The Achievement Goal Theory (AGT) of the last decade has been widely used in educational psychology as it can interpret students’ attitudes and behavior towards academic learning. AGT suggests the trichotomous model, which can be discriminated into mastery goals, performance-approach goals and performance-avoidance goals, as this model is the most recognizable and acceptable (Elliot & Harackiewicz, 1996). Although 2x2 model was also suggested (e.g., Elliot & McGregor, 2001) it is not supported by all the researchers and it is not observed often especially in students (e.g., Lee & Bong, 2016).

Mastery goals are the most beneficial goals as they are related to adaptive academic patterns in order to comprehend deeply the task. Specifically, students who are mastery-oriented often show high self-efficacy (e.g., Barron & Harackiewicz, 2001; Lüftenegger et al., 2017; Stavropoulou et al., 2023), and they also use strategies in order to master totally the learning task (e.g. Senko, 2019). Furthermore, students who select mastery goals try more and engage in the learning process (e.g., Senko & Dawson, 2017).

On the other hand, performance-approach goals are not such beneficial as mastery goals as its learning outcomes are inconsistent. Students who adopt performance-approach goals try to achieve higher than classmates and endeavor to earn positive comments. Performance-approach goals are associated with positive results when students engage with the task that leads to high learning outcomes and often high achievement (e.g., Al-Emadi, 2001; Church et al., 2001; Pekrun et al., 2009). On the contrary, it may also be related to dysfunctional learning patterns, such as negative emotions (e.g. Bong, 2009) and disrupting behavior (e.g., Sideridis & Stamovlasis, 2014). It is worth mentioning that there are also researches which support no relationship between this specific goal and adaptive learning outcomes (e.g., Senko & Dawson, 2017).

The third achievement goal is the performance-avoidance goal. When someone is oriented towards this goal makes an effort to avoid presenting lack of skills. It is also related to low performance and low interest (e.g., Wimmer et al., 2018). Moreover, this specific category is not associated with adaptive patterns (e.g., Elliot & Hulleman, 2017). Performance-avoidance goals are also associated to disruptive attitudes and cheating behavior (Elliot, 2005; Sideridis, 2005). This goal is also connected with negative emotions, such as anxiety (Kaplan et al., 2002).

The last decades a new perspective has appeared in the scientific field which suggests the multiple goal perspective, which means that a person can adopt two or more goals at the same time. Specifically, there are four types of multiple goals, the specialized goal, the additive goal, the interactive goal and the selective goal (Barron & Harackiewicz, 2001). According to research, the combination between mastery goals and performance-approach goals is the most beneficial combination as students gain mastery of the task from the first goal and high performance from the second goal (e.g., Barron & Harackiewicz, 2001; Schmidt et al., 2020).

Perceived teachers goals

Apart from achievement goal orientations that students adopt in an academic environment, the stimuli that they perceive play an important role (e.g., Anderman & Maehr, 1994). Teachers often adopt goals and promote them to the learning environment, and these goals can predict students’ behavior (Kaplan & Maehr, 1999; Urdan, 2010). Furthermore, teachers’ goals are associated to students’ goals (e.g., Maulana et al., 2016). When students perceive that their teachers promote clear rules and instructions, recognize errors as part of the learning process and promote deep understanding, effort and mastery of the task then students perceive that their teachers promote mastery goals and for that reason they are mastery oriented (Diseth et al., 2012). According to research, it has been observed that students who are mastery oriented perceive mastery goals from their learning environment (e.g., Bardach et al., 2020). It is also worth mentioning that perceived mastery goals may predict students’ engagement (Gonida et al., 2009; Stavropoulou et al., 2023).

On the other hand, when students perceive that their teachers promote those students who have high grades and social comparison is fostered then they perceive that performance goals are promoted. Specifically, when they perceive that accomplishment is fostered they tend to adopt performance-approach goals, whereas when they perceive that the avoidance of failure is promoted then they adopt performance-avoidance goals (Schwinger & Stiensmeier-Pelster, 2011; Wolters, 2004).

