Marketing research is the most important research activity for the organizations which touches and assists all vital areas.
Marketing research, as a art of social researches, has acquired matured academic status. It is not restricted to research on any specific marketing problem, but it applies to any phase of marketing.
The marketing research process is an application of the scientific method. The steps in these processes are highly interrelated, and one step leads to the next. The stages in the research process often overlap. Disappointments encountered at one stage may mean returning to previous stages, or even starting over.
Thus, it is something of an oversimplification to present a nearly ordered sequence of activities. Still, the stages of marketing research often follow a generalised pattern of seven stages.
The steps involved in marketing research are as follows:- 1. Defining the Problem 2. Planning the Research Design 3. Selecting a Sample 4. Collecting Data 5. Analysing Data6. Drawing Conclusions and Preparing a Report 7. Following Up 8. Situation Analysis 9. Obtaining Problem Specific Area 10. Problem Solution.
Steps in Marketing Research Process: Steps and Stages
Steps in Marketing Research – 5 Major Steps: Definition of the Problem, Situation Analysis, Obtaining Problem-Specific Data, Interpretation of Data and Problem Solution
The marketing research process is a five-step application of the scientific method.
1. Definition of the problem
2. Situation analysis
3. Obtaining problem-Specific data
4. Interpretation of data
Step # 1. Definition of the Problem:
Defining the problem is the most important and often the most difficult step in the marketing research process. In this step, the objectives of the research must be clearly defined. The manager must think about what decisions need to be made and must clearly specify what information is really needed to make them. The manager and the researcher should both be involved so that both agree on the major objectives of the research.
The problem definition step sounds simple but it is not so. A manager may assume that all of the questionable areas are obvious or that the researcher really understands what information is needed, this is also not the case. Hence the important questions may be ignored while less important questions may be analysed in depth. It is also easy to fall into the trap of mistaking symptoms for the definition of the problem.
Sometimes the research priorities are very clear, for example when a manager only wants to know if the targeted households have tried a new product and what percent of them bought it a second time. But usually it is harder than this. The manager might also want to know why some didn’t buy – or if they had even heard of the product.
There is rarely any time and money to study everything. The manager may have to narrow things down. Developing a priority list that includes all the possible problem areas is sensible. The various items on the list may need to be considered more completely in the situation analysis step before final priorities can be set.
When the marketing manager feels the real problem has begun to surface, a situation analysis is useful. A situation analysis is an informal study of what information is already available in the problem area. The situation analysis may help refine the problem definition and specify what additional information if any is needed.
The situation analysis usually involves informal talks with informed people. Informed people mean others in the firm, a few good gentlemen who have close contact with customers, or others knowledgeable about in the industry. In industrial marketing, where relationships with customers are close the customers themselves may be called; perhaps one of these people has already worked on the same problem, or knows about a useful source of relevant information. Their inputs may help to sharpen the problem definition, too.
The situation analysis is especially important if the researcher is a research specialist who doesn’t know much about the management decisions to be made of if the marketing manager is dealing with unfamiliar areas. They must be sure they understand the problem area, including the nature of the target market, the marketing mix, competition, and other external factors. Otherwise, the researcher may rush ahead and make foolish mistakes or simply “discover” what is already known by management.
This stage of research calls for determining the type of information needed and the most efficient ways to gather this information. A researcher can gather secondary data, primary data or both. Secondary data consists of information that already exists somewhere, having been collected for another purpose.
Primary data consist of originally collected information for the specific purpose at hand. Primary research involves the collection of primary data by the marketing researcher or agents of the Marketing Researcher directly from respondents. Because these data may be collected first-hand the process tends to be more costly and may be more time consuming than is the collection of secondary research data.
However, the use of primary data is sometimes mandatory when secondary data are unavailable. The primary data are usually much more relevant to what is being researched because of the unique situation or problem, or the timing.
Secondary research depends on secondary data or data obtained from sources other than directly from respondents. In other words, the research has already been accomplished. Most researchers usually begin by examining published secondary research sources to see what data already exist that have been collected for some other problem or some other purpose.
