Findings and Discussions

It is a common practice for companies the views of their customers or clients regarding the quality of services or goods they offer to them. These feedbacks are very crucial for decision making and improvement of the services or goods they offer (Kapustina & Babenkova, 2010). Getting feedback is also very crucial for improving the management since the management team will have the clue about the customers’ attitude towards their goods or services (Verhoef, Reinartz, & Krafft, 2010).

Customer satisfaction is a way of measuring whether the products or services offered meet or even surpass the expectations of the customers (Shestakov, Sidorov, Shefer, & Gichkina, 2010). Customer satisfaction will determine customers’ attitude towards the company or the service or product (Verhoef, Reinartz, & Krafft, 2010).

In this study, the main aim is to find out the satisfaction of customers for a vacation resort and the vacation resort factors. This investigation has been carried out with the regular customers of the resort. The customers reported their family size, their gender, overall attitude and their level of satisfaction under different choices.

The data used in analysis is from a secondary source. The sample consists of thirteen (13) variables describing each individual who gave their responses. The variables include Convenient Location, Internet Connect, Staff service, Business Facility, Clean room, Bed/mattress/pillow, Secretarial Service, attitude 1, attitude 2, attitude 3, family size, gender and age.

Correlation analysis among Convenient Location, Internet Connect, and Staff service, Business Facility, Clean room, Bed/mattress/pillow and Secretarial Service has been done. Similarly, a regression analysis has been done between the overall attitude and Convenient Location, Internet Connect, Staff service, Business Facility, Clean room, Bed/mattress/pillow, Secretarial Service and age. The data from the customers’ report have been analysed to draw some important conclusions and recommendations. Correlation and regression analysis have been done. The results have been obtained using SPSS. The results have been displayed as well.

Part A: Correlation Analysis

Correlation analysis is involves establishing the strength of relationships among the variables under study (Frankfort-Nachias, 2015). Relationship that exists among the variables can be described by observing the correlation coefficient (Stuart A., 2009). The coefficient of correlation will as well determine the strength of the correlation that exists among the variables. Correlation coefficient will also tell whether the relationship is negative or positive.

A correlation is positive when the value of the correlation coefficient is between 0 and 1 (Jackson, 2017). Similarly, a correlation is negative if the coefficient is between 0 and -1 (Lee, Eun, Park, & Chan, March, 2014). A negative or positive correlation can as well be classified as either strong or weak.

Correlation of overall attitude among the 3 groups and size of family

A strong positive correlation is that with a value falling in the range of 0.5 and 1 while a weak positive correlation has a value falling between 0 and 0.5 (Lind, 2008). Correlation coefficient of 1 shows a perfect correlation or perfect relationship between the variables (Jackson, 2017).  

A strong negative correlation has a value falling between 1- and -0.5 while a weak correlation has value falling between -0.5 and 0 (Jackson, 2017). A negative correlation or relationship is an indication of an increase in the value of one variable by given units that causes a corresponding decrease in the other variable. Similarly, a negative correlation implies that a decrease in the value of one variable causes a corresponding increase in the other variable by a given value (Lee, Eun, Park, & Chan, March, 2014).

In this section, a correlation analysis has been done to establish the nature of relationships that exists among the three levels of attitudes; attitude 1, attitude 2 and attitude 3. The analysis has been done in SPSS and the output is shown in the table below.

From the table below, the correlation coefficient between attitude 1 and attitude 2 is 0.907, the correlation coefficient between attitude 1 and attitude 3 is -0.864 and the correlation between attitude 2 and attitude 3 is 0.004. Clearly, this is a strong relationship correlation between attitude 1 and attitude 2 (0.907), a strong negative correlation between attitude 2 and attitude 3 (-0.916) and a weak positive correlation between attire 2 and attitude 3.

The results tell us that an increase in attitude 1 by one unit will cause a corresponding increase in attitude 2 by 0.907. Similarly, an increase in attitude 1 by one unit causes a corresponding decreases in the level of attitude 3 by 0.907. Finally, an increase in the level of attitude 2 by one unit causes a corresponding increase in the level of attitude 3 by 0.0044.

Correlations
	overall attitude1	overall attitude2	overall attitude3
overall attitude1	Pearson Correlation	1	.907**	-.864**
Sig. (2-tailed)		.000	.000
N	236	236	236
overall attitude2	Pearson Correlation	.907**	1	-.916**
Sig. (2-tailed)	.000		.000
N	236	236	236
overall attitude3	Pearson Correlation	-.864**	-.916**	1
Sig. (2-tailed)	.000	.000
N	236	236	236
Correlation is significant at the 0.01 level (2-tailed).

