Monday, November 27, 2017

Grebenik and Moser

This single case study will be conducted using a qualitative research method. This method is appropriate for this research project as it enables researchers to develop ideas and gain an understanding of society through 'studying the way that people make sense of their world as subjective individuals'. Silverman (2001) discusses qualitative research stating, "the methods used by qualitative researchers exemplify a common belief that they can provide a 'deeper' understanding of social phenomena than would be obtained from purely quantitative data".


A method needs to be chosen for data gathering, which is done in accordance with the type of research required. An interview is necessary for this research project as a questionnaire previously failed to provide sufficient information. There are various types of interviews, which Grebenik and Moser (1962, cited by Bell, 1999) believe to be placed on what they refer to as a 'continuum of formality'. At one end of this continuum we find the 'formalised interview', also known as a structured interview that uses a format such as a questionnaire. At the other end we find the 'informal interview', also known as the unstructured interview, which is "determined by individual respondents".

For the purpose of this research project the interview is to be conducted using the semi-structured interview technique. A semi-structured interview, which Bell (1999) refers to as the 'guided or focused interview', encourages "freedom to allow the respondent to talk about what is of central significance to him or her", but still maintains "some loose structure to ensure all topics which are considered crucial to the study are covered". (Bell, 1999)

Bell furthers this comment by saying that the structure is "established by selecting topics around which the interview is guided". Therefore it is necessary to consider the specific information required and develop some sort of structure. A structure devised in relation to the topic being researched also avoids collecting a mass of information that has to be worked through, which actually has no relevance to the topic being researched.

An interview structure or schedule consists of the questions that have been formulated from the topic(s) being researched. (refer to appendix 2) These questions also include prompts or probes to assist the interviewer should they wish to develop an area that the interviewee may have problems discussing. However, these prompts are only examples of questions that may be used. Often is the case that questions must be improvised in accordance with the interviewee's answers. This occurs as the interviewer may wish to question the respondent about a topic mentioned, enabling the expansion and development of answers and ideas being discussed.

It is important to remember that upon conducting the interview permission must be granted from the respondent stating that they understand how the data is to be used, and that they consent to the use of this data. (refer to appendix 3) It is also advisable to devise a list entitled 'interview preparation', to ensure that all important factors are completed before proceeding with the interview. (refer to appendix 4)

Bias When conducting an interview it is important for the interviewer to remain objective toward the topic being researched. However, this can be difficult sometimes as Sellitz et al (1962, cited by Bell, 1999) recognises, "interviewers are human beings and not machines". Bell (1999) continues that therefore, "their manner may have an effect on the respondents", which may bias their opinions and answers.

Bias can be a particular problem within a single case study as it may go unnoticed. This is unlikely to occur within a team of interviewers, as the data analysis would highlight any consistencies that appear bias. It can be very easy for an interviewer to 'lead' an interview, as just by asking a question with "a different emphasis and in a different tone of voice can produce very different responses". (Bell, 1999) If an interviewer does feel that they have a particular opinion upon a topic being researched within the interview they must ensure that they are aware of this bias and apply self-control to avoid affecting the respondents answers.

Reliability Bell (1999) describes reliability as, "the extent to which a test or procedure produces similar results under constant conditions on all occasions". Were this applied to a research project conducted by several interviewers then the reliability of the data could be recognised from the consistency of answers. As this is a single case study no comparisons can be made to check the reliability of answers. However, this study asks for the opinions of the respondent and "questions which ask for opinions may produce different answers for a whole range of reasons". (Bell, 1999) The interviewee may be affected by something they had previously seen on T.V. or read in a newspaper, or the surroundings may affect the answers given. For this reason the reliability of answers may be doubted.

Validity Clarke and Oxman (2003) describe validity as, "The degree to which a result (of measurement or study) is likely to be true and free of bias (systematic errors)", therefore stating that a valid measure is one that measures what it claims to and not something else. The application of External validity is most appropriate for this study as defined, "external validity concerns making generalisations about your results". This observes "how well the conclusions of your study apply to other people in other places at other times". This is the aim of the study to answer a question and yet be able to apply it to the population.

Validity and reliability are closely linked and through the improvement of external validity we can also improve the reliability of the information collected. Trochim (2000) suggests three ways that external validity can be improved: draw a good sample from the population, show data about similarities between various groups of people, places and times, and replicate the study to produce stronger results. Again not all of these can be applied to a single case study, but they would be valuable aspects to remember for future reference in studies involving several respondents.

No comments:

Post a Comment