This research is a quasi experimental research and design used is ARandomized Pretes-Postest Control Group Design. The subject of the researchdetermined by random toward experimental and control groups. The results of this studyshowed that there was differences class between implements the inquiry learning andconventional learning. Mean score N-Gain science of process skills students washighest in primary schools reached 0.29 A, while for the N-Gain lowest in primaryschools C of 0.15 As for the indicators of science-process skills, process skills of thehighest indicators in the experimental class there is the aspect of the hypothesis, whilethe lowest indicator contained in the aspect of the question.

Pembelajaran berbasis keterampilan proses sains menekankan pada kemampuan siswa dalam menemukan sendiri (discover) pengetahuan yang didasarkan atas pengalaman belajar, hukum-hukum, prinsip-prinsip dan generalisasi, sehingga lebih memberikan kesempatan bagi berkembangnya keterampilan berpikir tingkat tinggi (Houston 1988:208).

In the control class there areindicators that contained by aspects of the hypothesis while the lowest was on aspects ofobservation. The results visible experimental class science process skills results showedhigher levels of N-gain control of the class. The results in the experimental classsignificantly more improve students’ science process skills than conventional class. Keywords: Science Process Skills, Learning Science. Biblioteka eskw dlya kompas v15.