Observation

1
Observation

• Scientific observation is made under precisely
  defined conditions, in a systematic and objective
  manner, and with careful record keeping.

2
Need to consider.

• WHO to observe
• WHEN to observe
• HOW to observe
• HOW to record data

3
WHO to observe?

• Results are only generalizable to participants,
  times, settings, and conditions similar to those
  in the study in which the observations were made.
• Must have a representative sample.

4
WHEN to observe

• Time sampling
• Situation sampling

5
Time sampling

• Choose various time intervals for observations
• Random
• Systematic
• Or both
• When is time sampling NOT appropriate?

6
Situation Sampling

• Observing behavior in as many different locations
  and under as many different circumstances as
  possible.

7
HOW to observe

• Observation without intervention
• Observation with intervention

8
Observation without intervention

• Goal is to describe behavior as it normally
  occurs and to examine relationships among
  variables.
• Reactivity
• Demand characteristics

9
Observation with intervention

• Most psychological research uses this type of
  observation.
• Three types well talk about
• Participant observation
• Structured observation
• Field experiments

10
Participant Observation

• Beneficial because it allows access to
  places/situations that is usually not open to
  scientific observation.
• Pretending to be schizophrenic
• Pretending to be a black man
• Can also be problematic why?
• Disguised vs. Undisguised

11
Structured Observation

• More control/intervention than naturalistic
  observation, but less than in an experiment.
• The participant sets up a situation that can be
  observed.

12
Pros and Cons of Structured Observation

• Can observe behavior under conditions
  that are more natural than in the lab.
• - Failing to follow the same procedures EACH
  time can be problematic.
• - Uncontrolled (or even unknown) variables can
  interfere.

13
Field Experiments

• The observer manipulates one or more independent
  variables.
• This is the most extreme form of intervention in
  observational methods.

14
HOW to record data

• Qualitative
• Narrative record
• Field notes
• Quantitative
• Check list yes or no
• Frequency of occurrence
• Ratings on a likert scale
• Electronic recording and tracking

15
Data

• Data reduction
• Coding
• Mean (central tendency)
• Standard deviation (variability)
• Frequency counts

16
What about reliability?

• Interobserver reliability the degree to which
  two independent observers agree.
• How do we get this number?
• Percentage agreement (Number of agreements /
  Number of observations) X 100
• Correlate the two observers ratings

17
What about observer bias?

• Expectancy effects
• Ways to control observer bias
• Keep observes blind as to the hypotheses of the
  experiment
• Recognize that bias may be present