Observation Method (Part-1)
Observation Method (Part-1)
Subject PSYCHOLOGY
TABLE OF CONTENTS
1. Learning Outcomes
2. Introduction to Observation Research
2.1 Definitions of Observation research
2.2 Characteristics of Observation research
3. Observation and Naturalism
3.1 Tunnell’s Criteria
4. Applications in the field of psychological research
4.1 Observation as a technique
4.2 Observation as an overall research design
4.3 Research questions appropriate to be studied through observation
5. Development of Observational research
6. Approaches to Observation Research
6.1 Naturalistic observation and Observation in Contrived Settings
6.2 Disguised and Non- Disguised Observation
6.3 Structured and Unstructured Observation
7.Summary
1. Learning Outcomes
· Know about the technique of observation and it’s salience in psychological research,
distinguish between observation and allied methods and understand the basic tenets and
characteristics of this method.
· Learn about the technique of observation, the research questions appropriate to be studied
by this technique, the ways of conducting research with the application of this technique
and the ways collecting the data required through checklists and other means.
· Identify the fine differences between observation as a technique in research and
observation as an overall research design and between different approaches to this
technique.
· Evaluate each different approach to this method, including the different styles of
involvement of the researcher in the setting, the different ethics involved in the method
and the method of observation as a whole.
· Analyse the different types of observation, the fine details and differences in relation to
other techniques and what kind of different data recording methods would suit the
research question at hand.
The term observation derives from Latin, meaning to watch, to attend to, Dictionary definitions
(e.g. Oxford English Dictionary, 1989) tend to stress that it is concerned with the accurate
watching and noting of phenomena as they occur in nature, with regard to cause and effect or
mutual relations (note: the definition carries `in nature’ as opposed to manipulated experimental
settings).
Reber (1985), however, extends the definition that “All psychological methods involve
observation, but stresses a distinction that should be made between research that is controlled by
the manipulation of independent variables and research that is carried out by the use of
naturalistic observation”.
C.A. Moser (1958) expressed that “Observation can fairly be called the classic method of
scientific study.”
Observation is one of the oldest and the most fundamental research method approaches. From the
above definitions, it can be seen that `observation’ is the process of collecting impressions of the
world using all of one’s senses, especially visual and auditory, which can be used as an
independent `classic method’ of scientific study and also as a part of research, supplementing
other methods in an important way. However, in the layman’s perception, understanding and
usage of the term `observation’ is very different from the one used in the social research
processes. Everyone is an observer. However, observation stops being a part of everyday lives
and becomes a research method if it is systematically planned, recorded and the recordings are
checked for their validity and reliability with a particular intention for such an endeavour. These
factors simply distinguish between simply observing the world around you and collecting
research data through observation. Observation as a research method, thus, take place when the
researcher intentionally, with a specific purpose in mind, place themselves amidst certain people,
locations, situations and contexts to observe a phenomenon.
For example: Observing two kids playing just for the pleasure
of watching it is an example of everyday observation.
However, observing a group or a pair of kids playing in a particular context or setting, for
example a park in order to observe the “Aggression in the play interaction of children in the age
range of 6 – 11 years” is an observation which is a research method. Other less important
distinctions would be the seriousness, which, in a greater degree, observation as a research
method entails and the importance in furthering of knowledge that observation as a research
method helps in achieving.
The `characteristics’ of observation technique, in general and as a method used in the qualitative
realm of research are presented as follows:
1. An intent and purpose: Observation, as a scientific study seems to provide the characteristics
of objectivity, precision, specificity, systematic steps, verifiable, reliable and valid. For insuring
all of the above, the first and most important consideration is that the researcher (observer) must
know clearly what is to be observed. A scientific observation always has an intent or a purpose
behind it for it to be different from the everyday observation.
Thus, for example: If a researcher wants to study a problem through observation such as “Does
democratic style of teaching lead to better grades in school?” Here it is very important to measure
“democratic style of teaching” by defining it in a way it can be subjected to empirical testing,
Only when the researcher operationalizes the constructs or variables he wishes to measure, the
scope of `what to measure?’ and the purpose of observation would become clear.
