Coding can be done any way you choose. If you are talking about hand coding this is how I did it.

1. I read through my transcripts, while listening to the recording to make sure my transcriber did not miss any words or caught as many as possible. With recordings some words can be inaudible.

2. I had two highlighters with different colors. One for the question and the other for the words or phrases that stuck out.

3. Once finished, I opened a new word document and copied the question and listed the words or phrases under that question in numerical order and the page number.

4. After doing this to the entire document, I went back and read each word or phrase and then wrote the theme, I thought it best fell into. Hand coding was difficult at first but as I learned it became easier. I uploaded the transcripts into NVivo and re-coded them there. I got much better results with that program. If you do not have to handcode, I say skip it and use a program instead. Even though I hand coded, I was told to use a program because with handcoding you can miss a lot of information. It doubled my work so I would have just used the program had I to do it over again.

Find the way that is most comfortable for you, there is no one way to do it.

Here is something that might help you!

https://www.sagepub.com/sites/default/files/upm-binaries/24614_01_Saldana_Ch_01.pdf

Thank you Laverne A Davis for your immediate response. How you got Nvivo? Do you have any example to understand the concept clearly like how to make code and theme for the question

First of all: I would strongly recommend to use a software tool for coding, for example MAXQDA (I think it's still free for a limited period of time). This makes it easier to change or refine your codes during the process of data analysis.

Second: How you can or should make your codes also depends on how you did collect your data. In case of Grounded Theory you could have a look at Kathy Charmaz (2006) [Constructing Grounded Theory. A Practical Guide through Qualitative Analysis], who provides some examples in Chapter 3. There is one example in Mayring (2000) [http://www.qualitative-research.net/index.php/fqs/issue/view/28] - maybe you can also find other examples in FQS?

All the best,

Eva

Dear Chris,

During my dissertation writing process, I found these articles very helpful regarding analyses of qualitative data. . Hopefully, they will helpful for you too.

Best of luck for your research.

The following publications/link (see communications/referenced articles) may further help:

• Basit, T. N. (2003) Manual or electronic? The role of coding in qualitative data analysis, Educational Research, 45, 2, pp. 143-154.
• Fereday, J. and Muir-Cochrane, E. (2006) Demonstrating Rigor Using Thematic Analysis: A Hybrid Approach of Inductive and Deductive Coding and Theme Development, International Journal of Qualitative Methods, 5, 1, pp. 80-92.
• Saldaña, J. (2013) The Coding Manual for Qualitative Researchers. 2nd edn. London: SAGE Publications Ltd.
• How to Code Qualitative Data?: https://www.researchgate.net/post/How_to_Code_Qualitative_Data/1

Thanks everyone. I would like to do manually the analysis. I need any data of research in which researcher has done the qualitative analysis. Like how they coded the transcript, how they developed theme and sub themes? What is the interpretation?

I have done all my qualitative analysis by hand using Excel. First I transcribe all the interviews by myself and read and re-read thew transcipts several times to get a grip of the content. After that I copy the entire interview transcript to Excel with every answer and every answer in a separate cell. If the answers are too long to be visualized in one cell I divide it on severel cells. I put all the interviews in the same Excel book after each other to allow for sorting of the codes once I have them in place.

Then I copy alll meaning units into seperate cells in the next column and in the next column after that I write a condensation of the meaning unit. In the next coulmn after that I put the code for that meaning unit, that is a word or a few words telling about what the meaning unit is about, like a label. When I have done so for all the interviews I sort the codes in alphabetical order and gather similar codes together based on differences and similarities. After that I group the codes in bigger gropus, that is preliminary cathegories or themes. Maybe they are built up by sub-cathegories or sub-themes. That depends on the material.

I'm attaching a qualitative paper that I authored. The methods section should give you an idea of how to code unstructured text and help you understand some qualitative concepts. Hope this helps. :)

Just follow the step-by-step analysis by Braun & Clarke,2006,2012 Book chapter that Muhammad gave you and you will not get it wrong.Exercise patience, do not skip any step and read a lot of articles as you go along and you should be fine. Transparency is critical in explaining every step you take with a rationale and citations. Mariam`s article is a good example of how you put together an article and explains in simple terms what the audience wants to read & understand!!

Good morning Chris, sorry I just saw your question. NVivo is difficult to use but it is fun once you get used to it. You can auto code (I did not like because too many random words got chosen) or you can read every sentence and pull out words and phrases and put them under each node (theme) you think it best fits. It is a very time consuming process. What is great about NVivo is, while you are creating your nodes, the program is creating, charts graphs, pies, word clouds, trees etc. You choose whatever way you prefer to show your graphics. The best part I found, is that it also creates your codebook for you. Though codebooks can be created by hand in anyway you choose, creating it with NVivo takes 10 minues or less. Just be certain that this is the very last step that you do after all of your nodes have been chosen. I hope this helps.

Dear Chris,

Plz find enclosed raw data file of focus group discussion along with results. Hopefully, it will guide you that how coding and theme was done.

Dear Chris

i too manually coded the data on my ongoing doctoral work. Just to add my two cents worth to all the useful suggestions given above:

1. First step being to transcribe your audio files to a written document

2. I read through each of them several times to familiarize myself with the data

3. At the same time (unless its a grounded approach you are taking) you should be keeping in mind , the research objectives - what you hope to find in this study specifically, and the theoretical lens/es you are using.

4. With these two (your objectives and theoretical concepts) in the back of your mind, when you read through the transcriptions, you will be able to pick up interesting/important words, sentences, paragraphs etc. that you think will be relevant to address your research questions

5. These words/sentences/paragraphs are your initial codes. I created simple MS word tables - just two columns , for the participant's name and the codes selected

6. After getting all these initial codes in to a simple list / table, I then, re-read them several times , again with the research questions and my theories in mind --- at this stage, I could identify certain broader themes to fit the initial codes.

For e.g. one of my research questions was to explore identity formation through (traditional vs consumerist) values. So for instance, I initially identified any words/sentences/paragraphs that indicated a person's identity being shaped by traditional values (and not by consumerist values).

- The list of words/sentences/paragraphs were my initial codes. Then going through those initial codes in the next round, I found that these codes fitted in with different themes - e.g. traditional influence of parents and extended family / already achieved conflicting materialist goals due to their mature life stage (so no need to identify with consumerism) / status gained through alternative means such as education and career (so no need to identify with consumerism).

- These themes were what ultimately belonged to the general category 'identity formation through traditional values' . The same process continued for identity formation through consumerist values, and a similar approach was taken for my other research questions. I later wrote my findings based on these themes identified.

- Also remember that this has to be later theorized as per your theoretical lens/es (unless its a grounded approach). And do remember, this coding process can take a considerable amount of time and repeated work. I too, initially tried NVivo but was not successful. I found that doing it manually was personally much more convenient!

Hope this will help you with your work. Good luck!

There are some excellent diagrams throughout the pages of Miles and Huberman, second edition, that I used in my qual research-- each one called a called matrix. There are varous for differing types od data and methods used.

Thank you all for your valuable information.

Please refer to Leech N.L & Onwuegbuzie A. J (2007). An Array of Qualitative Data Analysis Tools: A Call for Data Analysis Triangulation.

This will guide you to analyse your data with ease.

Read the transcription>Code data which related with study objectives>Input in matrix table> gather similar findings>generate theme>analysis the findings

Read The Coding Manual for Qualitative Researchers by Johnny Saldaña. A vast array of coding choices.

I am sharing a list of steps I am presenting to my grad students in our qual analysis class.

Read the raw data>find the relevant findings based on the objectives>see the pattern>generate theme based on the relevant findings>see pattern of findings

Md Alamgir Hossain is this possible with classroom observation data?

yes