This research was done  due to several problems on the students’ ability and impact to the score in reading comprehension.The first problem because the students do their task alone, the second problem the students had low motivation to read the text, the last problem the students felt bored with learning model used. The researher expected wether the power of two strategy is able to give significant effect on students’ reading comprehension. This research was an experimental research. The researcher chose quasi experimental design by using the experimental class and control class. The population of this research was 1th grade of SMA N 1 Pangkalan. The sample was X MIPA 1 which was as treated experimental class and  X MIPA 2 was treated as control class. To determine the sample, the researcher used  cluster sampling because the class have similar characteristic in score of reading test and have similar problems in reading comprehension.The researcher used the reading test as instrument.To analyzed the data, the researcher used Liliefors test for normality test. Then to test the hypothesis, the researcher used t-test and   Based on the analysis of the data, the result was also indicated by the mean of experimental class pre-test score which was 60.36. It was higher than the control class pre-test score which was 49.29. The result was also indicated by the mean of experimental class post-test score which was 85.86. It was higher than the control class post-test score which was 61.86. It meant there was significant effect of using the power of two strategy on students reading comprehension. It was concluded that Ha was accepted and H0 was rejected. The students who were taught by using the power of two strategy better than the students who were not taught by using the power of two strategy. To sum up, the power of two strategy has a significant effect on students reading comprehension on descriptive text at 1th grade of SMA N 1 Pangkalan.

There are several problems found in this research, such the material used by teacher was not related to the major, textbook for  students was not provide the material related to their major, and the material only taken from the internet by teacher randomly. This research is research and development which uses the theory of Sugiyono. The stages in this development are research, developing, validation, and testing. The subjects in this study were 36 students from the tenth grade of Computer and Network Engineering. The need analysis shows, students want to learn reading material that is related to their major and useful when they work after graduation. For validation by expert, the English reading material is categorized as "Very Valid" with the percentage score obtained is 83%. The practical results that have been given to teacher and students, the English reading material is categorized as "Very Practice" with  score 92%. And the last, the effectiveness of the English reading material is effective, it gets a significant value of 0.00 which is <0.05,it means that the English reading material is effective. So overall, the developing of English reading material is  accordance with the needs of students majoring in computer and network engineering.