NVivo 2 Tutorials
NVivo 2 Tutorials
The following tutorials accompany each chapter of Handling Qualitative Data with
exercises in the previous version of QSR NVivo, revision 2. This version was replaced
by NVivo 7 in February 2006.
If you are working in NVivo 2, and at any stage wish to upgrade to NVivo 7, you can
import the project you have created in these tutorials.
Chapter summaries and software tutorials
Chapter 1 Setting Up
Chapter 2 Qualitative Data
Chapter 3 Data Records
Chapter 4 Up from the Data
Chapter 5 Coding
Chapter 6 Handling Ideas
Chapter 7 What are you aiming for?
Chapter 8 Searching the data
Chapter 9 Seeing a whole
Chapter 10 Telling it
Note that three of these tutorials have two stages provided this is so that you can open
either a project that is just beginning (as yours will be) or a later stage (to see how a
project develops). All these tutorials contain detailed instructions for using all the tools
in NVivo. To follow the instructions for any one of these tutorials will take about three
hours.
1. Select one tutorial called Stage 1. Click OK.
An Online Help file will appear, showing each of the steps in your tutorial, and giving
detailed instructions for you to go through its steps.
2. Do it! As you work through the tutorial, take notes of any processes you dont
understand or want to explore further. You can copy and paste from the tutorial
Online Help files as you create these notes. At the end of your tutorial session,
review what youve learned by clicking through the tutorial help file topics, which
show on the top of the help file screen. Check that you have understood the
following processes which you are about to conduct for your own project:
Creating a project
3. When you have reviewed your tutorial work, click Close Project to close the
tutorial.
You were working in tutorial mode, so changes were not saved. In the next stage, you
will be working with a real project yours so another task will be added - saving the
project you are about to create!
2. Choose Create a Project and let the Wizard take you through a typical project
setup.
3. Type in the name of your project and a brief description:
4. Click Next> then Finish to complete the creation of this new project.
5. The next screen displayed is the Project Pad of your own project. On the title bar
will be the name of the project you created.
Each Help file window will have further links to explanations of other windows or
functions. If you get used to moving around these files, youll be able to find your way
around the software rapidly.
2. In this case, to read more about the Project Pad, in the Help window, click on the
hyperlink from those words
3. Now click around the Project Pad on your screen to see how it gives access to the
places provided for storing your new project.
Always close your project before checking its contents via Windows Explorer.
Every project in NVivo has two main folders All Users and Database.nv1.
In the All Users folder you can store your Source Documents, photos or other material
you will link to (External Databites) and Results of your analysis. A folder called
Commands is provided for any automating processes.
Database.nv1 is the crucial folder whose files are in NVivo format. Files here
should NEVER be opened, renamed or moved elsewhere, since serious database corruption
may occur. (Thats what the file called Warning.txt says!)
The Project Pad must always be in focus when you are saving your project.
Chapter 1 begins with advice on setting up a project, and specifying its proposed
purpose, goal and outcome. There are no rules about the documents that will contain
these reflections and plans, but you are strongly advised to save them safely.
If you are working in NVivo you can write these early research design documents or
letters etc. in a word processor and import them to your new project. Or simply write
them within your project. And you can make proxy documents to represent other
material that you dont want to bring into the project at all but dont want to lose.
In NVivo 2.0, you will lose any embedded objects such as tables or photographs
from the file. Its just text that is being imported. So save separately any non-text items
in your file and use links to them to keep track of these. (See tutorial for Chapter 3.)
1. Create or locate a document youve typed up in the word processor. If you have no
document, write an informal one about what you are planning to do and call it
something like Research Brief. Save it in rich text (.rtf) format. Its a good idea to
save it in the Source Documents folder in your project so you know where you put
it!
2. In your NVivo project, on the Project Pad, click Documents, and choose Create a
Document. The first option is the default, Locate and import readable external
text file(s).
3. Click Next>.
4. Navigate to where your first RTF source file is located. Select the document, and
click Open. (If you have prepared more than one document, select them all to
import many documents at once.)
5. The Wizard offers options to Choose Name: leave the first option, Use the source
file name as document name and click Finish.
Your document is imported.
6. To see it, go to the Project Pad, then click on Documents and Explore Documents.
Find your first document, click on it an on the Toolbar click the top left Browse
icon. Your document will open and you can read the text.
7. If you wish, you can edit the text in the Document Browser. Keep writing, or
change colours and fonts etc just as though you were in a word processor.
