Comparing the Motivational Profiles of High-AbilityLow-Performing (HALP) Students & High-AbilityHigh-Performing (HAHP)Students

Under the No Child Left Behind (NCLB) Act, the federal government has made aggressive demands on K-12 educators and administrators to eradicate underachievement in their schools. While low student achievement is generally viewed as cognition-related (e.g., a lack of knowledge or skills), contemporary educational psychology recognizes the relationship between low student achievement and maladaptive motivational beliefs (e.g., a low expectancy for success). Although student underachievement is troubling in a general sense, student underachievement is particularly disturbing when students with high aptitude exhibit low performance. Yet not enough is known about the motivational characteristics of these high-ability-low performing (HALP) students, especially the nature of their motivational beliefs and how they interpret goals and expectancies in achievement settings. Contemporary achievement motivation theorists relate goals and expectancies to three major indices: (a) one’s choice of an achievement task, (b) one’s persistence on that task, and (c) the amount of mental effort one expends in completing that task (Eccles, Wigfield, & Schiefele, 1998; Pintrich & Schunk, 2002).

When high-ability students do poorly in school, this not only constitutes a dilemma for schools, it also creates a predicament for the individual and society as a whole. The societal consequences of ignoring HALP student underperformance are potentially harmful for a number of reasons: (1) the intellectual potential that HALP students possess may go untapped; in an increasingly competitive global market, high-ability individuals represent cognitive capital. (2) As a result of internalizing or externalizing their frustration with the consequences of failure, HALP students may be more likely to experience negative affective consequences such as depression or anger. Chronic underachievement can lead to pessimism, a cycle of defeatism, and entrapment in a psychological doom loop (Good & Brophy, 2000; Collins, 2001). (3) HALP students with low expectancies for future success may become disaffected with school and revert to maladaptive behaviors such as task avoidance, defiance, aggression towards peers, reluctance to assume responsibility for their behavior, disruption of the learning program, and/or delinquency. (4) Ultimately, HALP students may suffer from low self-efficacy and a distorted self-concept. What is known about these high-ability-low-performing students? Although the literature does much to illuminate the associations between adolescent goal orientations and achievement outcomes in academic settings, the findings shed little light on the HALP student phenomenon. The same holds true for the connection between adolescent expectancy-value constructs and achievement outcomes with respect to HALP students.

The intent of this study was, therefore, to determine the motivational characteristics of high-ability students who under perform in school with respect to goal orientation and expectancy-value theories of motivation. This distinct group of underachievers, which are referred to as High-Ability-Low-Performance (HALP) students, typically demonstrate high ability on standardized achievement tests, yet exhibit low performance in school as evidenced by low academic grade point averages. To better comprehend the phenomenon of HALP student underachievement, this study compared the motivational profiles of High-Ability-Low-Performing (HALP) students and their High-Ability-High-Performing (HAHP) counterparts on goal orientation (mastery, performance-approach, performance-avoid) and expectancy-value beliefs (expectancy for success, perceptions of competence, task value). By comparing the motivational characteristics that typify these two groups, this study sheds light on the differences between HALP students and HAHP students. The significance of goal orientation and expectancy-value beliefs with relationship to K-12 education is noteworthy in that dysfunctional motivational beliefs can initiate maladaptive patterns of behavior that lead to academic underachievement.

Connections between goal orientation and student achievement are well established. Recent research has divided performance goal orientations into a trichotomous framework which includes one mastery goal orientation and two distinct performance goal orientations: a performance-approach goal orientation and a performance-avoidance goal orientation (Elliot & Church, 1997; Middleton & Midgley, 1997; Midgley, et al., 1998). Goal orientation theory suggests that students who possess a mastery goal orientation focus on developing their competence in achievement settings. Midgley et al. (2000) associate a mastery goal orientation with adaptive patterns of learning such that students who exhibit a learning (mastery) goal orientation seek to develop competence in achievement settings through mastery and understanding of academic subject matter. In other words, mastery- oriented students are content-focused. Conversely, students who possess a performance-approach goal orientation are self-focused individuals who seek to demonstrate their competence in achievement settings for self-enhancement purposes. Midgley et al. (2000) associate a performance-approach goal orientation with both adaptive and maladaptive patterns of learning. Students who possess performance- avoid goal orientations also self-focused in that they seek to avoid the perception of incompetence in achievement settings by managing the perceptions of others. Midgley et al. (2000) associate a performance- avoid goal orientation with maladaptive patterns of learning.

Researchers have discovered that students who endorse a mastery goal orientation are more likely to invest greater effort in learning, hold an intrinsic value for learning, employ deep-level cognitive processing, have higher self-efficacy, and utilize self-regulated learning strategies (Ames, 1992; Middleton & Midgley, 1997; Anderman & Young, 1994; Pintrich & Schrauben, 1992).

The relationship between student achievement and expectancy values are also well established (Eccles & Wigfield, 1992, 1994, 2002; Dickhauser & Steinsmeier-Pelster, 2003; Whitley, 1997; Marshall & Brown, 2004; Pajares & Miller, 1994; Anderman et al., 2001). Atkinson (1964), who originally proposed the expectancy-value theory, suggested that expectancies represent cognitive appraisals of performance situations and pursuant judgments of one’s likelihood of success. Expectancy-value theories of achievement motivation with regard to academic achievement have primarily included three areas: (1) ability beliefs, (2) expectancies for success, and (3) task value. Eccles, et al. (1998) suggest that an individual’s expectancy constructs and subjective task value directly influence active choice, persistence, and mental effort. Expectancies refer to beliefs about how one will do on future achievement tasks. Expectancy-for-success beliefs center on an individual’s perception of competence with regard to one’s future possibility of success (Eccles & Wigfield, 1995). Ability beliefs (a.k.a. self-perceptions of competence) center on an individual’s perception of current ability. Self-perceptions of competence represent an individual’s evaluative judgments of their ability for executing a task. Task value represents the values portion of expectancy-value theory. Eccles et al. (1983) categorized achievement values into four psychological components: (1) attainment value, the importance of doing a task well, (2) intrinsic value, the satisfaction one derives from engaging in a task, (3) utility value, the usefulness of a task with regard to future gain, and (4) cost, the personal and emotional price one pays for engaging in a particular task.

