In order to explore the motivational strategies employed by Iranian English teachers, two questionnaires (see Appendixes 3 and 4), ‘Motivational Strategies Questionnaire’ (Cheng & Dörnyei, 2007), containing the same set of motivational strategies were used. The two questionnaires are different only in their rating scales. It means that one of them, Importance Questionnaire, includes six response options describing degrees of importance (“not important – very important”) and the other one, Frequency Questionnaire, describes the degrees of frequency (“hardly ever–very often”). Each of the questionnaires contains 48 items of which the participants should circle the desired number from 1 to 6. To do this, a 6-level Likert scale was used in both of the questionnaires. Farsi version of the questionnaires was administrated in this study. The selection of the strategies in these questionnaires is based on Dörnyei’s (2001a) systematic overview of motivational strategies. The original list contained more than 100 strategies, but this study uses almost 10 strategies which are included in this questionnaire by Hsing-Fu (2000), Cheng and Dörnyei (2007).
The selection of the 48 micro-strategies in the initial version of this instrument was based on the assumption that these strategies seem to be of considerable positive effect on the learners’ motivation if implemented appropriately in the language classrooms. In addition, they seemed to be transferable across diverse cultural and ethno-linguistic contexts and might fit in the Iranian EFL context. Ease of implementation the language classroom was also taken into account in the selection of the strategies.
The participating teachers were asked to rate, based on their past experience, each questionnaire item/strategy on a six-point Likert scale ranging from ‘Very important’ to ‘Not important’ in terms of how important they consider the use of each strategy for the purpose of enhancing their students’ motivation to learn English as a foreign language. The options that were given to participants to rate strategies on were assigned numerical values ranging from (1, Not Important) to (6, Very Important). As this study was more concerned with identifying how important EFL teachers in Ardabil province perceive some motivational strategies rather than with assessing their attitudes towards the actual use of these strategies, the researcher decided to list all the strategies in this survey in a positive format, as exactly all the previous studies did. For this reason, there were no negatively worded items in the survey and all the items were assigned the same set of values as mentioned before.
The internal consistency test of all the scales in the survey came up with a relatively good result. The Cronbach alpha coefficient for the whole items as one scale was 0.83. After excluding item 19, ‘Invite English-speaking foreigners to class recommended by some of participants in the pilot study, the coefficient fell to 0.82, which is not considered logical to exclude that item. The Cronbach alpha coefficient for the scale after grouping the survey items under the 10 variables/clusters was 0.81.
This study uses cross-sectional or one-shot research design as its overall plan. Two questionnaires were given out to respondents for the statistical representation of the findings in the study.
Prior to commencing the research project, an ethics clearance to conduct a field work research in Iran was obtained from the Shahid Rajaei Teacher Training University in Tehran province. Thus, the pilot study took place. The questionnaires were initially administrated to 20 participants and all the participants returned the survey. Those participants were the holders of a variety of qualifications and taught at different educational levels. Their age ranged from 25 to 40 plus. Respondents in the pilot study provided a lot of recommendations and suggestions. Some participants recommended omitting item 19, which relates to inviting some English-speaking foreigners as guest speakers to the class. They said that the real application of this item was very scarce in real educational situations in Iran. Some participants recommended translating the teachers’ questionnaire into Farsi due to the fact that the majority of those teachers are non-native speakers of English and not totally fluent in English. Despite the fact that our questionnaires were to be administrated only to teachers of English believed to have a solid competence in English, the researcher decided to translate the questionnaires into Farsi (see Appendixes 1 and 2) to eliminate even the slightest risk that foreign language competence would be a barrier for some participants.
3.5.2. Data collection
In this survey all the data was gathered from English teachers in Ardabil province. All parts of the province generally have in-service courses for English teachers. Based on the schedule which was given from each Department of Education in different regions of Ardabil province, the researcher or his friends took part in in-service classes and distributed the questionnaires, Importance Questionnaires or Frequency Questionnaires, randomly among teachers. All the teachers received a short briefing about the study. They were reminded of the fact that their participation in this study was voluntary and that they could withdraw at any time without having to give a reason. They were assured that the information they provided in the questionnaires would be treated very confidentially and would not be disclosed under any circumstances. They were then given Farsi versions of the questionnaires. Participants were given the choice as either to complete the questionnaires immediately after they received it or to take it home and deliver it, post it, or email it to the researcher. For those who preferred to complete the questionnaires immediately after they received, it took them from 10 to 15 minutes to finish responding to the questionnaires.
Since it was a major goal of this study to recruit as many participants as possible at this stage, the researcher decided to use “snowball sampling,” which, according to Dörnyei (2001a) involves a ‘chain reaction’ whereby the researcher identifies a few people who meet the criteria of the particular study and then asks these participants to identify further members of the population.
3.6. Data Analysis
The goal of the preliminary analyses was to screen our data to check the accuracy of the data entered in the data files as well as the reliability and normality of the clusters of variables tested in our surveys. The first step conducted in relation to this was looking for and amending the out-of-range and missing values provided by respondents. Some preliminary statistical analyses such as the internal consistency Cronbach alpha were finally conducted. The first procedure conducted was a prerequisite reliability alpha Coefficient test to measure the internal consistency of the clusters of variables used in the questionnaires. Item analysis was performed next and based on the alpha coefficients and inter-correlation values among the items in each scale, indices were computed. For reliability test, we went in line with what was suggested in the literature (e.g. Dörnyei, 2001a) that indices with alphas of0.60 and greater are considered reliable. We therefore used 0.60 as a cut-off value and regarded indices with such value as reliable and they were used in the main analyses. Conversely, indices with alphas of less than 0.60 were considered unreliable and were excluded from subsequent analyses.
The participants completed only one of the two questionnaires to avoid their inflating in the use of a strategy after indicating a belief in its importance. Such a precautionary measure does not imply any claim that the respondents to the Frequency questionnaire believe that the infrequently-used strategies are important. Respondents to the Importance questionnaire might well indicate a lack of importance for such strategies. This situation is taken into account in the method of data analysis that follows. The two questionnaires were processed separately. In this section, the terms “scale(s)” and “cluster(s)” refer to the term “macro-motivational strategy/strategies”. The data collected from participating EFL teachers on the importance and frequency of various strategies was submitted to a number of primary statistical analyses to make the dataset more manageable. The 48 micro-strategies in our final survey were originally grouped under 10 macro strategies/clusters based on their content similarities. This initial grouping followed Cheng and Dörnyei (2007), Dörnyei (2001a), and Dörnyei and Csizér (1998). The clusters were then rank-ordered based on the participating teachers’ responses in a descending order of the mean. The internal-consistency of the 10 scales was tested by means of a reliability analysis (alpha=0.82). Following Cheng and Dörnyei (2007), first, the mean of each item (i.e., micro-strategy) in the Importance Survey was calculated in order to obtain the importance rank order for the 48micro-strategies. Next, groups of conceptually related strategies (representing macro-strategies) were obtained.
Finally, descriptive statistics were computed to summarize the results and prepare a rank order of the perceived importance of the macro-strategies (see Table3.3).
The aim of the Frequency Survey was to identify those motivational strategies that were underused relative to the importance attached to them by the participating teachers. Following Cheng and Dörnyei (2007), three different measures were obtained (see Frequency section of Table 3.3). First, the mean frequency of use (M) was computed for each strategy. When calculating the