Alternative Hypothesis

An alternative hypothesis is one in which a difference (or an effect) between two or more variables is [Page 19]anticipated by the researchers; that is, the observed pattern of the data is not due to a chance occurrence. This follows from the tenets of science, in which empirical evidence must be found to refute the null hypothesis before one can claim support for an alternative hypothesis (i.e. there is in fact a reliable difference or effect in whatever is being studied). The concept of the alternative hypothesis is a central part of formal hypothesis testing.

Alternative hypotheses can be nondirectional or directional. If nondirectional, an alternative hypothesis is tested with a two-tailed statistical test and is stated in words to the effect that “A differs from B.” If directional, an alternative hypothesis is tested with a one-tailed statistical test and is stated in words to the effect that “A is greater than B” or “B is greater than A.” (The null hypothesis is stated in words to the effect that “A equals B.”)

An example in survey research would be a split-half experiment that is used to test whether the order of two question sequences within a questionnaire affects the answers given to the items in one of the sequences, for example, in crime surveys where both fear of crime and criminal victimization experience are measured. In this example, a researcher could venture a directional alternative hypothesis that greater levels of fear would be reported if the fear items followed the victimization items, compared to if they preceded the victimization items. Half the respondents would be randomly assigned to receive one order (fear items, then victimization items), and the other half would receive the other order (victimization items, then fear items). The null hypothesis would be that the order of these question sequences makes no difference in the answers given to the fear-of-crime items. Thus, if the null hypothesis is true, the researcher would not expect to observe any reliable (i.e. statistically significant) difference in levels of fear reported under the two question-ordering conditions. In contrast, if the directional alternative hypothesis is true (i.e. if results indicate significantly greater fear being reported when the fear items follow the victimization items than when they precede them), then the null hypothesis is rejected and support is accorded to the alternate hypothesis.

Another way of understanding the alternative and null hypotheses in survey research is to think about the crime survey example and the confidence intervals that can be calculated around the fear-of-crime measures in the two conditions. The null hypothesis would be that the 95% confidence intervals for the fear measures under the two question orders would overlap and thus not be reliably (significantly) different from each other at the .05 (alpha) level. A directional alternative hypothesis that states that reported fear of crime would be higher when the victimization items precede the fear items would be that (a) the confidence intervals would not overlap and that (b) the lower limit of the confidence interval for the fear items when the victimization items precede them would exceed the upper limit of the confidence interval for the fear items when the victimization items follow them.

...