Home / Resources / Psychological Measures / SCL-90-R
SCL-90-R:
Scoring and Interpretation
The Symptom Checklist-90-Revised (SCL-90-R) is a 90-item self-report inventory of current psychological symptoms. SCL-90-R scoring yields nine symptom dimensions and three global indices, led by the Global Severity Index, and follows the publisher manual, so the sections below explain what each index means, how to analyze scored data, what the factor-structure evidence implies, and how to run and report it with SPSS and R.
What the SCL-90-R Measures
The SCL-90-R is a broad self-report inventory of current psychological symptoms. Unlike a disorder-specific questionnaire, it summarizes symptom distress across nine content dimensions and several global indicators. The Global Severity Index (GSI) is the usual choice when a study needs one overall measure of distress (Derogatis, 1994).
It is used in psychiatric assessment, outcome monitoring and studies comparing general symptom burden. The nine dimensions are clinically descriptive, but research has questioned whether they are distinct latent traits once broad distress is accounted for. Dimension-specific hypotheses therefore need attention to the factor-structure evidence and to multiplicity (Urbán et al., 2014).
Structure of the SCL-90-R
- Items: ninety, each rated 0 to 4 for distress over a recent period, usually the past seven days. Confirm the period in the edition you hold.
- Dimensions: somatization, obsessive-compulsive symptoms, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation and psychoticism.
- Global indices: the Global Severity Index (GSI), the Positive Symptom Distress Index (PSDI) and the Positive Symptom Total (PST). Their denominators and rules differ.
- Metrics: record whether reported values are raw index values or manual-derived normative T-scores.
- Related instruments: the BSI and BSI-18 are distinct instruments, and a shared dimension name does not make their scores comparable.
The instrument itself is available from Pearson Clinical Assessment and the SCL-90-R administration and scoring manual.
How to Score the SCL-90-R
- Confirm the edition, the 0 to 4 response coding and the availability of the official item-level scoring documentation.
- Obtain the nine dimension scores and three global indices with the publisher’s authorized scoring rules, including the manual’s handling of missing responses.
- If the dataset arrives professionally scored, verify ranges, missingness, norm provenance and whether values are raw or T-scores.
- Analyze scores as continuous outcomes or predictors only when the research purpose and the distribution support it. The code below works on scored indices and does not reproduce the item key.
* Illustrative syntax for publisher-scored indices; no item key is reproduced.
DESCRIPTIVES VARIABLES=scl90_gsi scl90_depression scl90_anxiety
/STATISTICS=MEAN STDDEV MIN MAX.
EXAMINE VARIABLES=scl90_gsi BY treatment_group
/PLOT=NONE /STATISTICS=DESCRIPTIVES.
# data: already-scored indices; variable names are examples.
vars <- intersect(c("scl90_gsi", "scl90_depression", "scl90_anxiety"), names(data))
summary(data[vars])
# Compare scores only after checking their scale and norm provenance.
Interpreting SCL-90-R Scores
| Score band | Interpretation |
|---|---|
| Higher GSI | Greater self-reported overall symptom distress |
| Nine dimension scores | Describe symptom groupings, not nine diagnoses |
| Normative or T-score interpretation | Use only with the edition-specific norms and scoring reference |
Source: Derogatis (1994) manual and later psychometric research.
A higher GSI indicates more severe or pervasive symptom distress during the reference period. The other global indicators summarize different features of responding and should not substitute for the GSI. A paper that uses caseness thresholds or T-scores must name the manual and the norm population, not only the abbreviation.
The SCL-90-R cannot establish a diagnosis on its own. A significant difference between dimension means also does not show that all nine dimensions are psychometrically distinct in your sample.
Psychometric Properties
| Property | Published finding | Source |
|---|---|---|
| Design | Ninety items, nine content dimensions and global distress indices | Derogatis (1994) |
| Structure | In a representative Hungarian community survey of 2,710 adults, a bifactor model fit best among those tested, and the nine-factor model also fit acceptably | Urbán et al. (2014) |
| Norming | Clinical interpretation requires the applicable edition and reference population | Derogatis (1994) |
| Reliability | Estimate only for scores with a defensible item set and interpretation | Study-specific |
These are source-specific findings and cannot be assumed to hold as reliability or accuracy estimates in a new sample.
Using the SCL-90-R in a Study
In a multi-arm study, prespecify whether the GSI or a named dimension is the primary outcome. A single GSI endpoint avoids presenting an exploratory set of nine outcomes as nine confirmatory findings, and where several dimensions are examined, multiplicity should be controlled or described openly.
For repeated outcomes, model the dependence within participants and distinguish change on the original metric from change in normative T-scores. State whether scores were keyed by a professional scoring system, how missing answers were treated, and whether general distress or a specific domain carried the primary hypothesis.
Reporting the SCL-90-R
General psychological symptom distress was measured with the Symptom Checklist-90-Revised (SCL-90-R; Derogatis, 1994). The [GSI or named dimensions] was obtained using [published scoring system and manual edition], with higher scores indicating more self-reported distress. The analysis prespecified [primary outcome and time point], and missing responses were handled according to [rule].
References
Derogatis, L. R. (1994). SCL-90-R: Administration, scoring & procedures manual (3rd ed.). National Computer Systems.
Urbán, R., Kun, B., Farkas, J., Paksi, B., Kökönyei, G., Unoka, Z., Felvinczi, K., Oláh, A., & Demetrovics, Z. (2014). Bifactor structural model of symptom checklists: SCL-90-R and Brief Symptom Inventory (BSI) in a non-clinical community sample. Psychiatry Research, 216(1), 146-154. https://doi.org/10.1016/j.psychres.2014.01.027
Frequently Asked Questions
Is the GSI the same as the sum of all SCL-90-R responses?
No. Raw sums, averages and normative indices use different metrics, so document the scoring algorithm you used.
Do the nine dimensions represent independent diagnoses?
No. They are symptom domains, and how distinct they are varies by dataset.
Can I use BSI-18 scores as if they were SCL-90-R values?
No. The instruments and scoring metrics differ, so comparison needs explicit harmonization and validity evidence.
Does a high SCL-90-R score confirm a psychiatric disorder?
No. It reflects reported symptom distress and needs appropriate assessment for any diagnostic claim.
NEXT STEP
Need SCL-90-R 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