Data analysis techniques are best chosen in relation to the types of data you have collected. Your approach may be based on a particular model, such as grounded theory, statistical analysis or narrative case study based a combination of quantitative and qualitative methods. Be aware of what each can and cannot do. As with data gathering, you should consider your own expertise and the time you have available for the task.
Quantitative data analysis
Quantitative data will rely on correlation and regression methods – ‘t’ tests, analysis of variance, chi square as statistical outputs. Software applications for statistical analysis may be helpful here. Windows SPSS is one of the oldest and most popular of the quantitative data analysis packages and is available on the Warwick network applications launcher (in the Applications group).
Useful guides can be found on SOSIG and Leung’s “Making Sense of Quantitative Data” (see Resources section for web links.) For help with statistics, a good starting point is the Statistics Glossary produced by Easton and McColl (1997) as part of a national teaching and learning project. The University also provides access to “Statistics for the Terrified” through your office desktop ‘network application launcher’ (in the Applications group).
Qualitative data analysis
Qualitative data may include transcripts from questionnaires, interviews or focus groups. Where statistical software manipulates numbers, qualitative software manipulates words. On a small scale, this can be reviewed manually, extracting key themes emerging and providing anecdotal examples and quotes to emphasise and personalise particular points.
The main software applications to assist data analysis are Atlas Ti or Nudist Nvivo; possibly concordance packages might be helpful. Both are available on the Warwick system (in the network application launcher Applications group). They still needs human input to go through the text, code it, and enter these codes into the computer before you can use the software to manipulate the data. A useful guide on choosing between the two is provided by Barry (1998) and a good review of using software for qualitative analysis is by Audience Dialogue (2004).
If you are using a fill-in survey form to obtain data, a number of scanning software programs can be used to read filled-in bubbles, checkboxes and barcodes (e.g. REMARK) or simply convert text (and images) on a paper-based form into electronic form (e.g. OmniPage Pro).
Blending qualitative and quantitative data
A difficult area for most lecturers (and developers) is how to use quantitative and qualitative data in synergy, such that they are mutually reinforcing. Quantifying open responses can sometimes help give credibility to your qualitative statements and informed conclusions. For instance, you might be able to quantify the proportion of respondents who felt a particular way (perhaps expressed in different ways, but making the same basic points.) The quantification might indicate the strength of a trend, while the qualitative remarks provides depth and more detailed insight. Including quotes in an evaluation report can be very illuminating for the reader.