A Systematic Review on Teacher’s Expectations and Classroom Behaviors

Systematic Review Teacher Expectations Teacher Behaviors Student Achievement The aim of the present study was to scrutinize how teacher expectations are shaped and reflected in teachers’ classroom behaviors by presenting a holistic picture of teacher expectation literature that has significantly developed since 1968. To achieve this, a systematic review design was utilized in the study, and different academic databases, which were namely EBSCOhost, ERIC, Science Direct, Journal Park Academic, and HEC Theses Centre, were examined. Among 1.227 of the studies conducted, 32 research studies were included in the current review based on a set of inclusion and exclusion criteria after the identification, screening, and eligibility processes. After the content analysis carried out on the included studies, the review extracted certain factors shaping teachers’ expectations of students’ academic achievement, which were grouped as students’ readiness, skills and abilities, teacherand family-related factors, and school policies. In classes, teachers differentiated their instructional methods according to students’ ability levels, presented more group work opportunities, established more eye-contact, assigned cognitively harder tasks, and expected more quality work from high-expectancy students. Teachers also tended to decrease their interaction time by turning to another student when a low-expectancy student could not answer a question, and to know personal or academic strengths of high-expectancy students more than low-expectancy ones. Article Info: Received Accepted Published : 11-03-2022 : 06-06-2022 : 18-06-2022


Introduction
Students are directly influenced by their teachers' expectations (Weinstein, 2002), which consist of their beliefs and inferences about students' general behaviors and academic potential, mostly indicating how much they could achieve (Li & Rubie-Davies, 2017). These expectations are of high importance in educational settings to define standards for student evaluation because whereas one teacher's higher expectations bring about fostered learning outcomes in students, one's mediocre expectations may hamper students' further academic progress owing to students' lower academic self-image for themselves (Dusek & Joseph, 1983;Jordan & Stanovich, 2001;Rubie-Davies, 2006). The impacts of teachers' level of expectations on students' learning outcomes have also been tracked as the self-fulfilling prophecy or the Pygmalion effect in the literature put forward by Rosenthal and Jacobson as early as 1968.
In their leading study, Rosenthal and Jacobson (1968) falsely informed teachers about the students' results of an achievement test at the beginning of an academic year and their academic potential to achieve higher. Although the students were not selected based on the achievement test scores as declared, these students showed greater progress in test scores than the others in the same classroom. Teachers' expectations imposed by the researchers at the very beginning have shaped their behavior towards the students in the class, so the students have produced a positive 'self-fulfilling prophecy' effect (Rosenthal & Jacobson, 1968). This appears as an initially wrong assumption which is proven to be accurate following the emergence of new behavior in the environment (Merton, 1968). To be more precise regarding the presence of self-fulfilling prophecy in classrooms, firstly a teacher forms an opinion or expectation of a student's academic capability; then, the teacher demonstrates differential behaviors towards high-and low-expectancy students, which results in a confirmation of what the teacher has expected initially about the student (Gentrup & Rjosk, 2018;Jussim, Robustelli, & Cain, 2009), also named as sustained expectations by Cooper and Tom (1984). Rosenthal and Jacobson's (1968) study stands as a milestone since it reveals the role of affective factors in cognitive skills such as self-esteem, self-efficacy beliefs, motivation, and anxiety (Andres, 2002;McClure, Meyer, Garisch, Fischer, Weir, & Walkey, 2011).
As voiced by Smith (2011), "The human mind does not like to be wrong." (p. 33), and so teachers may form an expectation bias and retain it even if students start to perform differently (Bognar, 1982, as cited in Stegemiller, 1989. Therefore, teachers' classroom practices and behaviors may inevitably be formed by such expectations and beliefs. Rhem (1999) exemplifies teachers demanding simpler answers and tasks from low-achiever students. Cruickshank, Jenkins, and Metcalf (2009) also summarize common features of teachers with low expectations as being ignorant of task orientation, discipline, and students' response time in classes, lacking accurate and timely feedback, and positive reinforcement. On the contrary, when teachers have higher expectations from the students and believe in their academic potential, their instructional practices are moved to a higher level that presents a clear set of learning objectives, well-organized explanations relating to student interests, and more challenging learning opportunities together with remediation strategies -if necessary, more advanced thinking skills, and a more demanding curriculum (Cruickshank et al., 2009;Warren, 2002). Rosenthal (1974) also found out that teachers tend to build a more positive learning environment for high-expectancy students by nodding and smiling at them more. They are also inclined to evaluate these students' work more positively, present more praises and positive reinforcements, and behave in a more encouraging way (Babad, 1992;Madon, Jussim, & Eccles, 1997). Moreover, Rubie-Davies (2007) asserted that teachers with high expectations for their students have a tendency to provide more feedback, ask cognitively demanding questions more, and demonstrate more constructive behavior management techniques in their classes when compared to teachers with low expectations.
The literature is also home to many studies investigating the factors indicating how teacher expectations are shaped. Students' ethnicity (Rampaul, Singh, & Didyk, 1984) and socio-economic status may show parents' incapacities to academically assist their children and provide resources at home (Claassen & Mulders, 2003;De Boer, Bosker, & Van der Werf, 2010;Ditton, Krüsken, & Schauenberg, 2005). Gender of the students may also lead to differential teacher expectations. To illustrate, Timmermans, De Boer, and Van Der Werf (2016) point out that teachers have positively higher expectations for female students who are considered to possess better study skills and more engagement in school work whereas lower expectations of achievement for male students reported less successful in classes. Similarly, regarding teachers' expectations, female students are also favored in Driessen and Van Langen's (2013) study in terms of their cooperative and selfregulated learning skills.
Depending on the variety among these studies on teacher expectations and classroom behaviors, it is believed that a systematic review could prove more generalizable results by providing a fuller and more collective picture of the key findings of the previous studies. The current study also aimed to put the previously conducted studies forward in a more holistic way rather than focusing on their findings individually through a systematic review. For these reasons, the study aspired to illustrate how the teacher expectation literature has developed since 1968 by investigating how teacher expectations are shaped and reflected in their classroom behaviors. In this way, the study may increase the professional awareness of teachers regarding how they form their expectations of academic achievement and how these expectations are transmitted to their classroom behaviors and consequently to students. Thus, teachers might make an effort to regulate their differential behaviors towards high-or low-expectancy students so that they could avoid sustained expectations -if they have any, in the light of what the relevant literature has put forth thus far.
A meta-analysis study was carried out on teacher expectation interventions and their effects on student outcomes by De Boer, Timmermans, and Van Der Werf (2018). The researchers delved into 19 studies meeting their eligibility criteria after a literature search on PsycINFO and ERIC. They found out three kinds of interventions, namely changing teacher behaviour, awareness of expectancy effects, and teacher beliefs concealed in expectations. More importantly, they underlined the possibility of raising teacher expectations (De Boer et al., 2018). Tenenbaum and Ruck (2007) addressed the differentiation of teacher expectations for ethnic minority or European American students through another meta-analysis study. After working on 32 studies, they revealed that teachers tended to hold higher and more positive expectations for European American students than ethnic minority ones. It was also reported that teachers made fewer positive referrals and speech to ethnic minority students (Tenenbaum & Ruck, 2007). The first systematic review on this issue was published in 2018 by Wang, Rubie-Davies, and Meissel (2018) where the quantitative studies reporting only statistically significant findings were essentially included, and the effects of teacher expectations on student achievement were investigated by going back to the last 30 years only . The researchers conducted the study on PsycINFO and ERIC and included 144 articles meeting their inclusion and exclusion criteria. They put forth four main themes regarding influential factors on teacher expectations, transmission ways of teacher expectations, factors moderating teacher expectations effects, and teacher expectation effects on student achievement (Wang et al., 2018). Depending on the fact that these meta-analyses and the systematic review scrutinized merely quantitative studies on similar databases, the current systematic review aspired to specifically address qualitative findings in the literature, which might display more in-depth perspectives of teachers concerning their expectations of students and their classroom behaviors. In addition, unlike the other studies mentioned above, the researchers took 1968 as the starting point of the literature search for this review, when Rosenthal and Jacobson's study was first conducted.

Research Design
The present study employed the systematic review as its research design. The characteristics of this design can be considered as adopting a bias-free approach with the use of a rigorous and methodical way of literature search in a specific field (Hanley & Cutts, 2013). Petticrew and Roberts (2006) focus on the significant aspects of this method, which are to examine, compile and reunite research evidence in a critical way. To achieve this, having a set of eligibility criteria to choose studies from the literature, providing a systematic and reproducible method, and an organized way of synthesizing and presenting the findings of the included studies are musts of this design so as to answer the research questions formulated for the review (Systematic Review Module, 2018).

Research Questions
To achieve the aforementioned purposes, the present study sought answers to the following research questions: 1. What are the potential reasons shaping teacher expectations of students regarding their academic achievement in the research studies selected for the review?
2. How are teacher expectations transmitted, tracked, or reflected in teachers' behaviors in classrooms to students in the studies selected for the review?

Procedures of Literature Search
Since most of them were fed by similar sources, certain electronic databases for educational research were selected for the review, namely EBSCOhost -Academic Research Complete, ERIC, and Science Direct along with Journal Park Academic powered by TUBITAK ULAKBIM and Higher Education Council Theses Centre with the aim of reviewing the studies in both Turkish and international contexts. Another reason behind this selection was also the fact that it provided a better manageability and practicality for the researchers. While searching, various combinations of keywords were employed such as "teacher* expectation* AND student achievement", "teacher* expectation* AND academic achievement", "self-fulfilling prophecy in education", "Pygmalion in the classroom", "teacher* expectation* of students*" since they constitute the major terms in titles, abstracts, or keywords of the studies. For Turkish databases, "öğretmen beklentileri", "öğretmenlerin başarı beklentileri", "Pygmalion etkisi" were utilized. Depending on the options offered by these databases, peer-reviewed, full-text, open-access/archive, academic journals, research/journal articles, and theses and doctoral dissertations were selected interchangeably to narrow down the search to more reliable studies.
The number of initial results declined with the selection of the studies written in either English or Turkish and published after 1968. Within the scope of the current systematic review, the databases were last accessed on the 2nd of January 2020 by the researchers, and the total number of the recognized studies was 1.227. EBSCOhost-Academic Research Complete contributed to this number with 199 studies, ERIC with 535 studies, Science Direct with 489 studies, and HEC Thesis Centre with 4 studies. No studies including the target keywords were found on Journal Park Academic.

Inclusion-Exclusion Criteria
The duplicates were firstly removed from these 1.227 studies by using Mendeley Software (Elsevier, Mendeley Ltd., 2020). The studies without full texts were also excluded from the review. After this identification process, 319 studies were found appropriate for the upcoming screening process. In the next phase, the first criterion was to exclude the studies conducted on teachers' expectations of disabled, gifted, and ethnically diverse students since the current systematic review primarily focused on teachers' expectations of students' academic achievement naturally occur in regular classes because both the field of special education and ethnicity issues require a different academic perspective and expertise due to their complexity stemming from the combination of psychology, anthropology, and sociology disciplines (Good & Nichols, 2001;Odom et al., 2005). For this reason, teachers of such learners are expected to demonstrate different classroom behaviors and play varying roles and responsibilities (Eisenman et al., 2011) shaping their expectations of students in return (Gillung & Rucker, 1977).
Based on the first criterion, the researcher discarded 106 studies. The second criterion was to exclude the studies carried out with pre-service teachers and students based on their expectations of teachers or the teaching profession, teachers' accuracy of judgements, their expectations of themselves, or expectations of the implementations of a new method, or teaching approach because they do not reflect the essential aim of the current review. Herein, 21 studies were eliminated from the review.
In line with the purpose of the study, the researchers excluded 160 quantitative studies. Therefore, in the eligibility process, the researcher included qualitative and mixed design studies only based on the third criterion in line with the aim of the study, which was to display a multifaceted picture of teachers' expectations of student achievement and their classroom behaviors rather than presenting only statistically meaningful findings. Finally, the studies which collected qualitative data directly from in-service teachers were involved in the study as a result of the fourth criterion. All in all, 32 studies were included in the study in the final step. These processes can be examined through a three-step flowchart demonstrated in Figure 1.

Data Extraction and Analysis
By the nature of the systematic research, the researchers first examined the characteristics of the selected studies involving their aims, participants, research contexts where the data were collected, data collection instruments, sampling methods, and data analysis procedures. Apart from this data extraction process, the findings of the studies were analyzed through content analysis with the derivation of codes and themes determined in line with the research questions of the current review. In addition, the researchers checked their code agreements in the analysis but did not calculate the congruence.

Profile of the Selected Studies
Among the studies included in the review (n = 32), 90.63% of them (n = 29) were research articles published in academic journals, 6.25% of them (n = 2) were doctoral dissertations, and 3.13% (n = 1) of them were master's theses. When the school contexts were scrutinized, 20.51% (n = 8) of the studies collected qualitative data from elementary schools, 17.95% (n = 7) of them from secondary schools and primary schools (n = 8), and 12.82% (n = 5) of them collected data from middle schools and high schools (n = 5). Moreover, 5.12% (n = 2) of the studies were conducted at universities while 2.56% (n = 1) were carried out in a kindergarten. Four studies did not specify the school setting at all. Table 1 indicates more detailed descriptive information about the studies. Although the selected studies underlined that they utilized qualitative research paradigm in general, 33.33% of the studies (n = 13) did not clarify any specific qualitative research design, 23.08% of them (n = 9) employed mixed methods without specifying any typology from the relevant literature. Furthermore, 5.12% of them (n = 2) used ethnography, 10.25% of them (n = 4) used case study, and 5.12% of them used (n = 2) phenomenology. While one study utilized grounded theory, another one was an action research study as seen in Figure 2. Despite being limited, typical qualitative sampling procedures were explicitly stated to be used such as purposive maximum variation sampling (Göksoy, 2018), snowball sampling (Amini, 2016), convenience sampling (Niari et al., 2016), and criterion-based purposive sampling (Nutter, 2015).
All the studies in the review employed content analysis and derived codes and themes. While analyzing the data, some studies underlined the use of certain qualitative data analysis software such as N6 ( (Kususanto & Fui, 2012) were adopted as coding strategies. However, such details were rarely given in the included studies of the review.
After the systematic recognition of all these studies, common themes were derived in relation to the research questions. The first main theme was determined as the factors influencing teachers' expectations of academic achievement whereas the second main theme was teachers' reflected classroom behaviors. A table for themes and codes was also presented in Table 2 below. The following sections displayed the findings in a more detailed way with the relevant quotations.

