In achievement motivation theory, achievement goals represent subjective purposes (Pintrich, 2000a) or cognitive-dynamic focuses (Elliot & Church 1997) of competence-relevant behaviors for executing tasks. Portrayed as concrete representations of more abstract achievement motivational constructs, achievement goals are conceptualized as midlevel constructs situated between global motivational dispositions (antecedents of achievement goals) and specific behaviors (consequences of achievement goals) (Elliot & Church 1997).
Over the past three decades, approaches to achievement goals have undergone considerable development toward understanding motivated behavior in achievement settings. Using undergraduate samples, Elliot and Church (1997), Elliot and McGregor (2001), and Elliot and Murayama (2008) provide their own evidence for the location of achievement goals between global motivational dispositions such as fear of failure or need for achievement, and specific academic behaviors such as study strategies. Other researchers have identified such factors as classroom social environment (e.g., goal structure emphasized in a class, Wolters, 2004), general motives (e.g., need for achievement, Zusho, Pintrich, & Cortina, 2005) and competency expectancies (e.g., self-efficacy, Liem, Lau, & Nie, 2008; Vrugt, Oort, & Zeeberg, 2002) as direct antecedents of achievement goal adoption, with achievement goals directly and proximally influencing achievement-relevant consequences such as task scores, help-seeking behaviors, and self-regulation strategies, among others (Cury et al., 2006; Elliot & McGregor, 2001; Pintrich, Conley, & Kempler, 2003).
The dissertation attempted to examine the factor structure of the 2 × 2 achievement goal framework (Elliot & McGregor, 2001), the stability of achievement goal endorsement, and an achievement goal model with self-efficacy as an antecedent of achievement goals and Chinese performance as a consequence. In the following section I review related studies on (1) achievement goal theory, (2) the stability of achievement goal endorsement (3) achievement goals and self-efficacy, as well as (4) achievement goals and academic performance.
Achievement goal theory Dichotomous goal model
Various two goal models have been described and established by achievement goal theorists such as Ames and Archer (1988), Elliott and Dweck (1988), and Nicholls (1984). Ames and Archer (1988) emphasize mastery (i.e., the development of ability through task mastery) and performance goals (i.e., demonstrating ability relative to others). Elliott and Dweck (1988) distinguish between learning and performance goals. They suggest that learning goals, in which one seeks to develop competence, facilitate challenge-seeking and mastery-oriented responses to failure regardless of perceived ability. In contrast, performance goals (in which one seeks to gain favorable judgment for competence or avoid negative judgments) are described as causing challenge-avoidance and learned helplessness. Nicholls (1984) emphasizes task goals (developing ability in reference to one's past performance or knowledge) versus ego goals (demonstrating ability as capacity relative to those of others). Pintrich et al. (2003) suggest that despite differences among the dichotomous goal models behind these various terms, the concepts of mastery and performance have become the most commonly used labels in achievement goal research.
When clarifying and integrating mastery versus performance goal definitions, Pintrich et al.(2003) note that mastery goals emphasize competence, learning, and understanding tasks according to self-referenced standards of improvement, while performance goals focus on demonstrating competence and superiority according to comparative or normative standards.
Researchers such as Ames and Archer (1988), Elliott and Dweck (1988), and Nicholls (1984) describe mastery goals in terms of adaptive motivational patterns characterized by persistence in the face of failure, the use of increasingly complex learning strategies, and the pursuit of difficult and challenging tasks. Performance goals, however, are viewed as maladaptive motivational patterns characterized by greater propensity to withdraw from tasks, less interest in difficult tasks, and a tendency to seek less challenging tasks for which there is a greater likelihood of success.
In contrast, Harackiewicz, Barron and Elliot (1998) and Pintrich et al. (2003) do not view mastery and performance goals as opposite ends of a continuum, or as mutually exclusive in terms of their original concept formulations. Both research teams have reported Western-based research findings suggesting that mastery goals and performance goals are either unrelated (Ames
& Archer, 1988) or positively correlated (Pintrich & Schunk, 2002) in support of a multiple goal
perspective in which individuals pursue either a single predominant goal or multiple goals.
