# Thread: Simple Null Hypothesis Question

1. ## Simple Null Hypothesis Question

Ok this one is very simple I've no doubt but in my cabin fever (i'm at work at the moment to boot), I've confused myself.

The installation supervisor claims that the mean number of days between receipt of a complaint and resolution of the complaint is 20 days or less. If it is established that this claim is not true, the company will invest in a major overhaul of the organisation of the flooring department. Formulate null and alternative hypotheses regarding the data, and explain why the choice of null and alternative hypotheses is appropriate.

Now i was thinking

Null Hypo is (H : mu = 20) now thats corrent.

But my alternative Hypothesis was (H1: mu < 20)

Thoughts? I always get confused with the basics of setting a null and alt hypothesis.

Also in part B it goes on to ask

We wish to investigate whether there is sufficient evidence to conclude at the 5% level that the mean waiting time is longer than 20 days. Specify the formula for the test statistic that should be used, explaining the meaning of any symbols you use.

So based on part B i'm thinking that in part A the alternative Hypothesis is infact > 20 rather than < 20

2. Originally Posted by angelsu512
Ok this one is very simple I've no doubt but in my cabin fever (i'm at work at the moment to boot), I've confused myself.

The installation supervisor claims that the mean number of days between receipt of a complaint and resolution of the complaint is 20 days or less. If it is established that this claim is not true, the company will invest in a major overhaul of the organisation of the flooring department. Formulate null and alternative hypotheses regarding the data, and explain why the choice of null and alternative hypotheses is appropriate.

Now i was thinking

Null Hypo is (H : mu = 20) now thats corrent.

But my alternative Hypothesis was (H1: mu < 20)

Thoughts? I always get confused with the basics of setting a null and alt hypothesis.
I would have thought that appropriate hypotheses for this problem would be:

(H0: mu<=20), (H1: mu>20).

RonL

3. unfortunately no.

The null hypothesis is always = to the number specified.

The alternative hypothesis can either be

< or > or that funny = sign with a / line through it.