- Email: [email protected]
The differential impact of agency and pathway thinking on goal pursuit and university exam performance
Personality and Individual Differences 58 (2014) 20–25
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
The differential impact of agency and pathway thinking on goal pursuit and university exam performance Monique F. Crane ⇑ Macquarie University, Department of Psychology, Building C3A, Sydney, NSW 2109, Australia
a r t i c l e
i n f o
Article history: Received 3 June 2013 Received in revised form 18 September 2013 Accepted 24 September 2013 Available online 17 October 2013 Keywords: Hope theory Agency thinking Pathway thinking Exam performance
a b s t r a c t The present study examines the interaction between agency and pathway thinking on performance outcomes. The study used a repeated-measures design to examine the role of agency and pathway thinking on goal pursuit emotions (e.g., determination), secondary appraisal, and final exam performance in a group of university psychology students. Consistent with previous mental health research (Arnau, Rosen, Finch, Rhudy, & Fortunato, 2007; Cramer & Dyrkacz, 1998), the present findings suggest a dominant role for agency thinking in performance. Moreover, there was a reliable interaction between pathway and agency thinking in the prediction of goal pursuit and performance. The interactions consistently revealed that when agency thinking was high, pathway thinking was generally irrelevant to our various measures of goal pursuit. These findings challenge the additive role of agency and pathway thinking suggested by hope theory (Snyder, 2002). Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Dispositional hope has been identified as an important predictor of academic success and goals pursuit (Snyder et al., 2002). Hope is defined as, ‘‘the perceived capability to derive pathways to desired goals, and motivate oneself via agency thinking to use those pathways’’ (Snyder, 2002, p. 249). As the definition implies, hope integrates agency and pathway thinking. Agency is the motivation to pursue goals and the belief in one’s capacity to achieve desired goals (Snyder, 2002; Snyder et al., 2002). Pathway thinking is the development of routes to goal achievement. According to hope theory, the most hopeful individuals are those high on both pathway and agency thinking, both measured in the Dispositional Hope Scale (DHS; Snyder et al., 1991). Those high in hope are anticipated to be the most directed toward goal pursuit and goal success (Snyder et al., 1991). However, research examining the independent roles of agency and pathway thinking demonstrates findings somewhat inconsistent with the predictions outlined by hope theory (Snyder, 2002). In particular, hope theory anticipates that the best outcomes in terms of mental health and goal attainment would be for individuals with both high agency and pathway thinking; that make an additive contribution to overall dispositional hope. In contrast, to the proposed additive role of agency and pathway thinking previous research finds that pathway thinking plays a minimal role in predicting mental health outcomes (e.g., Arnau et al., 2007; Cramer & Dyrkacz, 1998). Empirical work suggests a principle role for agency over pathway thinking in ⇑ Tel.: +61 2 9850 8604; fax: +61 2 9850 8062. E-mail address: [email protected] 0191-8869/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.paid.2013.09.026
terms of mental health outcomes. The present study, sought to add further clarity to this debate by examining the independent and interactive role of pathway and agency thinking in relation to goal pursuit and goal achievement (i.e., university exam performance). 1.1. The role of agency and pathway thinking on performance and goal attainment To date, most studies of performance have been restricted to an analysis of overall hope. Previous work demonstrates that dispositional hope is positively related to students’ goal setting and attainment, as well as appraisals of future attainment (Snyder et al., 1991) educational and sporting achievement (Curry, Snyder, Cook, Ruby, & Rehm, 1997; Gilman, Dooley, & Florell, 2006; Snyder et al., 2002), and task performance (Peterson, Gerhardt, & Rode, 2006). A single study examines the independent roles of agency and pathway thinking in goal attainment. Feldman, Rand, and Kahle-Wrobleski (2009) found that goal-specific agency thinking, not pathway thinking, predicted goal attainment. Thus, akin to studies examining mental health, agency cognitions were more critical to goal attainment than the perception of goal pathways. 1.2. A possible interaction between pathway and agency thinking The above analysis of previous work has identified the possibly unique roles of agency and pathway thinking in predicting mental health and performance outcomes. Consistent with this idea, hope theorists have identified the possible unique independent roles of agency and pathway thinking (Snyder, 2002). Given the independence of agency and pathway thinking it is plausible for an individual
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
to be high in agency, but low in pathway thinking and vice versa (Snyder, 2002). An individual with high agency, but low pathway thinking may be motivated toward goal achievement, but fail to identify a clear strategy and thus motivation remains uncultivated (Snyder, 2002). Conversely, an individual with low agency and high pathway thinking may perceive a clear route to goal achievement, but lack sufficient self-belief and personal drive to motivate goal pursuit (Snyder, 2002). The various combinations of agency and pathway levels may have varying impacts on goal pursuit and performance. Hope theory does not specify how goal pursuit and attainment are impacted by mismatched agency and pathway thinking. However, it is implicit in hope theory that being high on either agency or pathway thinking is more beneficial than being low on both. Initially, hope theory implies no interaction between agency and pathway thinking. Snyder (2002) suggests that agency and pathway thinking have an equally antagonistic role when there is a mismatch in their use. However, the antagonism resulting from mismatched agency and pathway thinking has never been formally investigated. Moreover, given previous research demonstrating the principal independent effect of agency, a case can be made that when agency thinking is high, pathway thinking may be less crucial to performance outcomes. 1.3. The present study To date, no studies have examined the possible differential impact of dispositional agency and pathway thinking on performance. Curry et al. (1997) point to the importance of both goal related cognitions and emotions to goal success. Thus, in order to examine performance holistically this study examined objective performance outcomes (i.e., exam results), but also related secondary appraisals of goal attainment (i.e., perceived control over exam performance). Chang and DeSimone (2001) found that the hope construct was related to secondary appraisals, but not primary appraisal of exams. Moreover, the present study will examine positive approach emotions related to goal achievement (e.g., determined). Hope theory suggests that people high in hope, approach goals with a positive emotional state (Snyder, 1995). Research examining performance goals demonstrates that positive affectivity functions to increase goal directed behaviour promoting achievement of goals (Bagozzi & Pieters, 1998). The present study has two core aims. The first is to investigate the independent roles of agency and pathway thinking in predicting goal pursuit. The second is to examine the interaction between agency and pathway thinking in relation to the measures of goal pursuit. In particular, this paper seeks to explore whether agency and pathway thinking are additive in facilitating performance and to explore whether incongruent agency and pathway cognitions result in decreased performance. The following tentative hypotheses are made: H1. Agency thinking will moderate the relationship between pathway thinking and exam performance. When agency is high, pathway thinking will be unrelated to exam performance. In contrast, when agency is low there will be a positive relationship between pathway thinking and exam performance.
H2. Agency thinking is anticipated to moderate the relationship between pathway thinking and positive approach emotions (e.g., determined). When agency is high, pathway thinking will be unrelated to the experience of positive approach emotions. In contrast, when agency is low there will be a positive relationship between positive approach emotions and pathway thinking.
H3. Agency thinking is expected to moderate the relationship between pathway thinking and control over exam performance. When agency is high, pathway thinking will be unrelated to the experience of perceived control over exam performance. In contrast, when agency is low there will be a positive relationship between perceived control over exam performance and pathway thinking.
