Statistical Evaluation of Choosing Phone Provider - Case Study Example

Introduction The main objective of the study was to evaluate different factors and find out if it has an influence on the decision of the consumers on the choice of the phone provider. These factors included text messaging, emails, MMS pack, internet, camera quality GPS accuracy and age. Also the study was to determine the relationship between the satisfaction level of people from rural and urban areas as well as gender in relation to different factors mentioned above. The evaluation was carried out through inferential statistics that include t test and regression analysis. There were several hypotheses as shown below and all were tested. Hypothesis To determine if there is a significant difference between the satisfaction level of male and female on the choosing of the service provider To determine if there is a significant difference on the way people are satisfied between rural and urban areas To determine how text messaging, emails, MMS pack, internet, camera quality GPS accuracy and age influence the decision of the consumer on choosing the phone service provider Descriptive Statistics N Minimum Maximum Mean Std. Deviation v01 70 1 70 35.50 20.351 Test Messaging 69 1 5 4.17 1.028 Emails 70 2 5 4.19 1.040 MMS pack 70 2 5 4.19 1.067 Internet 69 1 5 3.78 1.223 Camera Quality 69 1 5 3.97 1.260 GPS Accuracy 70 1 5 3.69 1.234 Providers 70 1 3 1.96 .842 Ages 70 12 59 32.24 11.034 Gender 70 1 2 1.37 .487 Location 70 1 2 1.54 .502 Valid N (listwise) 68 Table 1. Descriptive statistics. As indicated in table 1 above we can see that the mean age of those who participated is 32.24.The text messaging, emails and MMS pack had the highest level of ranking of consumer satisfaction, in relation to the choosing of a mobile phone service provider. Inferential statistics. In order to test the hypothesis of whether there is a significant difference between the satisfaction level of male and female on the choosing of the service provider, I did t test to ascertain this as demonstrated in the next section. Independent sample t test for gender difference of satisfaction level Group Statistics Gender N Mean Std. Deviation Std. Error Mean Test Messaging 1 44 4.20 1.002 .151 2 25 4.12 1.092 .218 Emails 1 44 4.23 1.054 .159 2 26 4.12 1.033 .202 MMS pack 1 44 4.20 1.025 .154 2 26 4.15 1.156 .227 Internet 1 44 3.55 1.266 .191 2 25 4.20 1.041 .208 Camera Quality 1 43 4.00 1.234 .188 2 26 3.92 1.324 .260 GPS Accuracy 1 44 3.48 1.285 .194 2 26 4.04 1.076 .211 Providers 1 44 2.05 .806 .121 2 26 1.81 .895 .176 Table 2.Group statistics Independent Samples Test Levenes Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Test Messaging Equal variances assumed .100 .753 .326 67 .745 .085 .259 -.433 .602 Equal variances not assumed .318 46.493 .752 .085 .266 -.450 .619 Emails Equal variances assumed .182 .671 .432 68 .667 .112 .259 -.404 .628 Equal variances not assumed .435 53.458 .665 .112 .257 -.404 .628 MMS pack Equal variances assumed .266 .608 .191 68 .849 .051 .266 -.480 .581 Equal variances not assumed .185 47.652 .854 .051 .274 -.501 .602 Internet Equal variances assumed 2.705 .105 -2.196 67 .032 -.655 .298 -1.250 -.060 Equal variances not assumed -2.318 58.308 .024 -.655 .282 -1.220 -.089 Camera Quality Equal variances assumed .880 .351 .244 67 .808 .077 .315 -.552 .706 Equal variances not assumed .240 49.960 .811 .077 .321 -.567 .721 GPS Accuracy Equal variances assumed 3.158 .080 -1.871 68 .066 -.561 .300 -1.160 .037 Equal variances not assumed -1.959 60.065 .055 -.561 .286 -1.134 .012 Providers Equal variances assumed 2.214 .141 1.145 68 .256 .238 .208 -.177 .652 Equal variances not assumed 1.114 48.229 .271 .238 .213 -.191 .667 Table 3.Independent sample t test According to the t-test statistics, it is observed that the calculated t test for GPS accuracy, camera and internet have t values greater than the critical values. In this case we will reject the null hypothesis and conclude that there is a statistically significant difference between the level of satisfaction of male and female. On the other hand, text messaging emails and MMS pack has a calculated t test that is less than t critical value. In this case, we will fail to