Assessing Needs and Wants with Likert Scales¶

Strategy · All strategies
Evidence · 2 claims (2 for) · 3 studies (2 quant-synthesis, 1 design), q2–q4 · 2 of 3 report an effect size · 1 claim rests on one study

Description¶

This strategy uses short Likert-scale surveys (typically five-point, from "strongly disagree" to "strongly agree") to gather quantitative data on what learners need — prior knowledge, confidence, perceived relevance — and what they want — interests, preferences, motivations. Responses are aggregated to inform instructional decisions such as topic sequencing, pacing, grouping, and emphasis, and are typically collected at the start of a unit, at check-ins during it, and at closure.

Design Implications¶

Likert scales convert subjective learner states into comparable data, making them a lightweight form of Assessment that feeds instructional adjustment — the core move of Assessment for Learning [+M]. Their value depends entirely on what happens next: data that informs teaching improves learning, while data collected and ignored does not Formative assessment requires acting on the information gathered. [+S]. Item wording matters as much as the scale itself — ambiguous or double-barreled items produce noise that looks like data.

Context¶

Requirements¶

• 3–8 focused, single-idea items ("I can explain X" not "I understand and enjoy X")
• A consistent, labeled response scale; unlabeled midpoint categories invite inconsistent interpretation [Krosnick's survey-methodology work shows response-scale design systematically biases answers]
• A planned decision the data will drive — grouping, reteaching, topic choice, pacing
• Anonymity or low-stakes framing where honesty about confusion is the goal

Constraints¶

• Self-report measures perceived, not actual, understanding; learners with low skill often overestimate it, so scale data should be triangulated with performance evidence [-M]
• Acquiescence and central-tendency biases inflate mid-scale responses, especially with young learners or low-stakes contexts [~M]
• Ordinal responses are not interval data; computing means and treating a 4 as "twice" a 2 misrepresents the measurement [Jamieson (2004) documents common statistical misuse of Likert data] [-W]
• Repeated identical surveys produce response fatigue and drifting answers rather than genuine change data [~W]

Implementation Variability¶

• Pre-unit needs assessment: items on prior knowledge and confidence to calibrate starting points
• Interest/wants inventory: items on topic preferences and relevance to support choice and Activation of motivation
• Mid-unit pulse check: 2–3 items as a quick check-in to catch confusion early
• Exit self-rating: confidence ratings that learners compare against actual performance to build accurate self-assessment

Target Learners¶

• Adolescent and adult learners, who interpret scale items more consistently than young children [~M]
• Learners building self-assessment accuracy — comparing their confidence ratings to results exposes miscalibration and supports Self-Regulated Learning [+W]
• Reluctant participants, for whom anonymous scales surface needs that class discussion will not

Target Learning Goals¶

• Metacognitive monitoring: calibrating confidence against performance
• Affective and motivational diagnosis: identifying interest and perceived relevance to inform engagement design
• Instructional planning: locating the group's zone of need before teaching

Instructions¶

1. Define the decision the data will inform (regroup, reteach, choose examples, adjust pace).
2. Write 3–8 single-idea statements with a labeled five-point scale; avoid double-barreled and negatively worded items.
3. Administer anonymously before or during the unit; keep completion under three minutes.
4. Aggregate responses and identify the distribution, not just the average — a bimodal split calls for differentiated grouping, not a middle path.
5. Act on the data: adjust instruction, then tell learners what changed because of their responses, closing the Assessment loop.
6. Re-administer a parallel form at closure and have learners compare confidence ratings with actual performance to build calibration.

Claims¶

• The study used pre- and post-semester surveys with Likert-scale literacy-confidence items and open-text ethical reflection, thematically analysed [+W]

Related Strategies¶

• Entry and exit tickets — the performance-based complement to self-report scales
• Interest surveys and learner inventories — the "wants" side in more open-ended form
• Confidence ratings before answer submission — item-level rather than unit-level calibration
• Use percentage-consensus opinionnaires to locate shared values before redesign

Examples¶

• Pre-unit needs check in a science class: "I can explain the difference between weather and climate" (1–5), used to decide whether to open with the review lesson or skip it.
• Plickers and Mentimeter — tools for collecting anonymous scale responses in class and displaying the aggregate distribution instantly.
• Course-design feedback in higher education: mid-semester Likert pulse surveys (e.g., via institutional course-evaluation platforms) that instructors act on within the term rather than after it.

Key Sources¶

• Likert, R. (1932). A technique for the measurement of attitudes. Archives of Psychology, 22(140), 5–55.
• Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment in Education: Principles, Policy & Practice, 5(1), 7–74. doi:10.1080/0969595980050102
• Krosnick, J. A. (1999). Survey research. Annual Review of Psychology, 50(1), 537–567. doi:10.1146/annurev.psych.50.1.537
• Jamieson, S. (2004). Likert scales: How to (ab)use them. Medical Education, 38(12), 1217–1218. doi:10.1111/j.1365-2929.2004.02012.x