# Thread: help: what statistical test do I have to use?

1. ## help: what statistical test do I have to use?

Hello

I am writing my research proposal and I am using a questionnaire to find out what is the perspective of experienced occupational therapists regarding how newly qualified occupational therapists perform in the workplace. I do not have an hypothesis I simply want to find out what I have just highlighted above. I will be getting ordinal data. Now, because I don't have an hypothesis can I still use a statistical test? If so, which one?
Thank you

2. ## Re: help: what statistical test do I have to use?

The statistical tests are dictated by your data and your research questions/hypothesis, so I'd say in the absence of either of these determining a test is not possible. I mean the fact that you have ordinal data brings to mind some possibilities but without knowing more about the data and questions it (IMHO) is not possible.

3. ## Re: help: what statistical test do I have to use?

Why are you doing the research if you don't have a hypothesis?

4. ## Re: help: what statistical test do I have to use?

I am sure you have realised that I do not know anything about statistics and so on. The reason why I don't have an hypothesys is because I am trying to find out an opinion about something, I am not hypothetising about anything. Do you think that for this reason I could say that I am simply describing something and therefore a descriptive analysis should be enough. I mean, I know that with ordinal data you would use a non-parametric test but wouldn't I need varaiables to use these tests? and the variables would be in my hypothesis. But the aims of my study are to explore what is the perspective of experienced occupational therapists regarding how newly qualified occupational therapists perform in the workplace with a focus on graduates performance, preparation to practice, and factors they think assist or hinder their performance. I will be using a questionnaire with mainly closed-ended questions and few open-ended questions at the end of the questionnaire with demographichs. they will have to rate their answers on a 5 point likert scale. I don't know...........my brain is going to explode soon.

5. ## Re: help: what statistical test do I have to use?

Hello Junior,

You want to know 'what is the perspective of experienced occupational therapists regarding how newly qualified occupational therapists perform in the workplace'. I am not so sure what do you mean but since you are asking for what and how. I believe you can interview those experienced occupational therapists by asking them an open question like 'what is your perspective regarding how newly qualified occupational therapists perform in the workplace' and then try to find out if the people you interview have similar opinions or not. For that, you would have to set a criteria so you can summarize with a single word the explanation they will give you.

Say, for instance: excellent, very good, good, just OK, poor.

That is called a 5 point likert scale and it is widely used to ask for opinions in questionnaires and so.

Once you have the data from the people you interview. You can compute the means and standard deviations of the overall answers. In that way, you can see what is the overall perspective of the professionals you interviewed. You can do this with SPSS by going to analyses, descriptives. Also, you can do it with Excel.

I think that is pretty much what you can do... If you had more questions then you could run inter-rater and reliability statistics such as the Chronbach's Alpha and the Intra-rater correlations.

You may want to have a look at these links:
http://en.wikipedia.org/wiki/Likert_scale
http://en.wikipedia.org/wiki/Cronbach's_alpha
http://en.wikipedia.org/wiki/Intraclass_correlation

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts