illustrated additional dialogs of the speech acts under study (i.e.) were taken from: Functions (Matreyek, 1990); Passages series (Richards & Sandy, 1998); Interchange series (Richards et al., 2005); Top Notch series (Saslow & Ascher, 2005); Friends sitcom (Crane & Kauffman, 1994 to 2004); and instructional movies.
In the first step a sample PET was piloted with a number of students similar in characteristics to those of the target sample and its reliability was estimated. In the next step the piloted language proficiency test (PET) consisting of four parts including reading (35 items), writing (7 items), listening (25 items), and speaking sections was administered for the purpose of determining the homogeneity of participants. The four parts of the exam had the same value- 25% each and the total score was made by adding all the results together (the total score of the test equals to 50). Sixty eight female EFL learners at intermediate level received the test and according to the results of the test 55 learners whose scores fell within one standard deviation below and above the mean were selected to be divided into two groups for the purpose of the study.
In the next step the speech act test was piloted with 20 students and its reliability was estimated. Afterwards, the teacher made a multiple-choice discourse completion test (MCDCT) out of those speech acts which comprised of 20 items and was administered as the pretest to make sure if the students were homogeneous and lacked the knowledge of the parts of speech to be taught and therefore eligible to take part in this study and investigate if there was any difference in their ability in understanding the speech acts within the domain of pragmatic knowledge prior to the treatment. The pretest checked the participants’ knowledge of speech acts in both groups with respect to apology and greeting terms.
All the participants in both cooperative and competitive groups were exposed to the same material and same amount of time, the whole semester which lasted 10 weeks, two 90-minute sessions a week.
In cooperative group the students were put in groups of three or four and the teacher (the researcher, herself) tried to set up a friendly atmosphere in the class. Each group member knew that each correct or incorrect answer was of great importance for the rest of the members of the same group. Competitiveness was de-emphasized in each group while group work and cooperation was encouraged.
Every session the researcher provided the model dialogs. The participants listened to a dialogs two times. The first time they were asked to cooperatively identify the kind of speech acts used. The second time They were given 5 minutes to talk to each other and do their best to guess the age, social status, and the relationship between the speakers in the dialog(s).
Role play was a very useful technique which followed the model dialog. After the students analyzed the dialog, the teacher had them act out the dialog in their groups. The researcher gave learners enough information on the age, sex and social status of participants so that they would not blindly exchange a number of utterances. The participants helped each other in the process of role play and no one wanted to prove to be the best; rather, they were going to help each other to have a satisfactory outcome.
Sometimes a penalty was chosen for those groups who failed to accomplish the expected outcome. This was a performance of pantomime by a student who was chosen by the group.
The group members were not the same for different activities and this enabled the learners to experience working with different participants in their class.
At the same time, in the competitive group, the researcher divided the participants into the groups aiming at facilitating the teaching and learning process but the learners knew that even if they were in groups, they were assessed based on their individual efforts and outcomes. To do so, the researcher helped them from various groups differing from one session to the other. Contrary to the previous treatment for the cooperative group, this treatment included no group work.
The researchers provided the model dialog but this section enjoyed almost no group work and the participants listened to a dialog two times. The first time they were individually asked to identify the kind of speech acts used. The second time they only paid attention to their own listening and were asked about it later by their names being called.
Role play system was used in group with the teacher choice of partners. The outcome of that role play was assessed with praising one of the partners as the best one in regards with the amount of attaining the objectives.
Following the 10 weeks of treatment via employing cooperative and competitive methods, the posttest of MCDCT was administered. The results of both pre and post tests were analyzed and compared through SPSS version 20.
Since the participants were selected through convenient sampling and randomization was not possible for the researcher the design of the present study was quasi experimental, pretest posttest. Achievement of speech acts was the dependent variable and both the cooperative learning and competitive learning were the two independent variables. Participants’ gender (female) and language proficiency were regarded as control variables.
3.7. Statistical Analyses
In the present study, data analysis was implemented in the form of both descriptive and inferential statistics as well. Descriptive statistics like mean, and standard deviation were obtained to calculate the result of the homogenization test and the post test. The reliability of PET and MCDCT were obtained by KR-21formula. For inferential statistics an independent samples t-test was run to compare the cooperative and compatetive groups’ mean scores on the posttest of speech acts in order to probe the effect of the cooperative and competitive learning on EFL learners’ achievement of speech acts.