Similarly, Chan (2008) and Ng (2000) found statistically significant and positive correlations between mastery and performance goals in non-Western samples consisting of Hong Kong students aged 9 to 17 and Mainland Chinese secondary school students. Both researchers suggest that social endeavor in Chinese culture connects the two concepts, since the social goals of bringing honor to one’s family by working or studying hard can shape both mastery and performance goals.
Later, goal theorists (Elliot & Church 1997; Middleton & Midgley, 1997) criticize dichotomous goal perspectives and extend them to a trichotomous achievement goal framework.
According to Elliot and Church (1997), it may be unproductive to view all performance goals as maladaptive or in opposition to mastery goals. Middleton and Midgley (1997) also point out that dichotomous goals, mastery and performance, are commonly conceptualized as “approach”
motivational tendencies rather than “avoidance” motivational tendencies. These goal theorists, as well as Elliot (1997) and Elliot and Church (1997), note that activities in achievement settings may be either directed toward the attainment of success or the avoidance of failure. When reviewing the histories of approach and avoidance motivation theory, Elliot (1999; 2006) found that approach motivation is behavior directed by positive stimuli, whereas avoidance motivation is regarded as behavior directed by negative stimuli; in both cases the stimuli take the form of objects, events, or possibilities.
Trichotomous goal model
Elliot (1999), Elliot and Church (1997), and Middleton and Midgley (1997) all suggest that performance goals should be divided into two categories—approach performance goals and avoidance performance goals—because they have different effects on outcomes, and because some of them are not less adaptive, as predicted by traditional goal theory. This finding leads Elliot and Church (1997) to propose a trichotomous achievement goal framework, which they tested in the context of college classrooms. Their results provide strong support for the framework with three achievement goals: mastery, performance-approach, and performance-avoidance. Mastery goals are emphasized in the development of competence and task mastery, performance-approach goals are oriented toward the attainment of favorable
judgments of competence, and performance-avoidance goals emphasize the avoidance of unfavorable judgments of competence.
Middleton and Midgley (1997) tested their proposed trichotomous achievement goal model in the context of a middle school mathematics classroom. Their results give support to their model with three goals: task (developing ability), performance-approach (demonstrating ability), and performance-avoidance (avoiding demonstrations of lack of ability). Elliot (1999), Middleton and Midgley (1997), and Pintrich (2000a) are among researchers who believe that compared to dichotomous or oppositional goal categories, trichotomous achievement goal models reflect complex goal constructs more precisely.
2 x 2 achievement goal model
Following the logic of separating approach and avoidance performance goals, Pintrich (2000a) suggests that both versions of mastery goals may exist concurrently. He offers a 2×2 matrix that combines mastery and performance goals with approach and avoidance states. He defines two general aspects of achievement goals: general purpose or reason for engaging in a task, and standards or criteria that individuals use to judge their performance. His list of four goals consists of mastery approach, mastery avoidance, performance approach, and performance avoidance.
Based on Elliot and Church’s (1997) trichotomous achievement goal framework, Elliot and McGregor (2001) developed an advanced revision and extension known as the 2 × 2 achievement goal framework. It consists of two pairs of goals crossing over each other to form four achievement goals: mastery-approach, performance-approach, mastery-avoidance, and performance-avoidance. The feasibility of this model was examined by exploratory and confirmatory factor analyses and found empirical support for the differentiation among the four goals.
Elliot and McGregor (2001) and Elliot and Murayama (2008) posit that achievement goals contain components from two independent competence dimensions. The first, mastery versus performance, refers to competence as defined in terms of the standard used to evaluate it (Dweck, 1986; Elliot & Church, 1997; Nicholls, 1984; Pintrich, 2000a; Pintrich et al., 2003).
Mastery-based goals reflect a concern for developing competency and the use of self-referential
improvement standards, while performance-based goals reflect a concern for demonstrating competency in terms of social comparisons. The second dimension, approach versus avoidance, indicates how competence can be valenced. Approach-based goals focus on a movement toward positive stimuli such as competence and success, while avoidance-based goals focus on a movement toward negative stimuli such as incompetence and failure.
