We track anonymous visitor behavior on our website to ensure you have a great experience. Learn more about our Privacy Policy.

Conducting a Needs Assessment

Collecting data to understand a problem or challenge in context

Overview

In education, there is a tendency to rush to solutions before deeply understanding the problem or challenge. A needs assessment is a type of research project that intentionally seeks to clarify the underlying causes and identify potential ways to address them. By working through the steps described below, leaders can make more informed decisions and better identify opportunities for improvement.

Step 1: Start with Your Why

Before jumping in to the design of your needs assessment, make sure that you start by clearly articulating the purpose of your project and the objectives that you hope to achieve.

  • Purpose: Clearly articulate the problem, challenge, or trend that you want to understand
  • Objectives: Identify what you are trying to learn

Example: In a virtual school, teachers have noticed that students do not turn in their work when it has to be completed asynchronously. 

  • Purpose: Understand why students do not turn in their work
  • Objectives: (1) Understand what could be affecting the frequency and consistency with which students turn in their work asynchronously; (2) determine frequency with which different types of products or assignments are turned in asynchronously; and (3) identify any trends between how teachers ask students to complete their work and whether they turn it in.

Step 2: Articulate Your Research Questions 

Research questions are the driving force behind any measurement project – you are measuring because you need answers to questions! 

Good research questions are purposeful, focused, and actually researchable (learn more about how to develop good research questions). With a needs assessment, your research questions will revolve around WHY something may be occurring and HOW you know. Research questions should also align with your objectives and help you to achieve them.

Step 3: Identify the Measurement Tools for your Toolbox

Your research questions will drive the methods that you will use to collect and analyze data. For each research question, determine whether you need to collect numbersstories, or both in order to find an answer. Use the table below to understand potential tools for collecting data.

Quantitative Data - NUMBERS
What can be counted, calculated or compared
Qualitative Data - STORIES
What can be understood, observed, or explained
Research Questions:
  • How many/much…?
  • How often…?
  • What changed?
Research Questions:
  • What happened?
  • How did it happen?
  • Why did ___ occur?
Potential Measurement Tools:
  • Multiple-choice surveys
  • Assessment scores
  • Attendance data
  • Demographic data
  • Any other numeric data
Potential Measurement Tools:
  • Open-response survey questions
  • Interviews
  • Focus groups
  • Documents
  • Observations in the field

At The Learning Accelerator (TLA), we have developed a number of free and openly available tools that could be used as part of a needs assessment. 

Example: In a virtual school, teachers have noticed that students do not turn in their work when it has to be completed asynchronously. They want to understand why students do not turn in their work (purpose) and learn more about what could be causing this (objectives). After crafting their research questions ("RQs"), they identified potential ways to collect the data.

  • RQ1: What might be affecting the frequency and consistency with which students turn in their work asynchronously? This is a qualitative question requiring different stories or explanations. To collect this data, the district could use open-response survey questions or focus groups, but they also might try empathy interviews.
  • RQ2: What types of assignments do teachers ask their students to turn in asynchronously? This could be either a quantitative or qualitative question, depending on what you know about the general types of assignments. In addition to open-response survey questions or interviews, multiple-choice survey questions could also be used to identify trends.
  • RQ3: How do teachers structure assignments that ask students to turn in work asynchronously? This could be either quantitative or qualitative, depending on whether you know the types of assignments in advance. Consider how Likert scale questions could ask respondents to rate frequencies such as level of agreement or usefulness.

Step 4: Analyze Your Data

It can be easy to get overwhelmed by the amount of data that you collect. Before jumping into analysis, make sure that you organize your data as you collect it and then create a map so that you know what data helps you to answer which research questions. 

Remember that you might need to use a variety of tools – and even ask multiple questions to deeply understand your problem or challenge. While this may seem time-consuming, having good data will ultimately help you to identify and design more effective solutions.


Strategy Resources


Quality Drivers of Virtual/Hybrid Learning: Individual Self-Assessment

Using The Learning Accelerator’s Key Factors that Help Drive Remote Learning Quality as a framework... Learn More

Empathy Interview Question Template

When designing change in a district, it is essential that leaders hear directly from students... Learn More

Digital Equity Audit: Self-Assessment Tool

This Self-Assessment Tool is intended to foster productive dialogue within leadership teams in an iterative... Learn More

Know Your District Reflection Activity

When designing change, teams start with an in-depth analysis of what is currently working well... Learn More

badge equity image

Equity Focus

When you choose your measurement tools, make sure you also consider who will be using them. Do you need to offer surveys in multiple languages or at different reading levels? When you ask people to participate in a focus group, are there any power dynamics that need to be considered or dismantled? If you are using standardized assessment data, did you ask whether those measures truly represent the learning of the students you seek to understand?