Effect of Assessment Feedback on Dancers' Motivation

Table of contents

List of Tables

List of Figures

Acknowledgements

1. Introduction

2. Literature Review
Self-efficacy Theory
The relevance of promoting high levels of self-efficacy
Method to examine the effect of others in self-efficacy
Feedback
Feedback from the Self-efficacy framework
Other results: individual differences
Self-efficacy and Effort
Summary
Hypothesis

3. Methodology
Participants
Procedure
Instruments
Statistical Analysis
Ethical Considerations

4. Results
Preliminary Analysis
Hypothesis 1
Feedback Effect on Self-efficacy
Second Hypothesis
Effort Intentions, Feedback and Initial Self-efficacy

5. Discussion
Feedback effect on Self-efficacy Levels
Differences between pre and post manipulation per feedback condition
Non feedback
Lower evaluated feedback
Feedback per objectives.
Upper evaluated feedback
Self-efficacy Predicts Effort Intentions
The effect of initial self-efficacy levels on effort intentions following feedback
Upper evaluated feedback
Lower-evaluated feedback
Feedback per objective
Limitations
Further Research
Practical recommendations for teachers

6. Conclusion

7. References

8. Appendix
Appendix A
Appendix B
Appendix C
Appendix D
Appendix E
Appendix F PERSONAL CODE
Appendix G PERSONAL CODE
Appendix H

List of Tables

Table 1: Frequency and Percentage of Participants according to their Age, Course and Specialities

Table 2: Spearman Correlations between Pre Self-efficacy, Post Self-efficacy and Effort Intentions

Table 3: Percentage of Participants that Increased their Self-efficacy Levels in Post comparing to Pre Manipulation

Table 4: Previous Studies Examining the Self-efficacy Effect on Sports

Table 5: Initial Self-efficacy Levels Effect on the Feedback Response

Table 6: Effect of High Self-efficacy levels on Effort

Table 7: Mean Rank Values in Post Self-efficacy

Table 8: Kruskal Wallis p. value

Table 9: Pre to Post Self-efficacy Differences in the Upper-evaluated Group

Table 10: Pre to Post Sel-efficacy Differences in the Objective Group

Table 11: Pre to Post Differences in the Lower-evaluated Group

Table 12: Pre to Post Differences in the Control Group

Table 13: Model: Self-efficacy Levels as Predictors of Effort Intentions

Table 14: Linear Regression

List of Figures

Figure 1: Percentage of participants per condition who increased, decreased and remained constant their self-efficacy values between pre and post manipulation.

Figure 2: Mean ranks of effort intentions according to their initial self-efficacy levels (High, Medium and low SE) and their received (upper, objective and lower evaluated) feedback.

Abstract

The repercussion that term assessments have on students training in a dance conservatoire is a controversial topic. Several authors in sport psychology have shown the effect that manipulated feedback can have on self-efficacy and the existing relationship between self-efficacy and effort. However, these questions have not been previously studied in dance. Therefore, the aims of this investigation are 1) to examine the impact manipulated feedback can have upon the self-efficacy levels in dancers and 2) to analyze if this variable predicts the effort intention for the dancers to improve in the next term. As such, 61 students and 5 teachers from a dance conservatoire participated in an experimental study. Students were randomly divided into 4 groups. Teachers assessed their competence according the conventional criteria (scores from 0 to 10) during a flamenco movement. The first group received upper-evaluated scores, the second group lower-evaluated scores, the third group did not receive scores (control group) and the last group received one score per objectives (e.g. postural control, rhythm, etc) plus the accurate general score. Results show that objective feedback is the most beneficial strategy in terms of self-efficacy and lower-evaluated is the most detrimental for students. In contrast with the initial hypothesis, self-efficacy does not linearly predict effort intentions for next term. Therefore, this experimental study does not recommend the use of lower-evaluated and upper evaluated feedback; and suggest teachers to provide objective feedback due to its positive effect on self-efficacy. Among the limitations are the number of participants per group and the instrument used to measure effort intentions. Future research should be focused on examining the variables that influence on the effort intentions and assess the practicability of this new assessment strategy within real contexts. Finally, this study provides practical suggestions for dance conservatories teachers so they can implement them in assessment situations.

Key words: self-efficacy, feedback, effort intentions, assessment, dance conservatory.

Acknowledgements

I would like to thank my partner Juan and my family for their unconditional support throughout my masters journey.

Thank you to José Carlos Caracuel for guiding me both personally and professionally and for providing a critical view of my study, enabling me to reflect on my work.

Thank you to all my masters friends, especially Alex Balfour, Shaunna Meredith and Hannah Etlein-Stein for generously dedicating their time to review my use of English. Thank you Eleanor Quested, Ralph Cox for your time and suggestions. A large part of this study is because of your help.

I would like to thank every single person (teachers, students, parents, custodians, directors and students´ managers) from the Dance Conservatory for opening their doors to this study. Without your understanding and willingness to participate this investigation would not have been possible.

1. Introduction

Self-confidence in athletes has been frequently researched from the conceptual and methodological self-efficacy perspective (Bandura, 1997). The relationship between self-efficacy and sport performance has been analysed since the 1980s, using experimental as well as observational methodologies in different sports such as gymnastic, tennis and swimming (Lázaro & Villamarín, 1993).

According to the Self-efficacy Theory, self-efficacy is defined as “beliefs in one´s capabilities to organize and execute the courses of action required to produce given attainments” (Bandura, 1997, p.3). Efficacy beliefs are not judgements about possessing a set of skills to produce an action, but rather judgements of what can be accomplished with those skills. In other words, what one thinks one can do with one’s skills, not about what skills one has (Bandura, 1986 in White & Saks, 2007).

“Self-efficacy is a cognitive mechanism that mediates between selected self- appraisal information and peoples’ subsequent thought patterns, emotional reactions, motivation and behaviour” (Bandura, 1997, p.3). Variables that have been found to be related to self-efficacy, such as emotional reactions and attributions are important contributors to performance (e.g. Chase, 2001, and LaGuardia & Labbe, 1993 respectively). For this reason, self-efficacy judgements are extremely relevant and have been consistently studied both in general psychology and in sport psychology (Moritz, Feltz, Fahrbach, & Mack, 2000).

According to Bandura, self-efficacy beliefs are task specific rather than representing an overall trait; therefore they are changeable (Ilies, Judge & Wagner, 2010). The contextual characteristics of self-efficacy differentiate this from other constructs, such as perceive competence (García-Dantas & Caracuel, 2011a). Feltz, Short, and Sullivan (2008) explained that whereas self-efficacy would be “the belief that one can successfully organize and execute a course of action to reach a specific goal” (p.25); perceive competence would refer to “the belief of one´s ability in a certain domain or across a set of behaviours, developed as a result of cumulative interactions with one´s environment” (p.26).

There are four main sources of information that affect self-efficacy, namely, performance accomplishments, vicarious influences, verbal persuasion and physiological information (Bandura, 1997). The first part of the current study will focus on an aspect of verbal persuasion, particularly teachers´ feedback. This investigation, following Bandura´s recommendations and theory (1977, 1997), will analyse the effect that different types of assessment feedback have on the self-efficacy level. Therefore, feedback from dance teachers will be manipulated in order to test whether there is any difference in self-efficacy levels regarding to the received feedback (verbal persuasion).

Although there is literature that relate self-efficacy and immediate effort (Weinberg, 1981), non study have been found that analyses the relationship between self-efficacy levels and future effort intentions in both athletes and dancers. Therefore, the effect of self-efficacy on effort intentions to optimize the performance is analyzed in the second part of this study. A one item questionnaire will be used to ask participants how much effort they would invest in a hypothetical future situation in order to enhance the exercise that they have already executed. According to previous literature, the most efficacious athletes invested higher levels of effort and persevere in order to overcome obstacles (Bandura, 1997, p.3; George, 1994; White & Saks, 2007).

Additionally, the present study is needed because up until this point, the dance conservatoire system only allows teachers to provide one score that represents performance (from 0 to 10), per term to each student. As a result, the information tends to be vague and inaccurate as it is really difficult to synthesis the performance of a three month period in just one mark. Sometimes, anecdotally it is reported hat this difficulty leads some teachers to transform the real score to another either higher or lower with the aim of encouraging students to increase their efforts. However, these manipulations might not promote the expected outcomes as this practice has not been supported on research. Thus, this investigation is justified practically because it might provide interesting information about good practices that teachers could display during assessments.

As an alternative method of providing exclusively a general score, this study includes another option. This is giving specific scores per term objectives as a complement to the accurate general score. This method was designed with the aim of finding out whether this is a more constructive and beneficial strategy in terms of self- efficacy than only a general score. In case results find objective feedback strategy was beneficial in terms of self-efficacy, more accurate and specific feedback would be recommended to conservatoires dance teachers.

Therefore, the aims of the current study are to investigate 1) if different kinds of feedback (upper evaluated, feedback per objectives, lower-evaluated and non feedback) influence on students´ self-efficacy 2) whether self-efficacy predicts effort intentions to improve a specific task in dance students.

2. Literature Review

Self-efficacy Theory

Bandura´s (1977, 1997) theory of self-efficacy was developed within the framework of Social Cognitive Theory. From this perspective “individuals are considered proactive agents in the regulation of their cognition, motivation, actions and emotion rather than passive reactors to the context” (Feltz, Short, & Sullivan, p.5). People use forethought, self-reflection, and self-regulation to influence their own functioning.

Bandura (1997) refers to self-efficacy as a common cognitive mechanism that mediates between selected self-appraisal information, and people’s subsequent thought patterns, emotional reactions, motivation and behaviours. According to Feltz, Short, and Sullivan (2008) such motivated behaviours and thought patterns are important contributors to performance in sport. People with high self-efficacy levels pursue challenging goals, cope with pain, and persevere through obstacles. Athletes with low self-efficacy avoid difficult goals, worry about possible injury, expend less effort, and give up in the face of failure (p.5).

During the past decades self-efficacy construct has been meticulously studied in different populations such as children (e.g. Chase, 2001), drug addicts (e.g. Young, Oei, & Crook, 1991), or athletes (e.g. Whyte & Saks, 2007). The main sources of self- efficacy, the self-efficacy concept and its repercussion have been thoroughly explained. Bandura (1997) highlights four main sources which influence on self-efficacy: performance accomplishment, vicarious experience, verbal persuasion and physiological and affective states. A great number of studies have been published about the sources of self-efficacy and how these sources affect self-efficacy (e.g. Nease, Mudgett, Quinones, 1999; Weinberg, Gould, Yukelson, & Jackson, 1981). As an example, some researchers have manipulated information regarding an individual’s performance in order to measure the effect that verbal persuasion has on self-efficacy (e.g. Cofee & Rees, 2011).

Another line of research has focused on the effects that self-efficacy as an independent variable, presents in other cognitive, emotional and behavioural constructs. One of the pioneer studies based on a student population, was designed by Collins in 1982 (In Bandura, 1997). This author selected children with high and low self-efficacy at each of three levels of mathematical ability. They were then given difficult mathematical problems. The study demonstrated that within each level, children with higher self-efficacy levels were quicker to discard faulty strategies, solved more problems, chose to rework more of those they failed, and did so more accurately than children of equal ability who doubt their efficacy. Authors found that efficacy beliefs predicted interest in, and positive attitude toward mathematics, whereas mathematical ability did not. (p.214)

As it has been noted in the previous example, self-efficacy is not always the reflection of the real abilities but the perception of these capabilities. A talented and gifted dancer might have low self-efficacy because of negative past experiences. However, not only past accomplishments influence self-efficacy, repetitive negative comments can also lead to a decrease in dance students´ self-perceptions (García-Dantas & Caracuel, 2011, November). Particularly, dance teachers´ messages and feedback exert an important role on the students´ self-efficacy. As reported in a previous study, dance teachers´ behaviour influences directly on the students´ motivation. In that respect, results show that those teachers with higher frequency of positive reinforcement had the more motivated students (García-Dantas & Caracuel, in press). Teacher positive feedback behaviour can influence what the dancer believes they can achieve, which can influence the goals that they set and therefore, the effort to reach those goals (Feltz, Short and Sullivan, 2008, p.5). This finding justifies the analysis of teachers´ conduct to find out the strategies and methods that promote optimal levels of self-efficacy in dancers.

Although, it seems that there is a general consensus about the negative effects in cognitive (e.g. goal setting) and behavioural variables (e.g. effort) caused by a low self- efficacy (e.g. Bandura, 1977; Cantón & Checa, 2012) current studies argue that extremely high levels are usually counterproductive (Kim & Chiu, 2011;Nease, Mudgett, & Quiñones, 1999). Therefore, the lack of agreement about feedback effect on sport and the lack of information about self-efficacy in dance contexts justify the analysis of this variable in the population.

The next section will show some of the most popular studies that evaluated the effect that self-efficacy have on behaviour, cognition and emotions with the aim of providing a greater understanding of such an important psychological phenomenon for wellbeing and performance. Furthermore, it will be useful to contextualise the present investigation and justify the need of studies that promote adequate levels of self-efficacy in a great number of tasks.

The relevance of promoting high levels of self-efficacy

Research from the self-efficacy perspective has demonstrated that self-efficacy is an important determinant of individuals' motivation and behaviour. Thus, self-efficacy influences such variables as performance, goal level, goal commitment, and choice of activities in sport (Nease, Mudgett, & Quiñones, 1999).

Self-confidence is a strongly related construct with self-efficacy in that both terms are categorized as cognitive mediators of individuals´ motivation and behaviour within a goal context and both are conceptualized as what one can do with their skills, though the goal in self-confidence is more broadly defined than it is typically with self-efficacy (Feltz & Chase, 1998, as cited in Feltz, Short & Sullyvan, 2008, p.27). In spite of the relevance of the perceptions that individuals have on themselves, Hanrahan (1996) found that most of the dancers in her study, felt that their self-confidence was not always as high or as strong as it ideally could be. Even so, very few dancers felt that they could have any control over their self-confidence. It was seen as something that you either have or you don’t have. This would justify the need to design investigations that study effective strategies to increase the self-confidence and self-efficacy levels within this population.

Self efficacy and behaviour: persistence

Studies in sport psychology have shown that athletes with higher self-efficacy perceived competitive wrestling situations as less threatening than athletes with low levels of self-efficacy (Treasure, Monson, & Lox, 1996). Furthermore, Weinberg (1985) & Schmidt and DeShon (2010) have found that those with the highest self-efficacy levels had greater expectancy of success. Thus, high expectancy of success together with low threatening perceptions of a specific competitive situation, might explain why high self-efficacious athletes are also those with greater adherence. McAuley, Wraith, Duncan (1991) for example, found that high self-efficacious dancers were more intrinsically motivated toward attending aerobic dance classes than were their less self- efficacious counterparts.

The professional curriculum of dance in Spain is a 14 year long degree. Throughout this time, students attend a conservatoire for an average of four hours every day (García-Dantas & Caracuel, 2011b). The requirements of this discipline demand extremely committed students from a very young age. According to the Self-efficacy Theory and the previously mentioned studies, it seems necessary to promote high levels of self-efficacy in a variety of tasks so that in spite of the obstacles and difficulties that some movements can present, the adherence to the activity remains constant.

Self-efficacy and cognition: goal setting.

Self-efficacy and goal setting have been closely related in the literature. According to Bandura (1997) individuals set their goals depending on their self-efficacy levels. In that respect, Feltz, Short, and Sullivan (2008) stated that:

People with reasonably accurate efficacy judgments will undertake realistically challenging tasks as their confidence and efficacy believes develop. However, people who largely underestimate their capabilities hinder their development in the given area and chance losing out on rewarding experiences. For instance, teenagers who believe that they cannot learn to swim miss out the potential fitness and social rewards that come from swimming (p.15).

Therefore, considering the relationship between self-efficacy and goal setting, goal setting is another reason why enhancing self-efficacy is essential in dancers. This is even more important in dance conservatories where groups are numerous and sometimes it is difficult for teachers to pay the necessary attention to each student (García-Dantas & Caracuel, 2011b). Thus, dancers should control their own progress by setting their own goals (Ilies, Judie, &Wagner, 2010).

Self-efficacy and emotions: positive affect and anxiety

Positive affect

Anxiety, worry and depressive moods are some of the most typical emotional reactions that have been studied in relation to one´s self-efficacy (Feltz, Short, & Sullivan, 2008). Additionally, high self-efficacy and positive affect have been strongly related in the literature. According to Feltz, Short, and Sullivan (2008) “ a strong, resilient sense of self-efficacy can help one withstands the effects of competitive pressures” (p. 86).

Previous research has found that athletes who lost a competitive sporting event presented a significant decrease in positive affect levels after the competition. A positive sense of self-efficacy however can help prevent the decline in positive affect after a loss in sport. In order to test this, Brownm, Malouff, and Schutte (2005) designed an intervention that focused on self-efficacy promotion based on encouraging images related to three sources of self-efficacy: personal mastery, verbal encouragement and vicarious mastery. Results showed that the experimental group experienced less decline in positive affect (from pre to post), than a control group after losing a competition.

Nevertheless, it is still unknown the long term effects that this kind of programmes presents. It has not been found studies that investigate for how long this resilience is effective. Furthermore, we do not know how many events of failures are necessary to impact to those with high self-efficacy. In other words, where is the barrier from which individuals start the negative spirals of the learning helplessness process?

Cantón and Checa (2012) analysed the emotions and self-efficacy perceptions that football and karate players had experienced throughout past competitions. Authors reported a positive correlation between pleasant emotions and self-efficacy and a negative correlation between high arousal of unpleasant emotions (anger and fear) and self-efficacy in both sports. Correlations were only found in athletes who had a successful past experience, and not in those participants whom remembered their experience as being unsuccessful. These results proved unexpected to the authors who presumed to find no differences between experiences and so the authors continue to research this topic to discover the reason behind this.

A possible explanation why they might not find significant effects in unsuccessful experiences could be related to the way in which they measured self-efficacy. According to the Theory (Bandura, 1997), self-efficacy should refer to a specific task rather than a whole match or competition. In that sense, it is possible that within the same competition an athlete might feel confidence in certain moments but not in others. This fact could influence the lack of explanation of some results from this research, as the study measured self-efficacy through retrospective questionnaires that referred to entire matches as opposed to specific elements.

To summarise, the analysed studies highlight the importance of possessing high levels of self-efficacy due to its existing relationship to mood and positive emotions. Moreover, people with higher self-efficacy levels in specific tasks who fail, present less negative emotional reactions and less psychological distress and fatigue than low- efficacy participants (Brown, et al, 2005).

In contrast Bandura (1988 as cited in Feltz, Short, & Sullivan, 2008) explains that if people think that they cannot cope with a potential threat, they might experience disruptive arousal, which may decrease their believes that they can perform successfully (p.87). This is necessary when considering the rigorous and continuous assessments common to dance conservatoire environments (García- Dantas & Caracuel, 2011b). It is therefore of great importance that self-efficacy enhancements are provided in order to prepare students psychologically to cope with failures.

Anxiety

Many studies have consistently shown that self-efficacy can accurately predict athletes´ state anxiety (e.g. Marquez et al, 2002). Investigating the strategies that help to increase self-efficacy is beneficial because teachers and educators could incorporate them to their training and potentially reduce the dancers´ stage anxiety. This is particularly relevant as research has shown that dance students and professional dancers very often experience state anxiety (Walker & Nordin-Bates, 2010). Bakker (1991) showed in his study a greater predisposition to suffer anxiety in dancers than other populations.

Similarly, other paper based on a dance conservatoire which compared psychological variables between music and dance population found that dancers presented greater state anxiety, stress and lack of time to study than musicians experienced (Caracuel, García-Dantas, & González, 2011). This justifies the importance of promoting an optimum level of anxiety to improve performance and their wellbeing in every dancer.

Research has consistently demonstrated that anxiety has a direct relationship to performance. However, the relationship between self-efficacy and anxiety has been found to be stronger (Feltz, 1982; Haney & Long, 1995; McAuley, 1985). Therefore, the role of self-efficacy in determining anxiety levels is predominant and consistent in athletes. Thus, if a dancers´ self-efficacy level is low, it is expected that they will present a high level of anxiety. This is another important reason why sport psychologists consistently recommend the promotion of the self-efficacy levels in order to reduce state anxiety in athletes (Jaenes & Caracuel, 2001).

Performance as a result of a combination between behaviour, emotion and cognition

Up until this point, the important role that self-efficacy presents on adherence and participation in an activity, on success expectancy, on goal setting and mood and anxiety reduction have been briefly discussed. As these variables are related to performance (Burke and Jin, 1996), it is expected that increasing self-efficacy levels could be a beneficial strategy to optimize performance. In order to confirm this, Moritz, Feltz, Fahrbach, and Mack (2000) executed a meta-analysis of 45 studies and 102 correlations. The authors found that the average correlation between self-efficacy and individual performance in sport was .38, signifying a positive relationship with the former predicting the latter (Feltz, Short, & Sullivan, 2008).

Similarly, Felts, Chow, and Hepler (2006) found that high self-efficacy lead to enhanced performance, which in turn lead back to higher self-efficacy. Studies have also shown that self-efficacy is a stronger predictor of performance than anxiety (LaGuardia & Labbe, 1993; Weiss, Wiese, & Klint, 1989).

One interesting investigation that exemplifies the important role of self-efficacy was designed by Burke and Jin (1996). This study analyzed the value of different variables to predict total performance times in a Triathlon competition. There were physiological variables (e.g. height, weight, VO2max), history of performance and psychological constructs (e.g. self-efficacy, motivation and somatic anxiety). The results showed that when all variables were considered in the analysis, total performance was predicted more accurately by self-efficacy, performance history, and weight.

Therefore, promoting self-efficacy might well be an adequate strategy to optimize dancers´ performance. Enhancing the belief that dancers have in themselves could help them face tasks with more confidence. Thus, a greater confidence may lead to performance improvements such as movement precision, postural control or emotional expression.

Nevertheless, studies that use self-efficacy as an independent variable to predict performance should be cautiously designed. That is because, the main source of self- efficacy is past accomplishment, and therefore it makes sense that high self-efficacious individuals achieved the greater performances (Bandura, 1997). High successful athletes present greater probabilities to perform correctly again not only for the positive perception of themselves but also because they are physical, physiological and psychologically prepared to succeed. Thus, it would be a limitation from this study to consider self-efficacy as the responsible for leading them to win. In contrast, a bigger picture should be accounted because although self-efficacy has an important role in performance, at the same time past accomplishment is the main source of self-efficacy (Bandura, 1997).

See Appendix A for a summary of the previous analysed studies about the importance of self-efficacy.

Method to examine the effect of others in self-efficacy

Methodologically, it seems that the most common strategy to study the effect of verbal persuasion on self-efficacy regardless of real performance is via manipulated feedback. There are studies that provide manipulated feedback in order to influence the perceptions that individuals have of their own abilities. In general, results show that those who are said that have succeed tend to obtain better results (e.g. Hutchinson, Sherman, & Martinivic, 2008). Other studies opt for giving wrong information about the opponent. Thus, those who believe that are playing against the weakest rival presented higher self-efficacy levels (Weinberg, Gould, Yukelson, & Jackson, 1981). This induced self-efficacy sensation facilitates the separation between past performance and the ability perception in a specific task.

Bandura (1997) stated that the experimental strategy of providing manipulated or vagus feedback is an adequate method to measure the effect that self-efficacy presents on related variables such as performance. The next section will examine different studies that are based on the feedback effect on self-efficacy.

Feedback

When there is no external feedback during the learning process students have to trust on their interoceptive feedback which is the signals that the body send regarding to the adequacy of their movements. However, according to Sousa and Bandeira (1994) knowledge of results are expected to produce more significant positive effects in performance and learning than the exclusive intrinsic feedback from the individual. This means that external feedback, the one that is offered from external sources such as teachers, needs to be examined as it posses a remarkable value not only psychologically but also in terms of performance (e.g. Bandura, 1997).

Different methods of providing feedback are displayed in dance institutions. Specifically, this study focus on assessment feedback which is provided three times per course (December, February and June) in Spanish Dance Conservatoires. According to the organizational system, teachers are allowed exclusively to give students one score (from 0 to 10) to evaluate the dancers´ performance over the term. From non scientific evidence, it is known that most teachers provide lower-evaluated feedback, this is lower scores than teachers perceive that the students really deserve. Others use the opposite strategy and tend to give higher marks than deserved. These strategies are not based on research evidence and they are potentially used as methods for increasing effort or hypothetically motivate students.

Not every method of providing feedback produces the same effects (Sousa & Bandeira, 1994). The next section will describe studies from Social Cognitive Theory of Bandura (1977) that have analyzed responses that individuals present after receiving different kinds of feedback: ambiguous, upper-evaluated, lower-evaluated and accurate. Finally, this section will focus on studies that examine the influence of individual variables and the way in which people react to different feedback.

Feedback from the Self-efficacy framework

Non accurate feedback: unpredictability

Bandura (1997) states that if people are not given specific instruction regarding a task, they are not able to judge whether they have the necessary ability to perform it. This situational unpredictability might not allow them to anticipate the impediments and obstacles that they will have to face.

Throughout a school term, teachers provide students with a term score that represents their accomplishments in that given subject. However, within these scores students are provided with information regarding progress through the use of tests and modular assessments throughout the term designed to provide students with the necessary information and guidance in areas to improve. Students’ knowledge regarding both their results and the learning process might be generally higher than those in dance where a more overall score is given. Thus, it is expected that the impact of term assessment scores for dance students would be greater than for the school students.

The aim of the assessment score should represent what the students have achieved as well as presenting information regarding corrections and improvements for next term. However, providing exclusively one mark at the end of each term posses a low level of specificity about the movements they have to work on. For example, a dance student can excel in jumps but not have problems in controlling balance during pirouettes in their ballet class. Thus, a single score per term for ballet would not be sufficient as it would only provide general information without a definition and breakdown of criteria. Thus, only one score in a dance term might not be able to represent what the students have already achieved and those they should improve upon for the next term.

Lower and upper-evaluated feedback

Although Feltz, Short and Sullivam (2008) suggested that coaches have the persuasive power to enhance their athletes´ efficacy perceptions, it seems that the debilitating effects of this information are more powerful than the enabling effects (Bandura, 1997 as cited in Feltz, et al, 2008). In that respect, regarding on the feedback strategy teachers would be influencing self-efficacy in different ways. According to Feltz et al (2008) “feedback from coaches that emphasizes progress made tends to raise the efficacy expectations, whereas evaluations that highlight the shortfalls lower them” (p.10). As a result, some of them might come across what Seligman and Mayer (1967) called ‘‘learned helplessness spirals’’ which would decrease their self-efficacy (Cofee & Rees, 2011). This psychological phenomenon is originated when following failure/s on a given task, individuals believe that causes were beyond their control (uncontrollable causes) and unlikely to change (stable causes). This in turn might impact negatively on self-efficacy and performance promoting consequently higher learned helplessness.

