This research is conducted on the basis of researcher’s observation where the students are getting difficulties to understand the meaning of texts. The objective of the research is to find out the effectiveness of GIST strategy to teach students’ reading comprehension on narrative text. This research involved 60 students. Thirty students were in the experimental class and thirty students were in the control class. The researcher used cluster random sampling to determine a sample. An instrument test to collect the data used test. They are pre-test and post-test. The type of test is multiple choices test. The assessment of the test result was focused on students’ reading comprehension on narrative texts. Furthermore, the data was analyzed by using statistical analysis technique and the hypothesis used t-test.
Based on the computation of result, the average of pre-test score of experimental class was 69.67 and the control class was 64.83. Meanwhile, the average of post- test score in the experimental class was 81.17 and the one in the control class was 72.17. It was obtained that t-test was 3.606 and the t-table was 2.000 with the standard of significance of 5%. The t-test was higher than t-table (3.606 > 2.000). Because tcount ≥ ttable , it means that Ha was accepted and Ho was rejected. Based on the result, the researcher concluded that the use of GIST (Generating Interaction between Schemata and Text) strategy was effective to teach students’ reading comprehension on narrative text at SMA Negeri 13 Semarang in the academic year of 2016/2017.