How to analyze changes over time statistically

Question

I have narcotic prescription rates for each states for the last 10 years. I want to assess changes over time, but am unsure how to approach the data. I want to know what the general trend is, then divide the data into groups to compare whether certain programs have a greater effect on prescription rate decline.

I did an ANOVA, and found that there is a statistically relevant change between states that mandate prescriber access and those that don't. How do I evaluate this statistical difference?

I'm in grad school but never took a stats class (because my bachelor's is in writing). I'm really feeling out of my depth, so any help would be greatly appreciated.

Answer 1

Hi Katie,

Without having seen the data I would plot (scatter plot in excel) the number of prescriptions per year for each state vs time and examine the resulting graphs. You may not want to plot them all initially. Just pick 4 or 5. If the plots "look" linear then I would run a linear regression model on each data set. If the plots suggest a non linear trend I would an exponential model. One of these will provide the general trends for the states.

The ANOVA provides evidence that the means differ (or not) between the two categories. An the standard ANOVA output provides a critical F value (alpha=.05) which determines if the null hypothesis should be rejected or not.

Let me know if this is enough to get you started.

Jim