Does Gratitude Journaling Actually Work?
The founding trial gave the world three good things — and its benefits appeared at one month, not one week, and lasted six only in people who kept practicing. The gratitude visit's larger boost was gone by three months. Across 27 studies and 3,675 participants, the effects on depression and anxiety are small and shrink against active controls, and the meta-analysis's own authors recommend stronger treatments for people who are struggling. What the evidence supports, and what it doesn't.
By MyAudioBooks.ai ·
This free listen has rotated out. The story remains available to read.
The biggest boost in the whole study came from a letter. In two thousand five, five hundred seventy-seven adults signed up through Martin Seligman's Authentic Happiness website to test happiness exercises, and one group was asked to write a letter of gratitude to someone who had been kind to them and never properly thanked — and then to deliver it in person. At the immediate post-test, those participants showed the largest positive changes in the entire experiment. They were happier and less depressed, and the effect held at one week and at one month. By three months, they were back where they started.
The quiet winner was a much smaller exercise: writing down three good things each night, with their causes. It showed no dramatic first-week spike. Its benefits appeared at one month, and they were still there at three months and at six. Our verdict is that both results are true, and together they answer the question this book's title asks. Gratitude journaling works — modestly, slowly, and only as a practice, not a treatment.
At My Audio Books dot A I, you can create your own audiobooks from prompts, turn your documents into audio, all with one subscription, and store your items in your own personal library.
The Founding Experiment
The paper is called Positive Psychology Progress, published in the American Psychologist by Seligman, Tracy Steen, Nansook Park, and Christopher Peterson, and it is the origin of nearly every gratitude journal sold since. The design was more careful than the wellness industry usually admits. Participants were randomly assigned to one of five exercises or to a comparison exercise the researchers considered a plausible placebo: writing about early memories every night for a week. Everyone completed questionnaires measuring happiness and depressive symptoms before the exercise, immediately after the week, and then at one week, one month, three months, and six months of follow-up.
The sample matters too, because it explains both the study's reach and its limits. These were five hundred seventy-seven adults recruited from visitors to a book's website — people already interested enough in happiness to click on exercises for it. Almost two thirds were between thirty-five and fifty-four, about six in ten were women, and they were, on the whole, educated. This was not a clinical sample of people with diagnosed depression. It was a self-selected sample of people trying to feel a little better about their lives, and the results should be read in that population.
It is worth being precise about the instruments too, because everything you have just heard rests on them. Happiness was measured with the Steen Happiness Index, and depressive symptoms with the Center for Epidemiologic Studies Depression Scale — both self-report questionnaires, filled in online. Nobody was scanned, diagnosed, or observed. That does not make the results fake; self-report is the standard way to measure subjective well-being, and subjective well-being is the thing being claimed. It does mean the entire evidence base for the happiness industry is built on people describing their own states, which is exactly why the hard-headed meta-analytic follow-ups matter more than the original enthusiasm.
The exercises themselves are worth listing, because the study was not only about gratitude. Alongside the gratitude visit and three good things, participants wrote about a time they were at their best, identified their signature strengths, and used those strengths in a new way. The comparison structure lets you ask a much better question than does this work: which of these works, for how long, compared with simply journaling about your childhood?
The results split cleanly by duration. Two exercises — three good things and using signature strengths in a new way — increased happiness and decreased depressive symptoms for the full six months. The gratitude visit produced the large immediate effect already described and then faded to baseline by three months. The remaining exercises, and the placebo, produced positive but transient effects. And the detail that should change how you read every seven-day happiness challenge: the degree to which participants actively continued their assigned exercise on their own, beyond the prescribed week, mediated the long-term benefits. The week was not the intervention. The continuation was.
The Letter and the List
Put the two gratitude results side by side and you get a small theory of how these practices work at all. The gratitude visit is an event. You write to a specific person who shaped you, you name what they did, and then you read it to them. It is not really a journaling exercise; it is a relational detonation. The largest immediate effect in the study belongs to it, and it makes sense that a single courageous conversation can move a month of your life. It also makes sense that its effect decays: the conversation ends, and you go back to being a person who does not have it every day.
Three good things is the opposite instrument. It is small, repeatable, and slightly boring, which is precisely why it can become a practice. Each night you write three things that went well and, crucially, their causes. The instruction is not merely to notice pleasant events but to explain them — to practice locating your own role, other people's roles, and the circumstances that produced something good. That is a way of directing attention, night after night, until the direction becomes a habit of perception rather than an assignment.
Notice what this implies about the product being sold to you. A beautifully bound gratitude journal used for seven days is, on this evidence, a delivery system for the transient version of the effect. The durable version came from people who kept writing after the assignment ended. Our read is that the study is less an endorsement of gratitude journaling than an indictment of the way we consume it: as a purchase and a sprint, rather than as a repetition. The letter is worth doing for the month it gives you. The list is worth doing for the person it makes you by month six.
