Table of Contents

Executive summary

The model used in the SERVQUAL states that there are three variables to measure service businesses: service quality, product quality and price/quality ratio. Another study done by Nassrin (2009) has found that there is a difference in the level of satisfaction of students from different academic levels about university restaurants. We used these theories in our hypothesis and variables formulation and we found that that the level of satisfaction of students toward the restaurants differs between students of different academic levels. We proposed some recommendations for the managers to enhance the satisfaction level of students. We recommended them to do a qualitative research to know the specific needs of each of each of the academic level, and then use these findings to reshape the marketing and product mix of the Newrest restaurant.

Introduction

As the population is growing at a high and fast rate, and the world is recognizing a rapid-changing environment, the issue of customer satisfaction is becoming more and more critical as well as crucial while launching products into the market; this is done in order to satisfy customers’ needs and wants, and hence meet their requirements. In fact, with the adoption of new technologies and the increased awareness of consumers about the products and services offered in the market, the competition in the business market is becoming more and tougher. For that purpose, companies strive to capture target customers by differentiating their products and services from competition, and hence gain customers satisfaction. By focusing on this issue, companies recognize that customer satisfaction is, indeed, a key and a best leader to success throughout the world. After opening the doors to students in 1995, Al Akhawayn University had launched all its food products, but had never conducted a satisfaction survey or any kind of research in order to assess the satisfaction level of students towards its “Newrest” restaurant. Not to forget, the University’s management had never conduct a study to recognize how the students’ needs differ from a group to another. Consequently, failing to defining the students’ needs and preferences would lead to producing products that would not meet the students’ expectations. Hence, this would engender students’ dissatisfaction, that would in turn, lead to switching to other substitute products, such as food at Sodexho restaurant or going outside to downtown restaurants. Thus, it is as important as crucial for the university to know how the students are satisfied about the Newrest Restaurant.

Marketing research: Research Process:

Marketing research is defined by Carl McDaniel and Roger Gates as “the planning, collection and analysis of data relevant to marketing decision making and the communication of this analysis to management” (Carl McDaniel and Roger Gates, P5, 2007). Moiseenko claims that marketing research has increased in importance today and has become a primary factor of establishing a successful strategy because of the increased competitiveness (Moiseenko, 2002). Carl McDaniel and Roger Gates add that the increased necessity of customer satisfaction and retention have also contributed to the increase of the importance of marketing research (Carl McDaniel and Roger Gates, 2007). There are two types of marketing researches: they can be quantitative, qualitative or a combination of the two. The marketing research process consists of 8 distinct steps according to McDaniel and Gates (2007). Here is a graph in which the entire process is presented. Figure 1: the marketing research process [pic]

Research Context: Newrest restaurant at AUI

At Al Akhawayn University, the Newerst restaurant is an entire building located at the center of the university. The restaurant contains three sub-restaurant categories being: Pizzeria, International restaurant, Grill in the same building, as well as a cafeteria in another one. In fact, the main objective of the Newrest restaurant is to provide students with of variety of products and meals that would allow them save time and energy, rather than going downtown seeking food on a daily basis. From the students perspectives, since they have paid an amount of money during the registration period for the Electronic card “cash wallet”, better use it to pay for their daily food consumption. That is why it is highly important for the university to provide and offer the products that ensure the students’ satisfaction about the food served in the Newrest restaurant. On the other hand, from the university perspective, the Newrest restaurant is considered as a major source of income, that the university can exploit in order to improve its services quality, and hence its image. Moreover, if the university students express dissatisfaction about the Newrest restaurant offerings, it would affect negatively their whole experience in their carriers, which would engender a negative word-of-mouth and so affect depressingly the university’s brand image. Hence, it can be noticed that the students’ satisfaction about the university Newrest restaurant can be an important aspect in influencing the brand image of the university. In fact, customer satisfaction is becoming an important matter nowadays as the world knows a very rapid growth. As the business market is becoming more and more competitive, customer satisfaction is the major issue companies focus on while launching their products in the market. Serving customers well and satisfying them is at the heart of marketing as well as a necessity for a company’s survival in nowadays competitive market. One aspect of the importance of customer satisfaction is that when the products and services quality improves and customers are highly satisfied, this leads to positive word of mouth. This attracts more customers. As a result, the company’s profitability improves. One issue is that Al Akhawayn University does not do frequent surveys to assess the students’ satisfaction about its facilities, their campus life… As a result it does not know exactly whether students are satisfied with the university’s offers or not. It is only recently that Al Akhawayn University has started conducting students’ satisfaction surveys about: the housing services and the residents’ assistants. However, up to now, no specific research or customer satisfaction surveys have been conducted to investigate students’ satisfaction about the university’s Newrest restaurant in particular. Therefore, investigating this aspect of students’ university experience can be significant to the university, regarding the importance of the Newrest restaurant discussed above.

Problem definition

The purpose of this study is to investigate the level of satisfaction of Al Akhawayn university students with the Newrest restaurant with regards to their grade level. Therefore, our problem definition would be as follows: We know that such a question is very broad. Indeed, we have to know how satisfaction about a university restaurant can be measured, and the variables that influence it. Thus, we conducted a literature review in order to investigate the most relevant studies related to this topic and to analyze the theories that would be helpful in finding the correct variables or dimensions influencing students’ satisfaction and in developing our hypothesizes. Our literature review helped us find theories about the measurement tools of satisfaction, about the dimensions that influence students’ satisfaction such as product quality, service quality, and price-quality ratio.

