The Three Good Things Exercise: A Simple Positive Psychology Practice for Well-Being

Last Updated May 23, 2026

The Three Good Things exercise is one of the clearest examples of how positive psychology turns the study of flourishing into a daily reflective practice. Developed and tested by Martin Seligman and colleagues, the exercise asks people to write down three things that went well each day and then reflect on why those things happened. Its simplicity is part of its power. The practice does not require specialized equipment, complex theory, clinical language, or intensive training. It asks for attention, interpretation, memory, and repetition.

The exercise became influential because it showed that well-being is not only something researchers can measure after the fact. Under some conditions, it can also be shaped by small, repeated habits that redirect attention toward what is meaningful, supportive, fortunate, beautiful, relational, or effective in ordinary life. Three Good Things helped establish positive psychology interventions as a serious empirical field: structured activities designed to increase well-being, strengthen resilience, or reduce distress through intentional psychological practice.

Yet the exercise should not be misunderstood as forced positivity. At its best, Three Good Things does not deny suffering, injustice, grief, exhaustion, or fear. It does not ask people to pretend that life is easier than it is. Instead, it trains a more balanced form of attention. Human beings often remember threats, frustrations, losses, failures, and embarrassments more readily than small moments of care, progress, beauty, competence, relief, or connection. The exercise intervenes in that imbalance by making positive events visible enough to be named, interpreted, and remembered.

This article examines the origins of the Three Good Things exercise, the psychological mechanisms through which it may work, the evidence supporting its use, its relationship to gratitude and attention, its applications in education and well-being programs, and the limitations that should be kept in view if the exercise is used responsibly.

Origins of the Three Good Things Exercise

The Three Good Things intervention emerged from the early empirical development of positive psychology interventions. In a landmark 2005 study, Seligman, Steen, Park, and Peterson tested several brief exercises designed to increase happiness and reduce depressive symptoms by cultivating strengths, meaning, gratitude, and constructive reflection. Among the most influential was the “three good things” exercise: participants were asked to write down three things that went well each day and explain why they happened.

The intervention mattered because it was both theoretically modest and empirically important. It did not claim to solve the whole problem of suffering. It did not require people to adopt an entire philosophy of life. It asked them to practice a short, structured reflection that could shift the balance of attention. Yet the study found that this simple activity was associated with increased happiness and decreased depressive symptoms for some participants over time.

The exercise became an early flagship example of what positive psychology was trying to demonstrate: well-being can sometimes be strengthened through intentional activity. This was a meaningful departure from approaches that treated well-being mainly as a personality trait, a byproduct of external conditions, or the mere absence of illness. Three Good Things suggested that daily reflection, interpretation, and memory practices could influence how people experience and recall their lives.

It also became influential because it was accessible. Many interventions require specialized training, long sessions, clinical supervision, or institutional infrastructure. Three Good Things can be practiced with a notebook, a phone, a classroom handout, a private journal, or a simple evening routine. This made it attractive for education, workplace well-being, coaching, resilience programs, and individual self-reflection.

That accessibility is also why the exercise requires careful interpretation. Its simplicity does not make it trivial, but neither does it make it universally sufficient. It is best understood as one tool within the broader positive psychology intervention literature: useful for many people, limited for others, and most responsible when framed as a reflective practice rather than a demand for positivity.

Back to top ↑

How the Exercise Works

The structure of the Three Good Things exercise is intentionally straightforward. Each evening, the participant writes down three things that went well during the day. For each item, they also write a short explanation of why it happened. The exercise is often practiced daily for one week in research settings, though many people continue longer as a journaling or reflective habit.

The basic form is simple:

The “why” question is essential. A list of positive events can be pleasant, but causal reflection deepens the practice. Asking why something went well can reveal personal effort, relational support, institutional support, environmental conditions, luck, timing, skill, kindness, care, courage, restraint, or perseverance. This is where the exercise becomes more than positive recall. It becomes an interpretive practice.

For example, “I had a good conversation with a friend” becomes more powerful when followed by “because I made time to call, and they listened generously.” “I finished a task” becomes more meaningful when followed by “because I protected the morning and stayed focused.” “I enjoyed a walk” becomes more specific when followed by “because I slowed down, noticed the weather, and chose not to rush.”