Factors Influencing Teachers' Expectations About Students' Academic Achievement
In line with the first research question, the researchers defined a main theme named the factors influencing teachers' expectations of academic achievement under which there were certain sub-themes as students' readiness, skills, and abilities, teacher-and family-related factors, and school policies.

Students' Readiness, Skills, and Abilities
Within this review, the included studies revealed student-related factors shaping teachers' expectations of student achievement which were students' lack of pre-learning or cognitive readiness to learn new subjects (Göksoy, 2018;Timperley & Phillips, 2003), self-efficacy beliefs (Rubie-Davies et al., 2010), problems with the medium of instruction hindering comprehension and self-expression (Nishino, 2012;Timperley & Phillips, 2003), cognitive capabilities, maturity (Eder, 1981;Göksoy, 2018;Harris, 2012;Nutter, 2015), prior academic achievement (Li & Rubie-Davies, 2018;Niari et al., 2016;Rubie-Davies, et al., 2010), reading and writing skills in exams and expressions used in-class activities (Bergqvist, 2005;Harris, 2012;Li & Rubie-Davies, 2018;Timperley & Phillips, 2003). Students' interactions with their friends, response, and work quality during lessons were also reported as the other factors shaping teacher expectations of students' academic achievement (Rubie-Davies et al., 2010). Some of these findings were evidenced by the quotations below: "I do not believe that all of my students are able to learn all subjects. Because each student is different in terms of perception and learning levels. (T8)" (Göksoy, 2018, p. 210).
"I feel that it's very important not to push these children beyond their learning capabilities or their absorption levels at that time. You can't push these children into a place where they're not ready. Then you see what will happen … they'll flounder. (A teacher)" (Timperley & Phillips, 2003, p. 636). Moreover, students' interest and commitment to the subjects (Deuker, 2014;Fletcher, 2016;Göksoy, 2018;Ng et al., 2015;Niari et al., 2016), motivation in learning and test-taking (Deuker, 2014;Fletcher, 2016;Göksoy, 2018;Harris, 2012;Li & Rubie-Davies, 2018;Ng et al., 2015;Nishino, 2012), study skills and autonomy such as previewing the textbook before classes and doing exercises after classes (Deuker, 2014;Fletcher, 2016;Harris, 2012;Li & Rubie-Davies, 2018), and having academic discipline (Li & Rubie-Davies, 2018) also stood out as the student-related factors affecting teacher expectations as explained by a teacher: "I would be expecting them all to be researching independently. I would be expecting them all to be independently seeking me out for areas and asking questions as opposed to just coming to lessons and doing their homework. (T3)" (Deuker, 2014, p. 72).
On the contrary, students' off-task behaviors in class, insufficient abilities and short attention span to keep up with the classroom routines may lead them to be regarded as low-expectancy students (Eder, 1981;Goldenberg, 1992). Teachers tended to expect more from scienceoriented/stream students than art-oriented/stream ones who were only expected to pass the required exams and have more off-task behaviors in classes while science-oriented/stream ones were expected to achieve much higher scores (Kususanto & Fui, 2012;Yanisko, 2016). It was exemplified by the comments of a student from an art-oriented stream: "My teachers used to prejudge us as being incapable to score high. (S11)" (Kususanto & Fui, 2012, p. 112).
Even in the 21 st century, students' gender was also seen as another factor shaping teachers' expectations of students' academic achievement (Nutter, 2015;Williams, 2012). Sometimes highachiever female students were described as "a girl who does more than is required (Teacher X)" whereas no male students were defined like that (Jones & Myhill, 2004, p. 556). However, sometimes the high-achieving female students were considered "typical" while high-achieving male students were accepted as "atypical" in the eyes of their teachers (Jones & Myhill, 2004). Moreover, Janisch and Johnson (2003) also quoted a teacher who was surprised by male students' positive attitudes towards Shakespeare's work in their study: "I thought they would be frustrated with this topic. … but they became enthusiastic about the reading. It consumed them. When we read certain passages … you could hear a pin drop. They were that interested in finding out the emotions and feeling … even the boys." (p.302).

