Systematic Random Sampling - Samuels Greek Restaurant - Research Proposal Example

Introduction Market research is a valuable tool in any business. The information that it can provide helps businesses in various aspects and in various ways. Primarily, market research helps organizations understand the trends in the market most especially their target markets. However, conducting market research is not easy. There are many choices and considerations that must be made in order to ensure that the research will produce significant and usable information. Two of the most critical elements in any research are data collection and sampling. Both of the said processes have direct influences on the outcome of the research. Furthermore, data collection and sampling are both concerned with the elements to be studied, the population, and what data will be gathered from them. As such, their role in market research is very significant. Careful thought and consideration must be given to the choice of data collection method and sampling approach to be used in any research. Since there are several methods and approaches available, the researcher must analyze the scenario to ensure that the right method and approach will be used. Each situation may call for a different data collection method and sampling approach. Such choice relies on several factors. Primarily, the researcher must consider the problem at hand. He must determine whether the problem requires qualitative or quantitative research. This is important because the researcher must determine what type of data is involved or required. Once the type of research has been established, the researcher must then take into consideration other elements such as the population, budget, time allotted, and the number of personnel at hand. Basically, this means that the researcher should study the resources he has at his disposal for all of these are factors that help determine the most appropriate data collection method and sampling approach. This paper will attempt to highlight the process of choosing the appropriate data collection method and sampling approach for a given situation. In particular, this paper will shows what data collection method and sampling approach must be used for the market research to be conducted for Samouel's Greek Restaurant. Such research involves the following problems: Are employees being managed to maximize their productivity as well as commitment to the success of the restaurant' What are the best approaches to attract new customers, and to keep and grow existing customers' To arrive at the appropriate choices, the alternatives for both data collection method and sampling approach will be discussed. This discussion will include an analysis of the advantages and disadvantages of each method and approach as well why it is appropriate or not appropriate for the given situation. In essence, this paper will provide insight as to how to choose the right data collection method and sampling approach. Data Collection Method There are several data collection methods that a researcher may choose to employ. Also, the researcher may choose to combine some of these methods in order to gather the necessary data more effectively. The basic methods that may be used singularly or in combination to achieve specific research objectives are as follows: Internet surveys / Online Data Collection Mail surveys Interviews Real-time tracking Store checks and mystery shops Internet surveys/online data collection Conducting market research online allows the researcher to gather data in a simple, easy, and effective manner. With the internet, one can create interactive electronic surveys for specific markets or groups. Also, one can create surveys that reach wide market bases through the use of the Worldwide Web. Internet surveys are the most cost effective way to gather data internationally. Using various programming techniques, one can easily randomize data gathering. It is most effective in gathering data from a large segment of the population. However, there is a tendency for internet surveys to be non-representative of the entire population. This is due to the fact that when a researcher solely uses internet surveys, he fails to recognize that there are members of the population that do not have access to the internet. This then leads to the lack of representatives of the data he gathers. (Curasi, 2001) Using internet surveys to conduct market research is most effective for companies whose target market is composed of cyber-savvy individuals. For instance, online surveys are the primary way for United Parcel Systems to gather data regarding the market since 40 to 50 percent of its customers are internet users (James, 2000). Internet surveys should be used in conjunction with other types of surveys to ensure that the data gathered will truly be representative of the whole population. Mail Surveys Surveys are the most common method of gathering data for research. Even before the dawn of the internet, market researchers have relied on surveys to provide them with the data they need. Surveys are conducted in several ways. The researcher may opt to send the questionnaire through mail, conduct the survey through the phone, or hand out the questionnaires personally. Each manner of distribution has its corresponding pros and cons. Mail surveys are usually the least costly method to gather data. Also, using mail surveys allows the researcher to include visuals to help explain certain concepts. However, mail surveys are known to have one of the lowest response rates. According to Gartner (2002), "Mail questionnaires are sent back by only one out of five recipients." This in turn can lead to biased responses since the respondents can be largely self-selected. Mail surveys also take longer to complete than other methods. More importantly though, mail surveys have data limitations since open-ended questions are difficult to ask and incomplete answers are hard if at all possible to clarify. An alternative form of survey is the phone survey. One advantage of the phone survey is that it takes less time to accomplish than mail surveys. Also, phone surveys allow the researcher to clarify certain incomplete or obscure answers. However, the major drawback of the telephone survey is that there maybe problems with regards to the level of comprehension between the respondent and the researcher. According to Byrne (2002), "Sometimes the telephone discussion leads to problems of mutual comprehension between the interviewer and the respondent, particularly where one is less fluent than the other in the language being used. Telephone surveys can only really employ the simpler type of questions, particularly multiple choice or yes/no questions." Interviews Interviews are used just as commonly as surveys. More importantly, interviews are the primary way by which qualitative data is gathered since such type of data collection gives the researcher more flexibility to ask more complex questions. Surveys are most effective for gathering quantitative data since it allows the researcher to easily determine the number of people for or against a certain option or alternative. On the other hand, interviews allow the researcher to dig in deeper and know more in-depth information coming from the interviewee. Also, interviews allow the interviewer to establish rapport with the respondent eliciting cooperation and encouraging candour. However, interviews are very costly to conduct and take long to finish. This is especially true when one considers face-to-face interviews. One can opt to conduct interviews over the phone. This is a more cost efficient form of interview. Furthermore, it takes less time to complete phone interviews rather than conducting face-to-face interviews. However, one of the biggest disadvantages of phone interviews is that multiple barriers to cooperation exist. Gartner (2002) points out that, as more telemarketing and phone surveys are conducted; respondent cooperation is sometimes a problem. New technologies such as caller ID may increase these barriers. Interviews are a good way to gather qualitative data. By carefully choosing the respondents as well as the questions to be asked, the researcher can ensure that he gathers useful and pertinent data that is representative of the population under study. Real Time Tracking Real time tracking is used to measure various performance factors. This includes product satisfaction, customer service satisfaction, and product awareness. One may choose to measure such elements using various intervals, whether daily, weekly, or monthly. One advantage of real time tracking is that it provides up-to-the-minute feedback. For instance, the researcher can speak to the customer about recent experiences and gather responses that are fresh and detailed since the event is still fresh on the customer's mind. Data gathering is also very fast which allows quick analysis and decision-making. However, using real time tracking has several disadvantages as well. Gartner (2002) lists down the following disadvantages: Organization needed - In order to best take advantage of an ongoing monitor, lists of potential respondents need to be transmitted on a daily or weekly basis-consistently and accurately. Often this will require dedicating internal IS resources during the start of a project. Long start-up time - Real-time tracking studies often take longer to set-up than normal studies due to the high level of coordination required between the client and research vendor. Stable questionnaire - The questionnaire must be kept consistent across waves in order to make comparisons. The key issues must be identified before research begins. The use of real time tracking allows the researcher to gather up-to-date data. This maybe very useful when one is trying to measure performance of individuals or of products. Also, using real time tracking may be very helpful in helping an organization improve certain areas quickly based on customer feedback or performance measurements. Store Checks and Mystery Shops Store checks are usually used to determine how the staff functions on any regular or ordinary day. Mystery shoppers are employed to simulate the customer-supplier interaction to determine how an employee acts towards clients and customers. Both of these methods allow the researcher to identify the areas wherein employees may have strengths or weaknesses particularly when dealing with customers. The biggest advantage of such method is that it allows the researcher to view the business from the standpoint of a customer. This is a very unique perspective that will provide very useful data especially in cases wherein a research aims to improve customer relations for a company. Also, mystery shoppers maybe utilized to motivate and compensate superior performers. Lastly, using such method is very reliable especially if done regularly since it will provide a data regarding the typical situation and condition. The downside to using such method is that the quality of the data is highly dependent on the evaluator. This means that the bias of the evaluator may affect the data he gathers. Also, when not performed regularly, mystery shopping and store checks may create a false perception of reality which may serve as the wrong basis of judging the situation. Sampling Approach There are several sampling methods that a researcher may choose to employ for his research. However, one must remember that the goal of any method of sampling is to come up with a sample that is representative of the entire population in question. What this means is that the data gathered from the sample should be as similar as possible to the data that would be gathered from the entire population. Just as with data collection, there are likewise several factors that must be considered in choosing the sampling approach to be used. Also, a researcher must undergo several steps in order to identify the sampling approach suited for his study. In selecting a sampling approach, the researcher should have some knowledge of the target population. Aside from knowing who the target population is, he must likewise have estimates, if not actual figures, of the size of the entire target population. Using such knowledge, he can then may certain assumptions and use these to come up with estimates regarding the sample size required to achieve a reasonable estimate of population characteristics. Deciding on what sampling method to be used is basically the same process as deciding what sampling units will be chosen. The various methods of sampling are broken down into two major classifications, probability sampling and non-probability sampling. Probability sampling methods are basically methods wherein units are selected in such a way that each has a definable probability of being included. Birchall (2005) points out that, "Probability or random sampling gives all members of the population a known chance of being selected for inclusion in the sample and this does not depend upon previous events in the selection process." Basically, this means that when an individual is selected, his selection should not affect the chance of other individuals from being selected as well. Non-probability sampling on the other hand uses other criteria in selecting a sample. These may include voluntary participation or accessibility. However, using non-probability sampling, although unavoidable in some cases, always tends to lead to biases in the sample chosen (Birchall, 2005). Using non-probability sampling methods requires more thought in coming up with conclusions as these may not be an objective representation of the entire population. Probability Sampling Approach Probability sampling is the most ideal approach to be used in any research. There are four basic types of probability sampling. Simple random sampling Simple random sampling is a self-weighting sampling design. This means that all elements in a population have an equal probability of being selected to be part of the sample. There are several ways to accomplish such. However, the most basic way is to assign each member of the population a number, sort the data set according to the random numbers, and select the required sample size from any sequential part of the population, normally beginning from the first element and continuing until the desired sample size is reached. Systematic Random Sampling In systematic sampling, the elements of the population are arranged and every rth element is selected over the whole sampling frame. Usually, the researcher selects the first element and then moves on to the next for every interval r. In certain cases, the population may be arranged in a certain order that can be used for implicit stratification. For instance, a list of students may already be arranged alphabetically and the researcher may just pick every rth student. However, it must be noted that the researcher must employ caution in using such sampling procedure. According to Foreman (1991), "If the elements are in a random order, systematic sampling practically matches simple random sampling. However, if the population contains some (hidden) order or sequence, systematic sampling may yield a sample that consists of very similar elements that do not reflect the true population variation and thus the sample can lead to erroneous or biased results." In certain cases, proper arranging of the population can yield better samples than the use of a random order or the use of simple random sampling. Stratified Sampling "Stratified sampling is a variant on simple random and systematic methods and is used when there are a number of distinct subgroups, within each of which it is required that there is full representation" (Birchall, 2005). To construct a stratified sample, the population is divided into sub-populations otherwise known as strata. Such classifications are based on certain characteristics. Usually these include age, gender, and socio-economic status. From each stratum, elements are chosen using either random or systematic sampling. Cluster or Multi-Stage Sampling Cluster sampling is a form of random sampling that is more commonly used and in most cases, more practical. More importantly, it is used in certain situation wherein the researcher does not have a list of elements of the target population. Since such type of sampling is conducted by randomly selecting subgroups of the population, with the option to use a number of stages, it should produce results equivalent to a simple random sample (Birchall, 2005). Cluster sampling is used in the following situations: No list of the population exists. Well-defined clusters, which will often be geographic areas, exist. A reasonable estimate of the number of elements in each level of clustering can be made. Often the total sample size must be fairly large to enable cluster sampling to be used effectively. Conclusions and Recommendations Having discussed the possible alternatives to both data collection and sampling, the appropriate choice for the situation at hand may now be discussed. As previously mentioned, several factors are to be considered in deciding what method and type of data collection and sampling to use. Therefore, in order to identify the best choice, the situation at hand must be analyzed so as to identify the critical factors and considerations. With the case in question, the problems to be answered both require qualitative data. Moreover, since the research involves two different problems, the researcher must consider using different data collection techniques and sampling approach for each. There is no relation between the two problems. Therefore, they should be treated as two distinct and separate researches. The first problem concerning the way the employees are managed requires input from both the manager and the employees. Since the research aims to determine the level of productivity of the employees, the researcher must also consider gather data regarding the employees' performance. For this problem, it is recommended that the researcher conduct face-to-face interviews with all the employees of the restaurant. Since all the respondents are in one place, it will not be too difficult or expensive to do conduct such interviews. In the interview, the researcher should ask the employees the same setoff questions. This should include questions regarding their satisfaction with their job. It must be noted that job satisfaction is directly related to the level of productivity of employees as pointed out in several studies. Such interviews should likewise be coupled with the use of real time tracking as well as store checks or mystery shoppers. The former will allow the researcher to gather data regarding the punctuality of the employees as well as the manager. Also, the researcher can look to the number of complaints raised by customers with regards to the staff. Both types of data can be gathered by using real time tracking. Store checks on the other hand will allow the researcher to observe the conduct of the staff as well as that of the manager. In order to ensure that the researcher will gather reliable data, he must make visits every week for one month. Also, he must ensure that these visits are unannounced in order to ensure that the employees will act normally. With regards to the sampling approach for the first problem, no sampling is involved since the researcher will interview all employees and staff members of Samouel's restaurant. For the second problem, the researcher must gather several types of data. For one, he must determine the level of awareness regarding Samouel's Restaurant among consumers. Since the target market of the said establishment is the local area, the researcher should gather data regarding people's perception of Samouel's Restaurant. Also, the second problem requires data regarding the ways to attract more customers. The recommended data collection method for this problem is the mail survey. The questions should include those that ask the respondents regarding their awareness of the existence of Samouel's restaurant and their preferences with regards to Greek Food. Such information will help the restaurant's management in improving their food so as to meet the taste of their target market more. However, since mail surveys tend to have low response rates, it should be combined with interviews wherein the interviewers will be sent out around town to interview people. It is also recommended that the mail survey include flyers that will inform the respondents of what Samouel's restaurant has to offer. Also, it should include a discount offer. This will ensure that even if the respondent fails to return the questionnaire, he may be encouraged to visit the establishment since he was given a discount. Such discount offer should also be given to those who take part in the interview. In effect, the survey and the interview have served a double purpose. They gathered the necessary data and at the same time, they have spread awareness regarding the establishment. The suggested sampling method to be used is systematic random sampling. The researcher may use a list of known residents of the town, such as the voter's list, and pick every 10th element from such list. Since there maybe non-voters in the town, their non-representation will be addressed by the use of the interview. In the interview, the interviewer may pick every 10th person to pass his way to interview. In essence, this method is also systematic random sampling. Reference: Blankenship, A., Breen, G., Dutka, A. (1998). State of the art marketing research. New York: McGraw Hill Professional. Birchall, J.(2005). Sampling and samples. Market Research World. Retrieved 5 July 2006 from: http://www.marketresearchworld.net/index.php'option=com_content&task=view&id=23&Itemid=1&limit=1&limitstart=0. Byrne, M. (2002). Interviewing as a data collection method. British Market Research Association. Retrieved 5 July 2006 from BMRA website: http://www.bmra.org.uk/researchers-toolkit/ToolKitcontent.asp'category=3&subcategory=51&article=394&top=MRTech. Curasi, C. (2001). "Consumer research: Big brother is watching you." International Journal of Market Research. Retrieved 5 July 2006 from BMRA website: http://www.bmra.org.uk/documents/177.doc. Dillman, D. A. 2000. Mail and Internet Surveys: The Total Design Method. New York: John Wiley & Sons. Erreaut, G., Callingham, M. & Imms, M. (2002). Qualitative market research: Principles and Practice. California: Sage Publications. Foreman, E. K. 1991. Survey Sampling Principles. New York: Marcel Dekker, Inc. Gartner, A. (2002). Basic data collection methods. British Market Research Association. Retrieved 5 July 2006 from BMRA website: http://www.bmra.org.uk/researchers-toolkit/ToolKitcontent.asp'category=3&subcategory=51&article=390&top=MRTech. Hague, P. (2002).Market Research: A Guide to Planning, Methodology and Evaluation. 3rd ed. Virginia: Kogan Page Limited. James, D. (2000) "The future of online research." Marketing News, 3 Jan., pp. 1 and 11. Smith, S. & Albaum, G. (2004). Fundamentals of marketing research. California: Sage Publications. Read More

 

Read More