rm-lec-9

OBSERVATION

INTRODUCTION

Observation is one of the oldest methods of social research. It was initially employed by social anthropologists and ethnologists, with sociologists and other social scientists largely preferring surveys and other techniques. In time, however observation became popular outside anthropology and ethnology. Today, it is proclaimed as one of the central techniques of social research.

Observation entails gathering data through vision as its main source. It may be used as the only technique of data collection, or jointly with other techniques, such as intensive interviewing, documentary study or case studies. Although observation studies people, it focuses also on objects as products of human action or just as part of physical environments.

1          TYPES OF OBSERVATION

Observation can study all observable social phenomena, as long as they are accessible; it is obviously not possible to observe personal, sensitive issues or causes and consequences of social phenomena, nor past and future events. Observation takes several forms which, although basically similar, differ in a number of ways (see Box 10.1), for instance, in the degree of the observer's participation in the field of observation, and in the extent to which it is structured and standardized. This distinction results in the basic types of observation listed below

Participant and non-participant observation

In participant observation, researchers join the group they intend to study and observe it from the inside; ideally it is not known that they are researchers. For instance, researchers who wish to study homosexual behavior pretend to be (or are) gay, join homosexual groups and conduct their study from within. In a similar fashion, investigators who want to study the work conditions of factory labourers join them as labourers, working alongside them and observing them "from the inside'. As members of these groups, they can study, among other things, their structure, process, problems and attitudes, both directly and as experienced by the members of the groups.

In non-participant observation, investigators study their subjects 'from the outside'. Their position is clearly defined and different: from that of the subjects. Ideally, they are 'invisible', and remain unnoticed by the members of the group they observe. A typical example of non-participant observation is laboratory observation, where the subjects interact in a laboratory and the researcher observes them through a one-way mirror. Observing children in the school playground through a window is another example of non-participant observation. In both cases the observer does not actively participate in the group under study

Types of observation

Factor	Type of observation
Structure	Structured: entails strict design and control Unstructured: entails a flexible design and no control
Observer's role	Participant: the observer is a part of the setting Non-participant: the observer is not a part of the setting
Observer	Self observation: observer observing self, e.g. using videos Other observation: observer observing others
Focus	Human observation: focuses on people and their activities Physical observation: focuses on objects, e.g. artifacts and physical remains
Setting	Natural observation: observation in natural settings Laboratory observation: observation in laboratories
Observer's commitment	Active observation: observer is committed to the cause of the study (i.e. is ideologically, and sometimes personally, involved in and supports the overall purpose of the study) Passive observation: observer is not committed to the cause of study.
Depth of the study	Naive observation: simple, unstructured observation Scientific observation: systematic and highly structured
Observer's identity	Open observation: the observer's identity is known Hidden observation: the observer's identity is not known

Subjects are not always aware that they are being observed and the subject of a study in general. Subjects usually change their behaviour when they know that they are being observed, so researchers may take measures to prevent them realizing that they are being studied, and hence eliminate the effects of reactivity. Where this is not important, the identity of the observer and the existence of the study are made known. The question that arises in this context is one of ethics; the researcher should have a very good reason choosing the appropriate type of observation.

Many types of observation lie somewhere between the two extremes of participant and non-participant studies. In certain cases investigators are more 'observers' than 'participants'; in others they are more ‘participants' than 'observers'. Consequently, the difference between the various types often lies in the degree of participation/observation rather than in the nature of observation itself.

Structured and unstructured observation

Structured observation employs a formal and strictly organized procedure with a set of well-defined observation categories, and is subjected to high levels of control and standardization. It is also organized and planned before the study begins. Here, the researcher specifics accurately what is to be observed and by what means, and how the results of observation will be recorded.

Unstructured observation is loosely organized and the process of observation is largely left up to the observer.

Semi-structured observations lie somewhere between these two techniques; they may, for instance, be structured in their approach but unstructured in their settings. They are relatively common in social research and combine the advantages (and limitations) of both the structured and unstructured techniques observation.

