Nature of observation research

ADS

Instead of asking people questions, as a survey does, observation research depends on watching what people do. Specifically, observation research can be defined as the systematic process of recording patterns of occurrences or behaviors without normally questioning or communicating with the people involved. (Mystery shopping is an exception.) A marketing researcher using the observation technique witnesses and records events as they occur or compiles evidence from records of past events. The observation may involve watching people or watching phenomena, and it may be conducted by human observers or by machines. Exhibit 8.1 gives examples of some common observation situations

Conditions for Using Observation
Three conditions must be met before most types of observation can be successfully used as a data-collection tool for marketing research:

1.  The needed information must be either observable or inferable from behavior that is observable. For example, if a researcher wants to know why an individual purchased a new Toyota Sequoia rather than a Ford Expedition, observation research will not provide the answer.
2. The behavior of interest must be repetitive, frequent, or in some manner predictable. Otherwise, the cost of most forms of observation makes the approach prohibitively expensive.
3. The behavior of interest for many types of observation must be of relatively short duration. Observation of the entire decision-making process for purchasing a new home, which might take several weeks or months, is not feasible.

Approaches to Observation Research
Researchers can choose from a variety of observation approaches. They are faced with the task of choosing the most effective approach for a particular research problem, from the standpoint of cost and data quality. The dimensions along which observation approaches vary are (1) natural versus contrived situations, (2) open versus disguised observation, (3) human versus machine observers, and (4) direct versus indirect observation.

Natural versus Contrived Situations Counting how many people enter a Macy’s store during certain hours is a good example of a completely natural situation. The observer plays no role in the behavior of interest. Those being observed should have no idea that theyare under observation. At the other extreme is recruiting people to do their shopping in asimulated supermarket (rows of stocked shelves set up in a field service’s mall facility) so that their behavior can be carefully observed. In this case, the recruited people must be given at least some idea that they are participating in a study. The participants might be given grocery carts and told to browse the shelves and pick out items that they might normallyuse. The researchers might use alternative point-of-purchase displays for several products under study. To test the effectiveness of the various displays, the observers would note how long the shopper paused in front of the test displays and how often the product was actually selected. Today, many firms, such as Frito-Lay and Procter & Gamble, use online simulated environments.

Acontrived environment enables the researcher to better control extraneous variables that might have an impact on a person’s behavior or the interpretation of that behavior. Use of such an environment also tends to speed up the data-gathering process. The researcher does not have to wait for natural events to occur but instead instructs the participants toperform certain actions. Because more observations can be collected in the same length oftime, the result will be either a larger sample or faster collection of the targeted amount of data. The latter should lower the costs of the project. The primary disadvantage of a contrived setting is that it is artificial, and thus the observed behavior may be different from what would occur in a real-world situation. The more natural the setting, the more likely it is that the behavior will be normal for the individualbeing observed.

Open versus Disguised Observation Does the person being observed know that he or she is being observed? It is well known that the presence of an observer may have an influence on the phenomena being observed. Two general mechanisms work to bias the data. First, if people know they are being observed (as in open observation), they may behave differently. Second, the appearance and behavior of the observer offer potential for bias similar to that associated with the presence of an interviewer in survey research. Disguised observation is the process of monitoring people who do not know they are being watched. A common form of disguised observation is observing behavior from behind a one-way mirror. For example, a product manager may observe respondent reactions to alternative package designs from behind a one-way mirror during a focus group discussion.

Human versus Machine Observers In some situations, it is possible and even desirable to replace human observers with machines when machines can do the job less expensively, more accurately, or more readily. Traffic-counting devices are probably more accurate, definitely cheaper, and certainly more willing than human observers. It would not be feasible, for example, for Nielsen Holdings to have human observers in people’s homes to record television viewing habits. Movie cameras, audiovisual equipment, and software record behavior much more objectively and in greater detail than human observers ever could. The electronic scanners found in most retail stores provide more accurate and timely data on product movement than human observers ever could.

Direct versus Indirect Observation Some of the observation carried out for marketing research is direct observation of current behavior or artifacts. For example, the contents of 100 women’s purses in Portland, Oregon and Plano, Texas were examined as part of an observational study. Nearly every participant (99 percent) had something of a financial nature such as a credit or debit card, checkbook or wallet. Women who were married, college educated, and with high incomes carried the most cards. Ninety-eight percent of the women carried reward and membership cards. Office supplies, like paper and pens, were found in 93 percent of the purses. Other items discovered, in rank order, were: beauty and hair care, identification, receipts, cell phones and accessories, insurance cards, food and candy, healthcare   products, coupons, glasses, photos, trash, nail care products, feminine care products, tissues, hand sanitizers, food/drink supplies such as napkins and toothpicks, oral care products, religious items, weapons, keepsakes, and cameras. Only 8 percent carried a camera. In some cases, past behavior must be observed. To do this, the researcher must turn to some record of the behavior. Archaeologists dig up sites of old settlements and attempt to determine the nature of life in early societies from the physical evidence they find. Garbologists sort through people’s garbage to analyze household consumption patterns. Marketing research usually is much more mundane. In a product prototype test, it may be important to learn how much of the test product the consumer used. 

The most accurate way to find this out is to have the respondent return the unused product so that the researcher can see how much is left. If a study involved the in-home use of a laundry soil and stain remover, it would be important to know how much of the remover each respondent actually used. All of the respondents’ answers to questions would be considered from this usage perspective. Pictures can also be used to see what people have done in certain situations. For example, a global study conducted by New York-based GfK NOP created a massive visual database with the goal of better understanding global consumers. Part of that research was photographing people’s kitchens, which in many cultures is the “heart of the home.” Examples of the kind of understanding the researchers gleaned from the photos are shown in Exhibit 8.2.

Advantages of Observation Research
Watching what people actually do rather than depending on their reports of what they did has one very significant and obvious advantage: Firsthand information is not subject to many of the biasing factors associated with the survey approach. Specifically, the researcher avoids problems associated with the willingness and ability of respondents to answer  questions. Also, some forms of data are gathered more quickly and accurately by observation. Letting a scanner record the items in a grocery bag is much more efficient than asking the shopper to enumerate them. Similarly, rather than asking young children which toys they like, major toy manufacturers prefer to invite target groups of children into a large playroom and observe via a one-way mirror which toys are chosen and how long each holds the child’s attention.

Disadvantages of Observation Research
The primary disadvantage of many types of observation research is that only behavior and physical personal characteristics usually can be examined. The researcher does not learn about motives, attitudes, intentions, or feelings. Also, with the exception of online observation, only public behavior is observed; private behavior such as dressing for work or committee decision making within a company is beyond the scope of observation research. A second problem is that present observed behavior may not be projectable into  the future. The fact that a consumer purchases a certain brand of milk after examining several alternatives does not mean that he or she will continue to do so in the future. Observation research can be time-consuming and costly if the observed behavior occurs rather infrequently. For example, if observers in a supermarket are waiting to watch the purchase behavior of persons selecting Lava soap, they may have a long wait. And if the choice of consumers to be observed is biased (e.g., shoppers who go grocery shopping after 5:00 p.m.), distorted data may be obtained.

ADS