As well as this, Bandura (1997) acknowledges that upper evaluated feedback – providing higher scores than individuals really deserve- is a method to increase self- efficacy levels. However, he suggests using alternative strategies such as accurate and objective feedback rather than promote a false perception in students. Research related to this topic have shown a correlation between the upper-evaluated feedback with distress and false expectations, dejection, depression, low control rates and low self- efficacy (e.g. Kim & Chiu, 2011; Nease, Mudgett, & Quiñones, 1999).

Bandura and Cervone (1986, as cited in Feltz, Chow, & Kepler 2006) suggested that there is another important variable to consider when upper-evaluated feedback is provided: task difficulty. Thus, the authors showed that when athletes received positive feedback after performing a difficult task, lower self-efficacy beliefs resulted when performers felt that they would not be able to repeat the same level of effort again. Task complexity might modulate the effect on the way positive feedback affects the subject. Thus, Bandura and Cervone (1986) would critic the strategy of indifferently providing- evaluated feedback without considering the subjects´ perception about the task difficulty.

Although effects of lower-evaluated feedback have been previously examined, lower-evaluated effects are more popular within general psychology (Bandura, 1997). In that respect, Ilies, Judge, & Wagner (2010) suggested that after receiving negative feedback an individual who has negative emotional reactions might recall previous experiences with negative feedback. As the memory is primed with these negative cues it would be expected that many dance conservatoire students experience reduced task- based self-efficacy, therefore leading to lower goals for improvement as the athletes of this study.

Supporting previous statement, Kim and Chiu (2011) showed that feedback influences future self-efficacy such that negative performance feedback (lower evaluated scores) tends to lead to lower self-efficacy in a verbal task. Similarly, Hutchinson, Sherman, and Martinivic (2008) found that a negative performance feedback after an isometric handgrip task lead to a decrease in the self-efficacy levels.

Consequently, if a teacher always gives lower marks than students really deserve, (e.g. students that repeatedly fail, regardless of how much effort they invest) it is very likely the dancers will tend to perceive low control and high stability of this feedback. They do not perceive they have the capacity to modify it as changing the feedback is outside their control and possess a high grade of stability. In short, after receiving lower-evaluated feedback students might not be able to detect which objectives have fulfilled and which should improved. Thus, according to research, uncontrollability and stability perceptions will make it difficult for students to increase their effort to enhance their mark, diverting “…attention away from diagnostic task information that might enhance one’s ability” (Kim, Chiu, & Zou, 2010).

Nevertheless, not only negative feedback might lead to non beneficial consequences in terms of performance and psychological wellbeing. Authors have found similar results after providing upper-evaluated feedback to their participants. Thus, upper-evaluated feedback might lead to unrealistic positive self-assessments and individuals might tend to protect their public self-image by engaging in self-defeating behaviours (Colvin, Block, & Funder, 1995 in Kim & Chiu 2011; Kim, Zou, & Chiu, 2010). For example, the level of excessively positive self-assessments of personality characteristics forecast more maladjustment and relational problems. Consequently, although the immediate effects of upper-evaluated feedback in terms of self-efficacy might be positive, according to the literature long term this would not be necessarily a beneficial strategy.

In that respect, Cofee and Rees (2011) demonstrated that the belief of controllability and stability of the feedback can be related to the consequences after providing manipulated feedback. The authors found that after providing negative feedback (achieving a total of just 6 with three darts), levels of self-efficacy and performance in the dart task were maintained, so long as individuals were led to believe that causes of failure were under their control and/or were likely to change. Conversely, a large negative effect on self-efficacy and a medium negative effect on performance were observed for those individuals who were led to believe that causes of failure were out of their control and unlikely to change.

Therefore, according to the previous results it would be essential to encourage dancers to think that they are able to enhance this score throughout their effort and behaviour. The next section will examine the expected effects after providing detail and accurate feedback to participants.

Objective feedback

Corrections are only valid and useful if the students understand where the problem is located (Kimmerle & Cote-Laurence, 2003). In that respect, “to help students learn technique, one must be as clear as possible in presenting the material and try to help students know exactly what is expected of them” (Kimmerle & Cote-Laurence, 2003, p. 171).

Schunk (1991) explained that many students often receive persuasive information suggesting that they possess the capabilities to perform a task (e.g., "You can do this"). However, the author states that providing upper-evaluated feedback enhances self- efficacy, but this increase will not last if individuals perform inadequately in future tasks. As an effective strategy of his intervention programme he recommended teachers to provide feedback per objective because that incorporated specific performance standards. Results showed that this strategy raised efficacy and motivation better than general goals (e.g., "Do your best").

If the subject has more information about their performance in a specific movement, it will be used as a basis to correct the next execution because knowledge of the results presents three main functions: incentive, reinforcement and information (Sousa & Bandeira, 1994). The quality of the information given to the subject after the performance, particularly the accuracy of knowledge of result, seems to have a positive influence on the learning process and leads to changes in the behaviour. In contrast, Schmidt and DeShon (2010) show in their study that as long as teachers increased uncertainty the negative self-efficacy effect was emphasized. The results suggest that in order to reduce the potential for negative self-efficacy effects teachers should diminish the vagueness surrounding one’s performance by providing detailed and timely feedback.

Bandura (1997) stated that coaches should have high-quality diagnostics skills regarding their athletes´ strengths and weaknesses in order to develop a strong sense of efficacy. When students received a score extracted by objectives instead of just a general score, it would be expected they focus their attention not only on the general score but also on what they have to enhance. Moreover, they might increase their controllability perception because they will understand that although they have to work on some objectives, they already have others (Cofee & Rees, 2011).

Additionally, Kim and Chiu (2011) investigated the negative and collateral effect of providing manipulated feedback to lead performers to perceive their performance as low or high. Authors expected that self-effacement and self-enhancement, as compared with accurate self-assessment would be associated with higher depression levels. To test this hypothesis, Kim and Chiu (2011) analysed 95 students on a verbal task. They estimated on a percentile rank how well their performance was compared to the performance of other students in their school. In addition, they indicated how many questions they thought they had answered correctly. Finally, the participants filled out a Depression Inventory and a Self-Esteem Scale.

Results showed that the extent of deflated self-assessments was related to higher depression levels—participants who lower evaluated their relative performance to a greater extent were more depressed. Also as predicted, the extent of inflated self- assessment was also related to higher depression levels—participants who over evaluated their relative performance to a greater extent were also more depressed. Kim and Chiu (2011) also found that both deflated and inflated self-assessments were negatively related to self-esteem, whereas accurate self-assessments were associated with the highest level of self-esteem.

However, the main limitation of the reviewed literature is that although we know the negative effects on self-efficacy after providing upper and lower evaluated feedback, previous studies did not examine and therefore suggest any alternative assessment strategy to adopt in terms of increasing self-efficacy.

This alternative should be specific to the activity in question because it is not the same to assess and provide feedback to maths and art students using the same system.

The suggested alternative in this study is the use of accurate scores to promote a real perception of the performance. This is because the literature seems to agree that it is more positive to have a real perception of their performance in terms of low depression and self-esteem levels than having self-defeated and self-inflated conceptions about ourselves.

Consequently, it is expected that if teachers provided real feedback per objectives, which match dancers own perceptions of their performance, it may increase their controllability sensation. Furthermore, if they receive accurate scores; it is very likely they changed throughout the three terms, therefore promoting an unstable attribution of the causes of this feedback. Thus, it is expected that accurate and objective feedback per criteria would increase the controllability sensation and decrease the stability attributions, causing positive and adequate self-efficacy levels (Cofee & Rees, 2011).

Other results: individual differences

Considering the literature review previously exposed, it seems that the most adequate option is to provide accurate feedback and per objectives (e.g. Kimmerle & Cote-Laurence, 2003). Low-evaluated feedback might demotivate and decline the individuals´ self-perceptions (e.g. Kim and Chiu, 2011). On the contrary, although upper-evaluated feedback promotes short term the greatest self-efficacy levels, the long- term effects have been found to be psychologically negative (Bandura, 1997).

However, there are studies that find variables which affect the individuals´ feedback responses. This is the reason why not everybody acts in a similar way after a certain kind of feedback. Personal variables play an important role in the feedback acceptance (e.g. Griffin, Predmore, and Gaines, 1987) and this caused many authors question the feedback relevance as a source of self-efficacy.

In this respect, for example, it has been found that some athletes trust more interoceptive feedback from their own bodies rather than the feedback received from external sources such as coaches (Sousa & Bandeira, 1994). In that order, it would be possible to learn without knowledge of result reinforced by the fact that the practice itself helps to develop an error detection mechanism. Therefore, assuming this phenomenon, it could be expected that different strategies of providing feedback will not make any psychological difference for dancers.

Swann, Griffin, Predmore, and Gaines (1987) explained that only when the external feedback coincided with the interoceptive feedback the subject accepted it as their own, whereas external feedback will not be considered when it contradicts the interoceptive feedback. Coming back to the dancer population, according to these studies, it could be suggested that perhaps they would have learnt through their experiences the interoceptives sensations originated from wrong and right movements. As consequence, there might be dancers who mainly trust in their own assessment mechanism above their teachers´ judgements.

Self-esteem is defined as individuals´ judgements of self-worth and feelings of self-satisfaction (Jussim, Yen, & Aiello, 1995). Research has found that self-esteem influences how feedback affects individuals. More specifically, those with low levels of self-esteem are willing to accept more responsibility for negative feedback and are more likely to perceive that feedback as accurate compared to high self-esteem individuals (Jussim, Yen, & Aiello, 1995). This means that if a negative feedback is provided to an individual with high self-esteem they will not believe it because this is incongruent with their self-perceptions.

As well, initial self-efficacy levels seem to play an important role in the way in which feedback affects individuals (see Appendix B for a description of related studies). Nease, Mudgett, & Quiñones, (1999) found that depending on the initial self-efficacy level individuals will have a different feedback acceptance. In this study, the author found that those with low levels of self-efficacy judged positive feedback on their performance as less accurate over time, whereas those with high self-efficacy judged the feedback to be accurate over repeated trials. Moreover, individuals with high self- efficacy were less accepting of negative feedback than low self-efficacious individuals, who do not show to vary in their acceptance of recurring negative feedback. This suggests that “people with high self-efficacy who receive repeated negative feedback become increasingly frustrated with the notion that their efforts are unsuccessful and may begin to doubt the accuracy of such information” (Nease, Mudgett, & Quiñones, 1999). Those with low self-efficacy accept all information that confirms to them that their performance has not been successful. In this sense, it seems that both high- and low-self-efficacy individuals understand feedback in ways that are protective of their initial self-efficacy.

These results suggested support for Self-verification Theory (Swann, 1987 as cited in Nease, Mudgett & Quiñones,1999). According to this theory, Nease, et al (1999) explained that people prefer to be considered in a manner that is consistent with their self-concept, even when their views of themselves are negative. Previous research on Self-Verification theory has found that “people tend to endorse feedback about themselves as valid only when that feedback fits within their conceptions of self” (Swann, 1987 in Nease, Mudgett & Quiñones, 1999). The results of this study suggested that individuals actively perceive their judgements to repeated feedback according to beliefs that they already possess. By simply letting somebody know whether they are performing successfully or not may not be interpreted in the manner in which it is proposed as there are other variables (e.g. self-efficacy) that influence on the feedback acceptance.

In line with the previous research and studying another very different population, Whyte and Saks (2007) added that “self-efficacy both directly and indirectly, explained a significant amount of variance in the individual decision to withdrawal from or escalate commitment to a failing project…” (pg 39).Thus, the authors supported Nease and colleage study (1999) and concluded that initial self-efficacy level influenced the relationship between negative feedback and behaviour.

Another example that shows how people respond differently to feedback regarding to their individual characteristics has been recently published by Ortín, Garcés de los Fayos, Gosálvez, Ortega, and Olmedilla (2011). The authors analysed the effect that optimism had on the response of negative feedback. This study consisted of a swimming test executed by 66 swimmers. After the first trial all the subjects received negative feedback. Interestingly, after such negative feedback, pessimistic individuals´ performances significantly decreased between the first and second tests. In contrast, although not statistically significant, results indicate that all the swimmers whose performance improved in the second test had optimistic profiles. This means that optimistic individuals presented a greater persistence in the face of obstacles and kept on making effort more than pessimistic swimmers. In conclusion, not every body responded with the same intensity after the same kind of feedback.

Therefore, analysing the feedback (assessment) effect in the dancers´ population is recommendable because it a lack of consensus around this topic. In short, the literature does not provide a clear conclusion about what is the most effective feedback in terms of self-efficacy or effort. It seems that there are variables that can influence that relationship however it is interesting to examine whether any feedback method lead to any kind of general benefit for a great amount of students. In addition, it is interesting to continue examining the feedback effect in this population as dancers display different and unique characteristics which distinguish them from the rest of athletes. Thus, acknowledging the repercussion, if any, different strategies of providing feedback has on self-efficacy is well justified based on current literature.

Self-efficacy and Effort

Previous literature in sport psychology has verified the predictor value that self- efficacy presents in behaviour. One of the most studied variables has been effort measured by for example muscular endurance (Weinberg, Gould, & Jackson, 1979) or retrospective questionnaires (George, 1994). Literature has shown that those who trusted their capabilities of executing a certain task more were also the ones that put more effort into it (e.g. George, 1994; Weinberg, Gould, & Jackson, 1979;).