21
2. Method 2.1. Participants and design A convenience sample of Introduction to Psychology students was used. Of the 654 students enrolled, 481students (28.9% male; 71.1% female; Mage = 20.42; SDage = 5.33) attempted the Time 1 survey. Of these students, 306 gave permission to access their final exam grades, but 29 of these did not complete the exam. Thus, 277 participants were included in the analysis of exam performance (28% male; 72% female; Mage = 20.42; SDage = 5.33). The Time 2 survey was completed the week of the final exam. Only 98 students (20.4%) attempted this survey (22.2% male; 77.8% female; Mage = 20.95; SDage = 6.01). A chi-square confirmed that the Time 1 and Time 2 gender ratio was not statistically different (X2(1) = 2.401, p = .156).
2.2. Materials and procedure Time 1 data was collected as part of 26 tutorial classes. Time 2 data was collected via email invitation within the week of the final exam. All self-report measures except for gender, age, conscientiousness and neuroticism were measured at both time points. These measures were included as covariates, in the model of performance, because of their relationship to exam performance or dispositional hope (e.g., Chapell et al., 2005; Duff, Boyle, Dunleavy, & Ferguson, 2008). Dispositional hope was measured using the DHS (Snyder et al., 1991) a 12-item measure of hope consisting of four agency and pathway items, and four distracter items. Respondents were asked to indicate on a scale from 1(definitely false) to 4(definitely true) the degree to which these statements describe them across time and situations. The internal reliability of the pathway and agency sub-constructs was satisfactory (a = .69 and a = .70, respectively). Studies examining the factor structure of the DHS have yielded mixed results regarding whether a one or two-factor structure is most appropriate (Snyder et al., 1991). Having noted this, analyses on student populations appear to consistently suggest that a twofactor model comprised of the agency and pathway sub-constructs is a better fit to the data than a one-factor model (Babyak, Snyder, & Yoshinobu, 1993; Snyder et al., 1991; Roesch & Vaughn, 2006). A confirmatory factor analysis was carried out to confirm the twofactor structure of the scale. Conscientiousness and neuroticism were measured using the two item conscientiousness and neuroticism sub-scales from the Ten Item Personality Index (TIPI; Gosling, Rentfrow, & Swann, 2003). This measure consists of adjectives (e.g., disorganised) and the participant is required to indicate the degree to which these attributes describe themselves on a scale from 1(strongly disagree) to 7(strongly agree). The neuroticism scale demonstrated satisfactory internal reliability (a = .70); however, the internal reliability for the conscientiousness scale was below the satisfactory level (a = .62). The low alpha’s are expected given that only two items are used per dimension (Ehrhart et al., 2009). Positive approach emotions were measured using four positive approach emotions from the Positive and Negative Affect Scale (PANAS; Tellegen, Watson, & Clark, 1988): determined, excited, active, and attentive. Participants rated the extent that they had felt these emotions in the last week in relation to the psychology final exam. Ratings were made on a five-point response scale 1(very slightly or not at all) to 5(extremely). These items demonstrated satisfactory internal reliability (Time 1 a = .73 and Time 2 a = .76). Perceived control of exam performance (secondary appraisal) was designed for the purpose of this study. Four items measured perceived control of psychology exam performance (e.g., ‘‘I control
22
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
how well I will do in the psychology exam’’). Participants indicated the degree each statement applied to them on a four-point scale from 1(does not apply to me) to 4(applies to me completely). The scale demonstrated satisfactory internal reliability (Time 1 a = .80 and Time 2 a = .88). A principal components analysis was also conducted and confirmed that a single factor emerged at both time points. 2.3. Analysis strategy Hierarchical linear regressions were used to examine the three dependent variables: exam performance, Time 2 positive approach emotions and Time 2 perceived control. All predictor variables were mean centred and gender was dummy coded (0 = female, 1 = male). MODPROBE was conducted using the Johnson–Neyman technique (Hayes & Matthes, 2009) in order to investigate twoway interaction effects. MODPROBE allows the examination of the independent variable’s impact on the dependent variable at different values of the moderator (Hayes & Matthes, 2009). 3. Results 3.1. Preliminary analysis The agency and pathway sub-constructs were moderately positively correlated (r = .45, p < .005). Also interesting was that Time 1 agency thinking was related significantly to Time 2 positive approach emotions (r = .38, p < .005) and Time 2 perceived control (r = .46, p < .005), but had no relationship with final exam performance. Time 1 pathway thinking demonstrated no significant relationship with the predictors. The full set of correlations is provided in Table 1. Responders versus non-responders only varied on perceived conscientiousness (p < .002); a greater level of conscientiousness was found in responders (M = 5.38; SD = 1.43) compared to nonresponders (M = 4.80; SD = 1.27). 3.2. Confirmation of two-factor structure The relative fit of the one and two-dimensional DHS measurement model were compared. The two-dimensional measurement model was the best fit to the data. The two-dimension model yielded fit statistics that generally exceed benchmarks (comparative fit index [CFI] = .97; Root Mean-Square Error of Approximation [RMSEA] = .05; Tucker-Lewis Index [TLI] = .95). In contrast, fit indices for the one-dimensional model indicated less fit adequacy (comparative fit index [CFI] = .94; Root Mean-Square Error of Approximation [RMSEA] = .07; Tucker-Lewis Index [TLI] = .19). The Akaike Informa-
tion Criterion (AIC) for the models indicated that the two-dimensional model was a better fit to the data (one-dimensional AIC = 115.32; two-dimensional AIC = 90.3). 3.3. Analysis of exam performance A hierarchical linear regression examined the impact of the agency and pathway sub-constructs on exam performance. Age, sex, neuroticism and conscientiousness were initially included as covariates. Only age, neuroticism and conscientiousness were significant predictors and consequentially included in Step 1. Step 2 included Time 1 pathway thinking. Step 3 included Time 1 agency thinking and the final step included the two-way interaction. Table 2 presents the results for this analysis. The final model including the two-way interaction significantly improved on the more parsimonious models (R2change ¼ :014, Fchange (1, 269) = 4.442, p < .04). The final model revealed a significant main-effect for age (t = 2.91, p < .005; R2part ¼ :03; b = .17) indicating that on average older students had better exam performance compared to younger students. Both neuroticism and conscientiousness positively predicted exam performance (t = 2.85, p < .006; R2part ¼ :03; b = .17;t = 2.11, p < .04; R2part ¼ :01; b = .14, respectively). A significant main-effect for the agency sub-construct indicated that greater agency prior to the exam was predictive of better final exam performance (t = 2.18, p < .04; R2part ¼ :02; b = .15). As predicted (H1), there was a significant two-way interaction between Time 1 pathway and agency (t = 2.11, p < .04; R2part ¼ :01; b = .12). The interaction is illustrated in Fig. 1. The MODPROBE revealed that when agency thinking was or below .22 (mean centred) a significant negative relationship between pathway and final exam performance emerged. However, at higher values of agency there was no significant relationship between pathway and exam performance. 3.4. Analysis of positive approach emotions Twelve cases with more than 5% missing data on critical variables and five multivariate outliers were removed leaving 81 cases in the analysis. Time 2 positive approach emotions were examined using a hierarchical linear regression model; only Time 1 positive approach emotions was included as a covariate in Step 1. Table 3 provides the hierarchical regression for this analysis. The final model including the two-way interaction demonstrated a marginally significant improvement on more parsimonious models (R2change ¼ :03, Fchange (1, 76) = 3.75 p = .057). There was a strong significant positive main-effect for Time 1 approach emotions (t = 4.84, p < .001; R2part ¼ :21; b = .48). Consistent with predic-
Table 1 Bivariate correlations, means and standard deviations for measures.