Research Questions

Our Research questions are the following: 1. Is there a significant difference in the level of satisfaction concerning Newrest restaurant environment with regards to AUI students’ grade levels? 2. Is there a significant difference in the level of satisfaction concerning Newrest service quality with regards to AUI students’ grade levels? 3. Is there a significant difference in the level of satisfaction concerning Newrest food quality with regards to AUI students’ grade levels? 4. Is there a significant difference in the level of satisfaction concerning Newrest price with regards to AUI students’ grade levels?

Literature Review

The purpose of this literature review is to provide a benchmark and an examination of existing research and theories on customer satisfaction in general and restaurant services in particular. This information is necessary to back up our hypothesis and clearly define our variables concerning the satisfaction of AUI’s students with the service provided by NEWREST. Several theories have been developed during the last decades as the importance of the marketing concept grew where customers became the center of activities in most businesses and their satisfaction turned to be a key element in companies’ success. The literature review is based on research that has been conducted in developed countries. Yet, the factors analyzed that influence the customer satisfaction in restaurants are similar across nations, for example, all the studies done in Spain, China, Hong Kong and USA used basically the same factors to determine customer satisfaction. The factors that are considered are service quality and product quality. Although these researches used different conceptual frameworks and were conducted in different places, they go along with the Transaction-Specific Model that has been adapted and modified in each research. Inspired from the latter, the gap model provides a view of the relationship between customers’ satisfaction and expectations and management approach. Similarly, other models were developed as the SERVQUAL model developed in the mid eighties by Zeithaml, Parasuraman & Berry where the service quality was identified as a determinant of market share, return on investment and cost reduction (Anderson and Zeithaml 1984; Parasuraman, Zeithaml and Berry 1985). A model that criticized the SERVPERF was based on the work of Bolton and Drew (1991a) states that a customer's perception of service quality is based on his "preconceived" attitude about the service. The fourth model by Grönroos' on Perceived Service Quality model claims that expectations are a function of market communications, image, word of mouth, and consumer needs and learning, whereas experience is a product of a technical and functional quality, which is filtered through the image. In other words, Quality is a function of expectations, outcome and image

Transaction-Specific Model

[pic] Figure 1: Transaction-Specific Model This model was suggested by Teas (1993) and was later expanded by Parasuraman, Zeithaml and Berry (1994) – referred to as PZB. This model considers customers’ overall satisfaction with a transaction to be a function of their assessment of service quality, product quality and price. Transaction-specific satisfaction significantly influences overall customer satisfaction since customers are likely to consider specific aspects of the transaction such as product features (food quality and restaurant ambience), service features (responsiveness of the server), as well as price against the overall restaurant experience (Miguel, Silva, Chiosini, & Schützer, 2007). The full model was found to be significant (Dayang & Francine, 2009). They further state in the results suggested that the model explains adequately customer satisfaction and that service in restaurants should be based upon three major elements: service quality (responsiveness), price and food quality (reliability).

The 5-gaps model

Quality of service has been studied in the area of management. Marketing management has transferred its focus from internal performance such as production to external interests such as satisfaction and customers' perception of service quality. Based on this traditional definition of service quality, Parasuraman, Zeithaml, and Berry (1985) developed the "Gap Model" of perceived service quality. The gap model of service quality is an important customer-satisfaction framework. It evaluates the customer satisfaction through 5 gaps defined as the following: ✓ Gap 1 is the distance between what customers expect and what managers think they expect. ✓ Gap 2 is between management perception and the actual specification of the customer experience. ✓ Gap 3 is from the experience specification to the delivery of the experience. ✓ Gap 4 is the gap between the delivery of the customer experience and what is communicated to customers. ✓ Finally, Gap 5 is the gap between a customer's perception of the experience and the customer's expectation of the service.

SERVQUAL Model

According to Berry et al. (1991), SERVQUAL (service quality) dimensions are used as a basis for assessing customer satisfaction; they are measured by a 22-item scale using 5 quality dimensions. These five dimensions are: 1) Reliability which was defined as the ability to execute the promised service reliably and as requested, 2) Responsiveness: the willingness to help customers rapidly and provide timely service. 3) Assurance: the knowledge, confidence and courtesy of servants (employees) as well as their ability to communicate trust. 4) Tangibles: the appearance of amenities, equipment, personnel, and communication convenience. 5) Empathy: the caring, individualized attention provided to the customers (Dayang & Francine, 2009). As adapted from the research conducted by Andaleeb and Conway (2006), not all the five dimensions of SERVQUAL were used in this research. This is because not all the dimensions play an important role in determining customer satisfaction in the restaurant industry. In the restaurant industry, the customer’s risk is low given the purchase price, the outcome of the service and the alternatives available (Michael A. Jones, 2000). Hence, assurance is not as important in this industry. However, there is a possibility of only particular elements of the dimensions relevant to the study. Andaleeb and Conway (2006), acknowledges that elements of assurance – knowledge and courtesy – are important. The dimension empathy may not be applicable in the restaurant industry context as well, especially related to campus food. Empathy is defined in the SERVQUAL literature as “provision of care and individualized attention that is displayed to each customer” (Dayang & Francine, 2009). Students do not want devoted service that provides personal attention when all they want is to enjoy the food and the environment (Andaleeb and Conway, 2006). This model Indeed, Bolton and Drew (1991b) note that a consumer's current attitude is based on their residual attitude from a previous period about the service quality and their satisfaction or dissatisfaction with the service.

Traditional Macro-Model of Customer Satisfaction (Expectancy-Disconfirmation Model)

“Satisfaction is the consumer’s fulfillment response. It is a judgment that a product or service feature, or the product of service itself, provided (or is providing) a pleasurable level of consumption-related fulfillment, including levels of under- or over-fulfillment” (Oliver, 1997). The model, shown in the figure above, lies behind much of the research in customer satisfaction over the past decade. It considers that perceived performance often differs from the technical performance, especially when it comes to services where the marketing variables are complex, intangible, or when the consumer is unfamiliar with the service. It implies also that the comparison standards can come from numerous sources which are varying widely across individuals, situation, and by products’ types. This model is based upon the “perceived disconfirmation” which is the evaluation of perceived performance according to one or more standards. This latter can have either a positive or negative effect; a positive one implies generally satisfaction and a negative one implies dissatisfaction (Oliver, 1997). Likewise, the feeling of satisfaction was defined as the state of mind and attitude where customers show different satisfaction levels with the experience for each components of the product or the service. If such a feeling exists, the intent of purchase/repurchase is shown as an outcome. Other results could be positive word-of-mouth or complaints if no satisfaction is proved. The most dominant theory of customer satisfaction is the expectancy-disconfirmation model. It does not matter how the service provider believed the service was performed (Oliver, 1997). Instead, it is important for services business managers’ to recognize negative disconfirmation since it represents the largest threat to customer loyalty, word-of-mouth recommendation, repeat purchases, and other desirable customer responses

Loyalty Business Model

Another perspective considers additional aspects other than the process of service delivery. Grönroos (1984) noted that the quality of a service as perceived by customers consists of three dimensions: functional (the process of service delivery to customers), technical (the outcomes generated by the service to the customers), and image (how the customers view the company). Considering those dimensions, the quality of the service is dependent on two variables: the expected service and the perceived service. In addition, this theory studies the link between “relationship strength and customer loyalty”. Thus, based on this model, customer loyalty and satisfaction is determined by three factors: relationship strength, perceived alternatives and critical events in the service/product delivery. The relationship can be ended and satisfaction destroyed in one of these elements is broken. For instance, if the company handles poorly a critical step in the delivery of the service, or puts unexplainable change of price of the service provided, the loyalty relationship might be broken implying a gap in the satisfaction.

Hypothesis formulation

In our literature review, as discussed before we found that Schiffman and Kanuk (2009) define customer satisfaction as how customers perceive the performance of a product or service regarding their expectations. Schiffman and Kanuk (2009) emphasized on the fact that expectations are also a function of one’s reference. Because expectations are a function of one’s, background experience, etc, we can suppose that there is a difference in the level of satisfaction regarding Newrest Restaurant in AUI between different groups of students’ mainly academic level. This hypothesis is based on the fact that students from different groups have different frames of reference. To find which category of students to consider in our research, we looked at the study done by Nassrin (2009), about students’ satisfaction about the Xavier university. Nasrin, found that students satisfaction was significantly different between students of different academic levels. Based on this, and the variables identified before to measure customer satisfaction about shops, the hypothesis that we will try to test in this project are: 1. There is no difference in the level of satisfaction concerning Newrest restaurant environment with regards to AUI students grade levels 2. There is no difference in the level of satisfaction concerning Newrest service quality with regards to AUI students grade levels 3. There is no difference in the level of satisfaction concerning Newrest food quality with regards to AUI students grade levels 4. There is no difference in the level of satisfaction concerning Newrest price with regards to AUI students grade levels

Research design:

The research design we will choose is the quantitative research design. Indeed, our main purpose is to collect data that will enable us to measure AUI students’ satisfaction about the Newrest restaurant regarding the grade level (freshmen, sophomore, Junior, Senior or Graduate). Because we need to measure something about a sample to make the necessary inferences about the desired population, quantitative tools are thus needed. These variables are: ✓ Service quality ( responsiveness and assurance ✓ Food quality( reliability ✓ Environment (tangibles ✓ Price/quality ratio ( Value for the money in the SERVQUAL Moreover, we want to describe some basic elements about our population, and depict its level of satisfaction. Since the variables that will help us to correctly measure their satisfaction were already identified previously (SERVQUAL), all we need to do is simply to measure the sample’s satisfaction levels and to use it these findings to better describe it. Therefore, we are not looking to set a causal relationship between the variables set earlier and the level of satisfaction and that is the reason why we will use a descriptive study.

Methods of research:

In quantitative research designs, many methods of research exist such as experiments, surveys and observations. Because we are doing a descriptive study, we don’t need to do experiments, which is a method exclusively liked to causal studies. Since we do not have the necessary skills to conduct efficient observation, we have opted for surveys, more especially questionnaires, as the only method of research we will use. This method will also enable us to reach a high number of respondents in less time, and to be able to have a more precise idea about the level of satisfaction of the whole population through the Central Limit Theorem. Furthermore, since the population of AUI students know how to read and write, nobody will be excluded from our sample. The only difficulty here is to have a good questionnaire design, so that we will be able to collect efficiently the primary data of interest.

Sampling procedures:

Concerning the sampling procedure, we will first get a sample size of 200 students. In addition, probability samples allow the researcher to calculate the precision of the estimates obtained from the sample. We believe that 200 students out of 1500 students is very representative of the population of AUI students since it constitutes about 13.34% of the population. Moreover, we have chosen to use the simple random sample (SRS) because we believe that it is the best sampling procedure, since it allows every person in the population to have the same chance of being selected. Therefore, the sampling error or sample bias resulting from our sample will be greatly reduced. Nonetheless, such a sampling procedure is difficult to do in a practical way since we need to distribute the samples accordingly so that no group can possibly be excluded from the sample.

