Chapter 2: The Business Research Process

The Business Research Process: Chapter Objectives

Learning Outcomes

After studying this chapter, you should be able to

Define decision making and understand the role research plays in making decisions
Classify business research as either exploratory research, descriptive research, or causal research
List the major phases of the research process and the steps within each
Explain the difference between a research project and a research program

The Business Research Process: Chapter Overview

Chapter Vignette: The Changing Educational Environment

Students seeking a higher education today enjoy many more choices than did their parents. Universities offer new degree programs in varied and specific fields including areas like sports marketing and gaming management. However, it isn't simply the fields of study that may be new, but also the manner of study. Options for nontraditional students who have difficulty attending day classes or devoting years of study to obtaining a degree have grown exponentially. The University of Phoenix, Strayer University, and Nova Southeast typify institutions that specialize in catering to those seeking a nontraditional degree program. These competitive pressures have led even the most traditional universities to rethink the “sage on the stage” approach and conventional academic calendars.

The market for the MBA degree is particularly competitive. Students pursue their MBA either traditionally, in weekend-only programs, at night school, online, or in some combination. Over a quarter of a million U.S. students alone attend MBA classes of one form or another at any given time. In urban areas, such as the Dallas-Fort Worth area, there are sometimes a dozen or more institutions offering an MBA. Even in smaller communities, universities are facing decisions about their MBA offerings as the competitive set and nontraditional programs have entered the market:

How much should they adapt to the changing environment?
Should they offer courses online?
If so, who are they competing with?
Should they offer a weekend program?
Should they offer classes in multiple locations?
Can they better accomplish the mission of the university with an online MBA program?
If multiple formats and locations are offered in the program, should different faculty teach?
Is demand sufficient? That is, are there enough potential students to make this financially feasible?
Is there a potential perceived product quality difference between a traditional and a nontraditional MBA program?

The competitive MBA market typifies the landscape of many organizations. Clearly, universities could benefit from business research addressing some of these key questions. Each university maintains its own academic standard while still trying to attract enough students to make its MBA program feasible. The competitive landscape is filled with both potential opportunities and potential problems. Decisions made by university administrators will determine how successfully each school deals with the changing environment.

Introduction

This chapter focuses on the relationship between business research and managerial decision making. Business success is determined directly by the quality of decisions made by key personnel. Researchers contribute to decision making in several key ways. These include

Helping to better define the current situation
Defining the firm—determining how consumers, competitors, and employees view the firm
Providing ideas for enhancing current business practices
Identifying new strategic directions
Testing ideas that will assist in implementing business strategies for the firm
Examining how correct a certain business theory is in a given situation

The chapter introduces the types of research that allow researchers to provide input to key decision makers. Causality and the conditions for establishing causality are presented. Last but not least, the chapter discusses stages in the business research process.

The Nature of Business Problem

Certainty

Uncertainty

Ambiguity

Businesses face decisions that shape the future of the organization, its employees, and its customers. In each case, the decisions are brought about as the firm either seeks to capitalize on some opportunity or to reduce any potential negative impacts related to some business problem. A business opportunity is a situation that makes some potential competitive advantage possible. The discovery of some underserved segment presents such an opportunity. For example, eBay capitalized on a business opportunity presented by technological advances to do much the same thing that is done at a garage sale but on a very, very large scale.

A business problem is a situation that makes some significant negative consequence more likely. A natural disaster can present a problem for many firms as they face potential loss of property and personnel and the possibility that their operations, and therefore their revenue, will be interrupted. Most business problems, however, are not nearly as obvious. In fact, many are not easily observable. Instead, problems are commonly inferred from symptoms , which are observable cues that serve as a signal of a problem because they are caused by that problem. An increase in employee turnover is generally only a symptom of a business problem, rather than the problem itself. Research may help identify what is causing this symptom so that decision makers can actually attack the problem, not just the symptom. Patients don't usually go to the doctor and point out their problem (such as an ulcer). Instead, they point out the symptoms (upset stomach) they are experiencing. Similarly, decision makers usually hear about symptoms and often need help from research to identify and attack problems. Whether facing an opportunity or a problem, businesses need quality information to deal effectively with these situations.

Formally defined, decision making is the process of developing and deciding among alternative ways of resolving a problem or choosing from among alternative opportunities. A decision maker must recognize the nature of the problem or opportunity, identify how much information is currently available, how reliable it is, and determine what additional information is needed to better deal with the situation. Every decision-making situation can be classified based on whether it best represents a problem or an opportunity and where the situation falls on a continuum from absolute ambiguity to complete certainty.

Certainty

Complete certainty means that the decision maker has all information needed to make an optimal decision. This includes the exact nature of the business problem or opportunity. For example, an advertising agency may need to know the demographic characteristics of subscribers to magazines in

which it may place a client's advertisements. The agency knows exactly what information it needs and where to find the information. If a manager is completely certain about both the problem or opportunity and future outcomes, then research may not be needed at all. However, perfect certainty, especially about the future, is rare.