The first study that analysed how the changes in self-efficacy influenced effort was based on a competitive muscular endurance task (Weinberg, Gould, & Jackson, 1979). Authors in this study manipulated the information they gave to the participants. In the high-self-efficacy condition, participants thought they were performing against a weak and injured person. Conversely, participants in the low-self-efficacy condition thought they were performing against a track athlete who displayed greater leg strength. As stated, the participants were asked what they believed their opportunities of winning were and how confident they were in the prediction. This study showed that people with greater perceptions of their own abilities were more efficacious and persisted longer than the low self-efficacy group. In the second part of the study, both groups were told they had lost the first trial. The results showed that in the second trial, those who presented higher self-efficacy extended their legs for a longer period of time, whereas their peers with low-self-efficacy extended their legs for a shorter period of time (Weinberg, Gould, & Jackson, 1979).

Similarly, George (1994) retrospectively asked athletes through a questionnaire about effort levels they perceived they had invested while hitting during each game. Subjects responded to the question, "How much effort did you put into hitting in today's game?" Responses were indicated on a scale divided into intervals of ten percentage points, ranging from 0 to 100% effort. The author reported path analysis results showing that higher self-efficacy predicted greater effort in baseball hitting. According to George, these findings were consistent with previous research demonstrating that athletes with a strong sense of efficacy invested greater effort in an attempt to master a challenge, whereas those with a weak sense of efficacy beliefs usually exerted less effort when addressing difficulties (Weinberg et al., 1979).

Additionally, other studies have examined the relationship between self-efficacy and related variables to effort. For example, McAuley, Wraith, and Duncan (1991) examined relationships among self-efficacy cognitions, perceptions of success, and multidimensional intrinsic motivation of 265 participants in aerobic dance classes. Similar to George´s study in 1994, highly efficacious subjects were more intrinsically motivated towards aerobic dance than were their less efficacious peers. Thus, dancers with higher self-efficacy were willing to invest a greater effort in their classes than dancers with low-self-efficacy.

Attending dance and physical activity classes require high levels of commitment and consistent effort (Barr-Anderson, Young, Sallis, Neumark-Sztainer, Gittelsohn, et al., 2007). According to the previous analysed studies, if individuals trust their own capabilities, it will be likely that they present high levels of commitment and effort in dance.

Ortega, Olmedilla, Sainz de Baranda, and Gómez (2009) demonstrated that athletes with high self-efficacy tended to perceive the competition as a challenge that could be overcome, enjoy competition, and invest greater effort when faced with obstacles. Also, when athletes lost the competition they recovered their perception of efficacy quickly resulting in active participation within the game.

In conclusion, retrospective studies have shown that individuals with higher initial self-efficacy levels put more effort into a task reaching a greater effort compared to low self-efficacious individuals (Appendix C summarise previous studies about the self- efficacy effect on effort). However, from our knowledge what it is unknown is whether the actual self-efficacy level with regards to a future task will be able to predict the intention that students have of exerting effort to improve this task. Knowing the predictive value of self-efficacy on future effort intentions is important for dancers because most often the time period from which they receive their assessment score until they actually invest effort to improve their movements is prolonged. Therefore, providing prospective data is one of the aims of the present study.

Summary

The practical interest of this current investigation, as mentioned previously, emerged from the need of contribute to an understanding of effective assessment strategy in terms of increasing self-efficacy of dance conservatoire students. The theoretical interest of this investigation is grounded firstly on the lack of studies that consider the influence of feedback on dancers. Secondly, this investigation is justified from our knowledge that there are no prospective studies examining the predictive value of self-efficacy on future effort intentions to improve a specific task.

An experiment study is developed with the aim of mirroring a real evaluation situation and answering the following research questions:

- Is there any difference between feedback conditions (upper-evaluated, lower- evaluated, accurate and objective feedback and no feedback) in self-efficacy levels?
- Does self-efficacy predicts effort intentions in dance students?

Hypothesis

Therefore, considering the analysed literature and the Self-efficacy Theory as a main theoretical framework (Bandura, 1997) two main hypothesis are stated:

Hypothesis 1: There will be differences between self-efficacy levels after the manipulation of participants´ feedback. Those who receive lower feedback scores than deserved will have the lowest self-efficacy levels. In contrast, those who receive feedback per objective plus the accurate general score are expected to have greater self- efficacy levels.

Hypothesis 2: Post-feedback self-efficacy will positively predict effort intentions.

3. Methodology

Participants

With the permission of the heads of a Dance Conservatory of Spain, every participant volunteered to participate except one female dancer who voluntarily refused to take part in the study after acknowledging the experiment outlines. Therefore, 61 participants (56 female, 5 male, mean age 17.53, s.d. 3.25) studying at this Institution took part in the study.

The dance degree consists of three levels: basic (4 courses), professional (6 courses), and superior (4 courses). All of the participants belonged to the Professional level (1st to 5th course) of the Flamenco and Spanish Dance (folk) specialities. Descriptive data for the participants can be found in Table 1.

Table 1: Frequency and Percentage of Participants according to their Age, Course and Specialities

illustration not visible in this excerpt

In addition to this, five flamenco teachers (4 females and 1 male) participated in the experiment. They agreed to provide real feedback to each student and write the score down on a recording sheet.

Procedure

This experimental study was approved by Trinity Laban’s Ethics Committee and aimed at replicating the term assessment in a Dance Conservatory. The feedback manipulation allowed the establishment of predictions and cause-and-effect relationships (Thomas, Nelson, & Silverman, 2005, p.20). The present study was designed to achieve a great ecological validity - high capacity to emulate the real world (Thomas, Nelson, & Silverman, 2005, p.5) - as the experiment took place in a real dance context and thus mirrored the habitual assessment system that participants were used to. Furthermore, the chosen protocol was used in the hopes of reaching internal validity –the results could be attributed to the feedback manipulation- and a high external validity –an adequate generalization of the conditions and the results to the real Conservatory system and dancers´ perceptions.

The study conducted by Nease, Mudgett, & Quiñones (1999) was a reference for the design of the present investigation as the researchers in that study found differences in self-efficacy after participants received manipulated feedback following only three trials during a one hour session. In spite of this, the exact protocol of this experiment is summarised in the eight following points:

1. Consent forms: during the week of the experiment, teachers asked parents of students younger than 18 years old to give their consent to participate in the study. The same day of the experiment, all the students completed their own consent forms which were required to participate (Appendix D).

2. Identification: the researcher randomly put labels on the students with identification numbers. The numbers indicated their experimental condition, which were meaningful to the participating teachers but not meaningful to the dancers, who were blind to the manipulation.

3. Explanation: after the identification numbers (ID) were assigned, the researcher briefly explained the experiment to the dancers. They were told that a) they would complete some questionnaires regarding their perceived ability to perform the zapateado (flamenco feet movement; the step they were being tested on), b) they would execute the zapateado individually three times, c) following each time, the researcher would give them an individual card with the score that their teacher has consider they deserved, d) finally, they would be asked to answer questions similar to those used in the questionnaire previously administered.

4. Rehearsing and pre manipulation questionnaires: teachers demonstrated the zapateado that the students were required to perform for the experiment. Participants practiced it for a 5 minute period. Then, students answered the self- efficacy questionnaire with regards to their perceived ability to perform the mentioned zapateado (see questionnaire description in the Instrument section). They signed by using their identification number in order to be anonymous.

5. Zapateado and Feedback: All students made a line and individually showed their movement to the teacher. Afterwards, the teacher wrote on a recording sheet the real feedback that student deserved. Immediately after each zapateado, the researcher who was seated close to the teacher, transformed teachers´ scores according to the students´ respective conditions. Students were lead to think that the researcher’s role was exclusively to copy the scores from the recording sheet to their individual cards. However, the researcher gave students their own individual manipulated scores, which they were not allowed to show anybody else.

6. Manipulation: with the aim of the manipulation criteria to be consistent for all students, a guide was created in order to transform the real feedback:

- Condition one: upper-evaluated feedback (only one mark). The researcher provided three scores more than the teacher gave them. For example, the scores of 7 were changed to 10.
- Condition two: informational feedback with a different score per objective (sound clarity, memory, rhythm and postural control) plus an accurate general score. These scores represented their real performance, on a scale of 0 to10. In this case, the researcher did not manipulate the scores. Some teachers admitted to averaging the four objective scores in order to give the accurate general score.
- Condition three: lower-evaluated feedback (only one mark). The researcher provided three scores less than the teacher gave them. For example, the score of 7 was changed to 4.
- Condition four: control group. They did not receive feedback after the execution of the task. They received a blank card.

The students repeated the same sequence three times under the same condition (e.g. upper-evaluated). Therefore, participants executed the first zapateado, received their mark and waited to be asked to perform the movement again. The students were not allowed to rehearse between trials, they could only think about their feedback.

Participants were told that scores ranged from 0 to 10 (as they were used to). However, in case someone acquired a score of more than 10 or less than 0, the teacher individually told them that they were out of the scale in either a positive or negative direction. In this study nobody received less than 1 although four people from the upper-evaluated feedback (condition 1) received a score higher than 10.

7. Post manipulation questionnaires: After the three trials, students were seated and asked to answer again the dance adapted Self-Efficacy Questionnaire and the Effort item (see description in the Instrument section). This time the students were asked to answer the questions whilst imagining that they were about to perform a real exam of the zapateado next week. Thus, they were asked to state their self-efficacy of executing the step correctly and their intentions of rehearsing the zapateado during the week.

8. Conclusion: Finally, the purpose of the experiment was explained to the whole class. The researcher answered questions and doubts and led a short discussion about assessments strategies that appeared within the groups.

Instruments

The present study used two questionnaires in order to collect data: a five-item questionnaire which measured self-efficacy (pre and post manipulation) and a one-item questionnaire which evaluated (their) effort intention.

The Self-efficacy questionnaire was adapted specifically for this study, based on the diving self-efficacy questionnaire (Feltz, 1982 in Feltz, Short, & Sullivan, 2008, p.44) (Cronbach’s α=.87, M=38.31, s=5.86) It was administered to evaluate the confidence that participants had for doing a specific zapateado. Questions in the questionnaire referred to four zapateado competence criteria). These criterions emerged from informal and non-structured interviews with five flamenco teachers from the same Conservatoire. In addition to the four criterions, there was a question asking for students’ general confidence in performing the zapateado as a whole. Questions were to be measured using a likert scale (1 to 10). The pre-manipulation self-efficacy questionnaire (Spanish and English version) can be found in Appendix F.

After the feedback manipulation, another self-efficacy questionnaire was administered (Feltz, 1982 in Feltz, Short, & Sullivan, 2008, p.44). Although the items were exactly the same, this time participants were asked to hypothetically imagine how they thought they would perform the same zapateado if they had to execute it again the following week (Cronbach’s α =.91, M=42.32, s= 5.40).

On the same sheet, the effort questionnaire was included. This was adapted from the George (1994) study where subjects responded to the question, "How much effort did you put into hitting in today's game?" Similarly to the George (1994) investigation, dancers´ responses were indicated on a scale divided into intervals of 10 percentage points, ranging from 0 to 100% effort. The difference was that whereas George (1994) referred to the effort that athletes perceived they had exerted in the game (past), in the present study it was asked to what extent dancers were willing to invest an effort in order to enhance the same zapateado (future). Whereas George´s test properties for past effort in athletes were (M=8.51, s=0.71), the adaptation for this study, future effort intentions of dancers was (M=8.01, s=2.19). An example item of this questionnaire was “How confident you think you are that YOU CAN perform the zapateado next week with an appropriate sound clarity?”

Statistical Analysis

To begin with, a normality test was executed in order to check whether data followed a normal distribution. To test the first hypothesis, it was analysed whether there was a difference between the feedback conditions and the resultant self-efficacy throughout a One-way ANOVA analysis (parametric) or Kruskall Wallis (non parametric). This aimed at showing which kind of feedback (independent variable with four factors) provided greater self-efficacy (dependent variable). The experiment design allowed for differentiation between the real skills of a dancer that possibly could be related to a greater self-efficacy and the participants’ perception of these skills (self- efficacy). These perceptions were the initial results of the manipulated feedback by the teacher and not a result of past accomplishments.

In addition, a Linear Regression Analysis was executed between the self-efficacy (post trials) and the planned effort item. Thus, it was tested whether the role of self- efficacy was a predictor of the participants’ future efforts.

Self-efficacy data was also collected before the trials to test whether there were any differences between the conditions prior to the feedback reception. This was essential in acknowledging that the conditions were randomly assigned to the participants so the expected difference in self-efficacy after the treatment was due exclusively to the feedback (independent variable).

Ethical Considerations

Based on Nease, Mudgett, & Quiñones (1999) study in sport, this investigation reduced the trials to three and the period of time to a one hour session in order to emulate a previously approved, published study. Nevertheless, the fact that students were receiving manipulated feedback could have also promoted negative effects expressed by demotivation and frustration. Although previous studies under similar conditions did not mention any particular consequences, after explaining the aim of the investigation, some measures were taken as preventative strategies:

A parental consent form was required to participate in this study for those younger than 18 years old (Appendix E).

- The participants were classmates thus promoting a potentially more accepting atmosphere.
- The experiment mirrored a very typical exercise in flamenco training within Conservatoires therefore it was not expected to facilitate anxiety as a result of being a dance step that was unknown.
- The provided feedback was exclusively quantitative avoiding comments either positive or negative regarding the individual or the quality of their movement.
- It was highlighted several times that the feedback was only going to be used as part of the investigation, and teachers would not consider it for the real term assessment.
- The feedback was personally and individually provided, so students were not able to make comparisons between themselves in a competitive manner.
- The process was not video recorded to avoid unnecessary performance anxiety.
- At the end of the session, students were informed that this was an experimental study and they had received a non-real score in order to test for any difference in the questionnaires.

