Jump to content


Photo

Using Framework in NVivo

Framework matrices qualitative data analysis summaries

  • Please log in to reply
9 replies to this topic

#1 Kandyw

Kandyw

    Casual Member

  • Members
  • Pip
  • 5 posts
  • Gender:Female
  • Location:London, UK
  • Interests:Qualitative research; applied policy research; methods of qualitative data analysis; training and learning.

Posted 01 October 2011 - 07:27 PM

NVivo 9 now supports the Framework method of analysis developed at NatCen . We thought it would be helpful to share some insights about the Framework method for NVivo users who might be unfamiliar with the approach or who are keen to try out the full range of new functions the latest release of NVivo offers. Our researchers will be sharing their thoughts about and experiences of using the Framework method in NVivo 9 on this forum and we’re looking forward to a great discussion about the Framework approach with other users.

NatCen is the largest independent social research organisation in the UK; we conduct high quality applied policy research across a range of topics from crime to health, political attitudes to programme evaluation. In the 1980’s our qualitative researchers developed Framework as a method of data management that could support high quality qualitative analysis for applied policy research. The creators of Framework Liz Spencer and Jane Ritchie were struck by the dearth of analytical approaches which supported cross-case and within case analysis of qual data (critical for policy research), and which encouraged rigourous, transparent processes for ensuring the data was treated consistently and open to scrutiny. Read more here: http://www.natcen.ac...roach/framework

Framework differs from traditional qualitative approaches to analysis as it does not rely on coding and indexing alone. The Framework method encourages you to organise and manage your research through the process of summarisation, resulting in a robust, flexible and unique matrix output which allows you to analyse your data both by case and theme. It also assists with managing and interpreting data. It facilitates the systematic and comprehensive analysis of all qualitative data sets, from the straightforward through to the more complex.

In summary then, Framework is a matrix-based approach to QDA, which has three elements at its core:
  • It is a case and theme based approach to QDA which is built on the visual display of data in a matrix which is built on a hierarchy of themes and sub themes.
  • Data summarisation and synthesis that reduces the volume of data the analyst has to deal with yet maintains a direct link to the primary data and is fully grounded in the context, language and meanings captured in the primary data
  • A systematic process for creating and populating the matrices meaning the method can be used confidently by individuals or teams who want to collaborate
In this short series of blogs, we'll introduce some of the thinking behind these core elements and provide some hints and tips for getting started with Framework. We know there are lots of Framework users out there who already use NVivo, so we're hoping this short series of hints and techniques will encourage a wider discussion about how the Framework method develops for the future within NVivo. You might also be interested in this recent blog from our Director of Research about the future for qualitative research and our work with QSR.

#2 dambuster

dambuster

    Member

  • Members
  • PipPip
  • 34 posts
  • Gender:Male
  • Location:Brandon Manitoba Canada
  • Interests:Writing Fiction, Poetry re: Busines/Economic/Personal Development from Spiritual Perspective. Qualitative Research along these lines. Media as a research toolset.Developing use of NVivo as a tool for writing and editing of Fiction and Poetry.

Posted 16 October 2011 - 03:35 AM

Hi,
I greeted the framework tool addition to Nvivo with a great deal of relief as a method of quickly and efficiently organizing the large amount of data which was streaming into my project of a variety of types and needing a variety of stages of processing...so a matrix of “source items” in rows and “stages of processing completed” in columns with a capacity to hook on a related item on the right hand panel was genius...great stuff!

However, my excitement was short lived. There seems to be a degree of instability in the feature:

· Intermittent and widespread non-saving of materials in cells, after saving it.

· sudden program crashes when the “Control-V” paste function is used, (and through my dictation program’s paste function), where the program just completely shuts down, and you lose everything not saved, as well as some of the material in cells which have been saved.


This instability seems to be present both when I use the framework tool in a conventional and unconventional way. That is, once I experienced the problem, I did some mini-tests using it as it was designed to be used -- with conventional nodes bearing content to form the matrix, then I used dummy-nodes with and without content to form the matrix -- and the effect was much the same.

Strangely enough, the first time I used the framework tool, I had no problem -- which encouraged me to proceed with my use of this tool. It was when I tried to add new material into the original matrix that I first ran into troubles, and when I tried to make other matrices using the framework tool, the troubles continued.

I gather from the links to the company that had the original framework software, that there is an extensive process which goes along with a framework style of approach to qualitative research. I just wanted to say this in order to alert people who are shifting over to NVIVO, to be aware that there may be problems in the NVIVO version of the framework tool, which are not yet fixed, problems which they may not be experiencing within their earlier software for framework-based research.

I sure hope they get it fixed soon, because it promises to be a great addition to an already very good qualitative research support software... and the deluge of data continues to pour into my project... grin.

Cheers,
dambuster

#3 Kath McNiff

Kath McNiff

    Advanced Member

  • QSR Staff
  • PipPipPip
  • 182 posts
  • Gender:Female
  • Location:Melbourne Australia
  • Interests:NVivo, qualitative research, technical communication, social media

Posted 17 October 2011 - 02:27 PM

Hi dambuster (great name by the way),

That's an interesting idea - using a framework matrix to keep a handle on stages of processing. Did you first make nodes representing source items and stages of processing completed?

Love to hear more about your approach.

Sorry you're experiencing some problems - please let customer support know and they'll investigate it for you - support@qsrinternational.com

Cheers,
Kath

#4 dambuster

dambuster

    Member

  • Members
  • PipPip
  • 34 posts
  • Gender:Male
  • Location:Brandon Manitoba Canada
  • Interests:Writing Fiction, Poetry re: Busines/Economic/Personal Development from Spiritual Perspective. Qualitative Research along these lines. Media as a research toolset.Developing use of NVivo as a tool for writing and editing of Fiction and Poetry.

Posted 24 October 2011 - 02:04 PM

Hi Kath,
  • The only thing I got to work (for a while) was using "dummy" nodes.
  • I figured (rightly) that a node could be used without it actually being a live and functional node that is actually connected to anything. So I made some just for the purpose of generating a framework with cells I could enter info in. I had no intention of using them for "real" coding purposes. It was just to outsmart the software logic.
  • So I made up two folders to contain nodes -- one for rows and one for column labels. That way I could call up each folder and it would kick in the required nodes...and I could tweak each set easily if required, because they were segregated in their own folders.
  • Then a Made up a dummy node for each external, internal item I had in my sources...for the row or caeasily se items
  • I numbered them so they wouldn't keep jumping around when I added a new one and it went alphabetical on me.( Nvivo has this weird numbering system that requires "zero digits") I coded my numbers to that the 4 types of input would cluster when new items were added...but that could be over kill if the project has few types of items...mine has a lot of different types.
  • for the column labels I numbered them as well as I really needed them to stay in sequence
  • To start out I had a jumbled memo I had been using (in vain) to monitor which source I had processed and to what stage. So I linked parts of it to each relevant row, in such a way that it popped up in that nifty right hand window, so I could transfer the material to the relevant cell by copy and paste. I didn't try to keep the memo as I wanted to escape that system and use something more manageable...other wise I would have linked the two but that was overkill...I just wanted to keep my head straight ...or straighter that the memo system did. Once the old material was transferred I unhooked the memo from the rows leaving them free to hook on other stuff If needed, or whatever.
  • Column numbering was stretched by tens so I could enter sub-items if necessary without messing up the system, yet have sub sequences if needed...that explains the weird numbering in the list that follows
  • I just tried to figure out the features on the program and have one column for each feature, even if I didn't use it in the project, or if I wasn't there yet...without being silly about aaaaaalll the features.
  • 01-Internal created; 02 external created; 03 memo created; 04 framework matrix created; 21 coded; 31 explored; 41 analyzed; 51 synthesized; 61 modeled; 71 reported; 81 utilized.
  • I just put a "N/A" in boxes that weren't relevant, and kept notes in the ones which were....point form and brief...detail is all in my project log as I do it, this is just to have a rapid idea where I left off...
  • too bad the program didn't sustain the framework tool (yet)...I'm going nuts...grin...Nvivo is an amazing tool though...so when things go overboard, I just go out an play with my rabbits...
  • cheers,
  • dambuster.


