Information/Data collection methodologies

Information and Data Collection Methodologies 

There are various ways to collect information about your programme. We have provided you with an abridged version below.

Collecting Data On:

Methodology or Tool

Description

Activities/Output  

(the services and products offered by your organisation)

Monitoring system  A monitoring system generally takes a variety of forms, using procedures to document and count the occurrence of outputs and activities. Some of the tools that you might use are well known. For example, an ‘attendance register’ would be used to count the number of people participating in a training. Other forms and tools will have to be created by you; perhaps your system uses an online database with electronic forms that your staff complete as they deliver a service.  You can look at our monitoring section for guidance on developing a monitoring system for your programme. 

Outcomes and Impact 

(to measure the achievement of your outcomes and effectiveness of your programme)

These are also column headings (left = Quantitative methods, right = Qualitative methods): 

A quantitative approach (involves asking information in a way that requires counting[1] and will give you information that is measurable and comparable)

​A qualitative approach (provides you with information that will allow you to develop an in-depth understanding of the nature and characteristics of the phenomena that you are studying)

Interviews with individuals Structured interviews allow you to repeatedly collect the same information from different people on specific, predetermined questions or concepts (variables).  In-depth interviews ask open-ended questions about a general theme, which allows various sub-themes to emerge through conversation. This  allows you to develop a deep understanding of the theme or the participant's perceptions of the theme.  
Group discussions Focus group discussions are types of discussions led by a focus group moderator. They last approximately 60–90 minutes during which a group of 8–12 respondents are asked to comment on specific, predetermined questions or concepts using a structured moderators guide (questionnaire). Group discussions are informal, largely unstructured, in-depth group interviews where multiple respondents give input.   
Observation Observers record what they see and hear at the project site using an observation checklist. The observation may be of physical surroundings, ongoing activities, processes and discussions. This is clasically a qualitative method, also known as field research/ethnography. It involves the detailed observation of people in a specific setting/location (often over an extended period of time) to obtain an insider’s point of view. This is done in order to develop a deep understanding of the theme being studied. 
Case studies  Not done for quantitative research.  A large amount/variety of general information (such as information gathered from in-depth interviews/group discussions/observation) is compiled on a limted number of cases and studied in-depth. This analysis shows emerging patterns and themes that  allow a greater understanding of the subject being researched.  Important note: An anecdote/letter from a beneficiary is not a case study – it might be one piece of information in a case study.   
Surveys

A survey normally involves a questionnaire or test which is designed to collect information, that is easily countable, on very specific questions/themes (variables). These questionnaires can also be completed using interviews, often seen in the proccess of collecting data for a census.

The real issue in surveys is the fact that a smaller (sample) group of people are selected to be studied from a larger group and researchers hope to make generalisations about the entire group based on the information taken from the sample[2]

One way of doing this is to randomly select a large enough sample.  Random selection means that a participant’s chances of being chosen is a bit like like winning a lottery. Every person has one ticket and an equal chance in selection.  If you select enough people in this way you will find that the attributes of the chosen individuals are representative of the entire group from which they were originally chosen. 

Although there will be some limitations in terms of the strength of your study, you can also do informal small-scale sample (30-50 individuals) surveys where participants are not randomly selected.  There are various methods of selection, such as purposeful selection of people in different categories. This is done on the basis of easy accessibility and appropriate statistical tests. 

Not done for qualitative research

[1]This asks specific questions that do not require further elaboration such as "do you have a television – yes/no". This often provides pre-determined options that participants need to choose between, such as "best thing to do with a ball: a) kick it; b) throw it; c) hit it" or it asks participants to rate something on a scale (e.g. 1 - 5, with 1 representing terrible and 5 meaning wonderful).   

[2] Except when they do the census; in this case they actually try to count everybody (researchers would say a census represents a complete enumeration, or listing, of all units in a population)