Furthermore, according to Thomas, Nelson, & Silverman, (2005) this study considered the four participants´ rights—“the right of non-participation, the right to remain anonymous by using the ID number for the questionnaires, the right to confidentiality and the right to expect experimenter responsibility by informing them the study purpose immediately after the completion of the experiment” (p.84).

4. Results

Preliminary Analysis

A Kolmogorov-Smirnov K-S normality test was performed to check the data distribution. The data of the three main variables was non-normal: Self-efficacy pre- manipulation D(61)= 0.21, p<.05, self-efficacy post-manipulation D(61)= 0.16, p<.05, and future effort intentions D(61)= 0.20, p<.05. Therefore, non- parametric tests were performed.

In addition to recognizing that the data did not follow a normal curve, another question explored was whether the groups were homogeneous in terms of self-efficacy before the feedback manipulation. A Kruskal Wallis test demonstrated that there was no significant difference between the four groups H(3)=1.74; p=0.63; µ2=0.29. Specifically, participants in condition 1 (upper-evaluated feedback) presented a mean rank= 31.43; condition 2 (objective feedback) mean rank= 27.39; condition 3 (lower- evaluated group) mean rank =30.90; and condition 4 (no feedback) mean rank= 35.62. Although the groups did not present significant differences, the mean rank per condition differed considerably among them. For example, people in the objective groups presented lower initial self-efficacy levels than the other conditions. This fact lead to the inclusion of paired difference analysis separately per condition in order to examine whether there was any difference in self-efficacy between pre and post manipulation.

Table 2 shows Spearman´s Correlations between the three variables involved in this study. According to the data, there is a significant and positive relationship between self-efficacy pre and post manipulation. However, it has not been found any correlation between self-efficacy (both pre and post) and effort intentions.

Table 2: Spearman Correlations between Pre Self-efficacy, Post Self-efficacy and Effort Intentions

illustration not visible in this excerpt

* p < .05. ** p < .01.

Hypothesis 1

To test the first hypothesis, a Kruskal Wallis test (non-parametric test), equivalent to One-way Anova (parametric test) was executed. The hypothesis 1 predicted that there would be differences in self-efficacy post test depending on feedback. Those who received lower feedback scores than they deserved would have the lowest self-efficacy levels. In contrast, those who received upper-evaluated and accurate feedback per objective were expected to have greater self-efficacy levels. Non changes were expected to find in the non-feedback condition. (Please, find SPSS outputs in Appendix H)

Data analysis showed that opposite to the initial hypothesis, there were no statistical differences in self-efficacy post manipulation between feedback conditions H(3)=2.74; p=0.43; µ2=0.05. Specifically, participants in condition 1 (upper-evaluated feedback) presented a mean rank= 36.23; condition 2 (objective feedback) mean rank= 29.44; condition 3 (lower-evaluated group) mean rank=26.33; and condition 4 (no feedback) mean rank= 32.50.

Although statistical differences were not found between groups, a Wilcoxon test (paired sample t-test) was performed to analyse whether there was any change between pre and post feedback manipulation in self-efficacy levels within subjects. Data showed that after providing upper evaluated feedback self-efficacy levels were significantly higher (Mdn=9) than before (Mdn=8) z= -2.88, p=.004, r=.37. Similarly, feedback per objective promoted significantly higher self-efficacy levels after the manipulation (Mdn=8) than before (Mdn=7) z=-2.77, p=.006, r=.35. In contrast, those who received lower evaluated feedback scored did not present significant differences in self-efficacy between pre (Mdn=7) and post manipulation (Mdn=8) z= -1.27, p=.21, r=0.16. Finally, individuals from the control group significantly increased their self-efficacy levels z= - 2.44, p=.01, r=0.31 although no fluctuation in the median values between pre (Mdn=8) and post (Mdn=8) was seen.

In addition, further descriptive data is provided in order to explain more accurately the results. Figure 1 shows the percentage of participants who increased, decreased and remained constant their self-efficacy values between pre and post manipulation

Feedback Effect on Self-efficacy

illustration not visible in this excerpt

Figure 1: Percentage of participants per condition who increased, decreased and remained constant their self-efficacy values between pre and post manipulation.

Upon observation of Table 3 it can be noticed that condition 1 and 2 (upper and objective feedback) produced the highest increases in self-efficacy levels in a large percentage of participants. On the contrary, lower-evaluated feedback, as was hypothesized, lead to an increase of self-efficacy levels in a reduced number of individuals compared to condition 1 and 2 and even condition 4 (control group). The percentage of participants who decreased self-efficacy levels was similar in the objective group and the lower-evaluated group. However, none of the participants from the upper-evaluated group and from the control group diminished their self-efficacy. Self-efficacy levels remained constant in the control group where a large number of participants did not experience any change between before and after the experiment.

Table 3: Percentage of Participants that Increased their Self-efficacy Levels in Post comparing to Pre Manipulation

illustration not visible in this excerpt

It was observed that 64.7% of participants who received objective and accurate feedback increased their self-efficacy levels. Objective feedback is followed by upper- evaluated feedback with 62.5 % of participants reporting greater rates in this variable. Those students who did not receive feedback (control group) still increased their self- efficacy to 46.1%, this could be due to the fact that regardless of not receiving feedback they still believed they had a week to practice. As expected, only 33.3% of individuals from condition 2 (lower-evaluated) raised their perceived abilities to do the same zapateado next week.

Therefore, although there was no significant difference between groups in self- efficacy post manipulation, there were significant differences in condition 1, 2 and 4 between pre and post manipulation. Every condition displayed a significant increase in post self-efficacy levels compared to pre except for lower-evaluated feedback where significant differences could not be found. Although there were significant differences in self-efficacy between pre and post in the control group, median values did not vary as they did in upper and objective conditions. The expected trend was shown throughout descriptive analysis; therefore upper and objective feedback significantly increased self- efficacy levels whereas lower evaluated feedback did not.

Second Hypothesis

The second aim of the present study was to understand the predictive value that post self-efficacy presented upon the effort variable. Contrary to the hypothesis, self- efficacy levels did not predict effort intentions (F=.059; p>.05; R2 =.001; ß=.032). Neither changes in self-efficacy (pre to post) predicted effort intentions (F=.150; p<.05; R2 =.003; ß=.05). Nor were there significant differences between feedback conditions and effort intention, (H(3)=6.39; p=0.94; µ2=0.81). Thus, it seemed that others variables may have influenced the relationship among self-efficacy and effort intentions.

Therefore, as complementary information to the purpose of the present study, it seemed interesting to analyse how self-efficacy pre modulates the relationship between feedback condition and future effort. Considering the self-efficacy pre-manipulation percentile division (100-75%, 75%-25%, 25-0%), participants were divided (post hoc) into three groups: high self-efficacy group (self-efficacy higher than 8 score), medium self-efficacy group (7 and 8 scores), low self-efficacy group (less than 6 scores).

Figure 2 shows the mean ranks of the effort variable per self-efficacy group depending on their received feedback. Note that these results can not be generalised because for one, they are descriptive and the difference between self-efficacy pre and post manipulation are not related to effort (Pearson r=-.17, p>.05, r2 =.03). Furthermore, the number of people per condition was not controlled thus proving insufficient to provide any conclusion.

Effort Intentions, Feedback and Initial Self-efficacy

illustration not visible in this excerpt

Figure 2: Mean ranks of effort intentions according to their initial self-efficacy levels (High, Medium and low SE) and their received (upper, objective and lower evaluated) feedback.

Note. Upper-evaluated feedback (3 scores more than students deserve); Objective feedback (accurate scores per objective plus general score); Lower-evaluated feedback (3 scores less than deserved).

As Figure 2 shows, those individuals with medium self-efficacy (7 and 8) were the individuals who were willing to invest the greatest rates of effort in every feedback condition. In contrast, it seems that scoring either extremely high or low in pre self- efficacy lead individuals to similar rates of effort intentions usually lower than in medium self-efficacious participants.

After analysing data within self-efficacy groups, it was observed that the high self-efficacy group presented higher effort intentions levels (mean rank=9) when they received upper-evaluated feedback. On the contrary, when they received a lower score than deserved subjects expressed less effort intention (mean rank=4.3).

Medium self-efficacy pre group (7 and 8 scores) presented greater effort intentions when they received a real feedback divided per objectives (mean rank=20.38) followed by upper-evaluated feedback. However, similar to the high self-efficacy group, receiving a lower score than deserved affected the effort intention in a negative manner. Lastly, the low self-efficacy group (less than 6 score) presented greater effort intentions when they received lower evaluated feedback (mean rank: 6.67) rather than feedback per objectives and upper evaluated.

Therefore, although any statistical difference has been found due to the small and impaired size of participants per condition, it can be expected from the descriptive data that students’ self-efficacy of a task may influence how they respond in terms of effort to the received feedback. In addition to this, it seems that presenting a self- efficacy of 7 and 8 out of 10 promoted the greatest effort intention rates for every kind of feedback.

5. Discussion

The main practical interest of this investigation stemmed from the need to provide teachers with a term assessment strategy that promoted positive self-efficacy levels in students. The theoretical interest was stimulated by the lack of studies that analyse the relationship between feedback and self-efficacy, and self-efficacy and future effort intentions in dancers and athletes.

Results have shown that in general a great percentage of participants in this study increased their average self-efficacy levels after the manipulation. However, there was a significant difference for those individuals who received upper-evaluated and objective feedback and not for those who received lower-evaluated scores. Oppositely to the second hypothesis, the self-efficacy level did not predict future effort intentions, thus there was no association between high self-efficacy levels and greater effort intentions. As a complementary evidence and supported on descriptive and not inferential analysis, it was observed that possibly initial self-efficacy levels could have modulated the effect that different kind of feedback have caused in the effort intentions.

Feedback effect on Self-efficacy Levels

Based on previous literature it was predicted that receiving lower evaluated feedback would impact negatively in the dancer’s belief in correctly performing the movement (e.g. Kim & Chiu, 2011; Vance & Colella, 1990). Alternatively, it was predicted that providing a real and detailed feedback would increase the probability of scoring higher in self-efficacy (Schunk, 1991; Weinberg, 2003).

However, results did not statically support the initial hypothesis; data did not show significant differences among the feedback conditions. These results could have been explained by the discrepancies that groups presented in initial self-efficacy values. Thus, although feedback manipulation had produced remarkable differences between conditions, it is likely the analysis had not detected them. For example, although objective feedback would have produced a great increase between self-efficacy pre and post manipulation, we could have not detected it because participants from this group presented lower initial self-efficacy values than the other groups. This fact suggested the need of performing a paired sample test to compare differences between pre and post per condition.

Differences between pre and post manipulation per feedback condition

Non feedback

Every condition had a certain percentage of participants who increased their self- efficacy levels after the manipulation. This could be because the post self-efficacy questionnaire asked for the perception the students had of their capability to execute the zapateado correctly after a whole week of rehearsals. Thus, it make sense that people scored higher in post-test when they thought they would have time to improve the task than when they had to immediately perform the task with no rehearsal time. However, although in the upper-evaluated and objective condition the median self-efficacy value increased a score between pre and post-test, in the control group the median values kept constants.

If instead of asking for perceptions of future self-efficacy, the questionnaire had asked for immediate self-efficacy after the manipulation, it could be expected that the percentage of participants who increased their self-efficacy would be lower than the percentage found in this study. Possibly, more participants would have scored lower or identical values of self-efficacy in the post-test than in the pre-test.

Comparing the post self-efficacy values from the upper-evaluated and the objective group to the control group, it seems that the fact of providing no feedback did not produce positive effect in self-efficacy. This finding would also justify the relevance of analysing the most beneficial strategy of providing feedback in dance context and the moral responsibility of presenting this alternative to the institutions.

Lower evaluated feedback

Whereas upper-evaluated, objective and control condition significantly increased their self-efficacy pre to post, participants who received lower-evaluated feedback did not increase significantly their self-efficacy levels. It seems that providing lower scores than deserved promoted a great percentage of participants who did not trust their capabilities for performing correctly even after having a whole week to rehearse. Thus, considering the Self-efficacy Theory (Bandura, 1997), negative feedback would undermine one of the most important sources of self-efficacy –verbal persuasion. In that respect, when debilitative verbal persuasion is offered, self-efficacy level could decrease as occurred in a high proportion of subjects in this study.

According to Seligman and Mayer (1967) when subjects receive continued lower evaluated feedback, this impacts negatively upon self-efficacy. Causal attributions could be playing an important role in this result (Chase, 2001; Cofee & Rees, 2011). It is possible that, people under this condition could have perceived a high level of stability in their scores due to the similarity throughout the three trials. Moreover, although they thought they were performing as they used to, they could have felt a low level of control because they were receiving less score than usually. Modifying and increasing their marks was out of their control. Therefore, data of the present investigation would justify the examination of dance students´ causal attribution to find out if high stability and lack of control are predictors of low self-efficacy levels (Cofee & Rees, 2011).

To summarise, as the Theory indicated it seems that providing below evaluated feedback would lead to a decrease in the self-efficacy levels. Self-efficacy perceptions are extremely important to promote because they have been strongly related to cognitive process such as goal setting, emotional state such positive mood after failure and behaviour for instant effort. Providing below evaluated feedback would negatively affect self-efficacy producing undesirable effects in variables related to performance and wellbeing. This study provides information about the counterproductive effects of such a habitual practice in the dance conservatories. Giving lower-evaluated scores far from benefit students, reducing self-efficacy in a great percentage of dancers. Therefore, this experimental study would be useful to demonstrate to teachers that false reporting of students’ abilities does not promote, not only, self-efficacy but many of the variables related to this such as; goal setting and anxiety (Feltz, Short, & Sullivan, 2008).

Feedback per objectives.

