silently and sometimes loudly the lessons. Next, they were asked to do the related exercises to each lesson.
۳. Then two classes; control and experimental groups; were given the post-test to measure the effects of using three reading strategies on experimental group by calculating and comparing the mean scores of two groups. As the pre-test, the post-test was same in both experimental and control group.
۴. In order to test three hypotheses in the study the mean scores of two tests in control group compared with that of experimental group, the mean scores of pre-test in experimental group compared with their mean scores in post-test and also the mean scores of three sub-groups in post-test compared with each other.
This study had a comparative design. Measuring the effects of a treatment, it had the Pre-test/Post-test structure. It consisted of two groups: The experimental and the control group with the pre-test and post- test for each group. The participants were given the pre-test to ensure the comparability of them prior to treatment and to ensure the considered strategies. The post-test was to measure the effects of the treatment; to what extent the treatment truly resulted in learning. First, the researcher gave both groups the pre-tests. The results of pre-test helped the researcher to find out about the amount of learners’ prior knowledge and prior ability in reading. This was useful also for choosing appropriate reading texts for next step of study. Then, the control group was taught the reading comprehension lessons in the traditional way (as they had it before in their reading classes), while the experimental group was taught the same reading comprehension lessons by using three semantic mapping strategies before and after reading the text.
Next, both groups were given the post-test to investigate the effects of each instruction on each group. By examining the mean scores of both groups in the post-test, the effect of semantic mapping strategies was determined. By analyzing the mean scores of three sub-groups A, B, and C in the experimental group on post-test the best kind of semantic mapping strategies according to the constructor among the three introduced kinds determined. So the design had such a formula:
Control group: T T
Experimental group: T X T
T refers to tests and X refers to treatment.
CHAPTER FOUR: DATA ANALYSIS AND INTERPRETATION
The present chapter considered the statistical effect of using semantic mapping strategies on reading comprehension of EFL learners in Kerman. Generally, the results of this investigation were interpreted and analyzed here.
To obtain data related to the research questions, the researcher administrated the reading achievement pre-test in the second semester of the academic year 2010/2011 on Wednesday, 11 of April at Nabovat high school; just at the commencement of the treatment. Both the experimental and control groups sat for the same format of the pre-test. The needed data were collected from the 120 papers; 30 control group pre-test, 30 experimental group pre-test, 30 control group post-test and 30 experimental group post-test.
All the data collected from pre-tests and post-tests were analyzed in this chapter. First, the researcher considered the validity and reliability of the tests and semantic maps which used in the study. Then, there is a consideration of research variables with detailed tables and graphs. Next, the research hypotheses are considered by use of the Independent Sample T-test (according to Mackey and Gass (2005) it is a pragmatic statistic that is used to determine if the means of two groups are significantly different from one another) and analysis of variance (ANOVA; which enables researcher to compare the performance between more than two groups) within the Box-plot graphs. For each hypothesis, the statistical goal was to test the hypothesis and reject the null relationship or zero hypothesis (H0) by showing that there was a relationship. Null hypothesis predicts that there is no relationship between two variables; semantic mapping and reading comprehension. The SPSS analytic program was used which included the needed statistical tests.
According to Mackey and Gass (2005), determining the differences between the means of two groups, needs a t-test. Because the structure of this research is pre-test/post-test (the groups’ performances before treatment compared with their performance after treatment) a paired t-test was needed, it means that each group was paired with itself on the two tests.
ANOVA results provide an F value, which is the ratio of the amount of variation between the groups to the amount of variation within the groups (Mackey & Gass 2005). Comparing three sub-groups A, B and C in the effectiveness of three semantic mapping instructions, the researcher used ANOVA to see if group differences were due to chance or were sufficient to reject the null hypothesis.
۴.۲. Validity and Reliability of the Study
In order to have a valid study the researcher had to provide valid tests as pre-test and post-test and also valid reading comprehension lessons and semantic maps to teach to participants during the treatment period.
