In general, two basic mindsets can be distinguished with regard to the malleability of intelligence: the entity (fixed) mindset and the incremental (growth) mindset (Dweck, 1999, 2006). Some people believe that intelligence is a fixed trait: a person has a certain amount of intelligence, and it cannot be changed. The fixed mindset has been associated with performance goals and with helplessness-oriented strategies in response to failure and setbacks (Blackwell et al., 2007; Burnette et al., 2013), as errors are seen as confirming an individual's inability. Others believe that intelligence is malleable. In other words, intelligence can be changed through effort and persistence. This growth mindset is associated with learning goals, mastery-oriented strategies and beliefs in positive effort (Blackwell et al., 2007; Burnette et al., 2013).

To date, studies investigating the role of mindset in academic performance have focused largely on the mindsets of students with regard to their own intelligence. In this study, we address the mindsets of teachers concerning their general belief about the malleability of intelligence. The teachers' evaluation of students' achievements has consequences for the student's achievement motivation (Rheinberg, 1980, 1983, 2001). Teachers who compared students' learning outcomes with other students social reference norm (SRN), tended to attribute those academic achievements more often to stable characteristics of the student such as “ability.” Teachers who preferred intra-individual comparisons [the individual reference norm (IRN)] attributed academic achievements to highly variable causes such as “effort,” learning strategies and task characteristics. Dweck referred to these observations when she coined the terms fixed and growth mindset to describe the underlying beliefs people have about learning and intelligence (Dweck, 2006).

In the above mentioned study of Rheinberg (1980) on the influence of teacher's preferred reference norm on learning outcomes, he measured the following variables: students' achievement motive, IQ, and causal attribution of one's own academic success and failure at the beginning and at the end of the scholastic year. Additionally, at the end of the scholastic year, students rated the growth of their own academic competence during the year. Results showed that for teachers with an IRN low achievers at the beginning of the school year ended the year as moderate or even high achievers (Rheinberg, 1980). Contrarily, more than half of the students of teachers with SRN, who were low achievers at the beginning of the school year ended up as low achievers at the end of the school year.

Although it is important to note that definitions of the concept intelligence have been inconsistent and subject to change throughout the past century, the conceptualization of intelligence is not the subject of our current investigation.

In this study, we hypothesize that the mindsets of teachers concerning the malleability of intelligence influences their appraisals of student achievement and the feedback they give. These two aspects of the classroom behavior of teachers (i.e., appraisal and feedback) might bear an influence on the learning outcomes and behavior of their students.

In this study, we understand “appraisal of achievement” as referring to the actual assessment of learning outcomes, which can be either preceded or followed by teacher feedback. Beliefs about ability and beliefs about effort are both important within the context of appraisal and feedback (Kaplan and Swart, 1973). Feedback can be defined in many different ways (Smith and Smith, 2015). In this study, we understand it as “information (provided by the teacher) on the performance of the learner, in which both the process and the result are important, in order to promote learning and maintain or increase motivation” (Brown, 2004). We discuss these concepts in greater detail in the introductions to the two studies.

Previous studies have suggested possible associations between the gender of teachers and their pedagogical behavior and beliefs (Nosek et al., 2002; Almutawa, 2005). For example, previous studies of teacher behavior have indicated that female teachers tend to stimulate collaboration and class discussion, in addition to being be more student-centered, indirect and supportive of students than is the case for their male colleagues (Li, 1999). Gender differences can also be observed in teacher feedback. A previous study has demonstrated that female teachers provide more supportive and more expressive feedback (in equal amounts to boys and girls), as compared to male teachers (Duffy et al., 2001). In another study, female teachers gave more compliments and used less directive forms of feedback than male teachers did (Rashidi and Naderi, 2012). Gender has further been shown to affect teachers' beliefs on, for example, the nature of particular school subjects, the curriculum and conceptions of the teaching role (Li, 1999). Less is known, however, about the effects of teacher gender on beliefs concerning the malleability of intelligence, and the appraisal of achievement and feedback.

A second teacher characteristic, teaching domain, is likely to be related to mindset, given the tendency of people to assume that success in some domains (e.g., science, technology, engineering, and mathematics—STEM—subjects) depends upon innate ability, even more so than on dedication and perseverance. Women tend to be stereotyped as lacking such ability (Leslie et al., 2015). Characteristic “teaching domain” had to be taken into account for several reasons, such as the gender-specific character of some school subject (Vassilou, 2010), low proportions of women entering the STEM domain (Meelissen and Drent, 2009; Michels et al., 2014) and the presence of strong gender science stereotypes in men dominated science fields specifically (Leslie et al., 2015). Additional reasons to take domain into account are the more reported negativity in interactions between STEM teachers and their students (Watt et al., 2013) and the negative motivation of STEM teachers to choose teaching as a fallback career (fallback career: second choice career, when one has “failed to be accepted into the career of choice or otherwise unable to pursue their first-choice career,” Watt et al., 2013).

The primary aim of the article is to investigate how mindset influences (1) the appraisal of student achievement (study 1) and (2) the frequency and type of oral feedback that teachers provide in the classroom (study 2). Additionally, the effects of gender and teaching domain on appraisal of achievement and feedback are explored.

The ways in which teachers evaluate and appraise the achievements of their students might depend upon the reference norms (e.g., individual vs. social comparisons) they prefer to apply when evaluating learning outcomes. As established by Rheinberg (1980), the concept of teacher reference norm orientation (RNO) is defined as “a standard to which individual achievements are compared.” Such standards can be based on any of the various frames of reference that teachers can adopt. The reference norm can be seen as an effect of a teacher's mindset concerning malleability. According to Heckhausen (1974) three frames of reference have been distinguished: the SRN, the IRN and the criterial reference norm (CRN). When applying the SRN, teachers compare the achievements of a given student to those of fellow students. Teachers adopting the SRN believe that differences in ability amongst students are highly stable across time. Their appraisals are strongly dependent upon whether the learning outcomes of a given student are above or below the class average (Rheinberg and Engeser, 2010).

When applying the IRN, teachers compare the achievements of individual students to their prior achievements. Teachers adopting the IRN emphasize improvement, effort and learning, with a focus on the individual process of learning. The application of this reference norm has been shown to decrease fear of failure in students and brought many low achievers up into the high achieving range (Rheinberg and Engeser, 2010). When applying the third type of reference norm, the CRN, task-inherent properties are taken as the standard of comparison. In a study of Martinez et al. (2009) it was found that teachers evaluated student performances not in absolute terms but relative to other students in the school (SRN) and that they might adjust their grading for some students, perhaps with basis on perceived differences in needs and/or abilities.

Based on a recent study of teachers working with students with learning disabilities and otherwise academically-challenged students, Wilbert and Grúnke (2010) report that the reference norms used by teachers can affect the achievements and motivation of their students, with the INR having a positive effect on learning and the SNR inhibiting learning.

The current study addresses a gap in the research literature with regard to the relation between mindsets of teachers concerning the malleability of intelligence, teacher characteristics gender and teaching domain, and their appraisal of achievement. The IRN as mentioned above is growth-oriented. Teachers with an IRN appraise improvement, effort and learning, similar to growth-oriented teachers. We expect growth- oriented teachers to value increasing achievements positively, independent from the end marks students achieve. This in contrast to fixed oriented teachers; we expect them to value sufficient end marks, and we expect them to have less appraisal for students' personal increasing improvements. Therefore, we tested the main hypothesis that teachers with a more growth- oriented mindset are more positive in their appraisal of students' increasing achievements than those with a more fixed oriented mindset.

In addition, the effects of teacher characteristics “gender” and “teaching domain” on their appraisal of achievement were explored.

In this study, we investigate the association between the mindsets of teachers and their appraisal of student achievements. In line with our hypothesis, the results reveal a positive correlation between mindset and the appraisal of achievement for the increasing marks but not for non-increasing marks. Attention for increasing student achievements has been demonstrated to be of importance for students' motivation (Meece et al., 2006; Rheinberg and Engeser, 2010; Wilbert and Grúnke, 2010). In addition, we found that gender is associated with the appraisal of increasing marks, with women valuing these achievements slightly higher than men. In the following section, we present Study 2, exploring how mindset is related to the type and amount of feedback that teachers provide to their students in daily classroom situations.

In classroom situations, teachers generally provide feedback, to which students respond. The feedback that teachers provide to their students affects their learning behavior and learning outcomes (Hattie and Timperley, 2007; Lipnevich and Smith, 2009; Geyskens et al., 2012), and it has a powerful influence on motivation (Hattie and Timperley, 2007; Wilbert et al., 2010).

The literature contains a variety of classifications regarding types and descriptions of feedback. We highlight several of these classifications in the context of our study. One classification has to do with the forms, objects, descriptions, opinions, views, effects and goals of feedback (Sol and Stokking, 2009). In their scoring form for the Observation of Teacher Feedback Behavior, Sol and Stokking (2009) list result-oriented, process-oriented, instruction-oriented and “other” feedback. Emphasizing that it is important for students to understand the information they receive, Hattie and Timperley (2007) distinguish four levels of feedback: task, process, self-regulation and self-level.

Beliefs in general influence behavior (Cross, 2009). Earlier research shows that feedback both regulates and is regulated by motivational beliefs. External feedback has been shown to influence how students feel about themselves (positively or negatively), and what and how they learn (Dweck, 1999).

To monitor the progress of and the reflection on the students' learning process, teachers should provide formative feedback. This type of feedback provides information on performance to improve and accelerate learning (Sadler, 1998) and to adjust teachers' educational activities (Sluijsmans et al., 2013). The research on formative assessment and feedback was reinterpreted by Nicol and Macfarlane-Dick (2006) to show how these processes could help students take control of their own learning, i.e., become self-regulated learners. In this way, feedback not only contributes to the teaching process, but also to the learning process and the improvement of achievements (Arts et al., 2016; Schildkamp et al., 2016). In some countries high-ranking on PISA scores (e.g., Finland), self-regulating skills and the continuous providence of formative feedback to students is central to the learning process (Hill, 2011).

Feedback is closely related to assessment: earlier studies showed that tests focused on certification and selection, without forms of informative feedback were negative for the process of learning. Without informative feedback those tests diminished student's responsibility and motivation for learning (Sluijsmans et al., 2013).

Corresponding to the description of a growth mindset, growth-oriented feedback has been described as feedback that guides and motivates students, enhances their learning (Voerman, 2014), and keeps them persistent, resilient and focused on the process of learning. It provides specific information (Voerman, 2014) about the progress (and results) of students. Corresponding to the description of a fixed mindset, fixed feedback emphasizes basic qualities (e.g., intelligence or talent) and characteristics as fixed traits (Kamins and Dweck, 1999). It provides information about the results as such, and not about the process of learning.

In conclusion, and consistent with the conclusions of Hattie and Timperley (2007) and of Shute (2008), we regard feedback as information provided by the teacher concerning the performance of the learner, with both process and result being important, in order to promote learning and to maintain or increase motivation (Brown, 2004).

Based on the literature above, we distinguish between two dimensions of oral feedback: growth-oriented feedback and fixed feedback.

Growth-oriented feedback:

Fixed feedback:

Study 2 focuses on the oral feedback provided by teachers in relation to their mindsets. First we investigated which types of feedback interventions teachers generally provide in their classrooms (based on the Observation of Teacher Feedback Behavior instrument (Sol and Stokking, 2008) and our definition of growth and fixed feedback as described above. Then we hypothesized that teachers with a growth mindset would overall provide more feedback than teachers with a fixed mindset would (Hypothesis 2a), given that those with a growth mindset believe that feedback on efforts and strategies during the learning process is essential to the learning and achievement of students. For the same reason, we hypothesized that the growth- oriented teachers would provide more growth-oriented feedback (Hypothesis 2b) than fixed feedback.

Furthermore, we investigated the effects of two teacher characteristics, gender and teaching domain, with regard to the type and amount of feedback provided.

The second study was designed to investigate the link between the mindsets of teachers and the amount and type of feedback they provided in classroom situations. Personal praise or criticism (growth and fixed) was used in only 2.5% of all oral feedback interactions. Feedback concerning how (growth) and what (fixed) was provided in almost equal amounts (26.5 and 27.6% respectively). Growth-oriented and fixed feedback together comprised roughly half of all of the feedback interventions observed (with the other half being related to instruction or behavior). Contrary to our hypothesis, we found a significant negative correlation between mindset score and the total amount of feedback. In other words, teachers with a more growth- oriented mindset provided less feedback than teachers with a more fixed oriented mindset. Furthermore, we found an indication that male and/or STEM-teachers provided more growth feedback compared to female/non-STEM teachers. In the general discussion below, we address our findings from Studies 1 and 2 in an integral manner.