Because high ability is generally a predictor of academic success, it appears paradoxical when high-ability students perform poorly in school. This paradox describes high-ability-low-performance (HALP) students who perform well on standardized assessment tests, yet perform poorly in the classroom as evidenced by below average and failing grades on their report cards. When high-ability-low- performance students perform well on their standardized achievement tests (scoring in the Proficient and Advanced range) and under perform on their quarterly grades (getting C’s, D’s and F’s), motivational issues are a fruitful area to examine.

From a cognitive perspective, it appears counterintuitive when a highly capable student fails or does poorly in school. This is because performance is generally associated with knowledge-based conceptions of learning. When viewed through a motivational lens, however, HALP student underachievement makes more sense. This is because performance is at least partially determined by students’ active choice of activities (e.g., studying), persistence, and mental effort. The knowledge-based component of ability in this study was represented by students’ standardized test scores, generally regarded as demonstrations of academic knowledge and intellectual competence.

GPA, by contrast, represents the ability to complete actual tasks assigned by teachers. Whereas most standardized achievement scores are determined by a single assessment that occurs once a year, academic grade point average represents multiple assessments of student effort over time. Academic GPA represents one indicator of student success in motivationally-relevant areas: getting started, persisting, and putting forth the mental effort necessary to complete achievement tasks assigned by teachers. Teacher-assigned achievement tasks, which are typically accomplished both inside and outside of the classroom, generally involve strategic planning, organizational skills, self-regulation, sustained task engagement, and adherence to timelines. Academic GPA is a longitudinal composite of all these task-related achievement behaviors. From a motivational perspective, High-Ability-Low-Performance (HALP) students can be viewed as having a high degree of knowledge and ability, but maladaptive motivational beliefs.

This study characterizes HALP students as high ability as evidenced by high achievement test scores and low performance as evidenced by low cumulative grade point averages. Conversely, this study characterizes High-Ability-High-Performance (HAHP) students as high ability (as evidenced by high achievement test scores) and high performance (as evidenced by a high cumulative grade point average. This study examined the relationship between student achievement and motivation by using predictor variables such as goal orientation, expectancy-value beliefs, and academic behavior for both HALP and HAHP students.

Whereas poor performance due to lack of cognitive ability is viewed as a legitimate obstacle to success, poor performance due to lack of motivation is not (Weiner, 1995, Reyna & Weiner, 2001). Because students with cognitive disabilities possess stigmas whose causes are perceived as uncontrollable, assistance from support providers is more readily given. Types of assistance for students with cognitive learning disabilities can include special programs, smaller class sizes, one-on-one tutoring, specially trained teachers and aides, higher per-pupil spending formulas, and the likelihood that students with cognitive deficits will promote rather than be retained. High Ability-Low-Performing students, by contrast, can expect to experience anger and negative judgments from others due to the voluntary and controllable nature of their low performance (perceived as lack of motivation). For this reason HALP students may fail to evoke assistance from support providers. Subjecting HALP students to persistent admonitions, negative judgments, and retributive punishment for their chronic underperformance may do little to enhance their performance.

The motivational consequences of punishing and ignoring high- ability-low-performing students could be harmful for a number of reasons: (1) the intellectual potential that HALP students possess may go untapped. (2) As a result of internalizing or externalizing their frustration with the consequences of failure, HALP students may be more likely to experience negative affective consequences such as depression or anger and may come to avoid academic tasks. (3) HALP students may become disaffected with school and revert to maladaptive behaviors such as task avoidance, learned helplessness, disruption of the learning program, or delinquency. (4) HALP students may ultimately suffer from low self-efficacy and a distorted self-concept. Understanding the possible differences in the motivational beliefs of High-Ability-Low-Performance students could play a pivotal role in transitioning HALP students from low achievement to high achievement.

The purpose of this study was to compare the motivational profiles of High-Ability-Low-Performing (HALP) students and High-Ability- High-Performing (HAHP) students from a goal orientation and expectancy-value perspective. The existing literature suggests that conceptions of ability, learning goals, expectancies, and task value all play a role predicting academic success or failure in school. By examining some of the maladaptive motivational beliefs that characterize the motivational profiles of HALP and HAHP students, this study sheds light on possible areas that may increase academic success. The specific research questions were:

This was a comparative, descriptive study that examined the relationship of motivational variables to academic outcomes for two groups of students. A schematic of this study was conceptualized as a path model that began with the motivational predictor variables (goal orientation, expectancy-value), was mediated by academic behavior, and ended in academic achievement.

This study compared the motivational profiles of two groups of students: the first group represented students who exhibited high academic ability with respect to California Standardized Test (CST) scores, but low academic performance with regard to grade point average (GPA) and were referred to as High-Ability-Low-Performance (HALP) students. The second group represented students who exhibited high academic ability with respect to California Standardized Test (CST) scores, as well as high academic performance with regard to grade point average (GPA) and were referred to as High-Ability-High- Performance (HAHP) students. Although both groups shared similar cognitive ability (as evidenced by proficient/advanced California Standardized Test scores), the groups differed with regard to academic performance (as evidenced by low GPAs for HALP students and high GPAs for HAHP students).