In some cases, the problem has already been researched in its entirety, and the data are timely and directly relevant and applicable to the problem at hand. In such a case, use of secondary research data could save the time and money of putting together and conducting a primary research project.
In the observation method, the data are collected by observing some action of the respondent. No interviews are involved, although an interview may be used as a follow-up to get additional information. For example, if customers are observed buying a drink in cans instead of bottles, they may be asked why they prefer that one form of packaging to the other.
Information may be gathered by personal or mechanical observation. In one form of personal observation, the researcher poses as a customer in a store. This technique is useful in getting information about the calibre of the salespeople, or in determining what brands they push. Mechanical observation is illustrated by an electric cord stretched across a highway to count the number of cars that pass during a certain time period.
Another example is a study on food prices, conducted on behalf of a local government. Observers were sent to a large sample of research food stores to obtain price data on every brand available in 65 categories of food products.
The observation method has several advantages:
a. It can be highly precise.
b. It usually removes all doubt about what the consumer does in a given situation.
c. The consumers are unaware that they are being observed, so presumably they act in the usual fashion.
a. This technique reduces bias, but the possibility of bias is not completely eliminated as long as people are used as observers.
b. The technique is limited in its application. Observation tells what happened, but it cannot tell why.
Companies undertake surveys to learn about people’s knowledge, beliefs, preferences, satisfaction, and so on, and to measure these magnitudes or importance in the population. The most common means of gathering primary marketing data is the survey method. It usually leads to a broader range of data than observation or experimentation methods do. Surveys are used to plan product designs, advertising copy, sales promotions, and other marketing activities.
In the survey method, the researcher gathers facts directly from a sample of respondents.
The three principle kinds of surveys are:
(a) Telephone interview,
(b) The mail questionnaire, and
(c) The personal interview.
(a) In a telephone survey, a large number of interviews can be handled quickly at a comparatively low cost than either personal or mail surveys. By telephone the interviewer can talk with a number of family members during the same call and can clarify or elaborate on questions whenever necessary. This method is easy to administer, telephone survey may be timely. One disadvantage of the telephone survey is that interviews may be short. Lengthy interviews can be conducted satisfactorily over phone. Another limitation of the telephone is some people have unlisted numbers, so people do not have phones.
(b) Mail questionnaires also have many merits – Mail questionnaires can be presented in a more objective and controlled manner. It allows respondents to answer questions at their own pace. The haste and distraction of personal and telephone interviews can be avoided. Mailed questionnaires are more reasonable, short and the questions asked are very simple. The major disadvantage with mail questionnaires is the difficulty of obtaining returns.
(c) Personal Interview – When a long, complex questionnaire is used, a personal interview is most suitable. The personal interview provides an opportunity for the researcher to obtain complete answers and to judge the socioeconomic condition of the respondent. Interview bias may occur if the interviewer rewords the question so that the respondent provides anticipated responses. Respondent bias may occur, too, when the respondent relates to the interview what the Interview wants to hear. Personal interviewing is relatively expensive in terms of time and money.
Step # 4. Interpretation of Data:
The interpretation of marketing information is a significant nature of marketing research. The marketing information is properly interpreted and analysed. The information collected will have to be edited, coded, tabulated and analysed to interpret the facts and figures of the markets. One can see that getting a representative sample is very important. The most common method for getting a representative sample is random sampling, where each member of the population has the same chance of being included in the sample. Great care must be used to ensure that sampling is really random and not just haphazard.
The nature of the sample, and how it is selected, makes a big difference in how the results of a study can be interpreted. This should be considered as a part of planning data collection, to make sure that the final results can be interpreted with enough confidence so the marketing manager can use them in his planning.
Even if the sampling is carefully planned, it is also important to evaluate the quality of the research data itself. Besides sampling and validity problems, a marketing manager should consider whether the analysis of the data supports the conclusions drawn in the interpretation step. Sometimes the technical people pick the right statistical procedure and their calculations are exact but they offer a wrong interpretation because they don’t understand the management problem.
