Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
ANOVA or CORRELATION?
Message posted by A on December 10, 1999 at 12:00 AM (ET)
Hello,
Stats is very new to me. I am trying to analyze a table found in a particular research study comparing Depression of Institutional VS. Community-living older adults.
The table shows such variables as gender, education level, physical health, and personal meaning vertically along the left side of the table. Across the top of the table are two columns for community or institutional with Depression across the top of both columns.
I know that the first table, which I have not described, is a simple t-test comparing means and SDs for community vs. institutional. But this 2nd table I have described has me stumped. I was under the impression that a correlation could only
occur when all of the variables were measured. But I am not altogether sure that the table is an ANOVA, either. The authors title table 2 "Bivariate Correlations of Depression with Demographics and Predictor Variables". Can anyone out there help me?? Thanks SO much.
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: ANOVA or CORRELATION?
Message posted by JG on December 10, 1999 at 12:00 AM (ET)
The word correlation is sometimes used loosely to indicate any sort of relationship between factors. Also there are many types of correlations such as rank correlation etc. Remember that correlation measures in some way how two things vary or how variation in one is related to variation in the other.
Re: ANOVA or CORRELATION?
Message posted by Bill on December 10, 1999 at 12:00 AM (ET)
It sounds as if the second table presents correlations between depression and each variable (gender, ed level, etc.) within the institutional group and again within the community living group.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.