In study 1 we found a relation between mindset and appraisal of achievement, indicative for the influence of teacher's mindset on how they think about students. The results of our study indicate that overall growth-oriented teachers appreciate increasing marks more than fixed oriented teachers. This is an important finding, given that focusing on increasing achievement has been demonstrated to motivate students (Rheinberg and Engeser, 2010). Focusing on improvement can underscore a student's feelings of competence, which combine with feelings of autonomy and relatedness to form the foundation for intrinsic motivation (Ryan and Deci, 2000). In line with the conclusion of Wilbert and Grúnke (2010), our results indicate the necessity of pointing out improvements. This is especially true for students in the pre-vocational track of secondary education, as they are more likely to be academically challenged and less motivated in school than are students in other tracks. The Dutch system of secondary education has a classification system consisting of several different tracks: pre-vocational education (VMBO), higher general secondary education (HAVO) and pre-university secondary education (VWO). Students in the pre-vocational track exhibit large differences in learning rate, learning style and motivation (Harskamp et al., 2000). Although our results cannot be generalized to all students in this track, they do suggest that, compared to students from the other two tracks, these students tend to be more practical and application-oriented, to prefer educational methods involving concrete rather than theoretical techniques and working forms, and to encounter frequent difficulty in managing their own learning processes (e.g., planning, monitoring, executing/implementing, and evaluating; see Hamstra and Van den Ende, 2006). In combination with self-confidence with regard to ability and motivation, the ability to regulate and control one's own learning process is particularly likely to generate optimal learning outcomes (OECD, 2003). Focussing on growth is therefore of particular importance for students in the pre-vocational track, who generally tend to be less motivated than other students are.

Study 2 focused on the association between mindset and feedback. Feedback has been shown to influence the learning behavior and outcomes of students (Hattie and Timperley, 2007; Lipnevich and Smith, 2009; Geyskens et al., 2012).

Teacher feedback that focuses more on final results (i.e., whether they are sufficient or insufficient) and less on the learning process (i.e., whether there is improvement) might be less effective (Black and Wiliam, 1998; Hattie and Timperley, 2007; Shute, 2008), thus potentially decreasing the results and motivation of students. The literature provides increasing evidence for the importance of such growth-oriented feedback (Mueller and Dweck, 1998; Dweck, 2006; Skipper and Douglas, 2012; Gunderson et al., 2013). Given the frequent occurrence of feedback, their impact is likely to be important, thus underscoring to the importance of providing effective feedback. 26.5% of all feedback interventions were growth-oriented (e.g., pointing out learning questions and hints regarding strategies, possible improvements and alternatives for the choices that students had made) and surprisingly only 2.5% of the oral feedback interventions observed in our study contained comments reflecting personal praise/criticism on doing (e.g., “you tried very hard”). Contrary to earlier findings indicating that only half of teachers provide specific feedback (Voerman et al., 2012; Voerman, 2014), all of the teachers in our study (regardless of gender or teaching domain) provided one or more forms of such specific growth-oriented feedback. This discrepancy does not seem to be explained by differences in the definitions used, as Voerman's definition and examples of specific feedback (i.e., “provides information about the learning goal with reference to the task, the processing of the task, or self-regulation, while not being overly elaborate”) largely corresponds to with the concept of growth-oriented feedback. Our overall findings with regard to feedback suggest that teachers use growth feedback to only a limited extent (27.8% of all feedback provided by the teachers was identified as growth feedback).

Given the fact that for example in the educational field formative assessment, closely related to growth feedback, is on the rise (Kneyber and Sluijsmans, 2016), this in an important finding. Formative assessment can only be implemented successfully when teachers are able to provide growth process-oriented feedback.

Contrary to our hypothesis, teachers with a growth mindset generally provided less feedback. The results from our sample do not suggest that teachers with a fixed mindset do not guide their students during the process. On the contrary, they provide more feedback overall. Rattan et al. (2012) reported that teachers who believed math-intelligence to be fixed, tended to express their support and encouragement in unproductive ways. They found that those teachers' tended to express both support and encouragement (growth-oriented feedback) in a “comfort-oriented” manner, sending the implicit message: Its' ok- Not everyone can be good at math” (fixed oriented feedback). Contrarily to Rattan, our findings do not suggest that fixed mindset teachers provide support in this comfort-oriented manner. However, in future studies it will be important to explore to which extent such confusing messages occur and to explore the impact of this type of feedback. Not only Rattan, but also Rubie-Davies (2010) reported that teachers might send confusing messages. In her study, some teachers provided messages about positive student characteristics (e.g., trying hard, behaving well, relating well to others) while being negative about their expectations with regard to achievement. This could unintentionally decrease the students' motivation and causing them to have lower expectations regarding their own performance. When teachers report a growth mindset, but do not put their beliefs in action (that is provide growth feedback), students could become demotivated and have lower expectations for their own achievements (Rattan et al., 2012). The teachers in our sample with a growth mindset provided less feedback than those with a fixed mindset. One explanation could be that teachers with a growth mindset are less inclined to urge their students to achieve more or better, being more likely to appreciate their students' efforts as such. Furthermore, self-reported mindset and feedback behavior might be incongruent. This corresponds well to the concept of a “false growth mindset,” in which teachers might claim to have a growth mindset, but do not reflect it in their words or actions (Dweck, 2015).

