Data Analysis Software
Data analysis software is characteristically divided into:
- specialist applications designed for quantititative or qualitative data analysis (Datadesk, Clustan, Genstat, SPSS, SAS and others for quantitative; Nudist, NVivo, Atlas, Ethnograph and others for qualitative)
- 'generic' software which can be used for data analysis (Access, Excel, Word, Inspiration)
It may be the case that you have a larger and sophisticated data set which you wish to analyse; in this case, it may be worth learning to use one of the former. You should be aware, however, that this will involve time and effort; learning to use Nudist, for example, is a 'non-trivial' exercise.
Otherwise, you should be able to do most of the data analysis you need using MS Office tools - specifically, MS Word and MS Excel. The case for using these is that you may already be familiar with them, they are ubiquitous and they are the tools which you will be expected to use for administrative data handling (all those lovely SATS scores!) in schools.
A useful resource for those interested in qualitative data analysis is the electronic journal Forum: Qualitative Sozialforschung which in May 2002 published a "Special Edition on Using Technology in the Qualitative Research Process" (Vol 3(2)). This provides a number of practical examples of the use of both specialised and generic software for qualitative data analysis.
Using the Advanced Features of MS Word 1: Reviewing Tools
MS Word has a range of features which allow you to use it in support of ' iterative' data analysis in which you re-read, sort and attempt to develop 'categories' from transcripts, observation notes and interviews. The processes of 'marking up' texts, 'memoing', and sorting data into emerging categoris can be replicated using Word's 'reviewing' tools. You can make these appear by selecting View -> Toolbars -> Reviewing from the Word Menu. Two of these tools - the 'post-it notes' and the 'highlighters' are shown in use below:
Figure 1: MS Word's Reviewing Tools in Action
A nice feature allows you to view all the comments you have appended to a document - rather like footnotes - in addition to seeing them as 'pop-up' boxes as shown in the illustration above. This doesn't work for highlighters, however.
Using the Advanced Features of MS Word 2: Advanced Searching
The 'Find' function of Word (Edit -> Find) or 'CTRL + F' on the keyboard allows you to enter words or phrases which will then be highlighted. You can then skip through a document finding all the instances of the target by using the 'Find Next' button. This is useful if you are reviewing transcripts and want to find all the instances where a particular word was used or a particular person spoke:
Figure 2: A Search in MS Word
If you select the 'More' button, you will be given more search options. One which is very useful is the use of 'wildcards'. These allow you do run simple 'concordance' searches, so you can find, for example all the occurrences of a word or phrase spoken by a specific person, or of two words occurring in the same paragraph.
For example, if you are looking for occurrences of the words 'science' and 'computer' together a search with 'use wildcards' selected for 'science * computer' will be interpreted as 'look for 'science' followed by any number(the *) of characters followed by 'computer'.
Where this really scores is when you only want to search for specific things being said by specific people. For example: 'Interviewee: * science * computer' will allow you to find occurrences of the Interviewee using the selected word but will ignore the same things being said by the Interviewer. This depend on you formatting and labelling transcripts consistently, of course!
Figure 3: A More Complex Search with 'Wildcards' in MS Word
Using MS Excel for Planning and Recording Sequences of Events
Spreadsheets like Excel are excellent for planning the collection of structured data. You can use them for setting out questionnaires, preparing checklists and for observation schedules. You can save yourself time and effort by thinking through the entire process from data collection to analysis: you may be able to use the same basic framework for the different processes. Here is a simple example for collecting, storing and analysing structured interview data.
First, you can write up your questionaire schedule and save it as a 'document template' - you'll be needing it again later! Print out as many copies as have interviewees and then use them to keep track of which questions you asked and they answered. This will provide a useful way of cross-checking with audio-tapes. You may want to jot down notes on the sheet as well.
Figure 4: Setting out a Questionnaire Schedule in MS Excel
When you now come to transcribe the audio-tapes, you can use the template as the basis of a new spreadsheet on which to collate the interviewees' responses and any field notes you collected during the interview. Save this as a template as well - that way you can start a new one for each interview.
Figure 5: Questionnaire Data in MS Excel
The nice thing about spreadsheets is that you can add new columns as you wish, and as you analysis develops. So in addition to questions and answers, you can build up columns with emerging categories and supplementary questions.
Monitoring Trends in Data using Dynamic Charts
If you are collecting classroom data, you will sometimes want to keep an eye on developing trends rather than waiting until the end of the data collection. This is particularly the case if you are using a 'constant-comparative' approach to your research, or if you are undertaking 'action research' in which you may wish to adapt classroom strategies as the research develops.
A very useful feature of spreadsheets is their capability to present 'dynamic' data; in MS Windows this is also known as OLE or 'Object Linking and Embedding'. Here is a simple example. Let us assume that you are counting the number of undirected teacher questions answered by boys and girls - you might be using some kind of observation schedule. You set up a spreadsheet as shown below, and, even though you haven't collected all your data yet, plot a chart to show differences - but making sure the the 'chart area' which you select includes the blanks you intend to fill in later. Then you save the sheet and the chart.
Figure 6: Charting Spreadsheet Data (1)
After another visit to the classroom, you re-open your spreadsheet and enter the new data. Because of OLE, the chart will take its data directly from the sheet, including your new data, and will update 'automagically'.
Figure 7: Charting Spreadsheet Data (2)
It looks as though there might be an emerging trend there ... next lesson might be interesting!
Selecting and Presenting Data within your Dissertation or Essay
You are probably best off using MS Word as 'framework' within which you collect your data - photographs, sketches, scanned childrens' work, tables, charts etc can be 'embedded' within your document. Portions of the spreadsheets you may have used for interview data can be 'pasted' into word documents. By default, they will appear as 'tables', so you can then use the table converter (Table -> Convert ->Table to Text) to change them into nicely formatted paragraphs. Remember - never type anything more than once!
Figure 8: Embedding Spreadsheet Data in Word
You don't need to present all your data in the finished essay or dissertation. If there are large data sets it may be worth including them in an appendix, but in the context of most writing, carefully-chosen excerpts, summary tables and charts are probably worth more than page after page of transcripts and multi-coloured three-dimensional exploded pie charts.