Decision making in marketing

Marketing research is the initial step and forms the foundation for effective decision making in marketing. Kotler (2003, p.115) quoted Herbert Baum, CEO of Hasbro Inc., who mentioned that market research is crucial to a corporation's marketing process, and marketing decisions should not be made without some form of research as this could waste an awful lot of time and money.

Marketing research helps in many ways, such as, in marketing strategic planning and forecasting, marketing problem solving, control and monitor the marketing process to make the correct marketing decisions for the organization. The scope of marketing research activities could involve advertising research in media, consumers' motivation preferences; product research in packaging, new product development, competitors' product studies, estimating the size of a market, market share analysis, pricing strategies, product distribution channels and promotional tactics and store audits. These decisions fall into four categories of major marketing variables, often called the 4Ps and referred to as the 'marketing mix' (as illustrated in Figure 1.0). The several benefits of marketing research will allow marketers to identify the actions they need to take in order to create positive impact of packaging and shelf-positioning to make their products more visible to consumers to enhance sales through impulse buying. In some cases, managers use research to determine the heavy users of their product within the target market.

New product development managers use marketing research most. From qualitative research techniques that generate product ideas to concept testing, product prototype testing, and then test marketing, marketing research is the key to creating a new product. Marketing research revealed that heavy users of Miracle Whip consume 550 servings, or 17 pounds, of the product a year. These data were used to target a $30 million promotional campaign to core users, telling them not to "skip the zip." As a result, Miracle Whip's market share went up 2.1 percent, to $305 million annual sales.

Marketing research also plays an important role in the distribution function. It is used to choose locations for new stores and to test consumer reactions to internal store design fixtures and features.

Hester (1996, p.93) mentioned as technological advances have assumed greater importance in lowering product costs, expanding product capabilities and enhancing organizational flexibility has become more important to know what technologies the market competition are using. Researching on competition can help improve competitive advantages and identify how business can survive in the present era of a constantly changing market.

Producers of goods and services, such as Procter and Gamble and American Airlines are frequent users of the research data. They use marketing research data in a variety of ways to support the marketing decision-making process such as on determining how various target groups will react to alternative marketing mixes, evaluating the ongoing success of operational marketing strategies, assessing changes in the uncontrollable external environment and how those changes affect their product or service strategy, identifying new target markets and measuring the quality of customer service and level of satisfaction.

Manufacturers are moving into strategic partnerships with their vendors in order to implement just-in-time manufacturing. These alliances are based on fully integrated manufacturer-supplier logistics systems that get the component to the assembly line just when it is needed. Large retailers, such as Wal-Mart, have such relationships with their major vendors. Within the framework of strategic partnering, marketing research information is fed back to a manufacturer's supplier when consumers express opinions on the manufacturer's customer satisfaction surveys. For example, if Pioneer was supplying radios for Honda automobiles and when customers complain about a certain model was difficult to program, the research information would be shared with Pioneer.

Most major franchisors of consumer goods and services provide marketing research data to their franchisees and a common way of gathering data is from mystery shoppers. Franchisors also share marketing research information with their franchisees to support certain recommendations or actions. For example, when McDonald and Burger King launch a new promotional campaign, they will share with franchisees on the research data showing that the selected campaign theme was preferred by customers.

Marketing research links the marketing variables with the consumers and the environment, so to speak, marketers use research techniques to understand the customers and markets and marketing effectiveness. The new paradigm shift in management is characterized by its customer-centric approach, especially in the development of Customer Relationship management (CRM). The major research techniques in use are in-store observation, in-house observation, other observation, focus group research, questionnaires and surveys, in depth interviewing techniques, marketing experiments, and mystery shopper research. In fact, mystery shopping is applied to assess a company's or competitor's marketing effectiveness rather than to understand customers' needs or wants. With the McKinsey's consumer-driven model, new competencies involve project management, ensuring the consumer's voice is heard, reinvent the wheel in knowledge development, activate the research on consumer insights, provide consultancy advice to senior management.

Within larger organizations, pricing decisions are usually made by committees composed of representatives from marketing, finance, production, and perhaps other departments. A recent marketing research study conducted by Ford Motor Company took a look at some of the new ideas generated by its engineers in order to determine first whether target customers were interested in a particular feature and then, if the concept had appeal, whether consumers would pay the suggested retail price. Twenty-eight new technology features were evaluated, the research was conducted in both the United States and the United Kingdom. It was reported that features that are inexpensive but offer high utility appeared to be desirable by customers, like the Sun Tracking Visor and the Infinite Door Check, and consumers were willing to pay the manufacturer's suggested retail price for some features but not for others.

There are many factors outside of our immediate control that can impact the effectiveness of our marketing activities. These can include weather, interest rates, government regulations and many others. Understanding the impact these factors have on our consumers can help marketers to design programs that can take advantage of these factors or minimize the risk of these factors.

From time to time, other individuals besides marketing managers and senior management find a need for marketing research. Manufacturing also receives continual feedback, from customer satisfaction surveys. Finance departments use test market data to forecast revenue streams for 1 to 3 years. Similarly, repositioning research helps financial managers forecast revenue spurts from older products. Human resource managers may call on marketing research to survey employees about a variety of topics. Quality customer service requires that employees have a positive image of the company, an attitude that they then convey to customers. Companies are increasingly turning to marketing research to win their cases in court, for example, Pfizer went to court over physicians' perceptions about various antihistamines versus sales messages used by sales reps about the products.

Example 1 from Mouncey & Wimmer (2007, pp. 107-130): Unilever Belgium carries brands such as Becel, Bertolli, Calv and Lipton, Iglo-Ola frozen food, Magnum ice-cream and Lever-Faberg personal care products, Coral, Dove, Axe, Rexona products. It took them five years to transform the role of the market research department from 'data provider' to 'leadership for growth' starting in 1999. The department started with a mission on ensuring consumer understanding, provide added-value service, being a partner for strategy and decision-making. Then they aligned and motivated the team towards the vision and brought this mission to life within the company, gave ownership to the team members; coached the team towards the 'consultant making impact' role, translated the vision and mission into clear plans. In 2002, Unilever took up the change and implemented the successful 'Consultant Making Impact' (CMI) model to reconnect with the consumers. The survey was undertaken in 2000, an 'Insight Activator' tool was used, with understanding of the consumer insight from the research findings, Unilever redefined the task, enlisted a strong team, reviewed existing knowledge, generated creative ideas, and resulted in great success for the following brands (as illustrated in Table 1.0):

Example 2 from Mouncey & Wimmer (2007, pp. 195-208): Honeywell improved its marketing research by combining online surveys, traditional interviewing techniques and web-based reporting to solve data collection error and satisfy Six-Sigma quantitative demonstrated quality standards. Honeywell's internal marketing division, known as Industrial Automation and Control (IAC) established a research-driven customer value process (CVP) since 1995. The CVP survey process included a variety of measurements such as customer value assessment, reporting survey findings to keep the management abreast of service defects, measure performance against competition and tracking ability of the organization to deliver to customer-defined attributes which affect customer's purchase and repurchase decisions among Honeywell's customer base. With the Six-Sigma's data-driven philosophy, IAC practiced the '6Ms' process (Man [people], Machine [equipment], Material, Method, Measurement and Mother Nature [environment]) to continuously improve the business process. The Six-Sigma exercise, however, clearly pointed out that the data gathering and reporting process at that time had many uncontrollable variables which created numerous hidden rework loops that adversely affected the efficiency and effectiveness of the process. In 2001, Honeywell revamped its CVP process by blending the research tools with web-based research, and also continued to conduct traditional methods of telephone interviewing, and reassess with second telephone interviewing. They adopted a humanvoice e-interviewing Surveyguardian to counteract the sources of potential error in online research, guaranteeing the statistical gains in data and process quality. IAC also built in a comprehensive system based on a hierarchical model of customer behavior by giving a choice of products from different companies for the customers to pick, linear regression techniques were applied to establish the impact or weight of each critical expectation on perceived performance. With the research findings, not only was performance measured, more importantly, it resulted in customer delight, increased the speed of Honeywell's communication process with customers, reduced the report delivery time of reliable data to management. Overall, this technique delivered improved customer satisfaction and increased ROI. Evidence of the progress was visible in financial growth with 46% of customer value benefits sustained and 22% cost factors were saved in 2002.

