Defensive pessimism, self-esteem and achievement goals: A person-centered approach.
Although there is ample evidence that defensive pessimism is effective in the short term but extremely detrimental for emotional well-being in the long run (Norem & Chang, 2000), one of the issues that aroused more controversy is the role played by self-esteem in the functioning of this strategy.
So far, the few studies that have examined this issue show inconsistencies. Some research argues that the self-esteem of defensive pessimists is far from being high, as it is lower than that of individuals who use strategic optimism (Eronen, Nurmi, & Salmela-Aro, 1998), but not lower than the self-esteem of those who use other self-protective strategies such as self-handicapping (Rodriguez, Cabanach, Valle, Nunez, & Gonzalez-Pienda, 2004). From this viewpoint, it is postulated that defensive pessimists tend to think negatively, not only regarding the tasks to be addressed, but also with regard to themselves (Norem, 2001). Other work, however, suggest that defensive pessimists' self-esteem fluctuates (Martin, Marsh, & Debus, 2001; Yamawaki, Tschanz, & Feick, 2004), being low in some situations and high in others. Ferradas et al. (2016) have also shown that the self-esteem-defensive pessimism relationship is not linear, such that high self-esteem would be related to a greater use of defensive pessimism in males, whereas in women, this strategy would be more recurrent when self-esteem is low.
This second position opens the possibility of different profiles of defensive pessimists, such that some of them show high self-esteem and others, low. In fact, Valle, Cabanach, Rodriguez, Nunez, and Gonzalez-Pienda (2005) suggest the possibility that students with low self-esteem and those who have high self-esteem could both use self-protection strategies. The former, so that a new failure will not erode their already battered self-esteem; on the other hand, students with high self-esteem would need to protect themselves to preserve their high standard of self-esteem.
The previously reviewed studies have analyzed the self-esteem-defensive pessimism relationship by adopting a variable-based approach, which does not determine the possible existence of student profiles that combine these two variables. Under this approach, the present work has as its first aim the adoption of a person-centered approach to identify different profiles that combine defensive pessimism and self-esteem, an unpublished issue to date. Such an approach would provide a more realistic view of students' motivation (Schwinger & Wild, 2012). Drawing on previous research (with a variable-centered approach), we hypothesize the existence of two profiles of defensive pessimistic students: one with low self-esteem and the other with high self-esteem.
To test the validity of the identified profiles, as the second goal, we determine whether the groups found differ in their achievement goals. The few existing precedents essentially link defensive pessimism to performance goals (Elliot & Church, 2003; Rodriguez et al., 2004; Yamawaki et al., 2004). However, while some works (Rodriguez et al., 2004) argue that defensive pessimism is related to the desire to avoid negative social judgments (performance-avoidance goals), others (Elliot & Church, 2003) indicate that in this strategy, there is also an underlying interest to exhibit competition against others (performance-approach goals).
It is also unclear whether defensive pessimism is positively related to the desire to achieve new knowledge (learning goals). In this sense, several studies have shown a low correlation between this type of goals and defensive pessimism (Elliot & Church, 2003; Valle et al., 2007; Yamawaki et al., 2004). However, Martin et al. (2003) through qualitative interviews, observed some interest in learning in defensive pessimistic students, although it seems that it was more frequently used as a means to perform. In a similar vein, Ferradas, Freire, Nunez, Pineiro, and Rosario (2017) found greater use of defensive pessimism in students who combine learning with the two performance tendencies (approach and avoidance).
Whereas some works (e.g., Gebka, 2014; Phan, 2010) relate self-esteem positively with learning and performance-approach goals and negatively to performance-avoidance goals, we expect that the profile with high defensive pessimism and high self-esteem will use learning and performance goals to a greater extent. On the contrary, we expect that the profile with high defensive pessimism and low self-esteem will show higher levels of performance-avoidance goals.