Ten criteria of structured and unstructured observation

Structured observation	Unstructured observation
Employs a strict design Is a non-participant observation Employs high levels of standardization Focuses on aspects of the setting Works in laboratory or natural settings Mainly studies small groups Is a formal observation Is unobtrusive (unremarkable) observation Is direct observation Employs mostly a quantitative design	Employs a flexible design Is a participant observation Employs low levels of standardization Focuses on the whole setting Works in natural settings Studies small and large groups Is an informal observation Is also unobtrusive observation May be direct or indirect observation Employs mostly a qualitative design

Other types of observation

·         Human and physical observation: The former focuses on humans and human action; the latter focuses on objects.

·         Naive and scientific observation: Naive observation refers to everyday, unstructured observation which people use when they interact with others in social situations. Observation becomes scientific when it is systematically planned and executed, when it is related to a certain goal, and when it is subjected to tests and controls.

·         Natural and laboratory observation: The former takes place in natural settings, the latter in laboratories.

·         Open and hidden observation: In open observation, the participants are aware of the identity of the researcher as an observer and the purpose of the study; in hidden observation, subjects are not aware that they are being observed.

·         Active and passive observation: Active observation entails high involvement and commitment of the observer to the cause of the study; passive observation does not. In the latter, the role of the observer is strictly professional, concerned with the recording of data. In this ease, observation is a job to be done in an objective and neutral fashion.

·         Direct and indirect observation: Direct observation studies people individually or in groups. Indirect observation studies people through the products of their action. In the latter, researchers may observe, assess and analyze artifacts and traces in general left behind by people who are no longer accessible and through this draw conclusions about their lives.

·         Self-observation and other-observation:  In other-observation the actor observes other people; in self-observation observers observe themselves, for example by focusing directly on how they react in particular circumstances, or indirectly by watching videos showing their interaction, such as the manner in which they handled critical aspects of observation.

These types of observation are not mutually exclusive. An observation can, for instance, be participant, active and open at the same time.

2          THE PROCESS OF OBSERVATION

2.1       The research model

Observation takes the same form as the general research model introduced earlier in this book. However, the content of each step includes elements that are more or less influenced by the nature of observation. The following is a brief summary of the basic steps of research as employed in the area of observation, mainly by quantitative researchers. Qualitative investigators may use the same steps but their content will have to be adjusted to the principles of the underlying theoretical framework.

Selection and formulation of the topic

At this stage the object of observation is clearly stated. For example, it must be stated whether observation will focus on action, speech, attitudes or behaviour. No researcher enters a field of study without a firm idea of what is to be studied. As well as identifying the unit of study, the researcher often outlines the logical and normative structure of the study.

In quantitative studies and structured observation, the topic is precisely defined so that the observers are well aware of the specific elements of the object to be observed. In addition, specific categories will be developed, which will help the observer to categorize the material (i.e. behaviour, relationships and so on). Categorization can be based, for instance, on standards related to the purpose of the study, the degree of detail required, the interests of the researcher, the expected response of the observed and the nature of the situation. For instance, in a class observation, 'teacher centered', 'student centered', 'authoritarian', 'democratic’ 'sexist' and 'non-sexist' are some of the categories that can be employed. These categories will finally be operationalized by identifying criteria that indicate their presence. Type of language, type of speech, cone of voice and facial expression are a few examples.

These categories and indicators are developed after adequate exploratory work has been considered and before the observation has begun. Once these categories have been formulated and refined, they are employed in the same form throughout the study. This allows the researcher to categorize the units of analysis; it also allows comparisons to be made. Category construction will, finally, enable investigators to direct their attention to the elements close to the goal of the study, to collect valid and accurate data and, where multiple observers are employed, to produce more accurate, detailed and comparable findings.

In qualitative research, and where participant observation is employed the topic has to be chosen; the definitions, however, may not be very specific, and the categories - if employed not explicit or deterministic. In most cases, categories will be developed during observation. The process of observation in this context is flexible and open in its approach.

As noted above, during this step of research the investigator will choose the theoretical and methodological framework, and hence the type of observation: structured or unstructured, participant or non-participant and so on. With regard to the observation type, the investigator will also determine the role of the observer in the setting. In structured observation there is no flexibility in the role of the observer; here the observer will most likely be formal and detached.

In participant observation, there are many more options available. For example, complete participation (being a full participant), where observers are fully absorbed in the group under study; partial participation and observation (being partly participant and partly observer); and complete observation (being purely an observer).

Sampling procedures

Having established the details of the observation topic and unit, as well as the methodological parameters of the study, the researcher turns to the more practical aspects of the project. The next task to be completed is the choice of the subjects.

Where a strict quantitative design is employed, selection of the respondents is largely made by means of probability sampling. With respect to qualitative designs, for example where unstructured or participant observation is employed, the subjects are often chosen by means of purposive or theoretical sampling. In both contexts, sampling refers to more than just the selection of the subjects (see Mahr, 1995), as we shall see next.

Time

Observers have to decide when observation will be carried out. This is more significant in the case of participant observation, since certain times might offer different environments and experiences than others (days, weeks, months etc.). For this reason, choosing a definite time will have implications for the type, quality and quantity of information gathered. The structured observer does not necessarily need to comply with such requirements, since it is expected that observations will take place under controlled conditions (including time).

Duration

Having decided when observation will be conducted, the researcher will consider its duration. This entails firstly the length of each session (e.g. one hour during lunch time), and secondly the length of the study (e.g. every day for three months,). Hence, the duration of the study is: one hour every day during lunch time, for three months. The commencement of the study will be determined when considering the 'time' (see above).

Place

Sampling refers also to the place in which observation will be conducted. If schools, hospitals or clubs are to be observed, for instance, one has to determine where in these systems observations will take place: that is, in which room, level or specific location.

Type of event

The type of event to be studied also has to be determined. Will the researcher observe everything, some events, routine events, unexpected events or special events?

Arrangements

As well as the sampling procedures, the investigator decides about the arrangements for entering the setting and recording the data. Entry into the setting is relevant for participant observers and is a very important aspect of observation. It chiefly involves getting permission to enter the environment in question, which is often no easy matter. While it may be relatively easy to observe children playing in a public playground, it is quite difficult to gain entry into a school, goal, street gang, gay club or certain government committees. Arrangements have to be finalized before the process of observation can begin.

The observer

As in other research methods, so in observation, the researcher will decide on the person(s) who will collect the data. Apart from this, the investigator will assess the nature of observation and, alongside this, the attributes of the observers. Finally, this decision will also indicate whether the observer is expected to possess certain skills. Having addressed these issues fully, the researcher will proceed to the next point of the research mode.

Observer skills

The quality of the observer is often more significant in die context of observation than in other forms of data collection. This is because observation, particularly participant observation, relies very much on the attributes of the researcher to gather information in both quantity and quality. For this reason, observers must be chosen carefully. Their qualities may vary from case to case, particularly with reference to the type of observation chosen, but some qualities and skills are valued more highly than others (Pfeifer, 2000). Here are some examples of qualities required within the quantitative research paradigm:

·         general personal ability in terms of perception and memory

·         knowledge of the field of research and the surrounding (sub)culture

·         specific knowledge of the topic

·         previous experience of observation and other research

·         ability to handle crisis situations

·         flexibility and adaptability

·         respect for the boundaries between the observer and the observed

·         ability to feel the power of culture in everyday life

·         ability to get along with others

·         ability to follow instructions

·         ability to control personal biases and ideologies

·         honesty and trustworthiness

·         awareness of and respect for ethical standards

Observer skills vary from case to case, depending particularly on the theoretical and methodological context of the project. Participant observers working within a quantitative context will be expected to be aware of its requirements and to possess skills that may be different from those needed within a qualitative context.

Observer training

In most cases the nature of observation makes it necessary for the investigator to carry it out alone, particularly in participant observation and in qualitative and case-study research. In other cases more than one observer may be employed. Multiple observers usually observe their groups separately and produce data which are included in the final analysis. The use of multiple observers speeds up data collection but can also cause problems, especially with regard to inter-observer variability. Where more than one observer is employed and where the observer is not the investigator, training becomes essential.

When training is required, it often concentrates on issues that are central to the study, possible sources of distortion, aspects of the study that require further explanation, and most of all on technical observation skills. Pilot: studies will certainly show die way.  However, focusing on die points listed in Box 10.3 isthought by many writers to be very useful (sec, for example, Becker, 1989; Flisk ct al., 1991; Martin, 1988).

Even when only one observer will be involved in the study, training is important. What will be observed, when and how, are issues with which the observer must be very familiar. The extent of involvement is another issue. Also, becoming a genuine participant observer is a difficult task, and observers seldom reach that stage (Wolcott, 1992: 20). Training is always helpful.

Focus of observer training Observer training focuses on: ·         thorough understanding of the research topic ·         knowledge of the peculiarities of the population ·         understanding possible problem areas of the study ·         familiarity with the categories (where appropriate) and their effective use ·         introduction to ways of overcoming unexpected problems and conflicts ·         ability to follow instructions accurately and adapt them without causing bias or distortion of the data ·         adaptability and flexibility ·         ability to observe several subjects and categories at the same time

Collection of the data

During this stage the researcher executes all the instructions and employs all the techniques outlined in the research design. In practical terms, the observer approaches the subjects and collects the information. Only a few general points will be mentioned here.

Initiation

The initial duties of the observer are to prepare and introduce the appropriate setting, and to offer adequate instructions. More particularly, in structured observation the observer approaches the subjects and invites them to the laboratory, explaining their task in detail; this might involve offering a couple one hundred dollars and asking them to discuss how they would spend it. How much of the process of observation will be disclosed to them depends on the nature of the study.

If the structured observation takes place in natural settings the approach is similar. The subjects are, under normal circumstances, not informed of the fact that they are being observed, and no arrangements will be made regarding die setting. The observer visits the subjects and observes them, without them being aware of this. Recording will take place here or in laboratory observation in a manner determined by the investigator.                                 

In qualitative observation, for example participant observation, the choice of the respondents and the initiation of the study are rather different. As stated above, no random sampling procedures are used here, and the participant is more than just an informant. The observer enters the field but aims to remain invisible and, most of nil, not to affect the structure and functioning of the setting. In particular, the observer is expected to respect the observed, to be understanding and tolerant, and to be familiar with the lifestyle of the observed. The observer-observed relationship is close, based on cooperation, mutual understanding and mutual trust.

Data collection

In participant observation, data collection begins after entry to the setting is gained. Notes are taken in the way that best suits the circumstances. When the framework is qualitative, collection and analysis of data often take place simultaneously. Observation focuses on research units over a set period of time. In this sense data collection can relate to various time spans and therefore focuses on different time frames, generating different types of data collection. Below are a few examples.

·         Continuous observation: In its most common form, observation is continuous; this means it records occurrences for the entire duration of the event. For instance, the observer will record the activity level of camping children for a period of four weeks.

·         Time-point observation: Data collection can also focus on a specific point (time-point); here the observer is interested in what happens at a particular point in time; for example recording the place where elderly patients are at precisely 9.00 am. Are they still in bed sleeping, staring at the ceiling, reading, talking to others, walking around, helping others and so on? Time-point observation produces 'snap-shot' data, like a picture, separate from context or time frame.

·         Time-interval observation: Between continuous and time-point observation is time-interval observation. Here, data collection is focused on what happens between two set times, say, between 9.00 am and 9.10 am. We assume that the observer has good reasons to collect data during that period and record everything that is significant.

·         Event observation: This form of data collection relates to behaviours that occur as a result of other behaviours or events. The presence of an observation unit is conditional on the occurrence of another event. In our example, the activity levels of elderly patients in a geriatric ward are recorded after the doctor entered the ward. Here the observer is interested to see whether the patients are more/less active when the doctor is present than otherwise.

Recording

Recording of data is an issue that must be considered during the planning stage of the research; three issues are significant here: what will be recorded, when and how. This refers to the methods of recording, the events to be recorded, and the coding method.

Time focus of observation: four options ·         Continuous observation focuses on activities over the entire period of the study. ·         Time-point observation focuses on activities at a specific point in time. ·         Time-interval observation focuses on activities during a specific time interval. ·         Event observation focuses on the nature of specific events for as long as they occur.

Methods of recording

The method of recording varies from one type of observation to another. In quantitative research, recording of structured observation entails identifying, the presence and frequency of occurrence of the various categories. This requires concentration so that all categories are recognized and adequately recorded. The process is rather technical and mechanical, and allows quantifiable responses. Recording varies according to the type of events studied, the density of information and the size of the group.

 In qualitative research, the method of recording varies with the research topic and the observer's degree of familiarity with the available methods. In general, methods are constructed in a chronological or a systematic way. The most common methods of recording are: writing down information verbatim, in summary or in key words; tape recording conversations; video recording events; and taking photographs.

Note taking is the most common method, but is not always possible. For instance the information to be recorded may be too dense or there may be too many persons for notes to be taken, and the observer may not wish the subjects to be aware of the study. Apart from this, taking notes may divert the attention of the observers from the scene to the paper, causing them to miss part of what happens in the group. If circumstances do not allow note taking, the observer should write down key words or phrases as a guide, and complete the notes after observation, or leave the scene briefly to write up important notes.

Tape and video recordings are easier and certainly more efficient. The tapes can be listened to or viewed as many times as required, using more than one observer if desired, and this produces more accurate and more valid records. However, there are cases when recording is not possible or respondents object to it, and this limits its use. Apart from this, recording adds to the work of the observer – the task of writing up notes is merely postponed, and a lot of the 'recorded information is often of no use. The last of the methods we listed earlier, taking photographs, is of limited use.

Events

Recording is done within the context of a methodological framework. In quantitative studies observation is conducted on the basis of the observational categories developed by the investigator before the observation process. Such categories relate to items and subjects, indicating the existence of some characteristics, their content, trend and intensity. Observation may focus on the content of discussions, on feelings, facial expressions, aggression, patterns of communication and behaviour, or on general issues and items defined through the operationalisation process.

In qualitative studies observers may initially record whatever happens around them indiscriminately (that is, regardless of whether it seems at that point in time to be important or not), and keep accurate and detailed records and complete notes. Description of the setting, persons, discussions, relations and so on is the rule. Over the course of time, knowledge of the setting increases, and with it understanding of how relevant happenings are to the research topic. This leads to the establishment of screening mechanisms, which allow the observer to become more focused and more selective. Data collection and recording entail an element of analysis in their structure.                                             

Coding

Where observation categories have been developed and the items of observation are clear, specific and known in advance, codes can be used to record the data. Codes serve as symbols, a shorthand recording, where actions and behaviours are replaced by numerals or keywords. This makes recording easier, particularly when there are many items to be recorded and many people to be observed. If categories are distinct and easily identifiable mechanical devices can be used to record observation data. The interaction chronograph developed by Chapple, the audio-introspectometer develop by Thelen, and the interaction recorder developed by Bales and Gerbrauds are a few examples.

In quantitative research, codes are the result of careful operationalisation and accurate definition of the indicators. This process specifics accurately the aspects of behaviour that need to be observed in order for the object of study to be identified and assessed. Codes tell the observer what to look for and, in a way, what to ignore. They will tell observers what constitutes attention in primary school children and how to identify its indicators. While qualitative researchers will observe children at school in general, quantitative observers will focus on a specific aspect of school experience and will assess aspects of a child's behaviour.

Analysis and reporting

The way data are analysed and communicated to the public is an important aspect of the research process, and will be discussed in another chapter. At this stage it is important to note that in quantitative research, the frequencies of occurrence of the various categories identified during observation will offer the basis for a statistical measurement and analysis, employing appropriate statistical tests to establish relationships and allow explanations, comparisons and eventual predictions. In contrast, where qualitative research is employed, collection and analysis of the data are often interwoven and take place concurrently. This is the case in what we termed flexible qualitative design. The outcome of this analysis will help to generate the final report.

3          PARTICIPANT OBSERVATION

Participant observation is one of the central methods of observation in social research in general, and in qualitative research in particular. It is therefore important that a few additional comments are made on this method as employed within the qualitative paradigm. This is more relevant with regard to the major criteria that distinguish it from other forms of observation. The most important characteristics of this method are shown below.

Main criteria of (qualitative) participant observation Qualitative participant observation: ·         Demonstrates a commitment to studying everyday events, which are studied in terms of the way they are experienced and understood by the participants. ·         Is conducted in a natural environment. In this sense, observation remains natural and authentic. ·         Observation is designed to study social events under all conditions, bringing data close to reality, the people living in it and the way they construct and experience it. ·         Sets data collection within face-to-face interaction. ·         Conducts data collection in an unstructured mode. ·         Employ open and flexible methods. ·         Perceives reality as constructed through the interaction and communication of the participants ·         Addresses reality in an interpretative manner

More particularly, participant observation is characterized by the fact that it observes communication and interaction in an unstructured and natural manner, where the design is developed and modified while observation is carried out, in a face-to-face relationship, and in an open and flexible fashion.

