Introduction
The Le Petit, a small French wine bar and restaurant in the village of Surrey has contacted us to study and provide a report for increasing their profitability, employee efficiency, and customer satisfaction. It is understood that the present owner is a successful businessman in his own right. But the contract agreement states that he has no experience in the restaurant and bar business. Hence he has asked us to give a report which will detail ways by which the business can be improved. This paper will provide a plan of action that needs a study of customer preferences and the attitude and mindset of the employees of the firm.
Sampling and data collection
When undertaking a study two main methods of data collection are reported to. Data can be collected from primary and secondary sources. “Data collection is simply how information is gathered. There are various methods of data collection such as personal interviewing, telephone, mail and the Internet”. (Data collection methodology 2005). In most studies, it is possible to collect data from people who are directly affected by the topic of the study. Such a collection of people are is called population for the study. “In theory, to generalize findings, evaluators must first define the population”. (Designing the sample or Population for data Collection 1993). For example, if the study as in this case is about customer needs and perception with regard to a bar and restaurant business, it is possible to ask people from the population about this. They can be asked directly, by telephone, by email, by post questions regarding what they think is a good restaurant and bar. These are people who actually go to restaurants and bars and have preferences and likes about an eating place. Data collected from such persons are known as primary data. If done properly accurate information and knowledge about the topic of the study can be got from such studies or surveys. In other cases, it may not be possible to collect such primary data. For example, a study about international oil prices will need extensive travel and also interviews with experts in the area. This is expensive and time-consuming and not feasible, especially for students. In such cases, data is collected from reliably published literature and other studies on this area. Such studies are based on secondary data. There are many methods of data collection which will be given in detail in the next section.
There is a step to be taken with regard to the collection of data especially if it is taken from primary sources. It is practically not possible to interview or survey every single person of the population especially if the population is quite large. If the population consists of all the students of a particular class, it is quite easy to survey each and every one of them. But in a larger population like the one intended here (those visiting restaurants and bars), this is practically not possible and also not required. In such a case a representative group from this population is selected using many accepted methods. Sampling refers to the process of choosing the unit from the available population like people, organizations…etc. using different methods and techniques. A sample is “a group of people or elements selected from the population being studied.” (Glossary: Sample. 2007).
The methods and techniques used in sampling are also given in detail after which the correct method for this study will be selected. This report will also provide details of how this study will proceed. The individual persons of the sample are referred to as sample units or as even simply as a unit. The one thing that a person has to be careful of is not to select a representative population and also not to let sampling errors come into the study. If the selected sample consists of a large number of people who do not visit restaurants and bars, the results will not be relevant to the study. Sampling errors occur if the following situation happens. Suppose every single person in the population had participated in the study and the study arrived at a particular conclusion. If the results of the study of the sample come out different from the one mentioned above, sampling error will exist.
Data collection methods
The results from the study and survey of the sample will result in possession of information or data with the researcher. The researcher has to measure and record the results from this data. This process is known as data collection. As mentioned earlier, there are many scientifically accepted method of data collection which is given here in detail. There are primary and secondary data collection methods. Secondary data can be collected from journals, demographic data, other studies, etc. Primary data can be collected through surveys, interviews (personal, email, and telephone, internet surveys, mail surveys, etc. (Data Collection Methods). The data collection method needs to consider face-to-face interviews and questionnaires in order to reach a final conclusion upon the efficacy and effectiveness of research on the subject.
For this purpose, it is necessary that suitable respondents are selected after identifying the sample population using lots, in order to avoid bias and discrimination.
In this study, being a qualitative study, necessary mathematical and statistical data need to be provided in order to defend, or deny the hypothesis question. Thus, it is necessary that data be collected, based upon the reactions of respondents during the course of the survey and personal interviews. The study would involve around fifty (50) respondents, in the field to whom the interviewers would confabulate and ask questions relevant to the study.
This would also be followed by Group discussions in order to assess whether the respondents are capable of working in groups or sub-groups, it is important that important research data be collected from group discussions also, in order to arrive at the right solution regarding the hypothesis and its validation.
It is needed to be seen that proper permission and consent is taken from concerned authorities regarding moral and ethical aspects of this study, and also, in terms of the areas in which this study would prove useful.
It is also necessary to determine its limitations, in that whether further studies would also become necessary for further advanced studies on this topic. All research studies have their limitations, and this would not be an exception. However, it is necessary to identify, assess and focus upon deficient areas, whether it impacts present and future studies.
Another important aspect of research would lie in terms of the fact, that in qualitative assessments, much of the research would consider how one could improve one’s bar business.
Whenever we consider the question of customer satisfaction, it would be better to put ourselves in the mind of the customer. It is very easy since we are also customers in our day-to-day life so that you can easily understand the problems faced by customers when they approached you. Now what we want here is to find out the ways to attract more new customers and to maintain the current customers. Besides, it is also required to ensure that the wine bar is operated as effectively and efficiently as possible.