۴.۲.۱. Validity of the pre-and post-test. To determine the validity of the research instrument, both of the pre- and post-tests are given to a group of judges. The juries were two teachers with ten years experience each in teaching the basic and the secondary stages with B.A and higher Diploma in English teaching methods .In addition, the test was given to a supervisor with six years experience with M. A. degree. Also they were given to a teacher with M.A. degree in English teaching methods. They all suggested some changes in the post-test to be able to investigate the differences. The researcher modified the post-test accordingly.
۴.۲.۲. Validity of semantic maps. To investigate the validity of the semantic maps which were created by the researcher based on the lessons of the treatment, the researcher followed some models of semantic maps which were done by well-known researchers whose studies are trusted and published in well-known data bases, such as EBSCO and ERIC data bases. Moreover, the semantic maps were given to two supervisors with six years experience. They enrich the maps with some related ideas which in turn had a significant role on the students’ reading comprehension.
۴.۳. The Reliability of the Tests
In order to ensure the reliability of the pre and post-test of the study, the researcher administered a test-retest on an experimental sample of fifteen subjects who were taken from the population. In other words, the same tests were given to the same group of subjects at two points in time. Their tests were corrected. To arrive at a score by which reliability can be established, the researcher determined the Correlation Coefficient between the two test administrations. A statistical formula of Pearson Product-moment Correlation was calculated. It is a common means for determining the strength of relation between variables. The result was (.909) which was a very high internal contingency coefficient.
This was very appropriate for the reliability of the test and for the purpose of the study as shown in the following table (Correlation is significant at the 0.00 level):
Table 4.1. The correlation of test-retest
Pre , Post
۴.۴. The Conditions of the Research Variables
Here the conditions of variables in the study; pr
e-test and post-test scores in experimental and control group are considered.
۴.۴.۱. Pre-test and post-test variable scores in experimental group. The underlying table and related graphs demonstrate the scores of pre-test and post-test variables. Table (4.2) displays the measures of central tendency and numeric statistics of pre-test and post-test which was the outcome of the participants’ scores in experimental group in those tests. Graph (4.1) shows the learners’ performances in pre-test and graph (4.2) shows their performance in post-test. These graphs which are a figurative show of table (4.2), made the comparison of the results easier.
In table (4.2), N is the total number of participants in experimental group which was 30. The Mean or arithmetic average, which is the most common measure of tendency, for experimental group pre-test was 1.60 and for their post-test it was 4.03. It means that the average of their scores in post-test is higher than pre-test. The mean for each test was the sum of all scores in that test divided by the number of participants (N). The Median is the score at the center of the distribution; the score that splits the group in half. For pre-test in experimental group the Median was 2 and for post-test it was 4. The next measure in table (4.2) is the Mode which is the most frequent score obtained by group; the experimental group. The Mode of pre-test was 2 and that of post-test was 4.
The next measure in the following table is SD (standard deviation) which is the way of measuring variability. Simply put, the standard deviation is a number that shows how scores are speared around the mean; specifically. It is the square root of the average squared distance of the scores from the mean. In other words, it is the square root of the Variance which is the result of taking the differences between each score and the mean and squaring them, then adding up these squared values and dividing by the total number of participants (N). As it is obvious in table (4.2), the Standard Deviation of the post-test is smaller than pre-test. This means that in pre-test there was more variability. In other words, there was more homogeneity in post-test. The amount of minimum and maximum numbers in each test also shows, by itself, the higher amount of results in post-test. Graph (4.1) illustrates the statistics of experimental group in pre-test and graph (4.2) shows their statistics in post-test.
Table 4.2. Pre-test and post-test statistics in experimental group
Graph 4.1. Pre-test descriptive statistics in experimental group
Graph 4.2. Post-test descriptive statistics in experimental group
۴.۴.۲. Pre-test and post-test variable scores in control group. Table (4.3) and the related graphs are the results of control group performance in pre-test and post-test. Again, in order to compare the results easily, the statistics