This guide explains the ethical rules a teacher must follow when doing action research with learners in Department of Education (DepEd) schools. DepEd has no single research ethics order yet. The Research Management Guidelines, DepEd Order (DO) No. 16, s. 2017 promised a Research Ethics Committee and a policy on research ethics, and “in the interim” tells researchers to follow the DepEd Child Protection Policy (DO 40, s. 2012) when interacting with learners. The rules therefore come from several sources: the Research Management Guidelines, the Child Protection Policy, the policy on Indigenous Peoples education, the research portal order, the 2026 order on artificial intelligence (AI), the Data Privacy Act, and DepEd’s Action Research Toolkit (2019).

It is written for teachers planning research with learners, for school heads who endorse and supervise it, and for teachers preparing for the National Assessment for School Heads (NASH). It includes a sample consent letter, a sample assent script and a checklist, and ends with two practice items.

Key points

  • Ethics is checked first. A proposal that does not conform to research ethics is returned at initial screening, and anyone found to have violated research ethics “will be blacklisted from availing of any other research grant mechanism in the Department”.
  • The toolkit sets five protocols. Reduce the risk of harm, secure informed consent, safeguard confidentiality and anonymity, avoid deceptive practices, and ensure justice and fidelity.
  • Parents consent; learners assent. For learners, written informed consent comes from the parent or guardian; asking the learner’s own agreement, in words the child understands, is good practice and respects the child’s right to be heard.
  • Learners’ data stay confidential. Use codes instead of names, store records securely, and never enter learners’ personal information into an AI tool.
  • Research can surface abuse. If a learner discloses bullying or abuse during research, the Child Protection Policy applies: report it to the school head and the Child Protection Committee.
  • Indigenous communities give consent as communities. Research involving Indigenous Peoples (IP) learners and communities needs free, prior and informed consent through the community’s own processes, and the community co-owns research on its knowledge.

Where DepEd’s research ethics rules come from

Source What it requires
DO 16, s. 2017, Research Management Guidelines The highest ethical standards; “confidentiality in handling data of respondents and providing consent forms where necessary”; blacklisting for ethics violations, plagiarism and fraud; declarations of anti-plagiarism and absence of conflict of interest
DO 40, s. 2012, Child Protection Policy The interim reference for interacting with learners; the school Child Protection Committee; reporting of bullying and abuse
DO 62, s. 2011, National IPEd Policy Framework, with Annex 9 of DO 16 A rights-based approach for research with IP learners and communities; free, prior and informed consent; communal intellectual and cultural property
DO 14, s. 2022, E-Saliksik research portal Consent forms, agreements and photos left out of archived reports; only abstracts of studies on sensitive topics and vulnerable groups made public
DO 003, s. 2026, AI in basic education AI only for support tasks in research; no personal or sensitive information in AI systems; declaration of AI tools
Republic Act No. 10173, Data Privacy Act of 2012 The legal protection of personal information, which DO 003, s. 2026 cites for research data
DepEd Action Research Toolkit (2019) Three principles and five protocols for action research in DepEd

Three principles of DepEd research

The Basic Education Research Agenda (DO 39, s. 2016) and the Action Research Toolkit set out three principles:

  • Excellence. A topic that is relevant and researchable, appropriate methods, and findings “logical, coherent, and supported by data”.
  • Integrity. The highest ethical standards, especially when people take part: no harm, informed consent, no unusual risk, and “honesty and accuracy in the collection and analysis of data and the reporting of results”.
  • Openness. Collaboration with partners, and accurate and timely sharing of results “within the limits of confidentiality, safeguarding anonymity of participants”.

The five protocols for action research

Protocol 1: Reducing the risk of harm

Action research “should not bring harm to participants”. The toolkit lists the kinds of harm to guard against: “physical harm, psychological distress and discomfort, psychosocial disadvantage, financial problems, and invasion of privacy and anonymity”. If a study may cause discomfort, there must be a strong reason for it, written into the consent form with a plan to reduce the harm or to debrief participants.

In a classroom, harm is often quiet: a learner singled out as a “struggling reader” in front of classmates, a survey that asks about family income, an interview that brings up a painful experience, or learners in a comparison group who fall behind. Plan for each.

Informed consent is “a document that informs participants what their involvement in the study entails and what sensitive information might have to be asked from them”. It includes the purpose of the research, the methods, the possible outcomes, and “the demands, discomforts, inconveniences, and risks that the participants may experience”. For learners, it is signed by a parent or guardian.

Protocol 3: Safeguarding confidentiality and anonymity

“At no point should their identities and other personal information about them be divulged to other individuals without their permission.” Documents that bear names and sensitive information must be kept “in a secure place that is accessible only to the researcher/s”, and may be shredded “after one year of research dissemination”; dumping them in trash bins is “improper”.

Protocol 4: Avoiding deceptive practices

Researchers must make participants understand the objectives of the study and what they will do. If some information must be withheld to get valid results, participants “should eventually be debriefed of the true objectives of the study”. In school action research there is rarely a reason to withhold anything.

Protocol 5: Ensuring justice and fidelity

Justice is fairness: the researcher “adheres to the standards of impartiality, equality, and reciprocity”. In a study with a comparison group, justice means the comparison group keeps its regular lessons, and receives the intervention afterward if it works. Fidelity “refers to being loyal and truthful to the respondents by keeping promises to them such as the anonymity of the respondents’ identities and respecting their dignity as persons”.

Learners are minors, so their parents or guardians give informed consent. Asking the learners themselves, in simple words, whether they agree to take part (assent) is good practice in research with children and fits the Child Protection Policy’s call to ensure that “children’s right to be heard are respected”. A consent form should tell parents:

  1. who is doing the study, and why;
  2. what their child will do, for how long, and when;
  3. what information will be collected, and how it will be kept and used;
  4. any discomforts or risks, and how they will be handled;
  5. that taking part is voluntary, that they may withdraw at any time, and that saying no will not affect the child’s grades or treatment;
  6. that the child’s name will not appear in any report; and
  7. whom to contact with questions.

Dear Parent or Guardian,

I am your child’s Grade 1 adviser. This school year, I am doing a study to find out whether a daily fifteen-minute letter-sound activity helps Grade 1 learners learn to read. The study has been reviewed by our school head and will be submitted to the Schools Division Research Committee.

If you agree, your child will join the activity during the independent work period every school day for ten weeks, and will take two short reading tests, one in June and one in September. You may be invited to a short meeting with other parents in September.

Your child’s test scores will be recorded with a code, not a name, and kept in a locked cabinet. No child’s name or photo will appear in the report. There are no risks beyond those of ordinary class activities. If your child seems tired or upset, the activity will stop for that day.

Taking part is voluntary. You may say no, or stop at any time, and this will not affect your child’s grades or how your child is treated in class. If you do not agree, your child will continue with regular lessons.

If you have questions, please talk to me or to our school head.

I agree to let my child, ____________________, take part in this study.

Name and signature of parent or guardian: ____________________ Date: __________

A sample assent script for young learners (ours)

“I am trying a new way of learning letter sounds. I want to know if it helps children read. If you want, you can join a small group with me for a short time every day. We will play with letter cards and read short words. You can say no, and you can stop any time you want. Nobody will be angry. Do you want to join?”

Record each learner’s answer. A learner who says no is not included, even if the parent agreed.

Protecting learners’ data

  • Collect only what the study needs. A reading study does not need a learner’s family income.
  • Use codes, not names. Keep the key that links codes to names separate from the data, and in a secure place.
  • Store records securely. Paper in a locked cabinet; files on a password-protected device, not on a shared computer or a public drive.
  • Keep AI tools away from personal information. DO 003, s. 2026 states that “the input, processing, or storage of personal and sensitive information within AI systems is categorically prohibited”. AI may help only with support tasks such as analyzing processed data, refining grammar and checking citations, and every AI tool used must be declared.
  • Remove names before you share data. The Research Management Guidelines ask for an executive data set “for accountability and reproducibility”; submit it with codes only.
  • Keep supporting documents out of the report. E-Saliksik excludes consent forms, agreements and photos from archived reports; keep them in a separate file for the research committee.
  • Dispose of records properly. The toolkit allows shredding of identifying documents after one year of research dissemination.

Learners’ personal information is protected by the Data Privacy Act of 2012 (Republic Act No. 10173). When in doubt, ask your schools division office, or DepEd’s Data Privacy Office, before collecting or sharing personal data.

When research touches sensitive topics

Some research questions, such as studies of bullying, abuse, family problems, substance use or mental health, can harm learners if handled badly, and can surface cases the school must act on.

  • Know your reporting duty. Under DO 40, s. 2012, a bullying incident that comes to the notice of school personnel “shall be immediately reported to the School Head”, and the school’s Child Protection Committee identifies, refers and reports cases of abuse. A researcher who learns of abuse or bullying is school personnel first: the research does not suspend that duty, and the consent form should say so.
  • Plan support. Arrange with the guidance counselor or teacher, who is the Child Protection Committee’s vice chairperson, before you ask learners about painful experiences.
  • Ask only what you need. Prefer anonymous questionnaires to interviews for sensitive topics, and never ask a learner to describe an incident in front of others.
  • Expect limited disclosure. Studies on topics such as “trauma, or violence” or the experiences of “potentially vulnerable populations” are classified for limited public disclosure in E-Saliksik, with only the abstract posted.

Research with Indigenous Peoples learners and communities

The Research Management Guidelines add specific duties for research involving IP learners, Indigenous Cultural Communities (ICCs), Indigenous Knowledge Systems and Practices (IKSPs) and Indigenous Learning Systems (ILSs):

  • Community consultation at screening. The research committee’s secretariat discusses the proposal with “the recognized community representatives or elders” to validate its compliance with ethical requirements.
  • Free, prior and informed consent. Consent is sought from the community “prior to the planning of the research”, following “the community’s customary governance processes”, free of coercion and of any implied favors. The community decides the form of consent, and may revoke it if it feels its rights are violated.
  • Co-authorship and co-ownership. If the research focuses on a community’s IKSPs and ILSs with information generated from the community, the community is recognized “as co-author and co-owner of the research”, and its permission is needed to present or publish the study.
  • Communal property. IKSPs and their elements “shall remain as the communal property of the ICCs and shall not be owned or patented by the researchers or the Department of Education”.
  • Sharing back. A copy of the study and later publications is given to the community representatives.

Plagiarism, fraud and conflict of interest

Every proponent signs a declaration of anti-plagiarism, attesting that the proposal and final report are original and properly cited, and a declaration of absence of conflict of interest. The Research Management Guidelines define plagiarism as “intellectual theft”, and conflict of interest as situations in which “financial or other personal considerations may compromise” a researcher’s judgment. A conflict of interest is a ground for returning a proposal at screening, and anyone found to have committed plagiarism or fraud is blacklisted from DepEd’s research grants.

Under DO 003, s. 2026, “the submission of research papers generated, in whole or in part, by AI without significant human contribution and proper disclosure is strictly forbidden and shall be subject to applicable administrative sanctions”.

An ethics checklist before you collect data

  • The proposal describes the risks to learners and how they will be reduced.
  • Consent forms for parents and an assent script for learners are ready, in a language they understand.
  • Participation is voluntary, with no effect on grades, and learners may stop at any time.
  • Learners in a comparison group keep their regular lessons and will receive the intervention later if it works.
  • Data will be recorded with codes; the code key and signed forms will be stored securely.
  • No personal information will go into an AI tool, and any AI use is declared.
  • For sensitive topics, the guidance counselor and the Child Protection Committee know about the study, and there is a plan for disclosures.
  • For IP learners and communities, the community’s consent has been sought through its own processes.
  • The plagiarism and conflict of interest declarations are signed.
  • The school head has read the proposal before endorsing it.

What the school head does

  • Read for ethics before you endorse. The supervisor’s endorsement certifies the teacher’s capacity to do the study; a proposal with ethics problems will be returned.
  • Connect research to child protection. As chairperson of the school’s Child Protection Committee, make sure studies on sensitive topics have a plan for support and reporting.
  • Protect the comparison group. Agree with the teacher that learners who did not receive an intervention will get it later if it works.
  • Keep records secure. Provide a locked place for consent forms and identifying records, and a rule against sharing learners’ data on personal devices and in group chats.
  • Model honest reporting. Encourage teachers to report results as they are, including studies that did not work.

Common mistakes

  • No consent because “it is just a class activity”. If learners’ work or scores will be used as research data, get consent; the Research Management Guidelines expect consent forms where necessary.
  • Using names in tables and appendices. Use codes; names linked to answers identify learners.
  • Photos of learners in the report. Keep them out of the report, and out of social media.
  • Interviewing learners about painful experiences without support. Plan with the guidance counselor first.
  • Treating the comparison group as unlucky. Fairness requires that they receive the intervention later if it works.
  • Pasting learners’ data into an AI tool. It is prohibited.
  • Copying a sample proposal. Adapt ideas, cite sources, and write your own; the anti-plagiarism declaration is binding.

Practice: how the NASH may test this

Expect situational items in which a school head must balance a study’s goals with learners’ protection. The options differ in whether they protect learners, follow DepEd’s ethics rules and keep the research useful. For more items in this format, see the NASH practice questions with answers and the timed sets in the NASH Reviewer app.

Question 1

Indicator 1.4.3, Research and innovation. Career Stage 3, Situational, Moderate.

During an interview for her action research on learners’ study habits, a Grade 6 teacher learns that one learner is being repeatedly bullied by older learners. The learner asks her not to tell anyone. The teacher asks you whether she should keep it confidential, as promised in her consent form. What is the most appropriate action?

  • A. Keep the information confidential, since the consent form promised confidentiality and breaking it would damage trust in the study.
  • B. Ask her to report the incident to you as school head, refer the learner to the Child Protection Committee with care for the learner’s privacy, and revise the consent form to explain the limits of confidentiality.
  • C. Ask her to stop the study, since research that uncovers bullying is outside the scope of an action research on study habits.
  • D. Ask her to note the incident in the research data, so that the final report can recommend an anti-bullying program for the school.
Answer and explanation

Best answer: B. Points: A = 0, B = 3, C = 1, D = 2.

  • A (0 points, Prestructural). Research confidentiality does not override the school’s duty to protect a child who is being harmed.
  • B (3 points, Relational). It follows the Child Protection Policy, protects the learner’s privacy within the referral, and fixes the consent form so that future participants know the limits of confidentiality.
  • C (1 point, Unistructural). Stopping the study does nothing for the learner who is being bullied.
  • D (2 points, Multistructural). It treats the incident as data for a recommendation, but leaves the learner without immediate protection.

Why. This aligns with Career Stage 3 expectations because indicator 1.4.3 asks the school head to collaborate with school personnel in the conduct of research. DO 16, s. 2017 makes the Child Protection Policy (DO 40, s. 2012) the interim reference for interacting with learners, and DO 40 requires a bullying incident that comes to the notice of school personnel to be “immediately reported to the School Head”, with the Child Protection Committee handling referral. The toolkit’s first protocol is to reduce the risk of harm to participants.

Indicator 1.4.3: Collaborate with school personnel in the conduct of research and utilization of findings in facilitating data-driven and evidence-based innovations to improve school performance.

Sources: DepEd Order No. 16, s. 2017; DepEd Order No. 40, s. 2012; DepEd Action Research Toolkit (2019); DepEd Order No. 024, s. 2020.