In this section, the aim is to find out the nature of association between the overall attitude and the family size. This has been done in SPSS. The outcome of the analysis are contained in the figure below. Clearly, the correlation coefficient between attitude 1 and family size is 0.555, the correlation coefficient between attitude 2and family sizes is 0.557, the correlation coefficient between attitude 3 and family size -0.539.

From the above results, the association is positive correlation the overall attitude and the family size. This implies that as the overall attitude increases, there is a corresponding increase in the size of the family.

Further, a weak positive relationship between attitude 1 and family size (0.194). A weak positive association between attitude 2 and family size (0.207). A strong positive association between attitude 3 and family size (-0.522).

Correlations
	overall attitude1	overall attitude2	overall attitude3	family size
overall attitude1	Pearson Correlation	1	.907**	-.864**	.555**
Sig. (2-tailed)		.000	.000	.000
N	236	236	236	236
overall attitude2	Pearson Correlation	.907**	1	-.916**	.557**
Sig. (2-tailed)	.000		.000	.000
N	236	236	236	236
overall attitude3	Pearson Correlation	-.864**	-.916**	1	-.539**
Sig. (2-tailed)	.000	.000		.000
N	236	236	236	236
family size	Pearson Correlation	.555**	.557**	-.539**	1
Sig. (2-tailed)	.000	.000	.000
N	236	236	236	236
Correlation is significant at the 0.01 level (2-tailed).

Part B:  Regression

In this section, the aim is to establish the relationship between the overall attitude and the several levels of choices plus age. This has been achieved using a regression analysis. Regression analysis outlines whether there exists any association between an independent variable (s) and dependent variable (s) (Doda, Gennaioli, Groundson, & Sullivan, 2014). A regression output also outlines the suitability of the sample in making inferences about the population under study (Kapustina & Babenkova, 2010).

There are three tables of output as outlined below. The first table is the summary table. The value of R in the summary table is the regression coefficient between the dependent and the independent variables (Chesnokova, Barvenko, & Bradina, 2013). This value represents the gradient of the regression line connecting the dependent and the independent variables since this is a linear regression model (S, 2008). Coefficient of regression is 0.751. This value implies that an increase in the values of dependent variable by 1 unit causes a corresponding increase in the value of the independent variable (s) by 0.751.

Summary table has R square value. This is the value that tells of the suitability of the sample in making inferences about the population. The value of R square is 0.171. This value is equivalent to 56.3%. This is an implication that our sample explains 17.1% of the population. Hence this indicates that the sample is statistically appropriate for use in making any inference (s) about the population.

Model Summary
Model	R	R Square	Adjusted R Square	Std. Error of the Estimate
1	.413a	.171	.142	.373486961498210
a. Predictors: (Constant), Age, Choice_05, Choice_04, Choice_01, Choice_02, Choice_03, Choice_07, Choice_06

The second table shown below is the ANOVA table. ANOVA is Analysis of Variance. The purpose of analysing the variance to find out if there is any significant difference in the average values or the means of the variables under the study. This output has been produced at 5% level of significance. From the outcome below, the significance value or the probability value is 0.00. This is a value which is less than the level of significance used. Therefore, it is accurate to say there exists a significant difference in the average values or averages of the variables under this study.

ANOVAa
Model	Sum of Squares	df	Mean Square	F	Sig.
1	Regression	6.566	8	.821	5.884	.000b
Residual	31.944	229	.139
Total	38.510	237
a. Dependent Variable: OveralAttitudel
b. Predictors: (Constant), Age, Choice_05, Choice_04, Choice_01, Choice_02, Choice_03, Choice_07, Choice_06

The third and the final table is the coefficients table. This is a table that outlines the regression coefficients between the difference variables. From this table, the regression model can be developed. The purpose of developing a regression model is to be able to writrre the dependent variables in terms 7gof the independent variables.

Regression analysis is a form of hypothesis testing. Hypothesis is a given argument about a study topic whose truth can be tested (Jackson, 2017). Hypothesis testing forms the basis of the discussion and recommendations in a report. Hypothesis testing involves given steps that must be duly followed.

The first and initial step in hypothesis testing is hypothesis statement. After hypothesis is properly stated, the next step is to determine the best test that can best bring out the truth value of the hypothesis (Stuart A., 2009).

A rejection region or acceptance region is then set based on the significance level and the nature of the hypothesis- whether the hypothesis is one sided or two sided. A statistic or the probability value is then determined by actually conducting the test. Finally, decision is made based on the comparison between the probability value and the level of significance (Jackson, 2017).

  To begin with the initial step of testing hypothesis, hypothesis is tested in pairs. A null and alternative must be stated. An alternative hypothesis is viewed as the researcher’s point of view concerning the research topic. We may also say that an alternative hypothesis is the researcher’s original belief. An alternative hypothesis on the other hand is a statement that counters alternative hypothesis. Thus a null hypothesis is a statement that negates the null hypothesis (Stuart A., 2009).

In testing hypothesis, a null hypothesis must be stated first then followed by an alternative hypothesis. A null hypothesis has a denotation of H0 while an alternative hypothesis has a denotation of H1. In our case, the aim is to establish whether there is a relationship between overall attitude and the seven levels of choices plus age. In this case we attempt to ask the question: Is there any relationship between overall attitude and the seven levels of choices plus age? From this question, the following hypothesis is derived.

H0: There is no association between overall attitude and the seven levels of choices plus age.

H1: There is association between overall attitude and the seven levels of choices plus age.

As stated above, this hypothesis is tested by regression analysis. The test is done at 5% level of significance. This implies that the null hypothesis is rejected when the probability value is below 0.05. This is called the rejection region (Lind, 2008).

Probability value is 0.0087, and it is less than the probability value which is 0.05. This implies that we reject the null hypothesis that there is no association between overall attitude and seven choices levels plus age (Jackson, 2017). We make the conclusion that indeed there is a relationship between overall attitude and the seven levels of choices plus age (Jackson, 2017).

Having established that indeed there is a relationship between overall attitude and the seven levels of choices plus age, we can now model the regression equation that fits the outcome. This is the model that will enable us to write the overall attitude in terms of the seven levels of choices plus the age. The general form of the model is given by the following equation:

Overall attitude = B0 + B1 (Convenient Location) + B2 (Internet Connect) + B3 (Staff service) + B4 (Business Facility) + B5 (Clean room) + B6 (Bed/mattress/pillow) + B7 (Secretarial Service) + B8 (Age).

Where B0 is a constant representing the regression coefficient between the overall attitude and the independent variables. B1, B2, B3, B4, B5, B6, B7 and B8 representing the regression coefficient between the overall attitude and each of the respective independent variables. The values of these constants can be read directly from the table below.

From the table below, B0= 3.153, B1= -.251, B2= -038, B3= -.024, B4= -.123, B5= -0.058, B6= -0.238, B7= 0.047 and B8= 0.06. This implies that the resulting regression model is given by;

Overall attitude = 3.153 + 0.093 (Convenient Location) + -0.005 (Internet Connect) +0.064 (Staff service) + 0.043 (Business Facility) +-0.058(Clean room) + -112(Bed/mattress/pillow) + 0.061(Secretarial Service) + -0.05(Age).

In simple terms, given the seven different choices and age of an individual, the model above can be used to determine the overall attitude of the customer by correctly substituting the values.

Coefficientsa
Model	Unstandardized Coefficients	Standardized Coefficients	t	Sig.
B	Std. Error	Beta
1	(Constant)	3.153	.324		9.735	.000
Choice_01	.093	.022	.266	4.167	.000
Choice_02	-.005	.026	-.013	-.197	.844
Choice_03	.064	.026	.165	2.479	.014
Choice_04	.043	.017	.168	2.577	.011
Choice_05	-.058	.043	-.109	-1.354	.177
Choice_06	.112	.086	.260	1.306	.193
Choice_07	-.061	.081	-.143	-.745	.457
Age	.005	.008	.033	.544	.587
a. Dependent Variable: OveralAttitudel

Multi- collinearity in a regression model exists when two or more of the independent variables are highly correlated (Jackson, 2017). That implies that multi- collinearity exists when two or more variables have got a higher or larger correlation coefficient (Jackson, 2017). Multi- collinearity problem occurs when the estimates of independent variables are considered to statistically biased, inconsistent and inefficient in explaining the dependent variables (Jackson, 2017).  This is an implication that multi- collinearity problem exists when the estimates of the independent variables are considered to be statistically insignificant (Jackson, 2017).

From the regression table above, we can tell whether the estimates of the explanatory variables are insignificant by checking significance values (sig.) in the last column. A value is considered insignificant if it is more than value of the significance level which is 0.05 in our case (Jackson, 2017). Therefore, from the output above, it is clear that the significance values of explanatory variables are more than 0.05. This implies that these display multi- collinearity problem.

Conclusions

In The purpose of this report was to investigate the satisfaction factors to resort and the overall customer satisfaction based on the feedback of the regular customers of the resort (Kapustina & Babenkova, 2010). To achieve this objective, two tests or analyses have been done, correlation analysis and regression analysis on different variables.

Correlation analysis has been done on the three levels of attitudes, attitude 1, attitude 2 and attitude 3. From the results and findings in the previous sections, it clear that there is s general correlation or relationship among the three levels of attitudes. This implies that the management of the resort should take a keen note on dealing with the three attitude levels (Miles & Samantha, 2011).

The need to increase or decrease any given level of attribute can be guided by this findings. For instance, given attitude 1 and the management would wish to improve or increase the level of attitude 2, the resort should have measures that can improve or increase the existing attitude 1. This is simply because, the results and findings enables us to see that any increase in attitude 1 will cause a corresponding increase in the level of attitude 2. This is the same case for improving attitude 3 given attitude 1 due to the positive correlation that exists. On the other hand, given attitude and the management would wish to improve on attitude 3, the resort can think of measures that would decrease attitude 2. This is due to the negative relationships that exists.

Another correlation analysis was done among the three levels of attitude and the size of the family. It is again clear that there is a general relationship between the three levels of relationships and the size of the family. The relationship is a positive one. It is recommended that the management for any increase in the size of the family, the management should put measures to improve on the desired attitude since there is a direct positive relationship (Jing , et al., 2013).

Lastly, there is a regression analysis done to establish whether there is any relationship between the seven levels of choices, age and the overall satisfaction. The results clearly show that there is a general relationship. This is an indication that given the age of the customers and the seven levels of choices, the management can easily model the level of customer satisfaction (Naumenko & Tyutyunik, 2016).

References

Chesnokova, A. V., Barvenko, O. G., & Bradina, O. I. (2013). Identification Mechanisms of Contact Points with a Client as tools of Improvement of customer satisfaction. Marketing, 7-25.

Doda, B., Gennaioli, C., Groundson, A., & Sullivan, R. (2014). Enhancement of Customer Relationship Management through Web Social Network Integration. Customer Service Journal , 2-6.

Frankfort-Nachias, C. &.-G. (2015). Social Statistics for a diverse society. Thousand Oaks, CA: Sage Publications.

Jackson, S. M. (2017, November 24). Methods . Retrieved from [email protected] Colorado State University Department of English: https://writing.colostate.edu/research/stats/

Jing , W., Michael, S., Scott, C., Hanxi, Z., Gregor, V., & Bill, S. (2013). Customer-managed end-to-end lightpath provisioning. Part 15, 7-9.

Kapustina, L. M., & Babenkova, A. V. (2010). Assessment of Business Customers Satisfaction With the Product and Service of Pnevmostroimashina, JSC, on the B2B Market. Marketing, 2-4.

Lee, Eun, Y., Park, & Chan, S. (March, 2014). Does advertising exposure prior to customer satisfaction survey enhance customer satisfaction ratings? Marketing Letters, 11.

Lind, D. A. (2008). Statistical Techniques in Business & . Boston.: McGraw-Hill Irwin.

Miles, & Samantha. (2011). Stakeholder Definitions: Profusion and Confusion. EIASM 1st interdisciplinary conference on stakeholder, resources and value creation, IESE Business School. Barcelorna.

Naumenko, M. O., & Tyutyunik, A. M. (2016). Ways to provide quality customer Service in the business of restuarant management . Customer Relationships Journal, 2.

S, H. H. (2008). Developing an index for online customer satisfaction: Adaptation of American Customer Satisfaction Index. Marketing Letters, 10.

Shestakov, A. L., Sidorov, A. I., Shefer, L. A., & Gichkina, E. V. (2010). A quality management system for an institution of higher education: analysis of customer satisfaction. 

Stuart A., O. K. (2009). Kendall’s Advanced Theory of Statistics: Volume 2A- Classical Inference & the linear Model. 

Verhoef, P. C., Reinartz, W. J., & Krafft, M. (2010). Customer Engagement as a New Perspective in Customer Management. Journal of service, part 13, 2-4.