for school years Speaking Listening Autonomy Competence Relatedness Autonomy Competence Relatedness Year 1 vs Year 2 p = .327 p = .832 p = .076 p = .448 p = .910 p = .271 Year 1 vs Year 3 p = .951 p = .766 p = .678 p = .943 p = .151 p = .855 Year 1 vs Year 4 p = .567 p = .015 d = .57 p = .991 p = .253 p = .032 d = .51 p = .996 Year 2 vs Year 3 p = .391 p = .990 p = .049 d = .39 p = .456 p = .480 p = .232 Year 2 vs Year 4 p = .977 p = .079 p = .820 p = .740 p = .113 p = .913 Year 3 vs Year 4 p = .478 p = .203 p = .794 p = .218 p = .710 p = .911 Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 499 Disciplinary differences were also observed among all the speaking and listening needs (Table 12). In the case of psychological needs related to speaking, humanities and social science (HSS) students demonstrated higher degrees of autonomy, competence, and relatedness than their peers in science and engi- neering (Table 13), and the same trend was also observed in the listening needs between humanities and social science students and those of science and engi- neering students (Table 13). Table 12 Disciplinary differences (speaking/listening) Discipline N Autonomy Competence Relatedness M (SD) M (SD) M (SD) HSS 444 3.72 (0.89)/3.85 (0.89) 3.30 (0.74)/2.84 (0.71) 4.17 (0.80)/4.15 (0.75) Science 266 3.36 (0.95)/3.47 (0.93) 2.99 (0.79)/2.62 (0.67) 3.88 (0.93)/3.88 (0.89) Engineering 142 3.40 (0.96)/3.58 (0.97) 3.05 (0.91)/2.47 (0.78) 3.79 (0.97)/3.85 (0.90) Medicine 11 3.59 (0.71)/3.73 (0.81) 3.22 (0.54)/2.87 (0.63) 3.75 (0.82)/4.07 (0.87) F 10.018/10.151 9.422/12.099 10.530/7.996 p < .001/< .001 < .001/< .001 < .001/< .001 Table 13 Tukey post hoc results of disciplinary differences Psychological L2 speaking needs Psychological L2 listening needs Autonomy Competence Relatedness Autonomy Competence Relatedness HSS vs Sci. p < .001 d = .39 p < .001 d = .41 p < .001 d = .33 p < .001 d = .42 p = .001 d = .32 p = .001 d = .33 HSS vs Eng. p = .005 d = .35 p = .017 d = .30 p < .001 d = .43 p = .027 d = .29 p < .001 d = .50 p = .003 d = .36 HSS vs Med. p = .975 p = .991 p = .468 p = .980 p = .999 p = .992 Sci. vs Eng. p = .986 p = .902 p = .822 p = .730 p = .254 p = .988 Sci. vs Med. p = .886 p = .829 p = .972 p = .843 p = .715 p = .910 Eng. vs Med. p = .932 p = .929 p = .999 p = .966 p = .348 p = .872 Note. HSS = Humanities and social science; Sci. = Science; Eng. = Engineering; Med. = Medicine Study-abroad experience was also an influential factor, as those with over- seas learning experience significantly outperformed those without in all the speaking and listening needs (Table 14). This indicates that study-abroad expe- rience might also help to satisfy EFL learners’ psychological needs. Table 14 Study abroad experience (speaking/listening) Study abroad experience N Autonomy Competence Relatedness M (SD) M (SD) M (SD) +experience 115 3.98 (0.84)/4.03 (0.84) 3.58 (0.77)/3.17 (0.71) 4.27 (0.77)/4.23 (0.76) -experience 748 3.49 (0.93)/3.63 (0.93) 3.10 (0.78)/2.64 (0.70) 3.98 (0.90)/3.98 (0.84) t 5.307/4.314 6.192/7.547 3.362/3.046 p <.001/<.001 <.001/<.001 .001/.002 Cohen’s d .55/.45 .62/.75 .35/.31 Jian Xu, Xuyan Qiu 500 6. Discussion This study developed the scales of psychological L2 speaking and listening needs, explored their associations, and examined how they were influenced by demo- graphic factors. Findings of this study are discussed in accordance with three research questions. 6.1. Development and validation of the scales Responding to RQ1, our findings show that the parallel structured scales have been validated and that the three psychological needs correlated significantly within the speaking and the listening scales respectively (Van den Broeck et al., 2016). This further supports Ryan and Deci’s (2017) SDT and their claim that the three needs are antecedents of human motivation. This study also theoretically contributes to the understanding that basic psychological needs can be satisfied in L2 speaking and listening contexts. When these psychological needs are sat- isfied, the self-determined form of motivation can be enhanced, which, in turn, may generate desired or positive learning outcomes (Noels et al., 2019). Given the positive correlations among the needs within each scale, we can infer that enhancing the satisfaction of one need (e.g., autonomy) can also contribute to the fulfilment of other needs (e.g., competence, relatedness), which, in turn, motivate and promote L2 learning. In addition, from our results, the means of both L2 speaking and listening relatedness were above 4, suggesting that while communicating with others, Chinese university students perceive a higher con- nection or a stronger sense of belonging with peers and teachers. This can be explained by the fact that in the Chinese collectivist culture, students who have satisfying relationships with their parents are more likely to have a higher moti- vation to achieve better learning outcomes (Chow & Chu, 2007), and this satis- fying relationship may also be extended to their teachers and peers. However, Chinese university students proved to have a lower level of perceived compe- tence in L2 listening. The reason might be that Chinese university students do not have many opportunities outside the classrooms to engage in authentic and real-time communications to be competent listeners (Xu & Qiu, 2020). 