Achievement goal orientations between genders and ages

Differences in goal orientations have been identified between genders. Girls, in particular, cooperate more easily than boys, have more perseverance and establish more social relationships (Guan et al., 2006). In addition, an inclination and better performance of girls in artistic activities and languages is identified, while boys in science (Öztürk & Gürbüz, 2013; Wigfield & Eccles, 2002). Several studies show that girls tend to adopt higher mastery goals compared to boys, while the reverse happens for achievement goals (Luo et al., 2011; Meece et al., 2003; Roeser et al., 1996; Schwinger & Wild, 2012). However, Linnenbrink-Garcia et al., (2018) did not identify gender differences in mastery goals and performance goals.

Students' motivation even varies depending on the grade (Diseth & Samdal, 2014). When children start school, e.g. during their primary schooling, they tend to adopt mastery goals, which varies as the grade level increases (Anderman & Young, 1994; Urdan & Midgley, 2003; Xiang & Lee, 2002). Still, studies in younger age students present both the performance-approach goal and the mastery goal, while in older age students’ performance-approach and performance-avoidance goals are identified (Senko & Dawson, 2017). High school students are less mastery-oriented and more performance-oriented (Anderman et. al., 1999; Linnenbrink-Garcia et al., 2018; Urdan & Midgley, 2003), which can be explained considering the high demands of the final high school examinations for students' admission to higher education based on a competitive assessment system. Increasing the mastery goal from one grade to another has the potential to bring many benefits yet decreasing it is noticeable and affects the student's performance negatively (Urdan & Midgley, 2003). This occurs primarily during the transition from primary to secondary school but is even more pronounced with the transition to high school.

According to Anderman and Anderman (1999), a decline in mastery goals occurs between ages 5 and 7, while Freeman and Anderman (2005) argue the opposite. Bru et al. (2010) report that, in a sample of Norwegian students, they found no significant changes between ages 5 and 10. Gillet et al. (2012) report a decrease in intrinsic motivation among students aged 9 to 12, while after age 15, an increase in intrinsic motivation becomes evident. In this particular case, the transition from one grade to another plays an important role again. It has even been identified that students' motivation and individual achievement goals are more important for children attending lower secondary school (middle school) than for those attending higher grades in secondary school (high school) (Diseth & Samdal, 2014).

Students in the early stages of adolescence want to have autonomy in their choices. Obvious changes are therefore identified in individual achievement goals and during students' adolescence (Bong, 2009; Givens-Rolland, 2012; Meece & Miller, 2001). During adolescence, changes also take place in students' behavior and psychology (Urdan & Midgley, 2003). These changes can be enhanced by modifying the school context (Ames & Archer, 1988). During adolescence, students turn their attention mainly to extracurricular activities and this is a major reason why there is a decline in their motivation. According to Urdan and Midgley (2003), students who transit to high school usually become more focused on performance goal and neglect mastery goal. Often, motivation is reduced and behavior is differentiated in adolescents when the learning context changes and achievement goals are promoted compared to mastery goals that existed previously (Meece et al., 2003). However, Gottfried et al. (2001) report that motivation increases during the transition from middle school to high school. Intrinsic motivation increases in the first grade of high school when students begin to become oriented toward specific disciplines (Diseth & Samdal, 2014). This may be the case as in high school students focus on subjects they are most interested in and deemed necessary for their later life and future career (Gillet et al., 2012). When students perceive a transition from the learning goal to the achievement goal or a decrease in the learning goal, there is a decrease in achievement (Urdan & Midgley, 2003).

Assessment in schools related to subject also plays an important role. The transition from middle school to high school is accompanied by more rigorous and standardized assessment, which causes anxiety for students and leads them to adopt performance goals (Anderman & Midgley, 1997). Stamovlasis and Gonida, (2018) report that in Greek primary schools, assessment is not very rigorous, which is not the case in secondary schools, where the subjects, and in particular Modern Greek Language, become more demanding, while assessment becomes more and more rigorous and increases even more in high schools.

Aim of the study and research questions

The present study was to investigate the psychometric properties of the perceived teachers’ goals scale, adopted for Greek student population, the measurement invariance and the differences across genders and grades. The following hypotheses were stated:

1. - The adopted version of perceived teachers’ goals scale, holds satisfactory psychometric properties and factorial validity.

2. -The three dimensions the perceived teacher goals (mastery, performance-approach, and performance-avoidance), demonstrate satisfactory internal consistency coefficients.

3. - There is measurement invariance of the dimensions of the scale under study across genders and grades.

4. - There are significant differences between genders and grade-groups in the above three dimensions of the scale.

Method

Participants

The participants in the present study were 2,045 students (aged 13 to 17 years; 49.8% girls) attending junior (N = 1,342) and senior (N = 703) secondary education. Specifically, 543 students were in 7th grade, 502 in 8th grade, and 297 in 9th grade, while 387 students were in 10th grade and 316 in 11th grade. The participants came from 25 schools in Central Macedonia, Greece, with 3 schools from the private sector and 22 from the public sector.

Instruments

The trichotomous model was formulated by employing the Patterns of Adaptive Learning Surveys [PALS] (Midgley et al., 1998) to evaluate three distinct orientations: mastery-goal orientation, performance-approach orientation, and performance-avoidance orientation. Mastery-goal orientation assesses a preference for learning-oriented tasks (e.g., "I find satisfaction in class activities that enhance my understanding, even if errors occur"). Performance-approach orientation measures the desire for success and recognition (e.g., "I would feel a sense of accomplishment if I were the only one capable of answering the teacher's questions"). Performance-avoidance orientation captures the concern for avoiding failure and appearing incompetent (e.g., "It's important to me to avoid looking foolish in front of my classmates"). Each of these orientations is evaluated using a series of survey items. Additionally, PALS (Midgley et al., 1998) was utilized to assess perceived teacher goal orientations, encompassing perceived mastery goals, perceived performance-approach goals, and perceived performance-avoidance orientation. These dimensions include items reflecting students' perceptions of their teachers' instructional priorities. Each dimension of perceived teacher goal orientations is evaluated using a set of five items on the scale.

Procedure

The sample of the current research is an opportunity sample. Ensuring that the research complies with ethical standards, the researcher communicated with the principals of the schools to inform them about the purpose of the research. Having collected consent forms, the questionnaires were distributed to students. They answered in their classroom, while the students who did not participate in the research were engaged in other activities. The questionnaire we used was 7-point Likert scale which included 15 items (5 were related to perceived mastery-approach, 5 were related to perceived performance-approach and 5 to performance-avoidance). This scale has also been used in several researches (Friedel et al., 2007; Stavropoulou et al., 2023; Wolters, 2004).

Analyses

The dimensionality and the structure of Patterns of Adaptive Learning Surveys [PALS] (Midgley et al., 1998) was demonstrated by Exploratory Factor Analysis (EFA) via Principal Axis Factoring (PAF) and consecutively by confirmatory factor analysis (CFA), where multiple fit indexes, such as the chi-square (X ² ), the comparative fit index (CFI), and the root mean square error of approximation (RMSEA), with the usual acceptable values: CFI ≥ 0.95, TLI ≥ 0.95 and RMSEA ≤ 0.05, were used (Geiser, 2013). Reliability was measured as internal consistency via the coefficients Cronbach’s alpha and McDonald’s omega.

The measurement invariance followed, included four steps. In the first step, the configural invariance, which is the least restrictive model is assessed and served as the base line. In each of the next steps, a more restricted model is examined and compared with the previous one. The second model, is named the metric invariance and concerns the values of factor loadings in each group, testing that the meaning of the construct and the factor variances and covariance are similar across the groups under study. The next model, the scalar invariance, examines whether the item intercepts are equivalent across groups, and if this does not hold, a bias effect might be present, signifying a significant difference between the groups in perceiving the essence of the construct under examination. The last, the strict invariance model concerns the invariance of factor variances and tests if the residual error is equivalent across groups The comparison between invariance models is based on the X ² difference test, in tandem with the differences in some core indexes: ΔCFI < 0.01 and ΔRMSEA < 0.015 for non-rejecting the null hypothesis (e.g., Chen, 2007; Cheung & Rensvold, 2002).

Results

Perceived teacher goals in the classroom

Exploratory Factor Analysis (EFA)

Principal axis factoring (PAF) oblique/promax rotation reveal the number of the underlying dimensions. Bartlett’s test of sphericity (χ² = 6165.156, p < 0.001) and the Kaiser–Meyer–Olkin index (0.813) suggested adequate variance for applying factor analysis. Parallel analysis, along with the Kaiser’s Criterion and the corresponding scree plot suggested three-factor structure. The proposed structure is three—dimensional and includes the items (with loading greater to 0.40) showed in Table 1.

The three factors correspond to teacher goals: Mastery, performance-approach and performance- avoidance with eigenvalues 1.899, 1.693, and 1.669 respectively, while the corresponding portions of variance explained were 14.60 %, 13.0 %, and 12.80 % respectively, while the total variance explained was 40.05 %.

Confirmatory Factor Analysis (CFA) –The measurement model

CFA was applied to the dimensions of perceived teacher-goals in the classroom, and results for the single-factor model were: χ² = 2313,923, df = 54, p < 0.001, CFI = 0.569, TLI = 0.474, RMSEA = 0.146, SRMR = 0.121, NFI = 0.565. The three-factor model fitted better and satisfactorily to the empirical data possessing the following fit measure indices: χ² = 252.095, df = 63, p < 0.001; CFI = 0.975; TLI = 0.968; RMSEA = 0.040; 90 % CI of RMSEA = [0.035; 0.045]; SRMR = 0.038; NFI = 0.967; GFI = 0.998]. Comparison of the two models by means of a χ² test revealed that the three-factor model was substantially improved over the single-factor model (Δχ² = 2061.83, df = 9, p < 0.001). Thus, the hypothesis of the unidimensional structure was rejected. In addition, by inspecting the standardized residual covariance matrix, which had values smaller than two, the absence of possible model misspecifications was assured. The calculations were carried out in R (via JASP). Table 2 shows the CFA measurement model of the goal structures.

Reliability analysis

Reliability measures of the three factors were computed using Cronbach’s Alpha (α) and McDonald’s omega (ω): Perceived teachers’ Master Approach goal (α = 0.725/ ω = 0.729), Perceived teachers’ Performance Approach goal (α = 0.746/ ω = 0.749), and Perceived teachers’ Performance Avoidance goal (α = 0.715/ ω = 0.723). These reliability indices suggest that the present measurements with the PALS-G sub-scales have a satisfactory degree of internal consistency, as was also presented from Gonida et al. (2009) [Table 1]. Omega is actually superior to the traditional Cronbach’s alpha measure of reliability in realistic condition reflecting the ratio of true over total variance. It is more appropriate index for non-tau equivalent measurements and more accurate estimation of internal consistency, especially when the assumption that each observable has an identical contribution to the measurement of the latent factor, is violated. We just report both the traditional measure alpha and McDonald's omega for comparison reasons.

Table 1 also shows the correlation matrix of the three dimensions, along with the means and the standard deviations of each factor. Perceived teachers’ Master Approach goal is negatively correlated with Perceived teachers’ Performance Approach goal (r = -0.259, p < 0.001) and the latter is positively correlated with Perceived teachers’ Performance Avoidance goal (r = 0.337, p < 0.001).

Measurement invariance

Having ensured validity issues with CFA, measurement invariance was carried out for the two genders. The measurement invariance is a potential hypothesis in psychometrics and for the present scale and population it it addressed for the first time. Table 3 summarizes measurement invariance for gender. The chi-square difference (Δ.²) test, when comparing each of the invariance models: configural, metric, scalar, and strict invariance model with its predecessor, showed that p-values are statistically significant, which might imply a small biases, nevertheless, inspecting the differences in CFI, TLI, RMSEA and SRMR, which are negligible, it be concluded that overall the measurement invariance holds for gender.

Measurement invariance was carried out for the six grade-groups, ranging between grade A-gymnasium to grade B-lyceum. Table 3 summarizes measurement invariance for grades. The chi-square difference (Δ.²) tests showed statistically significant .-values, nonetheless, the differences in CFI, TLI, RMSEA and SRMR, are negligible, and thus the overall measurement invariance can be assumed for grades.

Testing the differences among individual characteristics

Differences among individual characteristics were tested via two-way analysis of variance (ANOVA) using the grade and gender as independent variables predicting the perceived teacher goals in the classroom. The analysis revealed statistically significant differences, however with low effect sizes. For the perceived teacher mastery goals, the main effects of gender (F = 12.68, p < 0.001, η² = 0.006) and grade (F = 10.47, p < 0.001, η² = 0.02) are statistically significant, while the interaction gender*grade is not. Figure 1 shows the perceived teachers goals in the classroom as a function of grade and gender. Regarding the perceived mastery goals, girls, in general, appear to perceived higher mastery goals and this appears constant across grades, while, except grade A Gym, boys have lower perceived teacher mastery goals in lyceum.

Regarding the perceived teacher performance-approach goals, the main effects of gender (F = 8.68, p< 0.01, η² = 0.004), grade (F = 18.45, p < 0.001, η² = 0.035) and the interaction term gender*grade (F= 2.68, p < 0.05, η² = 0.005), are statistically significant. Girls tend to declare lower perceived teacher performance-approach goals, while this perceived performance-avoidance goals goal decreases in gymnasium and decreases in lyceum. Regarding the perceived teacher performance-avoidance the main effects of gender (F = 35.47, p < 0.001, η² = 0.017) and grade (F = 19.29, p < 0.001, η² = 0.036) are statistically significant, while the interaction gender*grade is not. Girls tend to declare lower perceived teacher performance-avoidance, goals, while a consistent decline for both genders is observed from grades A-Gymnasium to B-Lyceum.

Discussion

The adopted version of the perceived teacher goals questionnaire, implemented via PALS (Patterns of Adaptive Learning Surveys) exhibits satisfactory psychometric properties and factorial validity. The study found that the three dimensions of the perceived teacher goals orientations, namely mastery, performance-approach, and performance-avoidance, demonstrated acceptable internal consistency (Hypotheses 1, 2, 3). In addition, measurement invariance across genders and grade levels was demonstrated, ensuring that the relevant concepts are similarly perceived across genders and grade groups (Hypothesis 4). The results indicate that the scale under study is based on a solid theoretical foundation and could be implemented as a valid means for measuring personal goals and perceived teacher goals in educational settings. Moreover, differences in the mean values of the six dimensions of PALS between the two genders and among grade-groups were found via analysis of variances (Hypothesis 5). Summarizing the current results, can be highlighted that girls adopt more mastery goals than boys, whereas boys tend to adopt higher performance-avoidance goals, results that are in line with previous findings (Diseth & Samdal, 2014; Luo et al., 2011). Performance-approach goals do not differ across genders. Differences among grade-groups were also documented. Mastery goals presented a variability across grades, with an increase from gymnasium to lyceum for both genders. Performance-avoidance goals also showed an increase in gymnasium and a decrease in lyceum for both genders, while performance-approach goals demonstrated a consistent decrease across grades.

Analogously, the perceived teachers’ goals our findings revealed that girls perceive that their teachers adopt more mastery goals than boys, whereas boys perceive that their teachers promote more performance-approach and performance-avoidance goals. The perceived performance-approach goals are higher in gymnasium and a lower in lyceum for both genders, while performance- avoidance a goal demonstrated a consistent decrease across grades. These finding suggest that boys and girls, as well as students from different grade levels, may exhibit distinct patterns of personal goal orientations and perceived teacher goals. Further analysis is needed to understand the underlying factors contributing to these differences, which could include societal norms, academic expectations, or motivational influences. The current study also has several limitations that stem from the use of an opportunity sampling procedure and from the fact that one set of cross-sectional data was implemented and the lack of additional test–retest reliability measures. Moreover, issues of content validity might be raised for scales with a small number of items.

However, the finding are in line and support previous reports on PALS implementation (Leondari & Gonida, 2008), while providing more established conclusions by using a large sample and including measurement invariance. The present survey, which has a methodological dimension, bridges the gap between theoretical conjectures and real-world application, by providing a certified instrument that secures valid measurement. Testing the differences across individual characteristics can provide valuable insights to educators and policymakers to design targeted interventions that foster positive goal orientations and academic success for all students. Additionally, researchers should investigate whether these differences persist over time and how they impact academic performance and overall comfort in educational settings.

References

Notes

^* Research article.

Author notes

^a Correspondence author. Email: gnstavro@edlit.auth.gr

Additional information

How to cite: Stavropoulou, G., & Stamovlasis, D. (2025). Perceived Teachers’ Goals Scale: Psychometric Properties, Measurement Invariance and Differences across Genders and Grades. Universitas Psychologica, 24, 1-13. https://doi.org/10.11144/Javeriana.upsy24.ptgs

Descargar