Theoretical Framework REVIEW OF RELATED LITERATURE

24 for data collection in survey research” Wiersma, 1969: 179. Backstrom and Cesar 1981: 187 state in their book that the questionnaire is “the physical form of all theories, hypotheses, and hunches that have gone into planning the survey.” Therefore, the writer used test and questionnaire as the instruments for collecting data.

B. Research Participants

The participants of this study were the fifth semester students of the English Language Education Study Program of Sanata Dharma University in the academic year of 20102011. They were chosen since they had learnt conditional sentence in Structure III course. As they had learnt conditional sentences, it was assumed that they would not make a lot of errors in the conditional sentences test. Because the fifth semester students of the English Language Education Study Program of Sanata Dharma University were in a great number, the writer selected a sample to represent the population as a whole. In choosing the sample, the writer chose the cluster sampling procedure. Wiersma 1995: 292 gives the definition of cluster sampling in his book as “a procedure of selection which involves the random selection of clusters from the larger population.” The writer chose this sampling because this sampling is in groups, not individuals, were randomly selected. Therefore, the writer did not need to make new cluster. Besides, the existing cluster consisted of the fifth semester students who had learnt conditional sentences in the same semester, and it was supposed to have the 25 same level of understanding conditional sentences. Although there would be difference, it would be no big difference among them. The population of the fifth semester students was divided into four structure classes. Because of the reasons explained above, the writer thought that cluster sample is the appropriate one. Therefore, in this research, the writer chose two classes as the sample of the research.

C. Research Instruments

The instruments used to gather the data in this study were a test and a questionnaire. Hague 1995: 2 mentioned the four objectives of questionnaire. “First, it is used to obtain the accurate information from the respondent. Second, it gives the design in interview. Third, it gives the standard format of fact, comment, and attitude recording. The last is to ease the data processing.” The questionnaire used as the instrument in this research consisted two parts. The first part was thirty items of close-ended responses. In this part, the participants were required to choose the number which showed their opinion on the statements given. The second part was one item of open-ended response. In this part, the participants were required to answer the question based on their own opinion with explanations on their opinion and suggestions. Those questions were related to the possible factors causing errors. The questionnaire was made by investigating five elements, such as students’ perceptions, abilities, motivations, teaching learning activities and also time and social influences. Ellis 1997: 4 identified two factors influencing the 26 language learners in learning second language such as internal and external factors. Therefore, the writer divided those five elements into two, which are internal and external factors. Ellis also gives the examples of each factor. Social conditions and input are external factors. On the other hand, Ellis 1997: 4-5 classified the cognitive mechanism and knowledge into the internal factors. Based on that theory, the writer classified those five elements such as: internal factors, which included the students’ perceptions, abilities, and motivations, and external factors, which included teaching learning activities and time and social influences. On the other hand, the writer also used the test to gather the data. A test is “a method to measure the students’ ability or knowledge and it also can provide the information about the achievement of group of learners” Hughes, 1989: 4. The test was arranged to measure the students’ understanding on conditional sentence. Therefore, the test was essential part of the research. Based on those facts, the writer chose a test and questionnaire as the instruments to gather the data. The test consisted of thirty items and it was divided into two parts. The first part consisted of fifteen items. In this part, the students were required to fill in the blanks with the correct answers based on the rule in conditional sentences. The objective of this part was that the students were able to recognize which type of conditional sentences used in the questions and demonstrate the correct pattern of conditional sentences. This would “show the participants’ ability to recognize the grammatical functions of words in sentences structures” Freeman and Long, 1991: 167. The second part consisted of fifteen items. In this part, the students 27 were required to make the conditional sentences based on the given facts. The objective of this part was that the students were able to recognize the correlation between the fact and the conditional sentences and demonstrate the correct pattern of conditional sentences based on the given facts. This kind of test “provided a better indication of students’ real achievements in learning” Ebel, 1979: 96. Because of its essential role in the research, the test should meet the requirements of measurement. Those are validity and reliability.

1. Validity

Validity is one of the important things in constructing the test. A test is said to be valid if “it measures accurately what it is intended to measure” Hughes, 1989: 22. There are four types of validity in test. Those are:

a. Content Validity

A test is said to have content validity if its content constitutes as representative sample of the language skills, structures, and so on with which it is meant to be concerned. The test would have content validity only if it included a proper sample of relevant structures. Therefore, in order to have high content validity, the test should contain the items which are intended to be measured on the test. The importances of content validity are: “first, the greater a test’s content validity, the more likely it is to be an accurate measure of what it is supposed to measure. Second, such a test is likely to have a harmful backwash effect” Hughes, 1989: 22-23. The details of the content of the test are presented in the Appendix 2.