Data Analysis suggestions

Once you have collected all your data you have to do something with it.  This has to be considered right at the beginning of the research process so that you are collecting the right data in the correct format for the method you intend using for analysis. It is necessary to organise and structure the data in order to reveal ‘something’ you can then discuss in your findings.

  Analysis of quantitative data can be done using a variety of statistical methods.  The simplest of these is a frequency table which will show how responses are distributed amongst the categories in your questionnaire.  Even the most complex analysis is now achievable using one of the statistical computer programmes available.A correlation technique would be appropriate if the purpose of the research is to identify any patterns in relationships between variables, whilst a cluster analysis would enable you to classify data.

Description of your data can focus on the dispersion of the values or their variation from the central tendency.  It is important that you understand the difference between the mode, mean and median and that you can discuss range and standard deviation in your findings.

To examine in detail the relationship between two variables you need to use a statistical analysis technique such as Pearson’s Product Moment Correlation which evaluates the strength of the relationship.  Using Spearman or Chi-square you can test whether there is any significant relationship between the two variables.A regression will show the strength of the cause and effect relationship between variables, for example do your grades increase in proportion to the time you spend in the library and the number of books you take out on loan?

The way that you present your data analysis will depend upon the technique you have used, but a bar chart or pie chart is much more attractive than a table of figures and the reader can immediately identify the most frequent response and any relationships between different variables.

  Qualitative data is more difficult to analyse although the techniques applied to quantitative data can also be used.  If you have collected a number of student questionnaires you could undertake a content analysis of the open question section to see if they have aired similar/different views.  These will be in their own words but you will be able to identify links between, for example, positive responses and evaluate them accordingly.

The presentation of this data is based on classification and recognising relationships between them.  The units you are working with could be the sentences you have taken from the questionnaire which all relate to a given topic.  If you find that many sentences fit into a few categories then review them to see if there are any other words you could use to sub-divide these categories to make the results more meaningful.  For example if a high proportion of students respond positively to a question about the library facilities look to see what it is they give as evidence, the book stock, environment, subject knowledge of the librarians etc… and break the analysis into smaller categories.