1. In the Project Pad, from the Documents tab, choose Create a Document and select
the final option, Make a new blank document.
2. Click Next>
3. Enter a name for your new document and add a description. Click Finish.
A new blank document appears in the Document Browser window. Note that it
looks very much like a word processor window. You are in NVivos rich text editor
so use heading styles, italics, bold or colour if you wish.
As noted in Chapter 2, its very important to date the entries in your project
log. You can do this from the Edit menu (Insert Date & Time).
4. Now write! For this exercise, follow the advice on starting a Log Trail document,
pp. 43-44 in Handling Qualitative Data.
5. Close the document as you would in a word processor, from the menu or by clicking
the corner X. Changes will be automatically saved.
3. Recording external data
Now for data that is not easily brought into NVivo. As noted in Chapter 2, qualitative
data is rarely homogeneous. Much of your data will not be textual.
For the project you proposed, find some data that you do not want to bring into the
computer, or cant e.g. books, photos, videos, or a box of newspaper cuttings. In
NVivo, such data can be represented by proxy documents. If the file is on your
computer its called an external file because it will not be in your project. If its not on
your computer, its termed non-file data.
If you are stuck for some external data, make a proxy for Handling Qualitative Data!
Any project requires work with literature, and a proxy document is a very useful way to
reference and keep connections with books and articles. You can type into the proxy
your summary of a chapter, with occasional quotes, link it to relevant data in your
project and use coding to find the passages when they are needed.
1. From the Document tab, choose Create a Document and select Make a Proxy
Document for non-file data. Click Next>
2. Type in the location, name and details of the book (or photos etc) you wish to
represent. Click Next>
3. Choose Use numbered paragraph headers, and then choose which units you want
numbered (e.g. chapters or pages for a book, tape counts for a video). NVivo will
make a proxy formatted with a paragraph for each of these units, so you can type
your summary or comments in the relevant place. Enter the number of those units
(e.g. first chapter 1 and last chapter 10).
4. Click Finish.
5. Now to find your new document. Return to the Project Pad and click Explore
Documents. You will now have (at least) three documents showing.
6. Select your new document in the Document Explorer, and click Browse. You have
a template into which you can type the summaries of your external data. Each
chapter (or other unit that youve specified) has its own paragraph, so you can put
the summary or your comments at the appropriate place.
7. As before, just type as though you were in a word processor. In the tutorials to
Chapters 3 and 5, youll learn how to link and code this proxy.
Note that the Explorer, like the familiar Windows Explorer, will allow you to sort
on any of the columns in the right pane, so you can view your documents in order of size,
amount of coding, or date created or modified.
4. Backing it up!
NVivo 2 has a single file backup and restore command which makes backing up your
project much easier. This single file can then be stored on a CD, Zip Disk, Network
drive or emailed to a supervisor or colleague.
1. Save and Close your project.
2. Via the NVivo Launch Pad, choose from the menu bar, Project/Backup
Project
3. Select your project and click the Backup button.
A good trick is to edit the file name to include a date so you can easily find the latest
version of your project.
A message will be displayed when the project backup is complete.
You havent created a secure backup until this backup file (*.qbk file) is copied
elsewhere! A backup on your hard drive is a copy not a backup what happens if your hard
drive dies or your computer is stolen?
Use this method to restore a project from backup or if you are transporting a project
between work and home. If your supervisor or colleague has NVivo 2 and would like to
see your project, send them the *.qbk file created in the previous exercise as an email
attachment. They save the file and use restore to open your project.
1. Via the NVivo Launch Pad, choose from the menu bar, Project/Restore
Project
2. NVivo asks you to locate the backup file. It defaults to the C:\QSR
Projects\Backups folder. If your backup is on a CD, Zip Disk, Network drive,
etc, you need to navigate to where it is stored.
3. Locate the Backup.qbk file and choose Open. Notice that the Project folder
name is the same as the project name.
4. Select Open project once restored and click OK.
Youll receive a message when the project has been restored successfully.
You now have a project with data, and a good routine for keeping it safe! In the next
tutorial, youll learn how to store different sorts of information.
If you know the attributes your respondents or sites will have (that is, the information
youll want to store), you can create them all now, with their values. But there is no
need to they can be created as needed.
If a document represents a person (e.g. an interview with Mary), you can record
information about that person by giving the appropriate values of attributes to the
document. If cases are represented by many parts of documents, this is handled
differently all those parts will be coded at a node and the attributes given to that node.
For this tutorial, we will store document attributes.
2. The first option is selected: Create a new Attribute. Type the name of an attribute
that will be relevant to your data e.g. Gender. Click Apply.
3. Select the new attribute (Gender) in the left-hand pane and click the Value tab.
Note, your attribute has values already! NVivo provides three ways of saying there
is no information about this document Unassigned, Unknown and Not Applicable.
Create a new Value is selected: so type in a value you know your attribute will
have (e.g. Female) and Click Apply, then continue for other values. Note that there
is a default value Unassigned. You can change that if you like e.g. if all the first
respondents will be women, make the default Female.
4. Close the Create Attribute dialogue. And back at the Project Pad, choose Explore
Document Attributes. This time, the table with your document names has at least
one attribute as a column. Unless you changed the default value, the cells will have
dots in them if you mouse over the cell youll find that means Unassigned.
5. Right-click on one of the cells and the values you entered will show, as in a
spreadsheet. Choose the value that applies to this document. If you so far have
imported no files for which it would be appropriate to record attribute information,
use the N/A value. (e.g., your Log Trail Document has no gender.)
6. When you find a document with the value of an attribute not yet created, simply
right-click in the cell, choose New Value and name the value needed. Note this new
value now appears each time you right click in a cell.
Note how often attributes appear in the menus and the attribute icon
appears
in the toolbars. You can get to the attributes of documents from most windows for rapid
checking (was this a woman speaking?) You can also check attributes for individual
documents via the document browser rather than going back to the Attribute Explorer.
This can be useful if you are coding a document and need to check information about this
group or participant. And attributes are always available in the search processes that will
be explored in the tutorial to Chapter 8.
Type a small table in Word or Excel, with the word Documents in the first cell,
and the names of your documents down the side.
Make sure that the document names are the same as the names of the documents
in your NVivo project that you want to give these attributes to.
2. Type names of attributes as the headings of the columns.
3. Type into this table the names of the values for each document, at each attribute. If
Gender
Age
Income
Education
Log Trail
Literature review
Interview 1
Female
Under 20
None
High school
Interview 2
Male
20-30
Middle
tertiary
Gender
Education
Log Trail
Literature review
Interview 1
Female
school
Interview 2
Male
tertiary
Age
Income
Under 20
None
20-30
Middle
High
5. Back at the NVivo Project Pad, go to the top menu bar, and choose
Documents>Import Document Attributes
6. Click the Choose button and navigate to find the file you saved. Read carefully the
window on your screen unless you change any options, NVivo will create any
documents, attributes or values that your project doesnt already contain.
7. Click OK and the import process will commence. A message will be displayed to
advise that the import was successful. Click OK. Now to check what youve done,
click on the Project Pad and select Explore Document Attributes.
8. Click around the table. To assign values, or create new ones, right mouse click.
Note that as discussed above, if the cases for which you have attribute
information are not represented by a single document, you have the option to import
attributes for nodes.
3. Using Sets
This tutorial finishes with the simplest way of managing your data records in NVivo,
Sets. Whilst these are very easily made and used, they are also ways of doing the most
subtle and powerful searching processes.
Sets can be created and changed in your project either in the Explorer or in the Set
Editor. Here we simply use the Document Explorer. If you want to find out more about
the ways you can work with sets, go to the chapter on Sets in the online Help. The Set
Editor, and the way it can scope searches, are covered in the tutorial to Chapter 8.
Sets icon in the left-hand pane. Right-click and choose Create Set.
4. Click in the name (currently Untitled set) and rename the set, e.g. Interviews.
(or go to Properties and rename it and add a description).
5. Click on the All Documents folder and drag the documents that are interviews
onto that set. Notice, they are still in All Documents! When you click on the set
you will see that it contains icons indicating short cuts to the interview
documents.
Because shortcuts, or pointers to the documents go into your sets, you can put a
document in any number of different sets. Think of other ways you could group your
documents. For example, Interview 1 might also belong in the set of documents from last
year.
6. Think of useful groups for your data records, and make sets for them whether
or not the records are yet in your project.
For most qualitative projects, a useful set is of documents that will contribute to
your log trail, your growing account of the project and how it is conducted. Visit the final
chapter to read about the task of doing a stock take of your writings about the project.
This will be much easier if theres a pointer to each relevant document in a set called Log
trail documents.
You now have a project with data documents and information about their context, and
have started managing those records in sets. Time to back up. Follow the instructions in
the previous tutorial to store a safe copy of your project.
In the next tutorial, youll turn to skills for discovery and handling of ideas that come
from the data.
Annotate it.
The document opens in a Document Browser. This window allows you to edit,
annotate, create memos and (see next chapters tutorial) code.
3. Annotating
3. Select a bit of text that requires a comment. In NVivo 2, you add bites of data
to your text via links called DataBites. Click the Make/Inspect DataBite icon
on the toolbar
4. Select the type of DataBite to create note that it could be an annotation, a link
to a file outside your project (such as a photograph) or to a web page.
5. To annotate, select Internal Annotation. Click Create Type your reminder
or comment and click Save and Close.
Back at the text you selected, youll notice it is now marked as the DataBite
anchor, coloured green and underlined.
6. Right-click inside this anchorand choose Inspect DataBite to view your
annotation.
4. Writing a memo
Memos are full status documents that NVivo sees differently from other documents only
because you label them as memos.
As described in Chapter 4, you can use memos to capture thoughts and insights you
have when you are working through your data. The computerized equivalent of a
notebook, they are great tools for noting down possible themes and concepts that you
want to go back and explore later on. But unlike the notebook, they can be linked
directly to the text that gives you an idea, or the place you store it.
Memos are editable rich text documents in NVivo and the simplest way to created them
is to ask to link to a memo from any other document. You can also create memos
simply by making a new blank document (see Chapter 2 tutorial) and in its Properties,
clicking the box that says it is a memo.
It is very easy to make a memo from the text of a document.
1. Place your cursor at a point where you would like to link your memo. Choose
Make/Inspect Doclink icon on the toolbar.
the
The In-Text DocLinks dialogue is displayed note that it will let you link to any
existing documents, or make a new memo.
2. Choose Link a New Memo and you have a new blank document that you can
edit as you wish. (this is the quickest way of making a document in NVivo!).
Write what you wish to record and go to the Edit menu to select Insert Date
and Time, to date the current entry. If you do this every time you add to the
memo, it will provide a good contribution to your log trail. When you have
finished writing, close the memo and you are returned to the DocLinks window.
Click Close.
3. Back at the document, youll notice you now have an icon
DocLink. Right-click on this icon to browse the memo
that represents a
4. Explore Documents and youll notice that your new memo document has been
named by the document it was created out of e.g. Interview 1 - Memo. It is
listed with all other documents but its icon has a notebook appearance.
5. Select the memo in the Explorer, then click the Properties button in the toolbar,
to rename the document to represent the theme, e.g Interviewing as a method
and perhaps change the colour of the memo icon. Now you can return to it at
any stage to add more ideas, or link to it from any other place in a document.
Now the data records are becoming richer, with your own ideas and the links to
related material.
This tutorial shows techniques for doing the more qualitative coding of the data by
topic, or to identify new categories and ideas in the data. In it you will learn to:
and most significantly, because this was the reason for doing all that work:
To review what you need to know about nodes and coding, go to Introducing NVivo, pp
6-7 and 12-13.
1. Focusing on a document
To start this tutorial, as with the previous one, you need to concentrate on one
document in your project.
It may be the document you worked with in the previous tutorial its interesting (but
not necessary) to combine the writing and linking techniques there with those of coding.
If you want to move to a different document, read it through before you begin coding on
computer Now, go back to the start of that document.
1. Making a node
Chapter 5 explains the uses of coding in qualitative research, and the many ways
researchers create categories, down from their research questions and designs or up
from the data. In NVivo, those categories and the coding you do are stored at nodes.
7. In the Project Pad, click Nodes and then Explore Nodes. The Node Explorer
opens, offering different areas for nodes free of organization or in catalogs of
trees. The next tutorial covers the managing of nodes in trees: here we will
code just at free nodes.
8. On paper, or on a white board, sketch a beginning list of the categories your
project is about topics or ideas about which you wish to collect data. (For
example, in this mythical project about the qualitative research class, an issue
under enquiry might be students fear of the flexible nature of the project too
few rules!)
9. Click on Free Nodes, right mouse click and select Create Free Node.
10. To rename the new node, as above, click on it (its called Free Node), and
click Properties on the Toolbar. The node properties box opens and you can
type the name of your node.
11. Continue creating and naming nodes for any topics or concepts your project will
require.
Note that the Properties box always offers the opportunity to describe the node
a good place to store hunches or first thoughts about the category. At any stage you can
print off a list of all your nodes with their descriptions to assess or discuss (just go to
Nodes on the top of the Project Pad and select List all Nodes and ask for descriptions
also.)
