People who finish a gratitude exercise still feeling low aren’t failing at positivity — the measured benefit is small on average and varies across studies

You tried to focus on what went well, then wondered whether the exercise changed anything real.

The numbers point to a modest answer: gratitude practices sometimes improve wellbeing, gratitude itself, or a related outcome, yet results vary across people and studies.

Research has found higher wellbeing after some gratitude interventions. Other work found no clear link, mixed results, or gains that depended on the group studied.

The strongest reading is careful and useful. Gratitude can support a positive shift for some people, while it cannot stand in for care, treatment, or a full answer to distress.

Before the sections open, the figures this page stands on — each one carrying its own source.

What the numbers actually measure

Your question starts with a simple concern. Does writing or saying thanks create a measurable change, or does it only feel like good advice?

Studies measure several different outcomes. They may ask about gratitude, happiness, life satisfaction, stress, sleep, mental health, or social connection.

Those outcomes overlap, yet they are not the same. A change in gratitude does not automatically prove a change in every part of wellbeing.

A 2023 meta-analysis of randomised controlled studies found that expressed gratitude interventions had a significant effect on psychological wellbeing.

The reported Hedges’ g was 0.22, with a 95% confidence interval from 0.11 to 0.33.

That result describes a small average difference between gratitude groups and neutral comparison groups. The small average difference does not mean every person felt better.

One university study during COVID-19 found higher wellbeing in the gratitude group than in the control group. The reported Cohen’s d was 0.74.

These findings answer different questions. One gives a pooled result across studies. The other describes one study with one group of students.

The number that fits your experience may therefore differ from the average. Your starting mood, setting, task, and response to the exercise all matter.

Pick the result that matches the question you brought here. The figures separate a measured effect from a broad promise about gratitude.

How large is the average wellbeing effect?

You may want a clear size for the change. The pooled result gives the best short answer, and it points to a modest average effect.

Hedges’ g of 0.22 came from a meta-analysis of expressed gratitude interventions. The comparison used neutral groups.

A small average effect can still matter to some participants. It can also feel barely noticeable to others.

Average results combine people who improved, people who stayed much the same, and people whose scores moved in another direction. The average cannot show each path.

The confidence interval ran from 0.11 to 0.33. This range shows uncertainty around the pooled estimate.

That uncertainty matters when you read strong claims online. A study result supports a measured link or group difference.

A measured link or group difference cannot support a promise of a set personal outcome.

Another study found a larger reported difference, with Cohen’s d of 0.74. It examined university students during COVID-19, so its setting differs from the wider meta-analysis.

Different samples and tasks can produce different numbers. You should read each figure with its study group and comparison in mind.

Thoughts arrive as first drafts. The lab below is where you edit one and feel the sentence loosen.

What happened after school gratitude programs?

If your interest began with a school setting, the six-week result may feel most relevant.

Students in gratitude intervention classes showed improved trait gratitude, mental health, and personal and social wellbeing.

The comparison involved waitlist or control classes. The reported result came after 6 weeks.

This finding covers several outcomes at once. It does not show that every student improved in every area.

A systematic review and meta-analysis of gratitude in youth and schools reached a more cautious conclusion.

Across the intervention research, gratitude-based programs were generally ineffective overall, and the review called for more research.

Those findings can sit together. One school study may report improvement, while a wider review finds that results across programs do not form a strong, steady pattern.

For the reader comparing these results, the setting matters. A class program may include discussion, social contact, teacher support, or other activities.

A school review also covers different ages, tasks, comparison groups, and study designs. The label gratitude intervention does not guarantee one shared experience.

The clearest conclusion stays narrow. Some school programs produced better scores after a short period. The full school literature remains mixed.

Compare the study group, outcome, and time point before treating one result as the answer for everyone.

A page should show its load-bearing numbers plainly. These are this one’s, receipts attached.

Can gratitude change mental health measures?

You may have arrived while feeling low, tense, or worn down. The research gives some encouraging findings, though it does not support a single mental health rule.

In one study of parents of troubled children, a combined Best Possible Self and gratitude writing intervention found significant decreases in stress, anxiety, somatic symptoms, and depression.

The study also found increased gratitude.

That study tested a combined intervention. Its results cannot show that gratitude writing alone caused every change.

A systematic review of gratitude interventions among workers found significant improvement in perceived stress and depression across eight gratitude list studies. Effects on wellbeing were inconsistent.

This difference is important for your reading of the numbers. A review can find improvement in some outcomes while the broader wellbeing result stays uneven.

Among adolescents with Type 1 diabetes, one pilot randomised trial found that glycaemic control increased slightly in the control group and stayed stable in the gratitude group.

The between-group difference at 12 weeks was 6.1 mmol/mol, with a 95% confidence interval from −2.6 to 14.7.

The same result was reported as 0.6%, with a 95% confidence interval from −0.2 to 1.3.

This finding concerns a specific health measure in a specific group. It does not establish a general mental health effect.

One palliative care pilot found no statistically significant pre/post changes for patients. Carers showed significant decreases in depression and total psychological distress scores.

For someone deciding whether gratitude practice fits, the evidence supports a measured trial with modest expectations. A difficult feeling still deserves attention in its own right.

A page should show what it grows from. Here is the tree — every leaf quoted, every receipt attached.

