One of the more useful aspects of effect sizes is that they can be transformed into an expected percentile point gain (or loss) for the strategy under investigation. The effect size reported by Tubbs (1986) of .50 is associated with a 19 percentile point gain. Thus, taking the findings at face value, one could infer that an average student in a group of students who were provided with specific learning goals would be at the 69th percentile of a group of students who were exposed to very general learning goals. Another way of saying this is that a student at the 50th percentile in a class that used nonspecific goals (an average student in that group) would be predicted to rise to the 69th percentile if he or she were provided very specific learning goals. In short, goal specificity is an important element to consider when trying to enhance student achievement.
In their 1990 meta-analysis of organizational studies, Locke and Latham found effect sizes that ranged from .42–.80 for specific instead of general goals (translating to a 16–29 percentile point gain). They argued that specific goals provide more concrete guidance for achievement that more general goals lack. A lack of concrete guidance creates ambiguity that students in school and laborers in the workforce simply have trouble translating into specific expected behaviors. Specific goals provide a clear direction for behavior and a clear indication of desired performance, and as such they serve as motivators.
More recently, Steve Graham and Dolores Perin (2007) conducted a meta-analysis of achievement in writing. They found five studies relating to goal specificity. Examples of goal specificity used in their study included a clearly established purpose in a writing assignment and the specification of product expectations. They found an average effect size of .70 for goal specificity, which translates to a 26 percentile point gain. Accordingly, Graham and Perin (2007) concluded that “assigning product goals had a strong impact on writing quality” (p. 464), but warned that although their conclusion was based on high-quality studies, their findings were drawn from only five studies and so should be interpreted cautiously.
Goal Difficulty
Students will perceive learning goals as more or less difficult depending on their current state of knowledge, their beliefs about what causes achievement, and their perceptions of their own abilities. Studies indicate that students are most motivated by goals they perceive as difficult but not too difficult. For example, Tubbs (1986) found an average effect size of .82 for difficult versus easy goals (translating to a 29 percentile point gain). The Locke and Latham (1990) meta-analysis found effect sizes of .52–.82 for difficult goals (a 20–29 percentile point gain), noting that “performance leveled off or decreased only when the limits of ability were reached or when commitment to a highly difficult goal lapsed” (p. 706). Goal difficulty may also moderate or change the effect of feedback on student achievement. For example, Avraham Kluger and Angelo DeNisi (1996) found that feedback as an instructional strategy is more effective when learning goals are at the right level of difficulty—challenging, but not too difficult.
Types of Learning Goals
In addition to their specificity and difficulty, learning goals vary in terms of their purposes and functions. Learning goals that emphasize mastery of content, or mastery goals, might enhance learning more than goals that specify attainment of a specific score, or performance goals. Noncognitive goals that involve students in cooperative tasks might have a unique effect of their own.
Mastery vs. Performance Goals
One well-investigated distinction regarding learning goals involves their overarching purpose; namely, mastery or performance. The first type, mastery goals, focuses on developing competence. The second type, performance goals, focuses on demonstrating competence by obtaining a specific score or grade (Kaplan, Middleton, Urdan, & Midgley, 2001).
This distinction between mastery goals and performance goals is subtle but profound in its implications. Performance goals will typically include a desired score or grade. For example, the following would be considered performance goals:
Students will obtain a grade of B or higher by the end of the grading period.
All students will be determined proficient or higher in reading by the end of the school year.
As these examples illustrate, performance goals don't describe content as much as they do a specific score or grade. Mastery goals, by definition, articulate the content that is to be learned. For example, the following are mastery goals:
Students will be able to use word segmentation and syllables to decode an unrecognized word.
Students will be able to compare ordinal numbers through the fifth position (that is, 1st, 2nd, 3rd, 4th, 5th).
Although each type of goal may be associated with increased student achievement, research indicates that mastery goals are typically associated with higher order learning and better retention than are performance goals, especially for more challenging content. For example, in his meta-analysis, Christopher Utman (1997) found an average effect size of .53 (a 20 percentile point gain) for mastery versus performance goals for grade school students completing a complex task. Research by Judith Meece (1991) revealed that teachers who used mastery goals in their classrooms promoted more meaningful learning, provided more developmentally appropriate instruction, and supported student autonomy more than did teachers with performance-oriented classrooms.
Noncognitive Goals
Much of the research on goals over the decades has focused on academic goals, sometimes referred to as cognitive goals. However, attention to noncognitive goals in education has increased in recent years. For example, a 2005 issue of Educational Assessment was devoted to noncognitive goals. In their introduction to the volume, editors Jamal Abedi and Harold F. O'Neil noted that “the affective (feeling) and psychomotor (doing) issues affect cognitive performance and are worthwhile domains of learning themselves” (p. 147). The remainder of the volume focused on the role of noncognitive goals such as motivation and affect in education.
Joseph Durlak and Roger Weissberg (2007) investigated the effects of after-school programs on noncognitive goals such as students' personal and social skills. They limited their analysis to programs that used “evidence-based” instructional strategies, which they defined as “well-sequenced” and “active.” Relative to well-sequenced, Durlak and Weissberg noted:
New skills cannot be acquired immediately. It takes time and effort to develop new behaviors and often more complicated skills must be broken down into smaller steps and sequentially mastered. Therefore, a coordinated sequence of activities is required that links the learning steps and provides youth with opportunities to connect these steps. Usually, this occurs through lesson plans or program manuals, particularly if programs use or adapt established criteria. (p. 10)
About active forms of learning, they noted:
Active forms of learning require youth to act on the material. That is, after youth receive some basic instruction they should then have the opportunity to practice new behaviors and receive feedback on their performance. This is accomplished through role playing and other types of behavioral rehearsal strategies, and the cycle of practice and feedback continues until mastery is achieved. These hands-on forms of learning are much preferred over exclusively didactic instruction, which rarely translates into behavioral change. (p. 10)
After examining ten studies that met their criterion of using evidence-based strategies, they concluded that after-school programs reduced problem behaviors and contributed significantly to student achievement and positive self-concept.
Durlak and Weissberg's study is noteworthy because it demonstrates that noncognitive goals