1. Exam performance 2. Age 3. Conscientiousness 4. Neuroticism 5. Time 1 pathway 6. Time 1 agency 7. Time 1 positive approach emotions 8. Time 2 positive approach emotions 9. Time 1 perceived control 10. Time 2 perceived control * **
p < .05. p < .005.
2
3
4
5
6
7
8
9
10
M
SD
.15 1.0
.25 .26* 1.0
.14 .10 .39** 1.0
.13 .07 .13 .26* 1.0
.20 .17 .59** .35* .45** 1.0
.07 .09 .41** .24 .08 .39** 1.0
.09 .05 .34* .03 .08 .38** .58** 1.0
.25 .08 .10 .08 .20 .39** .05 .06 1.0
.22 .02 .22 .06 .17 .46** .09 .14 .75** 1.0
27.90 20.42 4.90 3.23 11.99 12.07 9.72 9.68 3.28 3.42
6.07 5.33 1.42 1.52 1.49 1.85 2.68 2.95 .60 .61
23
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25 Table 2 Hierarchical regression analysis of exam performance. Step 1
(Constant) Age Neuro. Consc. Pathw. (P) Agenc. (A) PA DR2 DF * **
Step 2
B
SE
27.86 .19 .67 .99
.35 .07 .23 .25
Step 3
b
t
B
SE
.17 .17 .23
80.12 2.83** 2.85** 3.88**
27.87 .18 .63 1.00 .24
.35 .07 .24 .25 .23
.104 10.54**
Step 4
b
t
B
SE
.16 .16 .23 .06
80.11 2.79** 2.66** 3.93** 1.04
27.85 .18 .71 .67 .40 .56
.35 .07 .24 .29 .24 .25
.004 1.08
b
t
B
SE
b
t
.16 .18 .16 .10 .16
80.62 2.80** 3.00** 2.29* 1.67 2.26*
27.68 .19 .68 .61 .39 .54 .21 .014 4.44*
.35 .07 .24 .29 .24 .25 .10
.17 .17 .14 .10 .15 .12
78.54 2.91** 2.85** 2.11* 1.63 2.18* 2.11*
.017 5.11*
p < .05. p < .01.
Fig. 1. Interaction between agency and pathway in the prediction of final exam performance (out of 50).
Fig. 2. Interaction between agency and pathway in the prediction of positive approach emotions.
3.5. Analysis of perceived control over exam performance tion (H2), there was a marginally significant interaction between pathway and agency (t = 1.94, p = .057; R2part ¼ :03; b = .20). The nature of this interaction is presented in Fig. 2. The relationship between pathway thinking and positive approach emotions only approached significance at very high and very low levels of agency thinking, but the relationship was never significant at p < .05. Thus, the interaction appeared to be driven by the low perceived positive approach emotions at Time 2 when both pathway and agency thinking were low.
One case with more than 5% missing data and three multivariate outliers were removed leaving 94 participants in the analysis. Time 2 perceived control over exam performance was analysed using the same procedure as above. Time 1 perceived control and was the only covariate included in the model. The hierarchical regression analysis is present in Table 4. The final model including the two-way interaction demonstrated a significant improvement on the previous model
Table 3 Hierarchical regression analysis of positive approach emotions. Step 1 B (Constant) App. Em. Pathw. (P) Agenc. (A) PA DR2 DF **
s
p < .01. p < .06.
9.25 .52
.27 29.73**
Step 2
Step 3
Step 4
SE
b
t
B
SE
b
t
B
SE
b
t
B
SE
b
t
.27 .10
.52
33.71 5.45**
9.26 .53 .09
.28 .10 .17
.52 .05
33.49 5.44** .49
9.10 .47 .01 .30
.29 .10 .18 .18
.47 .01 .17
31.23 4.60** .02 1.65
9.26 .48 .16 .27 .21 .03 3.75s
.30 .10 .20 .18 .11
.48 .09 .15 .20
31.05 4.84** .84 1.47 1.94s
.002 .24
.03 2.73
24
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
Table 4 Hierarchical regression analysis of positive approach emotions. Step 1 B (Constant) Perc. Contr. Pathw. (P) Agenc. (A) PA DR2 DF * **
Step 2 SE
3.37 .76
.51 81.22**
.05 .09
Step 3
b
t
B
SE
.71
69.02 9.01**
3.37 .75 .02
.05 .09 .03
Step 4
b
t
B
SE
.70 .04
68.40 8.67** .47
3.34 .72 >.01 .06
.05 .09 .03 .03
.001 .24
b
t
B
SE
b
t
.67 >.01 .17
66.20 8.24** .04 2.06*
3.38 .73 .04 .06 .05 .04 8.01*
.05 .08 .03 .03 .02
.68 .10 .15 .23
67.37 8.71** 1.20 1.89 2.83*
.026 4.24*
p < .05. p < .01.
(R2change ¼ :04, Fchange (1, 89) = 8.01, p < .005). There was a strong significant positive main-effect for Time 1 perceived control (t = 8.71, p < .001; R2part ¼ :42; b = .68). The predicted (H3) two-way interaction between pathway and agency thinking also emerged (t = 2.83, p < .04; R2part ¼ :04; b = .23). The interaction is illustrated in Fig. 3. Consistent with H3, the relationship between pathway thinking and perceived control over exam performance was positive at lower values of agency at or below .85 (mean centred). Interestingly, at very high values of agency (at or above 2.44 mean centred) a negative relationship between pathway thinking and perceived control emerged. 4. Discussion 4.1. Summary of findings Collectively, the results suggest that agency thinking was the most reliable predictor of goal pursuit and actual performance. H1 was partially supported, as anticipated higher agency thinking resulted in a non-significant relationship between pathway and final exam performance. Thus, high motivation toward goal completion (e.g., I energetically pursue my goals) and anticipated goal completion (e.g., I meet the goals I set for myself) meant that pathway thinking became unrelated to exam performance. Inconsistent with prediction, when agency thinking was low there was a negative relationship between pathway thinking and exam performance. This finding is consistent with the suggestion that a mismatch between agency and pathway thinking may undermine performance
(Snyder, 2002). Alternatively, asking participants to consider their pathways at Time 1 may have elicited engagement in positive fantasies. Indulging in positive fantasies has been shown to elicit poorer performance compared to engaging in positive expectations (Kappes & Oettingen, 2011; Oettingen & Mayer, 2002). The examination of positive approach emotions generally confirmed prediction (H2). While the interaction was similar to that anticipated, it was driven by the lack of Time 2 positive approach emotions when both agency and pathway thinking were low. While this particular finding was consistent with the predictions of hope theory (Snyder, 2002); high agency and pathway thinking did not yield the greatest positive approach emotions. In contrast, pathway thinking appeared to play no role in the experience of positive emotions when agency thinking was high. It was only when agency thinking was low that pathway thinking tended to increase the experience of positive approach emotions. The findings for Time 2 perceived control again generally confirmed prediction (H3). When agency thinking was low, pathway thinking positively predicted perceived control over exam performance. In fact, when agency was high (>1.5 SD), pathway thinking negatively predicted perceived control over exam performance. The finding for high agency is curious and not observed for the other goal-pursuit outcome measures; therefore it may not be robust and perhaps an experimental artefact. Extending the work of Chang and DeSimone (2001), the current findings suggest that agency thinking has a dominant impact on secondary appraisal; pathway thinking appears predictive of secondary appraisal only when agency thinking is low. In contrast, the findings are generally inconsistent with the predictions of hope theory (Snyder, 2002) that suggest high agency and pathway thinking would provide the greatest level of hope and therefore perceived control over goal outcomes.