Teacher-related Factors
Secondly, teacher-related factors were also underlined in the included studies, which are teachers' lack of knowing their students' actual performances and their past teaching and learning experiences since teachers teach as being taught (Li & Rubie-Davies, 2018;Ng et al., 2015;Nishino, 2012). Moreover, teachers' competencies in teaching (Göksoy, 2018;Harris, 2012;Li & Rubie-Davies, 2018;Nishino, 2012) and their self-efficacy beliefs (Harris, 2012;Li & Rubie-Davies, 2018;Nishino, 2012) also came to the surface as the factors derived from the studies selected. To illustrate, there were teachers reporting that although they knew the potential in students, they professionally did not know what to do to reveal it with these words: "… I can see the potential in this kid, but I don't know how to draw it out, to maximize it. (A Math teacher)" (Harris, 2012, p. 138). Another teacher underlined the significance of professional development opportunities in Timperley and Phillips' study (2003)  Teachers' creativity, patience, and beliefs (Göksoy, 2018;Ng et al., 2015;Nishino, 2012) were also declared as the other factors having an impact on teachers' expectations of students' academic achievement since some teachers' notion was that every student could master the standards (Göksoy, 2018;Harris, 2012). However, some teachers expressed that while some of their students could easily contribute to society after being employed, some "will be rotting in jail" (Rubie-Davies, et al., 2010, p. 43). Similar to these arguments, some teachers believed that lowexpectancy students were "are hard to reach, … and only capable of working with a watered-down curriculum" (Calabrese et al., 2007, p. 287).
Furthermore, teachers' interaction with students and colleagues (Rubie-Davies et al., 2010) was also reported to influence their expectations in time. Teachers' expectations were also formed by their own experiences with the students and their colleagues' opinions about the students (Amini, 2016).

Family-related Factors
Family-related factors became also apparent based on the socio-economic status of students' families (Odabasi-Cimer & Cimer, 2010;Göksoy, 2018;Harris, 2012;Nutter, 2015;Robinson,1994;Stevens & Vermeersch, 2010), their cultural or social orientation (Odabasi-Cimer & Cimer, 2010;Harris, 2012), their insufficient interest in their children and their academic work (Calabrese et al., 2007;Timperley & Phillips, 2003), lack of academic resources provided at home (Harris, 2012;Timperley & Phillips, 2003), and students' home life (Calabrese et al., 2007;Göksoy, 2018;Harris, 2012;Wedin, 2010). For instance, one interviewee teacher mentioned these issues by expressing that "… Low socioeconomic students are going to perform lower because they don't have as many books in the home. Students don't know half the vocabulary you are looking at. …" (Harris, 2012, p. 137). Another quotation from Calabrese et al.'s (2007) study evidenced the same issue: "The kids that we usually can't reach or have the most difficult time with are the ones that we get no parental support from. (Mike)" (p. 287) Students from low socio-economic backgrounds were also considered more problematic regarding their attention span and working memory, which made them low-expectancy students for teachers (Nutter, 2015). What is more, students' motivation to learn and encouragement to go to university also built a base for teacher expectations depending on the level of their parents' valuing education (Rubie-Davies, et al., 2010). In relation to this, a teacher quoted in Robinson's (1994) study asked a crucial question "If the parent has little concern for the child, why should I?" (p.518).

School Policies
Finally, teachers mentioned that lack of instructional time (Harris, 2012) and limited instructional resources (Engin, 2018;Göksoy, 2018;Whittle et al., 2018) also had the potential to impact teachers' expectations of student achievement. It was also revealed that certain school administrators set different standards, goals, and programs for art-oriented/stream and scienceoriented/stream students, which was a situation influencing teacher expectations and students' self-esteem in return (Kususanto & Fui, 2012). Some studies specifically touched upon a stream hierarchy and emphasized lower-stream schools (e.g., vocational education schools) as having less hard-working and more problematic students, and hence, negatively affecting teacher expectations of student achievement when compared to higher-stream schools (e.g., technical or general education schools) (Rubie-Davies et al., 2010; Stevens & Vermeersch, 2010) as depicted below: "And I said, whoever told you, you could pick any goal you want really didn't realize that you were going to say you're going to be a brain surgeon, because . . . you can't tie your shoes. I mean, I don't want to rain on their parade, but sometimes it's not too realistic. (Teacher)" (Rubie-Davies et al., 2010, p. 42).
The same streaming factor was also described as an obstacle to the formation of self-esteem in students due to the student stereotypes determined by teachers (Kususanto & Fui, 2012;Rubie-Davies et al., 2010). In some schools, ability grouping in classes also had an impact on teachers' expectations of student achievement (Amini, 2016;Eder, 1981;Goldenberg, 1992;Nutter, 2015;Timmons, 2018;Williams, 2012). In addition, some school administrators wanted teachers to read students' records before they started teaching and there were teachers reporting that reading such kind of documents about students would prevent them to form their own expectations from scratch regarding students' academic achievement (Amini, 2016). In this issue, Amini (2016) quoted a teacher declaring this: "… I was told within the first week, I would have to spend time going through the student's Ontario Student Record and as a teacher… I felt I'm not gonna do that… because I don't want those biases to creep in." (p. 37).

