Understand

Analysing data to summarise it and look for patterns, relationships and trends is an important part of evaluation. Data can broadly be grouped into two categories – quantitative (numbers) and qualitative (text, images) data. Depending on the resources that are available to you, you may have to be content with a more informal approach to data collection and analysis. This can still provide you with useful insights.
 
There are many options and ways to analyse your data. This section of the evaluation hub describes some of the techniques that you may use to analyse your data.

Quantitative data

Using statistical methods can help you understand, describe, summarise and/or compare numerical data. You can then use the findings to strengthen your intervention.
 
To begin, prepare your data

• enter it in a database such as Microsoft Excel or equivalent. Specialist statistical software is available, although this can be costly and difficult to use for those new to evaluation and data analysis.

• Clean your data: remove blank responses, duplicates or obvious errors

• Make sure all variables are in the correct number format

The next step is to select appropriate statistical tests

• These will help you to organise and understand data

• Decide how you will use the data to describe outcomes

Commonly used statistical testing methods include:

• Frequencies: how often something happens

• Percentages: are a good way to compare two different groups or time periods

• Ratios: show a numerical relationship between two groups

• Mean, median, mode

  • Mean: Average

  • Median: Value in the middle of the data

  • Mode: the value that occurs most frequently

When the data is analysed, you will need to present it. Data needs to be presented in a clear way to make it easy to understand. One way to do this is to summarise it in graphs and charts. The below table suggests how you can best present different types of data:
 

Always include your sample base; the number of respondents that answered a question or number of people in sample. Include any limitations, such as small samples.
 
What does the data tell you?

• ​Draw out some key findings. Consider:

  • Are there any themes or patterns?

  • Is there anything that is surprising?

  • Is your data showing an improvement in outcomes? Or not?

 
Next, you will need to review and check

• Draw out key findings from your data and write notes on these to help construct your evaluation report
• Ensure analysis can be verified and that you have the evidence to justify your claims

Present and write up your report

Further reading

How to analyse Quantitative data for Evaluation

Find out more

Analysing quantitative data for evaluation

Find out more

Analyse data

Find out more

Interactive chart chooser

Find out more

Qualitative data

Qualitative methods collect data that is non-numerical form. Analysis of this type of data can be time consuming so it is important not to be tempted to collect too much. It’s important to ensure what you are analysing is meaningful and useful. So, keep in mind what questions you want to answer. This will guide the interpretation of your data.
 
How you choose to analyse  will depend on how much information you need to analyse. Regardless, it is important to read through the information and review it so that you are familiar with the contents. There are two main ways to do this:
 

Code and count

Useful for large amounts of information and can be done manually or with computer software. If you code your qualitative information, you are transforming it into something     that can be analysed through quantitative methods. Your code is a label that will describe the content information.
 
An introductory video can be found below.
 

Thematic analysis

If you opt for this method, you will be looking for themes and patterns. The process is similar to coding, whereby you create codes for your information to be able to sort and organise them into themes.

Next, you will need to review and check your data:

• Draw out key findings from your data and write notes on these to help construct your evaluation report
• Ensure analysis can be verified and that you have the evidence to justify your claims

Present and write up your report

Further reading

Qualitative Coding

Find out more

Thematic analysis

Find out more

Qualitative data analysis

Find out more

Visualise data

A key way to understand and communicate your findings is to visualise your data.  There are a vast number of ways in which you can do this but think about:

• What would you like to show?

  • By answering this it will help you understand what visualisation output will be relevant to show it. Do you want to compare, show a change or reveal relationships? You may need to test different types of visualisations to see what one works best

• Organise

  • Are you including all relevant data? It is important that all relevant data is represented in the visualisation so that it isn’t misleading.

  • Is the visualisation clear and concise.

• Test

  • Is it understandable to someone else? Ask a colleague if they understand the meaning. You may need to experiment with different graphic forms to see what the best way is to present your message.

 
Charts, tables, graphs, maps and infographics are ways in which you can represent your data in visual form. Below is a thought starter about how you may use each type

Source: Click Here

Further reading

Visualise data

Find out more

Data visualisation

Find out more

What next?