گسترش مدل های انگیزش پیشرفت سلسله مراتبی: نقش نیازهای انگیزشی برای اهداف پیشرفت و عملکرد تحصیلی

In the current study, we investigated the role of three basic motivational needs (need for power, affiliation, achievement) as antecedents of goals within the 2 × 2 achievement goal framework, and examined their combined predictive validity with regard to academic performance in a sample of 120 university students. Structural equation modeling analysis largely supported our postulated model, linking motivational needs indirectly to course grades through goals. Achievement goals were formed by a combination of different motives: need for achievement was a positive predictor of all four achievement goals, and need for affiliation was negatively related to performance-approach and performance-avoidance goals. Additionally, need for power was a positive predictor of performance-avoidance goals. Performance-approach goals had a direct (positive) effect on performance outcomes. In sum, our results integrate basic motivational needs with the achievement goals literature and extend therefore hierarchical achievement motivation models, by showing how basic human motives of achievement, affiliation, and power are related to goal striving motivation and performance outcomes in an academic setting.

In the field of achievement motivation, achievement goals belong to the most frequently studied variables with important implications for performance in various settings, such as schools, sports, or work (Payne, Youngcourt, & Beaubien, 2007). Achievement goals act as proximal antecedents of behavior that energize, direct, and guide behavior in achievement situations and are generally considered manifestations of more abstract dispositions (DeShon & Gillespie, 2005). One such disposition in particular, the achievement motive, has received considerable research attention (Thrash and Hurst, 2008 and Urdan, 1997) and has been integrated into hierarchical models of achievement motivation (Elliot, 1999). However, the role of other basic human motives as antecedents of achievement goals, especially with regard to social motives, is less well understood, and the interplay of goals within the 2 × 2 achievement goal framework (Elliot & McGregor, 2001) with motivational needs as predictors of academic performance has not been sufficiently investigated yet. The main aim of the current study was to extend hierarchical achievement motivation models with reference to three fundamental motivational needs. In detail, we investigated the role of need for achievement, affiliation, and power (McClelland, 1987) as antecedents of achievement goals. To provide a comprehensive test of the link of motives with goals, we applied the 2 × 2 achievement goal framework (Elliot & McGregor, 2001) by integrating approach and avoidance aspects of both mastery and performance goals in our study. Furthermore, we investigated their combined effect with basic motives on academic performance. 1.1. Motivational needs and achievement goals The 2 × 2 achievement goal framework (Elliot & McGregor, 2001) holds that achievement goals refer to either mastery (i.e. development of competence) or performance (i.e. demonstrating competence) goals, that can be further divided into approach or avoidance goals. Striving to demonstrate high ability and validate one’s competence in relation to others refers to performance-approach goals (PAP), whereas strategies that prevent the individual from showing a lack of competence or obtain unfavorable judgments from others indicate performance-avoidance goals (PAV). Similarly, mastery-approach goals (MAP) refer to the striving to master new tasks and the approach of success, whereas mastery-avoidance goals (MAV) focus on the avoidance of task-based or intrapersonal incompetence (e.g. avoid to do worse than before, or avoid losing one’s skills). Various variables have been suggested to act as antecedents of achievement goals. For example, prior studies seem to support the role of individual differences (in terms of stable disposition) as antecedents of (situation specific) goals (Bipp et al., 2008 and Payne et al., 2007). Given that several conceptualizations of goals draw from the achievement motivation literature, it seems valuable to examine in-depth the relationships of basic motivational needs with achievement goals on a theoretical and empirical level. Although Urdan (1997) already emphasized that definitions of achievement goals show resemblance to certain motives, to date a comprehensive empirical examination of the relationship of fundamental motivational needs and goals is lacking. Motives refer to fundamental, stable dispositions that direct a person’s behavior through the appraisal of action consequences. Building on Murray’s (1938) and McClelland’s (1987) motivation theories, three classes of needs, “the Big Three of motivation” (Schultheiss & Brunstein, 2001, p. 72), have received extensive attention in this research field: need for achievement, i.e. the need to do something better; need for affiliation, i.e. the need for friendly relationships; and need for power, i.e. the need to have an impact. Already Atkinson and Reitman (1956) highlighted that motives influence the striving for certain goals. More recently, Elliot and colleagues (e.g. Elliot, 1999 and Thrash and Hurst, 2008) took an important step by integrating the literatures on motives and achievement goals. Within hierarchical achievement motivation models, goals are seen as midlevel constructs that act as proximal antecedents of behavior through which individuals pursue their more general motives. Therefore, goals are linked to, but conceptually different from motives or needs that act as their antecedents. Elliot, Conroy, Barron, and Murayama (2010) proposed that the same goal can be used to realize different motivational needs, and that one specific motivational need can be channeled through different goals. Yet, only one motive has been incorporated in these models: need for achievement (or its subaspect fear of failure) has been shown to influence achievement outcomes indirectly via achievement goals (Elliot & Church, 1997). However, the connection of the other basic motivational needs with achievement goals is less well understood. We expected vital relationships of all three motivational needs (achievement, affiliation, and power) with achievement goals. First, prior research supports relationships of motive dispositions with personal strivings (in terms of broad goals) (Emmons & McAdams, 1991). Second, the distinction of approach and avoidance dimensions in motives (e.g. hope to approach success, fear to avoid failure; McClelland, 1953) suggests strong associations with specific goal dimensions. Third, social comparison and the ambition to outperform others plays a central role especially for performance goals (Elliot, 2005). Therefore, we expected that besides the need for achievement also social motives trigger the pursuit of achievement goals. In the following, we outline the expected connections for the extension of hierarchical achievement motivation models by discussing construct definitions, and presenting prior research findings. 1.1.1. Need for achievement (nAch) Definitions of nAch refer to the need to do something better or faster, accomplish a difficult task, overcome obstacles, or to prove ones talent. In various conceptualizations, a standard of excellence is included ( McClelland, 1953 and Murray, 1938) that can relate either to oneself (to excel) or to others (to compete or surpass others). As such, nAch refers to the need “to establish competence relative to the task, self, and others” ( Baranik, Stanley, Bynum, & Lance, 2010, p. 269), and incorporates personal mastery and the tendency to compete with others. Therefore, we expected a positive relationship of nAch with performance and mastery goals. Although theoretical conceptualizations suggested connection of this motive predominantly to approach goals ( Elliot & Church, 1997), and prior findings resulted in mixed results for the avoidance dimension ( Nie & Liem, 2013), a recent meta-analysis supports also positive connections of nAch with MAV and PAV ( Baranik et al., 2010). Elliot and Murayama (2008) offered a potential explanation for such an effect with regard to MAV: to accomplish or succeed in a task sometimes also demands to avoid errors or mistakes, which we assumed should also be the case if one is trying to perform not worse than others (PAV). Hypothesis 1. nAch is positively related to (a) MAP, (b) MAV, (c) PAP, and (d) PAV. 1.1.2. Need for affiliation (nAff) Already in the 1930s, Murray (1938) identified a social need central to human functioning that refers to building and maintaining warm and close relationships. People with high nAff enjoy the company of others, feel a sense of involvement and belonging with a group. Support for a link of this motivational need with achievement goals can be found in early goal conceptualizations ( Urdan & Maehr, 1995) that highlight the potential conflict between social and achievement motives, and emphasize the role of goals that refer to conformity with norms, or the need to gain approval from others. Similarly, more recent models refer to relationally-based variables (e.g. the fear of being rejected by others; Elliot, 1999) as antecedents of achievement goals. Given that other-based standards and social comparison play a central role for performance goals ( Elliot & Murayama, 2008), we expected nAff to be negatively linked in particular to those goals. First, people with high scores on nAff have been shown to prefer noncompetitive situations ( McClelland, 1987), so that we expected them to report a lower tendency to perform better than others (PAP). Second, Murayama and Elliot (2012) found that the dispositional preference to compete in contrast to cooperate with others (a sign of low nAff), activates both approach and avoidance goals, and therefore also triggers (e.g. through anxiety) the tendency to avoid performing poorly compared to others (PAV). The need to belong to a certain group (nAff) might therefore not lead to high PAV, as performing worse than others might actually be an effective strategy to build and maintain relationships in academic settings. Also, Payne et al. (2007) supported that Agreeableness (a related trait of nAff) correlates negatively with such goals. Hypothesis 2. nAff is negatively related to (a) PAP, and (b) PAV. 1.1.3. Need for power (nP) The power motive refers to the desire to feel strong, and to have impact on or control others. Typically, such a need is expressed in forceful or assertive actions, such as aggression, and in attempts to influence, persuade, impress, or dominate others (Schmalt & Heckhausen, 2008). Given the common grounds in terms of a focus on others, we expected nP to relate to both types of performance goals. First, we expected nP, as a need to feel superior, to positively relate to the tendency to prove one’s ability in relation to others (PAP). Indeed, Chiaburu, Marinova, and Lim (2007) found that status striving, a motivational orientation in terms of power or dominance, was positively associated with such performance goals. Second, we expected that this need also triggers the goal to avoid looking incompetent or less able than others, therefore high values on nP should be associated with higher PAV. Hypothesis 3. nP is positively related to (a) PAP, and (b) PAV. 1.2. Academic performance Only a few prior studies have investigated the combined predictive validity of basic motives and achievement goals. However, they only used a three-dimensional goal model or single motives (e.g. Elliot and Murayama, 2008 and Steinmayr and Spinath, 2009). Therefore, validity evidence regarding a comprehensive prediction model integrating also nAff, nP, as well as mastery-avoidance goals is much needed. We extended hierarchical models of achievement motivation ( Elliot and Murayama, 2008 and Thrash and Elliot, 2001) and tested the combined effect of the three basic motivational needs and the four goal dimensions on academic performance. Based on the role outlined for nAch in these models (cf. Elliot, 1999), we expected that the three motivational needs serve as antecedents of achievement goals (H1–H3), and therefore indirectly influence performance. For achievement goals, we postulated conform the model test by Elliot and Murayama (2008), direct effects of PAP and PAV on course grades. Hypothesis 4. (a) PAP are positively and (b) PAV are negatively related to performance. Hypothesis 5. nAch, nAff, and nP will have indirect performance effects through achievement goals.

