Annotated Steps for Hypothesis Testing

Research question? __________________________________________________________

1. Restate research question as two hypotheses: HA & H0 (our research hypothesis and our null hypothesis). Note: HA is what we are actually hypothesizing. H0 is our "dummy" null hypothesis that states, in effect "HA is not true."

HA :

H0 :

2. Determine characteristics (mean, standard error) of sampling distribution (the comparison distribution). Note: For the sampling distribution, assume the null hypothesis is true. For this week's purposes, it is the distribution of sample means. It is always a distribution of statistics.

Mean ____

Standard error ____

3. Find critical value (s) on the sampling distribution. Note: To do this, you need to pick a significance level (alpha) and decide whether to use a two-tailed or one-tailed test. Then check the normal curve table. If you are a visual thinker, sketch the curve, mark where critical value(s) fall, and shade the areas where you can reject the null. If you are a numerical thinker, just write down the Z-score you have to beat, and be sure you specify +, -, or +/- (two tailed).

Significance ____

Number tails ____

Critical value(s) ____

4. Calculate statistic, then turn it into a Z-score to see where it falls on the sampling distribution. Note: For Chs. 5&6, this is the sample mean. Statistics are always calculated using...the sample. But to get the Z, you use the mean and standard error of ... the sampling distribution.

Statistic _____

Z-score _____

5. Compare 3 & 4 and make decision about H0 : Reject or retain? What's the answer to your research question? Note: if you are a more visual thinker, look to see if your statistic falls in a shaded region. If you are more numerical, see if the Z-score exceeds the critical value(s).