In the present project two research instruments were applied, i.e. a motivation ques-tionnaire and an interview schedule. The initial version of the quesques-tionnaire con-tained sixteen scales that were created on the basis of Kormos and Csizér’s (2008) and Ryan’s (2005) motivation questionnaire. With the help of think-aloud protocols, professional insights and a pilot study the instrument was refi ned, its items were altered, deleted. Besides, new items and completely new scales were also added dur-ing the development phase. The fi nal version of the questionnaire contained eleven original and four newly added scales that added up to fi fteen scales.
Altogether, the scales involved 75 items which formed the main body of the questionnaire. Besides, altogether 16 items asked about the bio data and lan-guage learning history of the participants. The following scales were included in the questionnaire: Ideal L2 Self (six items), Ought-to L2 Self (own) (four items), Ought-to L2 Self (other) (four items), Learning Experience (fi ve items), Moti-vated Learning Behaviour (fi ve items), Written Language Use (three items), Lan-guage Contact (fi ve items), Parental Encouragement (four items), Attitude to-wards Learning Ukrainian/English (four items), Language Class Anxiety (four items), Language Use Anxiety (three items), Direct Contact with English Speakers (four items), Direct Contact with Ukrainian Speakers (six items), Cultural Inter-est (four items), Friends (fi ve items), English as an International Language (fi ve items), Local Relevance of Ukrainian (four items).
The interview schedule was developed on the basis of the fi ndings of the questionnaire study. It was divided into four parts, including a section on bio data (age, work, experience, job description, etc.) another one inquiring about students’
general interest towards languages, and a third one that contained guiding ques-tions referring to Ukrainian language motivation and the fi nal one that focuses on English language motivation. Items in the third and the fourth sections were creat-ed on the basis of the fi ndings of the regression analysis of the questionnaire data.
Therefore the Ukrainian part of the interview schedule asked about the role of parents, attitudes towards language learning as perceived by the respondents. The English section, on the other hand, included items inquiring about the perceived importance of friends, written language use and attitude towards language learn-ing. There were some items in both the Ukrainian and English sections asking the respondents their interpretation of the ideal and the ought-to self, and the language competence of the learners.
82
The original interview schedule was designed in Hungarian. But as some of the interviews were conducted in Ukrainian, the schedule was translated into Ukrainian. Parallel to the interview sessions, the schedule was continuously im-proved. Some of the questions were altered to make it more feasible for the re-spondents. Due to its continuous improvement the research instrument got its fi nal form only after the last interview.
3.2 Participants
Altogether there were 219 school leavers took part in the questionnaire survey.
But only secondary school-leavers studying in Hungarian educational establish-ments were asked to participate. They were between 16 and 18 studying in the 11th form of their school. The selection of respondents was conducted with the help of a multi-stage sampling procedure so as to provide a representative sample of learners of this age cohort. Thus, three groups of learners were created on the basis of the proportion of Hungarians as compared to the ratio of Ukrainians in the area where they study. Learners belonging to the fi rst group were studying in an area where the proportion of Hungarians is 25% and above (N=183). There were less learners in the second and third group because they represented areas where the proportion of Hungarians is between 10 and 25% (N=17) and less than 10%
(N=19) only.
There were seven respondents involved in the interview study. They were the ones who agreed to take part in the study, and allowed publishing the fi ndings alongside with the promised privacy issues. Altogether three (two teachers and the methodologist) of the seven respondents were asked the questions referring to bio data, general interest towards language learning, and items specifi c for English language learning motivation of secondary school learners. Another three were asked the same questions and items specifi c for Ukrainian language learning mo-tivation. There was only one respondent to whom all the questions present in the schedule were asked. He was a former offi cial of the regional department of edu-cation. As peer debriefi ng was present in each phase of the research, the interview study was developed and carried out by two researchers.
4 Data collection
The process of questionnaire data collection followed three steps. First, the headmasters of the selected educational establishments were approached with an offi -cial letter. Then, the date was approved and necessary arrangements with a teacher at the given school were made. As questionnaires were fi lled during one 45 minute lesson, there was a need to agree with the teacher having the class to allow the distribution and fi ll-in process to take place.
BEATRIX HENKEL, PHD
In the interview study, after prospective participants were selected, contact was established with them in person. As the present piece of research was carried out by two researchers, the respondents were approached with that researcher, whom they have already been in good terms with. First of all appointments were arranged, the interview schedule, if requested, was sent prior to the actual inter-view session. The interinter-views and possible follow-up discussions were recorded by Olympus WS-331M digital voice recorder. Before the interview the research-er and the respondent agreed upon the language of the intresearch-erview (Hungarian or Ukrainian). The interviews were precisely transcribed, and checked with the in-terviewees had the analysis meet-ups started. Between two interviews, though, the researchers met, so as to discuss problematic points, practicalities, and conduct necessary changes in the actual schedule.
Both data and investigator triangulation were applied to enhance the trans-ferability of the present study. Research data were triangulated, as naturally two sources of data were available for analysis, i.e. the questionnaire and the interview data. The interview data were analyzed by two researchers working independently from each other.
5 Data analysis
Questionnaire data were computed and analysed with the help of SPSS (Statistical Package for Social Sciences) 13.0. First, the descriptive statistical measures were calculated, namely the mean and standard deviation of each of the scales. Second, the internal reliability coeffi cients, i.e. Cronbach α-s were established. Third, sig-nifi cant correlations were identifi ed between the scales. Fourth, regression analy-ses were carried out to identify the underlying components of motivational dispo-sitions of learners of Ukrainian and English. In the present study to stay focused only the results referring to the four main dimensions will be discussed as they are the variables signifi cantly related to the criterion measure (which in our case was language motivation).
Analysis of the interview data as the technical literature suggests, it had started previously at the very beginning of the research period. Maykut and Morehouse (1994) qualitative data analysis principle guided the analyzing procedure. Therefore, meaning units were highlighted in the interviews. Each meaning unit received identifi cation codes and were then gathered under various headings (categories). Later on they were further analyzed and some of them were merged. The fi nal and the most diffi cult part of the analysing process, was to infer main categories and to fi nd a logical way of arranging the huge pile of data. If questions appeared, the respondents were contacted for clarifi cation or explanation.
84
6 Results and discussion