Confused about aspects of hypothesis testing

I encountered this question concerning hypothesis testing which goes as such:

At the 5% significance level, where a Z-test is used to test the null hypothesis  Ho: μ = μo against the alternate hypothesis  H1: μ < μo , it is discovered that there is sufficient evidence to suggest μ <μo ( the null hypothesis is rejected in favor of the alternate hypothesis).

If the same test is conducted at the 10% significance level, what conclusion should we expect?

Student X

The conclusion will remain the same.

Mathematically speaking, in the initial round of testing involving the 5% significance level, we can already surmise that the p value ( the area value alluding to the test-statistic) is lesser than 5%, ie p<0.05. Hence, when this significance level is widened, the p-value (note it remains unchanged) is obviously and definitely lesser than 10%, ie p<0.1. In this regard, we still accept the alternate hypothesis.

You can also consider an analogy to help you appreciate things better:

Imagine a public opinion poll conducted for a criminal trial shows that 5% of people feel the accused is guilty (they might be wrong of course, but that's besides the point), and a guilty verdict is delivered by the jury. 

If a re-surveying of public sentiment is carried out, and 10% of people now feel the accused is guilty, wouldn't the delivery of a guilty verdict be much more certain based on previous existential circumstances?

Hope this facilitates understanding. Peace.

Best Regards,

Mr Koh