Data Analysis



The purpose of analysis
•To identify patterns and themes
•
•To find support for / refute research questions
•
•To provide evidence for explanations
•

Forms of data
•
•
•Quantitative: numerical
•
•Qualitative: textual (language, visual)

Analysis of quantitative (numerical) data
•Descriptive statistics (often percentages, fractions or means) : the result of collecting,
classifying and summarising data: e.g. ‘75% of students in the study reported that that……’
•- Use of Excell spreadsheets in Microsoft Word etc
•
•Inferential statistics: the use mathematical formulae to make inferences about some aspect of the
research population by statistically testing the data from the sample.  For example, the use of
test to see how strong the difference between two data sets is.
•- Use of SPSS etc

Analysing quantitative data: a summary
•Code raw data according to the frameworks that you have designed for that purpose
•
•Reduce data to useful summaries
•
•Display data summaries meaningfully and accessibly: charts, graphs, bullet points.

Strengths and limitations of quantitative data
•Straightforward to code and analyse – is a mathematical exercise once coding frame is agreed
•
•Easy to make comparisons between numbers
•
•Can draw conclusions about large numbers – generalisability
•
•Can provide compelling statistic evidence for the support/rejection of a hypothesis
•
•Limitations: detail and individual voice is absent
•

Analysis of Qualitative (textual) data
•By hand
•Computer analysis (NU*Dist, Nvivo etc code and classify data using key word searches in the text)
•Top down: start with theory and look for evidence of it (hypothesis based coding)
•Bottom up: start with data and sift until themes and categories emerge (e.g. thematic analysis)
leading you to formulate a theory (e.g. grounded theory)
•
•

How to use qualitative data to provide a summary or sense  of findings
•Can be used illustratively: the pulling out dominant themes, the use of quotes
•Case studies can be presented as a ‘story’ part of which is the researcher’s commentary
•Data summary can have numerical format e.g. ‘one quarter or 25% of interviewees attributed their
success to the tutor’s skill’
•

Strengths and limitations of qualitative data
•Coding can be very complex process: in-depth detail can be overwhelming: how to select categories?
Assign data to categories?
•Difficult to compare non-numerical data
•Difficult to make generalisations although dominant themes can be explored
•Strengths: provides rich detail that can be used to ‘colour in’ quantitative data
•Absolutely essential not to reduce ethnographic work to numbers alone

Uses of all data
•Draw conclusions
•Weigh-up the case for/against research questions
•Able to make a response to/answer previous work/literature
•Should advance the field by providing digestible new knowledge that is supported by evidence
•Should add transparency not promote an obscuring of knowledge: know your audience