1
CHAPTER 4
OBSERVATION
CHAPTER OUTLINE AND OBJECTIVES
I. Overview
II. Sampling Behavior
When a complete record of behavior cannot be obtained, researchers seek to obtain a representative
sample of behavior.
The extent to which observations may be generalized (external validity) depends on how behavior is
sampled.
A. Time Sampling
Time sampling refers to researchers choosing time intervals for making observations either
systematically or randomly.
When researchers are interested in events that happen infrequently, they rely on event sampling
to sample behavior.
B. Situation Sampling
Situation sampling involves studying behavior in different locations and under different
circumstances and conditions.
Situation sampling enhances the external validity of findings.
Within situations, subject sampling may be used to observe only some individuals in the setting.
III. Observational Methods
Observational methods can be classified as direct observation or indirect observation.
IV. Direct Observational Methods
Direct observational methods can be classified as “observation without intervention” or “observation
with intervention.”
A. Observation without Intervention
The goals of naturalistic observation are to describe behavior as it normally occurs and to
examine relationships among variables.
Naturalistic observation helps to establish the external validity of laboratory findings.
When ethical and moral considerations prevent experimental control, naturalistic observation is
2
an important research strategy.
Online behavior can be observed without intervention.
B. Observation with Intervention
Most psychological research uses observation with intervention.
The three methods of observation with intervention are participant observation, structured
observation, and the field experiment.
Whether “undisguised” or “disguised,” participant observation allows researchers to observe
behaviors and situations that are not usually open to scientific observation.
If individuals change their behavior when they know they are being observed (“reactivity”), their
behavior may no longer be representative of their normal behavior.
Often used by clinical and developmental psychologists, structured observations are set up to
record behaviors that may be difficult to observe using naturalistic observation.
In a field experiment, researchers manipulate one or more independent variables in a natural
setting to determine the effect on behavior.
V. Indirect (Unobtrusive) Observational Methods
An important advantage of indirect observational methods is that they are nonreactive.
Indirect, or unobtrusive, observations can be obtained by examining physical traces and archival
records.
A. Physical Traces
Two categories of physical traces are “use traces” and “products.”
Use traces reflect the evidence of use (or nonuse) of items and can be measured in terms of
natural or controlled use.
By examining products people own or the products produced by a culture, researchers test
hypotheses about attitudes, preferences, and behavior.
The validity of physical trace measures is examined by considering possible sources of bias and
by seeking converging evidence.
B. Archival Records
Archival records are the public and private documents describing the activities of individuals,
groups, institutions, and governments, and comprise running records and records of specific,
episodic events.
Archival data are used to test hypotheses as part of the multimethod approach, to establish the
3
external validity of laboratory findings, and to assess the effects of natural treatments.
Potential problems associated with archival records include selective deposit, selective survival,
and the possibility of spurious relationships.
VI. Recording Behavior
The goals of observational research determine whether researchers seek a comprehensive
description of behavior or a description of only selected behaviors.
How the results of a study are ultimately summarized, analyzed, and reported depends on how
behavioral observations are initially recorded.
A. Comprehensive Records of Behavior
Narrative records in the form of written descriptions of behavior, and audio and video recordings,
are comprehensive records.
Researchers classify and organize data from narrative records to test their hypotheses about
behavior.
Narrative records should be made during or soon after behavior is observed, and observers must
be carefully trained to record behavior according to established criteria.
B. Selected Records of Behavior
When researchers seek to describe specific behaviors or events, they often obtain quantitative
measures of behavior, such as frequency or duration of its occurrence.
Quantitative measures of behavior use one of four levels of measurement scales: nominal,
ordinal, interval, and ratio.
Rating scales, often used to measure psychological dimensions, are frequently treated as if they
are interval scales even though they usually represent ordinal measurement.
Electronic recording devices may be used in natural settings to record behavior, and pagers
sometimes are used to signal participants to report their behavior (e.g., on a questionnaire).
VII. Analysis of Observational Data
Researchers choose qualitative data analysis or quantitative data analysis to summarize
observational data.
A. Qualitative Data Analysis
Data reduction is an important step in the analysis of narrative and archival records.
Researchers code behaviors according to specified criteria, for example, by categorizing
behaviors.
4
Content analysis is used to examine narrative and archival records and includes three steps:
identifying a relevant source, sampling sections from the source, and coding units of analysis.
B. Quantitative Data Analysis
Data are summarized using descriptive measures such as frequency counts, means, and
standard deviations.
Interobserver reliability refers to the extent to which independent observers agree in their
observations.
Interobserver reliability is increased by providing clear definitions about behaviors and events to
be recorded, by training observers, and by providing feedback about the accuracy of
observations.
High interobserver reliability increases researchers’ confidence that observations about behavior
are accurate (valid).
Interobserver reliability is assessed by calculating percentage agreement or correlations,
depending on how the behaviors were measured and recorded.
VIII. Thinking Critically about Observational Research
A. Influence of the Observer
The problem of reactivity occurs when the presence of an observer influences the behavior being
observed.
Research participants may respond to demand characteristics in the research situation to guide
their behavior.
Methods to control reactivity include concealing the observer’s presence, adaptation (habituation,
desensitization), and indirect observation (physical traces, archival records).
Researchers must consider ethical issues when attempting to control reactivity.
B. Observer Bias
Observer bias occurs when researchers’ biases determine which behaviors they choose to
observe, and when observers’ expectations about behavior lead to systematic errors in identifying
and recording behavior.
Expectancy effects can occur when observers are aware of hypotheses for the outcome of the
study or the outcome of previous studies.
The first step in controlling observer bias is to recognize that it may be present.
Observer bias may be reduced by keeping observers unaware (“blind”) of the goals and
hypotheses of the study.
5
IX. Summary
REVIEW QUESTIONS AND ANSWERS
These review questions appear in the textbook (without answers) at the end of Chapter 4, and can be
used for a homework assignment or exam preparation. Answers to these questions appear in italic.
1. Describe the types of sampling researchers use in observational studies and what the proper use of
sampling is intended to accomplish.
2. Explain the difference between direct and indirect observational methods and how the degree of
intervention can be used to distinguish direct observational methods.
3. Describe a research situation in which naturalistic observation can be useful when ethical
considerations prevent researchers from intervening to study behavior.
4. Explain why reactivity is a problem in observational studies.
5. What ethical issues are raised when a participant observation study is conducted online?
10
E. Identify one aspect of the procedures in this study that would likely limit the external validity of the
findings of this study.
3. An American psychology graduate student, raised in Germany until age 16, wishes to explore how
Germans and Americans use the social networking site, Facebook. She speaks fluent German and
has many social contacts both in Germany and the United States. Specifically, the student asks:
What are the differences in the way that Germans and Americans portray themselves on Facebook?
She contacts network friends in both countries and explains her project. She also requests that her
friends describe the project to other Facebook users in their country and help recruit people for her
study. After a few weeks she finds that she has more than 500 Facebook users willing to participate,
with the numbers from each country approximately the same. What specific advice would you give
her regarding each of the following questions related to her content analysis?
A. How would a qualitative analysis of the data differ from a quantitative one?
B. What might be coded as part of a quantitative analysis?
C. How might reliability of coding be assessed?
D. What characteristics of her sample might affect the external validity of her results?
12
13
• Rosenhan’s research illustrated the depersonalization and powerlessness experienced by
many patients in mental health settings. His research influenced the mental health field to
examine its practices and to improve conditions for patients.
2. Reading Research Critically
These two brief research summaries can be used for class discussion or as possible test questions.
The research summary and questions could be distributed to students in the class session prior to the
scheduled discussion to allow time for students to prepare answers, perhaps as a homework
assignment. The questions and answers are printed below. The summaries and questions are printed
on subsequent pages to facilitate photocopying.
A. Inattentional Blindness: Chapter 4 describes the findings for an observational study in which
Hyman et al. (2009) examined whether people talking on cell phones are more likely to miss key
events in their environment. This research summary provides more details of that study.
Answers to Reading Research Critically: Inattentional Blindness
(1) Hyman and his colleagues (2009) observed that people speaking on cell phones were less
likely to notice the unicycling clown than others who were not using a cell phone. One
interpretation of this finding is that talking on a cell phone causes inattentional blindness.
Comment critically on accepting this conclusion based on these data.