To conclude, marketing information collected serves to assist entrepreneurs to reduce business risks, spot current and upcoming problems in the current market, identify sales opportunities, and develop plans of actions. Success still involves the management to create an aspirational vision that announces the intended added value that marketing research will bring to the organization.

Marketing research may be conducted internally or purchased from external suppliers, Figure 2.0 illustrates the basic services provided by marketing research suppliers. Full service suppliers provide a full range of marketing research services from problem definition to report preparation and presentation. They often define the problem to start off with the research design, collect and analyze the data, then finally tender the written report. The services provided by the full service suppliers can be classified as syndicated, standardized or customized services. Limited service suppliers, on the other hand, focus their work in only one or a few phrases of the marketing research project and their services can be classified as data analytical services, coding and data entry, field services, or branded products (Kolb, 2008, p.64).

Wilson (2006, pp.13-14) has the same idea that full service suppliers are able to offer the entire range of qualitative and quantitative research approaches as well as capable of undertaking every stage of the research process for the buying client. Full service suppliers tend to be the larger research companies such as TNS (www.tns-global.com), BMRB (www.bmrb.co.uk), Ipsos (www.ipsos-uk.com) and Research International (www.resarch-int.com).

Burns & Bush (2010, p. 76) further elaborated that full services suppliers are typically larger firms that have the expertise as well as the necessary facilities to conduct research studies in their entities. ACNielsen, owned by VNU Inc., for example, earns its leading position by offering a diversity of services in the area of global media information, business publications, and the newspaper sector. IMS Health Inc., another giant in providing full services, offers market intelligence products and services, portfolio optimization capabilities, brand management services, and managed care and consumer health findings. The Kantar Group (TKG) operates around the world and has several research businesses, which give TKG the ability to conduct many different forms of research. Most of the research firms found in the Honmichl Global Top 25 and Honomichl Top 50 would qualify as full-service firms.

These full service companies are able to offer a wide range of expertise and services. They are in effect 'one-stop' research firms, however they may lack the depth of knowledge that has been acquired by specialized customer firms.

Limited-Service Supplier Firms, as mentioned earlier in the first paragraph, specialize in only one or a few stages of the marketing research project and some firms focus in one or a few activities. They may, for example, serve only one industry. Westat, for example, serves various government agencies, whereas IMS Health and DSS Research focus on healthcare. Wilson (2006, p.14) explained clearly that for example, a specialist supplier may only perform research in a specific market sector such as the automotive sector or children's products, or a geographic region such as the Middle East. Alternatively, the supplier may be a specialist in terms of the research techniques and may only do qualitative research or telephone research. Some suppliers may specialize in particular types of reporting approach, for example, certain suppliers may only focus on syndicated reporting services instead of carrying out a unique research project for a specific client, they research a market or product area and sell the resulting reports or data to a number of subscribing firms.

Some limited service firms use a single research technique or a special type of research technique, such as Shop 'n Check and Speedmark who engage only in mystery shopping program. The Pretesting Company uses devices such as the People Reader to measure how much time is spent reading an ad in a magazine or newspaper. Many new marketing research firms conduct research only via the Internet, for example, Greenfield Online (http://www.greenfieldonline.com), which conducts both quantitative and qualitative research via their internet-based marketing research panel.

Another Internet marketing research firm is Keynote (www.keynote.co.uk) which evaluates customers' online satisfaction and measures online behavior with a Web site. It works in the automotive, pharmaceutical, telecom, and other industries. Some cyber research firms, like ComScore and the competitor, Relevant Knowledge, focus on tracking the popularity of Web sites. Burns & Bush (2010 p. 78) gave some uncommon examples of firms that specialize in eye-testing or specific market segments, such as senior citizens or certain sports segments as in golf or tennis.

Problems and opportunities must be differentiated, although some of the same types of analysis are appropriate for dealing with either situation. A problem can be defined as anything that stands in the way of reaching an objective, whereas an opportunity is a chance to improve an overall performance. The failure of performing to meet the expectations of the firm's Mission/Objectives statement is a problem. When a firm faces a problem, it must have already identified objectives and have tried to accomplish them. Assuming that realistic objectives have been established, the management of the firm must first define the nature of the problem, then identify alternative courses of action and strategies, finally select a course of action to solve that problem. Researches on existing problems aim to find the roots of the problems, whereas researches on opportunity is involved in identifying the market situation for the use of entering into new markets or for ways to improve on current performance.

A change in any of the business operation normally requires a business unit to re-assess a firm's current strategy and capabilities to determine future strategy. Michael Porter's Five Forces Analysis model calls for the need of knowing what motivates the competitor, so by researching into a competitor's internal culture, value system, mindset and assumptions help in determining a more accurate and realistic understanding of a competitor's possible reactions. Company also needs to know its internal problem, and Porter's Five Forces model alerts firms to self analyze and rate the strength of each of the five forces so that they can work on the main competitive aspects. The five forces include 3 horizontal aspects of competition: 1) threat of established rivalry, 2) threat of substitutes when demand exceeds supply or vice versa, 3) threat of new entrants; and two forces from the vertical aspects of competition: 4) the bargaining power of suppliers and 5) the bargaining power of customers/buyers. It will be more useful to use Porter's five forces in conjunction with the SWOT analysis of the firm, which are Strengths, Weaknesses, Opportunities, and Threats. If we further study the six P.E.S.T.L.E. factors normally shared in project management operation, where P stands for political, E is economical, S is social, T represents technology, L for legal issues, E is environmental, the move into new market opportunities would be safer.

The reason for organizations to engage in marketing research is to identify and then find solutions for marketing problems. Research will help to identify problems which may actually be hidden but may arise in the near future, for example, situations on market share, company's branding, market potential forecasting and its characteristics, and business trends research. Then the findings of problem solving research will help decision-making in order to solve specific problems in marketing. Management should start off by stating the problem clearly which the firm must overcome in order to succeed, the problem could be underlying cause(s) of decline in sales/market share, or threat(s) to current or future success, therefore a SWOT analysis to find out the roots of the problem and how the situations could be rectified is important. Be prepared with plans to deal with any threats to its position, at the same time analyze the strengths and weaknesses of the competitors. Following the analysis, management should develop the strengths and resources of the company, ability to apply the current core competencies, and look for alternative solutions to tap into the potential market. Identifying the advantages and disadvantages of all elements is crucial to the success of the implementation stage. If the analysis job is conducted by analysts or marketers, recommendations for the best solution, how and when it can be implemented need to be constructed. Finally, the action plan should include a time frame and resources to work out all aspects of the marketing mix (Product, Place, Price, Promotion), in addition, the value, communication, people (staff, suppliers, partners) involved, systems and financial resources need to be tackled as well. Last of all, to achieve success and catch up with the fast changing market, action should be prompt.

Ocean Park (OP), opened in January 1977, is situated on the southern side of Hong Kong (HK) Island, covers more than 870,000 sq. metres of land, provided for free by the Hong Kong SAR Government. The Park is featured with world-class marine attractions, thrill rides, shows and panda habitat and aims to provide elements of entertainment, education and conservation to the public at an affordable price. OP has been one of the major tourist attraction points in HK, with the influx of tourists from China in the past decade, business had been safe and profitable; until 2003 when Disneyland announced its landing in HK, besides, new infrastructure will be built to facilitate smooth and efficient transportation to the park. The competitive forces in the macro environment caused a great impact to OP business and its future. Table 3.0 illustrates a brief SWOT analysis for OP at year 2004/05.

In view of the market threat and the problem of stagnant business, at no delay, Dr. Allan Zeman, newly appointed Chairman of OP at that time, announced a new Master Redevelopment Plan (MRP) in 2005, groundbreaking took place in November 2006, which signaled the start of this HK$5.5 billion project, and aim to have it completed in six years over eight phases until 2012 while the Park remains open throughout. MRP is a significant upgrade to the quality and availability of features at the Park, it includes redevelopment of the Waterfront brand new Aqua City which is four times the size of the existing Atoll Reef; a new Sea Jelly Spectacular with 1,000 sea jellies, the Headland will be transformed to be the Summit with Polar Adventure to feel the chill of the North and South Pole, and a journey down river rapids to discover the Rainforest, the Amazing Asian Animals themed-area will incorporate with the Giant Panda Adventure; a redeveloped children's area called Whiskers Harbor, a high-speed express train to transport visitors from the Waterfront to the Summit in less than 3 minutes. The helium balloon, SkyFair, will raise visitors 100 meters high up the air. The whole family can watch live shows of sea lions in the Whiskers Theatre and the nearby Dolphin University also offers visitors a chance to learn about the lives of dolphins. Hotels will be built as well, namely, the Spa Hotel, Ocean Hotel and Fisherman's Wharf Hotel.

Plans were achieved stage by stage, such as in 2004, Grand Prix motor ride was launched, 3 new cafes started operation in 2006 and most attractively was having 2 young pandas, Ying Ying and Le Le, joining An An and Jia Jia which facilitated the refurbishment of HKJC Giant Panda Habitat. Skyfair started to operate in 2007, Amazing Asian Animal center was launched in 2009 and in 2010, OP introduced express coach service for China mainland guests.

With the aggressive plan and commitments fulfilling, OP has earned more international recognition, such as, rankings by Forbes.com as one of the "10 Most Popular Amusement Parks in the World" in 2006. In 2007 recognized as one of the world's "50 Most Visited Tourist Attractions" in Forbes Traveler. OP gained the confidence of having local visitors because in 2008, 200,000 local students named OP as their favorite HK landmark in a survey by Yellow Bus Magazine. In 2008 ranked in the annual joint report of the Economics Research Associates (ERA) and Themed Entertainment Association (TEA) as the world's 16th, Asia's 5th and China's number one theme park. The Park has earned for its achievements in the field of animal conservation, and pioneered educational animal-oriented programs. This is what differentiates OP from its competitors.

The annual report in fiscal year 2005/06 revealed OP retained a competitive edge through its product differentiation and led to increased visitors with 4.38 million from 4.03 million visitors of preceding year. Thereafter, in 2006/07 with 4.92 million visitors, and 5.03 million visitors in 2007/08, and in 2008/09 with 4.83 million, down by 5% due to 2008 Olympic Games in Beijing and swine flu attack. OP's net current assets grew from HK$625.5M to HK$984.7M, and accumulated surplus grew from HK$620.5M to HK$791.8M (27.6% increase). By year 2008/09, its revenue was HK$243.7M, a 0.5% increase with a final surplus of HK$98.6M. OP realized its threat, analyzed its own problems, worked on its strengths, realized the opportunities and worked on its redevelopment scheme. This successfully made OP, a HK's own heritage, play an important role in HK's tourism industry.

Example 2 from Mouncey & Wimmer (2007, pp. 601-621): Market research has significantly improved the reformation of the Australian government public sector, especially in the Child Support Agency. Since the mid 1970s, all levels of government in Australia went through a 6-movements reform, which involved downsizing, Total Quality Management (TQM), re-engineering, reinvention, de-layering, and virtual integration, however these changes became a source of frustration for public sector employees. Until 1993, Australian government adopted market research techniques to re-model the structure of service delivery in the public sector and The Child Support Agency (CSA) is a significant example of utilizing the marketing research to upgrade its services. Founded in 1988, CSA administers the Australian Government's Child Support scheme, which aims to ensure that parents who are separated share the cost of supporting their children and CSA's role is to register, assess, collect and transfer the child support payments. In year 2002/03, CSA had 1.4 million clients and supported transfer of child support fund from separated parents to over 1 million children. CSA realized the parents were becoming increasingly dependent on them, this led to their thought of the necessity to increase client satisfaction and reduce costs while maintaining an excellent child support collection rate.

A major research program established the key factors that influenced voluntary payment of child support, such as relationship quality with the other parent, relationship and extent of contact with children, financial affordability of paying child support, and procedural support provided to parents from CSA. These research findings highlighted a need for CSA to strategize their operational change with the intent of maximizing parents' independence. CSA identified 6 challenges, which were influencing a combined 'one government' approach, build a community focus, encourage and support parental responsibility, refine agency service delivery, increase organizational capacity and productiveness, guarantee a transparent and accountable service.