Data collection:

This part is pretty difficult in a practical way since it pushes us to find the best strategy that will enable us to efficiently gather our data, which is in our case primary data, so that we will have the right data in our context. The right data in our context is characterized by the following criteria. First it should cover all the possible opinions of the population of AUI students. Second and finally, it should be valuable, which means, it should accurately answer to the problem definition. As a consequence, here is our strategy for our data collection that will respond to both these criteria. We will try to divide AUI with respect to the main zone where the profiles of different students are located. Here are the zones we came up with: the cafeteria, the restaurants, the main buildings of the three schools we have in AUI (SBA, SSHS, and SSE), the cafeteria, the sport complex, the labs, the shop and the library. Since we are 5 in the team and that we localized 10 areas with different types of students, we decided to give to each member 40 questionnaires to be filled by students and each member has to take place of 2 of the 10 locations. In order to avoid any situation where one student can answer more than once, we simply ask all the students to who we give the questionnaire if they have already filled the questionnaires linked to Newrest satisfaction. Concerning, the accuracy of our data, the only variable on which we can play is the questionnaire design that should be the most efficient possible.

FINDINGS AND ANALYSIS

Because we want to compare the means of five groups, we will use an ANOVA test Moreover, in this study, there is only one factor under study which is the grade level of students, with five levels (freshmen, Juniors, Sophomores, Seniors, Graduates). So, we will use more precisely a One-Way Between-Groups ANOVA test. Our decision will be based on the P value. If p value is lower than the level of significance α, we reject the null hypothesis. Since we are doing a marketing research, we have chosen α = 0.05. The hypotheses that we are testing are: 5. There is no difference in the level of satisfaction concerning Newrest restaurant environment with regards to AUI students grade levels 6. There is no difference in the level of satisfaction concerning Newrest service quality with regards to AUI students grade levels 7. There is no difference in the level of satisfaction concerning Newrest food quality with regards to AUI students grade levels 8. There is no difference in the level of satisfaction concerning Newrest price with regards to AUI students grade levels

In order to run ANOVA test correctly, our team checked the assumptions of ANOVA. The assumptions are according to: The dependent variable has a normal distribution • Each sample is randomly selected and independent • The populations are assumed to have equal standard deviations (or variances) • We will also look at the descriptive statistics of the level of satisfaction of students toward Newrest environment, service quality, price, and food quality.

Checking the assumptions of the one way ANOVA test

Normal distribution

In order to assess if the values of a sample are normally distributed to measures are used: the skeweness and kurtosis. When we state that the values of a sample are normally distributed we mean that the curve of the values is bell shaped. This reflects by having a large amount of the observations that are clustered around the mean.The following figure describes a normally distributed distribution.

Figure 1: A normal Distribution, bell-shaped curve

[pic]

Skewness

The measure of skewness helps determine the position of the peak of the distribution relative to the center of the distribution. If the measure of the skewness is positive, then we say that the distribution is right skewed. If the value of the skewness is negative, we say the distribution is left skewed.

Kurtosis

The measure of the kurtosis helps determine whether the values of the distribution are more located on the peak of the distribution or on the tail of the distribution (Rakotomalala, 2008). This measure informs about the relative flateness of the distribution. If the measure of the kurtosis is above zero, this means that the values of the distribution are concentrated around the peak. At the opposite if the measure of the kurtosis is below zero, this means that the values of the distribution are located around the tail.

Formula of Kurtosis and Skewness over their Standard error

According to Bai & Ng (2009), ‘’values of skewness and kurtosis have little inherent meaning, other than large values indicate greater asymmetry’’. The theory says that the absolute value of the measure of skweness over the standard error of skewness should be less than 2 to be said not to be significantly skewed. The same rule applies for kurtosis. By verifying those two formulas, we can see that the distribution do not significantly depart from being more or less symmetrical. In most case, there is little distribution that is perfect normal distributions. In reality, it is rare to find data that is perfectly normally distributed,

6) Evaluation of Normality of the Data

If the value of the ratio of the Skeweness over the standard error of Skeweness is between -2 and 2, we can conclude that the distribution is not significantly skewed and the result will support the assumption of normality for the research. Also, if the value of the ratio of the kurtosis over the standard error of the Kurtosis is between -2 and 2, we will conclude that the distribution for the group is not significantly Kurtoic and it will support the assumption of normality. The following table depicts the skeweness and Kurtosis of our different groups of students (Freshman, Junior, Sophomore, Senior, Graduate) according to each variable (Price, Service quality, Food quality, and Environment).

[pic]

