A population of flights can be classified according to two factors: duration (short, medium, or long) and purpose (vacation or business). We are interested in whether either of these factors has an effect on price. The table below gives mean prices for a balanced population and shows the presence of interactions. If a study is conducted and we think there may be interactions, we should:
A)find the F test for interaction first and, if the P-value is small enough to reject the null hypothesis of no interaction, assess the effects of distance separately for business and vacation flights.
B)test for effects of distance first and, if there is no distance effect, ignore the interactions.
C)find the F tests for distance effects and for type of trip and, if there is no effect for either, declare that all the means are the same.
D)drop observations that cause the interaction and repeat the analysis.

Question

A population of flights can be classified according to two factors: duration (short, medium, or long) and purpose (vacation or business). We are interested in whether either of these factors has an effect on price. The table below gives mean prices for a balanced population and shows the presence of interactions.  If a study is conducted and we think there may be interactions, we should:
A)find the F test for interaction first and, if the P-value is small enough to reject the null hypothesis of no interaction, assess the effects of distance separately for business and vacation flights. 
B)test for effects of distance first and, if there is no distance effect, ignore the interactions. 
C)find the F tests for distance effects and for type of trip and, if there is no effect for either, declare that all the means are the same. 
D)drop observations that cause the interaction and repeat the analysis.

Quizplus · Accepted Answer

When there are interactions present, it is important to test for the interaction effect first before assessing the individual effects of the factors. This is because the presence of an interaction means that the effect of one factor on the outcome is dependent on the level of the other factor. Therefore, ignoring the interaction may lead to incorrect conclusions about the individual effects of each factor. If the P-value for the interaction test is small enough to reject the null hypothesis of no interaction, then the effects of each factor should be assessed separately.

A Population of Flights Can Be Classified According to Two