Results showed a significant increase in self-efficacy after objective feedback. Therefore, upper-evaluated feedback is not a unique strategy in increasing self-efficacy in dance population. As predicted, accurate feedback per objective plus the general score instead of only a general score increased self-efficacy levels of many subjects (Weinberg, 2003). The reason behind this could be a greater control sensation as students would know how to guide their rehearsal and enhance their performance (Kim, Chiu, & Zou, 2010). It can be inferred that this feedback (accurate and specific per objectives) might lead them to think that to improve is feasible and that they were responsible for changing what they were doing wrong. Knowing those aspects that need improvement as well as those that they are doing well, might help them to set their own goals and consequently to increase the confidence of success (Bandura, 1977). According to Buckroyd (2000) “ if students are aware of what in particular they are being marked on and how each of the criteria is graded, then they have the opportunity to see where their strengths and weaknesses lie and consequently to direct their efforts to improve much more precisely” (p.90).

Results suggested that the strategy of providing one score per term objective (condition 2) would be more beneficial than either a lower-evaluated score (condition 3) and non-feedback (condition 4). Interestingly, upper-evaluated feedback (condition 1) produced the same positive change pre to post than an accurate and real feedback per objectives. Therefore, providing objective feedback plus the conventional general score would be a resourceful alternative that the teacher could implement in their assessment dynamics.

The strategy of providing objective feedback is not only efficient for students, but also for teachers (Buckroyd, 2000, p.91). Once they had already thought the marks per objective, teachers recognised to assign the general score in the objective group with greater confidence than when they had to directly provide one general score. Therefore, this strategy would help to be more unbiased and thus decrease the subjectivity levels which are traditionally associated to the assessment in dance (Buckroyd, 2000, p.92).

The conventional general score in the objective condition was also given but in this case as a complement of the objective marks. Institution requirements ask teachers to decide one general score per student which summarise the term performance. This score from 0 to 10 is used to differentiate from those who pass (above 4) to those who fail (below 5). Therefore, the objective alternative that this experiment suggested did not reject this option. Instead additional marks per objectives were designed to complement this general mark. Results seem to suggest that simply the fact of knowing the specific objectives behind this general score lead student to increase their self-efficacy levels.

The difficulty of this kind of feedback strives of knowing the adequate number of objectives that complement the general score. For example, four objectives have been selected for assessing a zapateado in the levels studied, but there are quite more objectives along the term that could be assessed within a subject. Therefore, future research should focus on examining the most adequate number of objectives to avoid unexpected effect after receiving long list of scores. It is possible that for instance 50 scores per term are a large number so students can positively respond in term of self- efficacy and goal setting.

Upper evaluated feedback

Upper-evaluated feedback also promoted significant increases in self-efficacy. In addition, 62.5% of participants increased their self-efficacy level compares to control group (46.1%). Previous studies that compares upper against lower evaluated feedback also found that upper-evaluated feedback predisposed to greater self-efficacy levels. However, it is important to question whether high self-efficacy levels directly lead to an optimum performance and psychological well-being such as depression and self-esteem (Kim & Chiu, 2011; Nease, Mudgett, & Quiñones, 1999).

According to Vancouver (2002 in Feltz, Short and Sullivan, 2008) “efficacy beliefs can be self-debilitating when measured on an intraindividual basis across time. High self-efficacy can result in a decreased performance because an individual may become overconfident and allocate fewer resources to the task at it is repeated across time” (pg 34). Thus, according to this author, when subjects have high self-efficacy levels they might be overly optimistic about the degree to which they are meeting their goals. This belief would decrease the difference between goal and perceived performance, and thus persons would display fewer sources to meeting goals (Feltz, Short, Sullivan, 2008).

According to these studies, it is possible that participants who have received upper-evaluated feedback felt so satisfied with their performance that they did not invest so much effort to improve it. Thus, this false perception could have lead them to exaggerate their abilities and therefore, to set inappropriate goals. Consequently, it is likely that this upper-evaluated group was not as prepared as the objective group to perform well next week. This hypothesis should be examined in future research.

Furthermore, according to previous literature having excessively high self- efficacy levels might promote dejection and depression (Nease, Mudgett, & Quiñones, 1999; Kim & Chiu, 2011). Peer comparison or receiving contrary feedback from different teachers could lead students to realise that the feedback does not correspond to the reality. This incongruence could affect their confidence and long term it could have devastating effects on the self-efficacy.

Hypothetically, if participants had not known that the assessment was part of an experiment and there were no real scores, they would have tended to compare their score with each other. As well, it is expected that the incongruence between the reality and their manipulated scores had promoted a lack of control sensation and insecurity with their scores which did not facilitated an appropriate actuation plan to improve the zapateado for next week. Therefore as Bandura (1997) suggested, providing upper- evaluated feedback is not an adequate strategy to increase self-efficacy preferring a real and detailed feedback for that purpose instead.

Results presented in Table 2 support the recommendation of providing objective feedback as the percentage of dancers that increased their self-efficacy was to some extent greater in the objective group (64.7%) than in upper-evaluated group (62.5%). This indicates that although in absolute terms the self-efficacy average from the upper- evaluated group was greater, a higher number of participants increased their self-efficacy levels under the objective condition. Due to the negative long term effects of providing upper-evaluated feedback, higher self-efficacy increases in a few participants might be less beneficial than a slight increase but homogeneous in more participants (Kim & Chiu, 2011; Nease, Mudgett, & Quiñones, 1999).

This study highlighted the positive, significant differences in self-efficacy pre and post in every condition except lower evaluated group. Thus, demonstrating the negative effect of providing lower scores on students’ self efficacy. This study will help to challenge the assessment believes of many teachers, who often present lower evaluated feedback in order to increase effort levels of their students and would provide descriptive evidence of the damaging effects of lower-evaluated feedback.

Self-efficacy Predicts Effort Intentions

The second hypothesis was focused on the predicted value that the self-efficacy variable presented upon the effort intentions. Previous studies have found that self- efficacy and effort (behavioural variable) were related and that self-efficacy perceptions predicted the effort that athletes invest to reach their goals (Weinberg, Gould, & Jackson, 1979; George, 1994). However, no previous study examined the predictive value that self-efficacy presented on a future effort intention.

Results did not confirm the initial hypothesis, thus dancers with greater self- efficacy levels were not necessarily those with higher effort intentions. This finding does not coincide with the Self-Efficacy Theory as this framework would expect greater effort intention (cognitive variable) as a consequence of higher self-efficacy (Bandura, 1997). Thus, theory would predict that effort intention as a cognitive variable to be positively related to self-efficacy. Those with higher levels of self-efficacy would present greater effort intentions. However, this study seems to show that there are modulating variables that affects this relationship such as the different understanding of the effort variable. In that respect, although some students might have considered that investing a great effort is associated to positive qualities and optimum performance, others might have thought that only those who have not the necessary capacities are the ones that need to exert effort. According to this, it could be expected that the most confident individuals had considered that they would need less effort to invest than those with medium and low self-efficacy levels.

Neither self-efficacy predicted effort intentions nor feedback predicted significant differences in the effort intentions. Therefore, the reasoning that some teachers have about providing lower-evaluated score than deserved aim to improve effort intentions for next term is not supported from the data. It seems that effort intention and feedback are not strongly related and there are variables that modulate that relationship. Further studies in this line would be needed in order to find out which variables teachers need to efficiently promote to increase effort intentions.

Several variables could have influenced the lack of significance of this prediction. First of all, it is possible that the regression was not entirely linear; medium self-efficacy levels might have caused the greater effort intention and not the highest or the lowest ones. As Vancouver (2001 in Feltz, et al 2008) suggested, extreme self- enhanced values of self-efficacy after receiving upper-evaluated feedback could have facilitated a low interest for investing effort and improve the task for next week.

Secondly, whereas some of the low self-efficacious dancers could have felt demotivation for investing greater effort because they might have thought that increasing the effort level would never lead them to improve their performances (McAuley, Wraith, Duncan, 1991); others could have thought that investing high levels of effort would be the solution to perform better next time (George, 1994). Also, it is possible that while certain dancers with high levels of self-efficacy might have thought that they did not need to invest effort to improve a task in which they are good at, others felt that there is always something to improve and effort is the only way to reach the perfection. Thus, the same level of self-efficacy could have originated different responses in terms of effort depending on further individual differences (e.g. Jussim, Yen, & Aiello, 1995).

Another possible explanation could be related to a different perception of the experimental situation. In that respect, due to ethical considerations researcher asked participant to imagine and not to believe that they would have to do the zapateado again next week. As a result, some of the dancers could have taken the experiment very seriously imagining the hypothetical case of repeating the same zapateado after a rehearsal week, whereas others could have had more difficulties imagining this hypothetical situation. Then, it is very likely the attitude towards the experiment could have interfered how students have responded to the effort item.

The previous possible explanations suggest that alternative questionnaires should be administering in future studies. Instruments with higher amount of items and a better recreation of the hypothetical situation could help to respectively increase the reliability and validity of the experiment. In order to design a more ecologically valid study, it would be useful to study variables that can in some way be identified as effort. For instance, number of hours rehearsing, frequency of repetitions during class, etc. Thus, we could examine whether students with different self-efficacy levels present differences in these variables..

The effect of initial self-efficacy levels on effort intentions following feedback

Regardless of the type of feedback, individuals with medium self-efficacy levels (7 to 8 out of 10) were those who were willing to invest the greatest rates of effort to improve the zapateado for next week. However, it seems that scoring either extremely high (9 and 10) or low (less than 6) in initial levels of self-efficacy lead individuals to similar rates of effort intentions, which appeared to be lower rates of effort than seen in medium self-efficacious participants. Therefore, it might suggest that although upper- evaluated scores might promote exaggerated self-efficacy increases, it might not be the best strategy when teachers pretend to encourage students for making greater effort (Bandura, 1997; Schmidt & DeShon, 2010).

Upper evaluated feedback

After analysing the graph in detail, it appears that students responded differently to the feedback regarding to their initial self-efficacy levels. Upper-evaluated feedback promoted the greatest effort intentions within the highest self-efficacious groups. On the contrary, providing higher scores than students really deserved created lower effort intentions in the low self-efficacious group. This result can be related to the congruence and acceptance degree (e.g. Nease, Mudgett, & Quiñones, 1999). In that respect, Self-Verification Theory (Swann, 1987) would suggest that high self-efficacious dancers might perceive positive feedback congruent with their own capabilities and would thus allow them to accept the feedback and respond positively to it. This positive feedback would reaffirm the high beliefs in themselves, and though it is suggesting that they are above their habitual confidence, it would appear to promote high rates of effort intentions for improving their actual performance.

This procedure may not work with low self-efficacious dancers however. Dancers who had lower levels of self-efficacy who then received upper-evaluated feedback may perceive the feedback as less congruent with their own perceptions, their interoceptive feedback, and with their past accomplishments. These variables could have affected the acceptance therefore diminishing the desires of exerting effort for improving the movement. It therefore appears that providing upper-evaluated feedback could have either positive or negative effects depending on initial self-efficacy levels.

Nevertheless, although it seems that initial self-efficacy levels were exerting an important role on how people react to upper-evaluated feedback, Self-Efficacy Theory did not recognise individual differences in the way people responded to it. Thus, when Bandura (1997) described the self-efficacy sources (e.g. vicarious experience) and associated them to cognitive, behavioural and emotional variables he did not consider individual differences that can affect how people respond to these sources.

Consequently, although data was descriptively analysed results suggested the influence of individual differences. Thus data would support those studies that highlight the role of variables such as optimism (Ortín, Garcés de los Fayos, Gosálvez, Ortega, & Olmedilla, 2011), initial self-efficacy (Ortega, Olmedilla, Sainz de Baranda, & Gómez, 2009) and self-esteem (Jussim, Yen, & Aiello, 1995) on the way people responded to this. For example, according to Ortín, et al (2011) the more pessimistic students would respond more negatively to lower-evaluated feedback than those more optimistic. Therefore, due to the reduce sample size in this study, individual differences in optimism might have contributed to the lack of significant differences found between feedback conditions.

Lower-evaluated feedback

Although previous studies have shown that negative feedback did not promote immediate greater effort in low self-efficacious individuals (George, 1994; Weinberg, Gould, & Jackson, 1979), the present study provides evidence that lower-evaluated feedback may have had a more positive effect with low-efficacious individuals. This may occur because the feedback would be congruent with their thoughts about themselves (Nease, Mudgett, & Quiñones, 1999) which could therefore encourage greater investment in future efforts. Nevertheless, this finding deserves a deeper analysis as this contradicts previous literature (Vance & Colella, 1990).

Lower-evaluated feedback may not have the same effect on high self- efficacious individuals however. When high self-efficacious individuals received lower-evaluated feedback, they presented the lowest levels of effort intention among every group and condition. It therefore appears that providing lower-evaluated scores to those with high self-efficacy would not encourage them to work harder, as many coaches believe. In fact, by doing this, teachers would actually reduce their effort intentions to enhance performance. Contrary to this study however, previous investigations have shown that high self-efficacious athletes who receive negative feedback invested immediate greater effort as the feedback encouraged them to excel themselves (Weinberg, Gould, Yukelson, & Jackson, 1981). Nevertheless, it seems that this study based on dancers can not confirm this previous evidence as the tendency is inversed. This conclusion should be taken cautiously as it has been said previously; there were a small number of participants per condition and self-efficacy level. However, results are useful as a finding to justify future inferential studies in this area.