Uncertainty

Uncertainty means that the manager grasps the general nature of desired objectives, but the information about alternatives is incomplete. Predictions about forces that shape future events are educated guesses. Under conditions of uncertainty, effective managers recognize that spending additional time to gather data that clarify the nature of a decision is needed. For instance, a university may understand that there is an objective of increasing the number of MBA students, but it may not know whether an online, weekend, or off-site MBA program is the best way to accomplish the objective. Business decisions generally involve uncertainty, particularly when a company is seeking new opportunities.

Ambiguity

Ambiguity means that the nature of the problem itself is unclear. Objectives are vague and decision alternatives are difficult to define. This is by far the most difficult decision situation, but perhaps the most common.

Managers face a variety of problems and decisions. Complete certainty and predictable future outcomes may make business research a waste of time. However, under conditions of uncertainty or ambiguity, business research becomes more attractive to the decision makers. Decisions also vary in terms of importance, meaning that some may have great impact on the welfare of the firm and others may have negligible impact. The more important, ambiguous, or uncertain a situation is, the more likely it is that additional time must be spent on business research.

Problems and Opportunities

Exhibit 2.1 depicts decision situations characterized by the nature of the decision and the degree of ambiguity.^¹Under problem-focused decision making and conditions of high ambiguity, symptoms may not clearly point to some problem. Indeed, they may be quite vague or subtle, indicating only small deviations from normal conditions. For instance, a fast-food restaurant may be experiencing small changes in the sales of its individual products, but no change in overall sales. Such a symptom may not easily point to a problem such as a change in consumer tastes. As ambiguity is lessened, the symptoms are clearer and are better indicators of a problem. A large and sudden drop in overall sales may suggest the problem that the restaurant's menu does not fare well compared to competitors’ menus. Thus, a menu change may be in order. However, it is also possible the drop in sales is due to new competition, a competitor's price drop, or new promotional campaign. Thus, research is needed to clarify the situation.

Similarly, in opportunity-oriented research, ambiguity is characterized by environmental trends that do not suggest a clear direction. As the trends become larger and clearer, they are more diagnostic, meaning they point more clearly to a single opportunity.

What Is Research Design

Exploratory Research

Descriptive Research

Causal Research

Uncertainty Influences the Type of Research

Business research is undertaken to reduce uncertainty and focus decision making. In more ambiguous circumstances, management may be totally unaware of a business problem. Alternatively, someone may be scanning the environment for opportunities. For example, an entrepreneur may have a personal interest in softball and baseball. She is interested in converting her hobby into a profitable business venture and hits on the idea of establishing an indoor softball and baseball training facility and instructional center. However, the demand for such a business is unknown. Even if there is sufficient demand, she is not sure of the best location, actual services offered, desired hours of operation, and so forth. Some preliminary research is necessary to gain insights into the nature of such a situation. Without it, the situation may remain too ambiguous to make more than a seat-of-the-pants decision. In this situation, business research is almost certainly needed.

EXHIBIT 2.1: Describing Decision-Making Situations View PDF

Business research can be classified on the basis of either technique or purpose. Experiments, surveys, and observational studies are just a few common research techniques. Classifying research by its purpose, such as the situations described above, shows how the nature of a decision situation influences the research methodology. The following section introduces the three types of business research:

Exploratory
Descriptive
Causal

Matching the particular decision situation with the right type of research is important in obtaining useful research results.

Exploratory Research

Exploratory research is conducted to clarify ambiguous situations or discover potential business opportunities. As the name implies, exploratory research is not intended to provide conclusive evidence from which to determine a particular course of action. In this sense, exploratory research is not an end unto itself. Usually exploratory research is a first step, conducted with the expectation that additional research will be needed to provide more conclusive evidence. Exploratory research is often used to guide and refine these subsequent research efforts. The Research Snapshot “Cute, Funny, or Sexy? What Makes a Mascot Tick?” illustrates a use of exploratory research. For example, rushing into detailed surveys before it is clear exactly what decisions need to be made can waste time, money, and effort by providing irrelevant information. This is a common mistake in business research programs.

Exploratory research is particularly useful in new product development.^²Sony and Honda have each been instrumental in developing robot technology.^³Making a functional robot that can move

around, perform basic functions, carry out instructions, and even carry on a conversation isn't really a problem. What Sony and Honda have to research is what business opportunities may exist based on robot technology. Exploratory research allowing consumers to interact with robots suggests that consumers are more engaged when the robot has human qualities, such as the ability to walk on two legs. Researchers noticed that people will actually talk to the robot (which can understand basic oral commands) more when it has human qualities. In addition, consumers do seem entertained by a walking, talking, dancing robot. These initial insights have allowed each company to form more specific research questions focusing on the relative value of a robot as an entertainment device or as a security guard, and identifying characteristics that may be important to consumers.

Descriptive Research

As the name implies, the major purpose of descriptive research is to describe characteristics of objects, people, groups, organizations, or environments.

Unlike exploratory research, descriptive studies are conducted after the researcher has gained a firm grasp of the situation being studied. This understanding, which may have been developed in part from exploratory research, directs the study toward specific issues. Later, we will discuss the role of research questions and hypotheses. These statements help greatly in designing and implementing a descriptive study. Without these, the researcher would have little or no idea of what questions to ask.

Descriptive research often helps describe market segments. For example, researchers used descriptive surveys to describe consumers who are heavy consumers (buy a lot) of organic food products. The resulting report showed that these consumers tend to live in coastal cities with populations over 500,000, with the majority residing on the West Coast. The most frequent buyers of organic foods are affluent men and women of ages 45–54 (36 percent) and 18–34 (35 percent).^⁴Interestingly, consumers who buy organic foods are not very brand-oriented—81 percent of them cannot name a single organic brand. Research such as this helps high-quality supermarkets such as Whole Foods make location decisions. Over half of Whole Foods’ food products are organic.

A diagnostic analysis seeks to diagnose reasons for business outcomes and focuses specifically on the beliefs and feelings respondents have about and toward specific issues. A research study trying to diagnose slumping French wine sales might ask consumers their beliefs about the taste of French, Australian, and American wines. The results might indicate a deficiency in taste, suggesting that consumers do not believe French wines taste as fruity as do the others. Descriptive research can sometimes provide an explanation by diagnosing differences among competitors, but descriptive research does not provide direct evidence of causality.

Causal Research

If a decision maker knows what causes important outcomes like sales, stock price, and employee satisfaction, then he or she can shape firm decisions in a positive way. Causal inferences are very powerful because they lead to greater control. Causal research seeks to identify cause-and-effect relationships. When something causes an effect, it means it brings it about or makes it happen. The effect is the outcome. Rain causes grass to get wet. Rain is the cause and wet grass is the effect.

The different types of research discussed here are often building blocks—exploratory research builds the foundation for descriptive research, which usually establishes the basis for causal research. Thus, before causal studies are undertaken, researchers typically have a good understanding of the phenomena being studied. Because of this, the researcher can make an educated prediction about the cause-and-effect relationships that will be tested. Although greater knowledge of the situation is a good thing, it doesn't come without a price. Causal research designs can take a long time to implement. Also, they often involve intricate designs that can be very expensive. Even though managers may often want the assurance that causal inferences can bring, they are not always willing to spend the time and money it takes to get them.

Causality

Ideally, managers want to know how a change in one event will change another event of interest. As an example, how will implementing a new employee training program change job performance? Causal research attempts to establish that when we do one thing, another thing will follow. A causal inference is just such a conclusion. While we use the term “cause” frequently in our everyday language, scientifically establishing something as a cause is not so easy. A causal inference can only be supported when very specific evidence exists. Three critical pieces of causal evidence are:

Temporal Sequence
Concomitant Variance
Nonspurious Association

Temporal Sequence

Temporal sequence deals with the time order of events. In other words, having an appropriate causal order of events, or temporal sequence, is one criterion for causality. Simply put, the cause must occur before the effect. It would be difficult for a restaurant manager to blame a decrease in sales on a new chef if the drop in sales occurred before the new chef arrived. If a change in the CEO causes a change in stock prices, the CEO change must occur before the change in stock values.

Concomitant Variation

Concomitant variation occurs when two events “covary” or “correlate,” meaning they vary systematically. In causal terms, concomitant variation means that when a change in the cause occurs, a change in the outcome also is observed. A correlation coefficient, which we discuss in a later chapter, is often used to represent concomitant variation. Causality cannot possibly exist when there is no systematic variation between the variables. For example,

if a retail store never changes its employees’ vacation policy, then the vacation policy cannot possibly be responsible for a change in employee satisfaction. There is no correlation between the two events. On the other hand, if two events vary together, one event may be causing the other. If a university increases its number of online MBA course offerings and experiences a decrease in enrollment in its traditional in-class MBA offerings, the online course offerings may be causing the decrease. But the systematic variation alone doesn't guarantee it.

Nonspurious Association

Nonspurious association means any covariation between a cause and an effect is true, rather than due to some other variable. A spurious association is one that is not true. Often, a causal inference cannot be made even though the other two conditions exist because both the cause and effect have some common cause; that is, both may be influenced by a third variable. For instance, there is a strong, positive correlation between ice cream purchases and murder rates—as ice cream purchases increase, so do murder rates.^⁵When ice cream sales decline, murder rates also drop. Do people become murderers after eating ice cream? Should we outlaw the sale of ice cream? This would be silly because the concomitant variation observed between ice cream consumption and murder rates is spurious. A third variable is actually important here. People purchase more ice cream when the weather is hot. People are also more active and likely to commit a violent crime when it is hot. The weather, being associated with both may actually cause both. Exhibit 2.2 illustrates the concept of spurious association.