3. Results Table 1 displays descriptive statistics and intercorrelations of the study variables. Similar to findings reported for the original instrument (Schultheiss et al., 2009), the three motives correlated positively in our sample, with the highest correlation between nP and nAff. Inspection of grades indicated no systematic differences between responders; N = 120; M = 6.84; SD = .52; and non-responders in the study; N = 76; M = 6.60; SD = .43; ANCOVA: F(1, 193) = .05; p = .83. Table 1. Means (M), standard deviations (SD), and zero-order correlations between study variables. Descriptives Correlations M SD nAch nAff nP MAP MAV PAP PAV Need for achievement (nAch) 18.18 6.43 (.88) Need for affiliation (nAff) 10.21 4.34 .62⁎⁎ (.79) Need for power (nP) 16.31 5.96 .69⁎⁎ .74⁎⁎ (.77) Mastery-approach goals (MAP) 3.58 .70 .13 −.06 .01 (.72) Mastery-avoidance goals (MAV) 3.07 .80 .20⁎ .00 .11 .62⁎⁎ (.75) Performance-approach goals (PAP) 3.24 1.00 .32⁎⁎ .08 .25⁎⁎ .33⁎⁎ .35⁎⁎ (.92) Performance-avoidance goals (PAV) 3.41 .96 .26⁎⁎ −.02 .21⁎ .30⁎⁎ .40⁎⁎ .69⁎⁎ (.95) Course grade 6.84 .52 .06 −.08 .03 .14 .13 .31⁎⁎ .14 Notes: N = 120. Values in brackets indicate Cronbach’s Alpha. ⁎ p < .05. ⁎⁎ p < .01. Table options The standardized path coefficients for the research model are presented in Fig. 1. The proposed model provided adequate fit to the data (Browne and Cudeck, 1993 and Kline, 2005). Although chi-square was significant at the 5% level; χ2(20) = 34.28, p = .02; the relationship with the degrees of freedom indicated sufficient model fit; χ2/df = 1.71. CFI and IFI indicated good model fit with values above .95 (CFI = .96; IFI = .97). Also an RMSEA of .077 (90% confidence interval: .028–.120) indicated reasonable model fit. Regarding the control variables, cohort was a negative predictor of grades, with lower performance of those students who took the course later in their study program. Furthermore, gender was a predictor of MAP; female compared to male students reported higher scores. Full-size image (21 K) Fig. 1. Standardized solution for hypothesized model. Mastery-approach (MAP), mastery-avoidance (MAV), performance-approach (PAP), performance-avoidance goals (PAV). Control variables and error terms are not presented in order to simply presentation. ∗p < .05, ∗∗p < .01. Dashed lines: non-significant path coefficients. Figure options Seven of our eight postulated connections of motives and goals (H1–3) were supported by the data. Supporting H1, nAch was a positive predictor of MAP (β = .17, p = .05), MAV (β = .20, p = .03), PAP (β = .31, p = .01), and PAV (β = .28, p = .02). Supporting H2, nAff was a negative predictor of PAP (β = −.27, p = .04), and PAV (β = −.41, p < .01). H3 was only partially supported: nP was a positive predictor of PAV (β = .32, p = .02; H3b), but the postulated path to PAP was only marginally significant (β = .24, p = .08; H3a). Together, the motives explained 13% of the variance in MAP, 9% of the variance in MAV, and 17% of the variance in PAP and PAV. Together, the model variables explained 25.7% of the variance in course grades (cf. Supplementary results on prediction accuracy in Appendix A; based on grade groups, 41.67% of students were correctly classified). PAP was a positive predictor of grades (β = .34, p < .01; H4a), but the path of PAV was only marginally significant (β = −.19, p = .09; H4b). 1 Therefore, H4 was partly supported. To test the indirect effect of motives on performance (H5), we compared the proposed model to an alternative model including direct paths from nAch, nAff, and nP to grades. The overall fit of this model was less satisfying; χ2(17) = 32.83, p = .01; χ2/df = 1.93; CFI = .96; IFI = .96; RMSEA = .088. None of those additional paths reached a meaningful significance level and a Chi-square statistic revealed no significant improvement in model fit; Δχ2(3) = 1.45, p = .69. Therefore, our results suggest that the proposed model is superior in terms of parsimony, and support H5. Given the differing effects on achievement goals, none of the motivational needs exerted a significant positive or negative indirect effect on grades (p > .12; n = 200 bootstrapped samples).