Quantitative Methods in Information Management - Case Study Example

Research Methods The following methods for conducting the survey have been suggested: A. Leave paper questionnaires near to the pay desks in the canteen with an invitation to anyone to complete one. B. Arrange for one of the canteen staff to conduct personal interviews with a random sample of employees in their workplaces. C. Send an email questionnaire to all staff (each member of staff has a work email address). D. Set up a display in the main entrance foyer of the central campus building, with questionnaires available on it for anyone passing to pick up. E. Send out a questionnaire with a prepaid reply envelope to a stratified sample of staff. The strata referred to in method E are the four council departments listed: Department Number of staff Education 1100 Business Studies 900 Chief Executive’s 320 Statistics 680 Your Name University/College Name January 14, 2008 Introduction The paper is a case study about research methods that can be applied in real time. The case is about a research to be conducted by a canteen head to understand the tastes and preferences of his customers and thereby increasing the customer base and improve the customer service. The various factors to be considered in determining a research design are discussed. The various methods of research design pertaining to the case study are discussed with their advantages and disadvantages. After an analysis of various research methods, a suggestion is given about selecting the research method to be stratified random sampling. According to this design, the methodology of arriving at the sample is also discussed in detail with formulae. The advantages of stratified random sampling over simple random sampling are discussed and finally, this method is suggested for carrying out the study. The aims of the survey are discussed and various reasons for selecting stratified random sampling method are discussed. The paper concludes by framing three open end questions and three closed end questions to fulfill the aim of the study. Target Population The definition of the population may be apparent from the management problem but there will always be confusion in determining the target population about whether the population consists of individuals, households, or families or a combination of these. In a communication study, the target population is person including individuals or households. In a observation study, the target population is non personal and the questionnaires are displayed at places where every one can access it. This is applied here in the Canteen survey. Good operation definitions are critical in choosing the relevant population. Researchers define the target population as all current existing members excluding their families. In the case, the target population is all currently enrolled students and employees on the main campus of NT University. The total population of the study includes 2500 staff and students. Research Bias: Selective Bias occurs when the researcher makes decisions on various aspects of research based on his reasoning. Due to selective perception, people analyze problems differently. Selective bias can be seen in the respondents, situation, the measurer and the data collection instrument. The selective bias leads to conclusions not pertaining to the study and there is a high level of deviation between the research and the interpretations. Response rate depicts the errors that occur due to the response made from the study. When the data reported differ from the actual data, response error occurs. Response error can occur during the interview or preparation of data for analysis. This is classified into participant initiated error, interviewer error etc. One study found that participants typically underestimated cash and other liquid assets by as much as 25 to 50 percent. These types of response errors will hugely effect the interpretation of results and make the research entirely obsolete. Consider the cases A, C and D, the advantages of these methods include collecting the response from all without any restrictions. There will be no interviewer bias as there is no interaction between individuals. The disadvantages include These methods include the selective bias occurs in collecting information only from the customers of the canteen and this will often lead to positive responses as these persons are already customers of canteen. So the views of others not coming to the canteen is ignored and this will result in result manipulations in the study. Here the situational factors also arise where the respondents fill the questionnaire together and there is a chance of a person influencing the other and thereby the response rate gets distorted. The instrument design will always raise issues. The participant initiated errors will be huge as the questionnaire is filled without any assistance. In Cases B and E, the advantages include the minimization of participant initiated error due to the presence of the interviewer. The errors that occur as a result of situation will get eliminated as the participant is provided full privacy to fill his response. The disadvantages include the interviewer error which includes failure to secure full participant cooperation, failure to consistently execute interview procedures, failure to establish appropriate interview environment, falsification of individual answers or whole interviews, inappropriate influencing behavior, failure to record answers accurately and completely, physical presence bias. Factors determining the research methodology The factors that are to be considered in determining the sampling method for this survey include: Cost: The economic advantages of taking a sample rather than a census are massive. The cost incurred for taking up various methods is to be analyzed and finally a method can be