Establishing evidence of nonspuriousness can be difficult. If a researcher finds a third variable that is related to both the cause and effect, which causes a significant drop in the correlation between the cause and effect, then a causal inference becomes difficult to support. Although the researcher would like to rule out the possibility of any alternative causes, it is impossible to observe the effect of every variable on the correlation between the cause and effect. Therefore, the researcher must use logic, or a theory, to identify the most likely “third” variables that would relate significantly to both the cause and effect. The researcher must control for these variables in some way. In addition, the researcher should use theory to make sure the assumed cause-and-effect relationship truly makes sense.

In summary, causal research should do all of the following:

Establish the appropriate causal order or sequence of events
Measure the concomitant variation between the presumed cause and the presumed effect
Examine the possibility of spuriousness by considering the presence of alternative plausible causal factors

EXHIBIT 2.2: The Spurious Effect of Ice Cream

Degrees of Causality

In everyday language, we often use the word “cause” in an absolute sense. For example, a warning label used on cigarette packages claims “smoking causes cancer.” Is this true in an absolute sense? Absolute causality means the cause is necessary and sufficient to bring about the effect. Thus, if we find only one smoker who does not eventually get cancer, the claim is false. Although this is a very strong inference, it is impractical to think that we can establish absolute causality in the behavioral sciences.

Why do we continue to do causal research then? Well, although managers may like to be able to draw absolute conclusions, they can often make very good decisions based on less powerful inferences. Conditional causality means that a cause is necessary but not sufficient to bring about an effect. This is a weaker causal inference. One way to think about conditional causality is that the cause can bring about the effect, but it cannot do so alone. If other conditions are right, the cause can bring about the effect. We know there are other medical factors that contribute to cancer. For instance, genetics, lifestyle, and diet are also plausible causes of cancer. Thus, if one smokes and has a genetic disposition, diet, and lifestyle that promote cancer, smoking could be considered a conditional cause of cancer. However, if we can find someone who has contracted cancer and never smoked, the causal inference would be proven wrong.

Contributory causality is the weakest form of causality, but it is still a useful concept. A cause need be neither necessary nor sufficient to bring about an effect. However, causal evidence can be established using the three factors discussed. For any outcome, there may be multiple causes. So, an event can be a contributory cause of something so long as the introduction of the other possible causes does not eliminate the correlation between it and the effect. This will become clearer when we discuss ways to test relationships later in the text. Smoking then can be a contributory cause of cancer so long as the introduction of other possible causes does not cause both smoking and cancer.

Experiments

Business experiments hold the greatest potential for establishing cause-and-effect relationships. An experiment is a carefully controlled study in which the researcher manipulates a proposed cause and observes any corresponding change in the proposed effect. An experimental variable represents the proposed cause and is controlled by the researcher by manipulating it. Manipulation means that the researcher alters the level of the variable in specific increments.

For example, consider a manager who needs to make decisions about the price and distribution of a new video game called the Wee Box. She understands that both the price level and the type of retail outlet in which the product is placed are potential causes of sales. A study can be designed which manipulates both the price and distribution. The price can be manipulated by offering it for $100 among some consumers and $200 among others. Retail distribution may be manipulated by selling the Wee Box at discount stores in some consumer markets and at specialty electronics stores in others. The retailer can examine whether price and distribution cause sales by comparing the sales results in each of the four conditions created. Exhibit 2.3 illustrates this study.

An experiment like the one described above may take place in a test-market. Test-marketing is a frequently used form of business experimentation. A test-market is an experiment that is conducted within actual business conditions. McDonald's restaurants have a long-standing tradition of test-marketing new product concepts by introducing them at selected stores and monitoring sales and customer feedback. Recently, McDonald's extensively test-marketed McCafé specialty coffees and beverages. These products were sold at a group of McDonald's outlets and feedback was used to refine the offering including the size of the cups, prices, and what types of extras to add to the drink (including sprinkles of chocolate, whipped cream, steamed milk, and chocolate, vanilla, and caramel shots). McDonald's could then monitor the effect on overall sales, as well

EXHIBIT 2.3: Testing for Causes with an Experiment
	Wee Box Sales by Condition
	High Price	Low Price
Specialty Distribution	Peoria, Illinois: Retail Price: $200 Retail Store: Best Buy	Des Moines, Iowa: Retail Price: $100 Retail Store: Best Buy
General Distribution	St. Louis, Missouri: Retail Price $200 Retail Store: Big Cheap-Mart	Kansas City, Missouri: Retail Price: $100 Retail Store: Big Cheap-Mart
Assuming that Wee Box consumers are the same in each of these cities, the extent to which price and distribution cause sales can be examined by comparing the sales results in each of these 4 conditions.

EXHIBIT 2.3: Testing for Causes with an Experiment View PDF

as cannibalization of regular coffee sales, in a real-world setting. Earlier, McDonald's had test-marketed Wi-Fi service in some outlets. Three different rival Wi-Fi service providers (the manipulation) were used in different locations and the cost, service, and customer feedback were used to select the best provider for use in McDonald's restaurants.