The participants were middle school students (grades 7-8) in an upper middle class suburban community in southern California. Of 1,075 students in this middle school, 17% of the students participated in the survey (n = 183) High-Ability-Low-Performing (HALP) students (n = 58) represented 42 % of the participants in this study, while High- Ability-High-Performing (HAHP) students (n = 80) represented 58 % of the participants.

Subjects chosen for participation in the study were divided into two groups: (1) High-Ability-Low-Performance (HALP) students, and (2) High-Ability-High-Performance (HALP) students. Both High- Ability-Low-Performance (HALP) and High-Ability-High- Performance (HALP) student designation required that participants exhibit High Ability (HA) on the English Language Arts (ELA) portion of their California Standardized Test (CST ELA = Advanced or Proficient). To measure students’ level of ability, the California Standards Test (CST) was utilized to determine participants’ English Language Arts and Mathematics proficiency. Through the California Standardized Testing and Reporting (STAR) Program, students in grades 2-11 are tested annually in various subject areas. Currently, the STAR program includes California Standards Tests (CST) in English language arts and mathematics in grades 2-11. The California Standards Test (CST) indicates the level at which students are achieving in relation to the state content standards. The five CST performance levels are Advanced (exceeds state standards), Proficient (meets state standards), Basic (approaching state standards), Below Basic (below state standards), and Far Below Basic (approaching state standards). Students at the Proficient or Advanced level meet state content standards in that area. High ability was defined as performing in the Proficient range (at grade level) or the Advanced range (above grade level) on the California Standardized Test (CST) in English Language Arts (ELA). Of this high ability ELA group (N =138), 83% were also designated high ability in mathematics (n = 115), with CST Mathematics scores in the Proficient or Advanced range.

After determining the students that did qualify as High Ability (HA) for this study (females = 87, males = 51), academic Grade Point Averages (GPAs) were then reviewed to determine which students were High Performing (HP) and which students were Low Performing (LP). Because Total GPA measures all subjects, including nonacademic subjects (e.g., P.E., art, music, dance, woodshop, chorus, home economics, etc.), only Academic GPA was used in determining students academic performance in school. Academic GPA measures students’ grades in their four academic core subjects (math, English, social science, science). Because the High Ability measure was predicated on high academic achievement scores on the California Standardized Test, Academic GPA was the logical performance measure with regard to academic achievement.

Goal orientation predictor variables included: (a) mastery, (b) performance-approach, and (c) performance-avoid subscales. Expectancy-value predictor variables included: (a) expectancies for success, (b) perceptions of competence, and (c) task values. Academic behaviors were based on teacher reports that provided information on participants with respect to: (a) citizenship, (b) work habits, and (c) comment codes. Academic achievement was based on students’ academic grade point average (GPA).

The Patterns of Adaptive Learning Scales (PALS) instrument was used to measure participants’ personal achievement goal orientations (Midgley, et al., 2000). Midgley et al., (2000) established reliability and validity for the PALS instrument using middle school students; goodness of fit indices confirmed that their model fit the data well (GFI=0.97, AGFI=0.95). This instrument included three personal achievement goal orientation subscales: (1) Mastery Goal Orientation, (2) Performance-Approach Goal Orientation, and (3) Performance- Avoid Goal Orientation. Each subscale was represented by five items in the survey; responses to survey items were made using a five-point Likert scale. A Cronbach’s Alpha analysis indicated that the mastery goal subscale (a = .863), and the performance-approach subscale (a = .873), yielded stronger reliability than the performance-avoid subscale (a = .718).

Survey items used to measure participants’ personal achievement expectancy-values were adapted from Different Self-Report Items Used to Measure Various Expectancy and Value Constructs (Pintrich & Schunk, 2002). The instrument used in this study included three personal achievement expectancy-value subscales: (1) Expectancies for Success, (2) Perceptions of Competence, and (3) Task Value. In order to measure all of the personal achievement expectancy-value constructs accurately, each subscale consisted of five similar sounding statements. Responses to survey items were made using a five-point Likert scale. The Expectancies for Success scale was comprised of five items. Sample items included, “I will do well on my report card” and “I expect to do better than most other students”. The Perceptions of Competence scale also consisted of five items including, “I’m just as smart as other students” and “I learn new concepts easily”. A Cronbach’s Alpha analysis indicated the following: expectancy-for- success subscale (a = .781), the perceptions-of-competence subscale (a = .672), and the task-value subscale (a = .614).

Although it should be noted that the Cronbach alpha coefficient of a scale should ideally be above .7, short scales with less than ten items commonly display Cronbach values that are lower than .7 (Pallant, 2004). In this study, each of the six subscales (mastery goal orientation, performance-approach goal orientation, performance-avoid goal orientation, expectancy for success, perceptions of competence, task value) included only five items. Of these six subscales, two exhibited alpha scores lower than .7 (perceptions of competence, task value).

In addition to academic grades, quarterly report cards and midquarter progress reports provided key information that described the academic behavior of students. Academic behavior was reported by teachers on Student Report Cards in three separate ways: (1) Student Citizenship, (2) Student, Work Habits and (3) Comment Codes. Teacher reports on students’ citizenship was mandatory whereas teacher reports on work habits and teacher comment codes (i.e., “Interested in learning-tries hard” or “Not working to capability”) were optional. Teacher reports on district Report Cards and district Progress Reports produced information regarding students’ Citizenship (O = Outstanding, S = Satisfactory, N = Needs Improvement, U = Unsatisfactory), and Work Habits (O = Outstanding, S = Satisfactory, N = Needs Improvement, U = Unsatisfactory). Teacher Comment Codes consisted of 19 descriptors of academic behavior, coded by letters A through T (e.g., A = Excellent student, I = Homework /Assignments incomplete, K= Poor motivation/attitude, etc.).