4          PHYSICAL OBSERVATION

Physical observation focuses on objects, some of which are part of the physical environment and others the product of human behaviour. There are two types of physical observation; the observation of objects used by the subjects or affecting subjects in some way, and the study of physical traces such as artifacts

4.1 Observation of objects 

This relates to observation of objects that have a significant impact on people’s lives. They are useful in at least two ways: first, they disclose information about people's attitudes and behaviour; and second, they verify, falsify or amplify information otherwise offered to researchers. These types of observation may be employed as the only method of study or in addition to another method.

Example:

·         Marital status of two people who are known to be sharing the same apartment as tenants.

·         Mother who is fighting to retain legal custody of her son, defending her case in an interview with a social worker at her home, may tell a convincing story but will find it difficult to win the case if the household environment reveals a different reality: a disorderly home, dirty surroundings, empty syringes next to her bed, empty whiskey bottles, evidence of heavy smoking and so on.

4.2       Observation of physical traces

Physical traces are products of human activity that provide information about certain social trends, habits, behaviour patterns and cultural configurations of a group of people or a community.

The concept of 'traces' has been used by researchers in two ways.

·         In some cases, 'traces' are physical objects only, such as household garbage in public spaces.

·         In other cases 'traces' are symbols marks or drawings. Drawings in caves and graffiti in toilets, for example, are 'traces' that provide important information about the people who use(d) those environments.

·         More specifically, the study of physical traces can offer information about the culture and life of past communities

There are several areas in which the study of physical traces can be very useful.

·         By observing the floor of a museum, one can draw some conclusions about the popularity of certain exhibits.

·         Social researchers also use physical traces to infer the habits of certain population groups or individuals. A researcher examined the household garbage bins of single mothers to ascertain the kind of food they eat.

4.3       Research process

Data collection and analysis follow the general pattern of the research model, beginning with the identification of the research topic, followed by the collection and analysis. Both qualitative and the quantitative methodology can be applied depending on the type of physical observation. When studying physical traces, unstructured observation can be applied, or in some cases, structured observation will be possible. b while interviewing, measuring their extent and quality as required by the research design. Physical observation is not as common in social research, and certainly not as prominent within observation, as human observation.

5          ETHICAL ISSUES

As in other methods of data collection, ethical issues are highly relevant and require due consideration. The nature of observation makes ethics a real issue, since observers have the opportunity to interfere directly with the personal life of the subjects, and since in many cases observation takes place without the subjects being aware of it. This becomes even more serious when the subjects do not know the identity of the observer, for example when the observer pretends to be a genuine member of the group under study.

For some writers ethics is not an important issue and should not hinder researchers from pursuing their research interests if they think that the investigation is carried out for a good purpose. They believe it is justifiable for them to conceal their identity if the research will eventually benefit society and perhaps the subjects. For most, however, ignoring ethical issues is not acceptable; observers should disclose their identity when entering the private domain of individuals, they should disclose their real intentions and the objectives of the research, and they should be honest about their intentions.

6          STRENGTHS AND WEAKNESSES OF OBSERVATION

Strengths (Mahr, 1995):

·         Observation provides information when other methods are not effective.

·         It employs a relatively less complicated and less time-consuming procedure of subject selection.

·         It can offer data when respondents are unable and/or unwilling to offer information.

·         It is conducted in a natural setting, and studies events as they evolve; this is particularly so in qualitative research.

·         It offers first-hand information without relying on the reports of others.

·         It allows the collection of a wide range of information, even when this information is thought to be irrelevant at the time of study. This is particularly true of participant observation.

·         It is relatively inexpensive.

Weaknesses (e.g. Becker, 1989; Mahr, 1995)

·         It cannot be employed when large groups or extensive events are studied.

·         It cannot provide information about future or unpredictable events.

·         It cannot offer data related to frequency of behaviour.

·         It cannot study opinions or attitudes directly.

·         It is inadequate when studying sensitive issues, such as sexual behaviour or family violence.

·         It is relatively laborious and time-consuming.

·         It is vulnerable to the observers' bias, and dependent on their selective perception and selective memory.

·         In participant observation the observer is a part of the situation that is being observed.