The Universidade da Coruna (Spain) has 17,227 students, of whom 1087 (6.3%) were selected through incidental sampling. Fifty-six cases were eliminated due to missing data. Also, using Mahalanobis' distance method (Hair, Anderson, Tatham, & Black, 1995), we identified three cases presenting outliers, so they were also eliminated. Thus, the final sample was made up of 1,028 students ([M.sub.age] = 21.36, S[D.sub.age] = 3.81). Of the participants, 86.3% were women. With respect to their qualification, 69.9% of the participants studied Educational Sciences, and 30.1%, Health Sciences.
We used in the Spanish version of the Defensive Pessimism Questionnaire (Norem, 2002). The 12-item instrument (e.g., "Considering what can go wrong helps me to prepare") has shown good reliability ([alpha] = .89). Students' responses were rated on a Likert scale (1 = never to 5 = always).
We used the Spanish adaptation of Martin-Albo, Nunez, Navarro, and Grijalvo (2007) of the Rosenberg Self-esteem Scale (Rosenberg, 1965). The 10-item instrument (e.g., "In general, I'm satisfied with myself") has shown adequate reliability (a = .88) in our study. Responses were rated on a Likert scale (1 = strongly disagree to 5 = strongly agree).
Using the Spanish adaptation (Jover, Navas, & Holgado, 2014) of the Goal Orientation Scale (Skaalvik, 1997), we assessed four types of achievement goals: learning goals (6 items; e.g., "It is important for me to learn new things in class"), performance-approach goals (5 items; e.g., "I try to get better grades than others"), performance-avoidance goals (6 items; e.g., "When I answer incorrectly in class, what worries me most is what my classmates think of me"), and work-avoidance goals (4 items; e.g., "In class, I prefer to do as little as possible"). The internal consistency of the factors ranged between [alpha]= 76 (work avoidance) and [alpha] = .85 (performance approach). The participants' responses were rated on a Likert scale (1 = never to 5 = always).
After obtaining permission from the relevant university departments, data were gathered in the classrooms where the students receive their university training, within the academic schedule. The participants were informed about the goals of the study, the voluntary nature of their participation, and the anonymity and confidentiality in the processing of the information obtained. The questionnaires were applied by trained personnel in a single session without time limit. In order to reduce the effect of the order of presentation of the instruments, a full counterbalance was performed. For this purpose, the questionnaires were administered randomly according to six combinations, each of which contained the three scales used in a different position.
The profiles of defensive pessimism and self-esteem were identified by a two-step cluster analysis (Everitt, Landau, Leese, & Stahl, 2011). In a first phase, hierarchical analysis was performed using the Ward method and the intragroup bonding method. The Ward method uses an F value to maximize the significance of the differences between clusters, making it the method with the greatest statistical power (Milligan & Hirtle, 2003). To check the stability of the solutions obtained, we performed a second hierarchical analysis, using the intragroup bonding method (Hair & Black, 2002). From the interpretation of the resulting dendrograms of these two methods, and in order to refine the allocation of subjects to each group and obtain the final group solution, in a second phase, we performed a non-hierarchical analysis (k-means). Finally, we conducted two separate MANOVAs to examine both the robustness of the identified profiles and the differences between the profiles in terms of achievement goals. We used the Games-Howell tests as post-hoc contrast tests. The magnitude of the effect was determined by Cohen's d statistic and partial eta squared. All the analyses were performed with the SPSS 24 statistical software.
Table 1 presents the descriptive statistics and correlations between the variables of the study. The skewness and kurtosis rates indicated that all the variables met the criteria of normality (see Finney & DiStefano, 2006). The correlation matrix also showed that all of the correlations were statistically significant. Bartlett's sphericity test results also indicated that the variables were sufficiently intercorrelated, [[chi].sup.2](15) = 2203.53, p < .001.
Identification of the profiles
The results of hierarchical cluster analysis (Ward's method and intragroup bonding method) showed two dendrograms with a similar structure when three and four groups were considered. Taking this information into account, as well as the change in the cluster coefficient, we decided that the best solution was to divide the sample into four groups. With reference to these four clusters, in a second phase, we performed an iterative procedure of k-means. After this analysis, 68 cases (6.61%) were reassigned to other groups, leading to better differentiated and more uniformsized segments.