(2) The summary of the Hyman et al. (2009) study indicates that they classified walkers into four
categories: walking with cell phone, walking alone (no electronics), walking alone with music
player, and walking with another person. What alternative hypotheses for inattentional
blindness can be ruled out when the performance of the last three groups (i.e., other than cell
phone use) is considered?
(3) Comment critically on two aspects of the possible conclusions of this study: (a) Can someone
argue, based on these findings, that use of different pronouns (e.g., I, we, us vs. you, your)
causes different interaction behaviors and marital outcomes? (b) What factors may limit the
generality of these research findings?
16
Reading Research Critically
Read the following description of a research study to answer the questions that follow. This description is
based on the second study in an article by: Hyman, I. E., Boss, S. M., Wise, B. M., McKenzie, K. E., &
Caggiano, J. M. (2009). Did you see the unicycling clown? Inattentional blindness while walking and
talking on a cell phone. Applied Cognitive Psychology, published online in Wiley InterScience
(www.interscience.wiley.com). doi: 10.1002/acp.1638
Inattentional Blindness
Hyman and his colleagues (2009) sought evidence in a natural setting to confirm laboratory findings that
people talking on cell phones demonstrate performance deficits; for example, people who talk on cell
phones while tested in driving simulators show poorer driving performance. A review of psychological
literature related to “inattentional blindness” revealed that when people’s attention is divided, they are less
likely to notice changes or new stimuli in their environment. Some investigators argue, however, that cell
phone use has become so automatic that most cell phone users are easily able to divide their attention
between their call and other automatic activities, such as driving. Hyman et al. decided to examine
another automatic activity: walking. They reasoned that if cell phone use and walking are two highly
automatic activities, people should have little trouble dividing their attention between the two tasks and,
therefore, should have little difficulty noticing an unusual event as they walked: the presence of a brightly
dressed, unicycling clown.
Hyman et al. chose a central square of a university to make their observations. Trained observers were
positioned in pairs at the two corners of a frequently used path that runs diagonally through the square.
Observations were made during a single 1-hour session. The observers interviewed all individuals who
crossed the square during the time the clown unicycled. They first asked the walkers whether they had
seen anything unusual when crossing the square and, if yes, what specifically they saw. If the walkers did
not mention seeing the clown, they were asked directly if they had seen the clown. Observers classified
individuals into one of four walking groups: cell phone user (n = 24), walking alone with no electronic
device (n = 78), walking alone with music player (n = 28), or walking with another person (and no
electronics; n = 21).
Results indicated that only 25% of the cell phone users reported seeing the unicycling clown. The other
groups did much better at noticing the clown: walking alone (51.3%), walking with music (60.7%), and
walking in a pair (71.4%). Hyman et al. (2009) interpreted these findings as evidence for inattentional
blindness among cell phone users. They highlighted the natural setting of their findings and that even
when attention was divided with a very simple task, walking, cell phone use was associated with
decreased likelihood of noticing an unusual event. They cautioned that people, when driving, may be
unaware of missing key events when talking on a cell phone until it is too late.
1. Hyman and his colleagues (2009) observed that people speaking on cell phones were less likely to
notice the unicycling clown than others who were not using a cell phone. One interpretation of this
finding is that talking on a cell phone causes people to be less attentive. Comment critically on
accepting this conclusion based on these data.
2. The summary of the Hyman et al. (2009) study indicates that they coded their observations into four
categories: walking with cell phone, walking alone (no electronics), walking alone with music player,
and walking with another person. What alternative hypotheses for inattentional blindness can be ruled
out when the performance of the last three groups (i.e., other than cell phone use) is considered?
3. One problem with the external validity of the Hyman et al. (2009) study is that their observations were
limited to one college campus, at one time, with one particular type of unusual event. Can you think of
reasons why these findings may not generalize to other situations involving cell phone use?
Reading Research Critically
Read the following description of a research study to answer the questions that follow. This description is
based on an article by: Simmons, R. A., Gordon, P. C., & Chambless, D. L. (2005). Pronouns in marital
interaction: What do “you” and “I” say about marital health? Psychological Science, 16, 932-936.
Pronouns in Marital Interaction
Researchers in social psychology have increasingly examined the words people use when they talk to
one another to measure the quality of their relationships. In addition, psychologists interested in marital
relationships have paid attention to relationship partners’ use of personal pronouns, such as we, us, our,
you, your, I, me, and my, to understand factors that affect relationship satisfaction and likelihood of
divorce. Simmons et al. (2005) noted that previous research lumped together all first-person pronouns
(e.g., I, me), but they suggested that these pronouns likely serve different roles in communication.
