Chapter 12 – Organizing data to known characteristics (QDA Miner)
Chapter 12 discusses the variety of ways organisation of data can happen and the importance of particular organizing tools to enable different levels and complexity of interrogation. Chapter 6 discussed basic structures like folders which enable simple tidying up and filtering. This chapter takes the subject further and focuses on the need to assign multiple variables or attributes to each respondent or case, so that comparing within or across cases can happen via combinations of data and subset characteristics if required. See all coloured illustrations (from the book) of software tasks and functions, numbered in chapter order.
Illustrating the potential for interrogation
Timing, when to put organisational structures in place
Organising whole documents
Organising parts of documents
Auto coding structures in documents
How to create variables – to assign e.g. socio-demographics to data
You may have already done several things earlier to organize data so some of these steps repeat what was covered then:
- As a natural process of setting up an empty project you may have created variables with the right ‘data type’ variables, e.g. Document, Image. In order to later house your data
- You will also have set up Cases and Groups/Descriptors which act a bit like folders
These are all aspects of organisation which will help you to interrogate or simply view the data required.
The ADD command in the VARIABLES menu is used to add new variables to the project file. This command displays a dialog box similar to the one used to create variables for a new data file.
The Variable Definitions dialog box will appear.
The first step involved in creating a new project file is to define the initial structure of the project file. This structure is defined by the list of variables that each case will contain. A variable may contain a document or an image to be manually coded, but can also consist of a numeric value, a date, a Boolean value (true or false), etc. A single project can contain up to 2035 variables per case. It is possible to create several document and image variables for each case. The ability to store many documents per case is especially useful when the project involves a set of several document types. A document variable may also be created to store notes or comments that are specific to a case. QDA Miner can later be instructed to search and analyze specific document or image variables or in all documents
In the Variable Definitions dialog box (see above), you can define various attributes for these new variables, such as their name and whether they will contain numeric, alphanumeric values, or dates.
VARIABLE NAME The first edit box at the top of the dialog box allows you to enter a variable name. Each variable name must be unique (within that project file). Valid variable names begin with a letter and may contain letters, numbers or underscore characters. Punctuation marks, blank spaces, accentuated and other special characters are not permitted. The maximum variable name length is ten characters.
DESCRIPTION The Description option is used to enter a variable label that describes in more detail the content of the variable. You may leave this column blank if you wish since it is always possible to add or edit a description later using the VARIABLES | PROPERTIES command.
DATA TYPE Each variable in the data file must have a type. QDA Miner supports the following types:
- Document This data type is used to store documents that will be manually coded. QDA Miner stores text in this data type using Rich Text Format (RTF). This format enables the use of different fonts and styles and paragraph formatting. Graphics and tables may also be inserted in the document. Numerous file formats may be directly imported into document variables, such as plain ASCII files (*.TXT), Rich Text files (*.RTF), MS Word documents (*.DOC), HTML (*.HTM or *.HTML) and WordPerfect documents (*.WPD)
- Image This data type is used to store graphics that will be manually coded. Numerous file formats may be directly imported into document variables, such as Windows bitmaps (*.BMP), Windows Meta Files (*.WMF), or Enhanced Meta Files (*.EMF), Compuserve Graphic Interface (*.GIF), Portable Network Graphics (*.PNG) or JPEG files (*.JPG or *.JPEG)
- Numeric Numeric variables can contain either integer or floating-point numbers. When you choose floating point numbers, you will be asked to specify the number of decimal places to display. Floating-point numbers are stored in the data file using double precision values (at least 15 significant digits). The option is used exclusively to control how numeric values are displayed in the Variables grid and in other locations and does not affect the internal precision of the variable
- Nominal/Ordinal Nominal or ordinal variables are used to hold a limited number of short strings used to describe specific properties of a case. For example, you may choose to use this variable type to identify the gender of the interviewee ("male" or "female") or its group membership ("manager", "employee", "client", etc.). You may also use this data type to hold ordinal ranking ("novice", "intermediate", or "expert") or responses to close-ended questions such as Likert scale items (e.g.: from "strongly disagree" to "strongly agree"). You should use this data type instead of short strings if you plan to analyse this variable or examine the relationship between its values and any other variable or coding of documents
When a nominal/ordinal data type is selected, you will be asked to first provide a list of values that this variable can take. To enter new values, click the EDIT button. A dialog box similar to the one below will appear:
You can start typing values (one value per line) in the large edit box. If the current variable uses the same values as another existing variable, you may also establish a link to this other variable so that they will share the same list of values. Click OK to confirm the setting of these values. Note: this list may later be modified to add new values or edit existing ones
- Date The date type holds a year, month and day. The display and data entry format used for dates is based on the Windows date setting
- Boolean The Boolean type stores a value that can be either true or false (or Yes or No)
- Short String Short string variables can contain up to 254 alphanumeric characters. When creating a short string variable, you must first specify the maximum number of characters this variable with hold
To create a new variable
- Enter a unique variable name
- Enter a description (optional)
- Select the data type for this variable
- Set the option associated with the chosen data type (see above)
- Click the Add button to add the defined variable to the list
To remove a variable from the list
- Select the row containing the variable you want to remove
- Click the Remove button
To change the position of a variable in the list
- Select the row containing the variable you want to move
- Click the up or down arrows located to the right of the grid until the variable appears in the desired location
- To append the variables to the data file
- When you have finished defining the structure of the new project, click the APPEND button to create these new variables and add them to the end of the current data set
The use of Variables and other organizing devices such as Cases, Group descriptors etc is all about the interrogation of data within and across those entities of data. Variables for instance can be selected as a filtering device in most forms of retrieval and querying process. See the next set of exercises associated with the discussion in Chapter 13 in the book.
Normand Peladeau 2014