reject the null hypothesis and conclude that there is no enough evidence to prove that there is a difference in the satisfaction level between male and female Independent sample t test for location (Rural and urban) difference of satisfaction level Group Statistics Location N Mean Std. Deviation Std. Error Mean Test Messaging 1 31 4.03 1.169 .210 2 38 4.29 .898 .146 Emails 1 32 4.25 1.016 .180 2 38 4.13 1.070 .174 MMS pack 1 32 4.00 1.218 .215 2 38 4.34 .909 .147 Internet 1 31 3.97 1.140 .205 2 38 3.63 1.282 .208 Camera Quality 1 32 3.75 1.391 .246 2 37 4.16 1.118 .184 GPS Accuracy 1 32 3.72 1.143 .202 2 38 3.66 1.321 .214 Providers 1 32 1.97 .861 .152 2 38 1.95 .837 .136 Table 4.Group statistics for location(Rural and Urban) Independent Samples Test Levenes Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Test Messaging Equal variances assumed .705 .404 -1.034 67 .305 -.257 .249 -.754 .239 Equal variances not assumed -1.007 55.422 .318 -.257 .255 -.769 .255 Emails Equal variances assumed .031 .860 .472 68 .638 .118 .251 -.382 .619 Equal variances not assumed .474 66.985 .637 .118 .250 -.380 .617 MMS pack Equal variances assumed 3.595 .062 -1.344 68 .183 -.342 .255 -.850 .166 Equal variances not assumed -1.311 56.472 .195 -.342 .261 -.865 .181 Internet Equal variances assumed .700 .406 1.138 67 .259 .336 .295 -.253 .926 Equal variances not assumed 1.152 66.475 .254 .336 .292 -.246 .919 Camera Quality Equal variances assumed 4.047 .048 -1.364 67 .177 -.412 .302 -1.015 .191 Equal variances not assumed -1.342 59.366 .185 -.412 .307 -1.027 .202 GPS Accuracy Equal variances assumed 1.487 .227 .204 68 .839 .061 .298 -.534 .656 Equal variances not assumed .207 67.942 .837 .061 .294 -.527 .648 Providers Equal variances assumed .071 .790 .105 68 .917 .021 .203 -.384 .427 Equal variances not assumed .105 65.310 .917 .021 .204 -.386 .429 Table 5. Independent sample t test for Rural and urban areas. Basing on the t test statistics as shown in table 5.It is clearly evident there is a significant difference of the way people are satisfied between rural and urban areas on camera, internet, MMS pack and text messaging. This is indicated by the calculated t value which is greater than the t critical value. Therefore we reject the null hypothesis which notes that there is no difference between the way people are satisfied in rural and urban areas. On the other hand, the results shows that there is no statistically significant difference on the way people are satisfied on GPS accuracy and emails between rural and urban areas. This is clearly shown by the t calculated value which is less than the t critical value. In this case, we fail to reject the null hypothesis and conclude that there is a statistically difference on the level of satisfaction between rural and urban people in relation to the choosing of service providers. Regression Model 1 Regression analysis was run to ascertain how the different factors influence the decision of the consumer on choosing the phone service provider. Providers, Test Messaging, Emails, GPS Accuracy, Camera Quality, Internet and MMS pack were included in the model as below to see how they can influence the consumer decision on choosing the phone service provider Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Providers, Test Messaging, Emails, GPS Accuracy, Camera Quality, Internet, MMS packb . Enter a. Dependent Variable: v01 b. All requested variables entered. Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .232a .054 -.057 20.720 a. Predictors: (Constant), Providers, Test Messaging, Emails, GPS Accuracy, Camera Quality, Internet, MMS pack ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 1466.317 7 209.474 .488 .840b Residual 25759.213 60 429.320 Total 27225.529 67 a. Dependent Variable: v01 b. Predictors: (Constant), Providers, Test Messaging, Emails, GPS Accuracy, Camera Quality, Internet, MMS pack Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 45.196 24.108 1.875 .066 Test Messaging 2.932 3.141 .150 .934 .354 Emails -.320 2.820 -.017 -.114 .910 MMS pack -4.870 3.424 -.260 -1.422 .160 Internet .809 2.728 .049 .297 .768 Camera Quality .479 2.670 .030 .179 .858 GPS Accuracy -.032 2.691 -.002 -.012 .991 Providers -2.470 3.975 -.103 -.621 .537 a. Dependent Variable: v01 According to the regression analysis above, it is observed that emails, MMS pack GPS accuracy and service providers have a negative influence on the consumer decision. On the other hand, Test messaging, internet camera quality has a positive influence on the consumer decision. Model 2 Age is taken as the independent factor alone in the model to see how it influences the consumer decision on the choosing of phone provider. Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Agesb . Enter a. Dependent Variable: v01 b. All requested variables entered. Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .257a .066 .052 19.813 a. Predictors: (Constant), Ages ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 1883.198 1 1883.198 4.797 .032b Residual 26694.302 68 392.563 Total 28577.500 69 a. Dependent Variable: v01 b. Predictors: (Constant), Ages Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 20.234 7.361 2.749 .008 Ages .473 .216 .257 2.190 .032 a. Dependent Variable: v01 Basing on the results above, it is clearly evident that age has a positive influence on the consumer decision of choosing the phone service provider. Model 3. Emails and internet was included in the model to see how they influence the consumer decision on choosing of phone service provider. Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Internet, Emailsb . Enter a. Dependent Variable: v01 b. All requested variables entered. Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .073a .005 -.025 20.597 a. Predictors: (Constant), Internet, Emails ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 150.877 2 75.438 .178 .837b Residual 28000.282 66 424.247 Total 28151.159 68 a. Dependent Variable: v01 b. Predictors: (Constant), Internet, Emails Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 34.239 11.277 3.036 .003 Emails -.897 2.549 -.046 -.352 .726 Internet 1.244 2.173 .075 .573 .569 a. Dependent Variable: v01 In regards to the above regression analysis, it is clearly evident that emails have a negative influence on the decision of choosing phone service provider while the internet has a negative influence. This does not change when it is related to the first model but the coefficients are relatively higher in the current model the model 1. Discussion According to the results, it is clearly evident that text messaging, emails and MMS pack had the highest level of ranking of consumer satisfaction in relation to the choosing of a mobile phone service provider. In regards to the inferential statistics, According to the t-test statistics, it is observed that the calculated t test for GPS accuracy, camera and internet have t values greater than the critical values. In this case we will reject the null hypothesis and conclude that there is a statistically significant difference between the level of satisfaction of male and female. On the other hand, text messaging emails and MMS pack has a calculated t test that is less than t critical value. In this case, we will fail to reject the null hypothesis and conclude that there is no enough evidence to prove that there is a difference in the satisfaction level between male and female It is clearly evident there is a significant difference of the way people are satisfied between rural and urban areas on camera, internet, MMS pack and text messaging. This is indicated by the calculated t value which is greater than the t critical value. On the other hand, the results shows that there is no statistically difference on the way people are satisfied on GPS accuracy and emails between rural and urban areas. In this case, we fail to reject the null hypothesis and conclude that there is a statistically difference on the level of satisfaction between rural and urban people in relation to the choosing of service providers. In regards to the regression analysis, it was observed that emails have a negative influence on the decision of choosing phone service provider while the internet has a negative influence. It was also observed that MMS pack, GPS accuracy and service providers have a negative influence on the consumer decision. On the other hand, Test messaging and camera quality has a positive influence on the consumer decision Read More