Observational Methods of Research , Qualities and Issues (Part I)

Observational Issues, Units and Methods

Observational Methods of Research , Qualities and Issues The Observer-Observed Relationship Qualitative Observational Methods: Participant Observation.

    Research observation involves the systematic selection, observation, and recording of behaviors, events, and settings relevant to a problem under study. Like self-report methods, observational methods include both unstructured methods that primarily yield qualitative data and structured approaches that yield mostly quantitative data. 

    Both approaches are discussed in this chapter. First, however, we present an overview of some general issue.

Observational Issues

    When nurse researchers observe an event such as a nurse interacting with a patient they have to know (at least broadly) what to observe. Researchers cannot absorb and record an infinite number of details; they need guidelines about how to focus the observations. In this section, we describe the versatility of observational methods and note some considerations for observing phenomena. 

    Phenomena Amenable to Observations Nurse researchers make observations of human behaviors or the characteristics of individuals, events, environments, or objects. The following list of observable phenomena is suggestive rather than exhaustive:

1. Characteristics and conditions of individuals. A lot of information about people's attributes and states can be gathered by direct observation, including both enduring traits such as physical appearance and more temporary conditions such as rashes. 

    Physiologic conditions can be observed either directly through the senses, or with the aid of observational apparatus, such as a radiograph.Examples of this class of observable phenomena include patients' sleep wake state, the presence of edema in congestive heart failure, alopecia during cancer chemotherapy, or symptoms of infusion phlebitis in hospitalized patients.

2. Activities and behavior. Many actions are amenable to observation and constitute valuable data for nurse researchers. Activities and behavior that indicate health status or physical and emotional functioning are particularly important. The following kinds of activities or behaviors lend themselves to observational . 

    Study : patients' eating habits, length and number of visits by friends and relatives to hospitalized patients, and aggressive actions among children in the hospital playroom.

3. Skill attainment and performance. Nurses are routinely called on to develop skills among clients. Skill attainment is often demonstrated behaviorally, and is therefore amenable to observation. As examples, nurse researchers might want to observe the following kinds of behavior: the ability of stroke patients to scan a food tray if homonymous hemianopia is present, diabetic patients' skill in testing their urine for sugar and acetone, or infants' ability to suck when positioned for breastfeeding. 

    Nurses' on-the-job performance and decision-making behaviors are also of interest to researchers.

4. Verbal communication. The content and structure of people's conversations are readily observable, easy to record, and, thus, a potential data source. Among the kinds of verbal communications that a nurse researcher may be interested in observing are information given by nurses to patients, exchange of information among nurses at change-of-shift report, and conversations among nursing home residents.

5. Non-verbal communication. People communicate their attitudes and emotions in many other ways than just with words. The kinds of nonverbal behavior amenable to observational methods include facial expressions, touch, posture, gestures and other body movements, crying or laughing, and extralinguistic behavior (ie, the manner in which people speak, aside from the content, such as the intonation, loudness, and continuity of the speech).

6. Environmental characteristics. People's surroundings can affect their behavior, and numerous studies have explored the relationship between observable environmental attributes on the one hand and human actions and characteristics on the other. Examples of observable environmental attributes include hospital noise levels, nursing home decor, safety hazards in an elementary school classroom, or the cleanliness of the homes in a community.

Units of Observation

    In selecting behaviors, attributes, or situations to be observed, researchers must decide what constitutes a unit. There are two basic approaches, which actually are end points of a continuum. The molar approach involves observing large units of behavior and treating them as a whole. 

    For example, psychiatric nurse researchers might study patient mood swings. An entire constellation of verbal and nonverbal behaviors might be constructed as signaling aggressive behaviors, for example. Most qualitative observational.

    Studies rely on observations that are reasonably molar. At the other extreme, the molecular approach uses small, highly specific behaviors as observational units. Each movement, action, gesture, or phrase is treated as a separate entity. The choice of approaches depends mostly on the nature of the research problem. 

    The molar approach is more susceptible to observer errors because of greater ambiguity in what is being observed. On the other hand, in reducing observations to concrete, specific elements, investigators may fail to understand how small elements work in concert in a behavior pattern.

The Observer-Observed Relationship

    Researchers can interact with individuals in an observational setting to varying degrees. The issue of the relationship between observers and those observed is important and has stirred much controversy. Two important aspects of this issue concern intervention and concealment. 

    The researchers make decisions in establishing a strategy for these considerations should be based on an understanding of ethical and methodological implications. Observational studies may involve an experimental intervention of the type described in Chapter 8, which deals with experimental design. 

    For example, a nurse researcher may observe patients' postoperative behaviors after an intervention designed to improve the patients' ability to cough and breathe after surgery. Sometimes, however, observational researchers intervene to structure research settings (called directed settings) without introducing an experimental treatment (ie, without manipulating the independent variable). For instance, researchers sometimes stage a situation to provoke behaviors. 

    Certain activities are rare in naturalistic settings, making it unnecessary to wait for them to happen. For example, several studies have examined the behavior of bystanders in crises. Because crises are unpredictable and infrequent, investigators have staged emergencies to observe helping behavior (or lack of it) among onlookers. Such studies, which are high on the intervention dimension, may be practical when there is little ethical issue all along the concealment and intervention dimensions.

Qualitative Observational Methods: Participant Observation

    Qualitative researchers collect unstructured or loosely structured observational data for some studies, often as an important supplement to self-report data. The aim of their research is to understand the behaviors and experiences of people as they actually occur in naturalistic settings. 

    Qualitative researchers seek to observe people and their environments with a minimum of structure and interference. Unstructured observational data are most often gathered in field settings through a process known as participant observation. Participant observers participate in the functioning of the social group under investigation and strive to observe, ask questions, and record information within the contexts, structures, and symbols that are relevant to group members. 

    Bogdan (1972) defines participant observation as “.research characterized by a prolonged period of intense social interaction between the researcher and the subjects, in the milieu of the latter, during which time data are unobtrusively and collected”

    Not all qualitative observational research is participant observation (ie, with observations occurring from within the group under study). Some unstructured observations involve watching and recording unfolding behaviors without the observers interacting with participants in activities. 

    If a key research objective, however, is to learn how group interactions and activities give meaning to human behaviors and experiences, then participant observation is the appropriate method. 

    The members of any group or culture are influenced by assumptions they take for granted, and observers can, through active participation as members, gain access to these assumptions. Participant observation is used by grounded theory researchers, ethnographers, and researchers with ideological perspectives.