# Observation in Action Research: Checklists and Field Notes

How to use observation as a data collection method in DepEd action research: types, sampling, checklists, field notes, anecdotal records and observer agreement.

This is the Markdown copy of [Observation in Action Research: Checklists and Field Notes](https://nash.com.ph/blog/observation-data-collection-action-research), an article on the NASH Reviewer Blog by Mark Anthony Llego, published Oct 28, 2026, updated Nov 4, 2026. Topics: Action research, School leadership. NASH Reviewer is an independent practice platform not affiliated with DepEd, BHROD, or NEAP.

Some of what an action research needs to know can only be seen: whether learners stay on task during a new routine, how often they ask questions, what happens in a reading group. Observation is the data collection method for it. The Department of Education's (DepEd) Action Research Toolkit covers it in section 4.4: the methods and types of observation, how to choose whom and what to observe, the tools for recording, analysis, and ethics. Below is that section, with what teachers usually need next: checklist items that can really be observed, simple recording forms, and a check that two observers see the same thing.

It is written for teachers planning action research, and for school heads and master teachers who observe classes.

## Key points

- **Observation records behavior as it happens**, in the toolkit's words "a methodical procedure of collecting data", using "all of his or her senses".
- **In school research, observe openly.** The toolkit describes covert observation, but with learners, DepEd's research ethics require permission from parents and the school.
- **Decide whom and what to observe before you start**: which learners (accessible, purposive or quota sampling) and which behavior (event, time or instantaneous sampling).
- **Write checklist items that can be seen.** "Raises hand before speaking" can be observed; "has positive self-esteem" cannot.
- **Keep description and interpretation apart**, as the toolkit advises: what happened in one column, what you think it means in another.
- **Check that two observers agree.** In our worked example, two observers agree in 17 of 20 intervals (85%), and Cohen's kappa is 0.63.

## Observation as a data collection method

The toolkit notes that observation is "relatively economical": "you only need paper and pen and good pairs of eyes and ears". It can also be "time consuming" and demand "physical and mental stamina". In the field, it involves:

- prolonged engagement in the setting;
- clearly expressed, self-conscious notes while observing;
- planned, methodical work; and
- standardized recording.

Observation fits both stages of action research: assessing the situation before choosing an intervention, and collecting data while the intervention runs.

## Methods and types of observation

**Overt or covert.** In *overt* observation, participants know they are observed and who the observer is. In *covert* observation, they do not, and the toolkit says the observer is "free to pretend to be a legitimate member of the group". In a school, covert observation of learners conflicts with the toolkit's own ethics paragraph, which requires permission from "parents, administrators and others concerned", and with the Research Management Guidelines ([DO 16, s. 2017](https://www.deped.gov.ph/wp-content/uploads/2017/03/DO_s2017_016.pdf)). Observe openly, and reduce the effect of being watched by spending time in the class first, as the toolkit suggests.

| Type | What it is | Strength | Limitation |
|:--|:--|:--|:--|
| Controlled | A set procedure, time and place, often with a structured coding instrument; the observer stays apart | Easy to repeat; observers' agreement can be checked; quick to code | The *Hawthorne effect*: people behave differently when watched |
| Naturalistic | Spontaneous behavior in its natural setting, such as the classroom, playground or library | "Ecological" validity: behavior flows naturally | No control of variables; observations are less reliable; needs trained observers |
| Participant | The observer is part of the group, as a teacher is in their own class | Deep understanding of the group | Hard to take notes; reliance on memory; more bias |

The toolkit judges naturalistic and participant observation "the most practical to be used in Action Research". A teacher observing their own class is a participant observer, so the bias the toolkit warns about applies: plan the recording form in advance, and when you can, ask a colleague to observe too.

## Whom and what to observe

**Choosing learners.** Random sampling is possible in controlled observation, but action research usually uses nonrandom methods:

- *accessible sampling*: whoever is present at the time;
- *purposive sampling*: learners with a particular characteristic, such as those identified as at risk;
- *quota sampling*: a set number from each subgroup, such as two boys and three girls.

**Choosing behavior.** The toolkit describes three ways:

| Method | What you record | Example (engagement during discussion) |
|:--|:--|:--|
| Event sampling | Every occurrence of chosen behaviors, and nothing else | Every learner response when the teacher asks a question |
| Time sampling | Behavior during a set period only | The middle 20 minutes of each hour |
| Instantaneous (target time) sampling | Behavior at moments decided in advance | Only during the teacher-led discussion part of each class |

A practical version for one learner's on-task behavior, widely used in classrooms, is *interval recording*: every 30 seconds, mark whether the learner is on task at that moment. Ten minutes gives 20 marks, and the share of marks on task is the learner's score.

## Tools for recording observations

The toolkit lists five:

1. **Field notes**: written descriptions of what was seen, heard and said, transcribed in full before analysis.
2. **Audio and video recordings**, transcribed for analysis. Recording learners needs their parents' written consent.
3. **Rating scales**: the intensity or frequency of a behavior.
4. **Checklists**: whether a behavior was present.
5. **Anecdotal records**: "detailed written accounts of important events and behavior without any interpretations".

### Writing checklist items that can be observed

The toolkit's practice checklist on student behavior includes items such as "positive self-esteem", "sense of empathy" and "thinks before acting". An observer cannot see these directly; two observers would mark them differently. Rewrite each as something an observer can see or hear:

| Hard to observe | Observable |
|:--|:--|
| Positive self-esteem | Volunteers to answer or read aloud |
| Sense of empathy | Helps a classmate without being asked |
| Follows rules | Raises hand and waits to be called before speaking |
| Easily distracted | Looks away from the task for more than 5 seconds |
| Disrespectful toward peers | Takes or hides a classmate's things; calls a classmate names |

### A field-notes page

Divide the page as the toolkit advises, so that what happened stays apart from what you think it means:

| Time | What I saw and heard (description) | My comments (interpretation) |
|:--|:--|:--|
| 9:02 | L4 and L7 read the passage together; L7 points to each word. L4 stops at "bulaklak" and waits; L7 reads it. | L7 may be helping too quickly; L4 needs time to sound out. |
| 9:05 | Teacher asks, "Who wants to read first?" Five hands go up, including L4's. | First time L4 has volunteered this week. |

Use codes (L4, L7) instead of names, and write the time beside each entry.

### An anecdotal record

| Item | Entry |
|:--|:--|
| Learner (code), date, time, place | L12, February 3, 2026, 10:15, classroom |
| What happened (facts only) | During partner reading, L12 read the passage three times without stopping, then asked the partner, "Can I read it to the teacher?" |
| What the learner said | "Can I read it to the teacher?" |
| Interpretation (separate, optional) | Growing confidence; first request to read to the teacher. |

## Checking that two observers agree

The toolkit names inter-observer reliability as a strength of structured observation. A simple check for action research: two observers record the same learner at the same time with the same form, and you compare their marks.

**A worked example (invented data).** Two observers marked one learner as on task (1) or off task (0) every 30 seconds for 10 minutes. Observer A marked 15 of the 20 intervals on task (75%); observer B marked 14 (70%).

| | B: on task | B: off task |
|:--|--:|--:|
| A: on task | 13 | 2 |
| A: off task | 1 | 4 |

**Percent agreement** is the share of intervals where both made the same mark: (13 + 4) ÷ 20 = 17 ÷ 20 = 85%. With the marks in B2:B21 and C2:C21: `=SUMPRODUCT(--(B2:B21=C2:C21))/20`.

**Cohen's kappa** corrects for the agreement expected by chance. Two observers who both mark most intervals "on task" would agree often even by guessing:

1. Expected agreement by chance: `=AVERAGE(B2:B21)*AVERAGE(C2:C21)+(1-AVERAGE(B2:B21))*(1-AVERAGE(C2:C21))` = 0.75 × 0.70 + 0.25 × 0.30 = 0.600.
2. Kappa = (observed − expected) ÷ (1 − expected) = (0.85 − 0.600) ÷ (1 − 0.600) = 0.63.

A widely used guide (Landis and Koch, 1977) calls 0.41 to 0.60 moderate agreement, 0.61 to 0.80 substantial, and above 0.80 almost perfect. If agreement is low, compare the intervals where you disagreed, make the definitions more precise, and practice together before collecting the study's data.

## Analyzing observation data

The toolkit pairs each tool with an analysis:

- **Field notes, recordings and anecdotal records:** qualitative analysis: read, find themes, classify the behaviors and statements, and count them in a frequency table ([Thematic Analysis for Action Research: Steps and Sample](https://nash.com.ph/blog/thematic-analysis-action-research)).
- **Rating scales and checklists:** descriptive statistics (mode, median, mean, range, standard deviation), and, when groups or times are compared, a t-test or F-test ([Paired and Independent t-Test in Excel](https://nash.com.ph/blog/t-test-excel-action-research)). For on-task shares from a handful of learners, see [Wilcoxon and Mann-Whitney Tests in Excel for Small Classes](https://nash.com.ph/blog/wilcoxon-mann-whitney-test-excel-action-research).

## Guidelines and ethics

The toolkit's three guidelines for observers:

1. **Be inconspicuous.** Spend time with the class before you begin, so that learners get used to you.
2. **Be objective and specific.** Describe behavior in clear, accurate terms, and write interpretations separately, labeled as such.
3. **Be systematic and thorough.** Notes should be precise, complete, logically arranged, and limited to what was seen and heard.

