What Do They Actually Need? An Investigation of English Learning Motivation of the Underprivileged Students

Motivation plays an indispensable role in education because it directs students’ behaviour toward goals; enhances cognitive processing which leads to effort and persistence in learning activities; and determines what learning behaviours should be enforced and brings learning outcomes. To the underprivileged students, what is the meaning of learning English? Why do they need to get English Language education? What drives them to learn English? To answer these questions, this paper has established a framework of learning motivation to profile the English learning motivation of the underprivileged students. Based on the proposed framework, a questionnaire has been developed. Over 2800 adolescents responded to the questionnaire. Data analysis included two procedures: Principle Components Analysis and Exploratory Factor Analysis were employed. According to the results found in this study, parental-teacher influence, extrinsic values and self-perception on own abilities are the most significant factors influencing the English learning motivation of the underprivileged.


Introduction
Motivation plays an indispensable role in education because it directs students' behaviour toward goals; enhances cognitive processing which leads to effort and persistence in learning activities; and determines what learning behaviours should be enforced and brings learning outcomes. To the underprivileged students, what is the meaning of learning? Why do they need to receive education? What drives them to learn? This paper aims to investigate the English learning motivation of this group of adolescents and shed light on classroom practices.

Towards a Comprehensive English Learning Motivation Framework
In the past, traditional psychologists were more concerned with what motivation is than with how this knowledge could be applied in social settings such as classrooms (e.g., Freud, 1926;Hull 1943). The development of motivation research has changed dramatically. As Eccles, Wigfield and Schiefle (1998) summarised, motivation research has gone from biologically based drive perspective to behavioural-mechanistic perspective and then to cognitive-mediational/constructivist perspective. In the 20th century, the importance of affect and less conscious processes has become the central theme. Researchers started to become interested in the contextual influences on motivation. Several significant motivation theories were proposed (e.g. Ames, 1992;Atkinson and Feather, 1966;Bandura, 1997;Covington, 1992;Eccles and Wigfield, 1995;Gardner and Lambert, 1972;Rotter, 1982;Weiner, 1984Weiner, , 1992. However, there have been no comprehensive framework to investigate learning motivation holistically. Wong (2007Wong ( , 2014 had developed a comprehensive framework for motivation, but it was only restricted to motivation to learn English as a foreign language. Therefore, a framework for investigating students' general learning motivation for a multi-contextual setting is needed. According to Cook & Artino (2016), there are four main types of theories for learning motivation. They are: competence, value, attributions, and social and cognitive process. This categorisation of motivation theories is known to be comprehensive which embraces most learning motivation theories. However, Cook & Artino (2016) did not explicitly identify specific theories under each category. Nor there is past study studying whether these four categories are distinct from one another. To fill this gap, attempts were made to examine these four categories and linked related motivation theories under each category in order to establish a comprehensive learning motivation framework for this study. For competence, it refers to whether the person perceive that he has the ability to perform the task. Theories include expectancy of success (Atkinson, 1974), self-efficacy (Bandura, 1997), learned-helplessness (Seligman, 1975), confidence and self-conception (Maslow, 1943). For value, it refers to the anticipated results of the learning task. Related theories include task value (expectancy-value theory), outcome expectation, self-determination, and intrinsic versus extrinsic motivation (Gardner and Lambert, 1972). For attributions, it refers to learners establish links between an observed event or outcome and the personal factors and underlying cause that led to this outcome. That is, if the learners perceive that the underlying cause is within their control, they will be likely to persist in face of initial failure. Theory related to attribution is attribution theory (Weiner, 1984(Weiner, , 1992. For social and cognitive process, it refers to interactions between an individual and a larger social context and it assumes this process is not observable. Related theory is Bandura's (1992) social-cognitive theory. However, this category can be confusing as it may overlap with the above three categories. Social cognitive process may include (a) understanding others, (b) understanding oneself, (c) controlling oneself, and (d) the processes that occur at the interface of self and others (Leiberman, 2007). For example, self-efficacy can be affected by factors including personal, behavioural and environmental factors (Bandura, 1997). In order to distinctly identify the core elements influencing learning motivation, this study suggested that social and cognitive process should be clearly defined the core elements of what social factors influence cognitive motivation process. According to research, the most direct and significant social influence' on learning motivation include: peers (Berndt, 1990;Plecha, 2002;Wentzel, 1998), teachers (Davis, 2003;Roorda, Koomen, Spilt, & Oort, 2011), and parents (Eccles & Harold, 1993;Henderson et al., 1994;Gutman & Midgley, 2000;Shumow & Miller, 2001;Wong, 2012). Related theories include social-cognitive theory (Bandura, 1997 (Atkinson, 1974) self-efficacy theory (Bandura, 1997) learned helplessness theory (Abramson, Seligman, & Teasdale, 1978;Seligman, 1975) Value expectancy-value theory (Eccles & Wigfield, 1995) Intrinsic versus extrinsic motivation orientation (Gardner and Lambert, 1972) Self-determination theory (Deci & Ryan, 1985) Attributions Attribution theory (Weiner, 1984(Weiner, , 1992 Locus of control theory (Rotter, 1982) Goal orientation theory (Ames, 1992) Social-cognitive process Social-cognitive theory (Bandura, 1997) • Peers (Berndt, 1990;Plecha, 2002& Wentzel, 1998 • Teachers (Davis, 2003;Roorda, Koomen, Spilt, & Oort, 2011) • Parents (Eccles & Harold, 1993;Henderson et al., 1994;Gutman & Midgley, 2000;Shumow & Miller, 2001, Wong, 2012 Based on the above literature, one can see motivation can be examined and investigated by using different approaches. Behaviourists like Gagne and Driscoll (1988) argued that environmental factors and situational variables affect human's motivation and its behaviour. Learning can be studied by investigating human behaviour and its consequences in the environment. Students' level of motivation can be measured by the amount of time learners engage in learning.