Teachers' Reflected Classroom Behaviors
As an attempt to answer the second research question, the reviewed studies were investigated to reveal the evidence of teachers' classroom behaviors reflected by their high or low expectations, which also became the second main theme of the current systematic review.
Depending on their expectations of student achievement, some teachers let their students know about their expectations directly by clearly stating the expected behavior (Amini, 2016;Engin, 2018;Nutter, 2015), setting the standards for success (e.g., rubrics), and making efforts to enhance student learning outcomes (Amini, 2016;McDonald et al., 2016). Teachers sometimes preferred talking one-on-one to students (Goldenberg, 1992;Timmons, 2018), communicating with parents through private conferences (Goldenberg, 1992), or weekly newsletters sent to homes (Nutter, 2015). As reported, especially one-on-one student-teacher conferences could provide empathy, trust, and understanding regarding teacher expectations on the side of both students and teachers (Nutter, 2015). When teachers had high expectations, they also had a tendency to increase the level and the number of tasks demanding more cognitive efforts from the students for better intellectual gains (Amini, 2016;Nutter, 2015;Rubie-Davies et al., 2010;Wedin, 2010;Yanisko, 2016), employ flexible grouping in their classes where students' own choices mattered and students were guided through goal-setting and self-management strategies (McDonald et al., 2016).
In the studies analyzed, depending on teachers' expectations of student achievement, teachers differentiated their instructional methods, materials, and pacing according to students' individual differences or learning styles (Amini, 2016;Göksoy, 2018;Janisch & Johnson, 2003;Mercuri & Ebe, 2011;Li & Rubie-Davies, 2018;Nishino, 2012;Nutter, 2015;Stevens & Vermeersch, 2010;Williams, 2012) as depicted by the words of a teacher: Therefore, as a consequence of a slackened instructional pacing, teachers complained about falling behind curricular requirements in their classes (Engin, 2018;Stevens, & Vermeersch, 2010). Furthermore, teachers provided scaffolding and structured support when they had low-expectancy students whereas they presented more collaborative and group work opportunities (Janisch & Johnson, 2003;Mercuri & Ebe, 2011;Yanisko, 2016) such as letting students check their answers with friends, edit their peers' papers or discuss the processes of problem-solving when they have high expectations of student achievement. They also assigned harder tasks, put much pressure, and expected more quality work in the presence of high-expectancy students (Jones et al., 2012;Rubie-Davies et al., 2010). More interestingly, it was exemplified in one of the studies that a highexpectancy student might also be ignored and monitored less by his/her teacher due to the beliefs that they would be fine by themselves, which made them a victim of great expectations in the end (Deuker, 2014;Goldenberg, 1992). Another example of the same issue could be found in Rubie-Davies et al.'s study (2010) where there were higher-stream students declaring their teachers expected them to know things in advance and provided less support in the learning process.
Similarly, teachers' time and efforts spent on each student, and classroom management strategies also differed in relation to their expectations of students (Goldenberg, 1992;Robinson, 1994). As depicted in Eder's (1981) findings, teachers had a tendency to ask more questions and employ more management acts while working with low-expectancy students because, in such classes, low-expectancy students violated each other's turns more often. Herein, Robinson (1994) found out that the way of managing off-task behaviors during lessons also changed according to teachers' expectations of student achievement because while teachers warned low-expectancy students in a harsher manner by saying "You wrote it all wrong" or "You don't listen to me" (p. 519), they tended to use more motherly words with a soft tone of voice while warning high-expectancy ones.
Furthermore, teachers' encouragement with a friendly style and interaction with students were also accepted as signs of teacher expectations (Niari et al., 2016). To exemplify, during classroom interaction, teachers might blame or label students as being low-achiever or low-expectancy (Sedova & Salamounova, 2016); and might start to decrease teacher-student interaction time by addressing the low-expectancy ones less (Robinson, 1994), or by turning to another student when a low-expectancy student could not answer a question (Sedova & Salamounova, 2016), and by knowing personal or academic strengths of high-expectancy students more than low-expectancy ones (Sedova & Salamounova, 2016). Additionally, teachers were most of the time aware of when and whom to push students to accomplish better academic performances (Rubie-Davies et al., 2010;Timperley & Phillips, 2003) since as teachers' expectations became higher, they asked critical thinking questions, spent more time on eliciting answers from high-expectancy students (Janisch & Johnson, 2003;Sedova & Salamounova, 2016), and questioned the reasons behind the correct answers more (Yanisko, 2016).
Based on their expectations, teachers arranged their classroom language and used 1st person plural or we language (Niari et al., 2016;Yanisko, 2016), students' first names, superlative adjectives, and possessive pronouns (Niari et al., 2016) more frequently. Moreover, when their expectations were high, some teachers established more eye contact with students and utilized more positive facial expressions and proximity to those students during classes (Niari et al., 2016). They also tended to appreciate students more (Yanisko, 2016) and feel happier when lowexpectancy students achieved higher than what they had expected (Fletcher, 2016;Goldenberg, 1992;Rubies-Davies et al., 2010;Yanisko, 2016). A teacher quoted in Fletcher's (2016) study described her/his feelings after low-expectancy students' performance as: "Students really surprised me and worked well on their writing activity. … Students like [Charlie], who are normally weak in writing skills, did well and never complained about having to write a recount." (p. 411) Another teacher also mentioned their regret regarding their low expectations of students' achievement by saying: "As wrong as it was for me to assume that these sixth graders should be at a lower level in their reading and comprehension, I expected little from them. When I saw that these students were actively engaged in several works by Shakespeare, I went home feeling a little guilty…" (Janisch & Johnson, 2003, p. 306) Additionally, one of the studies asserted that building rapport with students in the class was also required to learn about students' weaknesses and guide them better to take risks without being afraid of making mistakes (Nutter, 2015) in a well-established emotionally-free climate (Whittle et al., 2018;Niari et al., 2016) where student autonomy was also encouraged properly with enough guidance from the teacher (Mercuri & Ebe, 2011). What is more, teachers having high expectations of their students were found more inclined to be more self-aware and reflective in relation to what and how they were teaching in classes (Whittle et al., 2018;Ng et al., 2015), doing wrong or right (Ng et al., 2015), and finding ways to motivate students and draw the potential out of them (Amini, 2016;Harris, 2012). Similarly, they spoon-fed low-expectancy students by simplifying their teaching methods, re-arranging assignments and assessment procedures such as by revising crucial topics before exams, adjusting criteria to judge mastery of the expected standards (Harris, 2012;Nutter, 2015), presenting cognitively less demanding questions in the exams (Odabasi-Cimer & Cimer, 2010), and modifying their expectations so that low-expectancy students could achieve certain standards as well (Harris, 2012;Ng et al., 2015) because teachers did not actually mark students' exams, but their own expectations (Amini, 2016). They also varied exam questions in line with their varying expectations of student achievement as one teacher expressed with these words: "In the exams, I ask one or two questions they can answer easily and questions that can be counted as of medium difficulty, and then there are one or two difficult questions. I want to see how many of them can answer these questions. These last ones show who really studied." (Odabasi-Cimer & Cimer, 2010, p. 14).
Another teacher also carried out a modification for low-expectancy students and was quoted in Nutter's (2015) study: "When I put the project together, I knew I was going to have them do fewer of the requirements than the other students because I knew they wouldn't be able to handle it. … I knew I was going to need it changed a little bit to make it easier for them." (p. 107).
It was also underlined in one of the studies that either low or high, such teacher expectations were also noticed by students; therefore, when teachers had high expectations and an encouraging manner to help students achieve higher standards academically, students started to feel the same way and reflected the expectations of their teachers (Rubie-Davies et al., 2010).