The time between when the participants received the feedback and the hypothetical behaviour (greater or lower effort levels) might have played an important role on the response. Thus, it is likely that the known inclination to work harder after receiving negative feedback reported in previous studies was more behavioural than cognitive. For instance, Weinberg, Gould, Yukelson, and Jackson, (1981) reported that after negative feedback, high self-efficacious athletes presented higher actual leg strength than the low self-efficacious counterparts. However, the participants from this present study responded differently. Lower-evaluated feedback leads to high self-efficacious participants decreasing in their willingness to improve the task in the future. Whether this assumption is true or not, it requires additional studies which consider the time period that students have to respond to the task.

Feedback per objective

According to the descriptive data, feedback per objective was the most effective strategy in students with medium self-efficacy levels. Indeed, according to the results, individuals with medium self-efficacy levels and accurate feedback per objective promoted the highest levels of effort intention as compared to other conditions. Due to the greater percentage of students with medium self-efficacy levels (50%) than low (25%) and high (25%), it would be reasonable to consider the objective feedback as the most convenient strategy with which to provide feedback. In contrast, the less recommendable feedback strategy for both high self-efficacious (25%) and for medium self-efficacious (50%) students would be the lower-evaluated feedback. Thus, results suggest avoiding providing lower scores than students deserved, as this strategy did not encourage the effort intentions of the majority of students.

These conclusions, however, should be taken cautiously. First, results only showed tendencies and not significant differences. Moreover, confounding variables have not been controlled as this was a post hoc question. Future studies about this topic would be necessary to determine whether the objective feedback possesses the appropriate advantages in order to include it within the assessment dynamic of the dance conservatoires.

Limitations

There are several limitations in this study that should be highlighted to ensure careful consideration of the results. First of all, there were a reduced number of participants per feedback condition. An average of 15 people per group might be not enough to show significant differences between groups. When a sample size is small, individual differences are not compensated and as a result, the effect of the independent variable (feedback) might not be as significant as expected (Field, 2009, p.58).

The feedback conditions designed in this study should be considered as different but not necessarily opposite. We should have created more conditions to distinguish between specificity (scores extracted by objectives vs only a general score) and accuracy to the reality (upper and lower-evaluated vs real feedback). However, due to the sample size, only four conditions were examined to explore any differences among current practices (upper and lower-evaluated scores) and a new alternative.

Additionally, in order to interpret the results most effectively, time in the term should be considered. At the time of the study students were in their official assessment week which added to their general rehearsal schedules. Because both teachers and students seemed to display high levels of stress, it may have not been the most adequate period to perform the experiment. This might lead to different consequences: a) lack of motivation and interest from teachers to provide objective feedback to the students, as there were four criteria (thus more time thinking about the performance quality); b) due to the fact that students did not know that the received score would not count for their real assessment, it is possible that lower-evaluated feedback would not (negatively) impact on the their self-efficacy.

Another limitation that might have possibly affected the acceptance of the first null hypothesis was the task difficulty. Although everybody knew the zapateado in advanced, not all the courses had to perform the same one because there were different levels involved. Thus, whereas some students might consider the zapateado was easy, others might have found it difficult. As a result, this perception could have influenced the effect that feedback had on the self-efficacy levels (Bandura & Cervone, 1986, In Feltz et al, 2006).

Despite the fact that the researcher assistant asked students to be quiet and to not communicate their scores, occasionally some students were seen looking at each others´ cards or talking while they were queuing. This type of communication and comparing of scores could have possibly affected the number’s credibility with respect to the score and consequently the displayed reaction.

Homogenizing the number of females and males in this study would have been interesting in order to know whether there is any difference by sex. Perhaps it would have given information about how female and male dancers respond differently to feedback. However, the lack of males in the dance conservatoire 9:1 (García-Dantas & Caracuel, 2011a) makes this a practical limitation of this study.

Lastly, it is possible that the non-relationship found between self-efficacy post- manipulation and effort intentions was due to the one item questionnaire because some students asked for clarification. Some participants simply ticked low scores arguing that they would not have real time to rehearse during that week. In addition to this, as the effort item is a socially desirable variable, some of the participants could have stated firm intentions of investing higher efforts just because it is what they wanted others to believe. Consequently, it would be better to re-state the effort item and formulate it in an alternative way in order to increase validity.

Further Research

Theoretically, with the aim of improving the current research, it would be convenient to control variables that literature have found to influence the way students react to feedback (e.g. optimism -Ortín, et al 2011-, or initial self-efficacy -Chase, 2011). It would be also interesting to acknowledge the positive effects that presenting feedback per objectives has on the attribution style and/ or more specifically in a student’s performance competence. Thus, there would be additional reasons to justify the need or the lack of need of recommending accurate and objective strategies more than using general upper-evaluated or lower-evaluated scores.

As a future work it would be interesting to replicate the current investigation with a bigger sample and including additional feedback conditions. It would be useful if only one accurate score was provided to compare whether it was the specificity or the accuracy of the feedback what facilitate students´ self-efficacy.

Apart from that, results of the present study will be communicated to the teachers throughout an educational session in the hope of demystifying certain believes and to introduce the assessment per objective as an alternative to the lower-evaluated. Within the educational session it would be enlightening to investigate what teachers think about different kinds of methods for providing feedback as well as what they consider the practical advantages and disadvantages of these strategies. This would increase the knowledge concerning barriers and obstacles that teachers face during assessment periods and as a result, to propose real and practical methods for their daily routine as dance teachers.

The consequences of positive self-efficacy levels have been studied in sport. For example, it has been shown that high self-efficacious athletes tend to set more challenging goals, invest greater effort and perform better (e.g. LaGuardia & Labbe, 1993; Moritz, Feltz, Fahrbach, & Mack, 2000; Weiss, Wiese, & Klint, 1989). It is also known that confidence is positively related to motivation and negatively related to state anxiety and dropping out (García-Dantas & Caracuel, 2011a). However, we do not know yet any additional effects on the dance training. Therefore, in order to provide useful information for students and teachers, it would be interesting to learn the types of strategies that people with the higher self-efficacy levels develop to improve their performance. For example, do they rehearse with or without music? Do they practice the steps with the same speed or rather increase it progressively? Would they work with peers or alone? Would they use imagery? This type of information would give clues about efficient and effective strategies to rehearse in dance in terms of self-efficacy.

The last section will suggest some recommendations that have stemmed from the results of this study. These guidelines will help teachers to provide efficient and positive assessment feedbacks to the students from the Dance Conservatories in Spain.

Practical recommendations for teachers

1. Since self-efficacy has been related with cognitive, emotional and behavioural processes that help to optimize performance and wellbeing, lower-evaluated feedback should be avoided as this strategy has been found to be detrimental for self-efficacy.
2. Although upper-evaluated feedback can increase self-efficacy levels in short term, literature has showed the long term counterproductive effects of that. Upper- evaluated feedback can cause non beneficial comparisons among students, dejection and negative emotional reactions when they perceive that this feedback does not correspond to the reality. Furthermore, providing upper-evaluated feedback is not helpful to guide the training and set appropriate goals. Even though some students might wonder the reason why they have to invest effort once they have reached high scores without working.
3. Feedback per objectives: specific scores per objectives plus the general mark increases self-efficacy levels. Literature shows that this strategy increments the control perception above the training.
4. The assessment score matters: although all the students from the lower- evaluated feedback had previous knowledge of their weaknesses and strengths in their zapateado, the manipulated feedback immediately impacted on their self-perceptions of doing it. In other words, imagine a dancer who usually performs the zapateado with a high standard and is used to receive positive feedback from teachers. We have noticed that only with three lower-evaluated score in an experimental situation we can promote a confidence decrease in performing appropriately the zapateado next time. Therefore, we can assume that grades directly affect either positively and negatively students’ self- efficacy.
5. Neither self-efficacy or feedback condition predict effort intentions. The willingness of investing future effort seems to be affected by others psychological processes which require further scientific studies. Thus, we should not presume that lower-evaluated feedback influence effort intentions and neither low self-efficacious student will invest less effort than the rest. Therefore, as there are individual variables that explain the differences of effort intention, it is worthwhile to follow a homogeneous, accurate and objective feedback strategy for all the students. Therefore, this study would recommend the use of objective feedback which means specific score per term objectives plus the general real score.

6. Conclusion

This investigation was well justified both theoretically and practically by providing novel data about a seldom-studied topic in dance psychology. Generally, art education has been linked to the difficulty of assessing the artist’s competence as the intrinsic subjectivity of these disciplines limited impartial and equitable judgements (Buckroyd, 2000). However, teachers from dance conservatories are asked to provide feedback regarding performance quality to the students each term. As a result, assessment periods were associated with feelings of insecurity for many teachers. Many of them tried to combine objectivity with what they think this marks will caused on each student in terms of effort. These beliefs were not scientifically supported and the strategy of providing upper or lower evaluated scores directly depended on the teacher’s personal criteria. Thus, this study aimed to offer recommendations to the teachers to help them face the assessment process with more confidence and objectivity.

So that, it has been analysed the repercussions stemmed of different way of providing feedback in the students´ psychological responses, specially self-efficacy and effort intentions. Self-efficacy was selected because it was one of the variables that presented stronger correlation with emotion, cognition and behaviours linked to performance and wellbeing. Effort intention was chosen because it was the main teachers´ justification as to why gave lower-evaluated scores. From empirical evidence we knew that some teachers held the belief that lower-evaluated feedback lead to an increase in students´ effort intention for next term.

Although the number of participants was small, results demonstrated the damaging effect of lower-evaluated feedback on self-efficacy. Alternatively, it was found that the fact of providing a real general score followed by specific scores per objective promoted beneficial self-efficacy levels. Thus, this strategy would be suggested as the recommended method to the conventional lower-evaluated one.

Ultimately, this original study aims to provide a source for educational programmes regarding good practices for dance conservatory teachers. It is expected that if teachers introduce the suggested new assessment method, students will reach more beneficial self-efficacy levels. This will mean happier dancers who enjoy dancing with better self-perceptions. Furthermore, it is possible they will perceive a greater control about their training and this will lead them to optimize their performance improving large-scale the dancers´ level from the studied Dance Conservatory.

7. References

Bakker, E. (1991). Development of personality in dancers: A longitudinal study. Personality and Individual Differences, 12, 671-681.

Balaguer, I., Castillo, I., Duda, J. L., & García-Merita, M. (2011). Asociaciones entre la percepción del clima motivacional creado por el entrenador, orientaciones disposicionales de meta, regulaciones motivacionales y vitalidad subjetiva en jóvenes jugadoras de tenis. [Associations between motivational climate perception created by the coach, motivational regulation, and subjective vitality in younger female tennis players]. Revista de Psicología del Deporte, 20 (1), 133-148.

Feltz, D. L., Chow, G. & Kepler, T. J. (2006). Path analysis of self-efficacy and performance: Revisited. Journal of Sport and Exercise Psychology, 28, S66.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioural change. Psychological Review, 84, 191-215.

Whyte, G. & Saks, A. M. (2007). The effects of self-efficacy on behaviour in escalation situations. Human Performance, 20 (1), 23–42.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: Freeman.

Barr-Anderson, D. J., Young, D. R., Sallis, J. F., Neumark-Sztainer, D. R., Gittelsohn, J., Webber, L. et al. (2007). Structured physical activity and psychosocial correlates in middle-school girls. Preventive Medicine, 44, 404-409.

Brown L. J., Malouff, M. L., & Schutte, N. S. (2005). The effectiveness of a self- efficacy intervention for helping adolescents cope with sport-competition loss. Journal of Sport Behaviour, 28, 136-151.

Buckroyd, J. (2000). The student dancer. Emotional aspects of the teaching and learning of dance. London, England: Dance Books.

Burke, S.T., & Jin, P. (1996). Predicting performance from a triatlón event. Journal of Sport Behaviour, 19, 272-286.

Cantón, E., & Checa, I. (2012). Los estados emocionales y su relación con las atribuciones y las expectativas de autoeficacia en el deporte. [Emotional states and their relationship with attributions and self-efficacy expectative in sport] Revista de Psicología del Deporte, 21 (1), 171-176.

Chase, M. A. (2001). Children´s self-efficacy, motivational intentions, and attributions in physical education and sport. Research Quarterly for Exercise and Sport, 72, 47-54.

Cofee, P., & Rees, T. (2011). When the chips are down: Effects of attributional feedback on self-efficacy and task performance following initial and repeated failure. Journal of Sports Sciences, 29, 235–245.

Colvin, C. R., Block, J., & Funder, D. C. (1995). Overly positive self evaluations and personality: Negative implications for mental health. In Y-H Kim, C. Y. Chiu, &

Z. Zou (2010). Know myself: Misperceptions of actual performance undermine subjective well-being, future performance, and achievement motivation . Journal of Personality and Social Psychology, 99, 395-409.

Feltz, D. L., Chow, G., & Kepler, T. J. (2006). Path analysis of self-efficacy and performance: Revisited. Journal of Sport and Exercise Psychology, 28, S66.

Feltz, D. F., Short, S. E., & Sullivan, P. J. (2008). Self-efficacy in sport. Research and strategies for working with athletes, teams and coaches. Champaign, IL: Human Kinetics.

Field, A. (2009). Discovering statistics using SPSS (3rd ed.). London, England: Sage.

García-Dantas, A., & Caracuel J. C. (2011a). Factores que influyen en el abandono en los conservatorios de danza. [Factors that infuence on the dance conservatoires dropping out]. Revista Iberoamericana de Psicología del Ejercicio y del Deporte, 6, 79- 97.

García-Dantas, A. & Caracuel J. C. (2011, November). Interacciones entre estilo docente y motivación del alumnado en un conservatorio profesional de danza. [Interactions between teaching style and students´ motivation in a professional dance conservatoire]. Proceedings of Congreso Andaluz de Psicología del Deporte. Sevilla, Spain.

García-Dantas, A. & Caracuel, J. C. (2011b). Psicología y danza. [Psychology and dance]. En F. Arbinaga & J. C. Caracuel (Eds.), Intervención psicológica en actividad física y deportes minoritarios [Psychological intervention in physical activity and minoritary sports] (pp. 139-162). Madrid, Spain: Psimática.

García-Dantas, A., Caracuel J.C., & Peñaloza, R. (2011). Intervención formativa con profesorado de danza e influencia motivacional en el alumnado. [Formative intervention with dance teachers and motivational influence on their students]. Cuadernos de Psicología del Deporte. Manuscript submitted for publication.

George, T. R. (1994). Self-confidence and baseball performance: A causal examination of self-efficacy theory. Journal of Sport and Exercise Psychology, 16, 381-399.

Haney, C. J., & Long, B. C. (1995).Copping effectiveness. A path analysis of self- efficacy, control, coping, and performance in sport competitions. International Journal of Applied Psychology, 29, 73-89.

Hanrahan, S. J. (1996). Dancers’ perceptions of psychological skills. Revista de Psicologia del Deporte, 9-10, 19-27.

Hutchinson, J. C., Sherman, T., & Martinivic, N. (2008). The effect of manipulated self- efficacy on perceived and sustained effort. Journal of Applied Sport Psychology, 20, 457-472.

Ilies, R., Judge, T. A., & Wagner, D. T. (2010). The influence of cognitive and affective reactions to feedback on subsequent goals role of behavioral inhibition/activation. European Psychologist, 15 (2),121–131.

Jaenes, J. C., & Caracuel, J. C. (2001). Estado emocional y conducta deportiva: ansiedad competitiva en corredores de marathon. [Emotional state and sport behaviour: competitive anxiety in marathon runners]. Unpublish manuscript. Asociación Gallega de Psicología del Deporte de la Actividad Fisica y el Deporte en el III milenio (129-137) Pontevedra, Spain.

Jussim, L., Yen, H., & Aiello, J. (1995). Self-consistency, self-enhancement, and accuracy in reactions to feedback . Journal of Experimental Social Psychology, 31, 322-356.

Kim, Y.-H., &. C. Y. Chiu (2011). Emotional costs of inaccurate self-assessments: Both self-effacement and self-enhancement can lead to dejection. Emotion, 11, 1096- 1104.

Kimmerle, M., & Cote-Laurence, P. (2003). Teaching dance skills. A motor learning and development approach. Andover, NJ, J. Michael Ryan Publishing.

LaGuardia, R, & Labbe, E. E. (1993). Self-efficacy and anxiety and their relationship to training and race performance. Perceptual and Motor Skills, 77, 27-34.

Lázaro, I., & Villamarín, F. (1993). Capacidad predictiva de la auto-eficacia individual y colectiva sobre el rendimiento en jugadores de baloncesto. [Individual and collective self-efficacy as a performance predictor of basketball players]. Revista de Psicología del deporte, 4, 1-14.

Marquez, D. X., Jerome, G. J., Mcauley, E., Snook, E. M., & Canaklisova, S. (2002). Self-efficacy manipulation and state anxiety responses to exercise in low active women. Psychology and Health, 17, 783-791.

Mallaina, I., & García-Dantas, A. (2012). Variables psicológicas en función del curso en los conservatorios de danza: un estudio descriptivo. [Psychological variables regarding course in dance conservatoires: a descriptive study]. Manuscript submitted for publication.

McAuley, E., & Blissmer, B. (2000). Self-efficacy determinants and consequences of physical activity. Exercise and Sport Science Reviews, 28 (2), 85-88.

McAuley, E., Wraith, S., & Duncan, T. E. (1991). Self-efficacy, perceptions of success, and intrinsic motivation for exercise. Journal of Applied Social Psychology, 21 (2), 139-155.

Moritz, S. E., Feltz, D. L., Fahrbach, K. R., & Mack, D. E. (2000). The relation of self- efficacy measures to sport performance: A meta- analytic review. Research Quarterly for Exercise and Sport, 71, 280-294.

Nease, A. A., Mudgett, B. O., & Quiñones, M. A. (1999). Relationships among feedback sign, self-efficacy, and acceptance of performance feedback. Journal of Applied Psychology, 84, 806-814.

Ortega, E., Olmedilla, A., Sainz de Baranda, P., & Gómez, M. A. (2009). Relationship between the level of self-efficacy, performance indicators, and participation in youth basketball. Revista de Psicología del Deporte, 18, 337-342.

Ortín, F. J., Garcés de los Fayos, E. J., Gonsálvez, J., Ortega, E., & Olmedilla, A. (2011). Optimismo y ejecución en el deporte en situaciones adversas: Replicando a Seligman 1990. [Optimism and execution in sport in different situations: Replicating Seligman 1990]. Revista de Psicología del Deporte, 20 (2), 491-501.

Schmidt, A. M., & Deshon, R. P. (2010). The moderating effects of performance ambiguity on the relationship between self-efficacy and performance. Journal of Applied Psychology, 95, 572–581.

Schunk, D. H. (1991). Self-efficacy and academic motivation. Educational Psychologist, 26, 207-231.

Sousa, R. M., & Bandeira, M. A. (1994) Knowledge of results precision and learning: A review. Revista de Psicología del deporte, 1, 23-37.

Thomas, J. R., Nelson, J. K., & Silverman, S.J. (2005). Research methods in physical activity (5th ed.) Champaign, IL: Human Kinetics.

Treasure, D. C., Monson, J., & Lox, C. L. (1996). Relationship between self-efficacy, wrestling performance, and affect prior to competition. Sport Psychologist, 10, 73-83.

Walker, I. J., & Nordin-Bates, S. M. (2010). Performance anxiety experiences of professional ballet dancers: The importance of control. Journal of Dance Medicine and Science, 14, 133-145.

Weinberg, R. S. (1985). Relationship between self-efficacy and cognitive strategies in enhancing endurance performance . International Journal of Sport Psychology, 17, 280-292.

Weinberg, R. S. (2003). Goal setting in sport and exercise: results, methodological issues and future directions for research. Revista de Psicología del Deporte, 7 (8), 115-125.

Weinberg, R., Gould, D., & Jackson, A. (1979). Expectations and performance: An empirical test of Bandura´s self-efficacy theory. Journal of Sport Psychology, 1, 320-331.

Weinberg, R. S., Gould, D., Yukelson, D., & Jackson, A. (1981). The effects of pre- existing and manipulated self-efficacy on a competitive muscular endurance task. Journal of Sport Psychology, 3, 345-354.

Weiss, M. R., Wiese, D. M., & Klint, K. A. (1989). Head over heels with success: The relationship between self-efficacy and performance in competitive youth gymnastic. Journal of Sport and Exercise Psychology, 11, 444-451.

Whyte, G., & Saks, A. M. (2007).The effects of self-Efficacy on behavior in escalation situations. Human Performance, 20 (1), 23–42.

Young-Hoon Kim, Y. H., & Chiu, C. Y. (2011). Emotional costs of inaccurate self- assessments: Both self-effacement and self-enhancement can lead to dejection. Emotion, 11, 1096-110.

8. Appendix

Appendix A

Table 4: Previous Studies Examining the Self-efficacy Effect on Sports

illustration not visible in this excerpt

Appendix B

Table 5: Initial Self-efficacy Levels Effect on the Feedback Response

illustration not visible in this excerpt

Appendix C

Table 6: Effect of High Self-efficacy levels on Effort

illustration not visible in this excerpt

Appendix D

Informed consent form for students

I confirm that:

I am willing to participate in the “zapateado” experiment and cooperate in self- administering the questionnaires.

I am free to withdraw from the test or the experiment at any time.

I have been informed about:

All information is confidential and anonymous and used only for the purposes of this study. I will be assigned an identification number in order to preserve the questionnaires anonymity.

The experiment and their effects will be fully explained to me at the end of the class.

I will be able to access all information collected upon the completion of the study, upon request.

The Trinity Laban Research Ethics Committee has reviewed and approved the Project.

If there is an aspect of the study which concerns you, you may make a complaint via e- mail (anagardan@hotmail.com).

Personal ID Date. .. Gender…..AgeSignature…

Appendix E

INFORMED CONSENT PARENT PERMISSION LETTER

(model retrieved from http://www.fordham.edu/academics/office_of_research/institutional_review/sample_con sent_forms_25951.asp)

June 2012

Dear Parent or Guardian,

Ana García-Dantas psychologist and Dance Science student is conducting a research study entitled “feedback and self-efficacy” with Professional grade students at Dance Conservatory of …. I am interested in examining the importance of promoting the students´ self-efficacy. Ultimately, our hope is to learn the best way of providing term assessment feedback. With the permission of the Dance Conservatory and with the approval of Laban Ethics Committe, we are requesting that you allow your youngster to participate.

Participants in the study will be asked to learn a zapateado by the regular teacher. Afterwards, students will be provided a numeric score as a feedback. Participants will also complete a brief survey asking them how did feel after knowing that feedback. The total time to participate in the study will be approximately 60 minutes. Students who participate will complete the study during the regular flamenco class so there will not have to come extra time for the study.

ID instead of names will be used in filling out the students´ consents forms so all responses will be anonymous. No one at the dance conservatory will have access to any of the information collected.

Participation in the study is entirely voluntary and there will be no penalty for not participating. All students for whom we have parent consent will be asked if they wish to participate and only those who agree will complete the forms. Moreover, participants will be free to stop taking part in the study at any time. The experiment and their effects will be fully explained to the students. And participants, will be able to access all information collected upon the completition of the study, upon request.

Please give your permission by signing the consent form and having your youngster return it to his or her homeroom teacher tomorrow. Please keep this letter for your records.

Sincerely,

Ana García-Dantas

illustration not visible in this excerpt

I have read the attached informed consent letter and agree to have my youngster participate in the study entitled “Effect of assessment feedback on the dancer´s self- efficacy and effort intentions.”

illustration not visible in this excerpt

Parent’s or Guardian’s Signature Date

Appendix F PERSONAL CODE

INSTRUCTIONS: For the following zapateado´s criterium, circle the number that represents how confident you are that YOU CAN perform the zapateado correctly in the three trials that you are going to do:

illustration not visible in this excerpt

Spanish Version CÓDIGO

Instrucciones: te presentamos 4 aspectos importantes que hay que tener en cuenta para realizar un zapateado. Por favor, señala el número que exprese en qué grado crees que TÚ PUEDES realizar cada aspecto correctamente. Al final escribe un número que represente la confianza que tienes de realizar el zapateado en su conjunto de forma correcta.

Ejemplo: si estoy convencida de que voy a mantener un correcto control postural pondré un 10 y si no tengo mucha confianza en cómo me saldrá de limpio el zapateado pondré un 5.

illustration not visible in this excerpt

Appendix G PERSONAL CODE

Questionnaires (post-test)

INSTRUCTION: Please, keep in mind the average score that you have received over the 3 trials. Imagine hypothetically you will have to demonstrate the same zapateado next week but instead, the score will be part of your regular term assessment. So circle the number that better represent how would you address that situation:

1.- Intention of future effort: Adapted Effort one-item scale (George, 1994).

How much effort do you think you will put into practicing the zapateado during these forthcoming days? Please, tick SICERELY one of the boxes below:

illustration not visible in this excerpt

2.- Adapted Diving Self-Efficacy Questionnaire (Feltz, 1982 in Feltz., Short, & Sullivan, 2008, p.44).

For the following zapateado´s criterium, circle the number that represent how confident you think you are that YOU CAN perform the zapateado next week correctly:

illustration not visible in this excerpt

Thank you so much for your cooperation.

Spanish Version CÓDIGO

1.) Instrucciones: Por favor, piensa en las calificaciones que has recibido por parte de tu profesor/a. Imagina de forma hipotética que tienes que hacer ese mismo zapateado la semana que viene pero que la nota recibida formará parte de tu calificación de final de curso.

¿Cuánto crees que te esforzarías en practicar el zapateado durante los siguientes días hasta la fecha del examen? Señala una de las opciones. Ej: 0-10 es muy poco esfuerzo (pocas horas, baja intensidad, poca atención, poco interés en comparación con lo que necesitarías) y 90-100 sería si te esforzaras al máximo teniendo en cuenta tus necesidades y posibilidades.

illustration not visible in this excerpt

Por favor, señala el número que exprese en qué grado crees que TÚ PODRÁS realizar cada aspecto mencionado correctamente. Al final escribe un número que represente la confianza que tienes de realizar el zapateado en su conjunto de forma correcta.

illustration not visible in this excerpt

Muchísimas gracias por colaborar.

Appendix H

SPSS output from Hypothesis 1:

a) Difference between groups in Self-efficacy levels after feedback manipulation.

Table 7: Mean Rank Values in Post Self-efficacy

illustration not visible in this excerpt

Table 8: Kruskal Wallis p. value

illustration not visible in this excerpt

b) Differences between pre and post manipulation in self-efficacy levels

Table 9: Pre to Post Self-efficacy Differences in the Upper-evaluated Group

illustration not visible in this excerpt

Table 10: Pre to Post Sel-efficacy Differences in the Objective Group

illustration not visible in this excerpt

Table 11: Pre to Post Differences in the Lower-evaluated Group

illustration not visible in this excerpt

Table 12: Pre to Post Differences in the Control Group

illustration not visible in this excerpt

b) Hypothesis 2: Linear regression between Self-efficacy post manipulation and effort intentions

Table 13: Model: Self-efficacy Levels as Predictors of Effort Intentions

illustration not visible in this excerpt

Table 14: Linear Regression

illustration not visible in this excerpt