English Learning Motivation of the Underprivileged Students
The underprivileged in the paper refers to the students with low socio-economic status. Recent research has reaffirmed the relationships between being underprivileged and education, e.g. Faustina (2017) Aikens & Barbarin (2008) also found that schools in poor districts may negatively affecting students' learning progress and attainment. Specifically, the underprivileged students have less likelihood to have experiences which encourage development of fundamental skills like reading acquisition (Buckingham, Wheldall, & Beaman-Wheldall, 2013). In the United States, study found that children from lower socioeconomic status entered high school with average literacy skills five years behind those of average or higher income family students (Bergen, Zuijen, Bishop, & Jong, 2016;Reardon, Valentino, Kalogrides, Shores, & Greenberg, 2013). Students are found to be more present-oriented and found to have lower self-esteem and images which contribute to their low learning motivation for upward social mobility (Colyar & Stich, 2011;Wong, 2007bWong, , 2008aWong, , 2012Wong, , 2014aRamchandani, Sharma, Anekar, & Mishra, 2016) Diemer and Blustein (2007) also found that racial, ethnic, and socioeconomic barriers hindered students' career development.
However, research indicated that school conditions contribute more to the socio-economic differences in learning processes than family income or background do (Aikens & Barbarin, 2008). Researchers have argued that classroom environment plays an important role in outcomes. For instance, students who were randomly assigned to higher quality classroom earned more, were more likely to attend college, and lived in better neighborhoods in the future (Chetty, Friedman, Hilger, Saez, Schanzenbach, & Yagan, 2011). Also, teacher's quality of teacher training are correlated with children's academic achievement (Clotfelter, Ladd, & Vigdo, 2006;Gimbert, Bol, & Wallace, 2007).
Therefore, to help the underprivileged to learn with equal opportunities, despite their socio-economic status, it is necessary to investigate their English learning motivation hence appropriate policy and classroom practices may help to reduce some of these risk factors.

Research Questions
This study aims to answer the following questions: 1) What is the English learning motivation profile of the underprivileged students?
2) What are the most significant factors affecting the English learning motivation of the underprivileged?

Design of Study
In educational research, there are many ways to conduct reliable studies. These include naturalistic and ethnographic research, historical research, longitudinal research, correlational research, action research, ex post facto research, quasi-experiments and single-case research. Data are collected using questionnaires, interviews, accounts, role-playing, observation, tests, and personal constructs. This study will investigate the English learning motivation of the underprivileged students and seek how particular motivation components impact on students' learning motivation. In view of time constraints and the need for a broad-based investigation, it is believed that an adapted questionnaire will collect sufficiently-reliable data, as it reduces bias and is less intrusive -the researcher being unable to influence the responses with verbal or visual clues.