The Original Blessing Counters
The Seligman trial was not the beginning either, and the earlier work is where the modern gratitude journal really comes from. In two thousand three, Robert Emmons and Michael McCullough published Counting Blessings Versus Burdens, a set of three studies in the Journal of Personality and Social Psychology. Their manipulation was disarmingly close to what you have seen on a thousand Instagram posts: participants randomly assigned to focus on what they were grateful for, or on hassles, or on neutral life events, and then keeping weekly or daily records of mood, coping, health behaviors, physical symptoms, and overall appraisals of life.
The first study's results are the ones the wellness industry quotes, so quote them properly. The gratitude group felt better about their lives as a whole and more optimistic about the coming week. They reported fewer physical complaints. And they spent significantly more time exercising — nearly an hour and a half more per week than the hassles group, which is a behavioral effect, not just a mood report. Now the part that gets left out: in that same study, the gratitude condition did not appear to influence global positive or negative affect. Better life appraisal, more optimism, fewer complaints, more exercise — and no detectable change in overall mood. The founding document is already more nuanced than the products descended from it.
The third study is the one that should stop you dismissing the whole literature as a college-student artifact. Emmons and McCullough recruited adults with congenital and adult-onset neuromuscular diseases and randomly assigned them to a gratitude condition or a true control condition, explicitly to broaden who the findings applied to. Across the three studies, the abstract reports heightened well-being on several, though not all, outcome measures, and notes that "The effect on positive affect appeared to be the most robust finding." And the authors' own sentence about their thesis is the one every gratitude influencer should be required to read aloud: the relationship, they write, is "neither inevitable nor unequivocal." That is the founding scientist's own posture. Modest claim, real signal, no guarantees.
At My Audio Books dot A I, you can listen to this story and thousands of others that explore the hidden science and mechanics behind the headlines.
The Meta-Analysis
Sixteen years after the founding trial, the literature was large enough to grade. In two thousand twenty-one, Daniel Cregg and Jennifer Cheavens published a meta-analysis of gratitude interventions in the Journal of Happiness Studies, pooling twenty-seven studies with three thousand six hundred seventy-five participants, asking specifically about symptoms of depression and anxiety. This is the question the wellness marketing most often borrows, so it is the right one to put to the evidence.
Their answer is a small effect size — a standardized measure of how big a difference is — stated plainly. Gratitude interventions reduced symptoms of depression and anxiety by a standardized mean difference of about zero point two nine at post-test and zero point two three at follow-up. Their own summary sentence is blunt: "Our results suggest the effects of gratitude interventions on symptoms of depression and anxiety are relatively modest." Correcting for measurement unreliability did not change them. If you are used to hearing gratitude journaling described as a proven cure for anxiety, zero point two nine is a useful splash of cold water: it is a real signal, and it is small.
Then comes the finding that should discipline every positive-psychology headline. The measured benefits were larger in studies that used a waitlist comparison rather than an active comparison. In other words, part of the apparent power of gratitude interventions depends on what you compare them against. Compared with doing nothing and waiting, journaling looks good. Compared with another activity that might itself be expected to help, the advantage shrinks.
The previous meta-analyses had already shown the same uncomfortable gradient. Cregg and Cheavens describe the two thousand sixteen review by Davis and colleagues, which found effects ranging from zero point three one against a measurement-only control down to minus zero point zero three against a psychologically active comparison — essentially nothing, when the comparison group was doing something credible of its own. And they recall the earlier caution from Wood and colleagues in two thousand ten, who reviewed the intervention literature and concluded that "only a very small number show that gratitude interventions are more effective than genuine controls."
There is one more moderator result in the two thousand twenty-one paper that quietly changes how you should think about the practice. The authors did not find consistent evidence that the effects varied by risk of bias, by the severity of participants' depressive symptoms, or by the type of gratitude intervention used. Read that last part again: the kind of intervention did not reliably matter. The letter, the list, the journal — no technique emerged as the special one. Whatever the active ingredient is, it does not appear to be a proprietary method. It looks much more like the generic act of deliberately directing attention toward what is good in your life, on a repeating basis, which is available to you tonight for free.
The authors' conclusion is the most responsible sentence in this whole literature, and it deserves to be quoted in spirit rather than softened: the effects on depression and anxiety are modest enough that they recommend people seeking to reduce those symptoms engage in interventions with stronger evidence of efficacy. The people who study gratitude interventions for a living are telling you not to use them as your treatment. That is not a hostile reading of the research. That is the research.
The Strongest Case for the Journal
Now the defense, because a fair reading of these documents is not that gratitude journaling is useless. It is that it has been sold for the wrong job. Positive psychology was not founded to treat mental illness; the founding study was measuring happiness and depressive symptoms in a general, self-selected population of people trying to improve their lives. Asking whether gratitude journaling treats depression is a little like asking whether walking cures heart disease and then declaring walking a scam when the answer comes back modest. The intervention was always aimed at well-being, and on well-being the founding trial did find durable change.