Note that the node is now bold this shows it has coding. It is a useful habit to
scan the nodes you create, watching for the ones that are not bolded, and wondering why
they dont have coding.
You may wish to continue reading and looking for data about the categories for which
you already created nodes. But those prior categories will probably not suffice for what
you are now seeing in the data. You need to create nodes up from the text.
3. Creating new nodes up from the data
As you read on, you may see a category emerge. You can first create the new node in
the Coder (just right mouse on Free Nodes and create a new node as before in the
Explorer.) But it is quicker to use the coding bar on the bottom of the Browser.
1. Highlight some text you wish to code. Type the name for this node the
category or theme in the slot on the coding bar (to the left of the Code button).
Click Code. The node is created and the coding done.
2. Highlight a word in the text with which you would like to name a node. This is a
technique called in vivo coding. Its useful when the data provide specially
suggestive or meaningful concepts for example, here the student mentioned
needing recipes for research processes. Click the In-Vivo button on the left
of the coding bar. Again, the node is created and the coding done.
3. Continue, creating nodes as they are indicated in the text, for either topics you
wish to collect data on, or concepts you wish to explore. For each new category
you create, write a description in its Properties box.
4. Automating coding of sections
The tutorial for chapter 3 covered how to automate descriptive coding by importing
attributes in tables. Coding of text is a more subtle process, since it involves creating the
node, selecting the text to be coded there and doing the coding. But there are two ways
of automating coding of text.
Both are ways of sending the software to do a mechanical process, using either
your formatting or the words. Check the warnings in Chapter 5 about interpreting
mechanical coding! These methods may not be appropriate to the data you have so far in
your project.
In this section, we deal with section coding (go to the next section for text search).
You may have used headings and subheadings in your documents, (for example, in the
interviews in screen shots for these tutorials, questions being asked are always typed in
Word as a Heading 1.) NVivo recognizes these headings. The text between headings
can be coded automatically using section coding.
1. You need to make a node below which all the question nodes can be stored. This
will go in the Trees area of the Node Explorer. Click on Trees, and right
mouse to create a node.
2. Select the new Tree Node, go to its Properties and name it Questions.
3. In the Document Explorer, select the document with sections to be coded (or
All Documents, if they all have sections that are the same), right mouse and
select Code by Section.
4. The Section Coder opens, asking for the node that will be parent to your
section nodes; click Choose, select your Questions node and click OK.
The section coding is (probably!) reported as successful. Note, this does not mean
that the results are as you expected youll need to check that! Go to the Node Explorer
and click on the + at the Questions node you should now have a node for each section,
named by its header, and coding all the text under that header until the next header.
2. Type in a word to search for (one that will be found in your data.)
3. Ignore the advanced options, for now. Note that the finds will be saved as a
node. Click OK. Check the two lower panels on the Search Tool. The Scope of
the search is All Documents. (Leave it that way.) And you have the option to
Spread Finds. Change this to Enclosing Paragraphs
4. Now click Run Search. The search is completed and a window tells you where
the results are stored in a new node (called Single Text Lookup), under Search
Results in the Trees area of your Node Explorer. Check the Node Explorer and
you will see the text search node, complete with a description. NVivo has coded
there all the paragraphs where the text you specified was found.
As you get used to the Search Tool, you may do many such searches. A very
useful habit is to rename the nodes created to code the results, so you can recognize
them.
4. Select what context youd like to code to make sense of it. Your node is waiting
in the coding bar just click Code.
5. Remembering the emphasis on keeping context of qualitative data, you may
want to ask, where did this passage come from? To go to the document from
which its coded, right mouse click on a passage and select Browse/Edit/Code
Document this will jump you to the document at that passage and you can
keep working there.
Maybe you want to code some more context from the document, or place a memo?
Note if you annotate or place a link in the coded text, when you return to the node, those
links will show live there.
6. Now perhaps a new idea occurs this text suggests a different category? You
can code just as you did in the Document Browser: click Coder, make nodes
and do coding. You are now coding on from the first coding, a very important
process in refining qualitative coding.
Continue, reading and coding your data, browsing your nodes, expanding or revisiting
context of coded passages, coding-on to make new categories.
This concludes the tutorial for Chapter 5 of Handling Qualitative Data. You now have
data, ideas about your data and have started to relate them by coding data at those
ideas.
Dont forget to back up your project! The next task is to manage those growing ideas,
and this will change your project dramatically.