4.2. Theoretical contributions
Fig. 3. Interaction between agency and pathway in the prediction of perceived control over exam performance.
This paper sought to address two core aims. First, the paper explored the potentially dominant role of agency cognitions in facilitating goal pursuit. Second, to examine whether the two following proposals made by hope theory have merit: (1) agency and pathway thinking play an additive role in motivating goal pursuit and (2) whether incongruent agency and pathway cognitions result in decreased goal pursuit. While the nature of the interactions varied to some degree depending on the outcome variable, this does not detract from the central theoretical point of the analysis. Across all three measures of goal pursuit when agency thinking was high, pathway thinking tended to decrease in its significance as a predictor of goal pursuit and performance outcomes. There was no evidence, in this analysis, of an additive relationship between agency and pathway thinking as suggested by hope theory (Snyder, 2002). Moreover, the results consistently suggest that if an individual engenders
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
high agency thinking and low pathway thinking he/she will not necessarily suffer problematic outcomes in terms of goal pursuit or performance. In contrast, the findings for high pathway thinking and low agency thinking were not as consistent across outcome measures. For perceived control and positive approach emotions it appeared that high pathway thinking compensated for low agency thinking. Conversely, when it came to actual performance outcomes the combination of low agency and high pathway thinking resulted in the lowest performance. These findings suggest that the antagonism between the low agency and high pathway thinking may be specific to actual performance outcomes. However, high pathway thinking may be beneficial when agency is low in terms of the emotional and cognitive aspects of goal pursuit. 4.3. Limitations and future directions The present results are limited by the response attrition observed for the Time 2 survey. The high level of attrition is likely to be, in part, a consequence of the proximity of the Time 2 survey to the final exam. High attrition is likely to increase selection bias reducing the generalisability of the findings to other populations. Having noted this, the results still generally confirm prediction indicating merit in the findings. A further limitation is the use of first year psychology students, which again reduces the level of generalizability. Thus, the findings need to be interpreted with caution and future research should seek to replicate the findings in additional samples. The current study was conducted over a short time span. The 6 week period sought to cover a relatively stressful time for psychology students where the final exam was increasing in salience. As such, this time period is certainly worthy of study. However, it would be also useful to study longer time periods. A further limitation of this study is the level of situational ambiguity. Participants were in their second semester at university and there was perhaps a lower level of ambiguity than compared to students in semester one. Pathway thinking may be more important to goal pursuit and performance when there is a higher level of situational ambiguity where pathways to goals become necessary. Future research should seek to examine the interaction between the level of situational ambiguity and pathway thinking on performance outcomes. 4.4. Conclusion The findings suggest little support for the additive function of agency and pathway thinking in terms of increasing goal pursuit and performance. In contrast, agency thinking appears to provide the greatest benefit to goal pursuit. Moreover, inconsistent with the predictions of hope theory (Snyder, 2002), there was very little evidence to support the detrimental impact of incongruent thinking styles. Generally, these findings suggested that a closer investigation of the complex interaction between agency and pathway thinking is required by further research.
25
References Arnau, R. C., Rosen, D. H., Finch, J. F., Rhudy, J. L., & Fortunato, V. J. (2007). Longitudinal effects of hope on depression and anxiety: A latent variable analysis. Journal of Personality, 75, 43–64. Babyak, M. A., Snyder, C. F., & Yoshinobu, L. (1993). Psychometric properties of the Hope Scale: A confirmatory factor analysis. Journal of Research in Personality, 27, 154–169. Bagozzi, R. P., & Pieters, R. (1998). Goal-directed emotions. Cognition & Emotion, 12, 1–26. Chang, E. C., & DeSimone, S. L. (2001). The influence of hope on appraisals, coping, and dysphoria: A test of Hope Theory. Journal of Social and Clinical Psychology, 20, 117–129. Chapell, M. S., Blanding, Z. B., Silverstein, M. E., Takahashi, M., Newman, B., Gubi, A., et al. (2005). Test anxiety and academic performance in undergraduate and graduate students. Journal of Educational Psychology, 97, 268–274. Cramer, K. M., & Dyrkacz, L. (1998). Differential prediction of maladjustment scores with the Snyder hope subscales. Psychological Reports, 83, 1035–1041. Curry, L., Snyder, C. R., Cook, D. L., Ruby, B. C., & Rehm, M. (1997). Role of hope in academic and sport achievement. Journal of Personality and Social Psychology, 73, 1257–1267. Duff, A., Boyle, E., Dunleavy, K., & Ferguson, J. (2008). Erratum to: ‘‘The relationship between personality, approach to learning and academic performance’’ [Personality and Individual Differences 36 (2004) 1907–1920]. Personality and Individual Differences, 44, 532. Ehrhart, M. G., Ehrhart, K. H., Roesch, S. C., Chung-Herrera, B. G., Nadler, K., & Bradshaw, K. (2009). Testing the latent factor structure and construct validity of the Ten-Item Personality Inventory. Personality and Individual Differences, 47, 900–905. Feldman, D. B., Rand, K. L., & Kahle-Wrobleski, K. (2009). Hope and goal attainment: Testing a basic prediction of hope theory. Journal of Social and Clinical Psychology, 28, 479–497. Gilman, R., Dooley, J., & Florell, D. (2006). Relative levels of hope and their relationship with academic and psychological indicators among adolescents. Journal of Social and Clinical Psychology, 25, 166–178. Gosling, S. D., Rentfrow, P. J., & Swann, W. B. Jr., (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37, 504–528. Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41, 924–936. Kappes, H. B., & Oettingen, G. (2011). Positive fantasies about idealized futures sap energy. Journal of Experimental Social Psychology, 47, 719–729. Oettingen, G., & Mayer, D. (2002). The motivating function of thinking about the future: Expectations versus fantasies. Journal of Personality and Social Psychology, 83, 1198–1212. Peterson, S. J., Gerhardt, M. W., & Rode, J. C. (2006). Hope, learning goals, and task performance. Personality and Individual Differences, 40, 1099–1109. Roesch, S. C., & Vaughn, A. A. (2006). Evidence for the factorial validity of the dispositional hope scale: Cross-ethnic and cross-gender measurement equivalence. European Journal of Psychological Assessment, 22, 78–84. Snyder, C. R. (1995). Conceptualizing, measuring and nurturing hope. Journal of Counseling and Development, 73, 355–360. Snyder, C. R. (2002). Hope theory: Rainbows in the mind. Psychological Inquiry, 13, 249–275. Snyder, C. R., Harris, C., Anderson, J. R., Holleran, S. A., Irving, L. M., Sigmon, S. T., et al. (1991). The will and the ways: Development and validation of an individual differences measure of hope. Journal of Personality and Social Psychology, 60, 570–585. Snyder, C. R., Shorey, H. S., Cheavens, J., Pulvers, K. M., Adams, V. H., III, & Wiklund, C. (2002). Hope and academic success in college. Journal of Educational Psychology, 94, 820–826. Tellegen, A., Watson, D., & Clark, L. A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
The differential impact of agency and pathway thinking on goal pursuit and university exam performance Monique F. Crane ⇑ Macquarie University, Department of Psychology, Building C3A, Sydney, NSW 2109, Australia
a r t i c l e
i n f o
Article history: Received 3 June 2013 Received in revised form 18 September 2013 Accepted 24 September 2013 Available online 17 October 2013 Keywords: Hope theory Agency thinking Pathway thinking Exam performance
a b s t r a c t The present study examines the interaction between agency and pathway thinking on performance outcomes. The study used a repeated-measures design to examine the role of agency and pathway thinking on goal pursuit emotions (e.g., determination), secondary appraisal, and final exam performance in a group of university psychology students. Consistent with previous mental health research (Arnau, Rosen, Finch, Rhudy, & Fortunato, 2007; Cramer & Dyrkacz, 1998), the present findings suggest a dominant role for agency thinking in performance. Moreover, there was a reliable interaction between pathway and agency thinking in the prediction of goal pursuit and performance. The interactions consistently revealed that when agency thinking was high, pathway thinking was generally irrelevant to our various measures of goal pursuit. These findings challenge the additive role of agency and pathway thinking suggested by hope theory (Snyder, 2002). Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Dispositional hope has been identified as an important predictor of academic success and goals pursuit (Snyder et al., 2002). Hope is defined as, ‘‘the perceived capability to derive pathways to desired goals, and motivate oneself via agency thinking to use those pathways’’ (Snyder, 2002, p. 249). As the definition implies, hope integrates agency and pathway thinking. Agency is the motivation to pursue goals and the belief in one’s capacity to achieve desired goals (Snyder, 2002; Snyder et al., 2002). Pathway thinking is the development of routes to goal achievement. According to hope theory, the most hopeful individuals are those high on both pathway and agency thinking, both measured in the Dispositional Hope Scale (DHS; Snyder et al., 1991). Those high in hope are anticipated to be the most directed toward goal pursuit and goal success (Snyder et al., 1991). However, research examining the independent roles of agency and pathway thinking demonstrates findings somewhat inconsistent with the predictions outlined by hope theory (Snyder, 2002). In particular, hope theory anticipates that the best outcomes in terms of mental health and goal attainment would be for individuals with both high agency and pathway thinking; that make an additive contribution to overall dispositional hope. In contrast, to the proposed additive role of agency and pathway thinking previous research finds that pathway thinking plays a minimal role in predicting mental health outcomes (e.g., Arnau et al., 2007; Cramer & Dyrkacz, 1998). Empirical work suggests a principle role for agency over pathway thinking in ⇑ Tel.: +61 2 9850 8604; fax: +61 2 9850 8062. E-mail address: [email protected] 0191-8869/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.paid.2013.09.026
terms of mental health outcomes. The present study, sought to add further clarity to this debate by examining the independent and interactive role of pathway and agency thinking in relation to goal pursuit and goal achievement (i.e., university exam performance). 1.1. The role of agency and pathway thinking on performance and goal attainment To date, most studies of performance have been restricted to an analysis of overall hope. Previous work demonstrates that dispositional hope is positively related to students’ goal setting and attainment, as well as appraisals of future attainment (Snyder et al., 1991) educational and sporting achievement (Curry, Snyder, Cook, Ruby, & Rehm, 1997; Gilman, Dooley, & Florell, 2006; Snyder et al., 2002), and task performance (Peterson, Gerhardt, & Rode, 2006). A single study examines the independent roles of agency and pathway thinking in goal attainment. Feldman, Rand, and Kahle-Wrobleski (2009) found that goal-specific agency thinking, not pathway thinking, predicted goal attainment. Thus, akin to studies examining mental health, agency cognitions were more critical to goal attainment than the perception of goal pathways. 1.2. A possible interaction between pathway and agency thinking The above analysis of previous work has identified the possibly unique roles of agency and pathway thinking in predicting mental health and performance outcomes. Consistent with this idea, hope theorists have identified the possible unique independent roles of agency and pathway thinking (Snyder, 2002). Given the independence of agency and pathway thinking it is plausible for an individual
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
to be high in agency, but low in pathway thinking and vice versa (Snyder, 2002). An individual with high agency, but low pathway thinking may be motivated toward goal achievement, but fail to identify a clear strategy and thus motivation remains uncultivated (Snyder, 2002). Conversely, an individual with low agency and high pathway thinking may perceive a clear route to goal achievement, but lack sufficient self-belief and personal drive to motivate goal pursuit (Snyder, 2002). The various combinations of agency and pathway levels may have varying impacts on goal pursuit and performance. Hope theory does not specify how goal pursuit and attainment are impacted by mismatched agency and pathway thinking. However, it is implicit in hope theory that being high on either agency or pathway thinking is more beneficial than being low on both. Initially, hope theory implies no interaction between agency and pathway thinking. Snyder (2002) suggests that agency and pathway thinking have an equally antagonistic role when there is a mismatch in their use. However, the antagonism resulting from mismatched agency and pathway thinking has never been formally investigated. Moreover, given previous research demonstrating the principal independent effect of agency, a case can be made that when agency thinking is high, pathway thinking may be less crucial to performance outcomes. 1.3. The present study To date, no studies have examined the possible differential impact of dispositional agency and pathway thinking on performance. Curry et al. (1997) point to the importance of both goal related cognitions and emotions to goal success. Thus, in order to examine performance holistically this study examined objective performance outcomes (i.e., exam results), but also related secondary appraisals of goal attainment (i.e., perceived control over exam performance). Chang and DeSimone (2001) found that the hope construct was related to secondary appraisals, but not primary appraisal of exams. Moreover, the present study will examine positive approach emotions related to goal achievement (e.g., determined). Hope theory suggests that people high in hope, approach goals with a positive emotional state (Snyder, 1995). Research examining performance goals demonstrates that positive affectivity functions to increase goal directed behaviour promoting achievement of goals (Bagozzi & Pieters, 1998). The present study has two core aims. The first is to investigate the independent roles of agency and pathway thinking in predicting goal pursuit. The second is to examine the interaction between agency and pathway thinking in relation to the measures of goal pursuit. In particular, this paper seeks to explore whether agency and pathway thinking are additive in facilitating performance and to explore whether incongruent agency and pathway cognitions result in decreased performance. The following tentative hypotheses are made: H1. Agency thinking will moderate the relationship between pathway thinking and exam performance. When agency is high, pathway thinking will be unrelated to exam performance. In contrast, when agency is low there will be a positive relationship between pathway thinking and exam performance.
H2. Agency thinking is anticipated to moderate the relationship between pathway thinking and positive approach emotions (e.g., determined). When agency is high, pathway thinking will be unrelated to the experience of positive approach emotions. In contrast, when agency is low there will be a positive relationship between positive approach emotions and pathway thinking.
H3. Agency thinking is expected to moderate the relationship between pathway thinking and control over exam performance. When agency is high, pathway thinking will be unrelated to the experience of perceived control over exam performance. In contrast, when agency is low there will be a positive relationship between perceived control over exam performance and pathway thinking.
21
2. Method 2.1. Participants and design A convenience sample of Introduction to Psychology students was used. Of the 654 students enrolled, 481students (28.9% male; 71.1% female; Mage = 20.42; SDage = 5.33) attempted the Time 1 survey. Of these students, 306 gave permission to access their final exam grades, but 29 of these did not complete the exam. Thus, 277 participants were included in the analysis of exam performance (28% male; 72% female; Mage = 20.42; SDage = 5.33). The Time 2 survey was completed the week of the final exam. Only 98 students (20.4%) attempted this survey (22.2% male; 77.8% female; Mage = 20.95; SDage = 6.01). A chi-square confirmed that the Time 1 and Time 2 gender ratio was not statistically different (X2(1) = 2.401, p = .156).
2.2. Materials and procedure Time 1 data was collected as part of 26 tutorial classes. Time 2 data was collected via email invitation within the week of the final exam. All self-report measures except for gender, age, conscientiousness and neuroticism were measured at both time points. These measures were included as covariates, in the model of performance, because of their relationship to exam performance or dispositional hope (e.g., Chapell et al., 2005; Duff, Boyle, Dunleavy, & Ferguson, 2008). Dispositional hope was measured using the DHS (Snyder et al., 1991) a 12-item measure of hope consisting of four agency and pathway items, and four distracter items. Respondents were asked to indicate on a scale from 1(definitely false) to 4(definitely true) the degree to which these statements describe them across time and situations. The internal reliability of the pathway and agency sub-constructs was satisfactory (a = .69 and a = .70, respectively). Studies examining the factor structure of the DHS have yielded mixed results regarding whether a one or two-factor structure is most appropriate (Snyder et al., 1991). Having noted this, analyses on student populations appear to consistently suggest that a twofactor model comprised of the agency and pathway sub-constructs is a better fit to the data than a one-factor model (Babyak, Snyder, & Yoshinobu, 1993; Snyder et al., 1991; Roesch & Vaughn, 2006). A confirmatory factor analysis was carried out to confirm the twofactor structure of the scale. Conscientiousness and neuroticism were measured using the two item conscientiousness and neuroticism sub-scales from the Ten Item Personality Index (TIPI; Gosling, Rentfrow, & Swann, 2003). This measure consists of adjectives (e.g., disorganised) and the participant is required to indicate the degree to which these attributes describe themselves on a scale from 1(strongly disagree) to 7(strongly agree). The neuroticism scale demonstrated satisfactory internal reliability (a = .70); however, the internal reliability for the conscientiousness scale was below the satisfactory level (a = .62). The low alpha’s are expected given that only two items are used per dimension (Ehrhart et al., 2009). Positive approach emotions were measured using four positive approach emotions from the Positive and Negative Affect Scale (PANAS; Tellegen, Watson, & Clark, 1988): determined, excited, active, and attentive. Participants rated the extent that they had felt these emotions in the last week in relation to the psychology final exam. Ratings were made on a five-point response scale 1(very slightly or not at all) to 5(extremely). These items demonstrated satisfactory internal reliability (Time 1 a = .73 and Time 2 a = .76). Perceived control of exam performance (secondary appraisal) was designed for the purpose of this study. Four items measured perceived control of psychology exam performance (e.g., ‘‘I control
22
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
how well I will do in the psychology exam’’). Participants indicated the degree each statement applied to them on a four-point scale from 1(does not apply to me) to 4(applies to me completely). The scale demonstrated satisfactory internal reliability (Time 1 a = .80 and Time 2 a = .88). A principal components analysis was also conducted and confirmed that a single factor emerged at both time points. 2.3. Analysis strategy Hierarchical linear regressions were used to examine the three dependent variables: exam performance, Time 2 positive approach emotions and Time 2 perceived control. All predictor variables were mean centred and gender was dummy coded (0 = female, 1 = male). MODPROBE was conducted using the Johnson–Neyman technique (Hayes & Matthes, 2009) in order to investigate twoway interaction effects. MODPROBE allows the examination of the independent variable’s impact on the dependent variable at different values of the moderator (Hayes & Matthes, 2009). 3. Results 3.1. Preliminary analysis The agency and pathway sub-constructs were moderately positively correlated (r = .45, p < .005). Also interesting was that Time 1 agency thinking was related significantly to Time 2 positive approach emotions (r = .38, p < .005) and Time 2 perceived control (r = .46, p < .005), but had no relationship with final exam performance. Time 1 pathway thinking demonstrated no significant relationship with the predictors. The full set of correlations is provided in Table 1. Responders versus non-responders only varied on perceived conscientiousness (p < .002); a greater level of conscientiousness was found in responders (M = 5.38; SD = 1.43) compared to nonresponders (M = 4.80; SD = 1.27). 3.2. Confirmation of two-factor structure The relative fit of the one and two-dimensional DHS measurement model were compared. The two-dimensional measurement model was the best fit to the data. The two-dimension model yielded fit statistics that generally exceed benchmarks (comparative fit index [CFI] = .97; Root Mean-Square Error of Approximation [RMSEA] = .05; Tucker-Lewis Index [TLI] = .95). In contrast, fit indices for the one-dimensional model indicated less fit adequacy (comparative fit index [CFI] = .94; Root Mean-Square Error of Approximation [RMSEA] = .07; Tucker-Lewis Index [TLI] = .19). The Akaike Informa-
tion Criterion (AIC) for the models indicated that the two-dimensional model was a better fit to the data (one-dimensional AIC = 115.32; two-dimensional AIC = 90.3). 3.3. Analysis of exam performance A hierarchical linear regression examined the impact of the agency and pathway sub-constructs on exam performance. Age, sex, neuroticism and conscientiousness were initially included as covariates. Only age, neuroticism and conscientiousness were significant predictors and consequentially included in Step 1. Step 2 included Time 1 pathway thinking. Step 3 included Time 1 agency thinking and the final step included the two-way interaction. Table 2 presents the results for this analysis. The final model including the two-way interaction significantly improved on the more parsimonious models (R2change ¼ :014, Fchange (1, 269) = 4.442, p < .04). The final model revealed a significant main-effect for age (t = 2.91, p < .005; R2part ¼ :03; b = .17) indicating that on average older students had better exam performance compared to younger students. Both neuroticism and conscientiousness positively predicted exam performance (t = 2.85, p < .006; R2part ¼ :03; b = .17;t = 2.11, p < .04; R2part ¼ :01; b = .14, respectively). A significant main-effect for the agency sub-construct indicated that greater agency prior to the exam was predictive of better final exam performance (t = 2.18, p < .04; R2part ¼ :02; b = .15). As predicted (H1), there was a significant two-way interaction between Time 1 pathway and agency (t = 2.11, p < .04; R2part ¼ :01; b = .12). The interaction is illustrated in Fig. 1. The MODPROBE revealed that when agency thinking was or below .22 (mean centred) a significant negative relationship between pathway and final exam performance emerged. However, at higher values of agency there was no significant relationship between pathway and exam performance. 3.4. Analysis of positive approach emotions Twelve cases with more than 5% missing data on critical variables and five multivariate outliers were removed leaving 81 cases in the analysis. Time 2 positive approach emotions were examined using a hierarchical linear regression model; only Time 1 positive approach emotions was included as a covariate in Step 1. Table 3 provides the hierarchical regression for this analysis. The final model including the two-way interaction demonstrated a marginally significant improvement on more parsimonious models (R2change ¼ :03, Fchange (1, 76) = 3.75 p = .057). There was a strong significant positive main-effect for Time 1 approach emotions (t = 4.84, p < .001; R2part ¼ :21; b = .48). Consistent with predic-
Table 1 Bivariate correlations, means and standard deviations for measures.
