Data Analysis

Data analysis is about turning the raw data from your instruments into information that is in a useable and easily understood format. Analysis is also sometimes defined as the process of resolving complex issues into their simplest and most manageable elements. Ultimately, analysis should provide the researcher with the direction to make effective decisions.

As a researcher, would you prefer to know that respondent one answered yes to question 1, respondent 2 answered no, respondent 3 answered no etc, or would you prefer to know that 47% answered yes and 53% answered no.  Analysis makes your collected data more understandable.

The purpose of data analysis is to produce information that will assist decision makers. In deciding on how the data will be analysed, the researcher must consider what information is required and what information will be most meaningful and useful in the decision process.

In selecting a data analysis strategy, such things as research design, measurement scales and other characteristics of the data have to be taken into consideration. Different statistical techniques are appropriate for different types of data analysis.

There are two basic forms of data analysis

Descriptive
Descriptive data analysis does exactly what it says, it describes a phenomena or a set of data.  Descriptive data analysis can be in the form of a summary (65% said Yes, we can!) or in graphical form (such as a table or chart).   It can also take a series of observations and make a statement as to what those observations mean.  In the simplest form, 65% of respondents answered ‘Yes, we can!’ which means the majority of the people survey were in favour.

Inferential
This is where you interpret the data to make an informed assumption or interpretation of the facts.  So, in the example above, we could describe the data by saying 65% answered ‘yes, we can!’ An inference from this is that there is evidence to suggest that the respondents were influenced by Barack Obama in the way the viewed a certain subject.  The inference is because whilst we don’t know it for a fact, there is a reasonable assumption to be made that people who answered ‘yes, we can!’ may be aware of the use of that phrase by the Prez of the US!

What can you use to interpret data?

Statistics
These are the most common instruments of data analysis.  Statistics can help you describe your research findings in terms of concepts such as mean (average).  Statistics can also help to make reliable decisions using your data.  There are types of statistics that aim to prove the connection between numbers of observations, the strength of a relationship between variables or to predict the future using information from the past.  Statistics can also be used to test the reliability or significance of what you find out.

Models
Many fields have theoretical models that you can use to categorise or describe your data.  For example, below is one from my research.  This model, described by McCarthy and Jinnett, was their attempt to represent how people choose to participate in the arts.  I could use this model to interpret my own data about how people choose to participate in the art forms I am researching about.  My data could prove or disprove this model.  My data could also identify, for example, that in terms of contemporary dance that the most important motivation for an audience to attend a [performance is the ‘reaction to the experience’ because 75% of all my respondents ranked that as the most important reason they attend.

Figure 1. (McCarthy & Jinnett, 2001, p. 24)

Content analysis
Content analysis analyses the nature and frequencies of communication.  It can take the form of word counts (how many time sin the conversation did the respondents say ‘ouch’ as a measure of pain) or be as complex as identifying the meanings behind words.

Qualitative data analysis
Most of what we have focused on above has been numerical (quantitative).  Qualitative data also needs to be analysed.  From the use of summaries (where you might join responses together into broad or specific categories) through to more scientific forms (which can involve highly complex computer programmes) the nature of qualitative data analysis is to make sense of the huge volume of data you have and present it in a way that supports or denies your argument.

John Siedel – Qualitative Data Analysis – ftp://ftp.qualisresearch.com/pub/qda.pdf

There are three basic phases of qualitative data analysis.  The collect phase is about your research instruments and how you use to them to gather your data, the thinking phase is quite reflective and urges you to look at what you have collected and think what it means.  Finally, the notice phase is about finding these connections and what they mean.

There some data analysis exercises on a wiki space here http://etherpad.com/QaTPi4gsi4.  Have a go at them and share your ideas and interpretations with your colleagues.  Just click through and start writing.  If you are interested in how a wiki works, have a watch of Lee LeFevre’s video ‘Wikis in plain English’
http://www.youtube.com/watch?v=-dnL00TdmLY