Question 2

Indicator 1.4.2, Research and innovation. Career Stage 2, Situational, Moderate.

A teacher plans an action research on reading in a school where most learners belong to an Indigenous Cultural Community. She wants to use stories gathered from the community’s elders as reading materials and to publish the results. She has the parents’ signed consent forms. What is the most appropriate action?

  • A. Proceed, since the parents of every participating learner have already signed the consent forms required by DepEd.
  • B. Proceed with the reading intervention, but drop the plan to publish, so that the community’s stories stay within the school.
  • C. Before planning further, seek the community’s consent through its own processes and elders, agree on co-authorship and the use of the stories, and plan to share the results with the community.
  • D. Ask the division IPEd focal person to approve the study on the community’s behalf, so that the teacher does not need to consult the elders directly.
Answer and explanation

Best answer: C. Points: A = 1, B = 2, C = 3, D = 0.

  • A (1 point, Unistructural). Parents’ consent covers their children’s participation, not the use of the community’s knowledge and stories.
  • B (2 points, Multistructural). It avoids one problem, publication without permission, but still uses the community’s knowledge without the community’s consent.
  • C (3 points, Relational). It follows free, prior and informed consent through customary processes, recognizes the community as co-author and co-owner, and shares the results back.
  • D (0 points, Prestructural). An office cannot give consent for a community; the RMG requires consent through the community’s own governance processes.

Why. This aligns with Career Stage 2 expectations because indicator 1.4.2 asks the school head to facilitate evidence-based innovations from reliable research. DO 16, s. 2017 requires research involving IP learners and communities to follow the rights-based approach of DO 62, s. 2011 and the principles of free, prior and informed consent in its Annex 9: consent sought before planning, through the community’s customary processes; the community recognized as co-author and co-owner of research on its knowledge; and its permission needed before the study is presented or published.

Indicator 1.4.2: Utilize relevant research findings from reliable sources in facilitating data-driven and evidence-based innovations to improve school performance.

Sources: DepEd Order No. 16, s. 2017, Annex 9; DepEd Order No. 62, s. 2011; DepEd Order No. 024, s. 2020.

Frequently asked questions

Does DepEd have a research ethics committee?

DO 16, s. 2017 states that DepEd “will establish a Research Ethics Committee (REC) and issue a corresponding policy on evaluating compliance to research ethics”. Until then, researchers follow the DepEd Child Protection Policy (DO 40, s. 2012) when interacting with learners, and the research committees check ethics when they screen proposals.

If learners’ scores, work or answers will be used as research data, yes: the Research Management Guidelines expect consent forms where necessary, and the toolkit makes informed consent one of its five protocols. Get the parents’ written consent and the learners’ assent.

What are the five ethical protocols of DepEd action research?

From the Action Research Toolkit: reducing the risk of harm; securing informed consent; safeguarding confidentiality and anonymity; avoiding deceptive practices; and ensuring justice and fidelity.

Can I use AI tools to analyze my learners’ data?

Only for support tasks on processed data, and never with personal or sensitive information. DepEd Order No. 003, s. 2026 prohibits entering personal and sensitive information into AI systems and requires you to declare every AI tool used in your research.

What happens if a researcher violates research ethics?

Under DO 16, s. 2017, anyone found to have violated research ethics, or to have committed plagiarism or fraud, is blacklisted from DepEd’s research grant mechanisms. Using AI-generated research without significant human contribution and disclosure is subject to administrative sanctions under DO 003, s. 2026.

More on DepEd research

Sources

Written by Mark Anthony Llego for NASH Reviewer. Sources: DepEd Order No. 16, s. 2017 (Research Management Guidelines) and its Annexes 3 and 9; DepEd Orders No. 40, s. 2012, 62, s. 2011, 14, s. 2022, 003, s. 2026, 39, s. 2016 and 024, s. 2020; Republic Act No. 10173; the DepEd Action Research Toolkit (2019). NASH Reviewer is not affiliated with DepEd.