It’s time to crunch the numbers and look through any qualitative data you’ve collected, noting any trends and interesting responses.

Prepare your team

You’ve kept your team informed during each of the steps, and now it’s time to bring them back to the table and get their input on the data that has been collected.

Data analysis can be performed by a single person, a pair of colleagues working together or an entire team. It is helpful if everyone involved in the data collection is also involved in the data review but external eyes are also very important. Remember, you are the expert in your work but that doesn’t mean you are the expert in data analysis. This stage always benefits from the input of others to help validate your results. 

Prepare your data

Now it’s time to review and prepare your data so that it’s ready for you and your team to analyse. Expect to spend some time looking in depth at what you have collected to ensure that it meets your own quality standards and can help you deliver your goals. 

• Is it complete?

• What is missing?

• Can what you have collected tell you enough about what you want to know?

• Is it in a good format to enable collective review with your team?  

Data cleaning

Data cleaning involves the detection and removal (or correction) of errors and inconsistencies in a data set or database due to data corruption or inaccurate entry. Better Evaluation.

You’ll need to clean your data before you start analysing it. This can take a long time, particularly if you’re new to data analysis, but it’s like making bread. The effort you put into making sure the early steps are right will be crucial to the final outcome. The cleaner and better your data, the more accurate and reliable your findings. The steps below are not the only steps you can take to clean your data, but they will be a good start.

• Make a back-up of the original data.

• Remove unneeded columns (e.g. unnecessary data that has been collected.

• Find and remove duplicate responses.

• Ensure that data with multiple entries in one cell are split into multiple columns. You can do this by using the text to column feature on Excel.

• Decide what you will do with unfinished responses, if these are collected. If you don’t want to include them, remove them!

• Find missing data by searching for empty cells. Where data is missing (e.g. where respondents haven’t filled in a response to a question), either fill the space (e.g. with ‘no response’ or ‘missing’) or remove the column (e.g. if there are many missing responses that mean that you can’t draw meaningful conclusions).

• If the column title has been automated, make sure you replace this with a meaningful name (e.g. not question 3!), something related to the question asked.

• Consider giving unique identifiers to every response so you can cross-check responses at a later date. This is very helpful with qualitative data, too, when you include this as quotes in your report. You can ensure that you are not quoting only one or a few perspectives, but really giving a range.

• Fix inconsistent data. You might have given respondents a free text option to input something like their role, their location, etc. You can standardise this manually or automatically (e.g. replace all).

• Remove trailing or white spaces, that is, spaces at the end of responses.

Choose your data analysis tools

To crunch the numbers, you can use a free or standard tool like Excel or Google Sheets. If you have a background in social sciences or research, you may be familiar with other tools such as R, SPSS and STATA. 

To analyse qualitative data, you can use free or standard tools like Excel or Google. You can even print out your data, e.g. your written interviews, and code them manually with highlighters and coloured pens. Tools like Atlas.Ti can help you organise and code your qualitative data but these are often costly.

Analysing qualitative data

In this page we outline how to think about the data you collect in terms of patterns that help answer your research questions.

Analysing quantitative data

In this page we discuss how to analyse the numbers and draw conclusions about the validity of your sample.

A final step: write up your methodology

While it’s fresh in your head, write-up your methodology. Use your notes and document the decisions you took along the data analysis journey.

Next step

• Step 5. Interpret the data