Research methods in applied linguistics : quantitative,qualitative, and mixed methodologies
Citations
645 citations
561 citations
Cites background from "Research methods in applied linguis..."
...As a result, somewhat lower Cronbach alpha coefficients are to be expected, but even with short scales of three or four items, we should aim at reliability coefficients in excess of 0.70; a scale with a Cronbach alpha that does not reach 0.60 should sound warning bells (Dörnyei, 2007)....
[...]
...The resulting z-scores express how much each raw value is different from the group mean, and by equalizing the means, scores obtained from different sources are readily comparable (Dörnyei, 2007)....
[...]
...We are aware of only two published studies (Cheng & Dörnyei, 2007; Dörnyei & Csizér, 1998) that had the explicit objective to provide empirical data on the effectiveness of motivational strategies....
[...]
537 citations
Cites background from "Research methods in applied linguis..."
...Yet, this overlooks the question that Dörnyei (2007) refers to as often not mentioned in research projects, which is weather to make a single or multiple interview sessions....
[...]
...Dörnyei & Skehan, 2003; Nazari, 2007; Talmy, 2010) showed that interviews are one, if not the primary, method of investigating linguistic phenomena. More specifically, Block (2000: 757) states that the rationale behind using interviews in linguistic research is “to take research participants ‘at their word’ ...[that is] to offer presentation of data plus content analysis, but no problematization of the data themselves or the respective roles of interviewers and interviewees”. As noted by Paylenko (2007), when using narrative data, people’s description of their linguistic behaviour (e.g. language learners' reticence) may not always reflect the reality. He, therefore, maintains that interviewing might be the successful alternative technique in testing linguistic phenomena. However, although this may indicate that interviewing is more than just describing facts; presenting events might differ greatly depending on the language of telling (Richards, 2009). Yet, to overcome this, Mackay and Gass (2005: 174) state that “...depending on the research questions and resources available, interviews can be conducted in the interviewee's L1, thus removing the concerns about the proficiency of the [interviewer] impacting quality and quantity of the data provided”. Additionally, however, analysing interview linguistic data can be challenging with regard to the co-constructed nature of oral responses. That is, although some researchers would analyse the narratives as if they would do any oral or written analysis, this method tends to ignore the interactional influence on the presented data. Paylenko (2007) argues that analysing interviews using traditional approaches is more likely to create a distance between the interviewer and his/her interviewee, which would therefore produce an incomplete snapshot of the investigated phenomenon....
[...]
...That is, as Dörnyei (2007) indicates, a one-hour interview may approximately take up to six to seven hours to transcribe, and around fifty pages of transcript. However, to begin with, analysis can be done through coding. Two steps of coding have widely been discussed in social science research: a) generating meaningful data units; and b) classifying and ordering these units. Accordingly, such a process is expected to deliberately reduce the amount of data collected. Creswell (2009) notes that the analysis process should also be reflexive, i....
[...]
...That is, as Dörnyei (2007) indicates, a one-hour interview may approximately take up to six to seven hours to transcribe, and around fifty pages of transcript....
[...]
...Dörnyei & Skehan, 2003; Nazari, 2007; Talmy, 2010) showed that interviews are one, if not the primary, method of investigating linguistic phenomena. More specifically, Block (2000: 757) states that the rationale behind using interviews in linguistic research is “to take research participants ‘at their word’ ...[that is] to offer presentation of data plus content analysis, but no problematization of the data themselves or the respective roles of interviewers and interviewees”. As noted by Paylenko (2007), when using narrative data, people’s description of their linguistic behaviour (e.g. language learners' reticence) may not always reflect the reality. He, therefore, maintains that interviewing might be the successful alternative technique in testing linguistic phenomena. However, although this may indicate that interviewing is more than just describing facts; presenting events might differ greatly depending on the language of telling (Richards, 2009). Yet, to overcome this, Mackay and Gass (2005: 174) state that “...depending on the research questions and resources available, interviews can be conducted in the interviewee's L1, thus removing the concerns about the proficiency of the [interviewer] impacting quality and quantity of the data provided”. Additionally, however, analysing interview linguistic data can be challenging with regard to the co-constructed nature of oral responses. That is, although some researchers would analyse the narratives as if they would do any oral or written analysis, this method tends to ignore the interactional influence on the presented data. Paylenko (2007) argues that analysing interviews using traditional approaches is more likely to create a distance between the interviewer and his/her interviewee, which would therefore produce an incomplete snapshot of the investigated phenomenon. Yet, to resolve the issue, I would follow Sidnell (2010) who suggests using an interactional/conversational approach (e....
[...]
344 citations
285 citations
Cites background from "Research methods in applied linguis..."
...I have argued elsewhere (Dörnyei, 2007) that mixed methods research (i....
[...]
...I have argued elsewhere (Dörnyei, 2007) that mixed methods research (i.e., the meaningful combination of qualitative and quantitative approaches) offers a radically different new strand of research methodology that suits the multilevel analysis of complex issues, because it allows investigators to…...
[...]