Using Design of Experiment to Investigate the Rating Accuracy of Learning Effect

• Main Authors: YUE- JUN CHEN, 陳玥君
• Other Authors: Ming-Hsien Li
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/94821114273944407116

碩士 === 逢甲大學 === 工業工程與系統管理學研究所 === 100 === The purpose of this research is to investigate the rating accuracy and learning effect by using experimental design. There are two rating videos, I and II. For each video, there are three different operation contents. In videos I, the three different operations are (1) insertion of electronic components (2) make-up of stove heart, and (3) assembly of outer cover. In video II, the three different operations are (4) workbench of units set operation (5) template combination of operation and (6) electronic termination combination of operation. For each operation, there are five different accurate rating values. Afterwards transcribes again these two videos re-recording the videos of continuous playback according to the same operating (III and IV). Therefore, there are four videos (I ~ IV) in this experiment. To enable trainees to have the correct rating skill, the accurately rating value which is nearest to 100% of the operation is played in the beginning. After watching the video the exact rating value is announced so that the trainees could have proper sensing of accurate rating. In this way, the video play the operation 2 or 3 times, then request the trainees to write down the rating value right after their watching the rating videos, the accurately rating value will be announced immediately after the trainees filled out the pre-sentence rating value. The two control factors are gender and accurately rating value. The experimental results show that (1) accurately rating value factor in the six kinds of operating are significant different. (2) When the difference between the rating value and the normal speed (100%) are smaller, the trainees’ pre-sentence error is smaller. (3) The videos which follow the same order of play (III and IV) have smaller estimating error then the videos which do not follow the same order of play videos (I and II).