Opens in a new tab

Epworth Sleepiness Scale:
Scoring and Interpretation

The Epworth Sleepiness Scale (ESS) is an eight-item self-report measure of how likely a person is to doze in everyday situations. Epworth Sleepiness Scale scoring is a plain 0 to 24 sum, and the care goes into three places: the developer's rule that a form with a missing item is invalid, the excessive-sleepiness bands that start at 11, and the gap between self-reported and objectively measured sleepiness. Code and a methods paragraph close the guide.

ITEMS8
SUBSCALESNone
RESPONSE SCALEChance of dozing, 0 to 3
SCORE RANGE0 to 24
DEVELOPEDJohns, 1991

What the Epworth Sleepiness Scale Measures

Johns (1991) designed the ESS to measure general daytime sleep propensity, the usual chance of dozing off in eight everyday situations. It does not measure hours slept, sleep quality or whether someone meets criteria for a sleep disorder.

It is used in sleep clinics, population surveys and intervention studies as a short subjective indicator that can be compared over time. Subjective sleepiness is not interchangeable with an objective sleep-latency measurement, and the score alone does not identify the cause of sleepiness.

Structure of the ESS

  • Items: eight situations, each rated from 0 (would never doze) to 3 (high chance of dozing).
  • Subscales: none. The standard result is one total from 0 to 24.
  • Direction: no reverse-scored items. Higher totals mean greater subjective daytime sleepiness.
  • Missing items: the developer states that an ESS with one or more missing item scores is invalid, because missing scores cannot be interpolated (Johns, n.d.).
  • Versions: the adult ESS and modified or pediatric forms are not interchangeable. Use the scoring rules of the form you administered and name the language and its source.

Johns (1991) describes the scale. In the methods, state which version you administered and keep its item wording and instructions as published.

How to Score the ESS

  1. Check that all eight responses are coded 0, 1, 2 or 3.
  2. Sum the eight responses to obtain ess_total, which ranges from 0 to 24.
  3. Leave the total missing when any item is missing, in line with the developer’s statement that such a form is invalid. Report how many forms were left unscored.
  4. If a form records half values or two marked responses for one item, apply the rule set in your protocol and report it. The code below is for integer-coded items.
SPSS SYNTAX

* Illustrative syntax: ess1 to ess8 coded 0 to 3 (integer values only).
COMPUTE ess_total = SUM.8(ess1 TO ess8).
EXECUTE.
R

# data: one row per respondent; items ess1 to ess8 coded 0 to 3.
items <- data[paste0("ess", 1:8)]
stopifnot(all(vapply(items, function(z) all(is.na(z) | z %in% 0:3), logical(1))))
data$ess_total <- ifelse(complete.cases(items), rowSums(items), NA_real_)

Interpreting ESS Scores

Score band Interpretation
0 to 10 Within the normal range of daytime sleepiness
11 to 12 Mild excessive daytime sleepiness
13 to 15 Moderate excessive daytime sleepiness
16 to 24 Severe excessive daytime sleepiness

Source: the developer’s interpretation guidance (Johns, n.d.). These are descriptive bands, not diagnoses.

Higher totals indicate greater self-reported sleepiness. A total of 11 or more falls in the range described as excessive daytime sleepiness, but it does not establish a diagnosis such as sleep apnea or indicate the cause. The bands are a descriptive convenience, so analyze the total as a continuous score when the question allows it and keep the bands for description.

Medication, comorbid illness, shift work and sleep deprivation all influence the total. Subjective sleepiness should not be treated as equivalent to an objective sleep-latency test.

Psychometric Properties

Property Published finding Source
Original development An eight-item questionnaire of general daytime sleep propensity Johns (1991)
Reliability and structure Reliability estimates and a factor analysis were reported in an early sample Johns (1992)
Evidence synthesis A systematic review evaluated measurement properties, with evidence varying by setting and population Kendzerska et al. (2014)
Reliability in new samples Estimate internal consistency in your own data and report it Study-specific

These are source-specific findings and cannot be assumed to hold as reliability or validity estimates in a new sample.

Using the ESS in a Study

Analyzing the 0 to 24 total as a continuous outcome keeps more information than the bands. For repeated assessments in the same people, a mixed-effects model with participant-level random effects respects the dependence. If a threshold is used to define elevated sleepiness, give its source, prespecify it and report the proportion with a confidence interval.

Because incomplete forms are unscorable, missingness is a design issue and not a coding detail. Report how many forms were lost, compare completers with non-completers, and consider whether a sensitivity analysis is warranted. Comorbidity and medication can shift scores, so adjust for preselected covariates and avoid causal wording in observational designs.

Reporting the ESS

Daytime sleep propensity was assessed with the eight-item Epworth Sleepiness Scale (ESS; Johns, 1991). Each item was rated from 0 to 3 and the items were summed to a total (range = 0 to 24), with higher scores indicating greater subjective daytime sleepiness. Forms with any missing item were not assigned a total. The total was analyzed [continuously or in descriptive bands], using [model]. Internal consistency in this sample was [alpha or omega].

References

Johns, M. W. (1991). A new method for measuring daytime sleepiness: The Epworth Sleepiness Scale. Sleep, 14(6), 540-545. https://doi.org/10.1093/sleep/14.6.540

Johns, M. W. (1992). Reliability and factor analysis of the Epworth Sleepiness Scale. Sleep, 15(4), 376-381. https://doi.org/10.1093/sleep/15.4.376

Johns, M. W. (n.d.). Epworth Sleepiness Scale [Developer website].

Kendzerska, T. B., Smith, P. M., Brignardello-Petersen, R., Leung, R. S., & Tomlinson, G. A. (2014). Evaluation of the measurement properties of the Epworth sleepiness scale: A systematic review. Sleep Medicine Reviews, 18(4), 321-331. https://doi.org/10.1016/j.smrv.2013.08.002

Frequently Asked Questions

No. It indicates greater subjective daytime sleepiness, and it does not identify the underlying cause.

No. The developer states that an ESS with a missing item is invalid, because missing scores cannot be interpolated.

No. The ESS measures daytime sleep propensity, whereas sleep-quality instruments address how well a person sleeps.

Yes, when the research question and the model support it. Report the bands only as description unless a prespecified threshold is part of the question.

NEXT STEP

Need ESS Data Scored
or Checked?

OnlineSPSS is a statistical consultancy for researchers. Share your design and dataset, and a statistically trained Research Coordinator reviews it. The scope and one fixed project fee are confirmed before work begins.

No Commitment Required

Strictly Confidential

Fixed Project Fee

In-House PhD-Level Statisticians

Same-Day Response

Serving Researchers Since 2014