The exercise therefore links positive events to causes. That matters because people often dismiss good moments as accidents while interpreting negative moments as evidence. Three Good Things helps counter that imbalance. It makes beneficial causes visible and invites people to see that good events often arise from relationships, effort, choice, support, environment, and circumstances working together.

Back to top ↑

Why the Exercise May Work

The Three Good Things exercise may work through several overlapping psychological mechanisms. The first is attentional rebalancing. Human attention is highly sensitive to threat, loss, criticism, uncertainty, and failure. This negativity bias is often adaptive: it helps organisms detect danger and avoid repeated harm. But in ordinary life, it can distort retrospective judgment. A day containing several good moments and one frustrating moment may be remembered mostly through the frustration. Three Good Things counteracts that tendency by deliberately directing attention toward what went well.

A second mechanism is positive event salience. Many positive experiences are quiet. They are not always dramatic enough to demand attention. A kind email, a finished errand, a warm meal, a good sentence, a small success, a moment of patience, a friend’s text, a child’s joke, a quiet walk, or an act of restraint may pass quickly. When written down, such events become more cognitively available. The exercise increases their salience.

A third mechanism is causal reflection. The prompt “why did this happen?” encourages people to identify contributing factors. This can strengthen appreciation for relationships, effort, personal strengths, environmental support, and favorable conditions. It can also reduce the tendency to treat good events as meaningless accidents.

A fourth mechanism is gratitude and appreciative awareness. The exercise often reveals that many positive events are not purely individual achievements. They may depend on other people’s care, trust, labor, patience, instruction, humor, generosity, or presence. Recognizing these sources can deepen gratitude and relational awareness.

A fifth mechanism is memory consolidation. Written reflection can help encode daily positive experiences into memory. Repetition may make positive events easier to retrieve, changing the emotional evidence people draw upon when evaluating their lives.

A sixth mechanism is narrative integration. Over time, a person may begin to see daily life not only as a sequence of stressors but also as a field of support, contribution, meaning, and progress. The practice can help build a more balanced life narrative.

These mechanisms help explain why a simple exercise may have measurable effects. The practice is not magic. It works, when it works, by changing what is noticed, how it is interpreted, and how it is remembered.

Back to top ↑

Scientific Evidence

The Three Good Things exercise became widely known because it was tested in controlled intervention research. In the 2005 study by Seligman, Steen, Park, and Peterson, participants assigned to the exercise wrote down three things that went well each day and reflected on why they occurred. The study found increases in happiness and decreases in depressive symptoms for some participants compared with control or comparison conditions. The intervention became one of the most cited examples in the positive psychology intervention literature.

Subsequent research on positive psychology interventions and gratitude exercises has generally supported a cautious conclusion: brief reflective practices can produce modest improvements in well-being for many people, especially when practiced sincerely, repeatedly, and under appropriate conditions. Meta-analytic work has found that positive psychology interventions can increase subjective well-being and reduce depressive symptoms, though effects vary widely by intervention, population, delivery method, duration, baseline distress, and study quality.

The strongest interpretation is therefore balanced. Three Good Things is not a cure-all. It is not equally effective for everyone. It should not be used as a replacement for professional mental-health care when care is needed. But it has enough empirical support to be treated as a serious, low-cost, evidence-informed reflective practice.

Several factors may influence effectiveness:

The research evidence also raises a broader point about positive psychology. Interventions should be evaluated not only by whether they increase positive emotion, but by whether they are meaningful, sustainable, culturally appropriate, ethically framed, and useful in real-world contexts. Three Good Things has scientific value precisely because it can be studied, replicated, adapted, and critiqued.

Back to top ↑

Gratitude, Attention, and Cognitive Framing

Three Good Things is often grouped with gratitude interventions, but it is better understood as a hybrid practice involving gratitude, attention, memory, and interpretation. Gratitude is central because the exercise often makes visible the people, conditions, opportunities, and forms of support that contribute to positive events. A good thing may happen because someone was kind, because a teacher offered help, because a colleague followed through, because a family member showed patience, because a stranger made a task easier, or because a community or institution created the conditions for something to go well.

