Likert Scale Construction
Rensis Likert's Method for Constructing Summated Rating Scales · Also known as: Likert summated rating scale, Summated rating scale construction
Likert scale construction is a systematic methodology for developing attitude measurement instruments using summated rating scales. Introduced by Rensis Likert in 1932, it enables researchers to quantify latent constructs such as attitudes, beliefs, and psychological states by aggregating responses across multiple items. The method remains foundational to quantitative social and health sciences research.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Likert scale construction is appropriate for measuring attitudes, beliefs, psychological traits, and subjective experiences in diverse populations. It is commonly used in clinical psychology, occupational health, patient-reported outcomes, and social science research. The method is preferred when a construct cannot be directly observed and is multidimensional. However, it may be less suitable for measuring objective behaviors or in populations with very low literacy.
Strengths & limitations
- Combines multiple items to increase reliability and validity beyond single-item measurement
- Flexible and applicable to a wide range of constructs and populations
- Simple to administer and score, reducing respondent and administrative burden
- Well-established psychometric framework with extensive literature supporting best practices
- Assumes unidimensionality and equal item weighting, which may not reflect complex constructs
- Vulnerable to response bias (acquiescence, social desirability) that can inflate or deflate scores
- Ordinal responses are often treated as interval data, an assumption that may not hold in practice
Frequently asked
Should I always use a 5-point Likert scale?
No. The number of response options depends on your construct and population. Five-point and seven-point scales are most common, offering a balance between granularity and respondent discrimination. Scales with fewer points (3–4) sacrifice discrimination; those with more than 7 points rarely gain additional information and may confuse respondents.
What is the minimum number of items for a Likert scale?
While technically you can use 2 items, best practice recommends at least 3–4 items per subscale and ideally 5–10 or more for unidimensional constructs. More items generally improve internal consistency (Cronbach's alpha), though diminishing returns occur above 20 items.
Can I treat Likert scale scores as continuous variables?
This is debated. Likert responses are ordinal, but with sufficient items (typically 5+) and large sample sizes, the summed score approximates normality and parametric tests (t-test, ANOVA) are generally robust. For small samples or few items, consider ordinal methods or resampling.
How do I know if my items are valid?
Conduct exploratory factor analysis (EFA) or confirmatory factor analysis (CFA) to verify that items cluster into expected dimensions. Calculate item-total correlations (items should correlate r > 0.3 with the total); examine Cronbach's alpha (α ≥ 0.70); and assess differential item functioning (DIF) across demographic groups.
Sources
- Likert, R. (1932). A technique for the measurement of attitudes. Archives of Psychology, 22(140), 1-55. link ↗
- Kerlinger, F. N., & Lee, H. B. (1986). Foundations of Behavioral Research (3rd ed.). New York: Holt, Rinehart & Winston. ISBN: 0030652669
- DeVellis, R. F. (2016). Scale Development: Theory and Applications (4th ed.). Thousand Oaks, CA: Sage Publications. ISBN: 9781506330174
How to cite this page
ScholarGate. (2026, June 3). Rensis Likert's Method for Constructing Summated Rating Scales. ScholarGate. https://scholargate.app/en/psychometrics/likert-scale-construction
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Confirmatory Factor Analysis for ScalesPsychometrics↔ compare
- Content Validity RatioPsychometrics↔ compare
- Factor Analysis for Scale DevelopmentPsychometrics↔ compare
- Floor and Ceiling EffectPsychometrics↔ compare
- Guttman ScalePsychometrics↔ compare