## Creating data file with data from multiple years

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

rjaheiligers
Posts: 2
Joined: Thu Sep 15, 2011 8:11 pm

### Creating data file with data from multiple years

Hi!

The answer might be somewhere else, and if it is I'm sorry for opening a new topic, but I could not find it so far...

For my graduating thesis I have to work with SPSS for the first time and I have a (probably very simple) question. I have to perform a multiple regression analysis to determine the influence of several independent variables on one dependent variable. I have to perform this analysis using data from 7 years. So I have a sample of 100 firms with balance sheet data of 7 years. How do I construct my data file?

Do I have to make a separate variable for each of the 7 years (e.g. size_2001, size_2002, size_2003 etc) which still leaves me with a sample of 100 firms, but 7 times the number of variables? Do I enter each firm 7 times (e.g. Shell_2001, Shell_2002, Shell_2003) so that I have a sample of 700, but the original amount of variables?

Or is there an entirely different method of doing this?

Thanks already for the help, really appreciate it!
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

### Re: Creating data file with data from multiple years

depends on your hypothesis, but I think the first thing you describe (size-2001 etc.) is usually the way to go. The companies are the unit of analysis then.
rjaheiligers
Posts: 2
Joined: Thu Sep 15, 2011 8:11 pm

### Re: Creating data file with data from multiple years

My hypotheses are to test whether the independent variables influence the dependent variables. And I also have to see whether this changes over time.

What I've done now is indeed create each independent variable 7 times so I can perform the analysis each time with the independent variables of the year I need.

Thanks for confirming my idea!
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

### Re: Creating data file with data from multiple years

you might want to think about repeated measures of time course analysis

### Who is online

Users browsing this forum: No registered users and 1 guest