Discussion and Implications
The current review aspired to systematically investigate the potential factors building teacher expectations for students' academic achievement and teachers' reflected classroom behaviors as a result of their expectations. Student motivation and engagement in academic subjects previously stand as primary student-related factors forming the basis of teacher expectations (Thorburn, 2003;Wijnia, Loyens, Derous, & Schmidt, 2016). It is also seen that students' prior achievement is another student-related factor shaping teacher expectations in the review; however, there are some studies arguing that students' prior academic achievement makes no difference in teachers' expectations of students (Batten, Batey, Shafe, Gubby, & Birch, 2013). Moreover, students' study skills and tendencies to work independently give rise to teachers' having either low or high expectations of student achievement as asserted in previous studies (Coertjens, Donche, Maeyer, Van Daal, & Van Petegem, 2017). Teachers' characteristics (Rubie-Davies, 2007), beliefs, and previous experiences as a student and a teacher also shape what they expect from students (Pajares, 1992) since former teaching experiences with their ex-students and previous learning experiences play significant roles in the formation of their expectations of student achievement by providing them with both professional and experiential insights. Moreover, teachers' self-efficacy is supposed to influence teachers' expectations of student achievement (Rubie-Davies, Flint, & McDonald, 2012) in a way that when teachers have higher self-efficacy beliefs and higher self-esteem in implementing certain teaching methods and techniques, they have higher expectations regarding their students' achievement so they become more inclined to think that every student can reach certain standards in learning. On the contrary, teachers' low expectations of students may bring about academically lower selfperceptions of students themselves (Rubie-Davies, 2006), which creates a disadvantageous situation due to the so-called self-fulfilling prophecy for students.
In the light of the factors shaping teacher expectations, their reflections on teachers' classroom behavior were also investigated systematically in the review. As Brophy (1983) suggested, teacher expectations cannot directly influence students' academic progress; however, when teachers start to differentiate their classroom behaviours in line with their expectations, such differential behaviours of teachers are perceived by students, and then, self-fulfilling prophecy takes the lead, especially when there is a kind of special treatment towards good or high-expectancy students (Trouilloud & Sarrazin, 2003). Parallel to these arguments, while high expectations of success lead to high levels of student achievement, low expectations may tend to cause low levels of student achievement (Lunenburg & Ornstein, 2013) because when students feel discouraged by teachers' low expectations in time, they stay passive during the classes and hence, go on fulfilling the low student profile attributed by the teachers (Lefstein & Snell 2014). For these reasons, depending on their expectations of students, the detection of teachers' differential classroom behaviours is still worth exploring because these behaviours are generally known as being invisible on the side of teachers (Babad, 1993).
The present systematic review also puts forth that teachers have a tendency to change and adapt their teaching methods and techniques so that students can achieve the learning outcomes as expected. With the aim of promoting students' academic progress, teachers are expected to organize the learning environment in line with students' individual differences (Dennis, 2006). On the other hand, teachers' differential behaviours such as interacting more with high-expectancy students, waiting for more to elicit the correct responses from them, and labelling low-expectancy ones in classes, might create a learning atmosphere where students, especially low-expectancy ones, are hesitant about whether the teacher likes and respects them or not (Hamre & Pianta, 2001). They might also feel that their academic abilities are undervalued by teachers (Cooper, 1984). These arguments have similarly been verified in the literature such as Brophy's study (1983) revealing that teachers tend to give correct responses directly instead of asking probes to lowexpectancy students, let them sit at the back rows in classes, address them less frequently, reward even their inappropriate answers, establish less eye-contact with them, and smile less to them by paying insufficient attention to them. Therefore, this situation may pose an obstacle to the establishment of a cooperative and encouraging classroom environment for better learning opportunities.
It has also been revealed in the review that reflective teaching is in relation to teachers' expectations of student achievement because teachers start to evaluate their own teaching skills and performance after classes when they want to accomplish more in terms of students' academic progress. When teachers possess higher expectations for their students, they make more professional investments and try to foster their teaching skills in a more conscious and motivated way (Gorski, 2008;Hinnant, O'Brien, & Ghazarian, 2009). The current review has also found out that teachers have a tendency to assign more demanding tasks to high-expectancy students so as to provide them with more cognitive gains. This finding was also proven in the literature by Brophy (1983) and Mitman (1985) indicating that high-expectancy students are more often criticized by their teachers than low-expectancy ones with the aim of displaying teachers' high expectancies even against more challenging tasks.
It seems evident that teacher expectancies may cause differential teacher behaviors in classes, which might affect levels of student achievement in return as a result of the self-fulfilling prophecy (Skinner & Belmont, 1993). Therefore, especially teachers' classroom behaviors which are the reflections of their expectations of students may stand as a potential danger transmitting their low expectations of some students who may simply be considered low-achievers whose academic abilities are generally underestimated. On the other hand, some high-expectancy students may face this potential danger in a different form, which is being ignored due to the presence of their teachers' overconfidence in their academic capabilities. As a result, although they are supposed to achieve higher, such students may fall behind academically owing to their teachers' diminished care and attention during instruction.

