Professional Documents
Culture Documents
Chapter 8
Observation
AT-A-GLANCE
I. Introduction
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
LEARNING OUTCOMES
1. Discuss the role of observational technologies as marketing research tools.
2. Know the difference between direct and contrived observation.
3. Identify ethical issues particular to research using observation.
4. Explain the observation of physical objects and message content.
5. Describe major types of mechanical observation.
6. Summarize techniques for measuring physiological reactions.
SURVEY THIS!
Students are asked to review the questions about opinions and preferences for technological
products and to look at the results from this section. In what way has behavioral observation been
used to collect additional data—if at all? How might this information be useful to companies that
sell small electronic appliances? Are there other places in the survey where behavioral
observation has been combined with this traditional survey approach?
RESEARCH SNAPSHOTS
➢ All that Jazz!
Much can be gleaned about a person by examining his or her preferences. Pandora is
leveraging observational data suggesting a relationship between music preference and
politics to sell advertising to political campaigns. Consider two consumers, one who creates
a Bob Marley station with a touch of Daft Punk on Pandora, and the other who creates a
Yanni station with a touch of Frank Sinatra. How will these consumers vote? Probably no
surprise, but the Marley listener is going to vote democratic, while the Yanni listener leans
right. Knowing a consumer’s preferences provides powerful information because knowing a
few preferences allows one to predict many others.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
OUTLINE
I. INTRODUCTION
A. Scientists rely heavily on observation.
B. Observations play a key role in discovering ideas and developing theories.
C. Inductive learning begins with collections of observations.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
Observation is more than just looking around; it is an activity involving “watching” or paying
careful attention to what one sees. Scientific observation is the systematic process of recording
the behavioral patterns of people, objects, and events as they happen. No questioning or
communicating with people is needed.
2. What are the major advantages of unobtrusive observation over other types of data
collection?
Observation may provide data without distortions due to self-reporting. If the researchers use
unobtrusive observations, people will be acting naturally, rather than artificially, to please an
interviewer. The researcher does not need to rely on the respondent’s memory. Hence, data about
activities, such as how long a person spent shopping, may be more accurate. Further, because
there is no interviewer asking questions, there is no subjectivity associated with a respondent
giving socially desirable answers. Thus, a field worker inspecting a pantry will record the actual
brand purchased, not an answer that a prestige brand is purchased (when it is not). Although there
is no interviewer, it is possible that there will be some subjectivity in recording observation data.
Subjective perception operates in this area as well as in other areas of human activity. Nonverbal
messages may be recorded. Thus, data that cannot be verbally expressed can sometimes be
obtained. Often observation is the only way to get the data (as in the collecting of competitive
pricing information). And in some circumstances certain data (such as weather patterns) may be
obtained more quickly by observation than by any other technique.
There are several limitations to observational data collection techniques. First, while behavior can
be observed, the underlying reasons for that behavior cannot be observed. Analytical models are
good at predicting behavior, but they often fail to explain why that behavior occurs in the first
place. Observing behavior over a long period of time is often costly, which further limits our
ability to understand why a certain behavior occurs.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
Direct observation is a straightforward attempt to observe and record what naturally occurs; the
investigator does not create an artificial situation. Contrived observation is an observation in
which the investigator creates an artificial environment in order to test a hypothesis.
Conversation volume represents a measure of the amount of Internet postings that involve a
specific name or term. That name or term could be a brand name, a destination name, or the name
of an event. Numerous websites provide basic Web traffic statistics. Some sites provide services
for free and others for a fee. Alexa.com allows users to compare traffic among competing
websites and get basic demographic information about the consumers who frequent those sites.
Other web traffic statistics providers include statcounter.com, click.com and Google Analytics,
among others.
6. [Ethics Question] The chapter showed a photograph of a traffic monitoring camera. Do you
think the use of these cameras to issue speeding tickets is ethical? What types of behavior
might cameras like these capture that would help automobile designers produce products that
better match our needs as drivers?