Academic achievement for this study was represented by academic grade point average (GPA). Whereas the California Standards Tests (CSTs) represented student ability in this study, academic GPA represented actual academic performance. CST data was generated by students taking multiple choice assessments over a period of several days. One of the utilitarian purposes of standardized tests is to provide school administrators with data for the purpose of determining student placement (e.g., students scoring in the Advanced range on the CST are often eligible to be placed in honors classes, while students who score Basic or below range are generally placed in remedial classes). High academic ability did not, however, predict high academic achievement with regard to academic GPA. This is because the diagnostic assessment data that the CST provided was much less complex than the composite data that generated academic GPA. Academic GPA represented a qualitatively richer account of student achievement over time and across situations. The rigorous achievement tasks that constituted academic GPA required the deployment of motivational indices (active choice, persistence, mental effort) for students to succeed in school. This was because academic GPA was a summative composite of grades predicated on the completion of homework assignments, demonstration of new knowledge, test and quiz preparation, research, compositional writing, construction of projects, thematic presentations, and other complex undertakings that involved higher order thinking and execution skills.

Because the subjects in this study were minors, assent to participate in non-medical research was obtained from the participants and consent to participate in non-medical research was obtained from their parents. Data used to measure participants’ level of ability were obtained by looking at students’ California Standards Test (CST) results in English Language Arts and Mathematics proficiency. This information was contained on the school database at the site where the study was conducted. Participants’ cumulative academic GPA (used to measure academic achievement) and teacher reports on students’ academic behaviors (citizenship, work habits, comment codes) were also obtained from the school database for the site that was used for the study.

Because the purpose of this research was to compare the motivational profiles of High-Ability-Low-Performing (HALP) and High-Ability-High-Performing (HAHP) students, statistical techniques that explore differences between groups were performed. All quantitative data was analyzed using SPSS for Windows version 13.0. Surveys provided quantitative data on goal orientation and expectancyvalue constructs; teacher reports provided quantitative data on the academic behaviors of HALP and HAHP student groups. Independentsamples t-tests were performed to compare the mean scores and standard deviations between groups and variables. One-way analyses of variance (ANOVA) between and within groups were performed to determine the mean scores, F ratios, and significant differences between multiple groups (e.g., HALP/HAHP groups, gender groups, ethnic groups). Pearson Chi-Square tests determined the relationship between HALP/HAHP groups and all motivational variables explored in this study. A factor analysis of scales evaluated all subscales and determined the strength of relationship among survey items. Linear regressions created model summaries for all the predictor variables used for academic grade point average (the dependent variable) in this study. Linear regression coefficients were also used to evaluate each of the independent variables and standardize the values for each variable by converting them into a single scale. Additionally, a reliability of scales was run to determine the internal consistency of subscales using a Cronbach’s coefficient alpha analysis.

After determining that 138 of the participants met the High-Ability (HA) criteria, based on their Proficient to Advanced performance rankings on California Standards Test (CST), the group was then divided into Low-Performance (LP) and High-Performance (HP) cohorts based on academic Grade Point Average (GPA). The cut point used to separate Low-Performance (LP) from High-Performance (HP) students was the mean GPA score for all 138 high ability students included in this study. After determining a mean GPA score of 3.4, the High-Ability-Low-Performance (HALP) students became the group with an academic GPA range below the group mean (GPA > 3.40). After using the mean academic GPA as the cut point, the groups for the study consisted of 42% HALP (n = 58) and 58% HAHP (n = 80) students. The mean GPA variance between the HALP (m GPA = 2.87) and HAHP (m GPA = 3.79) groups yielded a statistically significant difference between groups (p < .001).

Of the 138 that participated in the study, 37 % were male (n = 51), 63 % were female (n = 87), 50% were seventh graders (n = 69), 50% were eighth graders (n = 69), 28.3% were Asian (n = 39), 9.4% were Hispanic (n = 13), 2.9% were African American (n = 4), and 59.4% were White (n = 82). The Asian participants in this study consisted of Chinese (n = 6), Japanese (n = 2), Korean (n = 10), Vietnamese (n = 3), Asian Indian (n = 5), Filipino (n = 5), and Other Asian (n = 8). Of the HALP student group (n = 58), 17% were Asian (n = 10), 14% were Hispanic (n = 8), 3% were African American (n = 2), and 66% were White (n = 38). Of the HAHP student group (n = 80), 36% were Asian (n = 29), 6% were Hispanic (n = 5), 3% were African American (n = 2), and 55% were White (n = 44). See Table 1  for HALP and HAHP student participants by gender, grade level, and ethnicity.

HALP/HAHP Participants by Gender, Grade Level, & Ethnicity

Note. HALP = High-Ability-Low-Performance students, HAHP = High-Ability-High-Performance students.

Means, Standard Deviations, & Correlation of Variables for Goal Orientation & Expectancy-Value for HALP Students & HAHP Students

Mean academic grade point average (GPA) comparisons between High-Ability-Low-Performing (HALP) students and High-Ability- High-Performing (HAHP) students revealed a much lower mean average for HALP students (M GPA = 2.87, SD = 0.45) than for HAHP students (M GPA = 3.79, SD = 0.17). Standard deviations for academic GPA was greater for HALP students (SD = .45) than for HAHP students (SD = .17). An overall analysis of variance (ANOVA) for grade point average (GPA) between High-Ability-Low-Performing (HALP) students and High-Ability-High-Performing (HAHP) students yielded a statistically significant difference between groups (p < .000) with F = 28.31. See Table 2  for mean scores and standard deviations for all goal orientation and expectancy-value variables examined in this study.