At the same time, the exercise does not require every good thing to involve another person. Some events arise from one’s own effort, restraint, skill, discipline, patience, curiosity, creativity, or courage. Others arise from nature, chance, beauty, quiet, rest, timing, or relief. This breadth makes the practice flexible. It does not reduce gratitude to social indebtedness. It cultivates appreciative attention toward the many conditions that make good moments possible.

The cognitive framing component is equally important. People do not respond only to events. They respond to interpretations of events. A difficult day can be interpreted as total failure, or it can be seen as a day that contained hardship alongside small moments of progress, support, and endurance. Three Good Things does not erase difficulty. It changes the frame through which the day is reviewed.

This connects naturally to Broaden-and-Build Theory, which proposes that positive emotions can broaden awareness and help build enduring resources. When positive events become more visible, people may become more able to notice possibilities, seek support, remember competence, and sustain hope. The exercise may therefore influence not only mood but also the cognitive field through which future experience is interpreted.

It also connects to meaning and purpose. A person’s “three good things” often reveal what matters: relationships, learning, service, beauty, health, faith, work, care, progress, patience, repair, or rest. Over time, repeated reflection can become a record of values. The exercise can show what a person actually finds life-giving.

Back to top ↑

How to Practice Three Good Things

The practice can be done in a journal, notebook, notes app, classroom worksheet, shared family conversation, coaching session, or private evening reflection. The standard version is written and individual, but the format can be adapted as long as the core elements remain: notice three good things, record them, and reflect on why they happened.

A useful daily entry includes:

• What happened? Name the event clearly and specifically.
• Why did it happen? Identify effort, support, timing, conditions, choice, skill, or kindness.
• What did it reveal? Optional: note what the event says about values, relationships, growth, or meaning.

Examples:

The best entries are often ordinary. The point is not to hunt for spectacular events. It is to notice value in daily life that might otherwise be missed. A small success, a peaceful meal, a repaired misunderstanding, an encouraging sentence, a helpful neighbor, a moment of patience, a completed errand, or a few minutes of quiet can all count.

Timing also matters. Evening reflection is common because it encourages a review of the day before sleep. But other schedules can work. Some people prefer morning review of the previous day. Some families or classrooms may use weekly reflection. Some individuals may adapt the practice during difficult periods by writing “three bearable things,” “three moments of support,” or “three things that helped me get through the day.”

The goal is not perfection. It is disciplined noticing.

Back to top ↑

Applications in Well-Being Programs

Because it is simple, low-cost, and flexible, Three Good Things has been used in many settings: positive psychology courses, resilience programs, coaching, workplace well-being, education, health-support settings, personal journaling, and community programs. Its appeal lies in the fact that it can be practiced individually or collectively, formally or informally, briefly or over a longer period.

In educational settings, it can help students practice reflective attention, gratitude, and emotional literacy. It can also support writing, discussion, and metacognition when used thoughtfully. But it should not become a mandatory positivity ritual that ignores school climate, stress, exclusion, or student distress. In positive education, the exercise works best when embedded in a broader culture of care, belonging, student voice, and developmental support.

In workplace settings, it may support recognition, morale, and reflective balance. But it can easily become problematic if used to mask overwork, insecurity, poor management, or unfair conditions. A workplace should not ask employees to list three good things while ignoring structural stressors that leadership has a responsibility to address.

In health-support or clinical-adjacent contexts, Three Good Things may be useful as a complementary practice for some people, particularly when integrated with appropriate care. But it should not be presented as treatment for serious mental illness, trauma, grief, or acute crisis. The exercise may be supportive for some individuals and inaccessible or invalidating for others depending on timing, severity, and context.

In family or community settings, the exercise can become a shared reflection practice. A family might ask each member to name one good thing at dinner. A community group might use it to open a gathering. A mentor might invite a student to notice progress. These adaptations can preserve the spirit of the exercise while making it relational.

The key distinction is whether the practice is used to widen awareness or to silence difficulty. Used well, it helps people see what is sustaining. Used poorly, it pressures people to ignore what is wrong.

