1. watching, listening, and recording: the process
of observation involves these three interrelated activities. Depending
on the type of research, the approach may either structured or unstructured.
In quantitative designs, observation is usually time limited and structured.
Criteria for watching, listening, and recording are selected before data
collection begins (a priori) and the system for recording is structured.
Checklists and rating scales might be used. Phenomena other than
those specified ahead of time are ignored. In qualitative research,
the watching, listening, and recording behaviors are more inductive in
nature. The researcher broadly defines the field of observation and
then moves to a more narrowed or focused approach as the data collecting
and analysis process unfolds. Your chapter gives the example of intelligence
to demonstrate the differences seen. In quantitative research, intelligence
would likely be measured by use of a structured paper and pencil test.
Measurements are taken and compared to predefined dimensions. In
qualitative research, the researcher might watch and listen to a person
in his natural environment with the hopes of revealing the meaning of intelligence
in his/her particular culture. If you wanted to compare one person
to another, the quantitative data collection would be appropriate.
If you were interested in attempting to develop new understanding of the
construct of intelligence, then the qualitative data collection method
described would be appropriate.
2. asking: asking questions in a research setting
is purposeful and systematic. Questions can be highly structured
with fixed responses or open-ended and unstructured. An example of
a highly structured, close-ended question would be, "This past week, how
would you rate your study skills: excellent, good, adequate, fair,
or poor?" This type of question would be useful in quantitative research.
In qualitative research, where the purpose in asking questions is more
for discovery and exploration, an open-ended, less structured question
would be more useful. An example of this type of question might be,
"Let's talk about your studying this week. How do you feel about
your studying?" This type of question does not give fixed response
choices and cannot be answered with a "yes" or "no". When you ask
questions, it can be accomplished through interviews or questionnaires.
a. Interviews: these are most often done with one
individual at a time, can be structured or unstructured, and may be in-person
or telephone. When groups of people are interviewed (five or more),
a "focus group" type of methodology is used. In this methodology, the interaction
between the interviewees is important. An audiotape is made, the
entire interview is transcribed and then analyzed. A structured interview
is one where the interviewer imposes tight control over the sequence of
questions and the content of the questions themselves. Interviewees
often are required to select one response from a fixed group of responses.
These types of responses include dichotomous (yes/no), multiple choice,
rank ordered, and graded responses (Likert scale: 0-5). Unstructured
interviews are used most often in exploratory studies and in qualitative
research. The interviewer presents the broad topic area and asks
a few probing questions. Follow-up questions emerge as a result of
the answers to the initial questions. Probes are neutral statements
that do not bias the interviewee to respond in a particular way. "Tell
me about that" would be an example. Sometimes, unstructured interviews
are used by quantitative researchers in an initial pilot study to identify
domains and responses that will be used in a later, more structured interview.
b. Questionnaires: these are written instruments that can
be done face-to-face or through the mail. The same things apply to
questionnaires as to questions used in interviews. They may be structured
or unstructured and use open or close-ended questions. There are
advantages and disadvantages to each. See your book's Tables 15-3, 15-4,
15-5, & 15-6 for the pros and cons of structured and structured
and open and close-ended questions. One particular kind is not better
than the others overall. The best type of question is determined
by the research design.
3. obtaining and examining materials: there are several
reasons why you might use already existing data. It might be accessible
when direct observation or interviews are not possible. The Hawthorne
effect is eliminated because there is no research person there to give
the subject "attention" (remember that the Hawthorne effect describes the
process by which a subject experiences just by participating in the project
itself) . The researcher may also get a better perspective by viewing
phenomena in the past and over time. This might not be possible with
a single data collection that happens at one point in time. Existing
data may also be helpful when sensitive materials are being analyzed.
a. unobtrusive methodology: this describes a process where
there is minimal investigator effect in the research setting. It
involves the observation and examination of data with little or no interaction
with the participants. If you are interested in alcohol consumption
in a neighborhood, you look through recycle bins and trash cans to count
empty containers. This methodology may be used in combination with
other means of data collecting. Say that you were interviewing people
living in a neighborhood where inhabitants described alcohol consumption
patterns. the data you acquired through the trash and recycle bin
search could help confirm the subjective reporting of the participants.
The investigator may also look at various forms of written data in this
type of methodology as well. These include diaries, journals, medical
records, clinical notes, meeting minutes, letters, historical documents,
newspapers, magazines, etc.
b. secondary data analysis: here the researcher reanalyzes
existing data sets. The purpose is to ask different questions of
the data than when it was analyzed in the original study. Let's say
that you collected data on alcohol consumption in teenagers. The
original study analyzed data from this perspective. Incidental to
the study at that time was information about teenage sexual activity.
Later, you or another researcher becomes interested in the connection between
alcohol consumption and sexual activity or is simply interested in the
sexual activity data. the data can then be reanalyzed with this new perspective
in mind. This is often true in longitudinal studies where large amounts
of data exist from multiple subjects over time. There is so much
data that no one researcher study or investigator can examine and analyze
it all at one time. It would be beyond the scope of any one study.
c. artifact review: this is a technique used to
ascertain the meanings of objects in research contexts. Say
that you were interested in the preparation process that clients underwent
prior to a planned hospitalization. you might ask them to keep a
diary or journal of the days surrounding the hospitalization. You
might also analyze the contents of the suitcase they bring to the hospital.
This analysis of the contents of the suitcase would be a type of artifact
review.
Summary: remember that there are three principles that guide the
action process of obtaining information: