Skip to content

Character Education

Principle · All principles
Evidence · 27 claims (18 for, 8 mixed, 1 against) · 13 studies (5 theoretical, 3 quant-synthesis, 2 causal, 2 associational, 1 qualitative), q1–q4 · 4 of 13 report an effect size · 26 claims rest on one study

Conditional relationship

For school-age learners (child, adolescent, pre-k-12), and in a narrower form for university students (postsecondary), character is expected to develop when a school or programme makes ethical understanding, caring response and concrete moral action part of everyday school life over a term-plus horizon: the same relationships and conflict-resolution routines across staff, regular democratic class meetings, service or collaborative projects tied to reflection, and moral problems worked through in specific situations inside ordinary subjects. That is expected to change observed behaviour (behavioural outcomes: conduct, violence, absence) and civic commitment more than a separate civic or moral subject on its own, or than teaching a list of rules by recitation, example, reward or punishment. The target is attitude-motivation and a complex-skill (judging and acting in a real situation), not the ability to state the right answer to a hypothetical dilemma.

The relationship is conditional on breadth (one enthusiastic classroom inside a punitive or competitive school is expected to reach less far), on whether learners act as well as discuss, on whether reflection connects the action to a moral question, and on which outcome is counted: the programme evaluations below find behaviour effects more often than effects on knowledge, attitudes and values or on achievement. No claim in the wiki compares this integrated, schoolwide approach with a separate character or civic course for the same learners. What the wiki has is (a) What Works Clearinghouse reviews of three branded programmes, each against usual practice or a narrower condition, and (b) philosophical arguments (Dewey, read by Liu 2014; the Naval et al. 2022 review of university civic learning) for why the integrated form should be preferred. So the design below is a default, labelled step by step.

This page folds in two thin pages (2026-10-07): one recommending character education grounded in virtue development to build civic competence at university, and one recommending, after Dewey, that character be cultivated indirectly through all of school life rather than in separate moral education courses. Their recommendations appear as steps 3, 5 and 7 and as rows in the situation table.

How converted siblings sit beside this: the Social-Emotional Learning pattern holds the sequence for explicitly teaching named social and emotional skills one at a time; character education overlaps with it in its programmes (Positive Action, Caring School Community) but owns the moral and civic side (judgement in situations, care, service, democratic participation). Community-Based Learning holds the conditional model for giving coursework a real role in a community; this page uses service as one of its means and does not repeat that model. Effective Classroom Management Plan holds the classroom-wide behaviour plan; this page's conflict-resolution framing of discipline sits inside it. Building Empathy holds structured perspective-taking.

Default design, while the relationship is untested

The earlier pages' guidance, kept as a concrete default a designer can act on and revise. Each step is a proposal unless a claim is linked beside it; quantities are untested proposals.

  1. Choose the scope first: a school, a grade or year group, or a programme, not one class. Agree a short set of shared commitments (cooperation, care, fairness) with all adults who meet the learners, including support staff, and how each will respond to a conflict. Schoolwide programmes with several components have WWC-reviewed evidence on behaviour against usual practice (Positive Action, behaviour [+M]; Caring School Community, behaviour [+M]); whether breadth itself is the active part is untested.
  2. Frame discipline as conflict resolution. When a rule is broken, the response is a short conversation about who was affected and what would repair it, not a punishment alone; distinguish moral matters (harm, fairness) from conventions (lining up, raising hands) when you explain a rule (Turiel's Social Domain Theory). Untested proposal, from the earlier page.
  3. Work moral problems in specific situations, not fixed virtue lists. Once a week (a proposal: 20–30 minutes), take a real problem from the class, school or a text, ask what each option would do to whom, and decide; teach a virtue in relation to the situation where it applies, and let learners see two virtues pull against each other. From the folded Dewey page: moral knowledge is a product of moral inquiry [+W] and virtues are taught in relation to specific situations [~W] (both philosophical arguments, no data). Kohlberg-style dilemma discussion is the tested neighbour: a meta-analysis of 55 studies is reported second-hand to show a moderate to significant effect (dilemma discussion [+W]); Liu argues for real problems over hypothetical dilemmas.
  4. Hold regular class meetings where learners help set rules and settle disagreements. A proposal for dose: weekly, 15–30 minutes, with a standing agenda (appreciations, one problem, one decision, a check on last week's decision). Early in the year the teacher chairs and models; by mid-year learners chair in rotation with the teacher as a participant. Untested proposal; the earlier page names democratic practice as "a significant, recurring mechanism", and no claim isolates it.
  5. Let learners act, not only discuss: tie a service or collaborative project to the curriculum and to reflection. One project a term (a proposal), with a defined need, a role for every learner, and a written or spoken reflection that links what happened to a moral question from step 3. A WWC review of one quasi-experiment found an ethics curriculum plus service learning ahead of senior English classes on knowledge, attitudes and values (Building Decision Skills [+M]), and ahead of service learning alone only on ethical awareness and ethical perspective (curriculum adds beyond service alone [~M]). Service learning in integrated projects is reported second-hand with self-reported gains (service learning [+W]); in Chicago high schools, service learning and in-class civic learning were the experiences most strongly associated with civic commitment two years later, controlling for prior commitment (Sporte & Kahne [+M], associational).
  6. Weave the moral question into ordinary subjects. In literature, history and science, teach the subject itself and, where the material raises one, name the question about the common good it raises; do not attach a moral lesson to every text. From the folded Dewey page and the university page: a dedicated civic subject gives necessary knowledge but is argued to be insufficient alone [+W] (authorial argument); didactic character education is criticised for de-emphasising reflection and judgement [+W].
  7. Make the adults the model. At university (the folded civic-competence page), staff and professors who hold a social view of their own profession act as role models; build leadership, teamwork and prosocial behaviour into curricular, extracurricular and research work. In school, adults show the conflict-resolution routine in their own disagreements. Untested proposal.
  8. Plan the arc across a year and hand over. Early sessions (first 6–8 weeks, a proposal): the teacher chairs meetings, names the moral question in each problem, and structures reflection with two or three prompts. Middle: learners chair and propose problems; prompts drop to one. Late: learners run a project and its reflection with the teacher on request. Move a group to the next stage when most meetings reach a decision the group then keeps without adult enforcement. The point of the arc is that reflection becomes a habit, which is argued to need supportive social arrangements (reflective methods internalised as habit [+W], theoretical). Thresholds untested.
  9. Count behaviour and civic commitment separately from knowledge, attitudes and achievement. Record conduct incidents, absence and participation, and survey commitment and attitudes, at the start and at the end of each year. Programme evaluations found behaviour effects without discernible effects on knowledge, attitudes and values or on achievement (Caring School Community: knowledge, attitudes and values [~M]; academic achievement [~M]), so one outcome does not stand for the others.

When a condition is not met: if the school will not adopt a shared approach, run steps 3, 4 and 6 in the classroom you control, expect a smaller reach, and say so; if there is no time for a project, use a classroom responsibility (a role learners hold for the whole term) as the action in step 5.

Fitting the design to a situation

The three facts that most change the decision:

  • How far the approach reaches. One classroom, a whole school, or a programme with its own curriculum. If unstated, ask: which adults will meet these learners each week, and will they respond to a conflict in the same way?
  • Which outcome counts. Conduct and violence, civic commitment, knowledge and attitudes about ethics, and achievement moved differently in the programmes reviewed. If unstated, ask: what would you see learners doing differently at the end of the year, and how will you record it?
  • The learners' age and setting. Primary, secondary and university evidence comes from different programmes and arguments. If unstated, ask: how old are the learners, and in what kind of institution?
If the brief says… Then change… Basis
Whole school, primary age (child), goal is fewer conduct problems Adopt a schoolwide programme with several components (classroom lessons, schoolwide climate, family involvement) and keep it for more than a year; record conduct, violence and absence from the start Positive Action, behaviour [+M] (two elementary studies, 56 schools, WWC domain average 0.51); Caring School Community, behaviour [+M] (two K–6 studies, +8 percentile points, rated with reservations)
The goal is reading and maths scores Do not plan character education as the route to higher test scores; if you add it, keep academic time intact and measure achievement separately Caring School Community: no discernible academic effects [~M]; Positive Action: Hawaii trial found no significant state-test difference [~M] (although the WWC domain average for achievement was positive, 0.37 [+M])
Secondary age (adolescent), goal is ethical judgement Pair a short ethics curriculum (deciding between competing values) with a service project and reflection; the curriculum is what adds ethical awareness and perspective over service alone curriculum plus service vs service alone [~M] (one quasi-experiment, twelfth graders)
Goal is self-esteem or general social responsibility through service Keep the service, but do not promise these outcomes; measure them if they matter Building Decision Skills: no significant effect on self-esteem, general social responsibility or anticipated participation [-M] (qualifies the second-hand service-learning gains [~W])
University students (postsecondary), goal is civic competence Keep a civic subject for its knowledge, and add service learning, project work and staff role models across courses and extracurricular life civic subject alone insufficient [+W] (authorial argument, Naval et al. review); service learning [+W] (reported second-hand)
Boys and girls may respond differently, or the target is violence Check results by group; do not assume a whole-group effect holds for each Positive Action reduced serious violence among boys but not girls [~M] (one grade 5 trial, Hawaii)
Only one teacher, one classroom, the school is punitive or competitive Run weekly class meetings and situational moral problems inside the class; keep the conflict-resolution response for your own room; tell families what you are doing; expect less reach untested proposal, from the earlier page's constraint on single teachers
Time is a single session or a short unit only Do not call it character education; use one real problem from the learners' own setting, one decision, and one reflection, and record it as a single activity untested proposal; the programmes reviewed ran for a year or more
Learners join remotely, some in the room, or phone-only Hold class meetings with a turn order and a shared written agenda; let remote learners propose problems in writing beforehand; make the service or responsibility something they can do where they are untested proposal
One adult for a large group, or scarce facilitator time (volunteers, a youth worker with a short slot) Rotate learner chairs early; keep a fixed 15-minute meeting format; let learners keep a short decision log the adult reads weekly untested proposal
Adults in a work setting (staff, professional ethics) Use real cases from their own work and decide on them; the profession's own code is the shared commitment; no class meetings untested proposal; the claims here test school-age learners or argue about universities
Learners have low confidence or low literacy, or the language of instruction is not their first Let learners show care and decisions by doing (roles, service, drama) before writing reflections; accept spoken or drawn reflections untested proposal
The brief names its own programme, rubric, values list or meeting routine Keep it; change only what the evidence here says matters: add action with reflection, make problems specific, and measure behaviour separately from attitudes untested proposal (wave-5 rule); moral inquiry over fixed principles [+W]
Problems discussed are private or sensitive (a family matter, a conflict between named learners) The learner decides whether it is discussed openly; use anonymised or fictional versions in meetings; repair conversations stay with those involved untested proposal
The brief wants a public product (a service project for a real partner, a performance) Keep the product whole and real; the reflection happens alongside, not as a replacement test untested proposal; see Community-Based Learning
High stakes: discipline decisions, exclusion Keep a record of conflicts and their repair; do not grade character untested proposal

Six of the sixteen rows rest on a claim from a programme evaluation; all of those come from WWC reviews of one or two studies per programme, in the US, in the age band named. The university row rests on an argument and a second-hand report. The rest are untested proposals.

Observation, state and explanation

What can be observed: conduct incidents and how they ended, attendance, whether a class meeting reached a decision and whether the group kept it, what learners said or wrote when asked what an option would do to whom, what they did in a service project, and survey answers about commitment. "Has good character", "cares about others" and "reasons morally" are inferred states. A learner who gives the right answer in a meeting has shown what they will say in that room, not what they will do in the corridor.

Response Plausible explanations Observation that could change the interpretation
Conduct incidents fall, survey attitudes do not change Behaviour changed through routines and adult consistency, not through belief; the survey does not measure what changed; ceiling on attitudes Ask learners to explain a recent repair conversation; compare classes with stronger and weaker routines.
Learners give model answers in meetings, behaviour unchanged Learners know what the adults want to hear; the problems are hypothetical, not theirs; no action follows a decision Switch to a real class problem with a decision the group then has to keep; watch whether it is kept.
A service project is enjoyed, reflection is thin Reflection prompts are generic; the moral question was never named; the project felt like an outing Ask one specific question (who was helped, who was left out, what would you do differently) and compare answers.
One group improves, another does not (for example boys and girls) Different starting rates; the programme content fits one group's concerns; measurement differs Check baselines by group and ask each what the programme felt like to them.
Effects fade the next year Routines stopped when staff changed; habit not yet internal; new setting without the same arrangements Record which routines continued and compare learners who moved to settings with and without them.

These rows are proposals, untested with learners as diagnostics. Keep the explanations that remain open rather than choosing the one whose name fits.

Evidence and qualifications

No claim in the wiki tests the integrated, schoolwide form against a separate course, or tests class meetings or conflict-resolution discipline on their own. The claims below test branded programmes against usual practice, or argue for the integrated form.

Learners, programmes and outcomes differ across these claims (US elementary schools, twelfth graders, Chicago high schools, universities), so do not rank them by effect label. All programme evidence is from WWC reviews of one or two studies each, against usual practice.

The evidence by question

Further evidence, not yet read against this model

Objective and learner-valued goal

The designer's objective is usually observed conduct and civic participation over a year, sometimes ethical reasoning or attitudes. Elicit separately what learners want: to be treated fairly, to have a say in rules, to stop a conflict that affects them, to do something that matters to people outside school. Agreement and divergence are both observations. A learner who values having a say will judge the programme by whether meeting decisions are kept; one who values fairness may read a schoolwide values list as one more set of adult rules.

For example, a secondary school may aim at fewer exclusions, while a group of learners wants the canteen queue to be fair. Taking the queue as a class-meeting problem, deciding a rule and keeping it serves both; record whether exclusions changed and whether the learners thought the decision was theirs.

What would revise this model?

The expectation should weaken if schoolwide programmes with class meetings, service and conflict-resolution discipline show no more change in behaviour or civic commitment than a separate character course with the same learners, or if the behaviour effects found by WWC reviews disappear in independent replications. If curriculum alone matches curriculum plus service, step 5's action component is a preference, not a requirement. If attitude and value measures keep showing no discernible effect while behaviour changes, the model should say it changes conduct and say less about character. Do not protect the model by calling every null "not schoolwide enough" after the fact.

Behaviour, civic commitment, ethical reasoning, attitudes, achievement and what learners do after they leave are separate claims. The present evidence establishes no dose for class meetings, no project length and no stage-change criterion.

Claims

Examples

Key Sources

  • Elkind, D., & Sweet, F. (2004). How to do character education. Retrieved from http://www.goodcharacter.com/Article_4.html
  • Berkowitz, M., & Bier, M. (2006). What works in character education: A research-driven guide for educators. Center for Character and Citizenship.
  • Narvaez, D. (2009). Moral complexity: The fatal attraction of truthiness and the importance of mature moral functioning. Perspectives on Psychological Science, 5(2), 162-181. doi:10.1177/1745691610362351
  • Nucci, L. P. (2009). Nice is not enough: Facilitating moral development. Merrill/Prentice Hall.
  • Minow, M., Shweder, R., & Markus, H. (Eds.). (2008). Just schools: Pursuing equality in societies of difference. Russell Sage Foundation.
  • Arduini-Van Hoose, N. (2020). Moral development. In Educational psychology. Retrieved from https://edpsych.pressbooks.sunycreate.cloud. CC BY-NC-SA 4.0.
  • Naval, C.; Villacís, J.L.; Ibarrola-García, S. (2022). The Transversality of Civic Learning as the Basis for Development in the University. Educ. Sci. 2022, 12, 240. https://doi.org/10.3390/educsci12040240
  • Xiangdong Liu. (2014). The Problem of Character Education and Kohlberg's Moral Education: Critique from Dewey's Moral Deliberation. Philosophical Studies in Education 45. https://www.ovpes.org/
  • Stewart Williams, J. E. (2020). Daisaku Ikeda's Philosophy of Value-Creating Global Citizenship Education and Africana Humanism: Africa as the Continent of the 21st Century. Journal of Interdisciplinary Studies in Education, 9(SI). https://ojed.org/jise
  • Sankaranarayanan Paleeri. (2014). Setting Objectives of Value Education in Constructivist Approach in the Light of Revised Blooms Taxonomy (RBT). i-manager's Journal of School Educational Technology. https://www.imanagerpublications.com
  • Norberg, K. (2003). Constitutive Values and Daily Practice in a Swedish School. Values and Ethics in Educational Administration, 1(3). http://www.ed.psu.edu/uceacsle/
Page metadata
Field Value
type principle
id character-education
aliases character-education-promotes-civic-competence, indirect-moral-education-through-school-life, nurture-ubuntu-spirit-students, integrated-constructivist-value-education-transaction, moral-practice-must-be-deliberative
title Character Education
description For school-age and university learners, making ethical understanding, caring response and moral action part of everyday school life (shared conflict-resolution routines, class meetings, service tied to reflection, moral problems in specific situations across subjects) is expected to change behaviour and civic commitment more than a separate civic or moral course or rule recitation; WWC reviews test three branded programmes against usual practice, finding behaviour effects more often than attitude or achievement effects, but no claim tests the integrated form against a separate course.
status review
canonical True
generated by claude/unspecified on 2026-10-14
trust unverified — no one has confirmed this page's content
sources 5 — naval-2022, xiangdong-liu-2014, stewart-williams-2020, sankaranarayanan-paleeri-2014, norberg-2003

Frontmatter is the page's machine-readable record — see the Schema & Guide.