Wednesday, May 8, 2013

After reading The Coding Manual for Qualitative Researchers by Johnny Saldaña

Most people would start off this blog post with: "I don't often get this excited about a research book."  I am not most people.  I do, often, get this excited about a book.  Still, it's worth a write-up.

The Coding Manual for Qualitative Researchers* is my first attempt to go mainstream.  What I mean by that is that I am not a methodologist.  Researchers use methodology in the same way that social climbers use name dropping.  It's oftentimes a contest to see who's in the know, who was educated more, and how you should be labeled.  Envision a group of researchers with turned-up noses at a cocktail party.  "You're a grounded theorist?  Oh.  I get it.  I have to go see about a thing..."

Fortunately this book never leaves you feeling like you showed up to a stats party in your ethnographer's dress.  It's basic.  It goes through first cycle coding methods (what you do when you first look at the material), second cycle coding methods (can be deceiving - you may have already looked at the material more than once before this), and gives useful ideas for memo-writing and pre-writing.

One useful part came toward the end of the book.  Saldaña mentions several focusing strategies that researchers can use when starting to write, one of which is called "The 'top ten' list".  He suggests extracting "no more than ten quotes or passages" and essentially arranging and rearranging them to give you further insights into your data.

Thus, we have what is, for this blog, my list #1.**

List #1: Top Ten Useful Findings from Johnny Saldña's TCMQR 

  1. Use simultaneous coding.  Because I was afraid of labels, I was already doing this in my research and not knowing what to call it.  It is essential to the way I use NVivo in my work.  Frequently my colleagues ask me to find organizations in our network that meet certain criteria.  For example, someone will find a grant opportunity that will apply to a set of network members, but not others.  They may ask me if I have heard of a rural affiliate with a great tutoring program, or a large affiliate that utilizes multiple university partners.  That's why it's essential that I code my interview data with multiple codes.  Simultaneous coding allows me to search the database of my interviews and find specific passages that refer to promising practices throughout our network.
  2. Memo writing is time-consuming but essential.  I don't know about you, but when I'm done coding a passage of interview questions, the last thing I want to do is write a memo.  Interviews are not always the most exciting things.  In fact, they are oftentimes quite repetitive.  To have to go through codes upon codes of data is exhausting.  Fortunately, Saldaña mentions that there are multiple ways to write memos.  I don't have to stick to the research questions in my memo writing.  I can think about the coding process (why am I using so many nodes for needs and not enough nodes for programs?); emergent patterns in the data (wait, is it possible that interviewees are mentioning needs but not aligning them with appropriate programs?); networks (are our affiliates in close proximity to each other doing better at matching needs with appropriate programs?); problems with the study (should I be asking better questions about matching needs to programs?); and future directions (maybe the next iteration of the study should focus only on practitioners who select appropriate programs). 
  3. Qualitative analysis is cyclical.  It's iterative.  It takes time, and patience, and the willingness to go back to your data with a fresh set of eyes to see if anything needs to be reworked.  All along, I thought that quantitative analysis was hard.  For me, qualitative analysis might be even more difficult.  I. Am. Not. A. Patient. Person.

    Note: Hold on to your hats; the next several items are about methodologies.
  4. Magnitude coding is useful for evaluating content.  I'm using this a lot in my work.  Initially, I referred to it as directionality.  Essentially, you take a bit of text and evaluate the content.  Take this sentence, for example:

    "If we lose this grant, we're going under."

    I can use simultaneous coding here and code for "grants," "cuts," etc.  I can also code it as "finances" and "negative".  Now, I realize I am applying a judgment here, but in my particular line of work, this is a good bit of information - we need to know if an interviewee feels that their organization is holding on by a thread.  It helps inform how we work with them.  That's the point of this blog; research informs practice.  Another sentence may be:

    "We worked with our local United Way and applied for a grant.  We received $1 million to support after-school tutoring."

    I can simultaneously code this with "partner," "non-profit," "tutoring," and "growth."  If an organization's financial health is important to me, I can also code it as "finances" and "positive".
  5. Use descriptive coding to make life easy for yourself.  This is essentially what I've been doing while analyzing the hundreds of interviews I've collected over the past several years.  You take a chunk of text and label it.  Done.  I love it.  (Wait, maybe I do love labels.)
  6. Evaluation coding is essentially the best of magnitude and descriptive coding.  This is my happy place.  I look at the data and think, what is this person saying?  Is it good, bad, or neutral? Then I use that information to inform my colleagues about what is working and what isn't.
  7. Provisional coding is useful if you already know going into a study what you want to highlight.  This is another type of coding that I use all the time.  My colleagues want to know about specific programs and initiatives that are common to our network, and having a set list of codes for these programs and initiatives can make the coding process easier.
  8. "[T]hematic analysis allows categories to emerge from the data."  This is something that is difficult for me, because it requires me to constantly step back from my precious nodes and allow broader themes to emerge.  Themes can emerge from what is being said, but also from what is not being said - and by whom.  If you're stuck in your analysis, and want to come up with broader, more dramatic theory, sometimes thinking of themes across the data will help out.
  9. Remember what is most important for the audience.  Or, as he puts it, "I appreciate being told early in a report what the 'headline' of the research news is..."  People don't have time, or patience (see #3 above).  Get to the point of your story, and quickly, or they'll flip the channel to a reality show.
  10. "[T]here are times when it's more powerful to end a presentation with tough questions..."  Although finding themes isn't my favorite thing, when I do finally get to the point, I like to tell everyone exactly what it means.  In my line of work, where practitioners and support staff aren't keen on research anyway, it's occasionally helpful to ask *them* to construct meaning.  In other words, rather than coming up with my own ideas about why some of our affiliates are challenge-focused and others are solution-focused, I should ask practitioners why this phenomenon is happening - and what they would do to solve it.
There's more, but I'll leave that for you to discover.

Saldaña, J. (2009). The Coding Manual for Qualitative Researchers. London: Sage.

*FYI, I used an earlier version - but I want you to have the most recent one!
**A caveat: This is what is useful to me, in my work - you will certainly find other useful tidbits.


  1. thank you for your comments. I am a doc student working on a qualitative dissertation, is there a contact number or an email that I could contact you?

  2. Thank you for share this informative post.

  3. We need to know if an interviewee feels that their organization is holding on by a thread. It helps inform how we work with them.
    qualitative research analysis


  4. Some sort of blog about learning importent document. If an organization's financial health is significant to me, I can likewise code it as "finances" in addition to "positive". Thanks for sharing that.analysis of qualitative data

  5. In my opinion, great content is above all. So focusing on creating good quality of content is the solution.


    Melbourne SEO services

  6. I read this article. I think You put a lot of effort to create this article. I appreciate your work.
    thesis Writing Service

  7. Your website homepage is very nice and important for me. Thanks Johnny for sharing this type post. This post is mainly Coding Manual for Qualitative Researchers relevant. I am waiting for your next post.

  8. Hi there, I found your blog via Google while searching for such kinda informative post and your post looks very interesting for me.writers have great importance in the writing company because they are the ones who solve customer's problems with respect to their needs. SEO specialist || Link Building Expert

  9. If you’ve just completed a large surveying activity, forum conversation or just have a large amount of written feedback and comments to make sense of, then you might need to utilise qualitative analysis techniques.