In 2008, there were about 26.2 million golfers in the United States, and of that number, women comprised 25 percent and represented one of the few growing segments in the long-stagnant golf market. Although women comprise a smaller percentage of all U.S. golfers, they purchase more than 50 percent of all golf products, excluding golf clubs, according to the Women’s Sports Foundation. This trend has led traditional golf brands to introduce women’s lines and open women’s-only golf stores around the country to cater to the needs of neglected female golfers
What TimeOut has learned is that with the passage of time women are becoming more and more serious about their golf game and wish more LPGA events were televised. Additionally, TimeOut discovered that women are extremely eager for new brands to hit the market, as traditional brands do not offer enough selection to meet their tastes. These women do not want to wear reformulated versions of men’s golf apparel nor do they want to scamper about the course in “cutesy” clothing, and finally, these women do not want to encounter other women wearing the same outfit. These ladies are hungry for more variety and are demanding it in the marketplace
Data obtained from panels not only provide information on market shares that are based on an extended period of time but also allow the researcher to examine changes in market share over time.’! As the following section explains, these changes cannot be determined from cross-sectional data
Relative Advantages and Disadvantages of Longitudinal and Cross-Sectional Designs
The relative advantages and disadvantages of longitudinal versus cross-sectional designs are summarized in Table 3.4. A major advantage of longitudinal design over the cross-sectional design is the ability to detect change at the individual level, i.e., for an individual respondent. This is possible because of repeated measurement of the same variables on the same sample.
Tables 3.5 and 3.6 demonstrate how cross-sectional data can mislead researchers about changes over time. The cross-sectional data reported in Table 3.5 reveal that purchases of Brands A, B, and C remain the same in time periods I and 2. In each survey, 20 percent of the respondents purchased Brand A; 30 percent, Brand B; and 50 percent, Brand C. The longitudinal data presented in Table 3.6 show that substantial change, in the form of brand-switching, occurred in the study period. For example, only 50 percent (100/200) of the respondents who purchased Brand A in period I also purchased it in period 2. The corresponding repeat purchase figures for Brands Band C are, respectively, 33.3 percent (100/300) and 55 percent (275/500). Hence, during this interval, Brand C experienced the greatest loyalty and Brand B the least. Table 3.6 provides valuable information on brand loyalty and brand switching. (Such a table is called a turnover table or a brand-switching matrix.12)
The main disadvantage of panels is that they may not be representative. Nonrepresentativeness may arise because of
1. Refusal to cooperate. Many individuals or households do Rot wish to be bothered with the panel operation and refuse to participate. Consumer panels requiring members to keep a record of purchases have a cooperation rate of 60 percent or less.
2. Mortality. Panel members who agree to participate may subsequently drop out because they move away or lose interest. Mortality rates can be as high as 20 percent per year. J 4
3. Payment. Payment may cause certain types of people to be attracted, making the group unrepresentative of the population .
Another dis advantage of panels is response bias. New panel members are often biased in their initial responses. They tend to increase the behavior being measured, such as food purchasing. This bias decreases as the respondent overcomes the novelty of being on 0Je panel, so it can be reduced by initially excluding the data of new members. Seasoned panel members may also give biased responses because they believe they are experts or want to look good or give the “right” answer. Bias also results. from boredom, fatigue, and incomplete diary or questionnaire entries.P
Microsoft: Experimenting with Usability
Microsoft performs meticulous usability research to enhance and develop its product portfolio in a way that is most beneficial to the customer. Usability research is aimed at increasing user comfort by making the product more intuitive to learn and remember Microsoft Usability Group is an important part of this effort. The group was conceived in 1988 to integrate user feedback into the design of the Microsoft development process and thereby into the ‘end products
In the Microsoft experiment.the causal (independent) was the Office suite. which was manipulated to have three levels: XP. 2003. and 2007. 1’be effect (dependent) variables were ease of use. capabilities, and the ability to enhance a computer user’s experience. The influence of other variables, such as user expertise and experience with Microsoft Office, had to be controlled. Although the preceding example distinguished causal research from other types of research, causal research should not be viewed in isolation. Rather, the exploratory, descriptive. and causal designs often complement each other
Visit and examiner sore of the recent projects conducted by Gallup. You will have to read through some of the reports posted on this Web site.
What type of exploratory research was conducted in these projects? Which methods were used? What type of descriptive research was conducted in these projects? Which methods were used? Did any project use an experimental design? If yes, identify the cause, effect, and control variables
Relationships Among Exploratory, Descriptive, and Causal F seorch
We have described exploratory, descriptive, and causal research as major classifications of research designs, but the distinctions among these classifications are not absolute. A given marketing research project may involve more than one type of research design and thus serve several purposes. Which combination of research designs should be employed depends on the nature of the problem. We offer the following general guidelines for choosing research designs:
1. When little is known about the problem situation, it is desirable to begin with exploratory research. Exploratory research is appropriate when the problem needs to be defined more precisely, alternative courses of action identified, research questions or hj potheses developed, and key variables isolated and classified as dependent or independent.
2. Exploratory research is the initial step in the overall research design framework. It should, in most instances, be followed by descriptive or causal research. For example, hypotheses developed via exploratory research should be statistically tested using descriptive or causal research. This was illustrated in the cause-related marketing example given in the “Overview” section. Exploratory research in the form of secondary data analysis and focus groups was conducted to identify the social causes that American businesses should be concerned about. Then a descriptive cross-sectional survey was undertaken to quantify the relative salience of these causes.
3. It is not necessary to begin every research design with exploratory research. It depends upon the precision with which the problem has been defined and the researcher’s degree of certainty
about the approach to the problem. A research design could well begin with dc-criptive or causal research. To illustrate, a consumer satisfaction survey that is conducted quarterly need not begin with or include an exploratory phase each quarter.
4. Although exploratory research is generally the initial step, it need not be. Exploratory research may follow descriptive or causal research. For example, descriptive or causal research results in findings that are hard for managers to interpret. Exploratory research may provide more insights to help understand these findings
The relationship among exploratory, descriptive, and causal research is further illustrated by the department store patronage project