6.2. Correlations between psychological speaking and listening needs When it comes to RQ2, the results further show that psychological L2 learners’ speaking and listening needs were correlated, indicating that satisfying learners’ psychological L2 speaking needs could lead to higher satisfaction of their L2 lis- tening needs and vice versa. From the learner motivation perspective, this finding Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 501 echoes existing literature that has suggested listening and speaking skills to be interrelated, integrated, and inseparable (Brown, 2004; Goh, 2014). A possible explanation for this correlation is that the actual communication process is a combination of listening and speaking in which one needs to receive, under- stand, and decode the interlocutor’s information and then generate message content, formulate and articulate a response (Vandergrift & Goh, 2012). This study reveals a symbiosis between L2 speaking and listening autonomy, compe- tence, and relatedness. Students’ psychological listening needs are likely to be satisfied when they manifest high levels of speaking autonomy, competence, and relatedness. In this respect, this study demonstrates the close connection between L2 speaking and listening, further supporting the need for oracy in- struction (Goh, 2014). Moreover, our study also illustrates the importance of social environment (i.e., L2 speaking and listening context) with respect to needs satisfaction (Ryan & Deci, 2017). This means that the needs satisfaction does not occur alone but simultaneously happen in L2 listening and speaking contexts. In addition, our findings support the essentiality of integrating listening and speak- ing into the concept of oracy to some extent. This is partially consistent with Qiu and Xu’s (2021) findings that L2 speaking and listening motivation were inte- grated and L2 speaking and listening should be taught and assessed together. 6.3. Demographic information and need satisfaction Moving on to RQ3, the results also revealed gender and disciplinary differences, and that those with varying university contexts, stages of schooling, and with and with- out study-abroad experience also reported different degrees of need satisfaction. In this study, female participants’ satisfaction with L2 speaking autonomy and relat- edness and L2 listening competence was significantly higher than those of males, whereas their degrees of L2 speaking competence and L2 listening autonomy and relatedness were similar. One explanation for this finding could be that females may be more willing to communicate in an L2 than males (Donovan & MacIntyre, 2004), and being willing to communicate has been positively correlated with self-deter- mined motivation and basic psychological needs (Joe et al., 2017). Therefore, fe- male students might obtain higher satisfaction of basic psychological needs in these linguistic areas. Also, in exam oriented L2 classrooms, such as China, where both male and female learners may have limited experience in interactive tasks and the lack of interaction causes both groups of learners to engage less inside and outside the classrooms (Xu & Qiu, 2020). Thus, in this learning environment, their sense of autonomy and relatedness may not obviously differ as shown in our study. The university geographic context was also linked to L2 learners’ satisfac- tion of psychological needs. In China, coastal provinces (e.g., Shanghai, Zhejiang), Jian Xu, Xuyan Qiu 502 in general, are more developed than inland provinces (e.g., Hubei, Xinjiang) which leads to disparities in higher education (Xiang et al., 2020). Beijing was listed as an independent category because of its role as a capital province and its economic development as an inland urban province. In our findings, the means for L2 speak- ing and listening autonomy and relatedness for the Beijing and coastal groups were significantly higher than those for inland participants, and Beijing students obtained a higher degree of competence for speaking and listening than inland counterparts, implying that the basic psychological needs for L2 learners from more developed areas could be better satisfied than those for inland students. This may be at- tributed to the