The relationship between the six motivational predictor variables in this study (Expectancy for Success, Self Perceptions of Competence, Task Value, Mastery Goal Orientation, Performance-Approach Goal Orientation, Performance-Avoid Goal Orientation) and the outcome variable (Academic Grade Point Average), was explored using the Pearson product-moment correlation coefficient (see Table 2  for HALP student variables). Pearson correlations for all High Ability Low Performing student variables indicated that expectancy for success was the strongest correlate to academic GPA. The Expectancy-for-Success variable had significantly high correlations with all other variables in this study. The highest correlation occurred between the PerformanceApproach Goal Orientation and the Performance-Avoid Goal Orientation variables. Both performance orientations are considered to be self-focused such that the individual performs for an extrinsic audience; performance-approach learners seek self-enhancement by outperforming others, while performance-approach learners seek to avoid appearing incompetent in the eyes of others (Midgely et al., 2000). There was also a significant correlation between Task Value and Mastery Goal Orientation for the HALP students in this study.

Pearson correlations for all High Ability High Performing student variables indicated that mastery goal orientation was the strongest correlate to academic GPA (see Table 2  for HAHP student variables). Although the Expectancy for Success variable had significant correlations with Self Perceptions of Competence, Task Value, Mastery Goal Orientation, and Performance-Approach Goal Orientation, Performance-Avoid Goal Orientation, the highest correlation occurred between the Performance-Approach Goal Orientation and the Performance-Avoid Goal Orientation variables. Curiously, the Pearson product-moment correlation coefficient between the performanceapproach and the performance-avoid goal orientation variables wasidentical between HALP (n = 58) and HAHP (n = 80) groups (r = .747, p < .000). Like the HALP students, the HAHP students displayed a significant correlation between Task Value and Mastery Goal Orientation. Overall differences between HALP and HAHP came in the link to academic GPA (outcome variable). A lower expectancy for success was associated with a lower GPA for HALP students, while a higher mastery goal orientation was associated with a higher GPA.

Research question one: “What are the goal orientation characteristics of adolescent High-Ability-Low-Performance (HALP) students as compared with their High-Ability-High-Performance (HAHP) peers?” High-Ability-Low-Performance (HALP) students exhibited a markedly lower mastery orientation mean score (M = 3.93, SD = .90) than their High-Ability-High-Performance (HAHP) peers (M = 4.26, SD = .54). Responses to performance-approach items were virtually identical between groups, with both groups reporting relatively low performance-approach orientation mean scores: HALP students (M = 2.65, SD = 1.04) and HAHP students (M = 2.67, SD = 1.02). Responses to performance-avoid items revealed a slightly higher performance-avoid orientation in HALP (M = 2.83, SD = .85) students than in HAHP (M = 2.71, SD = .88) students.

An ANOVA revealed a statistically significant variance between HALP and HAHP students for mastery goal orientation. A one-way between-groups ANOVA F-test revealed significant variance between HALP and HAHP students for mastery goal orientation (F = 6.84, p = .01). Pearson chi-square tests confirmed statistical significance for mastery goal orientation between HALP and HAHP students (p = .052).

The connection between ethnicity and Goal Orientation was also explored in this study. With the exception of the Chinese group (n = 6), goal orientation T-test comparisons between White students (n = 82) and all other ethnic groups (Hispanic, n = 13; Korean, n = 10; Other Asian, n = 8; Asian Indian, n = 5; Filipino, n = 5; African American, n = 4; Vietnamese, n = 3; Japanese, n = 2), revealed no significant differences (P > .05). Mean mastery scores between the White group (M = 4.0) and the Chinese group (M = 4.7), however, revealed significantly higher mastery goal orientation responses by the Chinese students on their surveys (p = .041). It should be noted that all the Chinese students were designated as high-ability-high-performance (HAHP) students in this study.

Research question two: “What are the expectancy-value belief characteristics of adolescent high-ability-low-performance (HALP) students as compared with their high-ability-high-performance (HAHP) peers?” High-Ability-Low-Performance (HALP) students exhibited significantly lower expectancies for success (M = 3.77) than their High- Ability-High-Performance (HAHP) peers (M = 4.17). Although HALP students also exhibited lower self-perceptions of competence than their HAHP peers, the difference was not significant: HALP students (M = 3.59), HAHP students (M = 3.75). Responses to Task Value items were significantly lower for High-Ability-Low-Performance (HALP) students than their High-Ability-High-Performance (HAHP) peers: HALP students (M =3.84) and HAHP students (M = 4.06).

An ANOVA was conducted to test the statistical variance of means between HALP and HAHP students revealed statistically significant variance between HALP and HAHP students for expectancy for success and task value. A one-way between-groups ANOVA F-test revealed significant variance between HALP and HAHP students for expectancy for success (F = 14.32, p < .001), and task value (F = 3.80, p = .053). Pearson chi-square tests confirmed statistical significance for the expectancy-for-success predictor variable in HALP and HAHP groups (p = .006).

The relationship between ethnicity and expectancy-value constructs was also examined in this study. With the exception of the Other Asian group (n = 8), expectancy-value T-test comparisons between White students (n = 82) and all other ethnic groups (Hispanic, n = 13; Korean, n = 10; Chinese, n = 6; Asian Indian, n = 5; Filipino, n = 5; African American, n = 4; Vietnamese, n = 3; Japanese, n = 2), revealed no significant differences (p > .05). Mean scores between the White group (M = 3.9) and the Other Asian group (M = 4.4), however, revealed significantly higher expectancy-for-success survey responses by the Other Asian student group (p = .037). Regarding gender, independent samples T-tests also revealed no significant differences between males and females for all three expectancy-value variables (expectancy for success, perceptions of competence, task value).

Research question three: “What are the academic behavior characteristics of adolescent high-ability-low-performance (HALP) students as compared with their high-ability-high-performance (HAHP) peers?” With respect to Citizenship behaviors, frequency and percentage data from teacher reports on student citizenship for math and English language arts revealed a great deal of contrast between High-Ability-Low-Performance (HALP) and High-Ability-High- Performance (HAHP) students. When comparing the citizenship profiles of HALP and HAHP students, reports by both mathematics and English language arts teachers revealed that a much lower percentage of HALP students were recognized for outstanding citizenship than their HAHP counterparts.