Most basic scientific studies in business (for example, the development of theories about employee motivation or consumer behavior) ultimately seek to identify cause-and-effect relationships. In fact, we often associate science with experiments. To predict a relationship between, say, price and perceived quality of a product, causal studies often create statistical experiments with controls that establish contrast groups.

Uncertainty Influences the Type of Research

So, which form of research—exploratory, descriptive, or causal—is appropriate for the current situation? The most appropriate type and the amount of research needed are largely a function of how much uncertainty surrounds the situation motivating the research. Exhibit 2.4 contrasts the types of research and illustrates that exploratory research is conducted during the early stages of decision making. At this point, the decision situation is usually highly ambiguous and management is very uncertain about what

EXHIBIT 2.4: Characteristics of Different Types of Business Research
	Exploratory Research	Descriptive Research	Causal Research
Amount of Uncertainty Characterizing Decision Situation	Highly ambiguous	Partially defined	Clearly defined
Key Research Statement	Research question	Research question	Research hypothesis
When Conducted?	Early stage of decision making	Later stages of decision making	Later stages of decision making
Usual Research Approach	Unstructured	Structured	Highly Structured
Examples	“Our sales are declining for no apparent reason.” “What kinds of new products are fast-food customers interested in?”	“What kind of people patronize our stores compared to our primary competitor?” “What product features are most important to our customers?”	“Will consumers buy more products in a blue package?” “Which of two advertising campaigns will be more effective?”
Nature of Results	Discovery oriented, productive, but still speculative. Often in need of further research.	Can be confirmatory although more research is sometimes still needed. Results can be managerially actionable.	Confirmatory oriented. Fairly conclusive with managerially actionable results often obtained.

Today, a new aspect called competitive intelligence (CI) is also associated with marketing research; it may be defined as a process for better understanding of a firm's competitors and the competitive environment to enhance its marketplace competitiveness. This process is obviously ethical with the legally permissible collection and analysis of competitor data, obtained from information databases, “open sources,” or through ethical marketing research. CI is also a continuous process of ethical collection and analysis of information with controlled propagation of actionable intelligence to the decision makers.

Stages in the Research Process

Alternatives in the Research Process

Research Objectives

Business research, like other forms of scientific inquiry, involves a sequence of highly interrelated activities. The stages of the research process overlap continuously, and it is clearly an oversimplification to state that every research project has exactly the same ordered sequence of activities. Nevertheless, business research often follows a general pattern. We offer the following research business stages:

Defining the research objectives
Planning a research design
Planning a sample
Collecting the data
Analyzing the data
Formulating the conclusions and preparing the report

Exhibit 2.5 portrays these six stages as a cyclical or circular-flow process. The circular-flow concept is used because conclusions from research studies can generate new ideas and knowledge that can lead to further investigation. Thus, there is a dashed connection between conclusions and reporting and defining the research objectives. Notice also that management is in the center of the process. The research objectives cannot be properly defined without managerial input. After all, it is the manager who ultimately has to make the decision. It is also the manager who may ask for additional research once a report is given.

EXHIBIT 2.5: Stages of the Research Process View PDF

In practice, the stages overlap somewhat from a timing perspective. Later stages sometimes can be completed before earlier ones. The terms forward linkage and backward linkage reflect the interrelationships between stages. Forward linkage implies that the earlier stages influence the later stages. Thus, the research objectives outlined in the first stage affect the sample selection and the way data are collected. The sample selection question affects the wording of questionnaire items. For example, if the research concentrates on respondents with low educational levels, the questionnaire wording will be simpler than if the respondents were college graduates.

Backward linkage implies that later steps influence earlier stages of the research process. If it is known that the data will be collected via e-mail, then the sampling must include those with e-mail access. A very important example of backward linkage is the knowledge that the executives who will read the research report are looking for specific information. The professional researcher anticipates executives’ needs for information throughout the planning process, particularly during the analysis and reporting.

Alternatives in the Research Process

The researcher must choose among a number of alternatives during each stage of the research process. The research process can be compared to a map. It is important to remember that there is no single “right” path for all journeys. The road one takes depends on where one wants to go and the resources (money, time, labor, and so on) available for the trip. The map analogy is useful for the business researcher because there are several paths that can be followed at each stage. When there are severe time constraints, the quickest path may be most appropriate. When money and human resources are plentiful, more options are available and the appropriate path may be quite different.

Chapter 1 introduced the research process. Here, we briefly describe the six stages of the research process. Later, each stage is discussed in greater depth. Exhibit 2.6 shows the decisions that researchers must make in each stage. This discussion of the research process begins with research objectives, because most research projects are initiated to remedy managers’ uncertainty about some aspect of the firm's business program.

