What is involved in structured observation

Structured observation is a tool used by researchers to collect quantitative data in order to measure and statistically test the validity of an assumption or hypothesis made about a social phenomenon. The method is ‘structured’ by virtue of the fact that the observer only records frequencies or durations of specific predefined occurrences. It is an ‘observation’ as questions need not be asked to the subject during the session, the observer, generally, just observes. For the first part of this paper I will discuss three components I feel are involved in method of structured observation; the objective, the system and the process.

In order to define a rigorous system and process, clear and concise objectives should be set for the study. The type of occurrences or behaviours to be recorded need to be set, these will form the basis for the categories in the observation schedule. The aim, and any expected benefits should be noted along with a specific hypothesis which will be tested by analysis of the data. All parties involved in the setting, facilitating, observing and recording need to be considered along with the related time, cost and geographical constraints.

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

The system should detail the method in which the data will be recorded and analysed. An observation schedule in the form of a chart or diary document should be produced for the physical recording of occurrences. This usually involves developing a clear and objective set of categories against which the observer will record some tally of frequencies. The categories should be distinct enough to avoid ambiguity, yet broad enough to cover all possible target behaviours. The definition of an ‘occurrence’ needs to be investigated; it could be represented by the basic frequency of target behaviours, a snap shot of behaviours at a predefined time intervals or a timeline of behaviours throughout the session. The method of data analysis should also be defined. The system of compiling results from multiple observers should be developed along with a contingency for unexpected or inaccurate recordings. The exact method of statistical analysis should be decided with some appreciation of the conclusions it might potentially be supporting.

The process, relates largely to the interactions involved in carrying-out an observation. The involvement of the observer needs to be agreed. A participant observer may sit with and even interact with the subject or group for a richer vantage point, whereas a non-participant observer would not interact or possibly not even be identified to the subjects. Depending on the number of subjects or the variety of target behaviours more than one observer may be employed. This raises issues of standardisation for which training sessions and possibly pilot sessions might be advantageous. The physical setting is important in that the observer requires good audio and visual range but should also remain as unobtrusive as possible. Finally a debriefing session may be useful, especially when more than one observer is employed. This will give those involved an opportunity to reflect on the session and their technique and at the same time highlight any issues that may affect the analysis of the data.

Part 2 – A brief summary of the first methodological philosophy you have selected.

I have chosen positivism as my first methodological philosophy to investigate. For the purposes of this part of the assignment and to offer some foresight for the next part, within the following summary of positivism I will discuss its history, conflict and bearing on quantitative research.

Positivism is a set of philosophical ideas that state that the only authentic knowledge is scientific knowledge. The name extends from the principal that positive facts alone should form the basis of knowledge, with a refusal to go beyond that which can be supported by empirical evidence. Although there is ongoing dispute over the forerunners of positivism, the French philosopher Auguste Comte (1798 – 1857) is widely recognised as “the father of positivism” (Lenzer, 2004, p xi). Prior to Comte’s dialogues, society, just simply, was. There was an understanding that the shape of society was not constant with a presumption that it was guided either by a higher, divine power or changed randomly through the grains of time (Babbie, 2008). Theological ideology predominated until enlightened thinkers began to believe in, and attempt to understand the fundamental nature of all reality. During this period major scientific discoveries were taking place. Physicians and chemists were not only developing theories but also scientifically proving concepts such of gravity, magnetism, force and mass. Comparing these discoveries to phenomena of social science, Hergenhahn makes an interesting point that like social science “none of these entities can be observed directly” (Hergenhahn, 2009, p424). The challenge for advocates of positivism and to some extent physicians was for science to use theory but yet avoid the inherent metaphysical speculation. By adopting scientific methods supporters of positivism believe that the fundamental components of social interaction can be reduced to a set of physiological rules.