Observation methods introduce a number of ethical issues. Hidden observation raises the issue of
the respondent’s right to privacy. One could argue that cameras are being used to catch speeders,
so they are not respondents at all because research is not taking place. Students’ responses
regarding the ethicalness of using cameras to issue speeding tickets will vary. Three questions
help the researcher feel comfortable about collecting observational data: (1) Is the behavior being
observed commonly performed in public where it is expected that others can observe the
behavior? (In this case, the answer is yes.); (2) Is the behavior performed in a setting in which the
anonymity of the person being observed is assured (meaning there is no way to identify
individuals)? (In this case, the answer is no because drivers can be identified by their license
plates.); and (3) Has the person agreed to be observed? (In this case, one could argue that the
answer is yes because they agreed to abide by the speeding laws when obtaining a license.).
The three questions above also apply to using traffic monitoring cameras to capture behaviors that
would help automobile designers produce products that better match our needs as drivers. Using
this type of observation to guide automobile design could include collision observation, perhaps
leading to safer automobiles. Observation of how consumers actually drive cars could lead to
improvements in safety or features that make driving an automobile more comfortable.
Initiatives are growing quickly in Europe following an E.U. court ruling supporting “a right to be
forgotten.” The ruling requires search providers like Google to respond to individual’s requests to
remove information from search results. As a result, Google is hiring a staff to review
individually the thousands and thousands of requests. However, Google thus far is reticent to
remove information appearing in published sources. Given the complexity of dealing with Google
on these matters, an industry of companies who assist individuals in these efforts has spawned
with France’s Reputation VIP leading the way.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
Students’ responses will vary. As an example, researchers might solve this problem by utilizing
aerial photography. After the photographs are developed and enlarged, a physical counting of
houses in the particular areas can take place. This will give an indication of the relative
population density of the areas and number of commercial enterprises. Another approach could be
to observe traffic counts on various roads and at intersections to determine the attractiveness of a
location. Satellite imagery would be useful to study traffic patterns.
9. What does it mean to use complementary evidence? Describe how observational data taken
from a users’ smartphone browser history could be supplemented with complementary
evidence to answer questions about some aspect of consumer behavior.
Observational study results may amplify the results of other forms of research by providing
complementary evidence concerning individuals’ “true” feelings. Researchers typically conduct
focus group interviews behind two-way mirrors from which marketing executives observe as well
as listen to what is occurring. This additional data source allows for interpretation of nonverbal
behavior such as facial expressions or head nods to supplement information from interviews.
10. Click-through rates for advertisements placed in websites typically are usually very, very low
(1 percent or less). What types of error might exist in using click-through rate data as a
measure of an advertisement’s success?
By using click-through rates, the researcher believes that those who did click on the ad are
representative of those who did not. If only those who clicked on an ad are included in the
measurement of an ad’s success, then a nonresponse error will occur, and those who did not click
on an ad would be called nonrespondents. However, just measuring click-through cannot
differentiate between those who clicked on an ad because they were interested or those who
mistakenly clicked on the ad.
11. Outline a research design using observation for each of the following situations:
a. A bank wishes to collect data on the number of customer services and the frequency of
customer use of these services.
Students will suggest several designs. One of the simplest might be to have tellers record the
number of customers and the times customers visit the teller windows. This might take place after
each customer is serviced. Or, it might be possible to have an observer standing by the doorway
to record information such as number of customers, amount of time spent in the bank, etc. Use of
the teller machine can be measured mechanically.
b. A state government wishes to determine the driving public’s use of seat belts.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
This research objective can be accomplished with direct human observation. A field worker
stationed at a traffic light or another location could easily observe if shoulder straps are being
used. It is also possible to determine seat belt usage from surveillance cameras set up at traffic
intersections.
c. A fast-food franchise wishes to determine how long a customer entering a store has to
wait for his or her order.
This makes an interesting class assignment. You might ask students to go into the field to answer
this. The assignment might read: “Go to your favorite fast-food restaurant. Observe and time how
long customers are required to wait to place their orders and how long it takes to be served their
food.”