The solution contained a first group (Cluster 1) of 149 students (14.49%) characterized by high self-esteem and a moderate level of defensive pessimism (HSE/MDP). A second group (Cluster 2) of 147 students (14.3%) with low self-esteem and low defensive pessimism (LSE/LDP). The third profile (Cluster 3), comprising 522 students (50.78%), was characterized by high self-esteem and low defensive pessimism (HSE/LDP). Finally, Cluster 4 was made up of 210 students (20.43%) with low self-esteem and high defensive pessimism (LSE/HDP). Figure 1 shows a graphic representation of these four profiles.
To analyze the robustness of the identified profiles, we performed a MANOVA, using defensive pessimism and self-esteem as dependent variables. The multivariate effect of the cluster was statistically significant, [[lambda].sub.Wilks] = .027, F(6, 2046) = 1741.43, p< .001, [[eta].sub.p.suo.2] = .836. Table 2 shows the means and standard deviations (raw and standardized scores) of the four profiles in the two classification variables, as well as the univariate tests for each variable.
Intergroup differences in Achievement Goals
Table 3 shows the means and standard deviations of each profile in the four achievement goals, as well as the univariate tests for each goal. Also included are the results of the post hoc contrasts (Games-Howell).
The MANOVA yielded statistically significant differences between the profiles of all four achievement goals, [[lambda].sub.Wilks] = .345, F(12, 2701.60) = 111.42, p < .001, [[eta].sup.p.sub.2]= .299. The effect size was medium in the case of performance-avoidance goals, and large in the other three goals.
Considering the post-hoc contrasts, in work-avoidance goals, the HSE/LDP group obtained significantly higher scores than the other groups, although the differences were only large compared to LSE/HDP (d = 1.74). In fact, the differences between the LSE/HDP group and the HSE/MDP and LSE/LDP groups were also large (d = 1.41 and d = 1.09, respectively). In learning goals, the LSE/HDP group obtained significantly higher scores, with differences with the other groups ranging between d = 0.62 (HSE/MDP) and d = 1.56 (LSE/LDP). In performance-avoidance goals, the LSE/LDP group obtained significantly higher scores, observing moderate (d = 0.51 with group LSE/HDP) to large differences (d = 0.97 with group HSE/LDP; and d = 1.06 with group HSE/MDP) with the other groups. As for performance-approach goals, there was a significantly higher score in both groups with low self-esteem (LSE/HDP and LSE/LDP), with large differences with the other two groups (between d = 1.23 and d = 1.80).
The main contribution of this work is the adoption of a person-centered approach to analyze the role of self-esteem in the functioning of defensive pessimism. This approach, unpublished to date, has allowed us to identify two profiles of defensive pessimistic university students: one with low self-esteem and the other, more moderate in the use of the strategy, which showed high self-esteem. We also identified two other profiles of students who do not use defensive pessimism, one with low self-esteem and the other with high self-esteem. These results reinforce those recently obtained by Ferradas et al. (2016), indicating that the need to protect oneself through defensive pessimism may involve students who like themselves and students who do not.
Considering that defensive pessimists generally fit the profile of a "good student" (Garcia, 1995)--efficient use of learning strategies (Suarez, 2014), high levels of self-regulation (Elliot & Church, 2003), good performance (Suarez & Fernandez, 2011)--it may be surprising that some defensive pessimists have low self-esteem. A possible explanation would be that, among good students, some of them are overly self-critical, and the fear of failing predisposes them to protect themselves through defensive pessimism (Pullman & Allik, 2008). Another plausible argument would incur in the trend of defensive pessimists to think negatively, both about the tasks they must face and about themselves (Norem, 2001). These negative self-appraisals would acquire a strategic value (Martin et al., 2003), helping defensive pessimists to "be on guard" to avoid an unfavorable outcome that would compromise their personal worth.