Back to top ↑

Limitations and Critiques

The Three Good Things exercise is useful, but it is limited. It is not universally effective, and it is not appropriate in every context. Some people may find it difficult during periods of acute grief, depression, trauma, burnout, crisis, or instability. When someone is overwhelmed, the instruction to identify positive events may feel hollow, forced, or invalidating if not offered with sensitivity.

A second limitation is that the exercise primarily targets attention and interpretation at the individual level. It cannot solve poverty, discrimination, overwork, institutional instability, unsafe schools, abusive workplaces, disability exclusion, or social isolation. It may help a person notice support or meaning within hardship, but it should not be used to make hardship appear acceptable.

A third limitation concerns repetition. Some people benefit from repeated practice. Others may find that the exercise becomes mechanical. Positive psychology interventions often depend on fit, timing, and motivation. A person who values writing may benefit more than a person who finds journaling burdensome. A person who appreciates structure may benefit more than someone who experiences the exercise as artificial.

A fourth limitation concerns cultural and personal meaning. Some people may prefer prayer, communal storytelling, poetry, gratitude letters, meditation, artistic reflection, or service as forms of appreciative awareness. Three Good Things is one practice among many. It should not be treated as the universal form of gratitude or reflection.

A fifth limitation is misuse by institutions. Schools, workplaces, wellness programs, and organizations may use simple well-being exercises to avoid deeper change. This is one of the major ethical risks surrounding positive psychology interventions. A good practice can become a bad intervention if used to individualize structural problems.

The right conclusion is not that Three Good Things should be rejected. The right conclusion is that it should be used modestly and responsibly. It is a reflective tool, not a universal remedy.

Back to top ↑

Context, Culture, and Responsible Adaptation

Three Good Things should be adapted with attention to context. The practice may look different in a classroom, a workplace, a family, a faith community, a recovery group, or a private journal. It may also need adaptation across cultures, languages, developmental stages, disability contexts, and life circumstances.

For some people, the language of “good things” may feel too bright or emotionally inaccessible. Alternatives may be more appropriate:

• three things that helped today;
• three moments of support;
• three things I endured or handled;
• three signs of care;
• three small moments worth remembering;
• three things I do not want to overlook.

These adaptations preserve the core mechanism while reducing the risk of forced positivity. They can be especially useful in grief, trauma-informed work, disability contexts, or high-stress environments where “good things” may feel too demanding.

Cultural adaptation also matters. Some people may interpret the “why” question individually, focusing on personal effort. Others may interpret it relationally, spiritually, communally, or ecologically. A responsible approach should not force all explanations into individual achievement. A good thing may happen because of family support, God’s mercy, ancestral memory, community care, public infrastructure, land, weather, rest, restraint, or another person’s kindness.

Responsible adaptation means honoring the underlying practice without narrowing its meaning. The point is not to manufacture happiness. The point is to notice what sustains life.

Back to top ↑

A Semi-Formal Framework for Three Good Things

The Three Good Things exercise can be represented semi-formally as an intervention on attentional, interpretive, and memory processes. Let daily well-being at time \(t\) be represented as:

The intervention aims to raise \(P_t\), \(G_t\), and \(C_t\) while partially offsetting the dominance of \(N_t\). This does not mean negative events disappear. It means they become less likely to monopolize the entire interpretation of the day.

A dynamic formulation can express repeated practice:

The “why did this happen?” component can be formalized as:

A simple intervention contrast can be represented as:

A moderation model can also be useful:

These equations do not reduce the practice to mathematics. They clarify its conceptual structure: Three Good Things works, when it works, through attention, interpretation, memory, repetition, gratitude, and contextual fit.

Back to top ↑

Data Design and Measurement Notes

A serious evaluation of Three Good Things should measure more than whether participants completed a journal entry. It should distinguish intervention adherence, positive event salience, gratitude, perceived support, stress load, life satisfaction, affect, depressive symptoms, and personal fit.

Several design principles follow:

• Measure both positive and negative outcomes. Increased gratitude does not automatically imply reduced distress.
• Track practice quality. Mechanical completion may differ from meaningful reflection.
• Assess fit. Some participants may find the exercise helpful, while others may find it artificial or invalidating.
• Include context. Stress, grief, trauma, workload, and social support can shape intervention effects.
• Use longitudinal data where possible. The intervention is based on repetition, so change over time matters.
• Protect privacy. Journal-style data can contain sensitive personal material.

The purpose of measurement is not to turn gratitude into surveillance. It is to understand when reflective practice supports well-being, for whom, through what mechanisms, and under what conditions.

Back to top ↑

R: Modeling Intervention Effects Over Time

The following R workflow illustrates how a researcher might model the effects of Three Good Things in a repeated-measures design. The example estimates change in life satisfaction and depressive symptoms across intervention and comparison conditions while also tracking gratitude, positive event salience, perceived support, stress load, and practice depth.

# Three Good Things intervention modeling workflow
#
# Purpose:
#   Estimate change in life satisfaction and depressive symptoms across
#   a repeated-measures Three Good Things intervention design.
#
# Notes:
#   This workflow is for research, teaching, and exploratory analysis.
#   It is not a clinical, diagnostic, therapeutic, workplace-screening,
#   employment-selection, or individual well-being assessment tool.

library(tidyverse)
library(lme4)
library(lmerTest)
library(broom.mixed)
library(emmeans)
library(performance)

# Expected columns:
# id, day, condition,
# life_satisfaction, depressive_symptoms,
# gratitude_score, positive_event_salience,
# perceived_support, reflection_depth, stress_load

df <- read_csv("data/three_good_things_panel.csv")

panel <- df %>%
  mutate(
    id = as.factor(id),
    day = as.integer(day),
    condition = as.factor(condition)
  ) %>%
  filter(complete.cases(
    life_satisfaction,
    depressive_symptoms,
    gratitude_score,
    positive_event_salience,
    perceived_support,
    reflection_depth,
    stress_load
  )) %>%
  mutate(
    day_c = as.numeric(scale(day, center = TRUE, scale = FALSE)),
    gratitude_c = as.numeric(scale(gratitude_score, center = TRUE, scale = FALSE)),
    salience_c = as.numeric(scale(positive_event_salience, center = TRUE, scale = FALSE)),
    support_c = as.numeric(scale(perceived_support, center = TRUE, scale = FALSE)),
    reflection_c = as.numeric(scale(reflection_depth, center = TRUE, scale = FALSE)),
    stress_c = as.numeric(scale(stress_load, center = TRUE, scale = FALSE))
  )

model_life_satisfaction <- lmer(
  life_satisfaction ~
    day_c * condition +
    gratitude_c +
    salience_c +
    support_c +
    reflection_c -
    stress_c +
    condition:stress_c +
    condition:reflection_c +
    (1 + day_c | id),
  data = panel,
  REML = FALSE
)

model_depressive_symptoms <- lmer(
  depressive_symptoms ~
    day_c * condition -
    gratitude_c -
    salience_c -
    support_c -
    reflection_c +
    stress_c +
    condition:stress_c +
    condition:reflection_c +
    (1 + day_c | id),
  data = panel,
  REML = FALSE
)

summary(model_life_satisfaction)
summary(model_depressive_symptoms)

performance::check_model(model_life_satisfaction)
performance::check_model(model_depressive_symptoms)

emm_life_satisfaction <- emmeans(
  model_life_satisfaction,
  ~ day_c | condition,
  at = list(
    day_c = c(-3, 0, 3),
    gratitude_c = 0,
    salience_c = 0,
    support_c = 0,
    reflection_c = 0,
    stress_c = 0
  )
)

emm_depressive_symptoms <- emmeans(
  model_depressive_symptoms,
  ~ day_c | condition,
  at = list(
    day_c = c(-3, 0, 3),
    gratitude_c = 0,
    salience_c = 0,
    support_c = 0,
    reflection_c = 0,
    stress_c = 0
  )
)

dir.create("outputs", showWarnings = FALSE)