Procedures
A pilot study was carried out before the final questionnaire was set. Students completed the questionnaire and were invited to comment on its language and content. Appropriate changes were made after several items were determined to be vague and other questions found to be statistically unreliable. Following the pilot study, students completed the questionnaire in class while monitored by the researcher. With all students gathered in the school hall, the researcher read the instructions from the questionnaire. Students were then assured that the information they provided would only be used for this study, and were given thirty minutes to complete the questionnaire.

Design of Questionnaire
The questionnaire consisted of 52 closed questions. As no previous study had done so, the researcher developed a questionnaire based on Cook and Artino's (2016) category of motivation theories.
Based on the discussion above, this paper attempted to establish a framework for English learning motivation in order to design a questionnaire to profile the motivation pattern of the underprivileged. In this framework, as suggested by Cook and Artino (2016), there are four main components and each component has its related motivation theories (see Table 2). Four questionnaire items were developed under each motivation theory (except for intrinsic and extrinsic motivation, two questionnaire items were developed respectively).   (Atkinson, 1974) If I try hard enough at school, I will be successful.
self-efficacy theory (Bandura, 1997) I have the ability to do well at school.
learned helplessness theory (Abramson et al., 1978;Seligman, 1975) Even there are difficulties ahead, I have the ability to overcome them.
Value expectancy-value theory (Eccles & Wigfield, 1995) Learning is valuable to me.
Intrinsic versus extrinsic motivation orientation (Gardner and Lambert, 1972) Learning is useful for my career development Self-determination theory (Deci & Ryan, 1985) I will put on appropriate amount of effort when I see the value of the goal.

Attributions
Attribution theory (Weiner, 1984(Weiner, , 1992 I know what caused my academic attainment results. Locus of control theory (Rotter, 1982) Whether to try harder and get good results is under my control.
Goal orientation theory (Ames, 1992) I will have a sense of successfulness if I have good academic results.

Social-cognitive process
Social-cognitive theory (Bandura, 1997) • Peers (Berndt, 1990;Plecha, 2002;Wentzel, 1998) I do not want to lose face in front of my friends, so I try very hard at school.
Consisting of 52-items, the questionnaire was based on a six-point rating scale and elicited responses from the underprivileged students regarding their motivation to learn. 6 indicated statements respondents strongly agreed with; 5 referred to statements respondents agreed with; 4 signified statements respondents tended to agree with; 3 referred to statements respondents tended to disagree with; 2 signified statements respondents disagreed with while 1 indicated respondents strongly disagreed with those statements. Included were items about different motivational constructs under each motivation theory categories, i.e. competence, value, attribution and social-cognitive process. Respondents' demographic information was collected, such as age, gender, school year attending, as well as parental background information on education, family income and family income per capita (total family monthly income / total number of family members). The questionnaire was administered to students in their native Chinese language to avoid any language barrier issues. A reliability test on all these 52 questionnaire items was run to test if there is an internal consistency of all the items set. Reliability coefficient Cronbach alphas) for the motivational components was high, with an alpha value of 0.948 which means the internal consistency of the 52 items set in the questionnaire was high.

Statistical Analysis and Procedures
To address the two research questions, data analysis included two procedures: First was a Principle Components Analysis of the 52 motivational items set in the questionnaire which aimed to screen out any factors with loadings lower than 0.4 within their own factors (Stevens, 1996), thus to perform data reduction. Second, to isolate which components or sub-components were significant to the participants' English learning motivation. Exploratory Factor Analysis (EFA) was employed, mainly due to its ability to elicit clusters of student motivational responses as well as spot inter-relationships within a set of variables. To observe to what degree the phenomenon is present in the population, the effect-size calculation is used. The effect size is used to express the  (Thompson, 2007). The Pearson r value is 0.886 which shows there is a positive relationship between variables.

Participants
This project invited 2 secondary schools to participate. One in Area A (School A) and one in Area B (School B)the poorest districts with the highest numbers of children in poverty in Hong Kong (Society for Community Organisation and its Children's Rights Association, 2017). All students were invited to take part in this project (N:2882) and their participation were on voluntarily basis. There were 6 grades in both schools. Among the valid responses, 1369 are girls and 1515 are boys of whom parents' education background were relatively low. Students were aged from 13-19. According to the collected data, it was found that although participants from these two schools may have family income between US$4501-6000 (17%). However, the monthly income per capita is low, which demonstrated the number of family members per household may be high and the participants were still living under the poverty line. The poverty lines for all household sizes rose in 2016. For single people, it was HK$4,000 (US$512), HK$9,000 for a two-person household and HK$15,000 for three people (South China Morning Post, 2017). Table 3 summarises the details of participants. University or above 0 (0%) 0 (0%) 0 (0%)