Are there more categories needed to ask the questions or explore the issues you
are tackling? If so, create the nodes.
3. Do they belong together in any logical order? Dont force it! On the paper or
whiteboard, sketch the shape of a starter catalog that looks logically sensible. If a
category doesnt belong with others, leave it free.
Looking top down, from your project design; what are the main areas of
enquiry? Revisit the categories you started the project with (Chapter 1).
Working with your data up categories, ask what they are a sort of, and if
some group logically, note this.
2. Rearranging nodes
At this stage, you can make major changes to your project. You did back it up at
the end of the last tutorial, didnt you?
You can cut or copy, and attach or merge, any node in NVivo, without losing any of the
data it contains. So if your nodes now could be catalogued, the steps are simple.
1. Back in your NVivo project, go to the Node Explorer and create nodes in the Trees
area for the main categories you have sorted your ideas into. These new tree nodes
will be parents for the subcategory children that belong there these are the
nodes that are a sort of the parent. Thus my node for fear of flexibility might go
with other categories about concerns about qualitative research). Create any
parent nodes indicated by this review.
2. Go to the Free Nodes you created by coding in the previous tutorial. Select one that
properly fits as a sort of one of your new parent nodes. Right mouse click and
Cut the node from the Free Nodes area.
3. Go to the parent tree node you want to catalog the cut node under. Right mouse and
select to Paste the node. Click the + and you will see your newly rearranged tree.
If you prefer caution, Copy, and then when youre sure its pasted correctly,
Delete the free node. You alternatively can drag the from Free Nodes to the new position
and this will copy.)
4. Continue cutting and pasting nodes as indicated by your review, until you are
confident the arrangement is logical and that it does not in any way force your
thinking about the categories or pre-empt things you might later find. (Chapter 6
advises on use of catalogs avoiding these traps.) If a node doesnt logically go in a
tree, leave it free.
5. Now code another document, working with the categories as they are now arranged,
so you can find existing ones and place new ones. You will find that as you work
with the nodes, now, it is very easy to cut or copy and paste or merge as the system
of nodes starts making sense. Now do these moves in the Coder (exactly the same
mouse-clicks as in the Explorer).
6. Visit the categories you have left free of the catalogue and review them.
A good habit is to review nodes left free often as the catalog forms, the
remaining Free Nodes are special. They are the categories that dont fit anywhere. Maybe
they are just irrelevant? But maybe they are the still unconnected ideas that will become
crucial in making sense of the more organized thinking about your project.
2. Listing nodes
1. Now to get a report on the node catalog you have created! Return to the Project
Pad, and from the Nodes menu, select List all Nodes (you can select to have
descriptions reported for each.
2. Read the report, and if it is useful, from the File menu select to save it. The default
location for the rich text file that is saved will be the Reports folder in your project
folder, eg. C:\QSR Projects\First Try at NVivo\All Users\Results
The catalog of your nodes that you create now will change, as your understanding
of whats going on in these data records grows. So the node system at any time will be an
excellent record of where you are at in the project. A highly useful technique for logging
progress is to save and archive a list of your nodes at regular intervals, or at significant
milestones in the project.
3. Make a new blank document and write a memo about this first stage of catalog
construction.
4. If you are working in a team, discuss and design a common catalog of categories
and discuss how you will use it to ensure consistency but allow innovation.
This concludes the tutorial for Chapter 6 of Handling Qualitative Data. You now have
not only data and, ideas about your data, but also an evolving management system for
those ideas. Time to back up your project!
As you continue relating data and ideas via links and coding, the ideas will be
accessible for review and reflection.
If the reorganization of your nodes is satisfactory, you will now be able to see the
logical groups of concepts and categories that are emerging in the project via the
catalogs of nodes. But the catalog does not show the theories you are growing about
how these categories relate. The next task is to represent the project visually in a model.
As you start working with models, carefully distinguish between the purposes of
node cataloging and modeling. A catalog shows how, logically, your categories are related.
In a model you show how, in the project data, or your hunches and predications about that
data, or the theory you are creating, those categories may relate.
1. Making a model
clicking on an item, then an arrow, then the item you want to link to.
2. Layering a model
You can make a model that is accessible layer by layer, to tell the story of your
interpretations of the data, or to compare different views. To do this, you simply
nominate which of the items in the model belong to which layer.