As Elliot and Church (1997) and Elliot and McGregor (2001) note in their trichotomous achievement goal and 2 × 2 frameworks, performance-approach goals emphasize demonstrations of skill and the attainment of favorable judgments of competency in relation to others.
Performance-avoidance goals focus on avoiding unfavorable judgments of competency and poor performance when compared to others. Mastery-approach goals focus on developing knowledge and skills, as well as enhancing competency and mastery in the form of intrapersonal or task-based criteria. Mastery-avoidance goals, which are the least studied in the achievement goal literature (Elliot, 1999; Pintrich, 2000a), focus on avoiding the loss of skills, abilities, or knowledge (and sometimes on avoiding misunderstanding), thus failing in terms of learning or task mastery. Elliot and McGregor (2001) provide two examples of mastery avoidance goals:
perfectionists who try to avoid making any mistakes whatsoever, and individuals in the latter parts of their careers (e.g., athletes and businesspersons) or lives (e.g., the elderly) who focus on not losing their skills, abilities, or memory. Elliot and Murayama (2008) suggest that mastery-avoidance goals emerge from both positive (the need for achievement) and negative sources of motivation (fear of failure), and note that the overall effect of mastery-avoidance goals remains unclear.
Under a multiple goal perspective (e.g., Pintrich, 2000b; Elliot & McGregor, 2001), goal theorists note that people often hold multiple goals simultaneously and so four goals are not independent. They therefore examine the intercorrelations among achievement goals. The empirical evidence on zero-order correlations among the four achievement goals is mixed.
Results from two Western-based research findings—using samples of American university students (Elliot & McGregor, 2001) and French secondary school students (Cury et al., 2006)—
suggested that mastery-avoidance goals were positively associated with mastery-approach and performance-avoidance goals. They also showed positive associations between performance-avoidance goals and mastery-avoidance and performance-approach goals, but no association between mastery-approach goals and performance-approach goals. Using a sample of
Taiwanese junior high school students, Cherng (2003) found positive associations between performance-approach and performance-avoidance goals, as well as between mastery-avoidance goals and both mastery-approach and performance-approach goals. He failed to find any association between mastery-approach and performance-approach goals and between mastery-avoidance and performance-avoidance goals, but did observe a negative association between mastery-approach and performance-avoidance goals (Cherng, 2003).
Mastery-avoidance goals and their related variables
Mastery-avoidance goals represent a fairly new construct to achievement goal theory. Some researchers (e.g., Ciani & Sheldon, 2010; Sideridis & Mouratidis, 2008) suggest that it may be still a conceptually problematic and somewhat controversial construct. In a sample of university baseball players, Ciani and Sheldon (2010) found mastery-avoidance goals were uncommon, and that high ratings may indicate misinterpretation of the items rather than actual mastery-avoidance goals. Sideridis and Mouratidis (2008) investigated nearly 400 elementary to middle school students selecting their most prominent achievement goal. Only 14 students chose mastery-avoidance goals as their primary goal. These results led Sideridis and Mouratidis (2008) to question the existence of mastery-avoidance goals in young students. The debate is likely because of ambiguity and counterintuitive nature of the mastery-avoidance goals (Elliot &
McGregor, 2001).
Pintrich (2000a) defines the mastery-avoidance goals as the reasons for engaging in tasks, as well as the standards or criteria that individuals use to judge their performance. The mastery-avoidance goals focus on avoiding misunderstanding, not learning or mastering tasks, and criteria for not doing things incorrectly relative to a task. Elliot and McGregor (2001) transform the definition of the mastery-avoidance goal to a construct in experiencing competence—defined as the absolute requirement of a task or one's own attainment.
