If I have posted this in the wrong forum, please forgive and move to the correct one.

Hi everybody. this is my first time posting.

I currently have a project for which I'd like to brush off my Stat and SPSS skills but I don't want to make a fool of myself so I need some help on how to set this sort of a task up.

I am attempting to show that when the number of American troops stationed in an autocratic country increases, what follows is the autocracy becoming a member of more Intergovernmental organizations where the majority of members are Democratic states. Then, with time, that autocracy becomes increasing democratic, presumably because it has more ties with democracies.

To Illustrate the Relationship is supposed to be something like this:
Increased TROOP LEVELS ----> Increased participation in IGOs formed by Democracies ----> Increased Democracy in Host Country

So thus far I have four pieces of data to work with:
1) Year
2) U.S. Troops Levels
3) Polity IV score (this is a data set that assigns each state a democracy score)
4) IO score (this is a data set that assigns each state a "network dependency" score for each state in relation to IGOs formed by Democracies).

Because I want to show that this sort of an increase in Democracy occurs because of an increase in U.S. troops in an Autocracy, I want to "control" (is that the right word?) for GDP per capita since increases in that are commonly associated with increase levels of Democracy.

So all in all I have five variables.

So in this case, I am assuming that my Dependent Variable is the "Polity" (democracy) score. My independent variables are U.S. troop Levels and IO score. Is that correct?

Now my problem is understanding how I can demonstrate that the One (i.e. "Troop Levels") leads to the other (i.e. Increased IO score) which then leads to another (i.e. increased Polity scores aka Democracy). Should I run two different sets of analyses, one in which I only only U.S. troops levels an an independent variable and one where I oly use IO score as an independent variable and then compare the trends? Or is there a way that I can incorporate both independent variables simultaneously in whatever analyses that I do?

Furthermore, as silly as this sounds, where does "Year" fall? Is this considered a variable at all? If so, what type?

I hope that I have not written things in a confusing way. If I have, I'd be more than willing to clarify what I mean.

If anyone could lend me some assistance, I would appreciate it.

Thanks in advance.