1. Go to Tools and Layers, then click New.
2. Type in a title and description for your new layer. Click OK.
3. Now from the View menu select Layer Palette. Select the First thoughts layer.
Go ahead to add the items and the arrows to model the first thoughts you are
taking into the project. If you click to untick that layer in the palette, the items
disappear.
4. If appropriate, make a second layer and in it place the story or map of your
early research. Perhaps the first interview changed your views? Note that if you
untick all layers in the Show Layers palette, then in turn tick each layer, you
can now show the evolving story.
Layering models in this way provides a very impressive presentation format for
qualitative research. Each stage can be shown and discussed, and at any point, since the
items are live to the data, you can show your audience the detail of the data. As an
exercise, present the modeled story of your project so far to colleagues.
This concludes the tutorial for Chapter 7 of Handling Qualitative Data. You now have
data, linked to other material and to ideas via coding. You have organized your ideas
logically in catalogs of nodes and modeled the beginnings of notions about how they
might relate in the data.
Dont forget to back up your project! At any stage you can of course delete or alter the
models, but when models represent the story of the project its highly advisable to leave
them as an honest record of where you were. They will become part of your log trail.
The next task is to use the ability to search your documents and your coding in order to
develop those growing ideas.
The result is a node, that codes all the finds in that selected spread.
To review what you need to know about searching text or coding, go to Introducing
NVivo, pp. 20-21. To review what you need to know about Assay, go to Introducing
NVivo, pp. 16-17.
1. Using the Search Tool
As your project progresses, you will return to the Search Tool for very many different
purposes. This is where you find occurrences of words, check who is using them, look
for patterns in your coding, test hunches and seek new ideas.
For now, identify an area of your project where you could pursue a hunch or question.
1. State what you are asking in ordinary language, (see pp. 152-153) and then plan
the searches that would ask it.
2. On the Project Pad, click
. Click around the buttons and press F1
for Help to obtain an explanation of what each search will do. Use the Help
Contents to locate the different descriptions under Searching/Doing It.
Read the description of the Boolean Searches.
3. Now in the Search Tool, click Boolean, and from the list at Operator, select
Intersection. This is the and search it will find for example:
any place where a text string is found and that text is coded at a node, and
the attribute gender = female.
4. Think of such a question that could be usefully asked in your data. Click on the
buttons to add to the dialog the text, node(s) or attribute value(s) you want
included in your and search. (The selections below will find anywhere text has
been coded at the node for recipes if it comes from a document with the
attribute gender = female.)
5. Now check the Scope for this search. (Leave it as All Documents.
6. Click Run Search. As for the text search you conducted in the tutorial for
Chapter 5, the results are saved as a node, stored at a Search Results node in the
Node Explorer.
7. Having the results as a node means that you can work with them like any coded
data, returning to the context of the coded segments. Browse the node where
your results are stored to find what was coded at that node, in documents given
that value of gender. Work with them to interpret the finds, perhaps write
about them in a memo.
Check the warnings in Chapter 8 about interpreting these very powerful searches.
Note that you cant claim you have found everything the women think about having
recipes for research processes. The node codes only what the documents that have been
given that value of gender, record was said in those interviews; and then only what was
coded (by you) at that node. If the search didnt find something it should have, its
because you didnt code or attribute fully.
1. Return to the Search Tool, and in the Scope panel, this time select a different set
of documents (e.g. your set of Interviews). Note you can also set the scope to a
set of nodes (e.g. to all the data coded at any nodes about learning problems, if
you had made a set of those nodes.)
2. Run the search again and check the difference. Note that the search results node
always has a description recording what scope was searched.
2. Select the items youd like to know about, and below the folders will appear the
proportion of your scope set to which this item applies.
3. Select some characteristics to make a table showing which documents in your
scope have these characteristics (e.g. which have the value of an attribute or are
coded at a node).
Assay is available in the Document and Node Explorers as well as from the Search
Tool so at any stage you can ask to see the characteristics of your data. Tables are a
very useful way of checking how your coding is progressing, for example, or which nodes
are being overlooked.
Of course you must back up your project, but this time, before you do so, tidy it
up. You probably have many search results nodes from playing with the Search Tool. Write
in a memo about what you learned and delete these records of searches you dont need to
keep. If any search results nodes are to be kept, a good habit is to rename them sensibly,
and move them (using cut and paste, as in the Chapter 6 tutorial).
This concludes the tutorial for Chapter 8 of Handling Qualitative Data. You now have
data, linked to other material and to ideas via coding, and know how to make catalogs
of nodes and models to show logical and theoretical relations, and how to search the
records and their coding.