In one survey, two wheeler buyers were asked to rank five scooters in order from most preferred to least preferred. One scooter was ranked first by slightly more respondents than any other scooter, so the researcher reported it as the “most liked scooter”. That interpretation however ignored the fact that most of the other respondents ranked the scooter last.
Step # 5. Problem Solution:
In the problem solution step, the results of the research are used in making marketing decisions. At the conclusion of the research process the marketing manager should be able to apply the research findings to marketing strategy planning. For instance, the mix of the four P’s. If the research does not provide the necessary information, to help guide these decisions, the research money probably is wasted. It is to be noted that this step is the logical conclusion to the whole research process. In fact it is the reason for the earlier steps. This final step must be anticipated in every one of the preceding steps.
Steps in Marketing Research – 7 Important Stages: Defining the Problem, Planning the Research Design, Selecting a Sample, Collecting Data, Analysing Data and a Few Others
The steps in these processes are highly interrelated, and one step leads to the next. The stages in the research process often overlap. Disappointments encountered at one stage may mean returning to previous stages, or even starting over. Thus, it is something of an oversimplification to present a nearly ordered sequence of activities. Still, the stages of marketing research often follow a generalised pattern of seven stages.
These stages are:
1. Defining the problem,
2. Planning the research design,
3. Selecting a sample,
4. Collecting data,
5. Analysing data,
6. Drawing conclusions and preparing a report, and
7. Following up.
Again, the stages overlap and affect one-another. In some cases, the “later” stages may even be completed before the “early” ones. Within each stage of the research process, the researcher faces a number of alternative methods, or paths, from which one must be chosen.
In this regard, the research process can be compared to a journey. On any map, some paths are more clearly charted than others. Some roads are direct, and others are roundabout. Some paths are free and clear; others require a toll. The point to remember is that there is no “right” or “best” path.
The road taken depends on where the traveller wants to go and how much time, money, ability, and so forth are available for the trip. Although there is no “right” path, the researcher must choose an appropriate one—that is, one that best suits the particular problem at hand.
In some situations, where time is short, the quickest path is best. In other circumstances, where money, time, and personnel are plentiful, the chosen path may be long and difficult.
Exploring the various paths marketing researchers encounter is the main purpose here, which will describe the seven stages of the research process:
Albert Einstein observed that “the formulation of a problem is often more essential than its solution.” This is valuable advice for marketing managers and researchers, who, in their haste to find the right answer, may fail to ask the right question. Too often, data are collected before the nature of the problem has been carefully established. Except in cases of coincidence or good luck, such data will not help resolve the marketer’s difficulties.
The old adage “a problem well-defined is a problem half-solved” puts all of this into perspective. Careful attention to problem definition allows the researcher to set proper research objectives. Determining what is to be accomplished determines the process to be used.
With the objectives of the research effort clear, the chances of collecting necessary and relevant data will be greater and the likelihood of gathering surplus data diminished. Errors or omissions at this early stage of the research process are likely to lead to costly mistakes later on, when it is harder to correct them.
(a) Problems can be Opportunities:
There are many occasions when the research process is not focused on a problem but on an opportunity. In this circumstance, for example, a toy maker who has developed a fabulous new item might face the “problem” of determining what age groups will most likely want the toy or which of several advertising media is the best to use.
In cases such as these, the problem definition stage of the research might well be called the opportunity definition stage. The point is that the problems addressed by marketing research are frequently “good” problems and not disasters.
(b) Don’t Confuse Symptoms with the Real Problem:
There is a difference between a problem and the symptoms of that problem. Pain, for example, is the symptom of a problem. The cause of the pain, perhaps a broken leg, is the problem. In marketing, falling sales are symptoms that some aspect of the marketing mix is not working properly.
Sales may be falling because price competition has intensified or because buyer preferences have changed. Defining the general nature of the problem provides a direction for the research.
The problem definition stage is likely to begin with the discovery that some problem exists because symptoms have been detected. If managers are uncertain about the exact nature of the problem, they may spend time analysing and learning about the situation.