As we can see, the assumption of normality can be assumed for the variable Food Quality and Environment. Indeed, all the groups’ ratios are between -2 and 2 for the skewness. Thus we can conclude the distribution is not significantly skewed. For the kurtosis, all the ratios of the kurtosis over the standard error are also between -2 and 2, thus we can conclude that the distribution is not significantly Kurtoic. For the distribution of the prices, all the ratios of the Skeweness over standard error of the Skeweness are between -2 and 2, except for the Freshmen group (SK/STD= 2.90). For this group the distribution is skewed. However, since the sample of Freshmen students is made of 47 subjects which is more than 30, according to the Limit Central Theorem we can assume that the distribution is normally distributed. However, the ratios of the Kurtosis over the Standard error of the Kurtosis are all between -2 and 2, thus we can conclude that the distributions for those groups are not significantly Kurtoic. To conclude, we will also assume that the distribution for the prices is normally distributed. For the distribution of the service quality, we can notice that 3 values of the ratio of the Skeweness over the Standard error of Skeweness are greater than 2. Indeed the ratio is equal to 3.29 for the Freshmen group, 3.03 for the Junior group, and 3.34 for the Sophomore group. Thus for those three groups we can conclude that the distributions are skewed. In addition all the values of the ratio of the Kurtosis over the standard error of the Kurtosis are between -2 and 2 for all the grade levels except for the Freshmen (Kurt/STD=3.55). For the distributions of Junior, Sophomore, Senior and Graduate students we can conclude that the distribution is not significantly Kurtoic. However, for the sample of the freshmen, we conclude that the distribution is Kurtoic. Nevertheless, since the samples for the freshmen, Junior, and sophomore are respectively equal to 47, 46, 32, we can assume that they are normally distributed according to the Central Limit Theorem. In fact, as long as the samples are made of more than 30 observations, we can assume normality of the distribution. To conclude, we assume that the distribution in each cell is normally distributed and we valid one of the assumption of the ANOVA test.

Homogeneity of variances

In order to check for equality of variances, our team used Levene’s test. Levene’s test is a homogeneity-of-variance test that is less dependent on the assumption of normality than most tests. For each case, it computes the absolute difference between the value of that case and its cell mean and performs a one-way analysis of variance on those differences. According to Isabel Parra-Frutos (2009), levene’s test is known for its high statistical power and is used to test for homogeneity of variances (source 1). If the levene’s test is significant with a P value < α, one can conclude that the two variances are significantly different, but if P value > α, the two variances are not significantly different. For the level of significance α, we will use α = 0.1 because we are conducting a marketing research study.

Leven’s test steps procedure:

The first step when conducting a levene’s test is to check for the assumption. We have first to check that the samples from the populations under consideration are independent and the populations under consideration are approximately normally distributed. The second step is to State the null and alternative hypotheses. The third step is to decide on the significance level, α. The fourth step is to determine the critical value and rejection region. The fifth step is Compute Levene’s statistic, and the last step is to make our decision The Levene’s test is known for its high statistical power and is used to test for homogeneity of variances (source 1). To test for the equality of group variances you use the following hypothesis. Our team members conduct levene’s test in SPSS in order to study each variable, which are service quality, food quality, environment, and price. Step1 After checking the assumptions, one can conclude that the samples from the populations under consideration are independent and the populations under consideration are approximately normally distributed. Step 2 -The null hypothesis: the variances of the level of satisfaction of food quality, service quality, environment, and price of the five group students (freshmen, juniors, seniors, sophomores, graduates,) are equal. -The alternative hypothesis: at least two variances of the level of satisfaction of food quality, service quality, environment, and price of the five group students (freshmen, juniors, sophomores, seniors, and graduates) are not equal. H0: δ² 1 = δ² 2 = δ²3= δ²4= δ²5 H1: δ² 1 ≠ δ² 2 ≠ δ²3≠ δ²4≠ δ²5 Step 3 We will use α = 0.05 Step 4 SPSS output: Test of Homogeneity of Variances Service Quality
|Levene Statistic|df1 |df2 |Sig. |
|2.295 |4 |195 |.061 |

Test of Homogeneity of Variances Price
|Levene Statistic|df1 |df2 |Sig. |
|.959 |4 |195 |.431 |

[pic] From the tables above, p values of the test of homogeneity of variances for the shop environment and price are respectively0.061, 0.431 and 0.255. Three of them are greater than the level of significance α (0.05), so we do not reject the null hypothesis. We conclude that there is not enough evidence at 5% level of significance that at least two variances are different for the five variables which are freshmen, juniors, sophomores, graduates, and graduates , so the assumption of homogeneity of variances for the shop environment, service quality, and price is accepted.

[pic] For the food quality, the p value is 0.017, which is lower than the level of significance α (0.05), so we reject the null hypothesis. We conclude that there is evidence at 5% level of significance that at least two variances are different. Therefore the assumption of equality of variances does not hold for service quality and food quality. Three of the Levene’s test enables us to support the assumption of equality of variances: for since the samples we are studying are large enough we will assume that the sample is robust not to violate the assumption for the last one, and we will also assume equality of variances to conduct the ANOVA Tests.

Independence of the samples

According to Garson (2007), “independent observations are assumed by many statistical procedures”. Independence can be verified by the Durbin Watson coefficient that uses studentized residuals. If the Durbin Watson coefficient is between 1.5 and 2.5, we can assume that observations are independent. The following table depicts the Durbin Watson coefficients for the observations of the four dependent variables under study: Food quality, Service Quality, Environment, and Price. [pic] Table: Durbin Watson Test for the Service Quality, Food Quality, Environment, and Price

As the above table shows the Durbin Watson coefficients for the Service Quality, Food Quality, Environment, and Price are all between 1.5 and 2.5. Thus, we can conclude that the observations are all independent. The assumption of independence is supported for the ANOVA test.