·         It offers no control measures to balance the bias, attitudes and opinions of the observer.

·         It cannot offer inductive generalisations of the results.

7 PROBLEMS OF OBSERVATION

Observation can be affected by a number of problems. Some of these are caused by the observers, others by the nature and purpose of observation as well as by methodological arrangements.

7.1       Sources of errors

Errors are in practice associated with all aspects of observation if studies are not effectively prepared and designed. The purpose of observation can cause errors if the observers do not understand it correctly. Likewise, arrangements for observation may make recording cumbersome and at times impossible, leaving important information unrecorded. This is most: obvious when codes are inadequate or inadequately defined. But most of all, the observer can be the source of many, errors, a point that motivates researchers to take adequate preventive measures. A brief list of such errors is given below:

The observer as a source of errors

Lack of ability: Errors may occur due to lack of ability, interest or willingness, as well as tiredness, boredom or attention problems.

Observer inconsistency: Inconsistency occurs in two ways: first, when the observer is unable to perform observations in exactly the same way throughout the investigation; and second, when several observers are employed, and they do not operate uniformly.

Observers’ non-verbal communication: This may affect the attitude and expectation of the observed and influence their behaviour accordingly.

Observer bias: This refers to observers' consistent tendency to perceive situations in terms of their personal ideology and bias, for example through selective choice of data and selective coding, producing a distorted reality,

Deviation: The behaviour of the observer and his or her relation with the observed may deviate from the prescribed and expected behaviour.

Deception: Observers might deceive the researcher; this is a problem that is difficult to detect and control, particularly in participant observation.

Lack of knowledge: Some observers may lack knowledge of the subject matter, the context of observation, or the categories employed in the research.

Problems in recording and analyzing data: Facts may not be recorded accurately and the analysis might be non-systematic and subjective.

Lack of familiarity with the observed group: The observer may not be familiar enough with the group to be observed.

Observer distortion: The settings observed are distorted by the fact that the observer is added to the structure under study (participant observation), or that the respondents have to operate in an artificial environment (laboratory observation).

Expectations: Expectations about responses can bias the outcome of observation.

7.2       How to prevent errors

The researcher should make a serious effort to prevent errors and, where these have occurred, to reduce their significance and eliminate their consequences. It is generally expected that the researcher should:

·         Take the necessary steps to construct a design that will avoid errors as much as possible. Such steps could include making the study object and goals clear and the categories specific and easy to understand, providing effective tools for recording the data and, where possible, offering an environment that will have the least possible negative effect on the research process as well as on the quality and quantity of the data.

·         Choose observers and research assistants who have the ability and interest to work on the project, that is, who possess the personal qualities required for the project, have the capacity to learn and adjust to the research needs as required, and are willing to work under diverse and demanding conditions with people of diverse nature and origin.

·         Train observers adequately so that they have full knowledge of all aspects of the project, its goals and problems, and particularly the categories and their accurate and meaningful application.

·         Supervise observers so that

1.    eventual deviations can be identified and controlled

2.    advice can be given when questions arise

3.    plans and categories can be improved through experience

Errors occur in all types of research, and observation is no exception. Errors are a part of any investigation. It is the cask of the researcher to make the necessary arrangements to prevent problems and, where errors are suspected, to adjust the analysis and interpretation of data accordingly.

Main Points

·         Observation is a method of data collection that employs vision as the on technique of collection.

·         There are several types of observation, including naive, scientific, participant non-participant, structured, unstructured, natural, laboratory, open, bidder active, passive, direct and indirect observation.

·         Quantitative researchers are more interested in structured observation; qualitative researchers prefer unstructured observational designs.

·         The steps of the research model employed in observation are similar to those of the standard model introduced earlier in this volume.

·         Sampling procedures in observation are similar to general procedures in this area.

·         Given the nature of the method, the observer is expected to be well qualified and have the required skills.

·         As a method of data collection, observation appears in a number of ways, including continuous, time-point, time-interval and event observation.

·         The study of objects and physical traces expands the scope of observation and strengthens its significance as a research method.

·         Observation is a unique method that can be employed in areas where other methods are unsuitable, and therefore offers several advantages.

·         Problems of observation can relate to a number of issues, for example, the observer, the purpose of observation, the tools used, the categories of observation and the expectations of the researcher.