The fact that, as our findings indicate, some students with high self-esteem use defensive pessimism can also be disconcerting. As opposed to the students who match the high self-esteem and low defensive pessimism profile, these students may value themselves, but these feelings would be characterized by their inconsistency. Therefore, if it is assumed that defensive pessimism is a strategy that is eminently "activated" in ego-threatening situations (Thompson & le Fevre, 1999), these students, who are insecure about their personal worth, would find in defensive pessimism a strategy to preserve their high, but weak, self-esteem when they feel threatened by a potential failure. This explanation would be consistent with the results that link defensive pessimism with unstable self-esteem (Martin et al., 2001; Yamawaki et al., 2004). However, given that the measurement instrument of self-esteem used in our study does not evaluate its degree of stability, the explanation remains a mere hypotheses that should be confirmed in future work.
In addition, our results indicate that the two profiles of defensive pessimists (HSE/MDP and LSE/HDP) differ partially in their achievement motivations. Thus, in comparison with the remaining identified profiles, the students of the LSE/HDP group show the highest levels of learning and performance-approach goals. On another hand, they also show fairly high levels (only exceeded by the LSE/LDP group) of performance-avoidance goals, as well as the lowest levels (differing broadly from the other groups) of work-avoidance goals. This finding confirms the description in other studies (e.g., Ferradas et al., 2017; Martin & Marsh, 2003) of the defensive pessimist as a student who is cognitively committed to the fear of failure but behaviorally focused on success. In this sense, as suggested by other works (Norem & Cantor, 1986; Suarez, 2014), it seems that, in defensive pessimistic students (in our case, in those with low self-esteem), dedication and effort shape the path to overcome their fears.
Defensive pessimists with high self-esteem seem to develop a different motivational achievement trajectory. These students show moderately high levels of learning goals (although lower than students of the LSE/HDP group), but also a moderately high desire (only exceeded by the HSE/LDP group) to avoid academic work. However, in comparison with the rest of the profiles identified, students of the HSE/MDP group do not seem particularly interested in comparing themselves with their peers (i.e., performance-approach and performance-avoidance goals). According to these results, we are faced with an atypical profile of defensive pessimists with regard to prior research linking this strategy to performance orientation (Elliot & Church, 2003;
Yamawaki et al., 2004). However, given that the two groups with high self-esteem (HSE/MDP and HSE/LDP) developed almost identical achievement motivations, it is possible that the moderate levels of defensive pessimism of the HSE/MDP group are indicative of a very occasional use of this self-protective strategy (only in certain tasks or markedly threatening situations). From this consideration, the scarce use of defensive pessimism would not influence the achievement goals of the students who have this profile. Future work should specifically analyze this assumption.
Altogether, the results of this study contribute substantially to the study of the relationship of defensive pessimism-self-esteem, identifying two profiles of defensive pessimists who are differentiated in their self-esteem. In addition, our findings indicate that these two profiles follow different motivational achievement trajectories. These contributions have important psychoeducational implications. Assuming that the emotions, cognitions, and motivations of the students constitute an integrated whole (Aritzeta et al., 2016), it is necessary to identify the factors that contribute to reducing the high levels of anxiety experienced by the defensive pessimists. Among them are control beliefs (Valle et al., 2015).
The limitations of the study should lead us to carefully consider these contributions. Firstly, the cross-sectional nature of the research carried out does not allow us to establish causal relationships between the variables. This issue must be addressed in future work through more appropriate designs. Secondly, the sampling procedure (incidental) limits the internal and external validity of the results. Therefore, our findings should be confirmed in future studies using more rigorous sampling techniques. Also, the fact of only including students of Educational Sciences and Health Sciences, together with the remarkable preponderance of women in these degrees, hinders the generalizability of the results. As a result, future lines of research could focus on gender differences in the relationship between self-esteem and defensive pessimism profiles and achievement goals. Finally, our study has only considered academic achievement goals. Successive works should also incorporate social motivations.
Aritzeta, A., Balluerka, N., Gorostiaga, A., Alonso-Arbiol, I., Haranburu. M., & Gartzia, L. (2016). Classroom emotional intelligence and its relationship with school performance. European Journal of Education and Psychology, 9, 1-8. doi:10.1016/j.ejeps.2015.11.001
Elliot, A. J., & Church, M. (2003). A motivational analysis of defensive pessimism and self-handicapping. Journal of Personality, 71(3), 369-396. doi:10.1111/1467-6494.7103005
Eronen, S., Nurmi, J., & Salmela-Aro, K. (1998). Optimistic, defensive pessimistic, impulsive, and self-handicapping strategies in university environments. Learning and Instruction, 8, 159-177 doi:10.1016/S0959-4752(97)00015-7
Everitt, B. S., Landau, S., Leese, M., & Stahl, D. (2011). Cluster analysis, 5th edition. New York: Wiley.
Ferradas, M. M., Freire, C, Nunez, J. C, Pineiro, I., & Rosario, P. (2017). Motivational profiles in university students. Its relationship with self-handicapping and defensive pessimism strategies. Learning and Individual Differences, 56, 128-135. doi:10.1016/j.lindif.2016.10.018
Ferradas, M. M., Freire, C, Valle, A., Nunez, J. C, Regueiro, B., & Vallejo, G. (2016). The relationship between self-esteem and self-worth protection strategies in university students. Personality and Individual Differences, 88, 236-241. doi:10.1016/j.paid.2015.09.029
Finney, S. J., & DiStefano, C. (2006). Non-normal and categorical data in structural equation modeling. In G. R. Hancock & R. O. Mueller (Eds.), Structural equation modeling. A second course (pp. 269-314). Greenwich, CT: Information Age.
Garcia, T. (1995). The role of motivational strategies in self-regulated learning. In R R. Pintrich (Ed.), Understanding self-regulated learning (pp. 29-42). San Francisco: Jossey-Bass.
Gebka, B. (2014). Psychological determinants of university students' academic performance: An empirical study. Journal of Further and Higher Education, 38(6), 813-837 doi:10.1080/0309877X.2013.765945
Hair, J. F, Anderson, R., Tatham, R., & Black, W. (1995). Multivariate data analysis (4th ed.). Upper Saddle River, NJ: Prentice Hall.
Hair, J. F, & Black, W. C. (2002). Cluster analysis. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding more multivariate statistics (pp. 147-205). Washington, DC: American Psychological Association.
Jover, I., Navas, L., & Holgado, F P. (2014). Goal orientations in the students of the Education Faculty of Alicante. International Journal of Developmental and Educational Psychology, 1(2), 575-584.
Martin, A. J., & Marsh, H. W. (2003). Fear of failure: Friend or foe? Australian Psychologist, 38(1), 31-38. doi:10.1080/00050060310001 706997
Martin, A. J., Marsh, H. W., & Debus, R. L. (2001). Self-handicapping and defensive pessimism: Exploring a model of predictors and outcomes from a self-protection perspective. Journal of Educational Psychology, 93(1), 87-102. doi:10.1037/0022-06184.108.40.206
Martin, A. J., Marsh, H. W., Williamson, A., & Debus, R. L. (2003). Self-handicapping, defensive pessimism and goal orientation: A qualitative study of university students. Journal of Educational Psychology, 95(3), 617-628. doi:10.1037/0022-06220.127.116.117
Martin-Albo, J., Nunez, J. L., Navarro, J. G., & Grijalvo, F. (2007). The Rosenberg Self-Esteem Scale: Translation and validation in university students. Spanish Journal of Psychology, 10(2), 458-467 doi:10.1017/S1138741600006727
Milligan, G. W., & Hirtle, S. C. (2003). Clustering and classification methods. In J. Schinka & W. Velicer (Eds.), Handbook of psychology: Research methods in psychology (Vol. 2, pp. 165-186). New York: Wiley.
Norem, J. K. (2001). Defensive pessimism, optimism, and pessimism. In E. C. Chang (Ed.), Optimism and pessimism: Implications for theory, research, and practice (pp. 77-100). Washington, DC: American Psychological Association.
Norem, J. K. (2002). El poder positivo del pensamiento negativo: utiliza el pesimismo defensivo para reducir tu ansiedad y rendir al maximo [The positive power of negative thinking: Using defensive pessimism to harness anxiety and perform at your peak]. Barcelona: Paidos Iberica.
Norem, J. K. (2008). Defensive pessimism as a self-critical tool. In E. C. Chang (Ed.), Self-criticism and self enhancement: Theory, research and clinical applications (pp. 89-104). Washington, DC: American Psychological Association.
Norem, J. K., & Cantor, N. (1986). Anticipatory and post hoc cushioning strategies: Optimism and defensive pessimism in "risky" situations. Cognitive Therapy and Research, 10(3), 347-362. doi:10.1007/BF01173471
Norem, J. K., & Chang, E. C. (2000). A very full glass: Adding complexity to our thinking about the implications and applications of optimism and pessimism research. In E. C. Chang (Ed.), Optimism and pessimism: Implications for theory, research and practice (pp. 347-367). Washington, DC: American Psychological Association.
Phan, H. P. (2010). Students' academic performance and various cognitive processes of learning: An integrative framework and empirical analysis. Educational Psychology, 30(3), 297-322. doi:10.1080/01443410903573297
Pullmann, H., & Allik, J. (2008). Relations of academic and general self-esteem to school achievement. Personality and Individual Differences, 45(6), 559-564 doi:10.1016/j.paid.2008.06.017
Rodriguez, S., Cabanach, R. G., Valle, A., Nunez, J. C, & Gonzalez-Pienda, J. A. (2004). Diferencias en el uso del self-handicapping y pesimismo defensivo y sus relaciones con las metas de logro, la autoestima y las estrategias de autorregulacion [Differences in the use of self-handicapping and defensive pessimism and their relationships with achievement goals, self-esteem, and self-regulation strategies]. Psicothema, 16(4), 625-631.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University.
Sanna, L. J. (2000). Mental simulation, affect, and personality: A conceptual framework. Current Directions in Psychological Science, 9(5), 168-173. doi:10.1111/1467-8721.00086
Schwinger, M., &Wild, E. (2012). Prevalence, stability, and functionality of achievement goal profiles in mathematics from third to seventh grade. Contemporary Educational Psychology, 37(1), 1-13. doi:10.1016/j.cedpsych.2011.08.001
Skaalvik, E. M. (1997). Self-enhancing and self-defeating ego orientation: Relations with task avoidance orientation, achievement, self-perceptions, and anxiety. Journal of Educational Psychology, 89(1), 71-81. doi:10.1037/0022-0663.891.71
Suarez, J. M. (2014). Optimistic and defensive-pessimist students: Differences in their academic motivation and learning strategies. Spanish Journal of Psychology, 17, e26, 1-8. doi:10.1017/sjp.2014.27
Suarez, J. M., & Fernandez, A. P. (2011). Assessment of students' affective-motivational self-regulatory strategies: The MSLS-VS. Anales de Psicologia, 27, 369-380.
Thompson, T, & le Fevre, C. (1999). Implications of manipulating anticipatory attributions on the strategy use of defensive pessimists and strategic optimists. Personality and Individual Differences, 26(5), 887-904 doi:10.1016/s0191-8869(98)00190-l
Valle, A., Cabanach, R. G., Rodriguez, S., Nunez, J. C, &Gonzalez-Pienda, J. A. (2005). Self-worth protection strategies in higher educational students: Exploring a model of predictors and consequences. In R. Nata (Ed.), New directions in higher education (pp. 99-126). New York: Nova Science.
Valle, A., Nunez, J. C, Cabanach, R. G., Rodriguez, S., Gonzalez-Pienda, J. A., & Rosario, P. (2007). Metas academicas y estrategias motivacionales de autoproteccion [Academic goals and self-protection motivational strategies]. Electronic Journal ofResearch in Educational Psychology, 13(5), 617-632.