Implications for Practice
Teacher expectations might be a potential obstacle to students' learning capabilities whether they are high or low. Whereas some students are capable of utilizing high expectations as a bar raised highly and making more efforts to reach it, some might feel underestimated and lose their confidence in attaining certain learning outcomes, especially when their teachers' low expectations become apparent. For these reasons, teachers should become more aware of what to expect from their students professionally and how they transmit these expectancies to their students with the professional understanding of knowing the possible positive or negative consequences of their expectations. Therefore, when teachers start teaching, their classroom behaviors should be scrutinized closely so that they do not pose more obstacles to some low-expectancy students' learning. Classroom observations or video recordings of the classes might be employed so as to raise teachers' self-awareness on the issue. Professional development can also be beneficial so that teachers can evaluate their teaching from the expectancies perspective and discuss the possible ways of changing their attitudes to avoid sustained expectations.

Implications for Further Research
The current systematic review showed that in teacher expectation literature there is a need for more on-site investigations such as conducting more ethnographies, case studies, or phenomenological studies which may allocate rooms for more observations, interviews, or focus groups to display fuller, more authentic and in-depth pictures concerning how teacher expectations are reflected in classrooms. Rather than focusing on the factors shaping teacher expectations, their genuine reflections to classrooms as teacher classroom behaviors, might be studied more frequently because these studies are the ones revealing the real effects of teacher expectations on students and their academic achievement. For a similar purpose, students' expectations can also be examined qualitatively so as to find out whether they are in compliance with their teachers' expectations, and learn more about their feelings and reactions to their teachers' expectations as affective filters in learning processes. Herein, teachers' self-efficacy beliefs, reflectivity, and perfectionism might be potential variables influencing their expectations of students by affecting their standards and definitions of achievement. Future studies and researchers might focus on possible relationships among these variables. Studies leading to model or theory building on factors influencing teacher classroom behaviors could also be helpful.