1. Exam performance 2. Age 3. Conscientiousness 4. Neuroticism 5. Time 1 pathway 6. Time 1 agency 7. Time 1 positive approach emotions 8. Time 2 positive approach emotions 9. Time 1 perceived control 10. Time 2 perceived control * **
p < .05. p < .005.
2
3
4
5
6
7
8
9
10
M
SD
.15 1.0
.25 .26* 1.0
.14 .10 .39** 1.0
.13 .07 .13 .26* 1.0
.20 .17 .59** .35* .45** 1.0
.07 .09 .41** .24 .08 .39** 1.0
.09 .05 .34* .03 .08 .38** .58** 1.0
.25 .08 .10 .08 .20 .39** .05 .06 1.0
.22 .02 .22 .06 .17 .46** .09 .14 .75** 1.0
27.90 20.42 4.90 3.23 11.99 12.07 9.72 9.68 3.28 3.42
6.07 5.33 1.42 1.52 1.49 1.85 2.68 2.95 .60 .61
23
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25 Table 2 Hierarchical regression analysis of exam performance. Step 1
(Constant) Age Neuro. Consc. Pathw. (P) Agenc. (A) PA DR2 DF * **
Step 2
B
SE
27.86 .19 .67 .99
.35 .07 .23 .25
Step 3
b
t
B
SE
.17 .17 .23
80.12 2.83** 2.85** 3.88**
27.87 .18 .63 1.00 .24
.35 .07 .24 .25 .23
.104 10.54**
Step 4
b
t
B
SE
.16 .16 .23 .06
80.11 2.79** 2.66** 3.93** 1.04
27.85 .18 .71 .67 .40 .56
.35 .07 .24 .29 .24 .25
.004 1.08
b
t
B
SE
b
t
.16 .18 .16 .10 .16
80.62 2.80** 3.00** 2.29* 1.67 2.26*
27.68 .19 .68 .61 .39 .54 .21 .014 4.44*
.35 .07 .24 .29 .24 .25 .10
.17 .17 .14 .10 .15 .12
78.54 2.91** 2.85** 2.11* 1.63 2.18* 2.11*
.017 5.11*
p < .05. p < .01.
Fig. 1. Interaction between agency and pathway in the prediction of final exam performance (out of 50).
Fig. 2. Interaction between agency and pathway in the prediction of positive approach emotions.
3.5. Analysis of perceived control over exam performance tion (H2), there was a marginally significant interaction between pathway and agency (t = 1.94, p = .057; R2part ¼ :03; b = .20). The nature of this interaction is presented in Fig. 2. The relationship between pathway thinking and positive approach emotions only approached significance at very high and very low levels of agency thinking, but the relationship was never significant at p < .05. Thus, the interaction appeared to be driven by the low perceived positive approach emotions at Time 2 when both pathway and agency thinking were low.
One case with more than 5% missing data and three multivariate outliers were removed leaving 94 participants in the analysis. Time 2 perceived control over exam performance was analysed using the same procedure as above. Time 1 perceived control and was the only covariate included in the model. The hierarchical regression analysis is present in Table 4. The final model including the two-way interaction demonstrated a significant improvement on the previous model
Table 3 Hierarchical regression analysis of positive approach emotions. Step 1 B (Constant) App. Em. Pathw. (P) Agenc. (A) PA DR2 DF **
s
p < .01. p < .06.
9.25 .52
.27 29.73**
Step 2
Step 3
Step 4
SE
b
t
B
SE
b
t
B
SE
b
t
B
SE
b
t
.27 .10
.52
33.71 5.45**
9.26 .53 .09
.28 .10 .17
.52 .05
33.49 5.44** .49
9.10 .47 .01 .30
.29 .10 .18 .18
.47 .01 .17
31.23 4.60** .02 1.65
9.26 .48 .16 .27 .21 .03 3.75s
.30 .10 .20 .18 .11
.48 .09 .15 .20
31.05 4.84** .84 1.47 1.94s
.002 .24
.03 2.73
24
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
Table 4 Hierarchical regression analysis of positive approach emotions. Step 1 B (Constant) Perc. Contr. Pathw. (P) Agenc. (A) PA DR2 DF * **
Step 2 SE
3.37 .76
.51 81.22**
.05 .09
Step 3
b
t
B
SE
.71
69.02 9.01**
3.37 .75 .02
.05 .09 .03
Step 4
b
t
B
SE
.70 .04
68.40 8.67** .47
3.34 .72 >.01 .06
.05 .09 .03 .03
.001 .24
b
t
B
SE
b
t
.67 >.01 .17
66.20 8.24** .04 2.06*
3.38 .73 .04 .06 .05 .04 8.01*
.05 .08 .03 .03 .02
.68 .10 .15 .23
67.37 8.71** 1.20 1.89 2.83*
.026 4.24*
p < .05. p < .01.
(R2change ¼ :04, Fchange (1, 89) = 8.01, p < .005). There was a strong significant positive main-effect for Time 1 perceived control (t = 8.71, p < .001; R2part ¼ :42; b = .68). The predicted (H3) two-way interaction between pathway and agency thinking also emerged (t = 2.83, p < .04; R2part ¼ :04; b = .23). The interaction is illustrated in Fig. 3. Consistent with H3, the relationship between pathway thinking and perceived control over exam performance was positive at lower values of agency at or below .85 (mean centred). Interestingly, at very high values of agency (at or above 2.44 mean centred) a negative relationship between pathway thinking and perceived control emerged. 4. Discussion 4.1. Summary of findings Collectively, the results suggest that agency thinking was the most reliable predictor of goal pursuit and actual performance. H1 was partially supported, as anticipated higher agency thinking resulted in a non-significant relationship between pathway and final exam performance. Thus, high motivation toward goal completion (e.g., I energetically pursue my goals) and anticipated goal completion (e.g., I meet the goals I set for myself) meant that pathway thinking became unrelated to exam performance. Inconsistent with prediction, when agency thinking was low there was a negative relationship between pathway thinking and exam performance. This finding is consistent with the suggestion that a mismatch between agency and pathway thinking may undermine performance
(Snyder, 2002). Alternatively, asking participants to consider their pathways at Time 1 may have elicited engagement in positive fantasies. Indulging in positive fantasies has been shown to elicit poorer performance compared to engaging in positive expectations (Kappes & Oettingen, 2011; Oettingen & Mayer, 2002). The examination of positive approach emotions generally confirmed prediction (H2). While the interaction was similar to that anticipated, it was driven by the lack of Time 2 positive approach emotions when both agency and pathway thinking were low. While this particular finding was consistent with the predictions of hope theory (Snyder, 2002); high agency and pathway thinking did not yield the greatest positive approach emotions. In contrast, pathway thinking appeared to play no role in the experience of positive emotions when agency thinking was high. It was only when agency thinking was low that pathway thinking tended to increase the experience of positive approach emotions. The findings for Time 2 perceived control again generally confirmed prediction (H3). When agency thinking was low, pathway thinking positively predicted perceived control over exam performance. In fact, when agency was high (>1.5 SD), pathway thinking negatively predicted perceived control over exam performance. The finding for high agency is curious and not observed for the other goal-pursuit outcome measures; therefore it may not be robust and perhaps an experimental artefact. Extending the work of Chang and DeSimone (2001), the current findings suggest that agency thinking has a dominant impact on secondary appraisal; pathway thinking appears predictive of secondary appraisal only when agency thinking is low. In contrast, the findings are generally inconsistent with the predictions of hope theory (Snyder, 2002) that suggest high agency and pathway thinking would provide the greatest level of hope and therefore perceived control over goal outcomes.
