Effectiveness of lab-work learning environments in and out of school: A cluster randomized study
Itzek-Greulich, H., Flunger, B., Vollmer, C., Nagengast, B., Rehm, M., & Trautwein, U. · 2017
grade Brctdeveloper-involvedunreplicated
Sample
67 ninth-grade classes, N = 1,854 students (same trial as Itzek-Greulich 2015, which analysed 68 classes / N = 1,773 on the achievement outcome)
Population
German ninth-grade students (age ~15), secondary schools; chemistry of starch unit.
Design
Companion analysis of the SAME cluster-randomised trial reported in Itzek-Greulich et al. 2015 (Learning and Instruction). Four arms: School only, SCOL (science centre outreach lab) only, SCOL & school, control. This paper adds motivational outcomes (state motivation measured during the unit; trait motivation measured before/after) alongside achievement, and separates the two. Same limitation as the parent paper: every treatment arm contains lab work, so the design identifies WHERE lab work happens, not WHETHER hands-on work adds anything to a content-matched non-practical lesson. Motivational measures are self-report; achievement is a researcher-designed aligned test.
Key findings
The most important single sentence in the K-12 practical-work literature for cost purposes: in a cluster-randomised trial, the out-of-school lab setting was more effective for MOTIVATION while teaching in school was more effective for ACHIEVEMENT. The premium setting bought feelings, the ordinary classroom bought learning. Second, the motivational effect was much stronger on STATE motivation (how students felt during the activity) than on TRAIT motivation (their stable disposition toward science) — i.e. the enjoyment was largely situational and did not convert into durable subject motivation. This is the archive's fadeout pattern showing up on the affective side. Note the design does not contain a content-matched no-practical arm, so it does not license any claim that lab work per se caused the achievement gains over the control.
Genetic confound
Low — class-level randomisation.
Replication notes
Same sample and same randomisation as Itzek-Greulich et al. 2015 and Itzek-Greulich & Vollmer 2017; these are three reports of one trial. Not independent replications of each other.
DOI / URL
10.1016/j.cedpsych.2016.09.005
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Achievement, school-based lab work vs out-of-school lab work | direction | teaching in school more effective for achievement | researcher-designed | posttest | active-alternative | end-of-treatment | domain-skill |
| State motivation (interest, joy, competence; boredom reversed) during the unit, out-of-school lab vs school | direction | out-of-school setting more effective for motivational gains | researcher-designed | during the unit | active-alternative | end-of-treatment | non-cognitive |
| Trait motivation (stable subject motivation) vs state motivation | direction | effects markedly stronger on state than on trait motivation | researcher-designed | pre/post plus in-situ | business-as-usual | end-of-treatment | non-cognitive |
Cited by
- Laboratory and practical work in school sciencemixedconf: mediumgc: low