Where it sits
The word travelled from the first to the second, and the research measures the second.
Tell it apart
| Aspect | Gratitude | Appreciation |
|---|---|---|
| Requires a giver | Yes | No |
| Direction | Outward | Inward |
| Can apply to a sunset | Awkwardly | Naturally |
| Creates obligation | Sometimes | No |
Appreciation is noticing value. Gratitude, in its narrow sense, points back to a giver — though broader definitions in the research drop that requirement. That structural difference — the benefactor — is what the interventions try to activate, and also why gratitude research is hard to separate from general positive-attention tasks.
What it doesn't mean
The pooled effects are small and the evidence is graded weak. Diniz and colleagues systematically reviewed 64 randomised controlled trials and meta-analysed 33. Typical results: gratitude scores about 3.7% higher, GAD-7 anxiety down about 7.8%, PHQ-9 depression down about 6.9%. The authors' own verdict is that results must be analysed cautiously — all outcomes were classified as high risk of bias due to lack of blinding and high attrition, and certainty of evidence was predominantly low or very low.
They shrank as controls improved. In Davis and colleagues' 2016 meta-analysis, the effect on wellbeing ran 0.31 against a measurement-only control, 0.14 against an activity-matched one, and -0.03 against a psychologically active comparison — the signature of an intervention whose benefit is mostly 'doing a structured pleasant task' rather than gratitude specifically.
What it really means
For about a hundred years, gratitude in English meant good will — a disposition toward someone else. The sense we use now, thankfulness as an inner state, arrives in the 1560s. The idea that it is a technique you perform on yourself for your own benefit is younger than most people assume, and it is the version the research actually measures.
So: does it work?
The most complete answer available is a systematic review of 64 randomised controlled trials, 33 of them pooled in meta-analysis. The results are real and small. Gratitude scores rose about 3.7%. Anxiety on the GAD-7 fell about 7.8%. Depression on the PHQ-9 fell about 6.9%.
Then the authors say the part that rarely travels: results must be analysed cautiously. Every outcome was classified as high risk of bias — no blinding, high attrition — and the certainty of evidence was predominantly low or very low.
There is a pattern underneath those numbers worth naming, because it recurs across wellbeing research. The effect looks largest against a waitlist — people who did nothing — and shrinks toward nothing against an active control, people who did some other structured pleasant task. That is the fingerprint of an intervention whose benefit is mostly doing a small deliberate thing regularly, rather than gratitude specifically.
Which is not nothing. A cheap, low-risk habit with a small real effect is a reasonable thing to do. It is just not what the phrase proven to rewire your brain is selling.
This page is educational and not a diagnosis. If low mood or anxiety is persistent, a doctor, therapist or counsellor is the right place to take it — not a notebook.
Where it comes from
Medieval Latin — thankfulness
Latin — thankful, pleasing
The word meant 'good will' for roughly a century before it meant 'thankfulness'. It described a disposition toward someone else, not an inner state you cultivate for your own benefit. The self-improvement framing is modern.
- mid-15th century
First attested in English meaning 'good will'.
- 1560s
The modern sense — thankfulness — is documented.
In real life
- A neighbour clears the snow off your path before you're even up. The warm thankfulness you feel — and the urge to bake them something back — is gratitude in its older, outward-facing sense.
- You appreciate a sunset; you feel grateful to the friend who drove you up the hill to see it.
- The modern technique version is the gratitude journal — three lines each night about things that went well — studied in trials that find small, contested effects.
Questions
Q01Does a gratitude journal work?
The meta-analytic evidence shows small statistically significant improvements in wellbeing, anxiety and depression measures, with high risk of bias and low certainty. It is cheap and low-risk, which is a fair reason to try it. It is not a substitute for treatment.
Q02Why do the effect sizes keep shrinking?
Because the comparison got harder. In Davis and colleagues' 2016 meta-analysis the effect on wellbeing ran about 0.31 against a measurement-only control but -0.03 against a psychologically active one. Blinding is also impossible in these designs and attrition is high, which is why reviewers grade the certainty of this evidence as low.
Q03Did gratitude research fail to replicate?
It is better described as replicating at a much smaller size than the popular coverage implied, rather than collapsing outright. This page is not a diagnosis — if you are relying on a journal in place of care, a mental health professional is the better call.
Sources
- Diniz et al., The effects of gratitude interventions: a systematic review and meta-analysis (Einstein São Paulo 2023), PMC.
- Etymonline — gratitude.