The third variable in this schematic affects the effect of the main independent variable, rather than the value of the independent and dependent variables. It is often graphically represented by a a diagram with an arrow pointing at another arrow. It means that we let the effect of one variable vary over the values of another variable. It can also be called a moderation analysis, or that we are investigating conditional effects. If we run two separate analyses, the control variables would also get different effects in the two models, and that might not be part of our theory. We can then also hold the effects of all control variables constant. We can then decide whether it is worth the effort to complicate our theoretical model by introducing these separate effects. We then get a sense of how much the effect differs between groups, and also whether the difference in effect is statistically significant. We then get the effect of becoming a parent among men, and among women.īut we can also do this in one single regression analysis. How to account for this in our analyses? One way is to run two separate analyses, one for each group we are interested in (in this case women and men). The effect of becoming a parent is for instance different among women and men women see a much larger drop in incomes. However, in the real world, we often find that relationships are different in different sub-groups of the population, or during different circumstances. Regression analysis is used to investigate relationships between variables, with or without control for other vairables. Regression analysis with interaction effects ¶
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |