Analysing your data
Once you’ve collected your data using the tools, you’ll need to analyse it. Here, we describe how to prepare qualitative data (from open-ended questions) for analysis and how to use two common analysis techniques for qualitative data. We also show you how to carry out a basic analysis of closed-response data.
Preparing Qualitative Data for Analysis
Before we analyse our data, we need to make sure there are no obvious errors. These could include repeated entries, typos, misspellings or similar. The following instructions use samples from a hypothetical dataset of teacher feedback (collected using Tool 9 – 3 Words) on an imaginary website, Europlanet for Schools .
For example, the raw data from our teacher feedback might look like Table 1.
What three words would you use to describe the Europlanet for Schools website?
|useful||easy to navigate||extensive|
|excellent||informative||can be frustrating to search|
|easy to use||Practical||hard to search|
|useful||Informative||easy to use|
We can see that there are a few minor issues with the data, for example:
- There are some small spelling mistakes e.g. relevent / relevant
- Some words are capitalised and some not e.g. Useful / useful
- Linked but slightly different words have been used e.g. challenge/ challenging (these are known as “stem words” in language analysis)
- Some people have entered a short phrase instead of a single word in some entries (e.g. “easy to use”). When creating word clouds, the software thinks of this as three separate words (“easy”, “to” and “use”) so we need to force it to recognise those terms as being connected.
- Some phrases / words have the same meaning but have used different words e.g. navigable / easy to navigate / good layout / navigatable
These differences will artificially reduce the frequency of such terms by splitting them over separate entries.
Step 1: Remove Duplicates
The easiest way to remove duplicates caused by capitalisation is to copy & paste the text into MSWord (keeping it in the table format) and then use the ‘case’ button to toggle all the text to ‘lowercase’.
You can then paste the table back into Excel.
Step 2: Connect Phrases
Highlight the text, then select Edit > Find > Replace and replace all spaces (“ “) with a tilde (“~”) sign (alt-n on a Mac) – so “easy to use” becomes
“easy~to~use”. This tells the word cloud software that those terms are now linked, but displays them as spaces for easy reading in the final word cloud.
Step 3: Standardise Words and Phrases
To check our data for further errors it is easier to look at the whole list of words it contains. If we input data to the www.wordclouds.com site, there
is an option to select ‘word list’ – clicking on this brings up a list of all of the words contained within the data file, as well as their frequencies (how often they occurred within the text). These can then be arranged into alphabetical order (this helps us to spot issues with the data more easily). The word list can then be copied back to the spreadsheet and spelling mistakes can then be easily corrected and linked words and similar phrases standardised, as shown in Table 2. The “Edited Word” column notes all the places where the original word needed editing.
The “Updated List” shows the same data with all the edits applied – so, for example, 1 count of “accesible” and 1 of “accessible” have been combined
into a single entry “accessible” with a frequency of 2.
When all errors have been corrected, we are ready to start analysing the data with our analysis tools, Word Clouds and Thematic Coding.
To find out more about analysing closed questions, see Tool 11 – Post event surveys and Tool 7 – Pre / Post Quizzes.
Back to Evaluation Toolkit homepage