Results
After data analysis of the 52-item questionnaire, descriptive statistical analysis showed that social-cognitive process served the main motivation for the underprivileged to learn followed by attributions, values and competence (See Table 4). In order to screen out any motivation factors lower than 0.4 within their own factors (Stevens, 1996), data reduction was performed. According to the analysis, 8 factors were generated: parent-teacher influence, Extrinsic values, Self-perceived ability, Peers, Self-expectations, attributions, intrinsic values, self-determination. Table 5 shows the loadings of each motivation components are greater than 0.3 (see Table 5). Among these factors, parent-teacher influence placed most influence on learning motivation of the underprivileged, followed by extrinsic values, self-perceived ability and peers.

Self-determination
Self-determination theory (Q25) .706 Self-determination theory (Q27) .687 Extrinsic motivation orientation (Q24) .552 Goal orientation theory (Q37) .402 In order to isolate which components were significant to the English learning motivation of the underprivileged and elicit interrelationship within the motivation components, exploratory factor analysis was employed. After data reduction, 62.57% of the total variance was accounted for by three motivation factors: teacher-parent influence, goal orientation and extrinsic values, and self-perception (see Table 6).

Discussion and Implications
According to the results found in this study, parental-teacher influence, extrinsic values and self-perception on own abilities are the most significant factors influencing the English learning motivation of the underprivileged.
Activation in the motivation process refers to the process whereby motivation is prepared or stimulated for a subsequent learning reaction or behaviour. For the underprivileged students, the activation may take high level of effort to stimulate the interests in learning because these students usually posses progressive decline in intellectual functioning, cumulative academic achievement deficits, and strong inclination to premature School termination or higher drop-out rate. Thus, it takes high level of energy to activate this motivation process. Among the motivation factors found in this study, parent-teachers placed the most influential and significant. But with support and encouragement from parents, mentors and teachers, the seemingly unmotivated students may become optimistically resilient because positive social interaction with adults is categorised as one of the key developmental needs of adolescents (Sale, 1991). At home, a sense of respect and trust should be built so that the underprivileged will feel at ease despite the tension they encounter socially and psychologically from external environment away from home. At school, teachers may help students in the school finding common ground despite their socio-economic backgrounds and exercise positive reinforcements in class. Education on equal opportunities and mutual respect is necessary. On the individual basis, one-on-one counsellor will be of great effectiveness in motivating the underprivileged as they are usually deprived of care and attention, individual attention provided by teachers on planning their future and giving them hope will serve as the most effective activating stimulant to learning.
In the process of persistence, according to Meier & Albrecht (2003), the persistence process begins with setting a goal, continues with implementation behaviour oriented towards the set goal, finally evaluative how behaviours should be altered to reach the identified goal. In the case of the underprivileged of this study, students set a purpose for learning under the guidance of the parents and teachers after the activation process, and the goals are extrinsic in nature because education will transform one's socio-economic status. For example, a stable life and career advancement are often significant motivators. When they see a direct pathway between school and a positive future, motivation to keep learning will be facilitated. However, motivation may decline in the process of implementation because of failure and inadequacy, reactivation by teachers and parents is imperative at this pointreadjusting the focus on achieving improvement rather than fearing failure. With the adjustment of focus with teachers and parents, the underprivileged students will evaluate their learning behaviour and attitude.
In the process of intensity, it is where self-perception on own abilities lies and motivation will sustain until the desired goal is achieved. After the process of persistence, self-efficacy, self-determination and perceived abilities are enhanced if the underprivileged students are able to readjust their learning behaviours in order to achieve the learning goals. Learning motivation is intensified. By continuing setting attainable goals under parent-teacher encouragement and help, readjusting learning behaviours towards set goals, the perception on own abilities will become positive and the underprivileged students will have confidence and motivation to continue learning English. Figure 1 demonstrates the process of learning motivation.

Conclusion
To facilitate and enhance the English learning motivation of the underprivileged, this study found teacher-parent influence is the most significant. Setting an extrinsic-based goal for learning will be of effectiveness than cultivating the pleasure of learning for this group of students. Last but not the least, feeling competent and having a sense of achievement will sustain the English learning motivation of the underprivileged.