As a matter of fact, the existing customers of yours may not be so satisfied as they seem to be. They do not complain just because they do not want to whenever they are requested to comment on your service, they simply appreciate it. Those remarks need not necessarily be true.
Some Facts
While going deep into the problem we come across certain surprising facts that need our urgent attention.
Most of the Customers do not Complain Directly
A study shows that only 8 percent of the customers complain directly.
Whereas 23 percent of them talk to the nearest employee. And the more surprising is that 69 percent of the customers suffer in absolute silence. So taking into consideration the number of complaints received or registered is a meaningless measure of customer satisfaction. Moreover, their silence before you does not mean that they will not complain to others. For them complaining is confrontational.
A Pyramid of Dissatisfaction
There was a research study back in 1999. It discovered that on average an unhappy customer will tell 10 people about their bad experience. In turn, these 10 people will each tell a further 5 people. That means a total of 50 people will have heard about it Customers very rarely complain to the provider Instead they will tell their friends who will, in turn, tell their friends and create a pyramid of dissatisfaction.
Collecting Honest Opinions
There is no problem without a solution. You need to ask your customer what they really think of you. And you should be prepared to take the rough with the smooth for that. Only those organizations that understand what makes their customers unhappy are able to put improvements in the future.
Sampling methods
The process of selecting the representative sample from a given population is called the sampling method. As data collection methods, there are many scientifically accepted methods for the sampling process also. The one assumption after the sample has been selected is that the selection will really be representative of the whole population. In other words, the sample should possess all the relevant characteristics of the full population. It is only common sense that the size of the sample will be directly proportional to the reliability of the data. If the sample size is too small, then sampling errors are likely to creep up into this study. Hence, apart from selecting the representative sample, the right sample size should also be selected. In this respect, the resources with reference to money, time, and manpower are relevant. This is because, if the researcher has only very little of the above, he will be forced to select a smaller sample. In most professionally conducted surveys there will be a large volume of resources and hence a bigger sample can be assembled. Resources will also depend on how much the company that needs the data is willing to spend. It is better to engage a company or agency that specializes in sampling and data collection rather than doing it by oneself. They have the expertise and the resources (apart from money) and hence will do a professional job. In short, sampling is a cost-effective way of collecting relevant information without wasting too many resources. At this time, it would be relevant to see how sampling is different from a census. In a census, the complete population is studied instead of selecting a sample. In short, each and every person in the population will be asked the same question with regard to a particular study. As is obvious, a census is extremely costly in terms of money spent, personnel required, and time needed to complete. In general, a census is done by governments to collect the population and other demographic data of the country. It is a well-planned and long-drawn-out process. Because of the resource factors mentioned above, a census will only be carried out very infrequently. The period in most censuses is once in ten years. The most commonly used methods of sampling are given below.
Simple random sampling
As the name indicates, this is the most simple and least complicated method of sampling. Here every member of the population has a chance of being selected as a sample. “A simple random sample gives each member of the population an equal chance of being chosen.” (Hunt and Tyrrell 2001). This is because the sample is selected at random without any preferences by the researcher. The samples are usually selected by using the lottery method. This particular study of the restaurant and bar business is proposed to be done using this method. The process will be explained in later sections of the paper.
Stratified sampling
This method is slightly more complicated than simple random sampling. Here the whole population is first categorized under some relevant parameter and each category is referred to as strata. After this is done a random sample is taken from each of these strata. Suppose a population is divided according to different age groups of say 20 to 39, 40 to 49, and 50 to 70. A random sample from each of these strata will be selected. The advantage of this method is that no category of the population will remain unrepresented. In the above example, there will be equal representation from each of the above age groups. This is advantageous over the previously mentioned methods because in a simple random sampling there will be a possibility that only a few and not all of the age groups be represented. “Stratified sampling is a commonly used probability method that is superior to random sampling because it reduces sampling error.” (Sampling Methods: Systematic Sampling. 2007).
Systematic sampling
Systematic sampling is also referred to as quasi-random sampling. This is still more complicated than the earlier two methods. The whole population is arranged in a particular order, say alphabetically. Then a number is assigned to each of the population. The whole population is then divided into a fixed number of groups. Then the Nth member from each group will be selected as the sample. An example will make the point clearer. Suppose the population number is 1000 and the sample to be selected is 100. The whole population with the number assigned can be divided into 100 groups. Then the fourth member of each group can be selected to form the sample. This method like stratified sampling will be more representative of the population. The problem here is that the population has to be very clear in order to assign those numbers. For example, the names of the members of the population will have to be known. “However, systematic sampling is less precise than stratified random sampling.” (Lillies 2004).
Cluster sampling
This method takes more time and effort when compared to the above mentioned methods. It is also called multi-stage sampling. The process is done in stages. “Cluster sampling is a sampling technique in which the entire population of interest is divided into groups, or clusters, and a random sample of these clusters is selected.” (Cluster Sampling). For example if the sample is to be selected from a big city. The whole city is divided according to some predefined criteria. It could be different areas of the city. Then some of the predefined areas are selected and the required sample is selected from these regions using the lottery method. A study for calculating the monthly family expenditure in a city can be done as follows. The city can be divided into areas, a few where wealthy people live, a few where middle class people live, and a few where poor people live. One or two areas from each of the above classifications can be used for selecting the sample. The selection can be done using simple random sampling.
Why sampling is done? The impracticality of studying the whole population due to time and other resource constraints is the main reason for selecting a sample. In most cases, the population is too large for a full study and hence the use of samples. A lot of savings in resources can be made and if the sampling is representative, then the results that are got will be the same as if the whole population was surveyed. Reduction of sampling errors and the selection of a representative population are the major factors to be considered.
Research design of this study for Le Petit
Le petit requires information in the following areas in order to maximize their profits, improve employee efficiency, and improve customer experience and satisfaction. Here, it is possible to conduct a primary data research by conducting a survey. This is because the population of the sample lives in the city, and can be identified with minimum effort. The population will consist of people who are in the habit of going to restaurants and bars. They will be in the age group of 20 to 70. This age group will be classified as 20 to 39, 40 to 59, and 60 to 70. As mentioned earlier it is proposed to conduct a stratified sampling so that representation from all the above age groups will exist in the sample. The whole process is expected to take a week. The company will employ ten persons for this purpose. The survey will be in the form of a questionnaire with closed questions. The team members will visit shopping malls, parks, and stand outside restaurants to select the sample. The first step will be to request a person whether he would be willing to take part in a survey if he or she is in the habit of visiting restaurants. The cutoff number will be those who visit at least once a month or more. If they answer in the negative then they are thanked and another person is asked. The negative can be with regard to lower number of visits (less than one per month) or unwillingness to take part in the survey. If they qualify and are willing to take part, they are asked the questions then and there and answers be marked by the concerned team member. It is expected that one person will need only ten to fifteen minutes for answering the whole questionnaire. The question of ethics is of paramount importance. The respondents will be assured of confidentiality of information before the questions are asked. Moreover only person appearing to be of sound mind will be part of the survey. Anyone who is mentally handicapped or is not in the right frame of mind will be avoided. This avoidance will depend on the observation of the team members. It is proposed that the sample size will be restricted to a hundred persons. The sample will consist of men and women in equal numbers. The same number of respondents will be selected from each strata (age classification) mentioned above. The team will cover approximately 20 persons each day. This could be time consuming as two eliminations have to be done. One is unwillingness to take part, and the other being not qualified to take part. If planned well, the part which relates to the questionnaires will be over in five days. The next day will be spent in classifying the data and arriving at a conclusion. One more day will be needed to prepare a report to the client.
With regard to the survey of the employees, the following procedure will be taken up. It should be noted that identification of the population is easy as each member is known by name and post occupied. A team of two persons will be sent to the Le Petit where each of the employees will be interviewed using a prepared questionnaire. Employees are free to talk and are not restricted by closed questions as in the case of the customer survey mentioned above. The employees will be classified as managerial and non-managerial staff. The management, as expected has given permission to interview the staff. The staff, in turn have agreed to cooperate fully. There are fifteen employees in the firm and two days will be needed to complete the interview. The interviews can be analyzed within two days and a report be prepared which will be presented along with the customer survey report to our client at the start of the next week. Employees will be assured full confidentiality and only their responses as a whole will be provided. No individual names will be used.
Conclusion
Le Petit has contacted our firm for conducting a survey with for the purpose of increasing profits and customer experience. Two separate studies are proposed, one using the general population and the other using the employees of the firm. It is our firm belief that the report generated out the data from the study will be of great benefit to the organization. If the suggestions are implemented properly, Le Petit can grow and become the favourite of the restaurant going public in the city.
Bibliography
Chapter 3 Designing the sample or Population for data Collection. (1993). Web.
Cluster Sampling. Metagova. Paris21. Web.
Data collection methodology: What is data collection. (2005). National Statistics. Web.
Glossary: Sample. (2007). Community Eye Health Journal, 20 (61), 17. Web.
HUNT, Neville., and TYRRELL, Sidney. (2001). Simple Random Sampling. Coventry University. Web.
LILLIES, Calla. (2004). Systematic Sampling. Grid Service Monitoring. Web.
Sampling Methods: Systematic Sampling. (2007). Stacpac. Web.