Data Objectives: The research objectives were derived from the decision objectives. That served as a basis of understanding to assure the client of relevant findings. That does not, though suffice to direct the project’s planning.
The statement tells the nature of the results expected from the study, what it should enable the client to conclude or predict. It states the particular marketing actions (in pricing and composing) whose effects are to be measured (in general terms, which can be defined more sharply). It does not tell what data are to be obtained, or from whom, to make the predictions. And it fails to specify the methods to be employed.
The data objectives are derived from the research objectives and comprise what we have observed to be lacking in the example. Their determination rests mainly on the researcher, to translate what the decision maker wants into a specific description of the needed data.
Let us first look at the general qualities that should be required of the data:
The measurements will be relevant to the decisions faced and will guide their key aspects. The data will be accurate in both
- Validity, that is, they will measure what they are supposed to, and
- Reliability that is, repeating the same methods would produce the same results.
That data can be obtained quickly enough and at an affordable cost.
Those three principles are obviously generalities, but they are essential. As the researcher works toward precise data objectives, those that are tailored to the client’s particular decisions, the task becomes more difficult. When clients have clearly described their perception of the task becomes more difficult. When clients have clearly described their perception of the problems and their decision methods The researchers in turn can put the data objectives on target for the client’s use.
Besides considering decision requirements, the researcher should consider the personal objectives and decision-making style of the clients. Unless these needs are met, clients may lack confidence in a style of the clients. Unless these needs are met, clients may lack confidence in study findings and reject them. Sometimes researchers are too fascinated with sophisticated methods and novel concepts to recognize what the client really wants.
Data Types: A nearly endless variety of data now exist or can be obtained, but only a few types are relevant to each study’s data objectives. Researchers have a substantial task a selecting the prices types of data to acquire. To be able to make this selection efficiently, one should sift through a number of data types to focus one the suitable once. For that reason, we now will describe numerous data classifications that indicate something of that background. It may seem tedious to read still more lists, but recognize that they are here as examples and not as things to memorize. We will just indicate what is done during it and first give clues to specifying data by describing two types of classification:
- The data’s nature and
- Its function in the ultimate interpretation and analysis
Nature of the Data
We are categorizing data here in general terms of their meaning. There are distinct differences among the meanings of facts, knowledge, opinions, intentions, and motives.
Facts include the measurements of anything that actually exists or has existed. Usually, facts describe tangible things. Although they can be intangible as long as they can really be determined. It may be a fact that Smith and Sons sold 417 new Plymouths last year and that this bettered the previous year by 55 units. It may be a fact too that Joe Smith dominates the partners who own the firm, but this fact might defy measurement.
Facts are ideal in the sense of possible measurement accuracy. The interpretations we place on real facts, through, may be inaccurate. We must beware of the danger that what we are dealing with the “quasi-facts – as we would call them –which are seemingly facts or reality. We are often learning that long – accepted truth were never valid. Many “facts” too are based on estimates or on samples that have a degree of unreliability. These latter may be used in research. But should not be treated as absolute truth Some of our fundamental data (like the gross national product) have to be based on quasi-facts.
It helps data specifications also to recognize that facts have many subtypes, and the relevance to us is those descriptive of people. We will mention four of them;
Demographic: – These are facts that describe the population to which the data refer, and in marketing, much of our description is economic, for example, that a family responding in a survey has an annual income of $21,700. The composition and age of family members or of one person also would be a form of demographics.
Sociological: – These data describe how people are organized in and relate to society, for example, groups or churches to which they belong.
Psychographics: – Facts that describe the life – style of an individual or of a group, in some respect pertinent to the study. For a hypothetical example, automobile sales people tend to live in apartments rather than in single homes (more often than average U.S. adults). That might be quantified, that 69 percent of them rent.
Behavioral: – What people do, how they actually behave, are facts of high impotence in many marketing studies. If a large proportions of new car buyers visits three to five dealers, all with the same make of car; this could be very significant in a dealer’s decision on how to treat them.
Knowledge (that is, what people know) also may be desired data, since that information (be it true or false) may be a determinant of what they do. Consumers’ knowledge or awareness of products or brands in an example that is an indicator of the effectiveness of past communication, when setting goals or deciding the scale of future advertising. Opinions are how people perceive something – what they believe about it and what those beliefs signify. The most potent form of opinions tends to be attitudes, which are mental sets or predispositions to act in some manner (e.g., to decide in advance not to accept the trade – in offered by the first dealer visited). Another form is images of what something is like (e.g., how a New Yorker envisions the New York Telephone Co., Opinions are significant, of course, as they affect behavior, and attitudes exert a general and consistent influence.
For example, Mrs. R believes that her balanced diet supplies plenty of vitamins and that she has no need to pay a premium price to obtain them in breakfast cereal. As a result, her behavior is to but those cereals with the lower prices per ounce.
Intentions are the acts that people have in mind to do, expectations of their behavior. The extent to which people intend to commit a particular marketing behavior and changes in these intentions may be key information. If Union Motors Corporation learns from a monthly data service that 14.1 percent of consumer families intend to buy a new car within the next 12 months, as against 14.4 percent in the previous report, it may receives its production plans.
Motives are the internal forces that cause people to behave as they do. Marketers would dearly like to have accurate data on the motives that impel buyers’ actions relative to the marketer’s product categories. Many motives are quite obvious or are subjects about which people will speak freely. The basic causation of behaviour instead may lie deeply and be difficult to draw out. You will meet some techniques, their, that may educe motivational data. That often cannot be done, so the researcher needs to select other kinds of data with which the motives might be inferred.
Functions of the Data : Also relevant in specifying data needs is to have it classified in terms of how it will be utilized, at the stage of analysis, when the researcher brings together various bits of data and synthesizes conclusions from them. As that implies, one needs to anticipate early the future analysis and synthesis of the data and how they would function in that process. This takes time and it is challenging. But it avoids wasted time and money gathering redundant data while making it likely that the collected data will fit the decision needs.
Let us begin by thinking about a simple experiment, in which one hypothesis (or possible cause) is to be measured in terms of its effects. If we have one cause (x) and one effect (Y), this hypothesis is simply expressed as Y = f (X)
To make this determination, the reader needs to conceive of and specify just two kinds of data: the causation (X) and the effect or payoff (1). These are our two first functional categories. The researcher in Union Motors wants to find out whether taking a demonstration ride (the cause) affects whether a car is bought (the payoff).
Considering this example further, we may recognize that the character of the person of family buying a car may have significant effects on the payoff. That is, some kinds of people or those in certain circumstances may be more affected by the causal variable than others. Therefore, the researcher needs to obtain an adequate description of them. With certain data These comprise our third functional classification of data, description. This category of data is needed also to describe the sample or cross section of the population that the study has covered.
As an example, consider again the proposed study for Union Motors. For interpretation purposes an important factor might be whether the subject(s) in each car-buying incident was an individual man or woman or some combination of two or more people. To check the sample, maybe the researcher wants to know the make of car(s) now owned by the shoppers(s). Thus descriptive data would play important functions and must be specified to be gathered.
There is still a fourth function to be served by the data; that of identification of the person who obtained an interview or made observations, the name or addresses the subject, or the location where the data were obtained.
Identification data for the Union Motors, study could be the date and the place of obtaining the information (e.g., at Smith and Sons on March 18, 1986) and the field staff member who made the interview (Sam ide). To summarize our terminology for functional classifications of data, they are these four
- Causation
- Payoff
- Description
- Identification
Sources of Data: There are numerous possible sources of data, and again we cannot list them in detail. Anyway, this is a determination that is special to each project. A step toward that determination is having, first, general classification of sources, which we now offer in several dimensions.
Secondary sources should first be considered, which refer to those for already gathered and available data (in contrast with primary data). There may be internal sources within the client’s firm. Externally, these sources may include books or periodicals, published reports, data services, and computer data banks.
Primary data may be obtained from individuals, from families’ representatives, or from organizations. There is increasing use of panels, which are groups of people (usually with some factor in common) that supply information. These may be one –time or ad hoc panels that are utilized for just one occasion. There are also more or less permanent panels that are used repeatedly, which tends to make the information more comparable over time. This also avoids the costs of recruiting new sets of people for each data gathering, valuable also for repeated measurements.
Location of the data sources is another option to be considered. Traditionally there have been three common types of locations:
- Where the subject lives (at home),
- Where the subject, their congregating in shopping malls has been a boon to gathering information.
- “Mail intercepts” in which people are importuned in shopping malls to participate in a study (typically in a small facility that a research agency maintains) have rapidly increased in usage.
As these several kinds of options suggest, today’s marketing researcher has a variety of source possibilities. This also must be stipulated in the data specification. Next we will discuss a different data decision: By what means of communication should it be obtained?
Communication Approaches: In deciding on which of the various means of communication means to choose, a researcher has much to consider. We introduce this subject with a simple division of communication media into two broad types: observation and questioning. Also we will consider the matters of whether to use structure and disguise. After that, each of the main media will be discussed and then their merits will be compared.
Through perceiving situations or actions, one can record and measure some descriptive facts. Observation includes both human means of perception and recording, which might be termed manual, and human means or mechanical. Manual is carried out by personal observers who see or hear the phenomena specified in the study.
Observation: A number of mechanical means of observation are in use today, some quite ingenious. Some have been around for many years, including attachments placed on television sets to record the time and channels when turned on to telecasts. Tape recorders, traffic counters, and photography have long been used in marketing studies. An example of the last is the eye camera. Set up in laboratory to record the gaze motion of the eyes of a subject who is looking at an advertisement.
A very notable development, in observation, stemmed from the adoption by food manufactures of the Universal Product Code (UPC) and its symbols (each one unique for a product) printed on labels. This was accompanied by electronic recognition devices to “read” or scan those codes at retail checkout counters. Projection of growth in scanners, by A.C. Nielsen Company, is that about 14,000 supermarkets will be using them when this book appears – all of them sources of quick and exact data on movement through supermarkets.”
Another important innovation, in observation methods, has been devise to “ cut into” cable television reception of selected homes. With this, a commercial under test may be substituted for the scheduled commercial being transmitted to the rest of the homes on that cable service. This created data for determining the tested commercial’s effects. One drawback has been that homes not hooked up to the cable were excluded. A more recent electronic creation enables commercials also to be cut into noncable households.
In this Nielsen service, both forms of scanning (of supermarket purchases at checkouts and of home viewing of commercials) are combined with a third observation measurement that uses a 2500-member consumer panel who present an identification card when they purchase at the supermarket checkouts. With this, their particular purchases can be specifically identified in the UPC and can be linked with the scanning of their TV viewing. Other variables in the sales environment, observed by the agency’s field personnel, are gathered that measure the promotional variables in each supermarket (e.g., displays, special pricing, and store advertising). These variables are reported along with the scanner information, enabling them to be analyzed for their effects on sales (for any segment of the consumer panel).
The Nielsen service was chosen as our example, but other agencies offer similar services. This illustrates the amazing innovations that are changing marketing research. Also it shows the extent of information and analysis that can be gained from observation alone.
Questioning: Attractive as observation may be as the medium for gathering data, it is very limited in its uses and its data In the majority of research problems, the required information can be gained only by asking for it. This is done mainly by interviewing, but also data may be obtained through self-administered questionnaires distributed by main and other ways. There are several means by which questioning can be carried out, which we will categorize as personal, telephone, and mail. Each will be discussed under separate headings.