Although the data on student Work Habits was sparse, due to the fact that teachers were not mandated to report student work habits, the data that did exist in this study suggested how some teachers viewed the work habits of HALP and HAHP students. Like the citizenship scores, math teachers were more likely to view the work habits of higher performing HAHP students as outstanding, and more likely to view the work habits of low performing HALP students as satisfactory or needing improvement. Concerning work habits, mathematics teachers were far less likely to identify HALP students as outstanding (1.7 %) as compared with HAHP students (17.5 %). HALP student work habits were much more likely to be described as satisfactory (13.8 %) as compared with HAHP students (0.0 %). Work habit reporting for English language arts was insignificant due to lack of reporting by English teachers on the work habits variable. Pearson chi-square tests for independence between variables also confirmed statistical significance for Work Habits variables in mathematics for HALP/HAHP groups (p < .001).

Another source of students’ academic performance data came by way of Teacher Comments. Teacher Comments provided descriptors on how students are performing in their classrooms. Teacher comments represented observations of the adaptive behaviors that contributed to academic success and the maladaptive behaviors that hindered academic achievement. Math teacher comments reported on HALP and HAHP student report cards in this study consisted of a mix of citizenship-related observations (e.g., “a pleasure to have in class”) and motivation-related observations (e.g., “not working to capability”). When comparing the profiles of High-Ability-Low-Performing (HALP) students with High-Ability-High-Performing (HAHP) students, math teachers were considerably less likely to make positive comments about HALP students (e.g., “a pleasure to have in class”) than they were for HAHP students (HALP = 19.0%, HAHP = 33.8%). With reference to completion of work (e.g., “homework/assignment incomplete”), HALP students displayed a far greater likelihood of not completing their work (HALP = 10.3%, HAHP = 0.0%). HAHP students, by contrast, showed no reported incidences of incomplete and missing homework comments.

Research question four: “Are goal orientation and expectancyvalue beliefs a significant predictor of academic grade point average?” Correlations of all variables in this study (see Table 2) indicated that expectancy for success was by far the variable most greatly associated with academic GPA (p < .001) for High-Ability-Low-Performance (HALP) students. All other variables in this study exhibited no significant correlation with academic GPA with respect to the HALP student group. Correlation data for the High-Ability-High-Performance (HAHP) student (see Table 2) clearly indicated that a mastery goal orientation was the greatest predictor of academic GPA (p < .01). None of the other variables in this study exhibited a significant correlation with academic GPA with respect to the HAHP student group. In other words, a low expectancy for success correlated with a low academic GPA for HALP students, while a high mastery goal orientation correlated with a high academic GPA for HAHP students.

General findings for all the motivational predictor variables examined in this study with respect to Pearson chi-square tests for independence between High-Ability-Low-Performing (HALP) and High-Ability-High-Performing (HAHP) students revealed significance (p < .05) in the areas of mastery goal orientation, expectancy-for- success, citizenship (English language arts), citizenship (mathematics) and work habits (mathematics).

Table 3  represents a linear regression model summary of findings for all the predictor variables used for academic grade point average in this study. The data presented in this table illustrates: (a) individual linear regressions for each individual variable (mastery, performanceapproach, performance-avoid, expectancy for success, perceptions of competence, task value), (b) cluster linear regressions for goal orientation (mastery, performance-approach, performance-avoid) and expectancy-value (expectancy for success, perceptions of competence, task value), and (c) a total linear regression for all predictor variables represented in this study. The R Square data explains how much of the variance in the dependent variable (academic GPA) is explained by the model (goal orientation and expectancy-value). The R Square value indicates that the expectancy-for-success variable accounted for the greatest variance (21%) in academic GPA, followed by mastery goal orientation (9%), perceptions of competence (7%), task value (6%), and performance-avoid goal orientation (2%). Combined goal orientation (mastery, performance-approach, performance-avoid) R Square value indicates that goal orientation accounted for 12% of variance in academic GPA. The combined expectancy-value (expectancy for success, perceptions of competence, task value) R Square value indicates that expectancy-value accounted for 21% of variance in academic GPA. The combined goal orientation and expectancy-value (all predictor variables) R Square value indicates that the total model accounted for 27% of variance in academic GPA. Because the R Square value tends to optimistically overestimate the true variance small sample sizes, the Adjusted R Square value adjusts for this error. The overall R Square yielded a 27% variance in academic GPA while the Adjusted R Square value suggested that the total model accounted for a 24% variance in academic GPA.

Table 4  evaluates each of the independent variables using linear regression coefficients. The SPSS Standardized Coefficient standardizes all values for each variable by converting them into a single scale. The Beta values in the Standardized Coefficient column compared all the different variables in this study. When all variables in this model have been controlled for, the Beta values indicate that expectancy for success accounts for the greatest variance in academic GPA. The Beta values in this study determined that the largest beta coefficient value belonged to the expectancy-for-success variable (P = .467), followed by performance-avoid goal orientation (P = -.202), mastery goal orientation (P = .089), performance-approach goal orientation (P = -.047), task value orientation (P = .018), and perceptions of competence (P = .010). Whereas beta values compare the contributions of each individual variable, the significance value determines whether or not a variable is making a statistically significant unique contribution to the equation. The expectancy-for-success variable is making a significantly unique contribution to the prediction of academic GPA such that the Sig. value is less than .05 (p < .001). It should be noted that although all other variables were greater than .05, the Sig. value is dependent on which variables are included in the equation and how much overlap exists between variables.