For example, they may discuss the situation with others, such as sales representatives who are close to the customers. A small-scale exploratory research investigation may be conducted to ensure that Stage 2, planning the research design, will not begin with an inadequate understanding of exactly what information needs to be collected. Exploratory research is optional and is not used in all research projects.
The problem is defined, as a series of research objectives related to the problem are stated. No decisions about the remaining stages of the marketing research process should be made until managers and researchers have a clear understanding of the objectives of the research about to be undertaken.
(c) Exploratory Research:
Exploratory research is sometimes needed to clarify the nature of the marketing problem. The kind of research is a first step that leads to more specific research efforts in Stage 2. Management may know, from noting a symptom such as declining sales or smaller order sizes that some kind of problem is “out there”.
Exploratory research may be used to try to identify the problem. Or management may know what the problem is but not how big or how far-reaching it is. Here too, managers may need research to help them analyse the situation. Providing conclusions is not the purpose of exploratory research.
Its purpose is simply to investigate and explore. Usually, exploratory research is undertaken with the expectation that other types of research will follow and that the subsequent research will be directed at finding possible solutions.
In any research situation, it is generally best to check available secondary data before beginning extensive data collection. Some work at a library with an internal data base may result in a saving of time and money.
However, there isn’t any set formula prescribing exactly how to analyse a situation. Sometimes checking secondary sources may not be the appropriate first step. A short series of interviews with a small number of customers may be in order.
If a fast-food restaurant were considering adding a low cholesterol menu or a line of tacos to its standard hamburger fare, marketing managers might begin their research by conducting some unstructured interviews with customers. Customers might surprise management with negative comments on the proposed additions. Exploratory research in this case could serve to identify problem areas or point to a need for additional information.
Although there are many techniques for exploratory research, our discussion highlights one popular method — the focus group interview—to illustrate the nature of exploratory techniques. Focus group interviews are loosely structured interviews with groups of 6 to 10 people who “focus” on a product or some aspect of buying behaviour.
During a group session, individuals give their comments and reactions to new product ideas or explain why they buy (or do not buy) certain products. Researchers later analyse those comments for meaningful and useful ideas, such as that a product is “too high-priced” or “looks like it would break easily”.
Focus group research is extremely flexible and may be utilised in many diverse exploratory studies. For example, focus groups were used when Arm & Hammer was trying to promote increased usage of banking soda by advertising additional uses for the product.
Historically, baking soda had been accepted as a mild cleanser for refrigerator surfaces. In focus groups exploring alternative usages for baking soda, the theme of sweetening and freshening repeatedly came up in the discussions.
At first, the researchers didn’t pay much attention to the idea of using baking soda to clean the air within refrigerators; they were too preoccupied with looking for cleaning ideas. However, once they recognised the potential of the idea of putting a box of Arm & Hammer inside the refrigerator to remove odours, they again turned to focus groups to determine how to advertise the concept.
“When we put the proposition to respondents directly, your refrigerator smells, and baking soda will cure that’ it didn’t go over at all. But when we came through the back door and worded the proposition in such a way that id didn’t imply the woman was a lousy housekeeper, they showed a lot of interest in their ideas.”
This led to Arm & Hammer’s classic advertising campaign promoting multiple uses for its baking soda. The-accompanying Competitive Strategy feature describes another successful example of focus group research.
Stage # 2. Planning the Research Design:
After the researcher has clearly identified the research problem and formulated a hypothesis, the next step is to develop a format research design. The research design is a master plan that specifically identifies the techniques and procedures that will be used to collect and analyse data relevant to the research problem.
The research design must be carefully compared against the objectives developed in Stage 1 of the process to assure that the sources of data, the data collected, the scheduling the costs involved, and so on are consistent with the researcher’s main goals.
At the outset, the researcher should determine if the data that will answer the research question has already been generated by others. In other words, they must determine if primary research is required. Researchers planning a research design must first choose between a secondary data study and a primary data investigation.
(i) Research Designs — Primary Data:
Researchers who find that no appropriate secondary data are available are faced with choosing from among three basic research designs for collecting primary data.