#5 dambuster

dambuster

    Member

  • Members
  • PipPip
  • 34 posts
  • Gender:Male
  • Location:Brandon Manitoba Canada
  • Interests:Writing Fiction, Poetry re: Busines/Economic/Personal Development from Spiritual Perspective. Qualitative Research along these lines. Media as a research toolset.Developing use of NVivo as a tool for writing and editing of Fiction and Poetry.

Posted 24 October 2011 - 02:11 PM

ps.sorry about the creative words in my message...I seem to have a meat-grinder in my porgram which changes my words into some interesting creations...grin...stu

#6 Kath McNiff

Kath McNiff

    Advanced Member

  • QSR Staff
  • PipPipPip
  • 182 posts
  • Gender:Female
  • Location:Melbourne Australia
  • Interests:NVivo, qualitative research, technical communication, social media

Posted 11 January 2012 - 10:23 AM

You can access a recorded webinar from NatCen about conducting framework analysis with NVivo  here: https://www.qsrinter...ork-with-nvivo/
I found it really useful – hope others might too.

#7 Pablo Gustavo Rodriguez

Pablo Gustavo Rodriguez

    Member

  • Trainers
  • PipPip
  • 45 posts
  • Gender:Male
  • Location:La Plata, Buenos Aires, Argentina
  • Interests:Social anthropology, social economy, local development, poverty and cyberanthropology.

Posted 29 January 2012 - 02:02 AM

As I use to work with traditional matrices in Nvivo and write summaries by row or column I created a framework matrix just to compare with my usual "method". I appreciate the function that allows linking fragments of summaries with fragments of text it refers to. But I find uncomfortable to work with big amounts of text within the tiny space of a cell. My projects have many source documents and nodes. In this particular framework matrix I have columns (themes) from A to Z, though only four cases (rows) wich represent international organizations. Sources are documents produced by these organizations.
The difficulty lies in display. If I want to have the associated view opened, I can only view one cell at a time and not all the text it contains, in my old SyncMaster with 1152x864 px resolution. Even if I hide the navigation view, reduce the zoom and enlarge the columns it is uncomfortable, because I have to scroll trough a large amount of text. Things could be better in wider screen monitors with higher resolutions and low zoom. I suggest allowing the user to change font size in framework matrices. Maybe from the properties window or from the options windows. It would be an aditional help.
***********************************************
Pablo Gustavo Rodriguez
Social anthropologist
NES-FTS-UNLP
La Plata, Buenos Aires, Argentina
[url="http://"http://www.analisiscualitativo.com.ar""]http://www.analisiscualitativo.com.ar[/url]
***********************************************

#8 Kath McNiff

Kath McNiff

    Advanced Member

  • QSR Staff
  • PipPipPip
  • 182 posts
  • Gender:Female
  • Location:Melbourne Australia
  • Interests:NVivo, qualitative research, technical communication, social media

Posted 01 February 2012 - 11:00 AM

Hi Pablo,

Yes, I can see the challenges of working with a lot of text in a grid format - here are some things you could try:

- Change the default font size for framework matrices (it applies to any new framework matrices that you create):
  • Click the File tab, point to Info, and then click Project Properties.
  • On the Framework Matrices tab, set your preferences for font size
  • Click OK.
- To change the font size on-the-fly, you can select text and change the size on the Home tab.

- Undock Detail view to make more room.

This help topic provides other suggestions - Customizing the display of framework matrices

Although NVivo prompts you to put your cases in the rows of the matrix you could put them in the columns instead (you can choose whatever nodes you like to populate rows and columns). Not sure if this helps your scrolling problem though.

When you did your summaries for the node matrices (before framework matrices came along) - where did you keep them - in memos or something outside of NVivo?

Hope some of this is helpful.

Cheers,
Kath

#9 docdempster.pd

docdempster.pd

    Member

  • Members
  • PipPip
  • 38 posts

Posted 03 August 2012 - 09:57 PM

I have just exported a framework grid. It is missing entries. Gone back and checked the original within NVivo, text is there. What is going on? How can I overcome missing data not showing up in excel charts.

this is quite scary.

any advice?