Descriptive statistics

Description of the Sample

The total sample was made of 200 Students. The students are from five different grade levels: freshman, junior, sophomore, senior, and graduate. The following tables summarize the characteristics of the sample studied. [pic] As the Table above shows 52% of the sample was made of women while, 48% was made of males. The questionnaire was answered by 96 males and 104 females. The sample was made of 47 freshman students, 46 junior students, 32 sophomor students, 57 senior students and 18 graduate students. Also, 101 students were from the school of business and administration, 66 were from the school of social sciences and 33 were from the school of engineering. The tables above also describe the percentages and proportions of students in the different grade level according to their gender and school. [pic] [pic] [pic] Table *: Descriptive Statistics of the frequency of eating at Newrest The above table depicts the descriptive statistics of the frequency of eating in Newrest restaurant of the students depending on their grade level. The question about the frequency of eating in Newrest was asked to the students on a scale of 1 to 5. The following table details the meaning of each of the scale levels.
|1 |2 |3 |4 |5 |
|Every day |At least 3 times a week |Once a week |Rarely |Never |

It seems from the mean level of each grade level that no group of students eats every day at Newrest restaurant while the restaurant is at proximity on campus. Also, no group of students seems to never go or go rarely to eat in Newrest restaurant. All the means are between 2.5 and 3.5 which mean that although students in different grade levels do not go everyday to the restaurant, they go between 1 and at least 3 times to eat there each week. There is no significant difference in the mean level of each group concerning the frequency of eating in the restaurant as it shown by the figures. Still the seniors seems to eat more often in Newrest (mean= 2.33) than the freshman, sophomores, juniors, and Graduate. It does not seem from the figures that the fact of being of a given grade level significantly affects how often a student goes to eat in the restaurant. Also, we should not forget that students do not really have the choice of eating in a lot of restaurant as Newrest is one of the two only restaurants that exist in the university. [pic] Figure *: Mean Frequency of Eating As the above Bar chart shows the senior are the ones who eat the most often in the restaurant, followed by the juniors, and the Graduate students then the freshman and finally the juniors.

Service Quality Descriptive Statistics

[pic] Table: Descriptive Statistics of the Service Quality The table above depicts the descriptive statistics of the Service quality of Newrest restaurant with regards to the grade level of the students at AUI. The answers made by the students were on a scale of 1 to 5, 1 describing that the student judged the service as being of very low quality and 5 describing that the student judged the service of Newrest as being very good. As the table shows, there is no apparent significant difference in the means of evaluation of service quality among different grade levels. Indeed, the highest mean is equals to 2.54 and the lowest mean 1.87. However, freshman judge the service quality as being lower (mean = 1.87) than the sophomores (1.97), followed by the Graduate students (mean = 2), followed by the juniors (2.02), and finally by the seniors students (mean=2.54). It seems that the seniors students that answered to the questionnaire believe that the service quality is a little lower than average (mean higher than 2), while the freshman believe that the service quality at Newrest is not good (mean lower than 2). In general, the students that were questioned at AUI judge that the service quality is low in Newrest restaurant (mean= 2.13) since the mean of the total students questioned is just a little above 2. The following bar chart gives a visual picture of what have been described earlier. As shown seniors seem to rate the service quality of Newrest higher than do the freshman. [pic] Figure *: Service quality with regard to Grade Level

Descriptive Statistics Food Quality

[pic] Table*: Descriptive Statistics of the Food Quality Investigation with regard to Grade Level The table above depicts the descriptive statistics of the Food Quality of Newrest restaurant with regards to the grade level of the students at AUI. The answers made by the students were on a scale of 1 to 5, 1 describing that the student judged the food as being of very low quality and 5 describing that the student judged the Food of Newrest as being very good. As the table shows, there is an apparent difference in the means of evaluation of food quality among different grade levels. Indeed, the highest mean is equals to 4.02 and the lowest mean 2.89. Freshman judge the food quality as being good (mean = 4.02) compared to the Juniors (3.43), followed by the Senior students (mean = 3.19), followed by the Sophomores (3.06), and finally by the Graduate students (mean=2.89). It seems that the Freshman students that answered to the questionnaire believe that the food quality is good, not average nor very good but good (mean = 4).) At the opposite the Graduate students believe that the food quality at Newrest is a little below average (mean = 2.89). In general, the students that were questioned at AUI judge that the food quality is above average (mean = 3.4) in Newrest restaurant. The following bar chart better describes what was just explained. As you can see it is obvious that the freshman judge the food quality as being better that the graduate or the seniors. One of the reasons may be that the freshman are still new and thus did not get used to the food yet. Another reason may be that the oldest students have more requirements regarding the quality of food they expect. The test of ANOVA will help us determine if there is a really a difference in the populations means. [pic] Figure *: Bar Chart of the Food Quality Level with regards to Grade Level

Descriptive Statistics for the Environment of Newrest Restaurant

[pic] Table *: Descriptive Statistics for the Satisfaction level with Regards to the Environment The table above depicts the descriptive statistics of the satisfaction level of the Environment of Newrest restaurant with regards to the grade level of the students at AUI. The answers made by the students were on a scale of 1 to 5, 1 describing that the student were not satisfied at all with the Environment of Newrest restaurant and 5 describing that the student were very satisfied with the Environment of Newrest Restaurant. As the table shows, there is no significant difference in the means of the satisfaction level of the Environment of the restaurant among different grade levels. Indeed, the highest mean is equals to 2.94 and the lowest mean 2.28. as the table shows the Freshmen score the highest (mean = 2. 94) compared to the Juniors (2.76), followed by the Sophomore students (mean = 2.69), followed by the Seniors (2.33), and finally by the Graduate students (mean=2.28). It seems that the Freshman students that answered to the questionnaire are neither satisfied nor dissatisfied with the environment of Newrest Restaurant (mean = 2.94).), while the Graduate students are almost not satisfied with the environment of the restaurant (mean= 2.28). In general, the students that were questioned at AUI about the restaurant environment were neither satisfied nor dissatisfied by the environment of Newrest Restaurant (mean=2.63). The following bar chart better describes what was just explained. As you can see it is obvious that the freshmen are more satisfied with the environment of Newrest restaurant than the other students. One of the reasons may be that the freshmen appreciate gathering to hang out in Newrest. Another reason may be that the oldest students are disturbed by the noise in Newrest restaurant or just have more requirements regarding the cleanliness of the restaurant. The test of ANOVA will help us determine if there is a really a difference in the populations means. [pic] Figure *: Barb Chart of the Satisfaction Level of Newrest Environment with regards to Grade Levels

Descriptive Statistics of the satisfaction Level of the Prices offered at Newrest

[pic] The table above depicts the descriptive statistics of the satisfaction level of the students concerning the prices of Newrest restaurant with regards to the grade levels. The answers made by the students were on a scale of 1 to 5, 1 describing that the student were not satisfied at all with the Environment of Newrest restaurant and 5 describing that the student were very satisfied with the Prices of Newrest Restaurant. As the table shows, there is no significant difference in the means of the satisfaction level of the students for the prices of the restaurant among different grade levels. Indeed, the highest mean is equals to 2.59 and the lowest mean 2.3. As the table shows the sophomores score the highest (mean = 2. 59) compared to the Seniors (2.42), followed by the Junior students (mean = 2.37), followed by the Graduate (2.3), and finally by the Freshman students (mean=2). It seems that the Freshman students that answered to the questionnaire are not satisfied with the prices of Newrest Restaurant and may think it is expensive (mean = 2), while the Sophomore students are almost neither satisfied nor dissatisfied with the prices of the restaurant (mean= 2.59). In general, the students that were questioned at AUI about the restaurant prices were not satisfied by the prices of Newrest Restaurant (mean=2.33). The following bar chart better describes what was just explained. As you can see the freshman are the least satisfied with the prices of Newrest restaurant compared to the other students. One of the reasons may be that the freshmen have less money than the other students in their cash wallets. The test of ANOVA will help us determine if there is a really a difference in the populations means. [pic] Figure *: Barb Chart of the Satisfaction Level of Newrest Prices with regards to Grade Levels

ANOVA test

Newrest environment

Step 1
State null and alternative hypothesis: H0 = μ1= μ2= μ3= μ4= μ5 H1= At least one of the means is different
Step2
Since we are doing a marketing research, we have chosen α = 0.05
Step 3 We will use more precisely a One-Way Between-Groups ANOVA test. The SPSS output is as follow: [pic] Step 4

Interpretation of results and decision:

The P value (.001) is lower than the level of significance 0.05. We reject the null hypothesis; there is difference in the mean level of satisfaction about Newrest environment with regards to AUI students’ grade levels. We can conclude that there is evidence at 5% level significance that the average level of satisfaction toward Newrest environment differs significantly with regards to AUI students grade levels.

Newrest service quality

Step 1 State null and alternative hypothesis: H0 = μ1= μ2= μ3= μ4= μ5 H1= At least one of the means is different
Step2
Since we are doing a marketing research, we have chosen α = 0.05
Step 3 We will use more precisely a One-Way Between-Groups ANOVA test. The SPSS output is as follow: [pic] Step 4 The P value (.015) is lower than the level of significance 0.05.We reject the null hypothesis , so there is difference in the mean level of satisfaction about Newrest service quality with regards to AUI students grade levels. We can conclude that there is evidence at 5% level of significance that the average level of satisfaction toward Newrest service quality differs significantly with regards to AUI students grade levels
Newrest Food quality
Step 1 State null and alternative hypothesis: H0 = μ1= μ2= μ3= μ4= μ5 H1= At least one of the means is different
Step2
Since we are doing a marketing research, we have chosen α = 0.05
Step 3 We will use more precisely a One-Way Between-Groups ANOVA test. The SPSS outpout is as follow: [pic]
Step 4 The P value (.001) is lower than the level of significance 0.05. We reject the null hypothesis, so there is difference in the mean level of satisfaction about Newrest food quality with regards to AUI students’ grade levels. We can conclude that there is evidence at 5% significance that the average level of satisfaction toward Newrest food quality differs significantly with regards to AUI students’ grade levels.

Newrest Price

Step 1 State null and alternative hypothesis: H0 = μ1= μ2= μ3= μ4= μ5 H1= At least one of the means is different
Step2
Since we are doing a marketing research, we have chosen α = 0.05
Step 3 We will use more precisely a One-Way Between-Groups ANOVA test. The SPSS output is as follow: [pic] Step 4 The P value (.231) is greater than the level of significance 0.05.We do not reject the null hypothesis, so there is no difference in the mean level of satisfaction about Newrest price with regards to AUI students grade levels. We can conclude that there is evidence at 5% level of significance that the average level of satisfaction toward Newrest price does not differ significantly with regards to AUI students’ grade levels