selected. The Industry reports state the following facts: Simple random sampling costs $25 per interview Geographic cluster sampling costs $20 per interview Self administered questionnaire costs $12 per respondent Telephone surveys costs $10 per respondent. Based on these criteria, the sampling method can be selected. Accuracy of results: Deming argues that “Sampling possesses the possibility of better interviewing, more thorough investigation of missing, wrong, or suspicious information, better supervision, and better processing with complete coverage”. The method selected must give the accurate results possible. Accuracy is the degree to which bias is absent from the sample. An accurate sample is one in which the under estimators and over estimators are balanced among the members of the sample. Precision: A sample must provide the estimate in a very precise manner eliminating the sampling errors. Precision is measured by the standard error of estimate, a type of standard deviation measurement; the smaller the standard error of estimate, the higher is the precision of the sample. Speed of data collection: This is a major factor to be considered while selecting a sampling method. The method must be able to collect the data at a faster rate so that decisions can be made at a rapid rate. Sampling’s speed of execution reduces the time between the recognition of a need for information and the availability of that information. In the case, a stratified random sampling can be adopted considering the cost perspective and the reliability, viability etc. Stratified random sampling Vs. Simple random sampling A stratified random sampling increases a sample’s statistical efficiency, provides adequate data for analyzing the sub population and enables different research methods and procedures to be used in different strata. Stratification is usually more efficient statistically than simple random sampling and is also useful when the researcher wants to study the characteristics of certain population subgroups. Researchers have several important variables and this method of stratification enables them to choose a lot of methods and variable of their choice. Simple random sampling is often impractical because it requires a population list that is often not available. It also fails to use all the information about a population, thus resulting in a design that may become waste and simple random sampling is very expensive to implement in both time and money. These are some of the improvements that stratified random sampling will have on simple random sampling. Stratified random sampling Methodology The following is the methodology to arrive at a total sample of 450. The steps in stratified random sampling include 1. Identify the total number of elements in the population. In the case it is 2500. 2. Identify the sampling ratio (k = Total population size divided by the size of the desired sample). In this case, k = (3000/450) = 6.67 (app). 3. Identify the random start. 4. Draw a sample by choosing every kth entry. The following concerns also must be included to avoid criticisms 1. Randomize the population before sampling. 2. Change the random start several times in the sampling process. 3. Replicate a selection of different samples. In the case, using stratified sampling, a sample of 450 is arrived as follows: Department Number of staff (B) Stratified sample = (B) / K Education 1100 165 Business Studies 900 135 Chief Executive’s 320 48 Statistics 680 102 Total 3000 450 (K=6.67) Recommended sampling method with reasons The sampling method to be used is stratified random sampling. The reasons include: 1. A stratified random sampling increases a sample’s statistical efficiency 2. It provides adequate data for analyzing the sub population 3. It enables different research methods and procedures to be used in different strata. Stratification is usually more efficient statistically than other methods and is also useful when the researcher wants to study the characteristics of certain population subgroups. Researchers have several important variables and this method of stratification enables them to choose a lot of methods and variable of their choice. A multi stage survey includes segregating various sub populations based on age, area etc. By conducting this survey, the response can be received from different categories. In this case, it will not be a cheap research as the questionnaire has to change for various categories to understand their tastes and preferences and this becomes a tedious work. It is unnecessary. This type would be more appropriate in a situation where the head of the canteen owns more than 3 canteens across the state. In this case, it is irrelevant to a certain extent. A stratified random sampling is more appropriate for the given scenario. Sample Questionnaire Improving service Measures to consider 1. Current customer service a. Cleanliness b. Time to deliver c. Quality of food Questions (Note: 1 – Very Poor 2.Poor. 3. Neutral 4. Good. 5. Excellent Encircle your choice) 1. Indicate the level of cleanliness in the canteen 1 2 3 4 5 2. Your level of satisfaction with the delivery time 1 2 3 4 5 3. How is the quality of food? 1 2 3 4 5 Extend the customer base Measures to consider 1. Word of Mouth 2. Awareness about promotions 3. Price Questions 1. Have you recommended the canteen to your friends Yes No If Yes, How Many? _________ If No, Comment_________________________________________________ 2. Have you availed the promotion offers in the canteen Yes No If No, Comment__________________________________________________ 3. Are you satisfied with the price charged by canteen Yes No If No, Comment__________________________________________________ Reference: 1. Cooper, Donald and Schindler, Pamela, Business Research Methods, Tata McGraw-Hill Edition, 2003. Pg. 32, Pg. 179, Pg 184-187, Pg 190-193, Pg 221-231. Read More