DATA ANALYSIS

The aim of data analysis is to help turn raw data into knowledge, which can then be used for decision-
making and other purposes. Data analysis can take place at any stage of a project or programme cycle.
There are many different types of data analysis. These include quantitative, qualitative and participatory
analysis. Many projects and programmes use a combination of different types of analysis.

Raw monitoring and evaluation (M&E) data is not normally rigorously applied methods. The purpose of
useful on its own. If it is to be useful it first needs to be participatory analysis may also be quite different –
analysed. The aim of data analysis is to help turn raw data – encouraging stakeholders to analyse their own
facts and opinions developed through formal or informal situations rather than coming to a conclusion based on
planning, monitoring, evaluation or research processes – an external viewpoint.
into knowledge. In turn that knowledge can then be used
for decision-making, or to ensure accountability to different Another way of categorising data analysis is as follows.
stakeholders (Britton 1998).
 Descriptive data analysis is only concerned with
Data analysis can take place at any stage of a project or processing and summarising data. This is often true of
programme. It can happen before a project or programme financial or administrative data analysis.
begins as part of the design phase, It can also happen  Theory driven data analysis is used to test theories of
during a project or programme, at the end, or a while after change, assumptions or hypotheses. The aim is to
it has finished. Data analysis can be carried out at many analyse data to see if it confirms (or not) the theory or
different levels – within or across projects, programmes, hypothesis.
sectors of work and organisations. In social development,
data analysis is often encouraged within communities as  Data or narrative driven analysis involves letting
part of a participatory development process. patterns emerge from data, and then developing
theories afterwards.

Different types of analysis Qualitative data analysis can normally be applied to any of
the three types described above. However, quantitative
There are many ways of categorising data analysis – far data analysis is rarely used with data or narrative driven
more than can be described in this paper. One way is to analysis. This is because most quantitative data analysis
categorise it according to the type of data collected. (Note techniques involve collecting predicted information for
that many organisations, projects and programmes use a specified purposes.
combination of different types of data analysis).
Within M&E, there are also many data collection
 Quantitative data analysis is used to analyse numbers methodologies which have their own data analysis methods
rather than words. It can range from simple exercises to built into the process. For example, the Most Significant
process and tabulate data through to very complicated Change (MSC) technique includes defined processes for
processes designed to accurately measure quantitative selecting, analysing and using stories of change.
changes with calculated degrees of precision. Contribution analysis has distinct methods for testing
alternative theories of change. And quasi-experimental
 Qualitative data analysis, on the other hand, is used to
methods involve rigorous processes for assessing changes
analyse words – quotes, cases, transcripts, reports –
within target and control groups, and drawing conclusions
and, sometimes, images. Qualitative methods rely on
afterwards.
rules and processes which are very different from those
of quantitative methods.
 Some M&E methodologies are designed to translate Analysis questions
qualitative data into quantitative information through
rating or scaling exercises. This involves developing Separate papers in the M&E Universe series deal with
ratings or scales based on qualitative analysis, and then processes used for quantitative and qualitative data
processing them through quantitative methods. analysis, as many of the processes are quite different.
However, many of the questions designed to be addressed
 Participatory data analysis can involve quantitative or through data analysis are similar. Some of these are listed
qualitative data analysis, and is often treated as a in the table on the following page (based partly on Gosling
separate case. This is because participatory data and Edwards 2003.
analysis follows different rules, and is usually based on
stakeholders’ sensemaking and consensus rather than
© INTRAC 2017
 Common Analysis Questions
Process questions
• What work (activities undertaken or outputs delivered) has been carried out?
• What work was planned but not done? Why was this work not done?
• What problems have been encountered? How were these problems addressed (if at all)? If they were not
addressed, why not?
• Which activities appear to have been particularly successful or unsuccessful? Why?
• Are there constraints to progress which could be addressed? If so, how?
• Are there constraints which cannot be helped? If so, what can be done to minimise their effects?
Change questions
• What changes have been realised? Were these expected or not?
• How do they compare with what was hoped for or anticipated?
• How important or significant are they? Are they likely to be sustainable?
• How have changes affected different groups?
• What made the changes happen? Which other factors (other than your own project, programme or
organisation) influenced the changes?
• Are there expected changes that have not happened? If so, why have they not happened?
Action questions
• Is the project, programme or organisation still on track to deliver its objectives? If not, what needs to
change?
• Are they still the right objectives? If not, how do they need to change?
• Are the activities and outputs still appropriate? Should some be stopped, or others added?
• How has the external political or socio-economic situation changed? How should the project, programme
or organisation adapt as a result?
• What are others doing (or not doing) that might influence the project, programme or organisation? How
should it adapt as a result?
Learning questions
• What lessons have been learned from implementing the work? How can these lessons be applied to
future work?
• What needs to be done differently in the future, based on what has happened in the past?
• What lessons are there for other projects, programmes or organisations?
• What lessons might there be for policy-makers or other decision-makers?