disparity in educational resources between the more developed and less developed areas reported in the literature (Xiang et al. 2020), as inland learners may have relatively fewer opportunities to communicate in English and develop their speaking and listening skills than their Beijing and coastal province peers, and they might be less confident about their language performance. There is also a trend of academic mobility or “brain drain” from inland to Beijing or coastal uni- versities, leading to an imbalance regarding teaching and research quality (Chen, 2016). Therefore, another explanation is that traditional teacher-centered and exam-oriented L2 classrooms may be more typical in inland provinces. In compar- ison, teachers in more developed areas may adopt more diverse teaching meth- odologies (e.g., learner-centered approaches such as task-based instruction), gen- erating more real-time communication experience for learners and satisfying their psychological needs in a better way. Students’ university schooling stage (Years 1-4) was also related to learn- ers’ satisfaction of needs. As expected, the means in speaking and listening com- petence for senior year students (Year 4) were higher than those of Year 1 stu- dents, which is likely due to their more developed English skills at the later stage of their university study. Similarly, Year 3 learners felt a stronger sense of relat- edness with their teachers and peers than Year 2 students, possibly because they were more familiar with their classmates and teachers after more than two years’ university studies. However, since no other significant differences were found in the schooling stage comparisons, these findings are inconclusive. The fourth factor that turns out to be influential is academic discipline, as the humanities participants were found to possess higher degrees of all the needs than their science and engineering counterparts in both speaking and lis- tening. Shaaban and Ghaith (2000) found that the major did not affect learners’ motivation to learn English as a foreign language, but regarding basic psycho- logical needs, the story could be different. Science and engineering students’ basic psychological needs in L2 speaking and listening may require more atten- tion than those for students of the humanities, social sciences and medicine. A possible reason is that learners from different disciplines may have different Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 503 learning styles and preferences. For example, Lau and Gardner (2019) found that L2 learners from soft science disciplines (e.g., humanities and social science) were more active in the learning process (being autonomous in L2 learning according to Lau, 2017) than their peers in the hard sciences. Their finding may also help explain the current result. English language learning itself belongs to the subjects of the humanities. While learning English, HSS students can find out interest, advantage, and/or learning goal. Moreover, although disciplinary differences have been regarded as crucial for teaching academic English due to different academic conventions in different disciplines, their relationship with L2 motiva- tion is worth exploring further. Finally, the lack of study-abroad experience has been reported to affect L2 performance and L2 listening motivation (Xu & Qiu, 2020), and our study further proposes that study-abroad experience strengthens L2 learners’ sense of auton- omy, competence, and relatedness in speaking and listening. Overseas learning experiences offered the learners real-time communication opportunities to de- velop their knowledge and the strategies of listening and speaking in the target language; hence, they might be more confident about their communication skills and have better developed English speaking and listening learning strate- gies, and greater knowledge of what to include when communicating than those without prior study-abroad experience (Xu & Qiu, 2020). 7. Concluding remarks This study examined basic psychological needs for L2 speaking and listening with two scales. Our findings provide a basis for some pedagogical implications as well as di- rections for future research. The limitations are also acknowledged in this section. Regarding pedagogical implications, teachers may focus on the instruction of integrated language skills and listening-and-speaking tasks, creating interac- tive communication experience (Goh, 2014) to achieve the satisfaction of L2 learners’ psychological speaking and listening needs, as the need satisfactions of the two skills are correlated. Teachers can also use the validated measure- ment scales to investigate the psychological needs of their own students so as to develop more insights into the motivational profiles of their