Incompetence is the central point of regulatory attention, with the main focus being on avoidance of negative possibilities. Elliot and McGregor (2001) provide examples which individuals are striving to avoid misunderstanding and so failing to learn course materials, striving to not make errors in business transactions, making a free throw in a basketball game, not leaving an incomplete crossword puzzle (i.e. someone dislikes/rejects to play a crossword puzzle because he
believe that he may be incapable of completing crossword puzzle to leave an incomplete one), not forgetting what one has learned (i.e., someone refuse to learn something new because he believes it may interfere/confuse what he has learned), and striving not to lose one's physical or intellectual capabilities (i.e., someone rejects to develop new capabilities because he believes these new capabilities may not performing as well as pervious excellent records and even damage or lose his existing capabilities). Pintrich (2000a) offers a prototypical exemplar, perfectionists who struggle to avoid making any mistakes whatsoever and individuals in the latter part of their careers or lives who focus on not performing worse than in the past, not stagnating, and not losing their skills, abilities, or memory.
While Elliot and McGregor (2001) examined the antecedents and consequences of the mastery-avoidance goals in an attempt to develop empirical profiles, their findings indicated mixed mastery-avoidance goal profiles. The results yielded that the mastery-avoidance goals were grounded in the fear of failure, low self-determination, perceived classroom engagement, entity (instead of incremental) view of competence, parental person-focused negative feedback, parental worry induction, and competence valuation. College students’ endorsement of the mastery-avoidance goals has precedent influences from parental socialization. Comparatively, parental socialization was not related to the endorsement of the mastery-approach goals.
The mastery-avoidance goals are associated with adaptive and maladaptive learning consequences. Elliot and McGregor (2001) show they are positive predictors of disorganized study habits, test anxiety, and subsequent mastery-avoidance, mastery-approach, and performance-approach goals. In a group of Taiwanese junior high school students, Cherng (2003) found that mastery-avoidance goals were positive predictors of self-handicapping, help-seeking, effort, persistence, and math grades. In the sport domain, mastery-avoidance goals have been linked to fear of failure (Conroy & Elliot, 2004), amotivation (Nien & Duda, 2008), and negative reactions to imperfection (Stoeber, Stoll, Pescheck & Otto, 2008). Other studies have identified positive associations between mastery-avoidance goals and perceived competence, enjoyment, effort, and physical activity (Wang, Biddle & Elliot, 2007), as well as perceptions of an enjoyable learning climate (Morris & Kavussanu, 2008).
When the mastery-avoidance goals are compared to three other goals, mastery- avoidance goals differ conceptually from mastery-approach goals regarding the valence of competence, from performance-avoidance goals regarding the definition of competence, and from
performance-approach goals regarding both the definition and valence of competence (Elliot &
McGregor, 2001). Empirical findings of Elliot and McGregor and Cherng (2003) revealed that mastery-avoidance goals were more negative than the mastery-approach goals, and more positive than the performance-avoidance goals. Mastery-avoidance and performance-avoidance goals have very similar antecedent profiles in terms of non-optimal variables—for example, fear of failure (Conroy & Elliot, 2004) and amotivation (Nien & Duda, 2008). Unlike performance-avoidance goals, and similar to mastery-approach goals, mastery-avoidance goals emerge from individual perceptions that a class (or some other scenario) is engaging and interesting (Elliot & McGregor, 2001). The mastery-avoidance goals share some negative characteristics with the performance-avoidance goals, but they differ from the performance-avoidance goals in that they are neither negative predictors of performance achievement (Cherng, 2003; Elliot & McGregor, 2001) nor positive predictors of health center utilization (Elliot & McGregor, 2001).
Elliot and McGregor (2001) attribute the mixed conceptual profiles of the mastery-avoidance goals to the combination of optimal (mastery) and non-optimal components (avoidance). Mastery has been always viewed as adaptive by educational psychologists (Dweck, 1986; Pintrich, 2000a; Pintrich et al., 2003) while avoidance maladaptive and how do we categorize the combination? Elliot and McGregor (2001) suggest that the adoption of these goals is most likely among individuals with non-optimal motivational dispositions in optimally structured achievement settings that challenge pursuit and foster intrinsic interest. They also suggest that empirical predictions regarding the mastery-avoidance goal antecedents and consequences are difficult to generate for two reasons. First, the mastery component likely results from optimal antecedents and the desire to foster positive consequences (similar to the mastery-approach goals), but the avoidance component likely results from non-optimal antecedents and causes negative consequences (similar to the performance-avoidance goals).
Second, it is hard to determine the relative strengths of the two components when combined, or the accurate manner in which each component functions in combination with the other.
Finally, optimal motivation and performance may require combinational types of goals.
Empirical evidence has indicated that pursuing one type of goal does not necessarily exclude pursuit of the other (Ames & Archer, 1988; Bouffard-Bouchard, Boisvert, Vezeau, & Larouche, 1995; Harackiewicz et al., 1997; Middleton & Midgley, 1997). Based on a multiple goal
perspective Shih (2005b) found that a group of Taiwanese elementary students who maintained high-mastery/high-performance-approach goals showed more adaptive learning patterns than students who maintained other types of multiple goals.
Measurement for the 2 x 2 achievement goals
Elliot and McGregor (2001) developed an achievement goal questionnaire (AGQ) to measure four goals in the 2 × 2 achievement goal framework. Item pools for mastery-approach goals, performance-approach goals, and performance-avoidance goals were chosen from their previous instruments (Elliot & Church, 1997); new items were designed for mastery-avoidance goals.
Three items were generated to represent each of the four achievement goal constructs. In the questionnaires, 3 items in each subscale assessed mastery–approach goals (e.g., “It is important for me to understand the content of this course as thoroughly as possible.”), mastery–avoidance goals (e.g., “I am often concerned that I may not learn all that there is to learn in this class.”), performance–approach goals (e.g., “It is important for me to do well compared to others in this class.”), and performance–avoidance goals (e.g., “My goal in this class is to avoid performing poorly.”). Participants responded to the extent which they believed ranged from 1 (not at all true of me) to 7 (very true of me). AGQ was tested in introductory-level undergraduates’ psychology classes in series of studies. The results of exploratory factor analysis (EFA), confirmatory factor analysis (CFA) empirically supported the separable and internally consistent achievement goal constructs; Cronbach alpha coefficients evidenced good reliability. AGQ was translated into Chinese version and used as the main measurement tool in my dissertation.
Stability of achievement goal endorsement
Do learners endorse the same goals or do they change goal adoption across time? In a review of theoretical perspectives regarding stability in achievement goal adoption over time, Fryer and Elliot (2007) note that achievement goals emerge from stable factors (e.g., personality traits such as achievement motives and temperaments) and remain grounded in these factors throughout goal pursuit and regulation processes. In addition, they claim that goal stability lies in the nature of the goal construct. When individuals face achievement tasks, they adopt goals and
develop cognitive frameworks for interpreting those tasks, experience task involvement, and react to competence-relevant information (Ames, 1992; Dweck, 1986). This framework can result in directional or biased perceptual–cognitive processes that foster subsequent goal seeking behaviors in a self-fulfilling way (Elliot & Harackiewicz, 1996).
Only a few articles focusing on this critical issue of goal stability and change have been available. Of these, some have addressed change in achievement goals across a sequence of similar tasks during several weeks of college classes (Senko & Harackiewicz, 2005; Fryer &
Elliot, 2007); some have examined shifts in goal endorsement for school within a school year (e.g., Bong, 2005; Seifert, 1996); still others have examined shifts in goal endorsement for school across the elementary to middle school transition (e.g., Anderman & Midgley, 1997). To my knowledge, there is still short of research about stability of achievement goal endorsement in Asian population.
My question of whether achievement goal endorsement changes in learning Chinese across secondary school year or whether it remains stable can be answered in several ways, depending on what type of change (or stability) one focuses on. Typical parameters are means, variances, and covariances, all of which may be subject to remain stable over time. There are at least three types of stability that can be examined in sample levels using longitudinal panel data: structural stability (or change), differential stability, and mean-level stability. Structural stability refers to the constancy of covariances among a set of constructs across time. In my case, structural change
My question of whether achievement goal endorsement changes in learning Chinese across secondary school year or whether it remains stable can be answered in several ways, depending on what type of change (or stability) one focuses on. Typical parameters are means, variances, and covariances, all of which may be subject to remain stable over time. There are at least three types of stability that can be examined in sample levels using longitudinal panel data: structural stability (or change), differential stability, and mean-level stability. Structural stability refers to the constancy of covariances among a set of constructs across time. In my case, structural change