Constructivism, that takes the view that the phenomena of the social world (including all one’s
knowledge of it) are not objective entities but are constructs of the mind arrived at (constructed)
through social interactions (e.g. Berger and Luckman, 1967) which thus emphasizes the meaning
attached to things and experiences by people, forms the base of the approach issued in
observational research. This form of research, studies these meanings people attach to things and
experiences and thus probe into their subjective reality.
Reflexivity, is the idea that research data and analysis is shaped by both the researcher and the
researched. The researcher is not a neutral agent of data collection or a blank slate, and carries his
own experiences, perspectives, assumptions to the task at hand and these contribute equally to
shape the research. Observation is characterized by `reflexivity’ at every step beginning from data
recording as only those observations that seem to be important are recorded by the researcher; to
the data analysis where the researcher adds meaning to the recorded data from his point of view
also.
However, it the observation is more structured, systematic and planned; it’s context validity can
also be checked, and it is also characterized by the chance of being replicable, but very rarely.
The major reason so as to why observational research is conducted in the friendly (not – so) real
life settings as compared to the facilitating confines of the controlled lab settings where
manipulation of variables can be seen is for the character of data collected- to be ‘naturalistic’;
which is the most valued feature of observational research.
Tunnell (1977) observed that naturalness of a research enterprise could be conceptualized in
three ways ;
(i) The criterion of naturalness could be applied to the behaviours being studied.
(ii) The criterion could be the treatment applied, and
(iii) The criterion could be the setting in which the research is conducted.
However, these three criterion may be combined in any number of ways, and these combinations
would reflect the amount to which the total study can be judged as more or less naturalistic.
Tunnell (1977) proposes a continuum, at one extreme end which has `unnaturalness’ under which
the studies that constraint behaviour and manipulate the variables (laboratory experiments) can be
placed and at the other and `naturalness’ describes those studies, that does not impinge on the
environment and occurs in natural settings. This classification bears consideration as `naturalness’
is an extremely valued commodity among field researches including Observational Research.
1. Natural Behaviour: The prime goal of all observational research is to study `natural’
behaviour. The inherent `generalizability’ of the findings of the study can be made with more
confidence if the behaviour elicited by the individual occurs in the natural flow of events in real –
life settings, not instigated in response to the demands of the research or in a manipulative and
controlled setting. This, in observational research, which is largely qualitative in nature,
`behaviour’ is thought to be natural to an extent that it is an inherent part of the individual’s
response pattern (i.e., not established to meet any demands of research) and is unconscious
(enacted without self-conscious awareness that he/she is being studies or observed).
There have also been instances in which, even in controlled and manipulated laboratory settings,
natural behavior was used as the `primary dependant variable’. This was done was shifting the
control and manipulation of the laboratory into the natural settings and then observing the `natural
behaviour’ elicited.
For example: Moriarty (1975) induced individuals at a crowded beach feel more or less
responsible for the welfare of another, by having his experimental `accomplice’ or `confederate’
ask a fellow sunbather to look at (watch) his radio while he went to the `broadwalk’ for a few
minutes. Following his departure another confederate approached the empty blanket and picked
up the radio, and if not stopped by the “watchperson” – ran away with it. Here, the natural
behaviour – participant’s responses to the apparent theft was the dependent measure.
The findings from this research were found to be more trustworthy and reliable than earlier
investigations that were conducted in strict laboratory settings.
2. Natural Treatment :
Tunnell (1977) defines natural treatment as a “naturally occurring, discrete event ….. that the
subject would have experienced … with or without the presence of a researcher.” Thus
definition, however, was found to be overly limiting the phenomenon of natural treatment which
is of no good to field research. However, a treatment can be considered a `natural treatment’ if
a. It plausibly could have occurred without the researcher’s intervention.
b. The participant is unaware of the fact that there was intervention of the researcher
(Crano,.W.D. Brewer, 2002).
Thus Moriarty’s (1975) treatment (a staged theft of a radio) qualifies as a `natural treatment’ as
a. Thefts do happen in beaches (which could have happened without the researcher’s
intervention
b. The participant was unaware that the theft was staged.