And its ethics: get permission from parents, administrators and others concerned; protect participants' rights; use code names; and keep field notes, recordings, rating scales and checklists "in a safe place", available only to the research team. Photographs and videos taken as evidence stay out of reports uploaded to DepEd's research portal under [DO 14, s. 2022](https://www.deped.gov.ph/wp-content/uploads/2022/04/DO_s2022_014.pdf). [DepEd Action Research Ethics: Consent, Assent and Privacy](https://nash.com.ph/blog/action-research-ethics-deped) covers consent letters.

**Research observation is not performance observation.** Classroom observations for teachers' performance evaluation are made for a different purpose and under different rules ([DepEd PMES for Teachers](https://nash.com.ph/blog/deped-pmes-teachers)). Do not reuse a teacher's evaluation ratings as research data without telling the teacher and getting their agreement.

## Common mistakes

- **Checklist items no one can see**, such as attitudes and feelings.
- **Mixing what happened with what it means** in the same sentence.
- **Deciding what to look for after observing**, which invites seeing what you hoped for.
- **Relying on memory** instead of recording at the time.
- **Observing covertly**, or recording learners without consent.
- **One observer, no check.** Even a short agreement check with a colleague strengthens the data.

## Frequently asked questions

### What is observation in action research?

A methodical way of collecting data by watching and recording behavior as it happens, with checklists, rating scales, field notes, anecdotal records or recordings. DepEd's Action Research Toolkit covers it in section 4.4.

### What is the difference between a checklist and a rating scale?

A checklist records whether a behavior occurred (yes or no); a rating scale records how strongly or how often it occurred, such as on a scale from 1 to 5.

### What is an anecdotal record?

A short, factual written account of a particular event and what a learner did or said, written without interpretation. Any interpretation is added separately.

### How do I check the reliability of classroom observations?

Have two observers record the same learner or class at the same time with the same form, then compute their percent agreement and Cohen's kappa. Kappa above about 0.6 is generally acceptable for action research.

### Can I observe learners without telling them?

Not in a DepEd school study. Parents and the school must give permission, and the learners should know they are being observed. Spending time in the class before collecting data reduces the effect of being watched.

## More on DepEd research

- Start here: [DepEd Action Research Guide: From Topic to Report](https://nash.com.ph/blog/deped-action-research-guide)
- [Interview and FGD Guide Questions for Action Research](https://nash.com.ph/blog/interview-fgd-guide-action-research)
- [Document Analysis in Action Research Using DepEd Records](https://nash.com.ph/blog/document-analysis-deped-records-action-research)
- [Validity, Reliability and Cronbach's Alpha in Excel](https://nash.com.ph/blog/cronbach-alpha-excel-questionnaire-reliability)
- [Thematic Analysis for Action Research: Steps and Sample](https://nash.com.ph/blog/thematic-analysis-action-research)
- [DepEd Action Research Ethics: Consent, Assent and Privacy](https://nash.com.ph/blog/action-research-ethics-deped)
- Related: [DepEd Instructional Supervision: Observing and Coaching](https://nash.com.ph/blog/instructional-supervision-school-heads)

## Sources

- Department of Education and the Basic Education Sector Transformation (BEST) Program. (2019). *Action Research Toolkit: Teacher's Guide*, section 4.4 (data collection through observations), with support from the Australian Government.
- [DepEd Order No. 16, s. 2017, Research Management Guidelines](https://www.deped.gov.ph/wp-content/uploads/2017/03/DO_s2017_016.pdf)
- [DepEd Order No. 014, s. 2022, Adoption of E-Saliksik: The DepEd Research Portal](https://www.deped.gov.ph/wp-content/uploads/2022/04/DO_s2022_014.pdf)
- Cohen, J. (1960). A coefficient of agreement for nominal scales. *Educational and Psychological Measurement, 20*(1), 37–46. https://doi.org/10.1177/001316446002000104
- Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. *Biometrics, 33*(1), 159–174. https://doi.org/10.2307/2529310

*Written by Mark Anthony Llego for NASH Reviewer. The observation data are invented. NASH Reviewer is not affiliated with DepEd.*

---

© 2026 Mark Anthony Llego, NASH Reviewer. All rights reserved. You may quote up to 100 words from this article with the author's name, its title and a link. In research, cite it as: Llego, M. A. (2026, October 28). *Observation in Action Research: Checklists and Field Notes*. NASH Reviewer. https://nash.com.ph/blog/observation-data-collection-action-research. Any other reuse of the article, its charts or its data needs written permission: email support@nash.com.ph. Terms of Service: https://nash.com.ph/terms#website-content-and-data