(b) Observation studies, and
Primary data are commonly generated by survey research. Survey results on one topic or another are reported almost daily by the news media. Most adult Americans have been stopped by interviewers at shopping centres or voting places or have received mailings or phone calls from survey takers. In general, a survey is any research effort in which data is gathered systematically from a sample of people by means of questionnaire. Researchers using surveys may collect data using personal interviews, face-to-face interviews, telephone interviews, or mailed questionnaires.
If the purpose of a research effort is to not actions that are mechanically or visually recordable, observation techniques can form the basis of that effort. Observation research involves the systematic recording of behaviour, objects, or events as they are witnessed.
Companies that sells peace on outdoor billboards are interested in traffic patterns, specifically the numbers of cars and people passing the billboard installations each day. Mass transit organisations may want to know the numbers of people riding each bus and at what stops most of them get on or off.
In both cases, the information could be recorded by either human observers stationed on street corners or in buses or by mechanised counters.
Observation can be more complicated than these simple nose- counting examples might suggest. “Mystery shoppers” can be used to check on the courtesy or product knowledge of retail salespeople. Researchers “disguised” as customers, store employees, or product demonstrators might subtly observe consumer reactions to prices, products, package designs, or display cases.
The customers are unaware their behaviour is being observed. The greatest strength of observation is that it permits the recording of what actually occurs in a particular situation. Its biggest weakness is that the observer cannot be sure why the observed behaviour occurred.
Still, in some cases, it is enough to know that something happened. The A.C. Nielsen Company, for example, uses mechanical observation to rate television shows. Its people meter is a recording device attached to a family’s television that uses a microprocessor to identify what family members are viewing and what station is being watched.
The questions of why a show is popular or why the ratings of old movies beat those of the president’s State of the Union address are often left to the critics.
Experiments have long been used by scientists attempting to discover cause-and-effect relationships. Almost every day we encounter news stories telling us about an experimental group of white mice that were exposed to some substance and then developed more cancers than mice in a group not so exposed.
The assumption, of course, is that the substance involved increased the chance of cancer. A properly run experiment allows the investigator to change one or two variable while observing the effects of those changes on another variable. Ideally, the experimenter holds all factors steady except the ones being manipulated, thus showing that changes can be caused by the factors being studied.
Clearly, the influence of environmental variables are more easily controlled by a scientist working with mice than by a marketing researcher dealing with human being and their reactions to changes. Yet marketing researchers can and to use experimental techniques. These may be used in marketplace, or field, experiments or in a controlled, or laboratory, atmosphere.
Advertisers often use “laboratory” settings to test advertising copy- One group of subjects is shown a television program that includes one version of the advertisement. A second group views the same program but with a different advertisement. Researchers compare the groups’ responses.
Research like this is conducted in a controlled setting, rather than a natural setting, to increase researchers’ control of environmental variables. Such an experiment is known as a laboratory experiment.
Scientists dealing with experiments in physics or chemistry might not be impressed with the experimental techniques used by marketers. Yet they are in fact true experiments within the limits imposed by dealing with people rather than animals or in animate objects. As with surveys, careful planning and execution are required, but properly conducted experiments are valuable sources of marketing information.
(ii) Research Designs — Secondary Data:
Formal marketing research studies do not always involve surveys or the generation of other new or original data. Previously gathered data, already in the researcher’s decision support system or in the library, may be adequate to begin a formal research effort. For example, a marketer of mobile homes might know that sales of this product rise as building permits issued for traditional homes decline.
Using government figures showing building permits issued and trends in home building, the mobile home seller can develop a quantitative model to predict marketing behaviour, basic the research design entirely on the analysis of secondary data.
(iii) Selecting the Research Design:
After considering the many research alternatives, the marketing researcher must pick the research design that will be used. Because there are many alternative ways to tackle a problem, there is no one “best” research design. Certain techniques are simply more appropriate than others.
Once the researcher has determined which research design will most likely yield information useful in solving the marketing problem at hand, the next step is to select a sample of people, organisations, or whatever is of interest.
