Most variance in students' momentary classroom emotions lies at the within-person level, yet person-aggregated state measures are fairly reliable (ICC(2) from 0.735 to 0.901)¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r2 · q3
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? ICC(1)s showed the majority of variance in momentary emotional experience was within-person, with ICC(1) ranging from 0.195 to 0.443 (Md = 0.255). → Keller 2025 q3 i? Person-aggregated emotional states were fairly reliable, with ICC(2)s ranging from 0.735 to 0.901 (Md = 0.82). → Keller 2025 (2)

Evidence¶

Keller 2025¶

Keller, M. M., Yanagida, T., Lüdtke, O., & Goetz, T. (2025). How Similar Are Students' Aggregated State Emotions to Their Self-Reported Trait Emotions? Results from a Measurement Burst Design Across Three School Years. Educational Psychology Review. https://doi.org/10.1007/s10648-025-09995-1

q3 · i? · associational · r2

Descriptive multilevel analysis of the experience-sampling data (2668/1543/1258 state datapoints across waves). The article reports ICC(1)s "ranging between 0.195 (anger at T3) and 0.443 (pride at T3)", comparable to prior classroom-emotion studies.

"the majority of variance in students' momentary emotional experience was on the within-person level as indicated by ICC(1)s ranging between 0.195 (anger at T3) and 0.443 (pride at T3) (Md ICC1 = 0.255)."

Keller 2025 (2)¶

Keller, M. M., Yanagida, T., Lüdtke, O., & Goetz, T. (2025). How Similar Are Students' Aggregated State Emotions to Their Self-Reported Trait Emotions? Results from a Measurement Burst Design Across Three School Years. Educational Psychology Review. https://doi.org/10.1007/s10648-025-09995-1

q3 · i? · associational · r2

Same study's reliability analysis of aggregated state emotions via ICC(2) (Spearman-Brown applied to ICC(1) with number of measurements). The article concludes aggregated states were "fairly reliable, i.e., larger than 0.70".

"findings suggest that using person-aggregated emotional states was fairly reliable, i.e., larger than 0.70, with ICC(2)s ranging from 0.735 (anger at T3) to 0.901 (pride at T3) (MdICC2 = 0.82)."

Discussion¶

Related Claims¶

• Most variance in students' epistemic emotions was situational (within-person) fluctuation, with 18–37% due to between-person variability. — possibly the same claim (merge candidate)
• Emotional variability in L2 learning episodes is primarily intraindividual (62–68%), with modest lag-1 inertia (r ≈ .17–.19) and weak long-term trends or weekly cycles — related
• Aggregated state emotions converge substantially but imperfectly with self-reported trait emotions, with averaged correlations ranging from r = 0.391 (shame) to r = 0.608 (boredom) — related
• Long-term stability of aggregated state emotions does not significantly differ from that of self-reported trait emotions over one- and two-year spans — related
• Aggregated state emotions show no superior incremental predictive validity over self-reported traits for academic outcomes; traits corrected for states were only systematically correlated with criteria for enjoyment and somewhat for pride — related
• Most variance in state achievement goals and emotions is situational (within-person) rather than between-person — possibly the same claim (merge candidate)
• SEEQ-S class-average ratings show strong aggregation reliability and high internal consistency, with somewhat lower internal consistency for TEEQ-S — related
• 30%-40% of true variance in each parent-child outcome was trait-like, while 60%-70% fluctuated situationally across time points — related