<![CDATA[The purpose of this study was to explain the effect of the prediction guide learning model on the ability to write a fantasy story in the text of seventh-grade students at SMPN 6 Percut Sei Tuan. The population in this study was all seventh grade students, with a sample size of 64. The assessment instrument used is the structure and linguistic rules of fantasy story text. The method used is the experimental method of a two-group post-test design. The sampling technique used was purposive sampling. The data collection techniques used were observation, testing, and documentation. The data analysis technique used was the data analysis requirements test, normality test, homogeneity test, and hypothesis testing. Based on the results of the data analysis, it was found that the data results were normally distributed. After doing the normality test, Lcount < Ltabel at a significant level, the results of the X variable data normality test are (0.1514 < 0.1568), and the results of the Y variable data normality test are (0.1529 < 0.1568). The results of the homogeneity test are Fhitung < Ftabel, namely 1.72 < 1.82, and the hypothesis test t0> ttabel, namely 3.56 > 2.039. The results of the research in the experimental class obtained by the average student reached 80.87, which is included in the good category. The acquisition of very good category scores was at a rate of 47%, while those in the very poor category only reached 6%. The research results in the control class obtained an average of 68 students included in the sufficient category. The acquisition of a very good category is in the percentage of 3%, while those in the category are very few and reach 31%. This proves that the Prediction Guide learning model has a significant influence on the ability to write fantasy story texts of VII grade students at SMPN 6 Percut Sei Tuan.]]>