The methods for selecting the sample are important for the accuracy and appropriateness of the study and depend on the appropriateness of the sample. Though sampling is a highly developed statistical science, its basic concepts are applied in daily life.
For example, the first taste (or sample) or a bowl of soup may indicate that the soup needs salt, is too salty, or is “just right.” Sampling, then, is any procedure in which a small part of the whole is used as the basis for conclusions regarding the whole. A sample is simply a portion, or subset, of a larger population. It makes sense that a sample can provide a good representation of the whole.
Sampling essentially involves answering these three questions:
(i) Who is to be sampled? Specifying the target population, or the total group of interest, is the first aspect of sampling. Suppose a department store manager who wants to analyse the store’s image in the community at large uses current credit-card records to develop a survey mailing list.
Who will be surveyed? Only current credit-card customers, not non-credit customers, and certainly not non-customers, though these groups may be important parts of “the community at large.” Managers want the results of the sample to give an accurate picture of the interest.
Population list from which the sample will be taken may be based on list of customers, telephone directories, membership lists, of automobile, registrations, and many other sources. Selecting a list is a crucial aspect of sampling. If the lists are inaccurate, the sample results may not be representative.
(ii) How big should the sample be? The traditional tongue-in-check response to this question—”big enough”—suggests the true answer. The sample must be big enough to properly represent the target population. In general, bigger samples are better than smaller samples. Nevertheless, if appropriate sampling techniques are used, a small proportion of the total population will give a reliable measure of the whole.
For instance, the Nielsen TV ratings survey, which appears to be highly accurate, involves sampling only a few thousand of the 94 million U.S. Households. The keys here are that most families’ TV viewing habits are similar and that the “Nielsen families” are selected with meticulous care to assure the representativeness of the sample.
(iii) How should the sample be selected? The sampling procedure—the way sampling units are selected—is a major determinant of the accuracy of marketing research. There are two major sampling methods – probability sampling and non-probability sampling.
When the sampling procedures are such that the laws of probability influence the selection of the sample, the result is a probability sample. When sample units are selected on the basis of convenience or personal judgment (for example, if Portland is selected as a sample city because it appears to be typical), the result is a non-probability sample.
a. Probability Sampling:
Simple random sampling, stratified sampling, and cluster sampling are the methods for drawing probability samples.
(i) A simple random sample consists of individuals’ names drawn according to chance selection procedures from a complete list of all people in a population. All these people have the same chance of being selected. The procedure is called simple because there is only one stage in the sampling process.
(ii) The random sampling errors is a statistical phenomenon associated with probability sampling. Chance variation causes such errors. Without increasing sample size, the researcher cannot avoid these statistical problems. However, if simple random sampling or other probability sampling techniques are used, the size of random sampling errors can be estimated and the possible range of error taken into account.
(iii) For a stratified sample, the researcher must first divide the complete list of all members of the population into subgroups, or strata. Strata are chosen based on characteristics relevant to the study. A simple random sample is then taken from each stratum. For example, research on Asian Americans might break out Koreans, Japanese, Vietnamese, Malaysians, and other subgroups. Independent random samples would-be drawn from each of these strata.
(iv) A cluster sample may be used when there are no lists of the sample population available or when it is necessary to minimize the cost of sampling. Suppose the purpose of a research study is to investigate behaviour in the western half of the United States.
Selecting Portland, Denver, and Salt Lake City as representative cities will substantially reduce the amount of time spent travelling in this region. The specific areas, or geographical clusters, are selected on a chance basis; then individuals within the clusters are sampled with a probability sampling technique.
b. Non-Probability Sampling:
There are three types of non-probability samples:
(i) The first is a convenience sample. Here, data are collected from the people who are most conveniently available. A professor or graduate student who administers a questionnaire to a class is using a convenience sample. It is easy and economical to collect the sample data this way. Likewise, the interviewer intercepting consumers at a shopping mall or the reporter interviewing the person on the street issuing a convenience sample.
(ii) With a judgment sample, judgment and experience are used in selecting the sample elements that will contribute most to the study. For example, test markets for new product trials are often selected because they appear to be typical cities.
(iii) A quota sample is another non-probability sample. Here, the sample elements are collected at the convenience of the interviewers, but the characteristics of the sample are matched against quotas. Thus, the researchers predetermine the pertinent sample characteristics (for example, 10 per cent should have incomes above $80,000), and the sample is chosen to match these characteristics.
Stage # 4. Collecting Data:
The problem has been defined, the research techniques chosen, and the sample to be analysed selected. Now the researcher must actually collect the needed data. Whatever collection method is chosen, it is the researcher’s task to minimize errors in the process—and errors are easy to make.
Interviewers who have not been carefully selected and trained, for example, may not phrase their questions properly or may fail to record respondents’ comments accurately.
Generally, the actual collection of the desired data is preceded by the pretesting of the collection method. A proposed questionnaire or interview script might be tried out on a small sample of respondents in an effort to assure that the instructions and questions are clear and comprehensible.
It may develop that the survey instrument is too long, causing respondents to lose interest, or too short, yielding inadequate information. The pretest provides the researcher with a limited amount of data that will give an idea of what can be expected from the up comings full-scale study. In some cases, these data will show that the study is not answering the researcher’s questions. The study may then have to be redesigned.
Once the researcher has completed what is called the “fieldwork” by gathering the data germane to solve the research problem, that data must be manipulated, or processed. The purpose of this process is to place the data in a form that will answer the marketing manager’s questions.
Data processing ordinarily begins with a job called editing, in which surveys or other data collection forms are checked for omissions, incomplete or otherwise unusable responses, illegibility, and obvious inconsistencies. The editing process may result in certain collection forms being discarded.
In research reports it is common to encounter phrases like “One thousand people were interviewed yielding 856 usable responses.” The process may also uncover correctable errors, such as – the recording of a usable response on the wrong line of a questionnaire.
Once the data collection forms have undergone editing, the data undergoes coding. That is, meaningful categories are established so that responses can be grouped into usable classifications. In a survey whose focus is on minority group responses, a race code might be used—for example – 1 – white, 2 – African American, 3 -Native American, 4 – Asian, 5 – Hispanic, 6 – other.
In another situation, in which only the responses of Hispanics are of concern, the code might be – 1 – Hispanic; 2 – non-Hispanic. Such codes are especially useful when computer analysis of the data is to be employed, but they are also helpful when the results are hand tabulated.
After the collected data has been edited and coded, the researcher is ready to undertake the process of analysis. Data analysis may involve statistical analysis or qualitative analysis, or both.
The type of analysis to be used should be based on the information requirements faced by management, the hypothesis, the design of the research itself, and the nature of the data collected. Qualitative, or judgmental, analysis should not be ignored here. Ultimately, the researcher and the marketing manager will have to interpret the results of quantitative analysis.
Statistical tools such as – the T-test of two means, the Chi- square test, and correlation analysis are popular stand-by used to analyse data. It may be surprising, in light of the availability of these and many other techniques, that a great number of studies use statistics no more sophisticated than averages and percentages.
Remember that the purpose of marketing research is to aid in the development of effective marketing decisions. The researcher’s role is to answer the question “What does this mean to marketing mangers?” Therefore, the culmination of the research process must be a report that usefully communicates research findings to management.
If the researcher is competent and the report makes good sense to management, decision makers are not likely to want complicated recounting of technical aspects of the research effort. Typically, management is not interested in how the findings came about. Except in special cases, management is likely to want only a summary of the findings.
Presenting these clearly, using graphs, charts, and other forms of art work, is a creative challenge to the researcher and any others involved in the preparation of the final report. If the researcher’s findings are not properly communicated to and understand by marketing managers, the research process has been, in effect, a total waste.
Stage # 7. Following Up:
After the researcher submits the report to management, he or she should follow up to try to determine if and how management responded to the report. The researcher should ask how the report could have been improved and made more useful. This is not to say that the report’s conclusions or suggested courses of action must be followed by management.
Deciding such things is, after all, the role of management, not of researchers. Marketing management, for its part, should let researchers know how reports can be improved or how future reports might of be better use.