M&E questions
• Are there questions about progress, change, actions or lessons that cannot be answered
through current M&E processes?
• What further evidence or information needs to be produced in order to make future
decisions?
• Are current indicators, methodologies and approaches appropriate? If not, how do they
need to change?
• Is there a need for new or further research, review or evaluation?

Enhancing capacity for data analysis To some extent, it is always possible to support people to
undertake better data analysis. But there are limits. Some
Data analysis is often the hardest part of M&E to do well. people are naturally better than others at data analysis.
Where it relies on following clearly defined rules and Some have more intuition; some are more experienced
processes, it may be relatively straightforward. But in other than others; and some are better able to handle complex
instances it can be very challenging. Data analysis often information. Capacity support for data analysis can help
relies on attributes such as experience, intuition, and an build on existing skills or abilities. But the ability to analyse
ability to handle complexity. This is especially true when complex data – whether used to help design projects or
interpreting findings in order to inform future plans. programmes, or assess changes and lessons with a view to
© INTRAC 2017
improving – may be partly down to natural ability or talent, This section of the M&E Universe contains advice and
rather than acquired skills. information on a range of different approaches, tools and
methods that can be used for data analysis. Ultimately,
Of course, the challenges of data analysis are very different however, data analysis always relies to some extent on
in different circumstances. For example, data analysis in a human interpretation, and is often subjective to at least
straightforward health project covering a single community some degree. It is not unusual for different stakeholders –
may be a relatively simple matter, largely relying on even skilled and experienced researchers or evaluators – to
mechanical processes. On the other hand, data analysis at examine the same data but come up with completely
the level of an international NGO working across many different analyses.
different sectors and countries is likely to be a much more
complicated affair, requiring considerably more
competency.

Further reading and resources

Further papers in this section of the M&E Universe deal with other topics related to analysis. These include quantitative analysis,
qualitative analysis, and the use of rating and scalar tools. There are also papers dealing with cost-benefit analysis, triangulation,
sensemaking and impact grids. These papers can be accessed by clicking on the links below. There is a further section of the
M&E Universe dealing with complex methodologies for data collection and analysis.

Quantitative analysis Qualitative analysis

Ratings and scales Cost-benefit analysis

Complex collection and analysis

Impact grids methodologies

Sensemaking Triangulation

References
 Britton, B (1998). The Learning NGO. Occasional Papers Series no. 17. INTRAC, July 1998.
 Gosling, L and Edwards, M (2003). Toolkits: A practical guide to assessment, monitoring, review and evaluation. Second
edition. Save the Children, UK.

Author(s): INTRAC is a not-for-profit organisation that builds the skills and knowledge of civil society
INTRAC organisations to be more effective in addressing poverty and inequality. Since 1992
INTRAC has provided specialist support in monitoring and evaluation, working with people
Contributor(s): to develop their own M&E approaches and tools, based on their needs. We encourage
Dan James, Alison Napier appropriate and practical M&E, based on understanding what works in different contexts.
and Nigel Simister

INTRAC Training M&E Universe

M&E Training & Consultancy M&E Universe
We support skills development and learning on a range of For more papers in
INTRAC’s team ofhigh
themes through M&E specialists
quality offer consultancy
and engaging and
face-to-face, For
themore
M&Epapers in
Universe
training in all aspects of M&E, from core
online and tailor-made training and [Link] development the series
M&E Universe
click the
through to the design of complex M&E systems. series
homeclick the
button
Email: training@[Link] Tel: +44 (0)1865 201851 home button
Email: info@[Link] Tel: +44 (0)1865 201851

© INTRAC 2017