Skip to main content
Log in

How beliefs explain the effect of achievement goals on judgments of learning

  • Published:
Metacognition and Learning Aims and scope Submit manuscript

Abstract

Research has demonstrated that compared to mastery-approach goals (i.e., trying to develop a mental ability), performance-approach goals (i.e., trying to outperform others) elicit higher metacognitive judgments, such as judgments of learning (JOLs). However, the underlying mechanism of the effects of achievement goals on JOLs remains unclear. Therefore, the present study conducted three experiments to examine the underlying mechanism of the effects of these goals on JOLs, based on the preexisting beliefs about achievement goals. The results of Experiment 1 showed that pre-study JOL in the performance-approach goal condition was higher than that in the mastery-approach goal condition, and Experiment 2 demonstrated that the difference of JOLs during learning (i.e., item-by-item JOLs) between the mastery-approach and performance-approach goal conditions remained constant from the beginning to end, suggesting that the beliefs about achievement goals affect item-by-item JOLs. In Experiment 3, participants read a description of a hypothetical experiment and estimated the performance of the group with mastery-approach and performance-approach goals. The results revealed that more participants believed that performance-approach goals have greater benefits than mastery-approach goals. Thus, the beliefs regarding achievement goals are one of the crucial factors affecting JOLs.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. Regarding the retrospective confidence judgments for previous tests, different results were obtained. Muis et al. (2016) showed that performance-approach goals were related negatively to bias between actual performance and confidence judgments (i.e., overestimate or underestimate), whereas mastery-approach goals were not. Additionally, Nietfeld et al. (2014) showed that both mastery-approach and performance-approach goals were not associated with bias.

  2. About 35-48% of participants were excluded in Experiment 1b, 2a and b, whereas about 15% of participants were excluded in Experiment 1a. This exclusion rates were due to include the worker participating in the previous experiments. After removing the worker participating in previous experiments, exclusion rates were approximately 12-15% in Experiment 1b, 2a and b.

  3. For exploratory purpose, Experiment 1 also conducted additional analysis for item-by-item JOLs in terms of serial position, as Experiment 2. The results showed that achievement goals and serial position had significant effects on item-by-item JOLs, whereas the interaction between goals and serial position was not significant (more details, see Appendix).

  4. The preset study calculated within-participant correlations between study times and JOLs to examine the effect of achievement goals on the utilization of the cues concerning study effort. The average of correlations were similar values in the mastery-approach goal condition (M = -.04, SD = 0.23, 95% CI [-.11, .02] in Experiment 2a and M = -.002, SD = 0.24, 95% CI [-.078, .002] in Experiment 2b) and in the performance-approach goal condition (M = -.04, SD = 0.27, 95% CI [-.12, .05] in Experiment 2a and M = -.07, SD = 0.26, 95% CI [-.150, .002] in Experiment 2b), and all of correlations were near 0, even considering 95% CIs. Therefore, participants might make item-by-item JOLs not based on study effort.

  5. The exluded participants did not sufficiently read the description. In fact, the present study also measured the reading time for the description about the hypothetical experiment, and the reading time for excluded participants (M = 30.30 sec, SD = 34.45, 95% CI [20.07, 40.53] in Experiment 3a and M = 28.05 sec, SD = 31.47, 95% CI [18.60, 37.51] in Experiment 3b) were shorter than those who were not excluded (M = 53.60 sec, SD = 32.73, 95% CI [44.67, 62.54] in Experiment 3a and M = 56.01 sec, SD = 83.31, 95% CI [30.05, 81.97] in Experiment 3b), ts > 2.09, ps < .05, ds > 0.44. Furthermore, given that the meta-analytic research concerning reading rate showed that the average reading rate for adults was 238 words per minute (Brysbaert, 2019) and the description included 167 words, reading times for the excluded participants were very fast compared to the average reading rate.

References

Download references

Acknowledgements

I thank Alan Castel for providing helpful comments on revision. This work was supported by JSPS KAKENHI, Grant Number 19K14384 (to Kenji Ikeda).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Kenji Ikeda.

Ethics declarations

Ethics approval

APA ethical standards were followed in the conduct of this study, and the present study was approved by Tokai Gakuin University ethics committee (approval ID: 2020–21).

Consent to participate

Before participation, all participants received informed consent and consented to participate the experiment.

Conflicts of interest

The author does not have any conflicts of interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Serial position effect on item-by-item JOLs in Experiment 1

For exploratory purpose, in Experiment 1, the present study conducted additional analysis to examine the effect of achievement goals and serial position on item-by-item JOLs, as Experiment 2 (see Table 8). The coding of the achievement goal condition was mastery-approach goals = -0.5 and performance-approach goals = 0.5, and serial positions were coded as 1–3 = 11, 4–6 = 9, 7–9 = 7, 10–12 = 5, 13–15 = 3, 16–18 = 1, 19–21 = -1, 21–24 = -3, 25–27 = -5, 28–30 = -7, 31–33 = -9, 34–36 = -11. Regarding random effects, the differences between participants as random intercept and slope for serial position were included, and the difference between the experiments as random intercept and slopes for achievement goals, serial position, and interaction were also included. The results showed that the effect of achievement goals was statistically significant (estimate = 6.51, 95% CI [2.33, 10.70], p = .002), demonstrating item-by-item JOLs in the performance-approach goal condition were higher than those in the mastery-approach goal condition. Additionally, the effect of serial position was also statistically significant (estimate = 0.30, 95% CI [0.21, 0.39], p < .001), and thus, JOLs decreased with later serial positions. In contrast, the interaction showed no statistical significance (estimate = 0.12, 95% CI [-0.07, 0.31], p = .22). Although it should be noted that this is additional analysis and did not include relatedness term because of the complexity of statistical model, item-by-item JOLs in the performance-approach goal condition were higher than those in the mastery-approach goals, regardless of serial position, as Experiment 2.

Table 8 Means, standard Deviations, and 95% confidence intervals of item-by-item JOLs and calibrations for each serial position bin in each condition

Serial position effect on metacognitive accuracy in Experiment 2

The present study also conducted additional analysis to examine the serial position effect on metacognitive accuracy (see Table 8). Calibration was computed as the subtraction of recall performance from item-by-item JOLs for each serial bins, and thus positive value indicates overestimate. The coding of the achievement goal condition and serial position was same as previous analyses. In Experiment 1, serial position effect, achievement goal effect, and interaction did not reach statistical significance (estimate = 0.72, 95% CI [-4.84, 6.27], p = .25, estimate = -0.08, 95% CI [-0.24, 0.08], p = .33, and estimate = 0.26, 95% CI [-0.06, 0.57], p = .12, respectively). Unlike Experiment 1, the results of Experiment 2 showed that serial position effect was statistically significant (estimate = 0.47, 95% CI [0.14, 0.81], p = .01), whereas the effect of achievement goals and interaction did not reach statistical significance (estimate = 2.72, 95% CI [-5.60, 10.70], p = .52 and estimate = -0.05, 95% CI [-0.90, 0.80], p = .92, respectively), suggesting that participants underestimate their performance as learning progress, regardless of achievement goals.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ikeda, K. How beliefs explain the effect of achievement goals on judgments of learning. Metacognition Learning 17, 499–530 (2022). https://doi.org/10.1007/s11409-022-09294-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11409-022-09294-y

Keywords

Navigation