write_csv(
  broom.mixed::tidy(model_life_satisfaction, effects = "fixed", conf.int = TRUE),
  "outputs/three_good_things_life_satisfaction_results.csv"
)

write_csv(
  broom.mixed::tidy(model_depressive_symptoms, effects = "fixed", conf.int = TRUE),
  "outputs/three_good_things_depressive_symptoms_results.csv"
)

write_csv(
  as.data.frame(emm_life_satisfaction),
  "outputs/three_good_things_life_satisfaction_estimated_margins.csv"
)

write_csv(
  as.data.frame(emm_depressive_symptoms),
  "outputs/three_good_things_depressive_symptoms_estimated_margins.csv"
)

practice_summary <- panel %>%
  group_by(condition) %>%
  summarize(
    mean_life_satisfaction = mean(life_satisfaction, na.rm = TRUE),
    mean_depressive_symptoms = mean(depressive_symptoms, na.rm = TRUE),
    mean_gratitude = mean(gratitude_score, na.rm = TRUE),
    mean_positive_event_salience = mean(positive_event_salience, na.rm = TRUE),
    mean_reflection_depth = mean(reflection_depth, na.rm = TRUE),
    mean_stress_load = mean(stress_load, na.rm = TRUE),
    .groups = "drop"
  )

write_csv(
  practice_summary,
  "outputs/three_good_things_practice_summary.csv"
)

This workflow is useful because it models both positive and negative outcomes. A reflective practice may increase life satisfaction without fully reducing depressive symptoms, or it may reduce symptoms only for participants who find the exercise acceptable and practice it with depth. The model also allows stress load and reflection depth to moderate outcomes rather than assuming the intervention works uniformly for everyone.

Back to top ↑

Python: Network Analysis of Gratitude and Well-Being

The following Python example treats Three Good Things as part of a wider gratitude and well-being system. It estimates a sparse partial-correlation network across life satisfaction, depressive symptoms, gratitude, positive event salience, perceived support, reflection depth, and stress load.

"""
Three Good Things network workflow

Purpose:
    Estimate a sparse network of gratitude, attention, support,
    reflection, stress, and well-being variables.

Use:
    Research, teaching, exploratory systems analysis, and intervention
    mechanism design.

Not for:
    Clinical diagnosis, therapeutic decision-making, employment selection,
    workplace screening, or individual well-being assessment.
"""

from pathlib import Path

import matplotlib.pyplot as plt
import networkx as nx
import numpy as np
import pandas as pd

from sklearn.covariance import GraphicalLassoCV
from sklearn.impute import SimpleImputer
from sklearn.preprocessing import StandardScaler

DATA_PATH = Path("data/three_good_things_network.csv")
OUTPUT_DIR = Path("outputs")
OUTPUT_DIR.mkdir(exist_ok=True)

cols = [
    "life_satisfaction",
    "depressive_symptoms",
    "gratitude_score",
    "positive_event_salience",
    "perceived_support",
    "reflection_depth",
    "stress_load",
]

df = pd.read_csv(DATA_PATH)

missing_cols = [col for col in cols if col not in df.columns]
if missing_cols:
    raise ValueError(f"Missing expected columns: {missing_cols}")

imputer = SimpleImputer(strategy="median")
X = pd.DataFrame(imputer.fit_transform(df[cols]), columns=cols)

scaler = StandardScaler()
X_scaled = pd.DataFrame(scaler.fit_transform(X), columns=cols)

X_scaled["appreciative_awareness_index"] = (
    X_scaled["gratitude_score"] +
    X_scaled["positive_event_salience"] +
    X_scaled["perceived_support"] +
    X_scaled["reflection_depth"]
) / 4

