Home / Resources / Psychological Measures / RSES
Rosenberg Self-Esteem Scale:
Scoring and Interpretation
The Rosenberg Self-Esteem Scale (RSES) is a 10-item measure of global self-worth. Rosenberg Self-Esteem Scale scoring depends on which of the two circulating item orders was used, so reverse scoring is set out for both, alongside interpretation, SPSS and R code, psychometric evidence and a model reporting paragraph.
What the Rosenberg Self-Esteem Scale Measures
The Rosenberg Self-Esteem Scale (RSES) measures global self-worth, meaning a person’s overall evaluation of themselves rather than their confidence in a specific domain such as work or appearance. Rosenberg introduced it in his research on adolescent self-image (Rosenberg, 1965). Because it is short and takes roughly two to five minutes, it is used in social, developmental, health and educational psychology, either as an outcome in intervention and longitudinal studies or as an explanatory variable in association models.
The scale has been studied far beyond its original setting. Schmitt and Allik (2005) analyzed translated forms administered in 53 nations (N = 16,998) and reported evidence broadly supporting a comparable structure across nations. Comparability of structure does not make raw scores interchangeable, though. Differences in item wording, response coding and cultural context all affect how a total should be read, which is why the scoring decisions below matter.
Structure of the RSES
- Items: ten agreement statements, five worded positively and five worded negatively.
- Item order: two orderings circulate. In the University of Maryland ordering, the negatively worded items are 2, 5, 6, 8 and 9. In the ordering reproduced from Rosenberg (1965), they are 3, 5, 8, 9 and 10. This guide follows the University of Maryland ordering.
- Response scale: four-point agreement, coded here 0 to 3, with higher numbers meaning more positive agreement before the five negative items are reversed.
- Subscales: none in the standard total. Factor-analytic work describes positive and negative wording effects, but the conventional score is a single sum.
- Total: one summed score from 0 to 30. An alternative 1 to 4 coding gives 10 to 40 and must be labeled explicitly.
- Recall period and unscored items: no recall period (the scale asks about general feelings about oneself) and no unscored items.
The instrument itself is available from University of Maryland Department of Sociology.
How to Score the RSES
- Confirm the exact item order of the form you used. The reverse-item numbers in this guide (2, 5, 6, 8 and 9) apply only to the University of Maryland ordering. If your form follows the Rosenberg (1965) ordering, reverse items 3, 5, 8, 9 and 10 instead.
- Code every raw agreement response from 0 (strongest disagreement) to 3 (strongest agreement) before any reversal.
- Reverse the five negatively worded items as 3 minus the response, then sum all ten keyed item scores to obtain rses_total, which ranges from 0 to 30.
- Apply a missing-item rule. The cited sources set no universal prorating threshold, so the code below requires all ten items to be present. If you use another rule, state it in the methods section.
* Illustrative syntax: variables must already be valid item scores (0 to 3).
* University of Maryland item ordering: reverse items 2, 5, 6, 8, 9.
COMPUTE rses2_r = 3 - rses2.
COMPUTE rses5_r = 3 - rses5.
COMPUTE rses6_r = 3 - rses6.
COMPUTE rses8_r = 3 - rses8.
COMPUTE rses9_r = 3 - rses9.
COMPUTE rses_total = SUM.10(rses1, rses2_r, rses3, rses4, rses5_r, rses6_r, rses7, rses8_r, rses9_r, rses10).
EXECUTE.# data: one row per respondent; items rses1 to rses10 coded 0 to 3.
# University of Maryland item ordering: reverse items 2, 5, 6, 8, 9.
data$rses2_r <- 3 - data$rses2
data$rses5_r <- 3 - data$rses5
data$rses6_r <- 3 - data$rses6
data$rses8_r <- 3 - data$rses8
data$rses9_r <- 3 - data$rses9
data$rses_total <- rowSums(data[c("rses1", "rses2_r", "rses3", "rses4", "rses5_r", "rses6_r", "rses7", "rses8_r", "rses9_r", "rses10")], na.rm = FALSE)Interpreting RSES Scores
| Score band | Interpretation |
|---|---|
| 0 to 30 | Continuous global self-esteem; higher = higher self-esteem |
| 15 to 25 (0 to 30 coding) | Often described as the “normal” range in guidance reproduced by the University of Maryland; scores below 15 are sometimes read as suggesting low self-esteem |
| Universal clinical cut-off | None validated across settings, cultures and scoring conventions |
Source: University of Maryland, “Using the Rosenberg Self-Esteem Scale”; Rosenberg (1965).
A higher RSES total indicates a more positive overall self-evaluation. No clinical diagnosis is attached to a single raw score, and the 15 to 25 “normal range” is a rule of thumb rather than a validated cut-off. Reported low, normal and high bands are not consistently validated across cultures and versions, so RSES interpretation is best anchored in the population actually under study.