students. More- over, they may consider students’ demographic information when designing tasks (e.g., gender), and try their best to satisfy students’ basic psychological needs in L2 communication. With respect to the methodological implication, our study successfully adopts both first-order and bifactor CFA models to verify the construct validity of basic psychological needs, which sheds light on future valida- tion studies in L2 contexts. In relation to conceptual contribution, our study consoli- dates the concept of oracy by showing the close relationship between psychological Jian Xu, Xuyan Qiu 504 L2 speaking and listening needs, which inspires future researchers to treat and study L2 listening and speaking in an integrated manner. The present study has limitations. Students’ questionnaire input were only self-reported data which might need to be triangulated with other data sources, such as observation, performance and so on. Furthermore, due to practical con- straints, data on students’ learning outcomes were not collected. Consequently, the predictive validity of the psychological L2 speaking and listening needs scale on language outcomes was not addressed. In addition, although the data cover- ing students from different regions of China were collected, the sample size was not very large compared with the student population in China. As to the suggestions for future validation studies, participants from dif- ferent countries or regions can be recruited to address how psychological needs vary across different cultural contexts. In addition, future studies may collect the data of students’ learning outcome in a larger scale to longitudinally examine the changes of psychological L2 speaking and listening needs. In summary, we hope that the findings of our study not only contribute to our understanding of the relationship between L2 speaking and L2 listening from the motivation per- spective but also reinforce the necessity of integrated listening-and-speaking in- struction in second and foreign language classrooms. Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 505 References Alamer, A. (2021). Basic psychological needs, motivational orientations, effort, and vocabulary knowledge: A comprehensive model. Studies in Second Language Acquisition. https://doi.org/10.1017/S027226312100005X Alamer, A., & Almulhim, F. (2021). The interrelation between language anxiety and self-determined motivation: A mixed methods approach. Frontiers in Education, 6, 1-12. https://doi.org/10.3389/feduc.2021.618655 Alamer, A., & Lee, J. (2019). A motivational process model explaining L2 Saudi students’ achievement of English. System, 87, 1-21. https://doi.org/10.10 16/j.system.2019.102133 Baker, S. C., & MacIntyre, P. D. (2000). The role of gender and immersion in com- munication and second language orientations. Language Learning, 50(2), 311-341. https://doi.org/10.1111/0023-8333.00119 Brown, H. D. (2004). Language assessment: Principles and classroom practices. Pearson Education. Carreira, J. M. (2012). Motivational orientations and psychological needs in EFL learning among elementary school students in Japan. System, 40(2), 191- 202. https://doi.org/10.1016/j.system.2012.02.001 Chen, Q. (2016). Globalization and transnational academic mobility. Springer. Chow, S. S. Y., & Chu, M. H. T. (2007). The impact of filial piety and parental in- volvement on academic achievement motivation in Chinese secondary school students. Asian Journal of Counseling, 14, 91-124. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Erlbaum. Cohen, J. (1992). Quantitative methods in psychology: A power primer. Psychological Bulletin, 112(1), 155-159. https://doi.org/10.1037/0033-2909.112.1.155 DeVellis, R. F. (2016). Scale development: Theory and applications (26th ed.). Sage Publications. Deci, E. L., Koestner, R., & Ryan, R. M. (2001). Extrinsic rewards and intrinsic mo- tivation in education: Reconsidered once again. Review of Educational Re- search, 71(1), 1-27. https://doi.org/10.3102/00346543071001001 Donovan, L. A., & MacIntyre, P. D. (2004). Age and sex differences in willingness to communicate, communication apprehension, and self-perceived com- petence. Communication Research Reports, 21(4), 420-427. https://d oi.org/10.1080/08824090409360006 Fromkin, V., Rodman, R., Hyams, N. (2007). An introduction to language (8th ed.). Wadsworth. Goh, C. C. (2014). Reconceptualizing second language oracy instruction: Metacog- nitive engagement and direct teaching in listening and speaking. Asian Jour- nal of English Language and Pedagogy, 2(1), 1-31. Jian Xu, Xuyan Qiu 506 Hair, J. F., Anderson, R. E., Babin, B. J., & Black, W. C. (2010). Multivariate data anal- ysis: A global perspective (Vol. 7). Pearson. Haynes, S. N., Richard, D., & Kubany, E. S. (1995). Content validity in psychological assessment: A functional approach to concepts and methods. Psychological Assessment, 7(3), 238-247. https://doi.org/10.1037/1040-3590.7.3.238 Jiang, Y., & Dewaele, J.