X_scaled["net_wellbeing_index"] = (
    X_scaled["life_satisfaction"]
    + X_scaled["gratitude_score"]
    + X_scaled["positive_event_salience"]
    - X_scaled["depressive_symptoms"]
    - X_scaled["stress_load"]
)

glasso = GraphicalLassoCV()
glasso.fit(X_scaled[cols])

precision = glasso.precision_
partial_corr = -precision / np.sqrt(np.outer(np.diag(precision), np.diag(precision)))
np.fill_diagonal(partial_corr, 0)

partial_df = pd.DataFrame(partial_corr, index=cols, columns=cols)

threshold = 0.08
G = nx.Graph()

for node in cols:
    G.add_node(node)

for i, source in enumerate(cols):
    for j, target in enumerate(cols):
        if j > i:
            weight = partial_df.iloc[i, j]
            if abs(weight) >= threshold:
                G.add_edge(source, target, weight=weight, sign=np.sign(weight))

degree = nx.degree_centrality(G)
betweenness = nx.betweenness_centrality(G, weight="weight")

try:
    eigenvector = nx.eigenvector_centrality_numpy(G, weight="weight")
except nx.NetworkXException:
    eigenvector = {node: np.nan for node in G.nodes()}

centrality = pd.DataFrame({
    "node": list(G.nodes()),
    "degree_centrality": [degree[node] for node in G.nodes()],
    "betweenness_centrality": [betweenness[node] for node in G.nodes()],
    "eigenvector_centrality": [eigenvector[node] for node in G.nodes()],
}).sort_values(
    ["eigenvector_centrality", "degree_centrality"],
    ascending=False
)

edge_table = pd.DataFrame([
    {
        "source": source,
        "target": target,
        "partial_correlation": data["weight"],
        "absolute_weight": abs(data["weight"]),
        "sign": "positive" if data["weight"] > 0 else "negative",
    }
    for source, target, data in G.edges(data=True)
]).sort_values("absolute_weight", ascending=False)

centrality.to_csv(OUTPUT_DIR / "three_good_things_network_centrality.csv", index=False)
edge_table.to_csv(OUTPUT_DIR / "three_good_things_network_edges.csv", index=False)
partial_df.to_csv(OUTPUT_DIR / "three_good_things_partial_correlations.csv")
X_scaled.to_csv(OUTPUT_DIR / "three_good_things_scaled_indices.csv", index=False)

print("\nCentrality summary:")
print(centrality)

print("\nStrongest edges:")
print(edge_table.head(15))

plt.figure(figsize=(11, 8))
pos = nx.spring_layout(G, seed=42, k=0.85)

positive_edges = [(u, v) for u, v in G.edges() if G[u][v]["weight"] > 0]
negative_edges = [(u, v) for u, v in G.edges() if G[u][v]["weight"] < 0]

nx.draw_networkx_nodes(G, pos, node_size=1800)
nx.draw_networkx_labels(G, pos, font_size=10)

nx.draw_networkx_edges(
    G,
    pos,
    edgelist=positive_edges,
    width=[abs(G[u][v]["weight"]) * 5 for u, v in positive_edges],
    alpha=0.75,
)

nx.draw_networkx_edges(
    G,
    pos,
    edgelist=negative_edges,
    width=[abs(G[u][v]["weight"]) * 5 for u, v in negative_edges],
    style="dashed",
    alpha=0.75,
)

plt.title("Partial Correlation Network of Three Good Things Variables")
plt.axis("off")
plt.tight_layout()
plt.savefig(OUTPUT_DIR / "three_good_things_network.png", dpi=300)
plt.close()

This type of analysis can reveal whether gratitude, perceived support, reflection depth, positive event salience, stress load, or depressive symptoms functions as a more central feature of the intervention system. That matters because Three Good Things may work through multiple connected pathways rather than through one simple mechanism.

Network models should not be interpreted as causal proof. They are exploratory tools for identifying patterns that may deserve further longitudinal testing, qualitative interpretation, or experimental follow-up.

Back to top ↑

Interpretation and Responsible Use

Three Good Things is a low-risk practice for many people, but responsible use still matters. Reflective exercises can become harmful if they are framed as obligations to feel grateful under conditions of grief, injustice, burnout, illness, abuse, trauma, or institutional neglect. The exercise should never be used to shame people for distress or to imply that suffering results from insufficient positivity.

The code examples above are designed for research, teaching, exploratory modeling, and intervention-mechanism analysis. They should not be used as clinical diagnostic instruments, therapeutic decision tools, workplace-screening systems, employment-selection tools, public-benefits eligibility tools, or individual well-being assessment systems.

Several principles follow:

• Do not force positivity. The exercise should invite reflection, not demand happiness.
• Respect distress. Some people need care, safety, rest, justice, or treatment more urgently than reflective exercises.
• Adapt language. “Three things that helped” may be better than “three good things” in some contexts.
• Protect privacy. Journal entries can contain sensitive personal material.
• Measure context. Stress load, support, grief, trauma, and instability can shape outcomes.
• Avoid institutional misuse. Schools and workplaces should not use the practice to avoid changing harmful conditions.
• Use as one tool among many. Three Good Things can support well-being, but it is not a complete well-being system.

A responsible approach treats the exercise as a modest but meaningful practice of attention and interpretation. It helps people notice what sustains them without asking them to deny what harms them.

Back to top ↑

GitHub Repository

The companion repository for this article organizes the R, Python, data-schema, and documentation materials into a reproducible workflow for studying the Three Good Things exercise. It includes sample data dictionaries, scripts for longitudinal intervention modeling, network-analysis outputs, validation notes, and guidance for responsible interpretation.

Back to top ↑

Conclusion

The Three Good Things exercise remains one of the most accessible and influential practices in positive psychology. Its power lies in a simple but profound shift: instead of allowing positive experiences to disappear unnoticed, the exercise asks people to name them, interpret them, and remember them.

Its scientific importance is equally clear. Three Good Things helped show that brief, structured psychological practices can influence well-being under some conditions. It also helped define the broader field of positive psychology interventions by demonstrating that flourishing can be studied not only as a state or trait, but as something shaped through habit, attention, meaning, gratitude, and memory.

The practice should be used with humility. It is not a cure for suffering, not a substitute for care, and not a remedy for structural injustice. But when practiced sincerely and responsibly, it can help people recover a more balanced relationship to daily life. It reminds us that flourishing often depends not only on what happens, but on what is noticed, how it is understood, and whether the good is allowed to become part of memory.

Back to top ↑

Related articles

• Positive Psychology article map
• Positive Psychology Interventions
• Gratitude in Positive Psychology
• Broaden-and-Build Theory in Positive Psychology
• Meaning and Purpose in Positive Psychology
• Hope Theory in Positive Psychology
• Flow and Optimal Experience in Positive Psychology
• Positive Education
• The Science of Flourishing: How Positive Psychology Measures Well-Being

Further reading

• Emmons, R.A. and McCullough, M.E. (2004) The Psychology of Gratitude. New York: Oxford University Press.
• Lyubomirsky, S. (2007) The How of Happiness. New York: Penguin Press.
• Peterson, C. (2006) A Primer in Positive Psychology. New York: Oxford University Press.
• Seligman, M.E.P. (2011) Flourish. New York: Free Press.
• Snyder, C.R. and Lopez, S.J. (eds.) (2009) Oxford Handbook of Positive Psychology. 2nd edn. Oxford: Oxford University Press.

Back to top ↑

References

• Bolier, L. et al. (2013) ‘The effectiveness of positive psychology interventions: A meta-analysis of randomized controlled studies’, BMC Public Health, 13, 119. Available at: https://doi.org/10.1186/1471-2458-13-119.
• Emmons, R.A. and McCullough, M.E. (2003) ‘Counting blessings versus burdens: An experimental investigation of gratitude and subjective well-being in daily life’, Journal of Personality and Social Psychology, 84(2), pp. 377–389. Available at: https://doi.org/10.1037/0022-3514.84.2.377.
• Peterson, C. (2006) A Primer in Positive Psychology. New York: Oxford University Press.
• Positive Psychology Center (n.d.) Positive Psychology: The science of the factors that enable individuals and communities to thrive. Available at: https://ppc.sas.upenn.edu/.
• Seligman, M.E.P. (2011) Flourish. New York: Free Press.
• Seligman, M.E.P., Steen, T.A., Park, N. and Peterson, C. (2005) ‘Positive psychology progress: Empirical validation of interventions’, American Psychologist, 60(5), pp. 410–421. Available at: https://doi.org/10.1037/0003-066X.60.5.410.

Back to top ↑