To summarize, both High-Ability-Low-Performing (HALP) and High-Ability-High-Performing (HAHP) groups displayed high academic ability with respect to standardized test scores (proficient or advanced), but significantly distinct (p < .001) mean academic grade point averages (HALP GPA = 2.87; HAHP GPA = 3.79). Grade point average represented the outcome variable in this study. Motivational predictor variables for GPA included goal orientation (mastery, performance-approach, performance-avoid) and expectancy-value beliefs (expectancy for success, perceptions of competence, task value). Academic behaviors (citizenship, work habits) were also examined. In the area of goal orientation, analysis of variance values for mastery goal orientation revealed significant differences (p = .010) between groups, with HALP students displaying lower mastery scores than HAHP students. In the area of expectancy-value beliefs, expectancy- for-success analysis of variance values yielded significant differences (p < .005) between groups, with HALP students displaying lower expectancy-for-success scores than HAHP students. Task-value also showed a significant ANOVA variance (p = .053), with HALP students exhibiting lower task-value scores than HAHP students. Pearson chisquare tests for independence between groups confirmed statistical significance for mastery goal orientation (p = .052) and expectancy for success (p = .006), as well as citizenship in math (p = .023), citizenship in English language arts (p = .049), and work habits in math (p < .001), indicating that HALP students engage in less adaptive academic behaviors than HAHP students. Linear regression calculations indicated that the single greatest predictor for academic achievement was the expectancy-for-success variable, which represented a 21% variance in academic GPA. Correlation data indicated that a low expectancy for success correlated with a low academic GPA for HALP students (p < .001), while a high mastery goal orientation correlated with a high academic GPA for HAHP students (p < .01).

The results indicate that High-Ability-Low-Performing (HALP) students display a significantly lower mastery goal orientation than High-Ability-High-Performing (HAHP) students. This implies that HALP students are less likely to focus on developing competence in achievement settings than HAHP students. The findings also indicate that HALP students are less likely to seek mastery and understanding of academic concepts than their HAHP peers. A lower mastery goal orientation also indicates that HALP students are less likely to invest the amount of effort necessary to achieve high grades. This is consistent with goal orientation theory which associates a lowered mastery goal structure with maladaptive learning patterns such as a lowered expenditure of effort (Midgley, et al., 2000). This lowered expenditure of effort on the part of HALP students was confirmed by teacher reports of students’ work habits. Teachers were significantly more likely to describe HALP student work habits as needing improvement or satisfactory and more likely to describe HAHP student behaviors as outstanding. Lack of effort was also reflected in comment codes where ten percent of HALP students were cited by math teachers as not completing homework and assignments, as opposed to zero cases among HAHP students.

Previous research suggests that students with high mastery beliefs are more likely to invest effort in learning, more likely to value learning, more likely to view setbacks as opportunities to improve competence, more likely to employ deep-level cognitive processing, more likely to have high self-efficacy, and more likely to utilize selfregulated learning strategies (Dweck & Leggett, 1988; Elliott & Dweck, 1988, Ames, 1992; Pintrich & Schrauben, 1992; Anderman & Young, 1994; Middleton & Midgley, 1997; Midgley, et al., 2000, Elliott & McGregor, 2001). These previous findings, support the supposition that High-Ability-Low-Performing (HALP) students’ lower mastery goal orientation translate to less effort invested in learning, more superficial cognitive processing, less self-efficacy, and a more self-denigrating view of setbacks as compared to their High-Ability- High-Performing (HAHP) counterparts. Less favorable teacher reports and significantly lower academic grade point averages also support these conclusions.

The maladaptive achievement behaviors exhibited by HALP students could stem from HALP students’ conception of intelligence. Implicit theories of intelligence suggest that academic engagement, persistence, and goal orientation are all predicated on what individuals believe about the fundamental nature of intelligence (Dweck, 1986; Leonardo & Gialamas, 2002; Grant & Dweck, 2003). A mastery learning orientation, for example, has been associated with an effortbased conception of achievement, centered on the belief that people can become intelligent through effort, learning, and mastery; believing that intelligence can be incrementally improved motivates the individual to increase their effort in achievement situations (Dweck, 1986; Dweck & Leggett, 1988). By believing that high ability is an effort-based construct, HAHP students’ higher mastery beliefs may suggest that they hold an incremental conception of intelligence. HALP students, by contrast, may hold greater entity beliefs about intelligence, believing that ability is a fixed entity over which they hold little control. Attributing low performance to low ability has been found to result in decreased intrinsic motivation and self-worth (Grant & Dweck, 2003). If HALP students believe that intelligence can’t be changed, they would be less willing to engage in challenging tasks in order to avoid failure and associated feelings of incompetence.

High-Ability-High-Performing (HAHP) students demonstrate significantly higher expectancies for success than High-Ability-Low- Performing (HALP) students. These findings indicate that higher expectancies for success displayed by the High-Ability-High- Performing (HAHP) group play a significantly beneficial role in their academic performance. Despite possessing similar ability levels to their HAHP counterparts, HALP students exhibited much lower expectancies for success. This lowered expectancy for success resulted in a 21% lower grade point averages for HALP students. This finding is consistent with previous research which has concluded that students with high expectancies for success have a higher likelihood of succeeding in school than students with low expectancies for success (Pajares & Miller, 1994; Eccles & Wigfield, 1995; Eccles, et al., 1998; Eccles & Wigfield, 2002; Dickhauser & Steinsmeier-Pelster, 2003; Marshall & Brown, 2004).

