Need help on Sample size

#1
Hello Everyone,

I am working on my dissertation and I need to know what is the sample size I need to have. I will conduct MANCOVA research to see if there are any differences attittutes toward their managers of the 4 group of professionals (engineers, professors, information technology workers, fashion designers). The dependent variables are gender (males and females), and independent variables are engineers, professors, information technology workers, fashion designers. I would like to set alpha at .05 and power is .80 and would like to have medium size effect. Would you please let me know what is the sample size I will need to conduct this research and the steps to get there. If you have any additional questions, please let me know.

Thank you!
 
#2
Anna,
First of all, you cannot have a dependent variable that is gender since you cannot control gender (I think that is predetermined at birth). By definition, dependent variables depend on all others in the model, but being a male does not depend on whether or not you are an engineer or a professor, perhaps being an engineer depends on whether or not you are a male. I think you mean that your dependent variable is the attitudes toward their managers, the predictor should probably be type of a professional and the independent variable that you will need in order to adjust the model is gender.
Second, in order to calculate power, you need an effect size. This is an estimate that you think you will get when you do your study. Usually you get this value from either a pilot study or some previous literature. Once you get an effect size, calculating the sample size is simple.

Jenny Kotlerman
www.statisticalconsultingnetwork.com
 
#3
thank you

Thank you for your response. You are right. I meant the independent variables: gender and type of professionals. Dependent variable is attitute toward manager. Do you know how much they will charge me for analysing the sample size for this research? Please let me know.

Thank you!