If this is an individual student project, there will be a common problem faced by most students.
They must define waiting time and time to receive an order. Questions, such as “Does the timing
start when the respondent opens the door or when the respondent has looked at the menu board
and then moves forward to wait in line?” must be answered. Other difficulties with the
observation process, such as two people who enter the restaurant at the same time from different
doors, will be noted. The professor can discuss these issues.
d. University food services wonder if a smartphone app with which students can order food
ahead of time will cut down on wait times and food waste.
The university could pilot test the app with a small number of students, and use that to observe
how long it takes the student to receive the preordered meal, and whether or not any of the meals
go unclaimed. If the pilot test is successful, the university can expand it to all students, and
continue to use it to make observations on students’ food consumption.
A package delivery service that wishes to observe its delivery workers from the point where they
stop the truck to the point where they deliver the package and return to the truck might use a
motion and time study. The observation of the delivery workers might require that the tasks they
engage in are broken down into component motions (for example, picking up the package,
walking to the building, handing the package to the customer, waiting for a signature, and so on)
so that the efficiency of the work process can be catalogued and evaluated. The observer may use
a stopwatch to time each of these discrete motions. Other verbal or behavioral factors, such as if
the employee says “thank you” to the customer, may also be observed and catalogued. This also
could be mechanically measured by following the electronic tracking of the packages when the
driver takes them off the truck and delivers them to the receiver as most packages are now
tracked electronically.
f. A political consulting agency would like to study the consumers’ reactions as they view
national candidates’ political commercials airing on cable and broadcast television.
A researcher can measure neurological events. Neurological activity can reveal how much
thought takes place and what types of feelings a person is probably experiencing.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
12. What is an artifact to a marketing researcher? How might one use artifacts to study the types
of things that fans at major sporting events might be willing to purchase when attending an
event? Can artifacts also be used to study ergonomics in the office? If so, how?
Artifacts are things that people make and consume within a culture that signal something
meaningful about the behavior taking place at the time of consumption. Researchers can comb
through a stadium or arena after a major sporting event to see the types of products that are left
behind that were not purchased at the event. Observation during the event can also reveal what
fans bring to the event. Combing the parking lots and surrounding area for artifacts left by
tailgaters can also tell researchers something. With respect to studying ergonomics in the office,
researchers can observe items workers bring in to make them more comfortable. For example,
pads on the back of chairs may indicate that they feel uncomfortable and experience back pain
while sitting at their desks. Thus, more ergonomically correct chairs may be in order.
In a scanner-based consumer panel, each household is assigned a bar-coded card, like a frequent-
shopper card, which members present to the clerk at the register. The household’s code number is
coupled with the purchase information recorded by the scanner. In addition, as with other
consumer panels, background information about the household obtained through answers to a
battery of demographic and psychographic survey questions can also be coupled with the
household code number.
14. What are the major types of mechanical observation? What types of observations might
Twitter potentially have access to that would be of interest to basic marketing researchers?
Video cameras, traffic counters, and other machines help observe and record behavior. Perhaps
the best-known marketing research project involving mechanical observation and computerized
data collection is ACNielsen’s television monitoring system for estimating national television
audiences. Nielsen’s PeopleMeter gathers data on what each television in a household is playing
and who is watching it at the time. Consumers visiting the Twitter site can provide useful data
regarding the time it takes to make a decision. Measuring how long users spend looking at a
Twitter post can provide information regarding decision making.
15. [Ethics Question] How can a marketing researcher determine if an observational data
collection involving hidden cameras is ethical?
Cameras are increasingly used in modern society to keep tabs on all sorts of behaviors.
Researchers sometimes ask and get permission to place cameras inconspicuously in consumers’
homes, offices, or even cars. As with any form of observation data collection, three questions can
help address the question of whether involving hidden cameras is ethical: (1) Is the behavior
being observed commonly performed in public where it is expected that others can observe the
behavior? (2) Is the behavior performed in a setting in which the anonymity of the person being
observed is assured, and (3) Has the person agreed to be observed?
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
This specific question raises at least three issues: confidentiality of the research, deception, and
invasion of privacy. These issues have to do with the intent of the research, if in fact this is
research. If the purpose of the research is to identify individual investors who might be emotional
about a possible decline in the market (hence, good sales prospects), then it could be argued that
this is not a scientific research study with the intent of aggregating the data and making general
conclusions. Assuming that this is a legitimate study, then it should be asked if “not telling” about
something is a deception. Is this different from telling a “little white lie”? If this information is
obtained without the subject’s consent or without a debriefing after the study, is it an invasion of
privacy? Students should be asked to take a stand.
This issue relates to informing subjects about the research. Will the subjects react differently if
they know they are being observed? If so, the researcher may justify the deception and invasion
of privacy because the research will suffer. Most researchers would suggest a debriefing.
However, others would argue that being informed about the two-way mirror will have very little
impact on behavior and the subjects should be informed.
c. A marketing researcher arranges to purchase the trash from the headquarters of a major
competitor. The purpose is to sift through discarded documents to determine the
company’s strategic plans.
Students’ responses will vary. However, three questions help the researcher feel comfortable
about collecting observational data: (1) Is the behavior being observed commonly performed in
public where it is expected that others can observe the behavior? In this case, while this behavior
can be performed in public, most individuals and companies do not expect their trash to be
searched nor would they consider it open to the public; (2) Is the behavior performed in a setting
in which the anonymity of the person being observed is assured (meaning there is no way to
identify individuals)? In this case, the answer is no because the trash being observed is a specific
competitor’s; and (3) Has the person agreed to be observed? In this case, it is highly unlikely that
the competitor would agree to this observation.
This is a serious invasion of consumers’ privacy rights. The only way to make this an ethical
practice is to obtain informed consent from consumers ahead of time.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
18. Look back to the chapter on qualitative research and find the definition for ethnography. Why
is observation such a big part of this important qualitative research approach?
Ethnography represents ways of studying cultures through methods that involve becoming highly
active within that culture. Participant-observation typifies an ethnographic research approach.
Participant-observation means the researcher becomes immersed within the culture that he or she
is studying and draws data from his or her observations.
19. What is an MRI device and how can it be used in research studying advertising effectiveness?
This tool allows for actual direct observation of what is going on in the mind of a respondent by
assessing and identifying where electromagnetic activity is taking place within the brain.
Marketing researchers employ the technology by having paid respondents view advertisements in
an MRI machine. The results help identify the patterns of cognitive and emotional reactions the
ads may generate.
RESEARCH ACTIVITIES
20. William Rathje, a researcher at the University of Arizona, Department of Anthropology, has
become well known for the “Garbage Project.” The project involves observational research.
Use http://www.ask.com to find information about the garbage project at the University of
Arizona. What is the name of the book that describes some of the key findings of the Garbage
Project? How do you think it involves observational research?
William Rathje started the project in 1971 and established it at the University of Arizona in 1973.
He has since retired. Students will probably come across a book written by Dr. Rathje entitled,
Rubbish! The Archeology of Garbage. The project entails sifting through garbage to understand
Americans’ consuming habits.
21. The Internet is filled with webcams. For example, Pebble Beach Golf Club has several
webcams (http://www.pebblebeach.com). How could a researcher use webcams like these to
collect behavioral data? In your short time viewing these webcams, are there any research
questions that you think might be addressed based on behaviors that can be observed in these
views? If so, what might one or two be?
The webcams at Pebble Beach are interesting, providing live views (still views, not motion
pictures) of various holes. Webcams can be used to observe how people behave when consuming
a service, such as playing a golf course.
22. Review the evidence presented regarding hand washing behaviors (or the lack thereof).
Report results based on the way the data were obtained. What accounts for the differences in
survey, observational and mechanical-based studies? Do a little research of your own on
hand-washing and compare to the results reported here. Can you think of other ethical ways
to “observe” this behavior that are not stated in this chapter?
