Sampling a classroom, and what a teacher has to keep in view, including figures from 2024 and 2025

Sampling is a way of choosing a part in order to understand the character of a whole, when surveying an entire population is hard. The population is the whole group under study; the sample is the part actually chosen from it. A full census asks too much time and money — surveying every child in early-childhood institutions nationwide would take excessive staff and budget — and it is often physically hard to reach every subject. The aim is not always to check each person one by one, but to grasp a tendency, as an opinion poll can follow a social current from a limited sample without surveying the whole public. Because a part is being used to understand a whole, results can change with who is chosen and how, so representativeness and the chance of bias have to be considered together with efficiency. Probability sampling gives everyone in the population a chance under a set probability: simple random sampling, equal chance, helpful for representativeness; stratified sampling, which divides the population by sex, age, region, or type of institution and then draws from each stratum, so that varied traits are reflected evenly; and cluster sampling, which selects groups first — a few kindergartens or daycares, rather than each child nationwide — and can cut time and cost. Non-probability sampling sets the sample by judgment or convenience: convenience sampling, such as one class at a kindergarten where the writer is on practicum, simple but possibly weak on representativeness; purposive sampling, which deliberately chooses subjects with traits that fit the aim; and quota sampling, which builds the sample to a set ratio in light of the population’s traits. Field application is easier, but subjectivity can enter. The usual process is setting the population, drawing the sample, collecting material by survey, interview, observation, or records, and analyzing it. When the subjects are young children, their language and the situation have to be considered together.

For a teacher carrying this out, the first task is a clear criterion: whether the aim is peer relations, language development, or play participation, so that it is clear which children, at which time, in which situation, will be looked at. Otherwise only conspicuous children are seen again and again, and quiet children can drop out. Second, prejudice can enter. A teacher may, without meaning to, look more at active children, children who often show problem behavior, or children who simply catch the eye, and an impression such as “this child is distracted” can make only the distracted scenes stick. The teacher should write down why a child was chosen, repeat the observation in several situations, and, if needed, check the criterion with a colleague. Third, ethics and protection. A young child cannot, like an adult, fully understand what it means to take part and decide for themselves. Records should support the child, not stigmatize or make the child look like a problem child. In 2025, the Ministry of Health and Welfare said child-abuse reports received had risen to 50,242, and that the role of mandatory reporters such as teachers was becoming still more important — which also means a teacher is in a position to notice small changes and warning signs. The child’s emotional state, privacy, and trust with parents have to be considered, including whether repeated recording burdens the child and whether records might be shared outside inappropriately. Sampling, here, is a matter of protection and responsibility before it is a research technique. A child can look entirely different by situation — quiet in morning free play, very active outdoors — so time, activity, peer grouping, and classroom atmosphere should be recorded together. In 2024 the Seoul Metropolitan Office of Education issued a kindergarten play-recording and evaluation casebook, because, in a reality where one teacher is responsible for many children and several duties at once, observation and recording needed more systematic support.

The scales a teacher actually uses, talking with parents, and a child’s own look back

A scale is a frame for classifying, comparing, and giving numbers under a set criterion. As the writer looked into it, scales are usually divided into nominal, ordinal, interval, and ratio. Nominal scales classify — sex, class name, type of institution — and in the field appear as basic living-habit checklists or observation sheets, marked does/does not tidy, or participates/does not, after a period of observation. Ordinal scales give order — high, ordinary, or low participation — and appear as social rating scales or peer-relation sheets, recording cooperation, self-regulation, and expression from 1 to 5, or as low, ordinary, or high, after repeated observation rather than a single scene. Interval scales have equal gaps between scores, as in test scores or standard scores, and appear in standardized language-development or developmental tests given under set items and procedures, then compared with peers. Ratio scales have an absolute zero, so amounts can be compared directly — height, weight, reaction time, number of performances, time on a task, number of blocks stacked, number of correct answers. In Han Seok-sil’s study, the purpose of early-childhood assessment was most often placed on understanding development and reflecting it in teaching-learning plans, and results were often used in parent counseling and curriculum improvement; that study is, without an invented year.

After a tool is used, counseling with parents should go toward understanding the child together. The teacher first organizes the result with everyday observation records, because a score alone does not explain the child’s actual appearance well. Before the meeting, the purpose should be explained — not to judge the child, but to find a direction of support. The talk should start with strengths, with actual scenes — “the child often went up to a friend first,” or “concentration was good at story time” — and areas needing support should be put softly, “this part seemed to need a little more support,” rather than stated as a verdict. Ways the home and the institution can help are then discussed, and a later follow-up is linked. Counseling is a process of aligning a direction of support, not an announcement of results. Self-assessment, as the essay frames the assigned heading on written evaluation, is a child’s looking back at an activity, a work, a way of taking part, or a feeling, and putting it in the child’s own words: “it was fun,” “it was a bit hard,” “it was better with a friend,” “I want to do better next time.” It is a process that lets the child take part in their own learning, not a one-way mark from the teacher. In Sim Yun-hui’s study, a child’s self-assessment in the course of portfolio evaluation is treated as having educational meaning, and creative productivity and reflective expression can appear in that process; the study is, without an invented year. Types of written evaluation and a fuller account of the teacher’s role under that later heading are not filled beyond this. The same Korean report, at fuller length, is on CampusDown.

Sources

1. Sampling is a way of choosing a part in order to understand the character of a whole, when it is hard to survey an entire population. In other words, because the whole the researcher wants to know is too large or too wide, a set of subjects is drawn and surveyed, and the tendency of the whole is grasped from that result. Here the population is the whole group that is the object of the study, and the sample is the part actually chosen from that population. Sampling is important because, in practice, surveying every subject is very hard. It takes a great deal of time and money, and the wider the range, the greater the burden of arranging and interpreting the material. Sampling can be seen as a basic procedure of research for obtaining, under limited conditions, results that are as valid and reliable as possible. The reasons sampling is needed can be gathered in a few points. First, a full census asks too much time and money. For example, surveying every child in early-childhood institutions nationwide would take excessive staff and budget. Second, it is often physically hard to reach every subject. Third, the aim of the study is not necessarily to check every subject one by one, but to grasp a certain tendency or trait. As in the opinion-poll case looked at, a social current can be followed from a set sample without surveying the whole public. But because sampling is a way of understanding a whole through a part, results can change depending on which subjects are chosen and how. So sampling is a method that raises efficiency and, at the same time, has to consider representativeness and the possibility of bias together. Kinds of sampling can be divided largely into probability sampling and non-probability sampling. Probability sampling is a method in which every subject in the population has a chance of being chosen as a sample under a set probability. Representative examples are simple random sampling, stratified sampling, and cluster sampling. Simple random sampling is a way of giving every subject an equal chance and choosing at random. Stratified sampling is a way of dividing the population by criteria such as sex, age, and region, then drawing a sample from each group. Cluster sampling is a way of first selecting groups rather than individuals, then surveying. By contrast, non-probability sampling is a method of setting the sample by the researcher’s judgment or practical convenience. Convenience sampling, purposive sampling, and quota sampling belong here. Convenience sampling is a way of setting the sample around subjects who are easy to reach, and purposive sampling is a way of deliberately choosing subjects who have traits that fit the research aim. So probability sampling is helpful for securing representativeness, and non-probability sampling can be said to have the trait that field application is comparatively easy. The sampling process usually goes in the order of setting the population, drawing the sample, collecting material, and analyzing the material. First, setting the population is the stage of deciding whom the researcher will regard as the whole object. This stage is important because the range of the study and the interpretation of the result change according to how the population is set. Next, drawing the sample is the stage of choosing, from the population, the subjects actually to be surveyed. At this point, the way the sample is set has a direct effect on the reliability of the study. In the material-collection stage that follows, needed information is obtained through surveys, interviews, observation, analysis of records, and the like. As in the case of research with young children that was looked at, it is important to gather material consistently, considering the traits and situation of the subjects. Finally, in the analysis stage, what has been collected is arranged and interpreted, and a meaning is drawn out. Simple random sampling, because it gives every subject the same chance and chooses at random, is helpful for securing representativeness. Stratified sampling is a method of dividing the population by criteria such as sex, age, region, and type of institution, then drawing a sample from each stratum. This method is needed in order to reflect evenly the varied traits that exist inside the population. Cluster sampling is a way of first dividing groups rather than individuals, then choosing some of those groups and surveying them. For example, instead of drawing each child one by one from the whole country, first selecting a few kindergartens or daycares belongs here. This method has the advantage of being able to reduce time and cost. Non-probability sampling is a method of choosing subjects by the researcher’s judgment or practical convenience. Here there are convenience sampling, purposive sampling, quota sampling, and the like. Convenience sampling is a way of drawing the sample around subjects who are easy to reach. For example, taking one class at a kindergarten where one is on practicum as the object of a survey belongs here. This method is very simple, but representativeness can be weak, because subjects who are easy to meet may not sufficiently reflect the whole. Purposive sampling is a method of deliberately choosing subjects who have traits suited to the research aim. It was a useful method when a particular behavior or trait has to be looked at closely. Quota sampling is a way of composing the sample to fit a set ratio, considering the traits of the population. So non-probability sampling has the advantage that field application is easy, but the point that the researcher’s subjectivity may enter also has to be considered together. When sampling is carried out, the first thing an early-childhood teacher should consider is clearly setting a criterion for what is being looked at and whom is being chosen. In a young-children’s classroom, even within a single day, play, conflict, cooperation, language, and emotional responses appear very differently. So it will not do, the writer thought, to judge the whole of a child easily from only a few scenes. Sampling is a process of trying to understand a whole through a part. Therefore the teacher should first clearly decide whether the aim of the observation is peer relations, language development, or play participation. Only then does a criterion arise for which children, at which time, in which situation, will be looked at. Otherwise only children who catch the eye will be seen repeatedly, and quiet children, or children with little expression, can drop out of the observation. The second thing to consider is the possibility that prejudice and preconception will enter. An early-childhood teacher, even without intending it, may come to look more at ordinarily active children, children who often show problem behavior, or children who unusually catch the eye. But this way can break the balance of the sampling, because the teacher’s impression can become the very criterion for choosing material. For example, if one first holds the thought “this child is distracted” and then observes, one may come to remember more strongly only the distracted appearance, rather than actually positive interaction or scenes of concentration. So the teacher should write down specifically the reason for choosing a subject, and should, as far as possible, observe repeatedly in several situations. If needed, checking the criterion together with a colleague can also help. Sampling looks like an objective technique, but in the actual field it can be said to be a matter in which the teacher’s gaze acts very largely. Third, questions of ethics and protection should be thought about carefully. A young child is a being who, unlike an adult, finds it hard to understand fully the meaning of taking part in research and to decide for themselves. Observation and recording should be for supporting the child, and should not become a way of putting a stigma or of making the child look like a problem child. In fact, the Ministry of Health and Welfare said that in 2025 the number of child-abuse reports received had risen to 50,242, and explained that the role of mandatory reporters such as teachers was becoming still more important. This also means that an early-childhood teacher is in a position to look sensitively at a child’s small changes and warning signs. Therefore, even in the sampling process, the child’s emotional state, the protection of privacy, and a relation of trust with parents should be considered together. It should be checked whether the work of recording a particular child repeatedly does not become a burden on that child, and whether there is a risk that the content of the records will be shared outside inappropriately. In this respect, an early-childhood teacher’s sampling can be said to be a question of protection and responsibility before it is a research technique. The teacher should consider together the accuracy of records and the variety of observation. Because a young child can show an entirely different appearance depending on the situation, a conclusion should not be drawn from only one scene. A child who had little to say in morning free play may be very active in outdoor activity. So when sampling is carried out, an attitude of recording together context such as time, activity, peer composition, and classroom atmosphere is needed. That the Seoul Metropolitan Office of Education issued a kindergarten play-recording and evaluation casebook in 2024 was also because, in a reality where one teacher is responsible for many children and performs several duties together, there was a need to support observation and recording more systematically. As that account shows, in observing young children it can be said to matter on what cycle and under what criterion one records. A scale, the writer came to think, can be called a frame for classifying a subject under a set criterion, comparing, and representing it in numbers. It came to be thought of as a criterion that decides how a subject will be understood in research or evaluation. As the writer looked into it, kinds of scale are usually divided into nominal, ordinal, interval, and ratio scales. A nominal scale is a scale that classifies subjects. For example, it can be said to be used in dividing categories such as sex, class name, and type of institution. An ordinal scale is a scale that shows order. For example, a way of dividing a child’s degree of participation into high, ordinary, and low can be seen as belonging here. An interval scale is a scale with the trait that not only the order but the gap between scores is the same. Test scores or standard scores can be said to be close to this. A ratio scale is a scale in which an absolute zero exists, so that the size of an amount can be compared directly. For example, height, weight, reaction time, and number of performances could be gathered as belonging here. In the early-childhood field, the writer came to feel, a scale does not remain only as theory, but is naturally reflected inside actual testing tools and observation tools. First, as an example in which a nominal scale is reflected, a young child’s basic living-habit checklist or observation record sheet can be raised. For example, it is a way of recording by dividing into categories such as “does tidy / does not,” or “participates in the activity / does not participate.” The method of carrying it out, the writer thought, goes in the manner of the teacher observing the child for a set period, then checking the relevant item. As an example in which an ordinal scale is reflected, a social rating scale or a peer-relation evaluation sheet could be thought of. It is a way of recording a child’s level of cooperation, self-regulation, and expression from 1 point to 5 points, or as low, ordinary, or high. It came to be seen as important that this tool be rated comprehensively after repeatedly observing several activity scenes, rather than judging from only one scene. As an example in which an interval scale is reflected, scores that come out of a standardized language-development test or a developmental test can be raised. Such a tool, it can be said, is carried out according to set items and procedures, scored, then interpreted by comparison with peers. As examples in which a ratio scale is reflected, there are measurement of height and weight, time to carry out a task, number of blocks stacked, number of correct answers, and the like. Because this is a way of measuring actual quantity and time directly, the writer thought of it as a tool from which comparatively clear figures can be obtained. In Han Seok-sil’s study as well, the purpose of early-childhood assessment was most often placed on understanding a child’s development and reflecting it in teaching-learning plans and improvement, and assessment results appeared as often used also in parent counseling and curriculum improvement. After a testing tool is carried out, the most important thing, the writer came to think, is to continue counseling with parents in a direction of understanding the child together. First, before speaking the test result at once, the teacher should arrange it together with everyday observation records, because it is hard to explain the child’s actual appearance sufficiently with a score alone. Before counseling, it can be said to be necessary to guide the parent on the purpose of the counseling, and to explain in advance that the result is not for judging the child but a process for finding a direction of support. When counseling begins, the writer felt it is good to speak first of the child’s strengths. For example, if actual scenes are explained together, such as “a look of going up to a friend first was often seen,” or “concentration was good at story-listening time,” the parent too can come to receive it more steadily. Next, areas that are lacking or that need help are more realistic, it can be said, when explained softly in the manner of “this part seemed to need a little more support,” rather than spoken conclusively. After that, methods the home can help with and methods the institution can support are talked about together, and if a follow-up counseling that checks again after a set period is linked, it can be seen as having become a more meaningful process. As the writer looked into it, early-childhood assessment results appeared as often used in parent counseling and in improvement of teaching and learning, so counseling could be called not an announcement of results but a process in which teacher and parent together align a direction of support. Self-assessment, as the essay frames the assigned heading, can be called an assessment in which a young child looks back for themselves at an activity they have tried, a work, a way of taking part, a feeling, and tries to express it. When this idea was first met, because of the word assessment, a scene of a teacher looking at a child and judging came to mind first. But as the content was looked at, self-assessment came to feel a little different in grain from a way in which a teacher one-sidedly assigns a result. That is because a young child too can, in their own language, show their own thought about their experience, in the manner of “it was fun,” “it was a bit hard,” “it was better because I did it with a friend,” “I want to do better next time.” So self-assessment could be seen as a process that makes the child take part in their own learning. As the writer looked into it, a young child’s self-assessment carried out in the course of portfolio evaluation was being treated as having educational meaning inside the assessment of a child’s development. Sim Yun-hui’s study saw a child’s self-assessment in the course of portfolio evaluation as holding an important meaning, and presented that in this process a child’s creative productivity and reflective expression can appear. Types of written evaluation and a fuller account of the teacher’s role under that later heading are not filled beyond this.

Last updated: 14 November 2026