Hypothesis Not Supported? Help Please

[Spreadbury]

if a person's study is statistically insignificant, does that mean their hypothesis isn't supported? (Question 22 of 2008 Neap, Multiple Choice)

[Gloamglozer]

So if someone's name isn't "Glockmeister", they can't help you? 

[Spreadbury]

well it's usually glockmeister chiming in with "in Uni psych...." which is probably more useful here. of course people other than glockmeister are welcome to help. I shall change the title.

[Gloamglozer]

well it's usually glockmeister chiming in with "in Uni psych...." which is probably more useful here. of course people other than glockmeister are welcome to help. I shall change the title.

How about now?

1.  What do you mean by "statistically insignificant"?
2.  Do you do learn about the null hypothesis in VCE psychology?

[Spreadbury]

the null hypothesis wasn't in the grivas textbook to my knowledge, so no I don't know it

the precise question was:
For research on the effects of 'memad' pill on memory performance, the calculated p value was greater than 0.05. The correct statistical conclusion would be:
A the results are statistically significant, therefore the hypothesis is supported
B the results are statistically significant, therefore the hypothesis is not supported
C the results are statistically insignificant, therefore the hypothesis is supported
D the results are statistically insignificant, therefore the hypotheis is not supported

[Gloamglozer]

Since the p value is greater than 0.05, this means that the sample mean has a low probability.  This means that there is evidence for an effect.  So that means that the hypothesis is supported.  C?

I'm not sure about this one.  I've just learnt this this semester and I haven't started studying...

[Spreadbury]

psychological standard for a statistically significant result has to be LESS than 0.05

[milkcarton]

Hi Spreadbury,
                    What was the hypothesis originally? If it stated that there is no relationship between two variables (a null hypothesis) , then yes; the hypothesis is supported. If the hypothesis suggested there was, I would personally say no. In both instances though, the results are statistically insignificant. I assume you don't have the solutions for the exam?

I hope that helped in some way.

Edit: Actually, reflecting back, my statement seems pretty contradictory. My apologies, we'll leave it to Grandmaster Glockmeister.

[Glockmeister]

The p=0.05 is a ubiquitous standard in any field that uses p-values.

As to you're question, for the purposes of this question, the answer is D.

But the idea that p-values indicates the probability of the null hypothesis being true (or that the hypothesis is not supported) is not correct, although this is a view that a lot of psychologists (even professional researchers) seem to misunderstand, which has been severely criticised by many psychologists (Cohen, 1994). If the p-value is greater than .05, strictly speaking you can only fail to reject the null hypothesis. You can't say anything about the truth of the null hypothesis at all.

References
Cohen, J. (1994). The earth is round (p < .05). American Psychologist, 49, 997-1003

[Spreadbury]

the null hypothesis was not mentioned in the question, so is the answer still D if it wasn't a null hypothesis?

[Glockmeister]

The null hypothesis is the negative proposition of the hypothesis. So if an alternative hypothesis is x will happen, the null hypothesis is that x will not happen.

[Spreadbury]

ok so if the hypothesis was "the memad memory performance pill will improve recall from a list of 15 words" and the p value was p>0.05, would the answer to this question still be D? (sorry to be insitent upon the same question)

[Glockmeister]

Yep pretty much.