Valle, A., Regueiro, B., Rodriguez, S., Pineiro, I., Freire, C, Ferradas, M., & Suarez, N. (2015). Motivational profiles as a combination of self-efficacy expectations and academic goals in university students. European Journal of Education and Psychology, 8(1), 1-8. doi:10.1016/j.ejeps.201510.001
Yamawaki, N, Tschanz, B. T, & Feick, D. L. (2004). Defensive pessimism, self-esteem instability and goal strivings. Cognition and Emotion, 18, 233-249. doi:10.1080/02699930341000004
Maria del Mar Ferradas Canedo, Carlos Freire Rodriguez, Bibiana Regueiro Fernandez, and Antonio Valle Arias
Grupo de Investigacion en Psicologia Educativa (GIPED) Universidade da Coruna
Corresponding author: Carlos Freire Rodriguez
Facultad de Ciencias de la Educacion
Universidade da Coruna
15071 A Coruna (Spain)
Received: May 28, 2017 * Accepted: November 21, 2017
Table 1 Descriptive statistics and correlation matrix of the variables of the study 1 2 3 1. DEF_PES -- 2. SE -.59 (***) -- 3. WAG -.49 (***) .54 (***) -- 4. LG .28 (***) -.12 (***) -.52 (***) 5. PApG 43 (***) -.46 (***) -.10 (***) 6. PAvG .11 (***) -.22 (***) .15 (***) M 2.35 3.41 2.70 SD 0.87 0.52 0.99 Skewness 0.83 -0.39 0.18 Kurtosis -0.49 -1.41 -1.00 4 5 6 1. DEF_PES 2. SE 3. WAG 4. LG -- 5. PApG -.18 (***) -- 6. PAvG -.37 (***) .56 (***) -- M 3.24 3.30 3.24 SD 1.00 0.93 0.87 Skewness -0.45 -0.51 -0.60 Kurtosis -0.65 -0.71 0.05 Note: PES_DEF = defensive pessimism; SE = self-esteem; WAG= Work avoidance goals; LG = learning goals; PApG = Performance-approach goals; PAvG = performance-avoidance goals Table 2 Descriptive statistics and univariate tests of the four profiles in self-esteem and defensive pessimism HSE/MDP LSE/LDP HSE/LDP n = 149 n = 147 n = 522 14.49% 14.3% 50.78% M SD M SD M SD Self-esteem Raw score 3.73 0.20 2.75 0.10 3.78 0.19 Defensive Z-score 0.62 0.39 -1.29 0.20 0.71 0.37 pessimism Raw score 2.69 0.37 2.09 0.53 1.74 0.27 Z-score 0.39 0.43 -0.30 0.60 -0.69 0.31 LSE/HDP n = 210 20.43% M SD [F.sub.(3,1024)] [[eta].sub.p. sup.2] Self-esteem 2.74 0.11 2861.51 (***) .893 Defensive pessimism -1.29 0.21 3.78 0.36 1746.19 (***) .836 1.64 0.41 Note: HSE/MDP = high self-esteem/moderate defensive pessimism; LSE/LDP = low self-esteem/low defensive pessimism; HSE/LDP = high self-esteem/low defensive pessimism; LSE/HDP = low self-esteem/high defensive pessimism. Table 3 Descriptive statistics and univariate tests of the four profiles in each achievement goal HSE/MDP (a) LSE/LDP (b) HSE/LDP (c) M SD M SD M SD WAG 2.84 0.72 2.57 1.14 3.11 0.79 LG 3.29 0.78 2.44 1.19 3.21 0.84 PApG 2.99 0.93 3.86 0.67 3.06 0.93 PAvG 2.81 0.88 3.86 0.60 2.93 0.87 LSE/HDP (d) M SD [F.sub.(3,1024) [[eta].sub.p.sup.2] Post hoc(n.s.) WAG 1.67 0.71 153.80 .311 (***) a-b LG 3.86 0.95 70.94 .172 (***) a-c PApG 3.44 0.45 45.30 .111 (***) a-c PAvG 4.17 0.33 182.66 .349 (***) a-c Note: HSE/MDP = high self-esteem/moderate defensive pessimism; LSE/LDP = low self-esteem/low defensive pessimism; HSE/LDP = high self-esteem/low defensive pessimism; LSE/HDP = low self-esteem/high defensive pessimism; WAG = work avoidance goals; LG = learning goals; PApG = performance-approach goals; PAvG = performance-avoidance goals (***)p < .001; n.s. = nonsignificant