Chapter 7 - Structured Observation PDF

Preview

Summary

Prof: Urvashi Soni-Sinha...

Description

Structured Observation

STRUCTURED OBSERVATION  Structured Observations vs. Surveys.  Direct observation over inferences from respondent reports.  Based on explicit rules for observing, categorizing and recording behaviour.  Limited access to some social spaces.  Smaller sample sizes.  Eliminates some problems found in survey research: o Misunderstanding the meaning of the terms used in the survey. o Problems of memory: what was done, where it was done, etc. o Social desirability effect in answering questions. o Threatening or embarrassing questions not answered sincerely. o Gap between stated and actual behaviour; e.g. Husband’s overestimating the time they devote to housework. OTHER TYPES OF OBSERVATION RESEARCH  Participant Observation: o Qualitative method; o Immersion in a particular social setting for an extended period of time; o To determine meanings that people give to their environment and their behaviour.  Non-participant or “unobtrusive”: o Observer does not take part in the activities of the observed group; o Observed group may or may not know that it is being observed; o Process may be structure or unstructured.  Unstructured observation: o No rules for observing or recording the observations; o Observer attempts to gather as much detail as possible, make thorough notes, and develop a narrative account of the behaviour at a later time; o Most participant observation is unstructured. STRUCTURED OBSERVATION  The rules for observation and for recording the data are contained in the observation schedule.  Example of Bales’ (1951) observation schedule for comparing smallgroup leaders and followers: o The actions are observed over a three-minute period. o The action observed every 15 seconds is noted and given a numerical value for the spreadsheet. o The values are tabulated and a score given. o The score is translated back to a description of the person’s leadership style. THE OBSERVATION SCHEDULE.  A clear statement of the research problem.

   

Specifies who is to be observed and which behaviours are to be recorded. Categories for recording behaviour must be mutually exclusive and exhaustive. o Pilot studies may help develop the categories. Classification scheme must be clear and easy to use. Minimal interpretation of behaviour.

STRATEGIES FOR OBSERVING BEHAVIOUR  Recording of incidents. o Wait until there is a behaviour of interest.  Time period: o Processes are observed and recorded at a designated period of time (e.g., everyday at lunchtime for two weeks).  Time sampling: o Observe the interactions that take place at designated intervals of time (e.g., every 15 minutes for three hours). RELIABILITY  Compared to survey research, structured observation offers: o More reliable information about events. o Greater precision regarding the duration and frequency of events. o Greater accuracy in time-ordering events. o More accurate reconstruction of large-scale social episodes.  Inter-observer consistency is a concern.  Intra-observer consistency is also an issue.  Training is usually implemented to counter these issues. VALIDITY  Measurement validity can be an issue. o Does the structured observation measure what it is supposed to measure?  The structured observation must be administered properly.  The researcher should try to determine whether there are any reactive effects. o People often behave differently because they know they are being observed. o Some research suggests this lessens as the researcher spends more time with the participants. VALIDITY: REACTIVE EFFECTS  Guinea pig or “Hawthorne” (1920) effect: o Participants intentionally adjust their behaviour to conform to what they believe is the purpose of the study; Work Conditions and Productivity.  Role selection: o Participants adopt an expertise role when asked questions or shift to a more passive role than usual.  Researcher presence as a change agent: o The physical environment is changed when a new person enter. o It will impact the social environment somewhat as well.  Trying to be helpful to the researcher. o Joining the experiment whole heartedly.



o Sometimes asking for approval afterwards. o This matters only if it impacts the actual phenomenon being studied. Varied reaction according to who the researcher is. o Expectations of the research subject may be communicated unconsciously.

FIELD EXPERIMENTS  The researcher engages in social interaction in its natural environment and watches to see what happens.  The participants are not aware that they are being studied.  In the extreme, researchers have adopted a social role and used deception to gain entry into social spaces that would not otherwise be allowed. o Limited opportunity for recording data. o At best a limited coding system can be implemented. CRITICISMS OF STRUCTURED OBSERVATION  The used observation schedule may be for a particular setting. o Unstructured observation done in advance of structured observation can help to eliminate this and develop a better schedule.  The intentions and meanings of the acts observed may be missed.  The effect of context on the behaviour may not get sufficient attention.  Structured observation produces many small bits of data, which may be difficult to piece together....