The practical case is also genuinely strong. The practice costs nothing, requires no prescription, has essentially no downside risk, and can be done privately at any hour. Most effective psychological interventions cannot say any of those things. A small effect from a free, safe, self-directed practice is not a bad trade; it is a good one, as long as the size of the promise matches the size of the effect. What you cannot afford is the opportunity cost of using it in place of something stronger when you need something stronger.
There is also a detail in the founding trial that makes the comparison question sharper rather than softer. The placebo exercise — writing about early memories for a week — produced positive but transient effects of its own. So did the other non-gratitude exercises. In other words, reflective writing itself appears to do something brief and pleasant for people, regardless of its content. If you feel better after a week of any structured journaling, you are experiencing a real phenomenon; you are just not necessarily experiencing gratitude. The specific claim of the gratitude practice has to be judged on what survives beyond that first reflective flush, which is why the six-month results and the active-control comparisons matter so much more than your first seven days.
There is also a fair methodological point in the journal's favor. When a meta-analysis finds that effects shrink against active controls, that does not prove the intervention is inert. It proves the comparison matters, and it forces a harder question: better than what? For a person doing nothing at all to support their own mental life, a gratitude practice is compared against the waitlist, not against cognitive behavioral therapy. We think the honest verdict has to hold both frames at once: against a credible alternative, the added benefit is doubtful; against the empty evening where your attention otherwise goes to rumination, the benefit is plausible, documented, and cheap.
And one more finding keeps the defense honest rather than comfortable. Cregg and Cheavens describe a study by Lyubomirsky and colleagues that used a more neutral comparison — writing about one's weekly schedule — and found the gratitude intervention had little effect on depressive symptoms. So the evidence is not that journaling always beats doing something ordinary. It is that journaling sometimes beats doing nothing, sometimes beats ordinary activity by a little, and occasionally, in the founding trial's own terms, coincides with improvements that last half a year. That is a real thing. It is just not a miracle.
Who Should Not Reach for the Journal
The most useful chapter in this book is the shortest one to state and the hardest one to sell. If you are trying to reduce symptoms of depression or anxiety, the meta-analysis's own authors recommend interventions with stronger evidence for those symptoms. That means a qualified clinician, a structured therapy with a real evidence base, a proper assessment — the unglamorous machinery of actual care. A journal can sit alongside that machinery. It cannot substitute for it, and the researchers who study it most closely are the ones saying so.
There is a second group who should be careful: people who use the practice to argue themselves out of their own experience. Three good things is an attention exercise, not a loyalty test for your own life. If the practice becomes a nightly instruction to find the good in something that is genuinely harming you, it has stopped being the intervention in the studies and become something else — a discipline of looking away. The founding paper asked people to notice what went well and why. It did not ask them to pretend that nothing went wrong.
And there is a third, quieter failure mode: the person who does the exercise for a week, feels nothing dramatic, and concludes that they are broken. On the founding trial's own timeline, that person stopped before the effect was scheduled to arrive. The benefits of three good things began to appear at one month, not at seven days. If you are going to test it, test it on the timescale the evidence actually used, and judge it by whether you can keep the practice, not by whether the first week felt profound.
How to Actually Do It
Here is the protocol as the documents prescribe it, not as the stationery industry does. Tonight, write down three things that went well, and write their causes. Do not skip the causes; they are in the original instructions, and they are the part that turns a list of pleasant moments into an explanation of your life. The train ran on time because you left ten minutes early. Your colleague helped because you asked. Your kid laughed because you put the phone in another room. That is the exercise: noticing, and then noticing your own and other people's agency in the noticing.
Do it for a week as an experiment, and then decide whether to keep going — knowing that the long-term benefits in the founding trial were carried by the people who continued on their own. If you stop at day seven and judge the whole enterprise, you have tested the version of the intervention that produced transient effects. The evidence for the durable version is literally the evidence of continuation. Give it a month before you evaluate it at all.
Do the gratitude visit too, but do it for the right reason. Write the letter to the person who was kind to you and was never properly thanked, and deliver it in person if you possibly can. Expect a real lift, and expect it to be a month-sized lift rather than a personality transplant. It is an event, not a practice, and its value is not only in your mood scores: the other person gets to hear it. That is a good in itself, whatever the follow-up data says.
And keep the categories straight, because this is the whole book in one sentence. Use the journal as a practice for building a life you notice, and use proper care for symptoms that need treating. A practice, not a treatment. The list is small, the letter is brief, the effects are modest and sometimes durable — and if you keep writing, the founding study suggests you may look up in six months and find that your attention has quietly changed sides.
At My Audio Books dot A I, you can create fiction, non-fiction, and turn your documents into audio, all stored in one place with a single subscription — plus get instant access to thousands of audiobooks and deep-dive investigations. Learn more today at My Audio Books dot A I.