Specifically, they hypothesized that frequent use of me may reflect decreased satisfaction and a passive
role of victimization in a relationship. In contrast, greater use of I may reflect positive self-disclosure and
perspective taking. Consistent with previous research, Simmons et al. also hypothesized that more
frequent use of pronouns indicating shared identity in a relationship (e.g., we, us, our) would predict
positive marital health, and more frequent use of second-person pronouns (e.g., you, your) would be
associated with negative interactions.
Another goal of Simmons et al.’s (2005) research was to observe partners’ behavior, including language
use, in an actual interaction (rather than, for example, written narratives prepared by couples). They
asked 59 couples in an outpatient therapy setting to complete a problem-solving interaction during which
they discussed for 10 minutes one of the top ten issues affecting their relationship. Couples were asked
to work toward a mutually satisfactory solution to their problem. The researcher left the room during the
discussion and the interactions were videotaped. Teams of coders classified each interaction utterance
into one of ten categories based on a well-established coding system for partner interactions, and
counted the occurrence in each category. In addition, nonverbal behaviors were classified into positive,
negative, and neutral categories and counted. Written transcripts from the videotapes were analyzed
using a computer program that counted pronoun use. Individuals also completed a questionnaire
assessing marital satisfaction.
Results indicated that relationship partners who used more second-person pronouns (e.g., you, your)
exhibited more negative behaviors during the interaction (e.g., criticism, justification), whereas those who
used more first-person plural pronouns (we, us) produced more positive problem solutions. Marital
satisfaction was positively associated with the use of first-person singular pronouns. The use of me was
not associated with any negative relationship variables in this study, contrary to their prediction. Simmons
et al. (2005) concluded that overall, their findings support hypotheses that pronouns used by relationship
partners are important indicators of marital health and interaction.
1. Describe the type of observational method used by Simmons et al. (2005). Based on this description,
identify the measurement scale used for these variables: interaction utterances, nonverbal behaviors
during the interaction, and counts for each type of pronoun. What descriptive statistics are
appropriate for these variables?
2. Explain how reactivity might have been an issue in this study. Is there any aspect of the procedure
that may have reduced reactivity?
3. Comment critically on two aspects of the possible conclusions of this study: (a) Can someone argue,
based on these findings, that use of different pronouns (e.g., I, we, us vs. you, your) causes different
interaction behaviors and marital outcomes? (b) What factors may limit the generality of these
research findings?
3. More Challenge Questions
These additional challenge questions can be used for class discussion or as test questions.
A. Use this description of a research study to answer the questions that follow:
Students in a research methods class did a study to examine when students would be more likely
to look for oncoming traffic when crossing a street. The researchers predicted that students would
be less likely to look for oncoming traffic when crossing at an intersection with a designated
crosswalk (i.e., with lines marking the crossing) than when crossing at an intersection without a
designated crosswalk (i.e., no lines). A team of six student researchers made the observations for
this study. The observers made their observations on Monday, Wednesday, and Friday for one
week, and on Tuesday and Thursday during a second week. On each day, they observed the
intersections for one hour mid-morning and for one hour mid-afternoon. Both intersections were
observed during each observation session. The observers were in a location such that they could
see the intersections, but could not be seen by the pedestrians who were being observed. The
observers included only student pedestrians who were walking alone, and they counted the
student as having “looked for traffic” if he or she looked in one or both directions before entering
the intersection. The observers made initial observations together to establish agreement
regarding what constituted “looking for traffic.” These observations were not included in the final
data. Over 200 students were observed crossing the intersection at the designated crosswalk and
over 100 students were observed crossing at the intersection without the designated crosswalk.
Their results supported the research team’s prediction: fewer students looked for traffic when
crossing at the designated crosswalk than when crossing the intersection without the designated
crosswalk.
(1) What type of observational method was used in this study?
(2) The students selected levels of an independent variable in their study. What is the
operational definition of this variable?
(3) What measurement scale is used for their dependent variable? Based on this, what
descriptive statistic should be used in this study to compare the values of the dependent
variable in the two conditions? Explain your answer.