The result was astounding, 51.1% of CSA clients were in the private collect category in 2004, as compared to the 1996 rate of 39%, reduced parent's dependence on CSA for the collection of child support and raised parents' child support responsibilities, while still providing safety net arrangements where private arrangements failed. This approach helped to maintain cooperative relationships between parents and led to increased voluntary acceptance of paying child support. Today, the external research program still includes research into client compliance, client satisfaction, services that support client independence and staff satisfaction. CSA allows staff to determine the defining characteristics of high-performers. The CSA's research which included both quantitative and qualitative research methodologies, formed the basis of segmenting clients into various groups, hence CSA worked on stopping the problems grow, eventually saved cost and time and assisted on parents to develop responsibility sense.

To conclude, in situations to realize problems, analyst should develop a complete database on the environmental factors, market factors, competition, revenue, cost and ROI. Decisions must be made on which opportunities, if any, will be sought. As new opportunities are identified, the analysis process could become standard procedure and a prerequisite to decision making with respect to stretching the firm's market share.

The process of measurement is a fundamental aspect of marketing research. It involves the use of numbers to represent the marketing phenomena under investigation. Measurement is defined as the assignment of numbers to characteristics of objects or events according to rules and errors can occur when the characteristics of the number system do not represent the relationships in the marketing phenomena being measured. There are four characteristics of the number system, namely equality of numerals, order of numerals, equality of differences, and equality of ratios, which all correspond to the four scales of measurement, that is, nominal, ordinal, interval, and ratio. A nominal scale consists of identifying and categorizing, with no implications of 'more or less'. An ordinal scale involves the determination of more or less but with no indication as to the distance or interval. An interval scale involves the determination of distance, while a ratio scale involves the additional characteristic of an absolute zero point. Each of these four types of measurement is important in marketing research. As we move from nominal to ratio measurement, the rules for assigning numerals to the marketing phenomena become increasingly more restrictive, affecting the statistical techniques for data analysis to be more detailed.

Generally, measurements in research possess some degree of error in that the numerical scale does not exactly represent the marketing phenomenon. The ways to classify these potential sources of error includes personal factors, such as mood, fatigue, health, personal preference etc. of the respondent which may influence the measurement; variations in the environment, variations in how the questions are administered and the influence of the interviewing method, such as telephone, personal or mail, the degree of ambiguity, difficulty of the questions and the ability of the respondent to answer them, last of all, errors made in the coding and tabulation process. So to speak, the total error of measurement consists of systematic error and random error.

Systematic error causes a constant bias in the measurements and a valid measure is one that is truthful, to illustrate the difference from reliability, assuming that we are measuring the income status of respondents that may affect their buying motive, however all respondents over-report their income by choosing a higher category of $10,000, and if all respondents make the same choice, the mean income of the respondents would be exactly $10,000 more than the mean of the true values. Although it is untrue and invalid, the data result is consistent and is deemed reliable. This upward bias in the measurement is a random error but is not systematic.

The reliability of a measure relates to the extent to which the measurement process is free from random errors. According to Silverman (2006), reliability is a standard for evaluating tests or measures that refer to the consistency, stability, and predictability of the research findings. The validity of a measure relates to the extent that the measurement process is free from both systematic and random error. Validity is concerned with the question: are we measuring what we think we are measuring? While reliability refers to consistency of the measuring device, validity refers to accuracy, and accurate estimations of what is being assessed. Validity is a broader and more difficult issue than reliability.

An easy measure of validity would be applying the test-retest method by measuring something at two different times and compare the scores so if the test is reliable, those who scored high the first time will also score high the second time, and vice versa, this correlation is called the coefficient of stability. Assuming that a sales manager believes there is a relationship between job performance and extrovert personality, then those who have high job satisfaction and extrovert personalities should exhibit high job performance scores; if they do not, we should question the construct validity of measures or question the validity of the hypothesized relationship. This method is challenged as situational or personal factors may have changed and impact the response in the second trial.

The second method is on content validity which involves a subjective judgment by an expert as to the appropriateness of the measurement, assuming that we are going to measure the image of retail stores in a grocery chain, a 20-items questionnaires is proposed, the content validity of these 20 items would be determined by having an expert assessing the appropriateness to measure store image, however if store safety and cleanliness, friendly atmosphere were excluded, the content validity could be challenged. The third method is on concurrent validity which involves correlating two different measurements of the same marketing phenomenon which have been administered at the same point in time. It is used to determine the validity of new measuring techniques by correlating with previously established techniques, assuming that our previous 20-item measurement technique of store image is valid and also assume that an alternative or shorter technique is used, concurrent validation would involve administering both techniques under identical conditions and correlating the two measures, so a high correlation would establish the concurrent validity of the new technique.

The fourth method is by using split-half reliability which involves dividing a multi-item measurement device into equivalent groups and correlating the item responses to estimate reliability, if a high correlation coefficient would mean that the items are measuring the same characteristic. Overall, reliability is a weaker concept than validity since it involves only random error. Consequently, reliability is easier to measure than validity and is reported more frequently in studies.

The measurement task in marketing is typically more difficult and the problem lies in the measurement of the behavior of people. This often involves measuring concepts or constructs believed to exist in people's minds. For instance, respondents may consider their beliefs about the trend in the direction of stock market are correct. It would be very difficult to say which beliefs are relevant to judging the accuracy of the respondents' answers and we do not have direct access to those beliefs anyway. Also there is possibility of variations in what participants tell they do and what they have actually done. This argument may be specifically true when asking people questions on a subject, but the answer could already have been affected through indirect societal influence. Therefore, respondents' attitudes, beliefs, perceptions, voices of opinions, habits and interests seem to be unstable; such people's inconsistency in responses will make it very hard to generate highly reliable measurement. The issues of measuring validity are thus more complex.

Another approach to assessing validity is to compare the answers of groups of respondents who are expected to differ on the underlying construct of interest, e.g., in Hong Kong, we expect Professional rep parties to be more conservative than the Liberal parties. Thus, we would expect registered professional parties to score higher, on average, than the liberal parties on an item designed to measure conservatism. Such evaluations depend critically on our theory about the relationships among certain variables. If, in the above example, professional parties are not different from liberal parties, it could mean either that the question was a poor measure of conservatism or that professional parties and liberal parties do not, in fact, differ on conservatism. In general, it is difficult or impossible to distinguish poor measurement from inaccurate theories in the assessment of validity.

Assuming we are conducting a survey of buyers to estimate the market share for brand X. For purposes of illustration, assume the market share is actually 10%. Let us also assume that four potential conditions exist regarding the influence of systematic and random error on our observed measure of market share, as follows:

In condition a) the expected value of the distribution of sample means is identical to the true market share of 10%, the low random error is reflected in a tight distribution of sample means. Here, repeated samples consistently produce means that are close to the true market share. In this condition, our survey results would be described as both valid and reliable. Condition b) has the same tight distribution of sample means, but the influence of the high systematic error has biased the expected value 3 points above the true share of market. Repeated sampling would produce means which are close to the 13% result. In this situation, our survey results would be reliable but not valid. In condition c), the expected value of the distribution of sample means is identical to the true market share of 10%, but the high random error causes the distribution of sample means to be highly dispersed. With the repeated sampling, one would find many sample means dramatically different from the true market share, hence this condition is described as neither valid or reliable. Finally condition d) has the same high random error but the systematic error is also high, causing the expected value of the sampling distribution to be 5 points below the true market share, hence this condition is neither valid nor reliable.

So to conclude, for a measure to be valid, it must be reliable, in other words, a measure cannot be valid if it is not reliable, and if it is reliable, then it may or may not be valid. Reliability is a necessary but not a sufficient condition for validity. Consequently, the validity of a measure is of main concern since it deals with both systematic and random error while reliability is a weaker concept since it involves only random error.

Please be aware that the free essay that you were just reading was not written by us. This essay, and all of the others available to view on the website, were provided to us by students in exchange for services that we offer. This relationship helps our students to get an even better deal while also contributing to the biggest free essay resource in the UK!