Page 270 - [Uma_Sekaran]_Research_methods_for_business__a_sk(BookZZ.org)

254 DATA COLLECTION METHODS

The above three advantages are perhaps unique to observational studies.

Drawbacks of Observational Studies
The following drawbacks of observational studies have also to be noted.
1. It is necessary for the observer to be physically present (unless a camera or
another mechanical system can capture the events of interest), often for pro-
longed periods of time.
2. This method of collecting data is not only slow, but also tedious and expensive.
3. Because of the long periods for which subjects are observed, observer fatigue
could easily set in, which might bias the recorded data.
4. Though moods, feelings, and attitudes can be guessed by observing facial
expressions and other nonverbal behaviors, the cognitive thought processes
of individuals cannot be captured.
5. Observers have to be trained in what and how to observe, and ways to avoid
observer bias.

Biases in Observational Studies
Data observed from the researcher’s point of view are likely to be prone to
observer biases. There could be recording errors, memory lapses, and errors in
interpreting activities, behaviors, events, and nonverbal cues. Moreover, where
several observers are involved, interobserver reliability has to be established
before the data can be accepted. Observation of the happenings day in and day
out, over extended periods of time, could afflict the observers with ennui and
introduce biases in the recording of the observations. To minimize observer bias,
observers are usually given training on how to observe and what to record. Good
observational studies would also establish interobserver reliability. This could also
be established during the training of the observers when videotaped stimuli could
be used to determine interobserver reliability. A simple formula can be used for
the purpose—dividing the number of agreements among the trainees by the num-
ber of agreements and disagreements—thus establishing the reliability coefficient.
Respondent bias could also be a threat to the validity of the results of obser-
vational studies, because those who are observed may behave differently during
the period of the study, especially if the observations are confined to a short
period of time. However, in studies of longer duration, the employees become
more relaxed as the study progresses and tend to behave normally. For these rea-
sons, researchers doing observational studies discount the data recorded in the
first few days, if they seem to be quite different from what is observed later.

Summary of Observational Studies
Observational studies have a formulated research purpose and are systematically
planned. Such studies can be structured or unstructured, with the investigator
being a participant or nonparticipant in the study setting. All phenomena of inter-
est are systematically recorded and quality control can be exercised by eliminat-