We have all conducted an investigation of sorts which involves us spending lovely amounts of time constructing a suitable hypothesis for us to test before we start rounding up our participants and having them do lovely tasks in the ‘name of science!’.
But why do we have to include a null hypothesis? Its purpose is to state that between specific populations there will be no different between them, but that should be the job of the hypothesis you lay out when constructing your study. It is my personal opinion (not the opinion of science) that the null hypothesis is just a tactic for researchers to save face in front of the rather large scientific community if their investigation goes pear-shaped.
The idea of saving face is based around the principle of losing perceived dignity and prestige to your peers; this was explored by a researcher called Eberhard (1967) who looked at this phenomenon who demonstrated the seriousness of losing face in Chinese society, which is a big deal, but this has been shown to be cross-cultural and can be seen in everyday life when we make a little mistake and prefer to cover it up rather than tell every (from as simple as spilling the milk to running away when you accidently push a friend over and they suffer a rather nasty injury).
My point is that scientists use a null hypothesis to save a little face to the scientific community by having their main hypothesis (X is more effective than Y) and then a hypothesis that covers them if the results isn’t as desired (X and Y are the same), as spending a substantial portion of their life conducting studies which yield little/nothing is very disheartening, and then having it mocked or ignored by your peers is a rather unpleasant and degrading thing to have occur. It is just an ego boost if they do make get nothing so they can say ‘well our null hypothesis was correct’. Which is pleasant, but science is far more cutthroat with so many people competing to ‘demonstrate the greater effect’.
So all-in-all the null hypothesis effect I deem to be a feature to sooth the egos of researchers so that if they make a mistake in their prediction of what they are looking for, as they are less likely to hide the study away after investing so much time. Especially as there have been a large quantity of studies hidden away because researchers feel ashamed for getting a poor result, which wont advance the course of science (and that’s a terrible, terrible thing).
References:
Eberhard, Wolfram. (1967). Guilt and Sin in Traditional China. University of California Press.
prpdh 
I disagree with your blog. You claim the null hypothesis is used to sooth the egos of researchers but I think it has a bigger role to play than that. The null hypothesis helps us see that our research is highly likely to be true. In research you need to carry out a statistical hypothesis test. Without the null hypothesis you would not be able to do this. During the statistical hypothesis test you aim to reject the null hypothesis rather than proving the hypothesis is true, (Howitt & Cramer, 2011). I argue in my blog post (http://psuc28.wordpress.com/) that you can never prove a hypothesis. I used the example of gravity, how can you test every atom that is in the universe to prove that gravity is universal? You can’t (yet!), therefore you need the null hypothesis.
————–
Howitt, D. & Cramer. D. (2011). Introduction to Research Methods in Psychology, third edition. Pearsons Education.
Pingback: Homework For My TA- Week 4 Comments « All About Psychology
Researchers have a null hypothesis because a hypothesis can never be proved. With a null hypothesis it says that no difference will be found which means there is two possible outcomes…there is either a difference or there isn’t. An importance of the null hypothesis is that it is falsifiable….you can never say that all pigs have curly tails because you will never see every pig in the world. However it only takes seeing one straight tailed pig to falisfy the theory. Researchers look to reject the hypothesis because proving it is near enough impossible!
Is there such thing as nothing? The null stats that nothing will happen which is always the possibility in scientific research despite how unlickly it may seem. Although few care about whether nothing will happen if i put a mouse in a maze(and highly unlikely -most likely a fluke of an un-usually smart-stubbed mouse) – i agree with the cutthrough attitued to get an effect but to be objective/falifyable we must accept this as a posiblity. Moreover, it would be strange that as scientist (well on day!) that we ignore the possibility of nothing. It also stops us having to choose between a positive effect or an negative significant effect because the possibility that there’s nothing.
Furthermore, as you can use a null hypothesis just to backup you hypothesis and say that there is an effect since there is no null effect (see my blog for more info on this ).
Pingback: Some homework for my TA— so far « whataloadofblog
Pingback: Homework for my TA :) « prpnw
I see your point however this is SCIENCE, surely a scientist’s first objective is to further knowledge rather than make themselves look good?? The real role for a null hypothesis is to make sure that your results aren’t produced by chance or by something else other than the variable you are testing. In order to back up your hypothesis you need to compare the results against the opposite situation. In statistics, the only way of supporting your hypothesis is to refute the null hypothesis. Rather than trying to prove your theory (the alternate hypothesis) right you must show that the null hypothesis is wrong. Therefore you have to assume that your alternate hypothesis is wrong until you find evidence to the contrary.
This is not done because scientist don’t want to look the fool but in the pursuit of making their work objective and scientific, this was the intention anyway, of Ronald Fisher in 1935 when first coined the phrase null hypothesis.
Pingback: Homework for my TA (week 11) « prpsjj