Survey results are subject to respondent bias, and many people may inflate the number of times
per day that they wash their hands in order to enhance how they are perceived. Human
observation is a stronger approach, but still presents some methodological and technical
difficulties.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
Mechanical observation is probably the best approach, as long as participants’ privacy is not
violated (e.g., using video cameras). Student responses to their own research into hand washing
will vary, but this could make for a very good class project.
Objective: To illustrate methods for assessing physiological measures and to encourage students
to think of other observational methods to use when assessing the usefulness of a website or other
marketing communication stimulus.
Summary: Mazda Motor Europe wanted details about how consumers were using its website and
whether finding information was easy. A research firm, Syzygy, used an eye-tracking device that
uses infrared light rays to record what areas of a computer screen a user is looking at. A small
sample of subjects was observed with the equipment, which measured each subject’s gaze,
recorded the location on the screen, and graphed the data. The results included three-dimensional
contour maps highlighting the “peak” areas where most of the computer users’ attention was
directed.
Questions:
1. What could Mazda learn from eye-tracking software that would be difficult to learn from
other observational methods?
This type of equipment can track actual eye movement rather than merely watching respondents
as they search a website for information. This gives a more precise indication of what they were
actually looking at. Other observations of respondents’ use of websites can track where they click
on a page, but this does not give information on where they looked before they clicked on
something.
As with all types of devices to measure physiological responses, it is assumed that physiological
reactions are associated with persuasiveness or predict some cognitive response. This assumption
has not yet been clearly demonstrated, however. Another major problem with physiological
research is the calibration, or sensitivity, of measuring devices. The eye-tracking software can
measure where a subject’s eye looked, but it cannot indicate why he or she looked there or if the
information was effective or not.
3. Along with the eye-tracking research, what other research methods could help Mazda assess
the usability of its website? Summarize your advice for how Mazda could use complementary
methods to obtain a complete understanding of its website usability.
Mechanical measures, such as how long a respondent spent looking at certain information and the
subsequent information clicked on, can be used. Additionally, the results of the eye-tracking
measures can be used to ask respondents why they looked at certain information on the website.
For example, they could have looked at it because it was interesting, it provided useful
information, or it was confusing and they were trying to make sense of it. Observation alone will
not explain behavior.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
Chapter 8: Observation
Objective: The purpose of this case is to illustrate some of the difficulties involved in conducting
an observation study where problems arise when field workers must be trained to carefully record
the available information.
Summary: Pedestrian field workers covered each block in the CBD area to obtain a physical
inventory of the existing commercial base. Information, such as the company name, address,
primary business activity, estimated frontage space, and other relevant information about each
building site or office was obtained. Standard Industrial Classification (SIC) codes for retailers
were recorded by the field workers. SIC codes for all other establishments were recorded by
research assistants. All the data were identified by census block.
Questions
Essentially, this is a sound research design. Using an observation study was the correct choice.
Virtually no other research method could obtain the desired information.
The observation form is adequate. However, there were some problems. There should have been
spacing below the activities section to further specify what type of business activity was taking
place. Without pretesting, though, it is difficult to say which changes should be made.
The information concerning vacant offices for sale or for rent could be difficult to obtain. The
estimated footage and estimated number of stories were probably recorded subjectively, which
could introduce error.
Because several field workers were involved in this project, it is important that they be trained
adequately because their need to interpret some information (i.e., type of business activity). A
second problem will likely be identifying firms and/or the nature of their business from the
exteriors of their offices. Many firms cannot be readily identified from outside or from a hallway
entrance. However, there could be some difficulty recording the activities of a firm that had only
a sign outside. How does one classify “Paragon Corporation”?
The main purpose of the study was to obtain an inventory, so this is primarily a descriptive study.
Therefore, a listing of establishments should be the primary form of data analysis. The data could
be coded by business activity, the company data categorized as vacant, retail trade, wholesale,
manufacturing, petroleum and related industries, service, etc. Census block, based on the address
of the firm, could also be coded, which can allow for a geographical distribution of firms to be
portrayed. Frequency counts can provide useful information for planners.
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.