The most common scoring error is applying reverse-item numbers from a differently ordered questionnaire, or mixing 0 to 3 and 1 to 4 totals. Before analysis, verify the version, the ordering and the direction of all ten variables against the survey codebook.
Psychometric Properties
| Property | Published finding | Source |
|---|---|---|
| Internal consistency | A validation synthesis of 28 studies found that reliability evidence was largely based on internal consistency, without a single transportable coefficient | Hubley et al. (2025) |
| Test-retest | The same synthesis identified little longitudinal reliability evidence; no single universal retest estimate is asserted | Hubley et al. (2025) |
| Factor structure | A meta-analysis of 113 samples (N = 140,671) supported a general factor alongside positive and negative item wording effects | Gnambs et al. (2018) |
| Cross-national validity | 16,998 participants across 53 nations; largely invariant factor structure across nations | Schmitt & Allik (2005) |
| Diagnostic accuracy | No single clinical screening cut-off or sensitivity and specificity convention | University of Maryland (n.d.); Rosenberg (1965) |
These are source-specific results and cannot be assumed to hold as reliability or accuracy estimates in a new sample.
Using the RSES in a Study
The standard RSES total works best as a single continuous variable, whether it serves as an outcome in an intervention or longitudinal design or as a predictor in an association model. Check the negatively worded items for unusual response patterns before you settle on that, especially in adolescent, multilingual and cross-cultural samples; the wording effects described in the factor-analytic literature are a reason to inspect the item structure when your research question depends on it. If you compare groups, such as countries or language versions, test measurement invariance before you interpret differences in raw means.
Report reliability from your own sample together with the 0 to 30 or 10 to 40 coding convention you chose, and audit the reverse scoring before you compute any internal-consistency estimate, because a wrongly reversed item depresses alpha and misleads. Plan sample size from the primary association or factor model rather than from a generic rule based on the ten items.
Reporting the RSES
Global self-esteem was assessed with the 10-item Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965). Responses were coded on a 0 to 3 agreement scale, and the negatively worded items (2, 5, 6, 8 and 9 in the University of Maryland item order) were reverse scored. Item scores were summed (range = 0 to 30), with higher totals indicating greater self-esteem. Cronbach’s alpha in the present sample was [α = XX].
References
Gnambs, T., Scharl, A., & Schroeders, U. (2018). The structure of the Rosenberg Self-Esteem Scale: A cross-cultural meta-analysis. Zeitschrift für Psychologie, 226(1), 14-29. https://doi.org/10.1027/2151-2604/a000317
Hubley, A. M., Ruddell, R. J., & Zhu, S. M. (2025). Cracks, gaps, and holes in validation practice as evidenced from a validation synthesis of the English version of the Rosenberg Self-Esteem Scale. European Journal of Psychological Assessment. https://doi.org/10.1027/1015-5759/a000877
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton University Press.
Rosenberg, M. (1989). Society and the adolescent self-image (Rev. ed.). Wesleyan University Press.
Schmitt, D. P., & Allik, J. (2005). Simultaneous administration of the Rosenberg Self-Esteem Scale in 53 nations: Exploring the universal and culture-specific features of global self-esteem. Journal of Personality and Social Psychology, 89(4), 623-642. https://doi.org/10.1037/0022-3514.89.4.623
University of Maryland, Department of Sociology. (n.d.). Using the Rosenberg Self-Esteem Scale. https://socy.umd.edu/about-us/using-rosenberg-self-esteem-scale
Frequently Asked Questions
Which Rosenberg items are reverse coded?
It depends on the item order of your form. In the University of Maryland ordering, items 2, 5, 6, 8 and 9 are negatively worded and reversed. In the ordering reproduced from Rosenberg (1965), the reversed items are 3, 5, 8, 9 and 10. Always match the item numbers to the questionnaire your participants actually completed.
Why do some reports use a 10 to 40 range instead of 0 to 30?
Some researchers assign 1 to 4 points to each response rather than 0 to 3. Both conventions are in use, and the totals differ by a constant of 10, so raw scores from the two conventions cannot be compared without recoding to a common scale.
Is a score of 15 clinically low?
Not on the evidence available. A score below 15 is sometimes read as suggesting low self-esteem, but no clinical threshold has been validated across settings, cultures and scoring conventions. Interpret a total against its scoring convention, your sample and the relevant literature.
Can the RSES be used as one continuous variable?
Yes. A single continuous total is the most common use of the standard score. Inspect the item structure, particularly the negatively worded items, when your research question depends on it.
NEXT STEP
Need an RSES Score Set Up
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