3. Natural setting:
A naturalist observational investigation refers to, usually, a study, conducted outside the
laboratory setting in the real – life setting.
Cook and Diamond’s (1972) definition for a naturalistic setting is “a context that is not
perceived to have been established for the sole or primary purpose of conducting research.”
Thus, the definition explains that the respondent’s perceptions of the setting and not the actions of
the researcher, defines the naturalness of the setting. Thus, Moriarty’s beach scene was a `natural
setting’ as people were unaware that the beach setting had been modified to allow for a
systematic study of the effects of the specification of responsibility on helping behavior, and thus
in their perspective, the setting was natural which is the only requisite for a setting to be defined
as a natural setting.
Thus, the three `natural’ criteria can show considerable interaction and become that feature of the
observational research study, that can lend it more credibility, more scope of generalizability and
thus more value, importance and acceptance as a `good research method’.
Also in Bandura’s (1965) `Bobo doll study’ of aggression, the degree of frustration of the child
just prior to observing the model was observed, the status of the role of the child and the
consequences of the model’s behavior were predicted using observation as a technique, using the
information of which, many independent variables were manipulated and their affect on the
dependent variables was observed.
In exploratory researches using any method, observation as a technique can be used to understand
the context of the phenomenon and the important questions regarding the phenomena to be asked.
Observation can be used in conjunction with interviews in order to observe and record the
inherent non-verbal eves of the participant and also when the interviewer possesses limited verbal
skills (like children and people with learning disabilities) and the like.
Thus, observation can be used as a technique in conjunction with many other instruments as it
adds comprehensiveness, meaning and scope for generalizing the data and the findings of the
research study.
An ‘observational investigation’ is a study carried out only using the method of observation for
collecting data and thus, analyzing it. It thus involves all the characteristics of observational
method such as the presence of an intent, it being carried out in a real-life-setting, usually away
from the labs and finally not involving manipulation of the variables which thus, enables the
formulation of strong correlations but not cause effect relationships between the variables.
The biggest weakness of such a study only employing the observational technique for the entire
process of research is that a cause – effect relationship between the variables measured cannot be
formed due to the vast dynamic interplay of variables in real – life settings where none of the
variables are manipulated. Thus, the reason for a particular phenomenon cannot be particularly
found and generalized.
However, in phenomena that require an exploratory approach, seeking to known about the field
no one treaded into, observational method is the only research method which positively effects
the study. In these studies, as the causes of behavior or the phenomena are more unimportant than
the general primary understanding of the phenomenon, observational method serves the purpose.
However, it is very rarely used as an individual method due to loss of comprehensiveness and
thus loosing out on important aspects of the phenomenon.
Observation, one of the methods of field research holds the strength of observing the
phenomenon as it happens, in the field setting, to get a full understanding of it and thus a deeper
insight into it.
Thus, with such purposes in mind, it can be said the observation is appropriate for research
questions that appear to defy simple quantification, which show study of social processes over
time and those which study attitudes and behaviours best understood within their natural setting.
The above three criteria answer the question “What research questions are appropriately answered
by the method of observation?”
(i) Firstly, observation is appropriate for questions that defy including simple quantification:
Observation, particularly the paradigm of qualitative research as a whole, rejects the idea
that all the subjective human experiences, individual differences and many intricacies that
help in better understanding of a phenomenon, cannot be quantified (always).
(ii) Observation is appropriate for research questions which study social processes over a
period of time.
While survey research and other longitudinal studies are ex-post facto researches where
the study occurs after the phenomenon had already occurred, in a reconstruction of
events, observation gives a picture though, from the beginning to the end of the
phenomenon.
For example: Observation is very useful to examine the rumblings and final explosion of
a riot as the events actually occur, rather than the afterward in a reconstruction of the
events.
(iii) Field research (observation) is especially appropriate for the study of those attitudes and
behaviour; which are best understood within their natural setting.
For example: Observation best suites the studying of dynamics of conversion of religion
at a revival meeting which requires a subjective experience to the data collected.
However “how many men and woman are more likely to convert is better studies through
a statistical analysis.
The term observation derives from Latin, meaning to watch, to attend to. Dictionary definitions
(e.g. Oxford English Dictionary 1989) tend to stress that it is concerned with the accurate
watching and noting of phenomena as they occur in nature; with regard to cause and effect or
mutual relations as approved to an experiment. Reber (1985) extends this definition and points
out all psychological methods involve observation, but stresses that a distinction should be made
between research that is controlled by manipulation of the independent variables and research that
is carried out using naturalistic observation. It was stressed, however, that the observational
method can be successfully used in laboratory settings, the crucial distinction that needs to be
made here is in terms of whether or not deliberate manipulation of variables is involved. Many of
the major developments in psychology came from the initial observation of a serendipitous event,
this as defined by Reber (1985) as the finding of one thing while engaged in a search for
something else. He uses Parlor as an example, including skinner’s search and superstitious
behavior, Piaget’s observation of systematic failure of children on intelligence tests which led to
PSYCHOLOGY PAPER No. : PSY_P3:TITLE: Qualitative Methods
MODULE No. : M15:TITLE: Observation Method (Part-1)
____________________________________________________________________________________________________
Thus, observation which is seen to permeate all research methods in psychology and other
spheres shows a history which sets its definition and scope as a research method.
In some observational studies, researchers observe and record behaviour without intervening
or affecting the behaviour in any way, in a real – life setting, as it occurs, naturally. The
participants are observed as they engage in ordinary activities in settings that have not been
artificially created or set up in the aim of carrying out research.
For example: Researchers have used naturalistic observation to study behaviour during riots,
crowd behavior, other mob events, littering, non-verbal behaviour, play interactions between
children and so on. Such an observation in the context of naturalism, is known as
‘Naturalistic observation’.
Naturalistic observation is not only used by psychologists, sociologists and other people who
study the human behavior, but is also conducted by researchers who are interested in the
behaviour of non-human animals in their natural habitats – ethologists and comparative
psychologists. Examples are studies of tool use by elephants, mating among ignana lizards,
aggression among monkeys (for eg., cheralier – skolnikoff & Liska, 1993). In reality, this
method was inherited by psychology from the ethologists largely (Lorenz, Tinbergen).
(i) The behavior which occurs is more natural and if the target is unaware of the observer,
(which is usually the case), the behavior is unaffected by anxity or the target’s need to be
socially desirable or to impress.
(ii) The study gains the status of a real and ecologically valid study as the observations tend to
more true in the real – life and thus more valid.
(iii) The places of phenomena where it would be unethical to experiment with or intervene or
manipulates variables (eg. In the lives of children and animals), this method is extremely of
use.
(iv) When people are not ready to cooperate with interview or questionna9ire methods and
when the researcher decides that full social content is necessary for the observed behavior
to carry meaning ; these are the places where naturalistic observation takes on an
advantage.
For example: “She waved him goodbye” would be more meaningful than recording “She
raised her hand”
Contrived settings:
Contrived setting involves the observation of behaviour in settings that are created for specifically
observing the behaviour and research purposes. Often these studies are carried out in laboratories
where the extraneous factors are controlled, also manipulated and the participants know that they
are being observed; even though the researchers are concealed behind a one way mirror.
In some cases, there is an amalgamation of both Naturalistic observation and contrived settings
observation. This is when a contrived setting, controlling or creating some factors or variables
specifically for research purposes is done in the “real world”. In these studies researchers setup
situations outside the laboratory to observe people’s reactions and behaviour. In such studies,
people are less or completely unaware of the fact that the situation was created and there is
someone observing them. These kind of studies use confederates who help in creating the
situation for the research
For example: In one study, researchers in the New York City stayed 103 accidents in which a
research confederate staggered and collapsed on a moving subway car. Sometimes the
confederate carried a case and acted as if he were injured or infirm, at other times he carried a
bottle in a paper bag and pretended to be drunk. Two observers then recorded by standers’
reactions to the “emergency”. The purpose of the study was to study the factors affecting helping
staged in an “emergency”.
(i) The extraneous factors are controlled and the variables identified can show a cause –
effect relationship (though rarely) or a high correlation.
(ii) There is lesser chance for observer bias as the phenomenon is anticipated to unfold and
the factors are controlled accordingly.
(iii) Structural data gathering instruments can be used as, if the instruments are exhaustive,
the data can be anticipated to fall in a particular direction.
(iv) Replication of the studies is possible as the same controlled, manipulative setting can be
recreated and thus more evidence can be anticipated to fall in a particular direction.
(v) Replication of the studies is possible as the same controlled, manipulative setting can be
recreated and thus more evidence can be found for against existing evidence.
PSYCHOLOGY PAPER No. : PSY_P3:TITLE: Qualitative Methods
MODULE No. : M15:TITLE: Observation Method (Part-1)
____________________________________________________________________________________________________
(i) Since, the study is undisguised – or the person being observed is aware that he/she is
being observed; they tend to conform to the social norms, are more conscious of their
behaviour and tend to take.
(ii) All the phenomena cannot be studies in a contrived setting, simply because those
conditions cannot be created artificially. For example: To study the behaviour of a person
in the context of a natural calamity (for eg., Tsunami).
(iii) The laboratory set up might yield results with low ecological validity, as they are
artificially created.
(iv) The control and manipulation established might be restricting many more dynamics that
might have enfolded, if the variables were not controlled.
(v) The unexpected aspects of the phenomenon, some new concepts, might be lost, due to
barring them by establishing greater control.
(vi) It might sometimes have a great number of or serious “ethical impediments” as
interference of the researcher takes place.
(i) In this kind of study, the degree of control required can be established or the variables
can be manipulated without actually the individuals being aware of it.
(ii) There is mostly no influence of the set up and the researcher/s in this kind of study, if the
individual does not recognize the contrived situation.
(iii) The individual, not aware that he is being observed due to no interference from the
researcher and also in the setting ; tends to exhibit his actual / original behaviour without
the impediments of faking to conform to the societal norms.
(iv) This study has both the advantages of a contrived setting (like replication, use of
structured data collection instruments) and a naturalistic setting (ecologically valid, actual
true behaviour obtained etc.).
(i) Any mistake by the confederates may reveal the whole plan and the purpose of the
research would not be fulfilled.
(ii) It might be unethical in two ways:
(a) Towards the confederate: The confederate might loose his life while doing a
risky act.
(b) Towards the participants: As they are not informed when they are observed. It
might also be “Intrusion into their private lives.”
(iii) It would involve a lot of risk or even mediocre amounts of it, if the situation is that to
create panic or anxiety in the individual.
Undisguised Observation:
Undisguised observation happens when the participant knows that he / she is being observed.
This thus leads to a problem of reactivity as people, knowing that they are being observed do
PSYCHOLOGY PAPER No. : PSY_P3:TITLE: Qualitative Methods
MODULE No. : M15:TITLE: Observation Method (Part-1)
____________________________________________________________________________________________________
Disguised observation:
When the researcher does not let the participant know that they are being observed for
research purposes, then the observation is known to be “disguised observation.”
For example: Festinger and his colleagues (1956) used disguised observation when studying
the doomsday group because otherwise they would not be even be allowed to observe the
group otherwise.
(i) The part of being observed by someone else is not known; which allows them to
exibit their own behaviour and reduce faking and socially desirable to the minimum.
(ii) There is reduction of error, increase in ecological validity
(iii) The findings would be more generalizable to the population; or through `evaluation’
of the findings and explain or explore new aspects of the phenomenon.
(iv) They, in terms of the behaviour elicited, are make appropriate or close to all the
naturalistic observation.
(i) They raise ethical issues, because of “invasion of privacy” and violation and also
violating the participant’s right of informed consent.
(ii) Will include “knowledgable” informants – people know who the participants, in case
the study becomes faith and true. It involves one extra work of training these
informants or speaking to them who themselves are not proper researchers and thus
stand a chance of adding to the error of the observational study.
Disguised observation can also be carried out in Laboratory settings, where the participants
are highly conscious of what they are doing. Researcher William Ickes proposed an approach
known as dyadic interaction paradigm which was often used to study dyadic, or two-person,
social interactions etc., in which a pair of participants reporting from experiment are escorted
to a waiting room and seated on a couch when the researcher excuses himself to complete
preparations for the experiment and leaves them alone. Unknown to the participants, their
behaviour is them recorded by a concealed videotape recorder, but this was analyzed or
observed only after the participants give permission to do so. Thus, this method stands
PSYCHOLOGY PAPER No. : PSY_P3:TITLE: Qualitative Methods
MODULE No. : M15:TITLE: Observation Method (Part-1)
____________________________________________________________________________________________________
Unstructured observation:
Researchers who decide to adopt the technique and may have a clear idea of the purpose of
observation, they may not be very clear about the details of observation. This kind of observation
is generally used in exploratory studies. The researcher `does not use predetermined categories
and classifications but makes observations in a more natural open – ended way ….. The logic
here is that categories and concepts for describing and analyzing the observational data will
emerge later in the research, during the analysis, rather than brought to the research, or imposed
on the data, from the start.’ (Punch, 2005).
As in grounded theory, the researcher `postpones the definitions and structures until a pattern has
been observed ….. and then continues with the fieldwork in order to elaborate these while the
data are still available for access ‘ (Bouling, 2002).
Unstructured observation can be useful to generate hypotheses and theories, but is difficult to
manage. Its qualitative in nature and this can sometimes be referred to as qualitative observational
research. Thus usually involves taking field notes and such unstructured means to collect data
where the field holds surprises.
(i) Not imposing any particular limit to the data on the behaviour in any way, this form of
observation is more open to new concepts and ideas that may be derived from the field.
(ii) New concepts may arise and new theories may develop which may ehlp to further
knowledge in an unrestricted manner.
(iii) It includes bias as structured observation have certain aspects to observe outlined which
depends strongly upon the researcher’s subjective bias or understanding what is important
and what is not.
(iv) It helps in greater understanding of a phenomenon in a more subjective manner which is
one quality of qualitative research.
Structured Observation :
Structured observation is when the research problem has been formulated and the assistant
observes have been trained and told what is to be precisely observed. However, this process
systemizes the process of observation to a great deal to make it objective and less influenced
by bias. This method uses recording techniques like checklists, ratings scales etc., which are
more structured and systematic with respect to what is to be measured.
(i) The fact that different observes can, and do, produce different accounts of situations can be
handled by using structured observation.
It is precisely this problem which is addressed by systematic (structured) observation
and its use of an observation schedule. The whole purpose of the schedule is to
minimize, possibly eliminate, the variations that will arise from data based on
individual perceptions of events and situations. Its aim is to provide a framework of
observation which all observers will use (Denscombe, 2007).
Thus, it gives objectivity to the research.
(ii) It can be replicated to some extent exactly.
(iii) It helps to not miss out any aspects which are of extreme importance for the study. It also
helps in quicker recording of the aspects of the phenomenon.
(iv) The observer bias is not much in the field, however while construction of the observation
schedule, the bias enters as only those observations anticipated to be important find a place
in the schedule.
Thus, the structuring of data or and observation can be done by structuring the data gathering
techniques which are structured and systematic like check-lists, rating scales, charts etc., which
will be discussed in detail later in the report.
7. Summary
Thus observation research used in the sense of a research tool as well as a research design in itself
can be seen to be characterized by the features of naturalism, exploration, intent and purpose and
various others. It is a method of research whose situations fulfill the ‘Tunnell’s criteria’. The
various types of observational research along with the history of the discipline are outlined above.
The applications, advantages and disadvantages of every type of observational research had been
presented for critical evaluation of the same. Thus, observation, an effective research method can
be applied to appropriate research questions in order to reveal interesting patterns of human
behavior.
PSYCHOLOGY PAPER No. : PSY_P3:TITLE: Qualitative Methods
MODULE No. : M15:TITLE: Observation Method (Part-1)