Post Hoc Test: The Tukey-Kramer Procedure

If the null hypothesis is rejected and the alternative hypothesis is accepted, then the researcher will proceed to a post hoc comparison of the means to determine which means are different (source 4). This method was found by John Tukey and then improved by Kramer (source 1). This method makes it possible To compare each pair of mean groups at the same time. We will use a significant level equal to 0.05.
The Environment of Newrest vs academic levels The ANOVA test showed that at least one the population means of the grade levels groups differ in terms of their satisfaction regarding Newrest environment. [pic] ( The p values freshmen vs. Junior and freshmen vs. sophomores are greater than α=0.05. Thus there population means are equal. Freshmen, Junior and sophomore students have the same satisfaction level with regard to the environment of Newrest. ( The p values of freshmen vs. Senior and freshmen vs. Graduate are lower than α=0.05. Thus there is a difference in the population means of freshmen with seniors and Graduates. Freshmen are more satisfied with the environment of Newrest than seniors and Graduate Students. (The p value junior vs. Graduates is greater than α=0.05. Thus their population means are equal. Junior and Graduate students have the same satisfaction level with regard to the environment of Newrest. ( The p values of junior vs. Senior is lower than α=0.05. Thus there is a difference in the population means of junior and senior. Junior students are more satisfied with the environment of Newrest than Senior Students. ( The p values sophomore vs. Senior and Graduate are greater than α=0.05. Thus there population means are equal. Sophomores, seniors and Graduates have the same satisfaction level with regard to the environment of Newrest. ( The p values senior vs. Graduate is greater than α=0.05. Thus their population means are equal. Seniors and Graduates have the same satisfaction level with regard to the environment of Newrest. To conclude, there is a difference in the population means with regards to the satisfaction level of the environment of the Newrest between the freshmen with the Graduates and the freshmen with the seniors. There is also a difference in the satisfaction level between the junior students and the senior students. Freshmen are more satisfied with the environment of Newrest than seniors and Graduate Students. Junior students are more satisfied with the environment of Newrest than Senior Students. All the conclusions are made at 5% level of significance

The Service Quality of Newrest vs. academic levels

The ANOVA test showed that at least one the population means of the grade levels groups differ in terms of their satisfaction regarding Newrest Service Quality. [pic]

The Service Quality of Newrest vs. academic levels

The ANOVA test showed that at least one the population means of the grade levels groups differ in terms of their satisfaction regarding Newrest Service Quality. [pic] ( The p values freshmen vs Junior, sophomores and Graduates are greater than α=0.05. Thus there population means are equal. Freshmen, Junior, Graduates and sophomore students have the same satisfaction level with regard to the service quality of Newrest. ( The p values of freshmen vs. Senior are lower than α=0.05. Thus there is a difference in the population means of freshmen with seniors. Freshmen are less satisfied with the service quality of Newrest than Seniors Students. (The p- values for junior vs. Graduates, Senior, and sophomore are greater than α=0.05. Thus their population means are equal. Junior, Senior, sophomore and Graduate students have the same satisfaction level with regard to the service quality of Newrest. ( The p values sophomore vs Senior and Graduate are greater than α=0.05. Thus there population means are equal. Sophomores, seniors and Graduates have the same satisfaction level with regard to the service quality of Newrest. ( The p values senior vs Graduate is greater than α=0.05. Thus their population means are equal. Seniors and Graduates have the same satisfaction level with regard to the service quality of Newrest. To conclude, there is a difference in the satisfaction level between Freshmen students and Senior students. Freshmen are less satisfied with the service quality of Newrest than Seniors Students. All the conclusions are made at 5% level of significance

Level of Satisfaction of the Food Quality of Newrest vs. academic levels

The ANOVA test showed that at least one the population means of the grade levels groups differ in terms of their satisfaction regarding Newrest Food Quality. [pic] ( The p values freshmen vs. Junior is greater than α=0.05. Thus there population means are equal. Freshmen, Junior and sophomore students have the same satisfaction level with regard to the Food quality of Newrest. ( The p values of freshmen vs. Senior, Graduate, and sophomores are lower than α=0.05. Thus there is a difference in the population means of freshmen with seniors, Graduates, and sophomores. Freshmen are more satisfied with the Food Quality of Newrest than seniors, sophomores and Graduate Students. (The p value junior vs. Graduates, seniors and sophomores are greater than α=0.05. Thus their population means are equal. Junior students have the same satisfaction level with regard to the Food quality of Newrest as sophomores, Seniors and Graduates. ( The p values sophomore vs. Senior and Graduate are greater than α=0.05. Thus there population means are equal. Sophomores, seniors and Graduates have the same satisfaction level with regard to the Food Quality of Newrest. ( The p values senior vs. Graduate is greater than α=0.05. Thus their population means are equal. Seniors and Graduates have the same satisfaction level with regard to the Food Quality of Newrest. To conclude, there is a difference in the population means with regards to the satisfaction level of the Food Quality of the Newrest between the freshmen with the Graduates, seniors, and sophomores. Freshmen are more satisfied with the Food Quality of Newrest than seniors, sophomores and Graduate Students. All the conclusions are made at 5% level of significance.

Bibliography

Dayang, A. M., & Francine, R. (2009). Influence of Service and Product Quality towards Customer Satisfaction. Retrieved April 10, 2010, from World Academy of Science, Engineering and Technology: http://www.waset.org/journals/waset/v53/v53-28.pdf

Michael A. Jones, J. S. (2000). Transaction-specific satisfaction and overall satisfaction: an empirical analysis. Emerald , 147-159.

Oliver, R. (1999). Value as Excellence in the Consumption Experience. In M. Holbrook (Ed.), Consumer Value: A Framework for Analysis and Research, 43-62, New York: Routledge.
Parra-Frutos, I. (2009). The Behaviour of the Modified Levene’s Test When Data is not normally Distributed. Retrieved April 10, 2010, from: http://www.springerlink.com/content/j737334768k47578/

-----------------------
Problem definition

Research Design

Method of research

Sampling procedures

Follow up

Write the report

Analysis of data

Collect data