Why do some studies disagree?

Your experience may differ from a friend’s, even when you both write gratitude lists. The research offers several visible reasons for different results.

Studies use different tasks. Some ask people to count blessings. Others ask for a letter, a diary entry, a spoken expression, or a broader positive psychology exercise.

They also use different comparison groups. A neutral activity, a waitlist, a memorable-events task, and a hassles task can create different contrasts.

In a Spanish study, the gratitude group differed from a hassles group in positive affect after the intervention.

The report said the difference seemed mainly influenced by a decrease in positive affect in the hassles group.

That detail changes the story. A group can look better partly because another group feels worse after its assigned task.

Personal response matters too. One study found that ambivalence about emotional expression inhibited the beneficial effect of gratitude on subjective happiness.

If expressing thanks feels forced, private, unsafe, or uncomfortable, the exercise may land differently. The study supports that pattern among people high in ambivalence over emotional expression.

Timing adds another layer. Some findings appear after 2 weeks, 6 weeks, or 12 weeks. A short follow-up cannot answer how long a change lasts.

These limits do not make every result useless. They tell you to match the claim to the exact task, group, outcome, and follow-up.

You have read enough about minds in general. This one maps yours — drawn live from your answers, with a citation under every claim.

What do gratitude studies show about relationships?

If your reason for trying gratitude involves another person, social findings may speak to that moment. Several studies measured connection, shared strength, or relationship-related behavior.

Participants who expressed gratitude in one study reported higher perceived communal strength than participants in all three control conditions at the end.

Another experiment found that grateful customers wrote employee-centered reviews because they wanted to help the firm. That result measured customer behavior in a business setting.

A study of gratitude and school belonging found that increases in gratitude were positively correlated with increases in school belonging among children.

A correlation shows that two changes moved together. It does not prove that one caused the other.

Research among young adults found gratitude significantly associated with happiness, life satisfaction, flourishing, positive affect, spirituality, and meaningful work.

Those findings describe associations. They do not show that gratitude alone produced each outcome.

Workplace studies also report links with job satisfaction, job performance, proactivity, disengagement, and exhaustion. The settings and measures differ across those studies.

For the reader thinking about a message of thanks, the evidence supports a social possibility. Expressed gratitude can coincide with stronger reported connection or helpful behavior in some settings.

It does not require a grand message. The research measures responses and associations, not a rule that every relationship will improve.

Remember the note you left? It has been waiting for you.

What should you expect from a gratitude exercise?

You may want to try the practice without turning your mood into a test. The evidence supports a simple expectation: a possible small shift, with no guaranteed result.

Some studies report higher gratitude after an intervention.

A university study found improved academic motivation after a 2-week online gratitude journal intervention, while the control group did not show that improvement.

A smartphone study found that the average frequency of practising thankfulness increased by more than 120% during the intervention weeks compared with the baseline weeks.

That number measures practice frequency. It does not measure a matching rise in happiness or mental health.

A brief gratitude intervention also reported increases in hedonic wellbeing, optimism, and sleep quality, along with decreases in diastolic blood pressure.

Those results came from that intervention and its participants. They do not set a personal target for your sleep, mood, or blood pressure.

Try noticing what the exercise asks you to do. Does it invite a real memory, a specific person, or a general list?

Notice your response as well. Relief, warmth, boredom, irritation, or no clear change can all tell you how the task lands.

Keep the conclusion close to the measure. If gratitude rises, call that a change in gratitude. If wellbeing rises, call that a wellbeing result.

Read each answer as a limit on the claim. Good research becomes more useful when the result stays attached to its original question.

If any of this pressed on something tender, the help below is real and free, there whenever you need it.

And before this page says its last, fold it into one sentence of your own — the when and the how, decided now.

Keeping faith with every receipt above: here is how the page itself works, device by device.

If part of your situation reaches past this page, the guides below cover the next step directly.

What is the fairest conclusion for this reader?

You arrived looking for numbers that could settle the issue. The strongest answer has two parts: gratitude interventions can help some measured outcomes, and the overall pattern remains uneven.

The clearest pooled figure is Hedges’ g of 0.22 for expressed gratitude interventions and psychological wellbeing. That is a small average effect.

Individual studies report stronger, narrower findings. A university sample showed Cohen’s d of 0.74 for wellbeing after a gratitude intervention.

School students improved in trait gratitude, mental health, and personal and social wellbeing after 6 weeks in one study. A wider school review found gratitude programs generally ineffective overall.

Workers showed improvements in perceived stress and depression across eight gratitude list studies, while wellbeing effects stayed inconsistent.

People who feel unsure about emotional expression may experience less benefit from expressing gratitude. Context, task, outcome, and comparison group shape the result.

So, does the practice hold for this reader? It can be a reasonable way to test a small personal change, especially when the activity feels genuine and manageable.

The numbers do not justify a promise that gratitude will cure distress, improve every health measure, or work equally for everyone. They support curiosity with limits.

That is the research-based position. Keep the claim as precise as the study, and your conclusion will remain as honest as the evidence.

This is general information about the mind, not therapy or a diagnosis. If things feel hard, please consult a professional. In a crisis, reach a free, confidential crisis hotline right away; findahelpline.com lists one for your country.