One of the popular criticisms of positivism is the argument that society does not abide by rational principles at all. It is thought that a positivist approach “arrives at a position by looking at science and then unjustifiably applies it to other things” (Wang, 1974, p7). In contrast to gravity, magnetism, force and mass, living beings interpret the society in which they live and at least to some extent construct their own realities. Nesfeild-Cookson describes the positivist antagonist and poet William Blake as stating:

“All they can do is to define life in terms of biochemistry, biophysics, vibrations, wavelengths and so on; they reduce ‘life’ to conceivable measurement, but such a conception of life does not embrace the most evident element of all: that life can only be known by a living being, by ‘inner’ experience”

(Nesfield-Cookson, 1987, p21)

For the purposes of the following parts of this paper it is important to give quantitative research a mention in this summary. Although there is no direct connection between positivism and quantitative methods it is understood that quantitative methods are positivist in conception and orientation. Quantitative research can involve such processes as measuring causal relationships, proving or disproving a hypothesis within a degree of statistical significance and developing models of society which lead to a predictable reality. This use of statistical analysis, evidence of causal or statistical relationships and using explicit transparent procedures are but a few of the quantitative tenets which are in parallel to the philosophy of positivism. (Bryman, 1988).

Part 3 – An assessment of the strengths and weaknesses of structured observation from this point of view.

Positivist methodology has been used widely in educational research techniques. For the last quarter of a century positivist techniques have been popular as a tool for studying classroom behaviour. With its roots firmly set within evidence based practice and its key perceived benefit in its pursuit and logical progression of ‘what works’, it has provoked no shortage of debate between researches of this and other ideologies. By investigating access to data, replicability and access to subjects I will explore some of the strengths and weakness of this methodology.

In the study of classroom behaviour arguably the greatest strength of a positivist approach is its direct access to the data. The observer can record details about student and teacher behaviours first-hand without having to rely on retrospective accounts of others. This method avoids both potential bias on the part of a researcher and also constructed, perhaps anticipatory accounts from subjects involved in a less structured observation or interview processes. As previously mentioned in part two of this paper, the observations in a positivist study lead to a (positive or negative) confirmation of a predefined hypothesis and therefore do require any subjective interpretation in order to draw conclusions, hypotheses or theories. This method therefore, offers the data independence from the researcher and importantly his or her personal beliefs, during the collection and analysis stages. The logical and scientific nature of structured observation leads to strong procedural objectivity.

Another advantage related to the data collection is one of replicability. As the data collection method of a research study approaches ‘pure observation’ the potential that the results can be replicated increases. The quantitative nature and the scientific processes behind the observation schedule allow for the data to be highly replicatable in future studies. This increases the validity of the conclusions drawn from the study and therefore its usefulness back in the classroom in terms of classroom decisions or possible improvements. The final advantage of structured observation in the classroom is access to students who might not otherwise be able to partake in a data collection activity. Early-years students and the mentally or socially unstable potentially would not be able to give responses or participate in dialogues with a researcher in a less structured interview or observation scenario. The structured observation in a classroom allows the observer to record data from the subject’s normal setting.

In today’s society there are many critics of structured observation and more generally, positivist methodology in educational research. I will explore three of the perceived weaknesses; inferences on a population, scope of measurability and accuracy.

Following a structured observation the data is analysed and statistically compared to a null hypothesis to determine the validity and confidence of any relationships between the variables. The statistical calculation compares the resultant data to a null hypothesis (being the values variables one would expect should there be no relationship) and, depending on the type of distribution, leads to a chi-squared statistic (I‡?) which mathematically describes the deviation (or spread) of the data. The I‡?statistic is then evaluated against the degree of freedom of the data. The result is a statistic that shows the level of confidence that the data collected displays some real causal relationship between the variables. Once this has been established, positivist researchers might conclude whether or not the sample is representative of a target population. Critics of this positivist approach might argue that the data collected is not robust enough to be able to cast inferences on a greater population. It could be disputed that in a classroom there is a multitude of influences on students and teachers which may lead to different data being collected on different occasions. Some examples of possible influences are; proximity to important exams, the developing relationship with a teacher and the varying levels of student alertness throughout a school day. Therefore the inference made about a target population raises the epistemological issue of the validity of a ‘universal state of truth’ based on observations of a certain number of positive instances. By the nature of quantitative methods only measurable occurrences can be recorded. The scope of a structured observation of classroom behaviour is therefore the second area of weakness. Some examples of occurrences which might not be measurable in a classroom are; the feelings of students, the motivation behind behaviours or the relationship between students. As well as these there maybe behaviours that are hidden or disguised such as verbal or physical bullying or cheating. These unobservable thus unquantifiable phenomena narrow the scope of a quantitative research project and raise some inadequacies in the positivist assumption that knowledge is derived from objective observation. The final weakness is of one of accuracy or more specifically; accuracy in the interpretation and recording of occurrences by an observer. Even after commenting on the objectivity of structured observation as a strength, it is important to understand that subjectivity does still exist. Each observer involved a structured observation perceives the occurrences in the classroom from the view-point of his or her uniquely constructed reality. There is therefore potential for behavioural occurrences to be interpreted differently by different observers and indeed differently from how the original interaction was intended by the student. As an example, a teacher might excuse the class from homework at which point one student might comment to another, “That’s wicked! Sir is so bad” to which the student might reply “Word, innit?” (slang for ‘isn’t it’). Based on the schedule of categories as developed by Bales (see Study Guide, p. 143), the table below shows; the interpretation an observer (OBS1) not conversant in contemporary childhood colloquialism, the ‘translation’ into Standard English (using my personal reality!) and the interpretation of a well-versed observer (OBS2).



by OBS1



by OBS2

That’s wicked

Shows tension

That’s great

Gives opinion

Sir is so bad

Shows antagonism

Sir is a great guy

Shows solidarity

Word, innit?

Asks for opinion*

Yes, I agree


* The original phrase would not be understood clearly. This might be coded as such by virtue of the perceived question mark.

Subjectivity is also a consideration in the interpretation of the observation schedule itself. Observers could interpret the boundaries of each category differently from how they were intended by the researcher which may lead to conflicting accounts of the observation.

Part 4 – A brief summary of the second methodological philosophy you have selected.

I have chosen interpretivism as my second methodology to discuss. Interpretivism is otherwise known as anti-naturalist or anti-positivist, one can therefore deduce that it was developed in opposition to positivism, with its roots usually outside the realm of quantitative methods. Much of the tradition and origins of interpretivism can be found in writings by Max Weber (1864-1930). My summary will focus on the nature and goal, focus of interest, the elements involved, the subject-researcher relationship and the desired information of interpretivism.

Interpretivists aim to gain understanding human behaviour from the points of view of the people involved, rather than explaining human behaviour by causal relationships (Henn et al, 2006). The main focus is on gaining a rich understanding the human experience in specific social and cultural contexts. The methodologies employed through interpretivism attempt to understand social phenomena largely through qualitative techniques. It is important to make clear that the choice of techniques are not mutually exclusive or in any way predetermined by the choice of methodology. However it is the case that qualitative techniques are usually better suited to interpretative studies and the converse is true for quantitative techniques and positivist studies. The ultimate choice of methodologies is usually a function of the research problem. Unlike the world of natural science the social world is not fixed, realities are continually changing. It is felt that the positivist methods akin to those used by natural scientists are not applicable to social realities. The social world is believed by interpretivists to be too complex to be able to infer causal relationships based on data and statistical analysis of data from a sample. The elements involved in this methodology are clearly distinct from the elements of positivist methods. The process of the interpretivist researcher is a holistic approach beginning with an observation. This is usually a highly interactive process in which the researcher is immersed and in direct contact with the subject. By means of field notes the researcher is able to develop patterns and definitions of the social realities being studied. Finally the interpretivist researcher can form relevant hypotheses and theories. Importantly we can see that adopting interpretive methodology, the process leads to the formation of a hypothesis whereas a positivist study begins with a hypothesis to be tested. This is described as an inductive process compared to the deductive process of positivism. Research using these methods is usually small-scale, focusing on achieving intensive data from some key individuals. The researcher may form a rapport with the subject, empathising and encouraging in order to passively elicit rich dialogue and descriptions. The desired information of interpretivism is not of a positivist single or authoritative ‘truth’ but rather insight into the ongoing story of the subjects. More concisely, the desired information is thorough documentation of the subjects’ perspective of reality (Butin, 2010).

Part 5 – An assessment of the strengths and weaknesses of structured observation from this second point of view.

To the hardened interpretivist researcher, trading in-depth social insights for a table of frequencies would seem against the core principles of interpretivism. However by employing quantitative methods such as structured observation one does not necessarily have to sacrifice the qualitative aspects of research. It is important to remember at this point that it is the research project determines the techniques used, not the methodology. I will explore the strengths of structured observation within a qualitative research project by looking at the benefits of; greater perspective, reflection and increased validity.

Qualitative research can be positivist, it is clear to see that within qualitative research, points of view and social realities are documented, so it could be reasoned that dialogues which lead consistently to one outcome would be interpreted as a ‘truth’ (Lin, 1998). Within a research project to study classroom behaviour, carrying-out a structured observation at the outset of the project could help the researcher develop a deeper understanding and perspective of the students and an insight into their genuine (not constructed through dialogue) realities. In the pursuit of a rich understanding it would be hard to deny that this extra time spent in the field would be of negative value to the project. Although the nature of interpretivist research is relatively unstructured, this additional technique would allow the researcher to understand the scope and perhaps best select the most appropriate qualitative techniques. During the study, structured observation can be used to reflect on, confirm or strengthen some of the hypothesis and theories that might be starting to develop through the course of the qualitative techniques. Even though the researcher might not complete a full analysis of the data to the point of inferring truths about a population, the added insight that the quantitative data offers would offer a valuable comparison to the dialogues already taken from the students. Wider validity and generalisability are perhaps not at the forefront of a pure interpretivist research project. It may however be of some benefit to a research project to draw more valid inferences on a wider population. Kelliher describes this restriction, “While interpretive research is recognised for its value in providing contextual depth, results are often criticised in terms of validity, reliability and generalisability” (Kelliher, 2005, p 123). The inclusion of a structured observation would offer the project scope for inference on a population. Interestingly a structured observation could even be employed objectively to test the validity of a hypothesis made as a result of the qualitative study itself.

The methodology of structured observations does not fall in line with the classic paradigms of interpretivism. From an interpretivist point of view I consider there to be three main weaknesses in using a structured observation to study classroom behaviours; student and teacher reactivity, shallowness of data and the inaccuracies of the process.

It is apparent that subjects behave differently from normal when involved in an overt structured observation. An example of this reactivity can be witnessed in the video footage of start of ‘Frank’s lesson’ (DVD 2, Section 3d). At the beginning of the class there is a hum of excitement followed by several embarrassed glimpses at the camera throughout the rest of the lesson. Students therefore, clearly show an interest in the camera, and although they were presumably briefed to ignore the camera they may have felt uncomfortable and perhaps more introverted for the course of the lesson. As Mertler points out in his dialogue on action research, “The simple presence of a teacher as a ‘data collector’ with a notebook and pencil, or perhaps a video recorder in hand- can change student behaviour” (Mertler, 2006, p 93). As well as a natural reactivity to the observation, students and teachers may conduct themselves differently due to the presumption that they are being judged in some way by the observer or researcher. The observer may witness increased effort on the part of students or teachers and any usual negative behaviour may be hidden. I have already mentioned that any normal physical or verbal bullying or intimidation may be suspended during the observation. Covert observation could be a solution to student or teacher reactivity however there are ethical uncertainties concerning the right of awareness and consent which limit its use in the classroom. The type of data achieved through a structured observation can be regarded by the interpretivist researcher as but a small snap-shot of the overall activity, void of ‘meaningful’ information. Stokking et al discount positivist data collection by describing the techniques as “Meaningful behavioural entities are broken down into meaningless pieces” (Stokking et al, 1999, P 104). The observation of a classroom would understandably capture measurements of what students and teachers are doing but would not explain why they do it, the underlying motivations and meanings of the behaviour is still inaccessible. The highly structured nature of the observation schedule used to record frequencies of behaviours could result in some unexpected or spontaneous behaviours left unrecorded. These instances of unrecorded behaviour may therefore be silenced by the data collection process and thus ignored during the analysis stage of the project. With these inadequacies in mind interpretivist researchers feel that using data from a structured observation to form some replication of reality would be tenuous. Interpretivists believe that there are inherent inaccuracies within the process of recording data from structured observations. The accuracy of physically recording data depends on the understanding of the observation schedule and the skill of the observer. It may be that the intended boundaries of the categories, or the categories themselves, may have been misunderstood. As well as misunderstanding the schedule, the behaviour and interactions may be misinterpreted by the observer. These inaccuracies would mean behaviours might be coded under inappropriate categories causing the results of the project to be misaligned with reality. There are also limitations to what the observer can actually witness. Some interactions might be outside the observer’s range of vision and some dialogue might be too quiet for them to hear. As well as these limitations of the setting the observers might have been temporarily distracted or have fallen behind in coding and therefore fail to notice other important interactions.

Part 6 – A conclusion exploring the implications of the comparative assessment you have carried out.

Guided by a set of seemingly mutually exclusive tenets I have summarised the history and application of the two methodologies when applied to the data collection technique of structured observation in the classroom. Under the quantitative framework, researchers agree on the importance of defining and adhering to a methodological protocol. The data is shaped and constructed by measurement, either of time or frequency. Objectivity and the reliability of the data are assured by the methodological rigor. The analysis is carried out statistically and the results are often highly generalisable to a target population. Under the qualitative framework researchers do not plan or predict all the methods to be use in a study, instead, they develop their methods based on the ongoing needs and requirements of the study. The project and methods are continually reflected upon and the succeeding methods are decided by the direction of the project. The data achieved through qualitative research are in the form of rich descriptive insights into the social reality of the subjects involved. The analysis is carried out draws conclusions about the realities of the specific subjects involved in the study.

Almost every facet of the methodologies are divided by the differences between the study of the natural world and the study of the social world; researcher involvement based on detachment and objectivity vs. rapport and relationship, theories that result from statistics vs. insight, explanations achieved through the formation of causal laws vs. deep descriptions, conclusions with their strength in reliability and validity vs. trustworthiness and rigour and knowledge founded from what is directly observable vs. understanding.

Purists from each methodological camp have made powerful statements promoting their ideologies; “There is no other knowledge about the world, only what is provided by these sciences” (Ajdukiewicz , 1973, p 63) Ajdukiewicz proclaims, conversely von Wright contests that “understanding human behaviour and their intentions behind it demands a degree of empathy with our subjects” (von Wright, 1993, cited in Henn, Weinstein, Foard, 2006, P 15). With such uncompromising extremes it is easy to assume that there could not possibly be any common ground, however there is a growing school of thought concerning ‘bridging the gap between positivism and interpretivism’. There exist various examples of how a mixture of techniques has been used collectively in research (Lin, 1998). Lin analyses the differences between a quantitative research project and a qualitative research project both conducted on the same topic by two practiced researches. She concludes that the differences between the projects stem from the differences one asks of the data and the conclusions one wishes to draw.

One interesting theory as developed by Karl Popper (1902-1994) is one of falsification. In Bharadwaj’s paper on integrating the two approaches he describes positivism and interpretivism as two extremes with the Sophisticated Methodological Falsification model somewhere near the middle (Bharadwaj, 1996). Based on a Lakatosian SMF framework, Bhararadwaj goes on to suggest that “while absolute truth may not be achievable by science, scientific research programs should in the long run lead to ever more true and fewer false consequences, and thus have increasing plausibility.” This shift in methodology permits a wide variety of methodological approaches to be employed in coalition with a goal of working towards an acceptable ‘truth’ based on the somewhat abstract scientific and sociological theories of social studies.

Much of the debate between the methodologies of positivism and interpretivism result from a difference of opinion on epistemology. The plethora of definitions presented within the pages of publications on research of the various techniques is of course fair and just however it seems almost unequivocally divided and decided by the researcher’s epistemological background. Lin points out that the difference between qualitative and quantitative work extend from the dimensions of the project rather than methodological paradigms “The differences in interpretivist and positivist qualitative work thus are differences in the questions one asks of the data and the types of conclusions one wishes to draw.” (Lin, 1998). I have already mentioned more than once in this paper that the tools that a researcher selects should be a function of the project itself rather than a result of a philosophical debate. The two methodologies (positivism and interpretivism) are closely aligned with the two forms of research techniques (quantitative and qualitative). I believe that the description of this relationship indirectly deepens the divide between the two methodologies. Throughout this paper and many other publications, techniques are describes as ‘using qualitative methods’ or ’employing quantitative methods’. Perhaps if we can conscientiously shift these descriptions to have a greater focus on ‘qualitative data’ and ‘quantitative data’ research projects might become free from the ideological debate of methodologies and techniques and focus more directly on what is needed to achieve the results required. Even Webber, one of the leading thinkers on interpretivism, held that researchers should conduct research that is value-free in their pursuit of conclusion. It seems to me that a project driven purely by a researcher’s methodological preconceptions constitutes the ultimate bias.

You Might Also Like

I'm Alejandro!

Would you like to get a custom essay? How about receiving a customized one?

Check it out