Hi,

your problem is beyond elementary statistical level. You should contact some people of the Statistical Department of your U, preferably some one familiar with Likert Scales and may be IRT (Item Response Theory).

If you don't have a strong back up in the field you will go astray.

As I see your study, you have conducted an opinion survey asking about people's intentions. You can't prove your hypotheses. That's not what statistical analysis does.

You can disprove the null hypothesis, though, which, I assume, would be that agreement with your statements would be randomly distributed amongst your choices (some would, some wouldn't but the responses would be equally distributed).

If this is a correct assumption, then I suggest a simple chi-square test comparing your response distributions vs equal numbers in each of the cells. I assume you have a large enough sample to answer this question. Consult with your advisor regarding the need for a power analysis, what your null hypothesis should be, and whether you need to correct for multiple comparisons.

Now if all of your hypotheses basically test the same assumption, you could compute the sums of each of your statements, compute the internal consistency across your 5 statements to get a reliability coefficient and, if the reliability is high (say >0.7) then you could compare the sum to your null hypothesis using just one chi-square test.

Good luck with your study

In short, you could first conduct factor analysis and "save" the variables, and then use linear regression!