Analysis of the results of the conversation
The processing and analysis of interview data is a complex and time-consuming process, the characteristics of which are determined both by the goals and objectives of the study, and by the features of interview construction, the use of different types of questions, etc.
Depending on the information received (mainly qualitative or mainly quantitative), an interview was sent, the methods of processing will differ. Thus, the quantitative processing of data obtained on closed questions can be reduced to constructing the frequencies of the selected variants of answers, to a description of the group of subjects as a whole or to the comparison of the answers among different subgroups of respondents on their basis. However, this type of processing is mainly used for mass surveys, while interviews that involve the collection of qualitative data require a meaningful interpretation of respondents' responses that go beyond simply correlating the frequencies of the most frequently encountered answers.
In any case, the question of how to handle the results should be asked at the planning stage of the research, rather than after
of how the interviews were conducted. The matter is that the chosen method of analysis can influence how the interviewer will behave during the conversation: will he make more or less clarifications, will he focus on TC or other aspects of the answer, etc.? For example, if further categorization of respondents' answers is assumed in the pre-planned categories, the interviewer will try to bring the respondent to the answer in these categories.
There are different approaches to the analysis of qualitative data - for example, narrative analysis, phenomenological analysis, construction of sound theories, etc. They were developed within the framework of different schools of humanitarian thought, the specifics and detailed exposition of which can be found in authors working in the corresponding traditions. We will consider the main steps in the analysis of interview data in a simplified form.
One of the first recommendations for a qualitative analysis of the interview is that the treatment should begin as early as possible (ideally, immediately after the first interview). This is due to the fact that the analysis of qualitative data takes a very long time.
The first step in processing qualitative interviews is decryption of the conversation record (since most often the interview is recorded with the help of audio or video fixing), i.e. the creation of a written protocol, the transfer of all the replicas of the interviewer and the respondent spoken during the interview to a text document, so that the resulting text fully and literally reflected the entire course of the conversation. Further all data analysis is carried out on the basis of these transcripts, therefore it is important that they be made qualitatively. Interpretation of interviews is a long and complex process, which can take many times more time than conducting the interview itself.
When decrypting, especially if it is performed by several people, it is necessary to say in advance how really detailed it should be. Whether it is necessary to decipher all the sounds made by the respondent, including interjections; whether to indicate pauses in the speech, and, if so, whether to separate more and less long pauses, and to what degree of accuracy they should be measured; whether it is necessary to record in the protocol absolutely everything that the respondent says, or, if he has moved away from the topic of the interview, this piece of the answer may be missed, etc. At first glance this seems insignificant, however such questions will necessarily arise before the person who will do the decoding. And the answers to these questions depend on how the results are to be processed, i.e. should be known to the researcher and in no case should be given to the decipherer by the decryption (if the decipherer is not concerned with the researcher himself).
Sometimes, in order to fulfill confidentiality requirements, some mention of the names of own, circumstances and events that can be used to identify the respondent's identity should be removed from the protocol already at the decryption stage.
It is important to understand that deciphering an interview record and actually interviewing is not the same thing. The text can not fully reflect the entire situation of the conversation. Therefore, sometimes in the further analysis of the interview, the researcher may still need to address directly to the audio recordings.
After that, you can proceed to the analysis of the material received. In fact, the processing of qualitative interviews is a content analysis that involves the allocation of categories in the analyzed material. Processing can then be reduced to obtaining quantitative data on the frequency of occurrence of certain categories or continue to qualitative generalization of these categories in the category of higher order, after which attempts are made to build on the basis of the material some theory explaining them.
The analysis of the received material begins with its coding. Coding is the selection of categories grouping data, and attributing to the parts of the text representing these categories appropriate labels (short notations). For example, codes "Excuse your negative behavior", "Self-control manifestations", "Orientation to the opinions of others" can be highlighted. and the like.
For encoding, the coded system is developed. in which all the codes used, their designations (labels) should be listed, their contents formulated, examples of utterances pertaining to the corresponding code are given. This category system should reflect important concepts from the point of view of research, topics, events, processes. For example, in a study on adolescents' adaptation processes in a new team (when they move to a new school), such categories can be "Contacts with members of a new team", events that are of interest to the researcher may include "First encounter with the team", "Interaction with the leaders of the group ", as the key topics can be considered reasoning about attempts to establish contact with the new group, explaining the reasons for successful or unsuccessful contacts, etc.
In the coding system, the links between categories can be reflected. For example, they can represent a hierarchical structure: some categories will be more general, some - private. For example, the Contacts with members of the new team can include the subcategories Positive Contacts and Negative Contacts & quot ;. Codes can be related to both the content and the form of the text (for example, "Self-irony"), if such a form is important to the researcher.
The initial list of codes can be developed in different ways.
First, the coding scheme can be developed on the basis of literature analysis, borrowed from works that were conducted on similar topics, thought out in advance at the planning stage of the interview.
However, with such a priori given coding systems can be difficult: they can be ill-suited for analyzing the empirical material you have received. Some codes may be unsuitable or not suitable for categorizing the answers of your respondents. Or, on the contrary, you can get information, the codes for which is not provided in the system used.
Secondly, the code system can be developed completely based on the interviews received. The researcher repeatedly reread the transcripts of the interview and, based on the available material, highlights important topics, events and concepts, which later on is included in the code system, fully based on the received data.
Typically, the final code system includes both categories originally designed by the researcher, as well as codes developed on the basis of the received data. It is important that the codes used are directly related to the subject matter under investigation and related to the hypotheses being tested.
Technically, the encoding process can be carried out both on paper printouts of transcripts of the interview (when the necessary parts of the text are marked as related to certain codes using markers of different colors, marks in fields, etc.), and by computer programs.
Fragments of the text that are to be encoded are not necessarily complete finished sentences - they may be parts of sentences, and a whole paragraph. Ultimately, this is determined by the content of what was said and the specifics of the code. So, for example, two parts of one sentence can refer to different codes. Example: "At first, almost no one spoke to me, and it was very frustrating for me." Here the first part of the phrase refers to the description of the circumstances of the interaction in the new group, and the second - to the emotional feelings of the respondent about them.
Interpretations of the source code should be avoided when encoding. In a well-conducted interview, all the meanings put by the respondent in words should be clarified in the course of communication. However, in some cases, interpretations may be necessary. It is necessary to carefully approach the coding of complex and ambiguous utterances and, if necessary, even return to the original audio recordings that can provide additional information about the interview process, the intonations of the respondent, etc., thus clarifying what has been said.
It is important to understand that all the text of the interview can not be coded without exception. To do this, only the most important elements are selected. Some of the statements will remain uncoded, and the remaining statements can be considered as a basis for formulating further research questions.
Too coding should not be subject to too general phrases, repetitions, statements that are not relevant to the topic being studied, and so on. Such fragments of text will not carry a meaningful load in the analysis of results.
In addition, sometimes coding systems can allow one and the same piece of text to be assigned to different codes (for example, to the content and formal). For qualitative analysis, this is a permissible situation.
The analysis of qualitative data is an iterative process, involving a multiple return from later stages of analysis to earlier ones. For example, in the encoding process it may become obvious that the coding system should be supplemented with new codes. In this case, it is necessary to revise already encoded interviews taking into account the changes in the coding system. In the process of coding, it may become necessary to generalize (combine) some codes into one or, conversely, the desire to divide the more general code into several more particular ones.
Also, in the coding process, you can mark separately successful quotations, concisely summarizing what has been said, accurately reflecting the speaker's thoughts, etc., which can later be used in describing the results for quoting.
After the primary data (transcripts of the interview) have been encoded, i.e. the parts of the text (answers, sentences, words and expressions) were assigned the appropriate labels, it is necessary to group all the code fragments encoded in one code in one place. The ego can be carried out manually or using specialized software. This will allow us to more fully cover the content of the code, isolate common points or analyze the differences in the answers in this category for different types of subjects (subjects of different sex, age, etc.).
Typically, encoding begins with developing very low-level categories and assigning as many interviews as possible to them, and then the source code is summarized in a higher-level category. At the first steps of processing, dozens of specific codes can be singled out, which are later generalized into more and more general categories, which in the end should be significantly less. Thus, in the process of data analysis, there is progress from specific fragments of respondents' answers to generalized scientific categories.
Analyzing the system of categories, it is necessary to try to establish links between them, to identify their specifics, to compare the features of their manifestation among subjects of different categories. The full path of analysis is always ultimately determined by your goals and objectives. The result of the analysis can be the construction of theories with respect to the processes under study; in the end, the analysis should allow you to answer the research questions.
For example, in a study devoted to studying adolescents' adaptation processes in a new collective, the stages of this process, possible outcomes, descriptions of the behavior of adolescents and the effectiveness of adaptation, etc. can be described.
Descriptions and theories based on the results obtained may differ in the degree of generalization, depending on the sample used and the research tasks. So, they can be associated with one
a specific case, or refer to the characteristics of the entire population.
The results obtained can be compared with what is described in the literature on this topic. In the above example, the results can be correlated with the characteristics of adolescence described in age psychology, with the results of studies on the same subject conducted by other authors, etc.
