Theme
3BB The lecture and the flipped classroom
INSTITUTION
Chang Gung Memorial Hospital - Taiwan
Chang Gung University-Taiwan
Several factors had been proposed to affect the outcome of flipped classroom learning. However, scientific publications which investigated the influencing factors for the outcome of flipped classroom learning are still few. This work retrospectively analyzed the available data in a flipped classroom learning program to determine the significant predictors of final outcome.
A flipped classroom re-education program called TOTR (teaching on the run) was implemented to improve the teaching skills of clinical teachers in our hospital since July 2014. Till Dec 2014, three courses of TOTR with a total of 45 clinical teachers attended and completed this program. The available information for analysis included age, gender, duration of clinical teaching, professional category, TOTR session, pre-class test scores and post-class test scores. The pre-class score was used to represent effort and achievement in pre-class learning, and post-class scores were used to represent the final outcome of TOTR learning.
Our results showed that achievement in pre-class learning is an important determinant of the final academic outcome in flipped classroom learning. It means that if a learner cannot have a good achievement in pre-class learning, the final outcome might not be optimal. Thus,efforts should be made to improve the achievement in pre-class learning. Then, a better final outcome can be expected.
Another interesting finding is that more senior participants were associated with a lower final test score. This might suggest that senior learners prone to resist to this mode of new teaching style.
In flipped classroom learning, facilitators should monitor the learning process and give the learner assistance whenever necessary during the pre-class learning stage. Thus, a good final learning outcome can be guaranteed.
Active communication and orientation may be necessary for senior learners before the implementation of flipped classroom style learning.
When divided into 3 groups according to the post-class scores, participants with highest post-class scores had also highest pre-class scores (p<0.001, linear trend p<0.001, Table 1) and shortest teaching duration (p=0.015, linear trend p=0.004).
Groups |
low (n=15) |
middle (n=15) | high (n=15) | p value | linear trend p | total (n=45) |
Final test score | 53.8±10.18 | 71.07±5.2a | 83.87± 5.54a,b | 0.000 | 0.000 | 69.58±14.37 |
Male | 5 | 5 | 4 | 0.902 | 0.697 | 14 |
Age | 36.33±8.34 | 36.13±6.11 | 33.4±6.52 | 0.452 | 0.261 | 35.29±7.03 |
Teaching Duration | 11.93±8.82 | 7.53±4.73 | 5.13±3.78a | 0.015 | 0.004 | 8.2±6.67 |
Job Category | 0.303 | 0.138 | ||||
Physician | 0 | 4 | 3 | 7 | ||
Nurse | 9 | 7 | 6 | 22 | ||
Others | 6 | 4 | 6 | 16 | ||
TOTR session | 0.529 | 0.273 | ||||
1st | 2 | 4 | 6 | 12 | ||
2nd | 6 | 4 | 3 | 13 | ||
3rd | 7 | 7 | 6 | 20 | ||
Pre-class score | 48.2±13.44 | 62.67±10.47a | 73.73±13.75a,b | 0.000 | 0.000 | 61.53±16.26 |
Difference | 5.6±11.29 | 8.4±12.74 | 10.13± 14.1 | 0.621 | 0.336 |
8.04±12.61 |
ANCOVA analysis showed only pre-class score, but not age, gender, teach duration, job category or TOTR session, was a significant predictor of the post-class test scores (Table 2).
Table 2. ANCOVA analysis to determine the predicting variables of post-class scores
Source | Type III Sum of Squares | df | Mean Square | F | Sig. | Partial Eta Squared | Noncent. Paramete | Observed Powerb |
Corrected Model | 5347.478a | 8 | 668.435 | 6.432 | .000 | .588 | 51.452 | .999 |
Intercept | 990.959 | 1 | 990.959 | 9.535 | .004 | .209 | 9.535 | .852 |
Gener | 94.339 | 1 | 94.339 | .908 | .347 | .025 | .908 | .153 |
Job Category | 326.208 | 2 | 163.104 | 1.569 | .222 | .080 | 3.139 | .311 |
TOTR session | 486.756 | 2 | 243.378 | 2.342 | .111 | .115 | 4.683 | .443 |
Age | 10.291 | 1 | 10.291 | .099 | .755 | .003 | .099 | .061 |
Teaching Duration | 119.990 | 1 | 119.990 | 1.155 | .290 | .031 | 1.155 | .182 |
Pre-class score | 1920.702 | 1 | 1920.702 | 18.481 | .000 | .339 | 18.481 | .987 |
Error | 3741.500 | 36 | 103.931 | |||||
Total | 226937.000 | 45 | ||||||
Corrected Total | 9088.978 | 44 |
Linear regression analysis showed that post-class score positively correlated with pre-class score (R = 0.691, p = 0.000, Figure 1). If participants were divided into high, middle and low groups according to pre-class score, the low pre-class score group had most significant improvement in scores after classroom activity, less significant in middle group and no change in high group (Figure 2). The post-class score of the low pre-class score group was still significantly lower than those of the high pre-class score group, despite the improvement after classroom activity is most significant in the low pre-class score group. (Figure 2)
Figure 1. Relationship between post-class and pre-class test score Figure 2. Change in scores after classroom activity