5.3 Unstructured, Semi-Structured and Structured Interviews

welcome to another video in the course where we are learning how to write a proper research disease now in the previous video we have started to discuss primary data collection methods and it was observation now in this video we are going to discuss our next option of primary data collection method and it's going to be interviews and interviews are used really a lot especially by bachelor and master students in business administration so let's go for the official definition of what a research interview is so the research interview is a purposeful conversation between two or more people requiring the interviewer to establish a rapport to ask concise and unambiguous questions to which the interviewee is willing to respond and listen intently now I think we all get the idea what an interview is but a research interview is something a little more specific and a little more advanced if we will not do it really properly because we cannot allow any bias to a cure you know bias for instance can be that we are leading the interviewee in some direction and our result will be biased or generally my idea is that it is a bit harder to conduct a proper research interview than we can imagine therefore we need to learn a few things about interviews before we really tried moreover there are three kinds of research interviews that we can conduct and we need to select the one that fits with our research nature so we'll go through it now and so that you can choose which type of research interview fits your research the best first of all we have structured interviews these really are very standardized interviews in this case you as a researcher coming to an interview and you are bringing yourself literally a paper on which there are given fixed standardized questions in a predetermined order and nothing will change they will ask all of these questions and you will record the interviewees answers to them I mean on one hand it has sort of something fit because as the interview is standardized you know your role and your only task is sort of to just conduct the answers to these questions because later on when we will be talking about unstructured interviews which is the complete opposite then we also need to have some sort of a charisma and really a knowledge of the context that you are researching so we standardized interviews these are quite simple to conduct second of all we have semi structured interviews and I think you can guess it correctly we again as a researcher are coming to an interview with a predetermined list of questions but in this case we are not that fixed to these questions and the order of questions can change we can also take away some questions and do not ask them and also we can maybe add some questions if we during the interview find out that some particular topic is very interesting so there is a semi structured interview and finally we are going for an unstructured interview this is sort of a free conversation when we as a researcher we come to an interview we sit down with the interviewee and the only thing we do is that we say some that's a topic for instance accounting we are researching companies and we just sit down with the head of accounting of the company and we'll let this person to freely talk and this person is going to give you some subjective feelings or some perceptions of the events that have happened so you see these are the three formats of interviews that we can conduct now we need to learn which of these three works well with which research nature do remember many videos ago when we have been choosing research nature it was exploratory descriptive or exploratory now it depends which of these three we have chosen we will have to accordingly choose our type of research interview so let's begin if we have chose an exploratory research nature meaning that we are trying to explore this new curing phenomena then we should go for unstructured interviews and maybe we can use semi structured interviews so just think about the situation you are exploring this new phenomena which has not been researched yet so of course the obvious choice is to try to interview people and let them talk freely because you as a researcher do not help that much knowledge of the context of the phenomena that you are researching so unstructured interviews are the best here also you can use semi structured but here I have a strong advice for you what I have seen many times is a very successful primary data collection metal was sort of two rounds of data collection under exploratory research nature so the first round of data collection was unstructured interview maybe one or two during this unstructured interview you as a researcher are learning about the phenomenon so what say you will learn some information that will allow you later on to construct questions for your second round of data collection which will already be semi structured interviews so you do two rounds one instruction and one semi structure this is a very common and useful practice second of all we have the descriptive research nature or descriptive study here it is strongly recommended to use structured interviews because when you remember back in the days when we were choosing our research nature the descriptive research nature is about examining the relationships between variables that are occurring within a phenomena and now within this structured interview you really can construct the questions the way that maybe each question represents one relationship between two variables or one question represents a variable then another question represents the other variable and then third question represents the relationship between these two so really structured interviews work very well with descriptive research nature and finally we have explanatory study here we can use both semi and structured interviews but over have a little bit different use so previously conducted studies because if you want to do exploratory study you have to have a background of a lot of studies that have been conducted before your work takes place so all these previous studies have identified some variables and some relationships so you know these already you just need to explain them that's why your study is called exploratory now you can use structured interviews to help statistical analysis of these relationships so that's why the structured interviews work well now also the semi structured interviews work well if you want to use qualitative explanation of these relationships so both of these work very well with exploratory studies all right now we have gone through all three types of interviews that we can help and we have also discussed which of these types works well with which of the research natures now there is a bit of downside with interviews as a primary data collection method and those are some issues that are connected with this method so let's discuss this first of all there is issues with reliability if you remember it is maybe 20 videos back in this course we have discussed reliability and validity of research reliability basically refers to the fact whether when someone else on some different place may be under some bit different context in a different time would conduct the same interview at the same research with the same aim and same methods whether the results of you and this other person would be similar or may be the same so there's a reliability of research if they would be the same then your research is considered reliable now if you are using interviews especially unstructured interviews then this is a real threat to a reliability of your research because let's say that you interview 10 people or you interview even 20 people which is considered quite a lot because maybe one is going to take you over only half day so 20 interviews is a lot even though when you contact such large amount of unstructured interviews if someone else in some other company would conduct 20 interviews the responses that you do these two researches would get would be very dissimilar that means that the unstructured interviews are really a bit of a threat to a reliability the solution to this usually is that you try to make your population smaller you do not try to generalize that much with your results so let's see if you are conducting 20 unstructured interviews you do not try to claim that your results are applicable globally and that all companies around the world can rely on your results no you will make your generalizability a little bit smaller maybe for your country or adjust for your region so that is the first threat to a reliability second of all we have the interviewer and interviewee bias I think you can pretty easily imagine this when two people are talking I'd say it's a business meeting both of them need to behave professionally and the same is of course true for a research interview many biases cannot cure for instance you are interviewing employees of a company and now manager of this company says that he wants to be attending these interviews so he is sitting in the corner of the room where the interview is takes place now the employees are not maybe going to be dead honest with their answers when they know that the manager is listening to them as well Finley connected with this is that you need to record your interviews let's say on some mobile device or something like that often people do not like it and just the fact that they know that you are going to be recording them can cause a bias to a cure and the same is true for you as a researcher if you will be asking the questions in a leading way or you will be forgetting to ask some questions well that's also a bias so you need to be really careful that no interviewee no interviewer bias will occure and final Fred we have already discussed a bit and that is a threat or issue of generalizability of a research that uses interviews as a primary data collection method because usually the practice is to conduct 10 20 50 interviews and all of a sudden you would like to generalize to a population maybe to the world market which can be thousands of companies so when you are going to conduct interviews you need to justify the selection of your sample most likely you are going to have nonprobability sampling techniques and that means that you really need to be sure that you clearly state why have you selected these exact cases and so that when you interview them your research results or the outcomes of this will be generalizable to some extent of course when you are using interviews it's a qualitative study mostly and you will be unable to use the statistical analysis so there is nothing wrong about it just that interviews are very specific data collection method and you need to justify the selection of your sample so those were the interviews and in the next video we are going to discuss questionnaires which is really a different type of primary data collection method

Channel: MeanThat & Authentic Data Science

Views: 86,154

Rating: 4.8501563 out of 5

Keywords: research, research thesis, writing a thesis, writing academic paper, writing academic research, research design, research approach, literature review, research strategy, primary and secondary data, sampling of research, interviews, research interview, unstructured interview, semi-structured interview, structured interview

Id: gzwGEBJGz8s

Length: 12min 16sec (736 seconds)

Published: Sat Apr 09 2016