As indicated by the results from Study 1, teachers with a growth mindset tend to value scores that reflect improvement more highly than teachers with a more fixed mindset. For the growth- oriented teachers, it is not the outcome but the process that is the most important. At the same time, school-based education aims to ensure teaching and learning processes lead to the achievement of certain goals in terms of academic achievement (Rijksoverheid, 1963; Darnon et al., 2012). Therefore the focus of growth-oriented feedback, should include interventions on both process and result.

In our study, we measured the mindset as a general belief, we did not distinguish teacher's mindset in specific situations or toward individual students. Another explanation for the finding that teachers with a more growth-oriented mindset do not provide more growth-oriented feedback, might be that the type of feedback is dependent on both the mindset of the teacher and the characteristics of the individual student. Jager and Denessen (2015) reported that teacher beliefs and their causal attributions toward different low achieving students showed a large within-subject variance. Causal attributions such as attention, effort and interest were described inconsistently for different low-achieving students. Jager and Denessen suggested that attributions are not mere teacher variables but should be studied at the student-specific level.

The relationship between mindset and feedback is even more intricate, however, as students with different mindsets (i.e., growth and fixed) may respond differently to feedback. Compared to the fixed mindset, the growth mindset has been associated with a more effective response to feedback regarding an occasional failure (Dweck, 2007). It might be important to make teachers more explicitly aware of their mindsets, their feedback styles and students' mindsets. If teachers are aware of their own mindsets concerning intelligence, and if they are provided with information on how to provide growth-oriented feedback, this is likely to enhance the effectiveness of their students' learning processes.

The gender of teachers has been shown to be associated with several aspects of their classroom behavior and feedback (Li, 1999; Duffy et al., 2001; Rashidi and Naderi, 2012). The current results indicate that male and female teachers differ in terms of the appraisal of achievement, with women tending to value increasing achievements slightly more than men do. However, our findings suggest that female (vs. male) teachers provide a significant smaller proportion of growth-oriented feedback.

Based on the belief that success in some domains (e.g., STEM subjects) depends upon innate ability (Meelissen and Drent, 2009; Michels et al., 2014; Leslie et al., 2015), we predicted that teachers working in such domains would be more oriented toward a fixed mindset. In our study, we found no associations between teaching domain and appraisal of achievements, but contrary to our expectations, we found that STEM teachers provide a higher proportion of growth-oriented feedback then non-STEM teachers. Possibly, the type and amount of feedback needed is domain specific. In the STEM-domain more process feedback might be needed due to the nature of the STEM-domain, regardless of the teachers' (mindset) beliefs. It should be noted that the associations between gender, domain and feedback should be interpreted with caution, as gender and domain overlap in the current sample.

Several limitations of the study should be mentioned. One has to do with the limited sample size in Study 2. However, this sample did not differ from the larger sample regarding the key characteristics appraisal of achievement and mindset on malleability of intelligence. Because of the small sample (N = 22) and the number of predictive variables (mindset, gender and domain) it was not feasible to conduct multiple regression analyses.

In addition, our research was conducted on teachers working in one particular track of the Dutch educational system (i.e., pre-vocational secondary education, or VMBO), which includes 55% of all secondary students in the Netherlands (Van Schaik, 2013). Teachers in the pre-vocational track work with students with specific learning characteristics, and they must therefore comply with specific demands with regard to capability, including practice-oriented learning, attention to vocational subjects, integration of several subjects, attention to the learning processes of individual students, a student-oriented approach, customization, and attention to social and emotional development (Van der Rijst et al., 2011). Such demands, especially those having to do with the necessity of attending and customizing instruction to the learning processes of individual students, might be particularly attractive to teachers who believe in growth.

Furthermore, the Theory of Intelligence Questionnaire (Dweck, 2006) is an explicit, self-report measurement. There might therefore be a discrepancy between the responses that teachers entered on this instrument and their actual behavior in the classroom. The participants might have been unwilling or unable to report on their beliefs (Cunningham et al., 2001; Gawronski and De Houwer, 2014), or they might have been biased by a tendency toward socially desirable responses (Hornstra et al., 2010). A different or additional method of measuring mindsets might help to diminish the risk that the results reflect a false growth mindset. Finally, Jager and Denessen (2015) reported that teacher beliefs and causal attributions toward different low achieving students showed a large within-subject variance. A limitation in our study was the absence of mindset of individual students, although results from study 1 suggested that teacher's mindset impacted feedback behavior.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.