Читать книгу Ugandan Children's Literature and Its Implications for Cultural and Global Learning in TEFL - Stephanie Schaidt - Страница 41

A semi-structured questionnaire with mainly open items was used at the beginning and at the end of the reading project. According to Cohen, Manion et al. (2011, p. 382), “[t]here is a clear structure, sequence, focus” in semi-structured questionnaires “but the format is open-ended, enabling respondents to reply in their own terms”.

In my study, the number of open questions predominated because the research project does not primarily aim at measurement and generalisation but at the perspectives of individual cases. Only socio-demographic data were collected through closed questions, and strength of response (e.g. to what degree they enjoyed reading) was indicated by the students on Likert scales. The responses to these items were later statistically analysed and compared across participants.

Open questions permit free, individual answers and do not limit the participants in their responses by means of pre-set categories. Thus, they may provide the researcher with comparatively deep answers. As Cohen et al. (2011, p. 392) note, “an open-ended question can catch the authenticity, richness, depth of response, honesty and candour which […] are the hallmarks of qualitative data”. It has to be acknowledged, however, that the open character of questions can also overstrain the respondents because writing down their answers requires time and reflection. Another challenge posed by questionnaires with open questions is that the qualitative nature of the data makes comparison between respondents difficult and, therefore, the process of analysis complex (ibid., p. 382). Thus, open or word-based questionnaires are particularly suitable for smaller samples and for studies which require personal, individual comments (ibid., p. 392), as in the case of my research project.

In order to make sure that all the students understood the questions and were able to write down their responses, the questionnaires in my study were given in German. They were completed in silence in the classroom in my presence. This was helpful because uncertainties could be addressed immediately and it also ensured a good response rate. All 45 students who participated in the study completed the first questionnaire; only one student failed to hand in the final questionnaire because he was absent in the last lesson.

The first questionnaire (see Appendices A and B), which was completed before the reading project started, asked for students’ associations and expectations concerning Uganda and their interest in the country. The students were asked to write down their associations with Uganda, list a few sources for their information about African countries, name aspects concerning Uganda they are particularly interested in and state whether and why they would (not) participate in a student exchange with a Ugandan school. In addition, the questionnaire inquired after the reading preferences of the students and their expectations regarding Ugandan children’s fiction. The students were asked to what degree they enjoyed reading literary texts in German and English in their free time and for the frequency with which they read such texts in the two languages. Students were also questioned for topics/literary genres they are most and least interested in and asked for their expectations concerning Ugandan children’s fiction. Furthermore, socio-demographic data about the students were obtained through the first questionnaire. The students were asked for their sex, year of birth, age, nationality, their own/parents’ country of birth, the language they first learnt and the language spoken at home.

The final questionnaire (see Appendices C and D) was completed after the reading project. In this questionnaire, students were asked to reflect upon the reading project and the books they had read. They were encouraged to suggest improvements for the extensive reading project and to elaborate on the narrative text they liked the best or disliked the most.

Questionnaires were used in this study because they are target-oriented and economical in terms of time. Given time-constraints it would not have been possible to interview all the students who participated in the extensive reading study before and after the study. Questionnaires make it “possible to get a range of responses from many people on a limited number of items” (Nunan & Bailey, 2009, p. 319). In my study, they permitted insights into the prior knowledge, expectations and reflections of all 45 students. Furthermore, responses in questionnaires are often more honest and authentic than in interviews and thus more reliable. As Cohen et al. (2011, p. 395) state, “the lack of face-to-face contact between the researcher and the respondents in a questionnaire might facilitate responses to sensitive material”. Another argument in favour of questionnaires is that when they are used with interviews, the two research instruments complement one another very well.

You may sometimes wish to employ a research design in which you first use questionnaires to get a broad cross section of information or opinions. You could subsequently use interviews to get more detailed data from a subgroup of your sample. (Nunan & Bailey, 2009, p. 319)

After collecting data through questionnaires and reading diaries, I made use of the “sample-resample procedure” (Nunan & Bailey, 2009, p. 319) and interviewed 24 of the 45 participants in the study to get more detailed information concerning the subject matter from them.

There are two major problems researchers face when using questionnaires in a study. First of all, even questionnaires with predominantly open questions can only ask for a limited amount of information. Participants cannot explain their thoughts and feelings in detail. Therefore, questionnaires do not provide very deep data and there is need for validation with other research instruments. Another problem, which, however, also applies to other research instruments is, that it is not possible to tell how accurate, truthful and realistic participants’ responses in questionnaires are:

Respondents give inaccurate answers for a number of different reasons. They give inaccurate answers both consciously for reasons of their own, and also without any conscious realization that the information they are giving is inaccurate. The researcher must be aware of these inaccuracies, try to minimize them and, where necessary, take into consideration the bias and inaccuracy in the data. (Brace, 2008, p. 195)

Respondents might forget to include important information in questionnaires, or deliberately withhold it, exaggerate or lie. The latter often occurs when social desirability bias arises. Although the anonymity of questionnaires eliminates much of the bias, they are not completely resistant to it and this has to be considered by the researcher.