Results and Discussions
The present chapter deals with purpose and null hypotheses. In this chapter details of the data analysis and results of the study based on both descriptive and statistical analyses are reported. Then discussion of the results will follow.
4.1 Pilot study of Preliminary English Test (PET)
A sample PET was piloted in the first phase of the study. Table 4.1 presents the results. The results indicated that the mean was 38.26 and the SD was 7.39.The reliability of the test then was calculated as 0.89 based on KR-21 method which is an acceptable reliability. Table 4.1 shows the descriptive calculations related to the PET pilot study.
Table 4.1 Descriptive statistics of PET pilot study
Reliability based on KR-21
4.2. Subject-Selection Statistics
The piloted PET test was administered to 68 students. Based on the mean (40.72) and SD (4.80), 55 students whose scores fell within one SD above and below the mean were selected. Table 4.2 below represents the descriptive statistics of the subject selection procedure.
Table 4.2 Descriptive Statistics of subject selection
4.3 Pilot study of MCDCT
The researcher- made MCDCT was piloted to a group of 20 students similar to the students of the main study. The reliability of the test was then calculated as .74 based on KR-21 method. The following Table 4.3 reports the statistics related to the piloting process of pre/posttest.
Table 4.3 descriptive statistics of pilot st
udy of MCDCT pre/post test
Reliability based on KR-21
4.4. Proficiency Test (PET)
Upon the main administration of the piloted PET, 55 participants who scored one standard deviation above and below the mean were selected as the target sample of the study and were randomly assigned to two experimental groups. Table 4.5 presents the descriptive statistics of the PET scores for both cooperative and competitive groups. As displayed in Table 4.4 the mean scores for cooperative and competitive groups on PET were 39.39 and 40.67 respectively.
Table 4.4 Descriptive statistics of PET by groups
Std. Error Mean
An independent samples t-test was run to compare the cooperative and competitive groups’ mean scores on PET in order to prove that the two groups enjoyed the same level of general language proficiency prior to the main study.
The results of the independent t-test as shown in Table 4.5 below (t (38) = 1.39, P = .164 .05, r = .22 it represents a weak effect size) indicated that there was not any significant difference between cooperative and competitive groups’ mean scores on the PET. Thus it could be concluded that the two groups enjoyed the same level of general proficiency prior to the main study and were homogeneous.
Table 4.5 Independent samples t-test of PET scores
Levene’s Test for Equality of Variances
t-test for Equality of Means
Std. Error Difference
95% Confidence Interval of the Difference
Equal variances assumed
Equal variances not assumed
It should be noted that the assumption of homogeneity of variances was met (Levene’s F = 9.13, P = 3.122 .05). That is why the second row of Table 4.5, i.e. “Equal variances not assumed” is reported.
Figure 4.1 presents the graph of the two groups’ PET scores.
4.5. Pretest of Speech acts
To ensure the homogeneity of the participants in regards with their knowledge of the selected speech acts the MCDCT was administered to participants prior to the treatment. The students refused to answer the questions None of the participants showed any evidence of prior knowledge of the selected parts of speech situations provided in the test.
4.6 Post test of speech acts
The null hypothesis of the study was formulated as “there is no significant difference between the effect of cooperative and competitive learning on EFL learners’ achievement of speech acts”.
At the end of the treatment, the MCDCT test was administered to the participants to investigate any possible difference between the two groups. Table 4.6 reports the results.
Table 4.6 Descriptive statistics of speech acts posttest by groups
Std. Error Mean
As can be seen from the Table, the mean for the cooperative group turned out to be 16.32 outperforming the competitive group with a mean score of 14.04.
An independent samples t-test was conducted to investigate whether the mean difference was significant or not. However, before running the inferential statistics, some assumptions had to be checked.
4.7 Testing Assumptions
Four assumptions should be met before one decides to run parametric tests; 1) the data should be measured on an interval scale; 2) the subjects should be independent that is to say their performance on the test is not affected by the performance of other students, 3) the data should enjoy normal distribution and 4) the groups should have homogeneous variances (Field; 2009). The present data were measured on an interval scale and none of the subjects’ performed independently on the tests. The assumption of normality is also met. As displayed in Table 4.7 the values of skewness and kurtosis are within the ranges of +/- 2 (Bachman