Women who smoke are as likely to have low birth wt babies as women who don smoke: is this a null hypothesis?
A null hypothesis states there is no difference… so in your case the null hypothesis is “there is no difference in birth weight of babies from women who are smokers compared to women who are non-smokers” Then, by statistical test you either reject the hypothesis (and suggest another), or accept the hypothesis. You don’t prove a null hypothesis. The way you have phrased the question is to look at only one category of babies… those with low birth weight… and not all babies.
Correct. “Null hypothesis” means “no difference”. This is how you would propose the null hypothesis for a test between smoking and non-smoking pregnant females. You would collect the birthweight data (Dependent variable), use smoking status as the grouping variable. If there was a significant difference in birthweights, then you would reject your null hypothesis if the alpha (p-value) of your test was less than 0.05.