When describing the method of processing interviews and the results obtained in a report or article, the whole process of creating codes, coding, generalizing codes, etc. should be described in as much detail as possible. It is necessary to indicate on the basis of which codes were developed, how many people used to encode, how the process of revision of the code system occurred (if there was), etc. All this will allow the future reader to create a detailed idea of exactly how the qualitative material was analyzed, and therefore, on what grounds do the researchers draw conclusions in their work. In addition, the coding system with codes, their values and examples of statements that are coded accordingly should be cited.
All the formulated conclusions and assumptions must necessarily be accompanied when presenting the results with citations from the protocols. However, the citation must obey certain rules concerning both its formal and content side.
When using quotes from transcripts of interviews in publications or reports, it is necessary to receive the verbal or written consent of the respondent. At the same time, in some cases, the respondent can be sent to the negotiation for the text itself, which is supposed to be published. The interview is conducted verbally, and when responding, the respondent uses oral speech. Such a speech, transferred to paper, may look pretentious, and therefore in some cases, the editorial staff is needed to translate the oral speech of the respondent into a form acceptable for writing. Transfer to oral speech verbatim can literally create an impression of a low educational or intellectual level of the speaker.
The attitude to the editing of the original oral text during publication may vary depending on the specifics of the study. For example, research with a linguistic bias is very important to preserve the original wording, because in most cases the transformation of spoken text into written text can make more convenient for perception - the main thing is that in this case was completely preserved expounded by the respondent's position.
In addition, when quoting you need to carefully consider the context in which this or that quotation is given, and make sure that its reduction does not create additional meaning. For example, if you cite a quotation stating that the respondent has devoted a lot of time to improving his educational level, but his words contain some mistakes that can cast doubt on his words, the citation of such a quotation can create a reader
The feeling that you are sarcastic about his words. It is necessary to track this kind of disagreement by choosing quotes.
As for the formal side, the quotes are usually quoted (or stand out somehow differently so that it is clear where your words end and the quoted words of the respondent begin). A quote can not be given in its entirety, if the ejection of some of its elements does not violate the general meaning of what has been said. In this case, the ellipsis must be placed in place of the cut out part of the phrase. In some cases, in order to make the quote more understandable to the reader, it may need to add some missed words that were implied in the context of the interview, but they are not obvious when considering the phrase outside of this entire context. In this case, the added words are indicated in square brackets.
Since the analysis of qualitative data is a process largely tied to the researcher, and the results of such analysis are subject to subjectivity, it is very important to ask how you can verify the reliability and validity of the results.
Several people can be attracted to improve data processing reliability. This will make the system of codes more objective due to their discussion at the stage of allocation. It can also help increase the depth of analysis, since the data will be viewed from different perspectives.
In cases where the ego is particularly important, an independent coding scheme can be applied, when different systems develop their code systems in parallel. This allows us to find out how subjective the process of categorization has been due to the ability to compare the resulting copying systems.
Using multiple encoders will make it possible to check the reliability of the encoding by counting the consistency coefficients (α Cronbach, correlation coefficients), which will enable us to evaluate how well the code system is constructed, whether the codes were well described and whether they were equally applied by different encoders.
A reliability assessment can be performed on a subsample of data when re-encoding by other people is used only for a portion of the source material. In any case, the ego will allow you to assess how recognizable the texts are in the categories used and how much they are evaluated by different people, but it will save time.
To increase the validity of the results obtained, one can turn to a comparison of the obtained data with the results reported by other authors. You can also suggest commenting on the analysis of results and the interpretations made to the respondents themselves. Sometimes respondents can give valuable comments, for example, pointing out that important elements for them were not taken into account and analyzed or the like. However, if the respondent does not agree with your interpretations, you should not assume that they are incorrect, because you and the respondent look at the problem from different positions.
Also We Can Offer!
- Argumentative essay
- Best college essays
- Buy custom essays online
- Buy essay online
- Cheap essay
- Cheap essay writing service
- Cheap writing service
- College essay
- College essay introduction
- College essay writing service
- Compare and contrast essay
- Custom essay
- Custom essay writing service
- Custom essays writing services
- Death penalty essay
- Do my essay
- Essay about love
- Essay about yourself
- Essay help
- Essay writing help
- Essay writing service reviews
- Essays online
- Fast food essay
- George orwell essays
- Human rights essay
- Narrative essay
- Pay to write essay
- Personal essay for college
- Personal narrative essay
- Persuasive writing
- Write my essay
- Write my essay for me cheap
- Writing a scholarship essay