Research Objectives

Exhibit 2.6 shows that the research process begins with research objectives . Research objectives are the goals to be achieved by conducting research. In consulting, the term deliverables is often used to describe the objectives to a research client. The genesis of the research objectives lies in the type of decision situation faced. The objectives may involve exploring the possibilities of entering a new market. Alternatively, they may involve testing the effect of some policy change on employee job satisfaction. Different types of objectives lead to different types of research designs.

In applied business research, the objectives cannot really be determined until there is a clear understanding of the managerial decision to be made. This understanding must be shared between the actual decision maker and the lead researcher. We often describe this understanding as a problem statement. In general usage, the word problem suggests that something has gone wrong. This isn't always the case before research gets started. Actually, the research objective may be to simply clarify a situation, define an opportunity, or monitor and evaluate current business operations. The research objectives

EXHIBIT 2.6: Flowchart of the Business Research Process

cannot be developed until managers and researchers have agreed on the actual business “problem” that will be addressed by the research. Thus, they set out to “discover” this problem through a series of interviews and through a document called a research proposal.

It should be noted that this process is oriented more toward discovery than confirmation or justification. Managers and researchers alike may not have a clear-cut understanding of the situation at the outset of the research process. Managers may only be able to list symptoms that could indicate a problem. For example, employee turnover is increasing, but management may not know the exact nature of the problem. Thus, the problem statement often is made only in general terms; what is to be investigated is not yet specifically identified.

Defining the Managerial Decision Situation

In business research, the adage “a problem well defined is a problem half solved” is worth remembering. Similarly, Albert Einstein noted that “the formulation of a problem is often more essential than its solution.”^⁶These phrases emphasize that an orderly definition of the research problem provides direction to the investigation. Careful attention to problem definition allows the researcher to set the proper research objectives. When the purpose of the research is clear, the chances of collecting the necessary and relevant information, and not collecting surplus information, will be much greater.

Managers often are more concerned with finding the right answer rather than asking the right question. They also want one solution quickly, rather than having to spend time considering many possible solutions. However, properly defining a problem can be more difficult than actually solving it. In business research, if data are collected before the nature of the problem is carefully thought out, they probably will not yield useful information.

Much too often research is conducted without a clear definition of the objectives. Researchers forget that the best place to begin a research project is at the end. In other words, knowing what is to be accomplished determines the research process. An error or omission in specifying objectives is likely to be a costly mistake that cannot be corrected in later stages of the research process.

Exploratory Research

Exploratory research can be used to help identify and clarify the decisions that need to be made. These preliminary research activities can narrow the scope of the research topic and help transform ambiguous problems into well-defined ones that yield specific research objectives. By investigating any existing studies on the subject, talking with knowledgeable individuals, and informally investigating the situation, the researcher can progressively sharpen the focus of the research. After such exploration, the researcher should know exactly which data to collect during the formal phases of the project and how to conduct the project. Exhibit 2.6 indicates that managers and researchers must decide whether to use one or more exploratory research techniques. As Exhibit 2.6 indicates, this stage is optional.

The business researcher can employ techniques from four basic categories to obtain insights and gain a clearer idea of the problem: previous research, pilot studies, case studies, and experience surveys. These are discussed in detail in later chapters. This section will briefly discuss previous research and focus group interviews, the most popular type of pilot study.

Previous Research

As a general rule, researchers should first investigate previous research to see whether or not others may have already addressed similar research problems. Initially, internal research reports should be searched within the company's archives.

Previous research may also exist in the public domain. The first place researchers will likely look today is online. The Internet and modern electronic search engines available through most university libraries have made literature reviews simpler and faster to conduct. A literature review is a directed search of published works, including periodicals and books, that discusses theory and presents empirical results that are relevant to the topic at hand. While a literature survey is common in applied research studies, it is a fundamental requirement of a basic research report.

Suppose, for example, that a bank is interested in determining the best site for additional automated teller machines. A logical first step would be to investigate the factors that bankers in other parts of the country consider important. By reading articles in banking journals, management might quickly discover

Definition of Business Research Design

Selection of the Basic Research Method

The “Best” Research Design

Sampling

Gathering Data

Processing and Analyzing Data

Drawing Conclusions and Preparing a Report

After the researcher has formulated the research problem, he or she must develop the research design as part of the research design stage. A research design is a master plan that specifies the methods and procedures for collecting and analyzing the needed information. A research design provides a framework or plan of action for the research. Objectives of the study determined during the early stages of research are included in the design to ensure that the information collected is appropriate for solving the problem. The researcher also must determine the sources of information, the design technique (survey or experiment, for example), the sampling methodology, and the schedule and cost of the research.

Selection of the Basic Research Method

Here again, the researcher must make a decision. Exhibit 2.6 shows four basic design techniques for descriptive and causal research: surveys, experiments, secondary data, and observation. The objectives of the study, the available data sources, the urgency of the decision, and the cost of obtaining the data will determine which method should be chosen. The managerial aspects of selecting the research design will be considered later.

