Needs Assessment Methods on a Small Budget

Who This Page Is For

You have a small budget and about two weeks, and you need to learn what a community needs. You can run focus groups, interviews, a survey or some observation, but you cannot run everything. This page helps you choose a mix you can finish, and tells you where each method tends to let you down.

The Short Answer

For most small teams, pair a short survey for breadth with a few key informant interviews for depth. Add a walking or windshield survey if you need to see how the place is used. Leave out digital mapping unless you truly need information in layers that you cannot get from tables you already have. Each method has a blind spot, so two methods that fail in different ways beat one done perfectly. Whatever you choose, set limits on what you will collect before you start. Collecting too much is as much a problem as collecting too little, because the team gets lost in the data.

How They Differ

The real split is between methods that give you numbers and methods that give you explanations.

Method	What it gives you	What to watch
Focus group	Spoken, open-ended, qualitative answers, plus nonverbal cues	Needs more time per respondent than an individual survey, because of recruiting and the session itself; quieter people may not speak
Interview	In-depth views from people who know the topic	Poor for straight numeric data; interviewees may have their own agendas
Survey	Closed-ended questions, scored as numbers	Needs a pre-set question list and a sample chosen in advance
Public forum or listening session	A wide range of views at once; working groups kept under 30 to 40 people, with larger crowds split up	Needs a facilitator who is known and respected, neutral on the topic and good at listening
Windshield survey	Systematic observation from a moving vehicle	Needs at least two people, so one can drive safely while the other records
Walking survey	Systematic observation on foot	Two or three people; groups of five or six stand out
Mapping (GIS)	Layers of spatial information you can add or remove	Needs hardware, software, accurate geographic data and trained people

Focus groups earn their cost when you want depth and want to watch how people react. A written survey scales better, but it cannot follow up on an odd answer. A community listening forum can cover four things: issues and concerns, barriers and resistance, community resources for change, and recommended alternatives and solutions. That is a lot of ground for one meeting, so decide which of them you need. Observation shows what people do rather than what they say, though you may need to repeat it at different times of day, week or year to see real activity.

How to Decide

Start with the kind of answer you need. If you need a rate or any numeric picture across many people, use a survey. If you need opinions, assumptions or expert knowledge, use interviews or focus groups.
Match the conversation method to your time. Focus groups cost more time per respondent than individual surveys, because you have to recruit people and then run a session of its own, so plan hours for both. If you hold them, spread invitations across a representative sample of the target population, such as different facilities or business types, and not only the most vocal or extreme voices. If time is very short, telephone interviews work, kept to about ten minutes at most.
Add observation if place matters. A windshield survey suits a team with a vehicle. A walking survey suits a team that wants to do what residents do: ride public transport, eat at local restaurants, shop at local stores. A small neighborhood or rural village can be covered in a day by one person, while a large city takes several days and several teams.
Set the sample by the size of the group. For a small or medium-sized group, include every single person. For a larger community, around 5,000 people or more, use a percentage chosen in advance. If the group is under 30, do not calculate a rate from it; the number is too small to give a credible picture.
Be slow to reach for GIS. If you need only a little data and it is not tied to one place, tables such as census figures may give you the same answer in an afternoon. Mapping pays off mainly before you start, in planning, and when you refine the work. If you do use it, remember that it only works with four things in place: hardware, software, accurate geographic data and trained staff. If one of them is missing, mapping will not work.

Worked Example

An illustration, not a real case. A small team at a community center in a rural village wants to find local assets that could support youth services. Money is tight and the deadline is close.

First, three team members do a walking survey over one day. They shop at local stores and eat at local restaurants, and they note the physical structures, services and businesses that could help young people. Because they dress and behave like residents, they do not stand out.

Next, they pick a few key informants, such as teachers and local leaders, and interview them by telephone, keeping each call to about ten minutes. Before any recording starts, the interviewer asks permission and states on the recording that it was given. The team also asks itself what each person might want from the conversation, since interviewees can have an agenda.

Then they survey the village. It is a small group, so they ask every person instead of sampling. The draft starts with clear instructions, which are often forgotten but set the tone for respondents, and it is tested first on people similar to the ones who will answer.

Along the way they keep a list of what they found: people, buildings, services and businesses. They plan to review that list from time to time so it stays current.

Finally, the team limits what it collects to what youth services need, and it holds a community meeting to report back to everyone who gave their time.

Needs Assessment Methods for a Small Team and Budget