4.2. Theoretical contributions
Fig. 3. Interaction between agency and pathway in the prediction of perceived control over exam performance.
This paper sought to address two core aims. First, the paper explored the potentially dominant role of agency cognitions in facilitating goal pursuit. Second, to examine whether the two following proposals made by hope theory have merit: (1) agency and pathway thinking play an additive role in motivating goal pursuit and (2) whether incongruent agency and pathway cognitions result in decreased goal pursuit. While the nature of the interactions varied to some degree depending on the outcome variable, this does not detract from the central theoretical point of the analysis. Across all three measures of goal pursuit when agency thinking was high, pathway thinking tended to decrease in its significance as a predictor of goal pursuit and performance outcomes. There was no evidence, in this analysis, of an additive relationship between agency and pathway thinking as suggested by hope theory (Snyder, 2002). Moreover, the results consistently suggest that if an individual engenders
M.F. Crane / Personality and Individual Differences 58 (2014) 20–25
high agency thinking and low pathway thinking he/she will not necessarily suffer problematic outcomes in terms of goal pursuit or performance. In contrast, the findings for high pathway thinking and low agency thinking were not as consistent across outcome measures. For perceived control and positive approach emotions it appeared that high pathway thinking compensated for low agency thinking. Conversely, when it came to actual performance outcomes the combination of low agency and high pathway thinking resulted in the lowest performance. These findings suggest that the antagonism between the low agency and high pathway thinking may be specific to actual performance outcomes. However, high pathway thinking may be beneficial when agency is low in terms of the emotional and cognitive aspects of goal pursuit. 4.3. Limitations and future directions The present results are limited by the response attrition observed for the Time 2 survey. The high level of attrition is likely to be, in part, a consequence of the proximity of the Time 2 survey to the final exam. High attrition is likely to increase selection bias reducing the generalisability of the findings to other populations. Having noted this, the results still generally confirm prediction indicating merit in the findings. A further limitation is the use of first year psychology students, which again reduces the level of generalizability. Thus, the findings need to be interpreted with caution and future research should seek to replicate the findings in additional samples. The current study was conducted over a short time span. The 6 week period sought to cover a relatively stressful time for psychology students where the final exam was increasing in salience. As such, this time period is certainly worthy of study. However, it would be also useful to study longer time periods. A further limitation of this study is the level of situational ambiguity. Participants were in their second semester at university and there was perhaps a lower level of ambiguity than compared to students in semester one. Pathway thinking may be more important to goal pursuit and performance when there is a higher level of situational ambiguity where pathways to goals become necessary. Future research should seek to examine the interaction between the level of situational ambiguity and pathway thinking on performance outcomes. 4.4. Conclusion The findings suggest little support for the additive function of agency and pathway thinking in terms of increasing goal pursuit and performance. In contrast, agency thinking appears to provide the greatest benefit to goal pursuit. Moreover, inconsistent with the predictions of hope theory (Snyder, 2002), there was very little evidence to support the detrimental impact of incongruent thinking styles. Generally, these findings suggested that a closer investigation of the complex interaction between agency and pathway thinking is required by further research.
25
References Arnau, R. C., Rosen, D. H., Finch, J. F., Rhudy, J. L., & Fortunato, V. J. (2007). Longitudinal effects of hope on depression and anxiety: A latent variable analysis. Journal of Personality, 75, 43–64. Babyak, M. A., Snyder, C. F., & Yoshinobu, L. (1993). Psychometric properties of the Hope Scale: A confirmatory factor analysis. Journal of Research in Personality, 27, 154–169. Bagozzi, R. P., & Pieters, R. (1998). Goal-directed emotions. Cognition & Emotion, 12, 1–26. Chang, E. C., & DeSimone, S. L. (2001). The influence of hope on appraisals, coping, and dysphoria: A test of Hope Theory. Journal of Social and Clinical Psychology, 20, 117–129. Chapell, M. S., Blanding, Z. B., Silverstein, M. E., Takahashi, M., Newman, B., Gubi, A., et al. (2005). Test anxiety and academic performance in undergraduate and graduate students. Journal of Educational Psychology, 97, 268–274. Cramer, K. M., & Dyrkacz, L. (1998). Differential prediction of maladjustment scores with the Snyder hope subscales. Psychological Reports, 83, 1035–1041. Curry, L., Snyder, C. R., Cook, D. L., Ruby, B. C., & Rehm, M. (1997). Role of hope in academic and sport achievement. Journal of Personality and Social Psychology, 73, 1257–1267. Duff, A., Boyle, E., Dunleavy, K., & Ferguson, J. (2008). Erratum to: ‘‘The relationship between personality, approach to learning and academic performance’’ [Personality and Individual Differences 36 (2004) 1907–1920]. Personality and Individual Differences, 44, 532. Ehrhart, M. G., Ehrhart, K. H., Roesch, S. C., Chung-Herrera, B. G., Nadler, K., & Bradshaw, K. (2009). Testing the latent factor structure and construct validity of the Ten-Item Personality Inventory. Personality and Individual Differences, 47, 900–905. Feldman, D. B., Rand, K. L., & Kahle-Wrobleski, K. (2009). Hope and goal attainment: Testing a basic prediction of hope theory. Journal of Social and Clinical Psychology, 28, 479–497. Gilman, R., Dooley, J., & Florell, D. (2006). Relative levels of hope and their relationship with academic and psychological indicators among adolescents. Journal of Social and Clinical Psychology, 25, 166–178. Gosling, S. D., Rentfrow, P. J., & Swann, W. B. Jr., (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37, 504–528. Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41, 924–936. Kappes, H. B., & Oettingen, G. (2011). Positive fantasies about idealized futures sap energy. Journal of Experimental Social Psychology, 47, 719–729. Oettingen, G., & Mayer, D. (2002). The motivating function of thinking about the future: Expectations versus fantasies. Journal of Personality and Social Psychology, 83, 1198–1212. Peterson, S. J., Gerhardt, M. W., & Rode, J. C. (2006). Hope, learning goals, and task performance. Personality and Individual Differences, 40, 1099–1109. Roesch, S. C., & Vaughn, A. A. (2006). Evidence for the factorial validity of the dispositional hope scale: Cross-ethnic and cross-gender measurement equivalence. European Journal of Psychological Assessment, 22, 78–84. Snyder, C. R. (1995). Conceptualizing, measuring and nurturing hope. Journal of Counseling and Development, 73, 355–360. Snyder, C. R. (2002). Hope theory: Rainbows in the mind. Psychological Inquiry, 13, 249–275. Snyder, C. R., Harris, C., Anderson, J. R., Holleran, S. A., Irving, L. M., Sigmon, S. T., et al. (1991). The will and the ways: Development and validation of an individual differences measure of hope. Journal of Personality and Social Psychology, 60, 570–585. Snyder, C. R., Shorey, H. S., Cheavens, J., Pulvers, K. M., Adams, V. H., III, & Wiklund, C. (2002). Hope and academic success in college. Journal of Educational Psychology, 94, 820–826. Tellegen, A., Watson, D., & Clark, L. A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070.