本研究參考Campione和Brown(1985)的「漸進提示評量」模式，以自編之「浮力概 念」補充教材為主，透過「前測-學習-遷移-後測」的實施程序，實際檢視動態評量應用在 學科領域的區辨、助益及預測等三方面效益。研究者以高雄市勝利國民小學六年級學生為對 象，依「國民小學五年級自然科成就測驗」成績選出高成就、中成就及成就各30名（共90名 ），並根據瑞文氏智力測驗成績粗略配對，爾後隨機分派至實驗組及控制組（每組高、中、 低成就者各15名)，兩組受試者除了接受前測及後測外，實驗組並接受「學習階段」及「遷 移階段」的動態評量，而控制組則接受學習題的教學並自行練習遷移題。研究結果發現： (1)動態評量能有效區辨不同程度的受試者在「學習能力」以及在「遷移能力」上的差異。 (2)動態評量比靜態評量更具有區辨力。(3)實驗組娙歷各階段的動態評量後，其後測成績高 於前測成績。(4)實驗組經歷各階段的動態評量後，其後測成績高於控制組的後測成績。(5) 實驗組經歷各階段的動態評量後，受試者間後測成績的差異減少。(6)「動態評量提示量」 能增加「國民小學五年級自然科成就測驗」成績對「判斷物體密度及浮沈情況解題能力測驗 」對近期學業成就(數學及自然兩科)的預測力。(7)「動態評量提示量」能增加「國民小學 五年級自然科成就測驗」對近期學業成就（數學及自然兩科）的預測力。綜合言之，本研究 結果支持動態評量應用在學科領域的區辨、助益及預測等三項效益。

　　　　　This study referred to the model of Graduated Prompting Assessment introduced by Campione and Brown (1985). The learning materials with the topic of buoyant concepts were developed by the researchers. The purpose of this study was to apply the procedure of pretest-learning-transfer-posttest to examine the dynamic assessment effects of differentiation, acquisition and prediction on learning specific content area materials. The subjects were selected from the 6th graders of Kaohsiung Municipal Sheng Li Primary School. Based on the test scores of the 5th Grade Science Achievement Test(SAT), subjects of three levels of achievement (high, medium, and low)were selected from the pool of the 6th graders. Thirty subjects for each level were randomly assigned to experimental and control group respectively with matching their scores of the Raven's Progressive Matrices. Both groups took pretest and posttest of Buoyant Problem Solving Ability Test (BPSAT). The experimental group got dynamic assessment tests in addition. In the experimental condition, the subjects received 15 minute dynamic assessment in learning stage and 12-50 minute dynamic assessment in transfer stage. In the control condition, the subjects received 15 minute instruction to answering the questions in learning stage and 50-60 minute exercise to finishing the questions without any help in transfer stage. The major findings were as following: 1.The dynamic assessment could effectively differentiate the differences both on learning ability and transfer ability among various achieving levels of subjects. 2.The dynamic assessment had better effects of differentiation for individual differences than the static assessment had. 3.The posttest of experimental group in BPSAT were significantly higher than their parallel pretest after receiving the dynamic assessment treatment. 4.The posttest of experimental group in BPSAT were significantly higher than those of the control group after receiving the dynamic assessment treatment. 5.The variances of posttest of experimental group in BPSAT were minimized compared to those of pretest. 6.Dynamic assessment measures could contribute the unique variances to pretest and posttest in BPSAT, as the SAT is the control variable. 7.Dynamic assessment measures could contribute the unique variances to the performance of academic achievement in Mathematics and Science, as the SAT is the control variable. In conclusion, these results including differentiation, acquisition, and prediction support the advantages of dynamic assessment on learning specific content area materials.