Analysis and Interpretation

Analysis and Interpretation

Addition in the t- test explanation in section 4

Interpretation of significance

When the significance level is small (p <0.05), we can reject the null hypothesis and conclude that the two averages are not from the same population. However, there is always a chance that this conclusion is false. It is possible that the null hypothesis is true and that your results are among the possibilities unlikely. In fact, this is what tells you the degree of significance: what is the probability that a difference at least as great as that observed appear when the null hypothesis is true.

When the significance level is too high (p> 0.05) to reject the null hypothesis that means are equal, two explanations are possible:

There is really no difference between the two averages (something you cannot prove) or there is a small difference that you cannot detect;

There is a significant difference between the two groups, but you have not detected! How is this possible? One possible reason is that the sample is small and that several values ??are consistent with the null hypothesis. Therefore, the result does not appear as "unusual." For example, although there was a significant difference between users and non-Internet users, with 5 subjects in each group, you would not be able to detect it because the difference observed between the two groups can be compatible with several values ??of the population, the value 0.

The ability to reject the null hypothesis also depends on the variability of the observed values. If you have a lot of variability in the sample, the range of possible values ??for the real difference is population-wide.

One Sample T-test (Significance test for different types of calls)

Significance test for mobile

Assuming the hypothesis:

H0: The variables taken for the usage of BSKYB mobile are not significant.

H1: The variables taken for the usage of BSKYB mobile are significant.

The results of the analysis of the variables show that the variables are highly significant as their t- values are higher than the tabulated and also t he level of significance is also less than 0.05. The results show that there is an ethnic difference found in using the mobile service of BSKYB i.e. their usage depends upon whether they are white or non - white. Another thing that is analysed in this analysis that there is a significant difference in the consumption of mobile in the urban areas as compared with the rural areas. The overall section results were significant but few questions were taken for the analysis which was significant for the study.

One-Sample Test

Test Value = 0

t

df

Sig. (2-tailed)

Mean Difference

95% Confidence Interval of the Difference

Lower

Upper

Free Time SKY

15.802

560

.000

.308

.27

.35

Contribution of population not being white British, Asian, black or Chinese category

68.025

560

.000

1.258

1.22

1.29

Contribution of population being white British category

68.895

560

.000

2.176

2.11

2.24

Urban or Rural

48.192

531

.000

.814

.78

.85

Significance test for Free phone

Assuming the hypothesis:

H0: The variables taken for the usage of BSKYB free phone are not significant.

H1: The variables taken for the usage of BSKYB free phone are significant.

The results of the analysis of the variables show that the variables are highly ...