The next task is to use the ability to search your documents and your coding for
different purposes, to show and interrogate patterns.
demographic characteristic, for example, or how one group of nodes pattern onto
another, as represented by your coding.
In the following example, the matrix created will show how the coding at selected nodes
is patterned by the education level recorded for the interviewee.
Start this tutorial by asking which of the seeing processes you are exploring in your
project could be assisted by matrix display. As before, write the question in plain
language first.
My question here, (in this highly inadequate early stage in my project!) might be: does
having done qualitative research in previous courses show any association with peoples
ideas of problems in using qualitative methods? Ive an attribute for previous courses,
and values allocated (return to the tutorial for Chapter 3 if you need to revise
attributes!):
2. Making a matrix
1. Open your project and the Search Tool. Click the Boolean button.
2. Choose the operator Matrix Intersection. Clear any Items that may be
displayed from a previous search so as the right-hand pane is empty.
3. Click on Choose Nodes to find text referenced by each of the nodes you want to
make rows in your matrix. Select the nodes you require and for each click
Add. (To select all the children of a node, click on the parent node and click
Children.) Check that the right box now shows the nodes you wish to have as
rows. Click OK.
4. Back at the Boolean Search dialogue, click the Choose Attribute Values button
in the second pane. Click on Document Attributes and in the left-hand pane,
select an attribute and move the values wanted to the right-hand pane by clicking
the
button.
5. Click OK.
6. Notice that the search will Return a set of matrix nodes and that the name of the
node will be /Search Results/Matrix Intersection.
7. Click OK to return to the Search Tool dialogue. There, click Run Search.
8. In the Search Completed box, click OK.
The table will display, with cells showing the number of documents coded at each. You
can change this to number of characters or number of coded passages. Examine the
table and ask whether the display surprises at all. And be very skeptical of what it
shows, especially if your study is still as small as the one illustrated here!
In your class report or project write-up, you may wish to include this table. When you
have the table version that you wish to export on the screen, select File>Export. This
will export a file that will be saved in the Results area of your project and can be opened
in Word or Excel to show the table.
You can very quickly build up your node system by running matrix searches. Each
cell in a matrix is a node. Ensure that you do some node housekeeping after running these
searches.
This concludes the tutorial for Chapter 9 of Handling Qualitative Data. The final
tutorial covers other aspects of reporting, ways of getting the results of your analysis
out of NVivo and into the report.
2. Select the report you want, view it on the screen and if you wish, when you close
the window, save it as a file.
3. Go to the folder in your project for Results and open the file in Word.
In Chapter 3, where Sets are described, one suggestion was to make a set of all
the documents that will contribute to telling the project, with a pointer to each relevant
document. If you dont yet have a set called Log trail documents, you may find it useful
to make one now. Put into it all the memos, diaries and reflective notes that will contribute
to your log trail. Now you can report on just those documents, or search all those
documents at any time, by setting the scope of the search tool to that set.
4. Now report on a particular document. Click on that document, and right mouse
If you like sitting under trees to think about your data, or wish to have the text
of annotations for your projects report, make a Text Report on the document and select
to show the DataLink details as endnotes: this will give all the text of the document and
also your annotations, all in .rtf format so you can copy into your Word report.
2. Reporting on nodes
Listing of nodes is covered in Chapter 6. But there are very many other ways of
reporting on these critically important containers for your ideas.
To report on nodes, the process is symmetrical with the document reporting process.
1. Right mouse click on any set, or the set of All Nodes, and view the options. You
can list the nodes (with descriptions, optionally) a good way to start. Or you
can get tables that profile information about them (for example dates of creation)
or coding patterns.
2. Now, just as for documents, you can report on any node the information about
it or the text coded at it. Right mouse click on any node and select from the
options:. You wanted the text coded at the node? Select Make Coding Report.
Even though the node browser does not allow editing, it will allow the fast keys
such as Control-C for copy.
2. Use the Search Tool to make a node that codes only the quotations you want
for example only those coded at one node and also another, or from cases with a
particular characteristic. Report on the search node and save paper and time!
3. At the Project Pad, from the Project menu, select List Project Information.
The report that results tells the bare bones facts about your project!
This concludes the tutorial for Chapter 10 of Handling Qualitative Data, and the series.
Go back to the introduction to this series for advice on further sources of information
about research processes using the software, and ways of working further with it.
Happy researching and dont forget to back up your project!!