Home / Resources / Psychological Measures / SHS
Subjective Happiness Scale:
Scoring and Interpretation
The Subjective Happiness Scale (SHS) is a four-item self-report measure of how happy people consider themselves overall (Lyubomirsky & Lepper, 1999). Subjective Happiness Scale scoring is a short calculation with one trap, the reversed fourth item, so the page sets out that keying, how to read a mean score without clinical bands, SPSS and R syntax, and a methods paragraph you can adapt.
What the Subjective Happiness Scale Measures
The SHS captures a global judgment of one’s own happiness. It does not track momentary mood, and it does not ask respondents to rate how often particular emotions occurred (Lyubomirsky & Lepper, 1999). Two items ask people to rate themselves in absolute terms and in comparison with their peers, and two describe a happy and an unhappy person and ask how far each description fits the respondent.
The development work used many different samples and reported reliability and convergent validity evidence for the scale (Lyubomirsky & Lepper, 1999). The format suits surveys that carry many constructs, but four items leave little room for error. One miskeyed item changes the score for every respondent, so the keying and the response anchors deserve a check before any model is run.
Structure of the SHS
- Items: four, each answered on a 7-point scale with its own paired anchors.
- Subscales: none. The standard output is one global happiness score.
- Direction: item 4 is reverse scored. Higher scores mean greater self-rated happiness.
- Recall period: none. Respondents give a general self-assessment and not a rating of a recent window.
- Versions: the original four-item scale and translated adaptations exist. Anchors in an adaptation can differ, so compare scores across languages only after checking them.
The instrument itself is available from Sonja Lyubomirsky’s academic website.
How to Score the SHS
- Check that shs1 to shs4 contain integers from 1 to 7.
- Reverse item 4 by computing 8 minus the recorded response.
- Take the arithmetic mean of shs1, shs2, shs3 and the reversed item 4, only for records with all four valid responses. This is the SHS score, with a range of 1 to 7.
- After a survey export, confirm which column holds the developer’s item 4. A platform may reorder or relabel items, and the reversed item is defined by its content and not by its column position.
* Illustrative syntax: shs1 to shs4 already hold responses coded 1 to 7.
COMPUTE shs4_rev = 8 - shs4.
COMPUTE shs_mean = MEAN.4(shs1, shs2, shs3, shs4_rev).
EXECUTE.
# data: one row per respondent; items shs1 to shs4 coded 1 to 7.
cols <- paste0("shs", 1:4)
stopifnot(all(vapply(data[cols], function(z) all(is.na(z) | z %in% 1:7), logical(1))))
data$shs_mean <- ifelse(complete.cases(data[cols]),
(data$shs1 + data$shs2 + data$shs3 + (8 - data$shs4)) / 4, NA_real_)
Interpreting SHS Scores
| Score | Interpretation |
|---|---|
| Toward 1 | Lower self-rated global happiness |
| Toward 7 | Higher self-rated global happiness |
| Clinical or diagnostic bands | None established for this scale |
Source: Lyubomirsky and Lepper (1999); no diagnostic cut-off is part of the scale.
A mean closer to 7 reflects greater self-rated happiness, and a mean closer to 1 reflects less. The scale is not a measure of psychiatric illness. Without reference data for your population, avoid sorting individual respondents into “happy” and “unhappy” groups at an arbitrary dividing score.
Interpret group differences and change in light of the design, with confidence intervals, and not through a single number. A high or low score does not by itself explain why a group differs.
Psychometric Properties
| Property | Published finding | Source |
|---|---|---|
| Internal consistency | Reported as high across the development samples | Lyubomirsky & Lepper (1999) |
| Temporal stability | Test-retest results supported stable global happiness judgments | Lyubomirsky & Lepper (1999) |
| Validity | Convergent and discriminant evidence against other affective and personality measures | Lyubomirsky & Lepper (1999) |
| Cross-cultural use | Item functioning and anchors can differ by language, so test invariance before pooling | Study-specific |
These are source-specific findings and cannot be assumed to hold as reliability estimates in a new sample.
Using the SHS in a Study
The SHS mean can serve as a continuous outcome, predictor or moderator, subject to a look at its distribution, because a mean of four ordinal items is bounded at 1 and 7 and often skews toward the upper end. If it is an outcome, prespecify what counts as a meaningful change and report the estimate with its confidence interval instead of labeling any shift an improvement.
With four items, a coefficient is sensitive to a single weak item, so estimate reliability from your own data and report item distributions and missingness. When groups from different cultures or languages are compared, examine measurement invariance before attributing a difference in means to the groups themselves.
Reporting the SHS
Global subjective happiness was measured with the four-item Subjective Happiness Scale (Lyubomirsky & Lepper, 1999). The fourth item was reverse coded, and the mean of the four responses was computed for participants with complete data (range = 1 to 7), with higher scores indicating greater self-rated happiness. Internal consistency in the present sample was [α = XX], and missing responses were handled by [method].
References
Lyubomirsky, S., & Lepper, H. S. (1999). A measure of subjective happiness: Preliminary reliability and construct validation. Social Indicators Research, 46(2), 137-155. https://doi.org/10.1023/A:1006824100041
Frequently Asked Questions
Which SHS item is reverse scored?
Item 4 in the original four-item order. On the 1 to 7 scale the recoding is 8 minus the recorded response.
Is the SHS scored as a sum or a mean?
The score is the mean of the four keyed responses, so it stays on the 1 to 7 metric. A raw four-item sum is not the standard score.
Can the SHS diagnose depression?
No. It records a global happiness judgment and is not based on diagnostic criteria, so it cannot classify depression.
Does the SHS have severity bands?
No universal bands exist. Use comparisons that the study design supports, and do not invent a cut-off.
NEXT STEP
Need SHS Data Coded Correctly
Before Analysis?
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