Whereas expectancy-for-success beliefs highlight an individual’s future possibility of success, perceptions-of-competence beliefs center on an individual’s perception of current ability (Eccles & Wigfield, 1995). This study found that High-Ability-Low-Performing (HALP) and High-Ability-High-Performing (HAHP) students exhibited virtually identical mean scores on the perceptions-of-competence subscale, indicating that both HALP and HAHP student groups have confidence in their ability to learn and understand new concepts. Curiously, this finding reveals that although HALP students’ maintain optimistic self-appraisals of current ability, they concurrently hold pessimistic expectancies for future success. This would suggest that perceptions-of-competence beliefs are not obstacles to academic achievement for HALP students, but that expectancy-for-success beliefs are, and that believing one can do well is not the same as believing that one will do well. This study does not, however, uncover why HALP and HAHP students begin with similar self-perceptions of current ability yet diverge so significantly with regard to expectancies for future success.

The task-value findings indicated that the High-Ability-Low- Performing (HALP) group exhibited significantly lower mean scores on the task value subscale than High-Ability-High-Performing (HAHP) group. A chi-square test further revealed a statistically significant relationship between the mastery goal and the task value variables used in this study, implying that HAHP students, who are more likely to endorse a mastery goal orientation, are also more likely to hold higher task value for achievement tasks than HALP students. Studies involving junior high and high school students revealed that students who value and display interest in subject area content are more likely to report deeper processing strategies, greater self-regulatory strategies, and higher academic achievement (Wolters & Pintrich, 1998; Greene, DeBacker, Ravindran, & Krows, 1999). Research also indicates that task-related behaviors are influenced by an individual’s degree of interest for a given task (Eccles & Wigfield, 1992, 1994; Schiefele, 1999; Dickhauser & Steinsmeier-Pelster, 2003).

Teachers in this study had clearly differentiated perceptions of HALP students and HAHP students. High-Ability-Low-Performing (HALP) students were viewed by their teachers in a more negative light than their High-Ability-High-Performing (HAHP) peers. HALP students were viewed as less compliant, less likely to complete work, and less likely to put forth the effort necessary to succeed. Unfavorable teacher appraisals of HALP student performance may very well create a negative feedback loop for HALP students such that teacher expectations can play a role in a student’s academic success such that inferences about the abilities of students can lead teachers to treat students differently (Good & Brophy, 2000). If teachers hold lower expectations for High-Ability-Low-Performing (HALP) students and higher expectations for High-Ability-High-Performing (HAHP) students, HALP students may very well respond to decreased teacher expectations with decreased self-expectancies for success, a decreased motivation for learning, and decreased expenditures of effort, all ending in decreased grade point averages. Negative teacher inferences about particular students or particular groups of students can also lead to stigmatization (Reyna & Weiner, 2001). Once stigmatized, HALP students could expect to become the recipients of negative teacher judgments. These negative appraisals could translate into lowered teacher expectations, less praise, more admonition, and more retributive consequences for HALP students. This cycle of defeatism (doom loop) can become self-sustaining over time and across situations (Collins, 2001).

Another stigmatizing factor that may turn teachers, parents and other support providers against High-Ability-Low-Performing (HALP) students is their high ability. Whereas poor performance due to lack of cognitive ability is viewed as a legitimate obstacle to success, poor performance due to lack of motivation is not (Weiner, 1995, Reyna & Weiner, 2001). Because students with neurological disabilities possess stigmas whose causes are perceived as uncontrollable, pity is elicited from support providers and assistance is given (Reyna & Weiner, 2001). HALP students, by contrast, can expect to experience anger and negative judgments from others due to the perceived controllability of their stigmas- a lack of motivation. For this reason HALP students fail to evoke positive attitudes and assistance from support providers. Whereas positive attitudes lead to assistance for those with cognitive disabilities (e.g., special programs, smaller class sizes, one-on-one tutoring, specially trained teachers and aides, higher per-pupil spending formulas, and the likelihood that students with cognitive deficits will promote rather than be retained), anger leads to retribution (e.g., detentions, failed classes, summer school, behavior contracts, and retention) for HALP students. Such negative teacher reactions most likely do little to enhance HALP students’ intrinsic motivation. Viewed as willful, HALP students can expect to receive much less assistance than either special needs or HAHP students. Put another way, students who are perceived as incapable of doing well are supported, while students perceived as capable, yet unwilling to do well, are not supported.

The implications of this study for public educators are considerable such that if a significant number of high-ability students endorse low expectancies for success at one middle school site in California, then the problem is most likely generalizable to most other middle schools throughout America. The scope and scale of this problem could be enormous. If educators are to fulfill their obligation to close achievement gaps between high-performing and low-performing subgroups in public schools, then the motivational obstacles that High- Ability-Low-Performing (HALP) students experience can no longer be disregarded. Previous expectancy-for-success research offers suggestions that may be helpful in raising HALP students’ expectancies for success. Marshall and Brown (2004) suggest that because low expectancies for success have been found to have a greater impact on achievement-related situations than high expectancies for success, convincing people that they wouldn’t fail would have greater practical implications than convincing people that they would succeed. This would imply that alleviating HALP students’ pessimism would be a stronger motivator than fostering optimism.

Additional steps need to be undertaken to better understand the motivational deficits that occur between High-Ability-Low-Performing (HALP) students and High-Ability-High-Performing (HAHP) students, particularly how and why these motivational deficits develop in HALP students. Although it is clear that HALP and HAHP groups exist, it is unclear at what point these groups diverge, or whether transitions between groups occur. A future avenue of inquiry could be a longitudinal study of HALP and HAHP students to better understand how stable or unstable motivational profiles remain over time. Additionally, further investigation is needed to determine whether or not there are predictable occurrences (e.g., school transitions, the onset of adolescence, family dissolutions, etc.) that impact the motivational profiles of high-ability students. Finally, with respect to social reproduction, an investigation of the relationship between HALP student expectations and the expectations of their parents would be worth pursuing. In any event, recognition of the HALP student phenomenon represents a conversation that is just beginning.