-M. (2019). Language anxiety in Chinese dialects and Putong- hua among college students in mainland China: The effects of sociobiograph- ical and linguistic variables. Journal of Multilingual and Multicultural Devel- opment, 40(4), 289-303. https://doi.org/10.1080/01434632.2018.1515213 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 Dif- ferences, 53, 133-144. https://doi.org/10.1016/j.lindif.2016.11.005 Kline, R. B. (2016). Principles and practice of structural equation modeling (6th ed.). The Guillford Press. Lau, K. (2017). ‘The most important thing is to learn the way to learn’: Evaluating the effectiveness of independent learning by perceptual changes. Assess- ment & Evaluation in Higher Education, 42(3), 415-430. https://doi.org/ 10.1080/02602938.2015.1118434 Lau, K., & Gardner, D. (2019). Disciplinary variations in learning styles and prefer- ences: Implications for the provision of academic English. System, 80, 257- 268. https://doi.org/10.1016/j.system.2018.12.010 Murphy, J. M. (1991). Oral communication in TESOL: Integrating speaking, listen- ing, and pronunciation. TESOL Quarterly, 25(1), 51-75. https://doi.org/ 10.2307/3587028 Noels, K., Lascano, D., & Saumure, K. (2019). The development of self-determi- nation across the language course: Trajectories of motivational change and the dynamic interplay of psychological needs, orientations, and en- gagement. Studies in Second Language Acquisition, 41(4), 821-851. https://doi.org/10.1017/ S0272263118000189 Noels, K., Pelletier, L., Clément, R., & Vallerand, R. (2000). Why are you learning a second language? Motivational orientations and self-determination the- ory. Language Learning, 50(1), 57-85. Oga-Baldwin, W., Nakata, Y., Parker, P., & Ryan, R. (2017). Motivating young lan- guage learners: A longitudinal model of self-determined motivation in ele- mentary school foreign language classes. Contemporary Educational Psychol- ogy, 49, 140-150. https://doi.org/10.1016/j.cedpsych.2017.01.010 Oxford, R. (1990). Language learning strategies: What every teacher should know. Newbury House. Second language psychological speaking and listening needs: Scale development, symbiosis, and. . . 507 Qiu, X., & Xu, J. (2021). Defining oracy: Second language listening and speaking mo- tivation in higher education and the role of demographic factors. Psychologi- cal Reports, 1-29. https://doi.org/10.1177/00332941211054775 Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on hedonic and eudaimonic well-being. Annual Review of Psychol- ogy, 52(1), 141-166. https://doi.org/10.1146/annurev.psych.52.1.141 Ryan, R. M., & Deci, E. L. (2017). Self-determination theory: Basic psychological needs in motivation, development, and wellness. Guilford Press. Shaaban, K. A., & Ghaith, G. (2000). Student motivation to learn English as a for- eign language. Foreign Language Annals, 33(6), 632-644. https://doi.org/ 10.1111/j.1944-9720.2000.tb00932.x Sung, C. C. M. (2019). Investments and identities across contexts: A case study of a Hong Kong undergraduate student’s L2 learning experiences. Journal of Language, Identity & Education, 18(3), 190-203. https://doi.org/10.1080/ 15348458.2018.1552149 Tian, L., Han, M., & Huebner, E. S. (2014). Preliminary development of the adoles- cent students’ basic psychological needs at school scale. Journal of Adoles- cence, 37(3), 257-267. https://doi.org/10.1016/j.adolescence.2014.01.005 Van den Broeck, A., Ferris, D. L., Chang, C. H., & Rosen, C. C. (2016). A review of self- determination theory’s basic psychological needs at work. Journal of Man- agement, 42(5), 1195-1229. https://doi.org/10.1177/0149206316632058 Vandergrift, L. (2005). Relationships among motivation orientations, metacog- nitive awareness and proficiency in L2 listening. Applied Linguistics, 26(1), 70-89. https://doi.org/10.1093/applin/amh039 Vandergrift, L. (2007). Recent developments in second and foreign language lis- tening comprehension research. Language Teaching, 40(3), 191-210. https://doi.org/10.1017/S0261444807004338 Vandergrift, L., & Goh, C. C. M. (2012). Teaching and learning second language listening: Metacognition in action. Routledge. Xiang, L., Stillwell, J., Burns, L., & Heppenstall, A. (2020). Measuring and assessing regional education inequalities in China under changing policy regimes. Ap- plied Spatial Analysis and Policy, 13(1), 91-112. https://doi.org/10.1007/ s12061-019-09293-8 Xu, J., & Qiu, X. (2020). Study abroad experiences count: Motivational profile of EFL listeners and its impact on top-down and bottom-up processing. Ap- plied Linguistics Review. https://doi.org/10.1515/applirev-2020-0037