In business research, the most common method of generating primary data is the survey. Most people have seen the results of political surveys by Gallup or Harris Online, and some have been respondents (members of a sample who supply answers) to research questionnaires. A survey is a research technique in which a sample is interviewed in some form or the behavior of respondents is observed and described in some way. The term surveyor is most often reserved for civil engineers who describe some piece of property using

The objective of many research projects is merely to record what can be observed—for example, the number of automobiles that pass by a proposed site for a gas station. This can be mechanically recorded or observed by humans. Research personnel known as mystery shoppers may act as customers to observe actions of sales personnel or do comparative shopping to learn prices at competing outlets. A mystery shopper is paid to pretend to be a customer and gather data about the way employees behave and the way they are treated in general. How often are store policies followed? Are they treated courteously? Mystery shoppers can be valuable sources for observational data.

The main advantage of the observation technique is that it records behavior without relying on reports from respondents. Observational data are often collected unobtrusively and passively without a respondent's direct participation. For instance, Nielsen Media Research uses a “people meter” attached to television sets to record the programs being watched by each household member. This eliminates the possible bias of respondents stating that they watched the president's State of the Union address rather than Gossip Girl on another station.

Observation is more complex than mere “nose counting,” and the task is more difficult than the inexperienced researcher would imagine. While observation eliminates potential bias from interviewer interaction, several things of interest, such as attitudes, opinions, motivations, and other intangible states of mind, simply cannot be observed.

The “Best” Research Design

It is argued that there is no single best research design. As such, the researcher often has several alternatives that can accomplish the stated research objectives. Consider the researcher who must forecast sales for the upcoming year. Some commonly used forecasting methods are surveying executive opinion, collecting sales force composite opinions, surveying user expectations, projecting trends, and analyzing environmental factors. Any one of these may yield a reliable forecast.

The ability to select the most appropriate research design develops with experience. Inexperienced researchers often jump to the conclusion that a survey methodology is usually the best design because they are most comfortable with this method. When Chicago's Museum of Science and Industry wanted to determine the relative popularity of its exhibits, it could have conducted a survey. Instead, a creative researcher familiar with other research designs suggested a far less expensive alternative: an unobtrusive observation technique. The researcher suggested that the museum merely keep track of the frequency with which the floor tiles in front of the various exhibits had to be replaced, indicating where the heaviest traffic occurred. When this was done, the museum found that the chick-hatching exhibit was the most popular. This method provided the same results as a survey but at a much lower cost.

MORE INFO COLLAPSE ALL

To the Point

You cannot put the same shoe on every foot.

—Publius Syrus

Sampling

Although the sampling plan is outlined in the research design, the sampling stage is a distinct phase of the research process. For convenience, however, we will treat the sample planning and the actual sample generation processes together in this section.

If you take your first bite of shrimp po-boy and conclude that it needs Tabasco, you have just conducted a sample. Sampling involves any procedure that draws conclusions based on measurements of a portion of the population. In other words, a sample is a subset from a larger population. If certain statistical procedures are followed, a researcher need not select every item in a population because the results of a good sample should have the same characteristics as the population as a whole. Of course, when errors are made, samples do not give reliable estimates of the population.

Gathering Data

The data gathering stage begins once the sampling plan has been formalized. Data gathering is the process of gathering or collecting information. Data may be gathered by human observers or interviewers, or they may be recorded by machines as in the case of scanner data and Web-based surveys.

Obviously, the many research techniques involve many methods of gathering data. Surveys require direct participation by research respondents. This may involve filling out a questionnaire or interacting with an interviewer. In this sense, they are obtrusive. Unobtrusive methods of data gathering are those in which the subjects do not have to be disturbed for data to be collected. They may even be unaware that research is going on at all. For instance, a simple count of motorists driving past a proposed franchising location is one kind of data gathering method. However the data are collected, it is important to minimize errors in the process. For example, the data gathering should be consistent in all geographical areas. If an interviewer phrases questions incorrectly or records a respondent's statements inaccurately (not verbatim), major data collection errors will result.

Processing and Analyzing Data

Editing and Coding

After the fieldwork has been completed, the data must be converted into a format that will answer the manager's questions. This is part of the data processing and analysis stage. Here, the information content will be mined from the raw data. Data processing generally begins with editing and coding the data. Editing involves checking the data collection forms for omissions, legibility, and consistency in classification. The editing process corrects problems such as interviewer errors (an answer recorded on the wrong portion of a questionnaire, for example) before the data are transferred to the computer.

Before data can be tabulated, meaningful categories and character symbols must be established for groups of responses. The rules for interpreting, categorizing, recording, and transferring the data to the data storage media are called codes. This coding process facilitates computer or hand tabulation. If computer analysis is to be used, the data are entered into the computer and verified. Computer-assisted (online) interviewing is an example of the impact of technological change on the research process. Telephone interviewers, seated at computer terminals, read survey questions displayed on the

Be sure to fully understand the differing roles of exploratory, descriptive, and causal research:
- Exploratory research provides new insights—the domain of discovery in philosophy of science terms—and often sets the groundwork for further investigation.
- Descriptive research describes the characteristics of objects, people, or organizations. Much of business information is based on descriptive research.
- Causal research is the only research that establishes cause-and-effect relationships. Most commonly, causal research takes the form of experiments such as test markets.
A major flaw in business research is to not give due diligence to exploratory research (especially secondary data and qualitative research). Instead, researchers often move too quickly to collecting descriptive data.
A second, and related, flaw in business research is to fail to carefully define the research objectives.

Data Analysis

Data analysis is the application of reasoning to understand the data that have been gathered. In its simplest form, analysis may involve determining consistent patterns and summarizing the relevant details revealed in the investigation. The appropriate analytical technique for data analysis will be determined by management's information requirements, the characteristics of the research design, and the nature of the data gathered. Statistical analysis may range from portraying a simple frequency distribution to more complex multivariate analyses approaches, such as multiple regression. Later chapters will discuss three general categories of statistical analysis: univariate analysis, bivariate analysis, and multivariate analysis.

Drawing Conclusions and Preparing a Report

One of the most important jobs that a researcher performs is communicating the research results. This is the final stage of the research project, but it is far from the least important. The conclusions and report preparation stage consist of interpreting the research results, describing the implications, and drawing the appropriate conclusions for managerial decisions. These conclusions should fulfill the deliverables promised in the research proposal. In addition, it's important that the researcher consider the varying abilities of people to understand the research results. The report shouldn't be written the same way to a group of Ph.D.'s as it would be to a group of line managers.

All too many applied business research reports are overly complicated statements of technical aspects and sophisticated research methods. Frequently, management is not interested in detailed reporting of the research design and statistical findings, but wishes only a summary of the findings. If the findings of the research remain unread on the manager's desk, the study will have been useless. The importance of effective communication cannot be overemphasized. Research is only as good as its applications.

Learning Objectives	Key Concepts
Define decision making and understand the role research plays in making decisions	Decision making occurs when managers choose among alternative ways of resolving problems or pursuing opportunities. Every business decision can be classified on a continuum ranging from complete certainty to absolute ambiguity. Research is a way that managers can become informed about the different alternatives and make an educated guess about which alternative, if any, is the best to pursue.
Classify business research as either exploratory research, descriptive research, or causal research.	Exploratory, descriptive, and causal research are three major types of business research projects. When the decision is very ambiguous, or the interest is on discovering new ideas, exploratory research is most appropriate. Descriptive research attempts to paint a picture of the given situation by describing characteristics of objects, people, or organizations. Causal research identifies cause and-effect relationships.
List the major phases of the research process and the steps within each.	The sixmajor phases of the research process discussed here are 1) defining the research objectives, 2)planning the research design, 3) sampling, 4) data gathering, 5) data processing and analysis, and 6) drawing conclusions and report preparation. Each stage involves several activities or steps.
Explain the difference between a research project and a research program.	A research project addresses one of a small number of research objectives that can be included in a single study. In contrast, a research program represents a series of studies addressing multiple research objectives.

View PDF

Key Definitions

The Business Research Process: Multiple Choice Questions

1. Which of the following represents a situation that makes some potential competitive advantage possible for a business?

a. business threat

b. forward linkage

c. symptom

d. business opportunity

2. Which of the following means that the decision maker has all information needed to make an optimal decision?

a. certainty

b. ambiguity

c. concomitant variation

d. non-spurious association

3. In which situation do symptoms exist, but are subtle and few, making problem identification difficult?

a. problem-focused decision making and conditions of high ambiguity

b. problem-focused decision making and conditions of low ambiguity

c. opportunity-oriented research and conditions of high ambiguity

d. opportunity-oriented research and conditions of low ambiguity

4. All of the following are types of business research EXCEPT _____.

a. exploratory

b. selective

c. descriptive

d. causal

5. ____ research aims to clarify ambiguous situations or discover ideas that may amount to true business opportunities.

a. Exploratory

b. Preliminary

c. Clarifying

d. Descriptive

6. Which type of business research addresses who, what, when, where, why, and how questions?

a. causal research

b. exploratory research

c. descriptive research

d. proscriptive research

7. Which type of association is said to occur when any covariation between a cause and effect is indeed due to the cause and not simply due to some other variable?

a. nonspurious association

b. spurious association

c. concomitant association

d. temporal association

8. The goals that researchers intend to achieve by conducting research as referred to as _____.

a. results

b. causal inferences

c. research outcomes

d. research objectives

9. When the researcher has only one or a small number of research objectives that can be addressed in a single study, that study is referred to as a _____.

a. research project

b. research program

c. research assessment

d. research snapshot

10. When a researcher conducts numerous related studies that come together to address multiple, related research objectives, we refer to this as a research _____.

a. agenda

b. project

c. program

d. conglomeration