#10 Meg Callanan

Meg Callanan

    Member

  • Members
  • PipPip
  • 35 posts
  • Gender:Not Telling

Posted 08 August 2013 - 08:49 PM

I’ve been working with the NVivo Framework matrices feature for a couple of years now and wanted to share a couple of approaches I’ve used. I’d also be really interested in hearing how other users have worked with the framework matrices feature:

 

Using coding and the auto summarise feature to generate framework matrices:

I recently worked on a project which involved thirty half-hour telephone interviews with staff delivering a service. These interviews formed part of a scoping stage for the project and the primary aim was to map the range and diversity in provision so that twelve in-depth case studies could be selected for the main stage of the project.  The study had a reasonably tight deadline and the interviews were fairly short and focused, so we decided to audio record the interviews and write up short (usually 3-4 page) summaries of each interview.

 

Because this data was already in a summarised format, we decided to code the data in NVivo (using a simple coding frame e.g. context of provider, approach to delivery, views on delivery etc). Once the coding was completed, I used the ‘auto summarise’ feature to display the coded data in a matrix with each column representing a different theme of interest, and each row representing a respondent.  This worked really well for this study because the coded data was already concise so fitted well into the matrix output without the need for further summarization. By having the data displayed in the matrix, I was able to see quickly the variation in delivery across the 30 service providers (by reading down the ‘approach to delivery’ column), while also retaining a sense of each individual case by reading across each row. I don’t tend to use the ‘auto-summarize’ feature with verbatim interview transcripts because the quantity of verbatim data is too much to fit neatly into a matrix output (the cells are so large it defeats the point of having a matrix output!).

 

One tip I’d recommend if you do want to auto-summarise coded data, is to make sure the original coded case node is in a format suitable for conversion (choose a small font, and remove double line spacing for example) as the auto-summarise feature puts the coded data into the matrix in the format it was coded in.

 

Using framework matrices without coding

As Kandyw mentions above, the Framework approach does not rely on coding, but instead encourages the summarization of your data into a matrix display that facilitates both ‘across case’ and ‘within case’ analysis.

 

Coding the data as a pre-cursor to summarization can be useful because:

a) It can be a way of piloting the thematic framework you have developed for your data management – helping you refine the structure;

b) For some types of data it facilitates the use of the ‘auto summarize’ feature (as detailed above);

c) Once data has been coded you can select the option to only see ‘cell coding’ in the associated view on a framework matrix and then summarise the coded data.

 

However, I will not always code my data prior to using the Framework matrices features. For example, a recent study I worked on that was evaluating a health service involved nine case studies. In each of the case studies a range of staff (both clinical and non-clinical) and patients were interviewed with a final dataset of over 100 in-depth interview transcripts.  To facilitate the management of this amount of complex data, we used the Framework Matrices feature in Nvivo. Each individual interview was assigned a case node and was displayed in as a row in the Framework matrix. Each column in the matrix represented a different theme that was relevant to the study (some themes were relevant across all respondent types e.g. views of service impact, while other themes were specific only to staff e.g. implementation of service, or specific to patients e.g. motivation for seeking treatment).

 

The data management phase of the project involved summarising data from each case into the matrix cells and creating summaries links for quick reference back to the original data. The first few cases were used to pilot the thematic framework we had developed which was then amended (see KandyW earlier post in this forum for more detail on developing a thematic  framework). Once happy with the structure all the cases were summarised into the matrix.  No coding was done prior to this, as the process of deciding where data belonged in the thematic framework was part of the summarisation process.

 

A particular strength of the matrix output is the ability to quickly compare across cases, while also retaining a sense of the whole case. By displaying the patient data alongside the data from the staff, it was straightforward to compare views and experiences from these different perspectives. We also used a node classification to facilitate data organisation. Attributes were created to differentiate between each case study area, to differentiate between staff and patients, and to identify which treatments patients received. This allowed us greater flexibility when viewing data in the matrices as it was quick and easy to filter or order the data according to these attributes. While summarisation does involve an investment of time, by the end of this process, a complex large data set had become a managed set of framework matrices which can  then be used to facilitate analysis.

 

I’d be really interested in hearing how others have used the framework matrices features, and any hints / tips other users may have to make the most of these features. 







Also tagged with one or more of these keywords: Framework, matrices, qualitative, data analysis, summaries

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users