Hi everyone,
I have a data set with 55 rows (countries) and 8 Columns ("Growth on 7 different types of expenses 2019-2020" + " one for the population").
I want to see whether there is a significant difference between these 7 types of expenses while I am controlling for the population.
I know that ANCOVA might be a good option for this, however, how should I setup the table? should I create a table with 385 rows (7 type of expenses * 55 countries) then create a new column called expenses and also one for the population? then I am repeating the countries, is that okay?
Any suggestions which analysis I should use?
I can also get raw data in $, like expenses for 2019 and expenses for 2020, if that helps.
Cheers,
Hassan
Dear Hassan Kalantari ,
You can carry out an ANOVA test or just run a regression including dummies (7 dummies without a constant or 6 dummies and a constant) and test if all the dummies are simultaneously zero (null hypothesis).
Formally and in Stata:
y = D1b1 + .... + D7b7 + e
reg y D1 D2 D3 D4 D5 D6 D7, nocons
Now test
Ho: b1 = ... = b7
Ha: No Ho
test D1 D2 D3 D4 D5 D6 D7
I hope it helps,
Nacho
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA