Statistical Analysis - T test and ANOVA?
Does the dependent variable always has to be continous?
Because I am now currently analysis a set of data, and wanna know whether person's age would have effect on their perference of having an ID card.
So the DV is perference on ID card (For and Aganist) and the IV would be the Age Group (which has six levels)
All the books said that the DV has to be continous but in this case both of my DV and IV are catergorial.
So wht should I do?
Thx
Answer:
are you suggesting that age causes preference?
seems to me that neither is the appropriate analysis for the variables you have. given than you can't manipulate either, you're actually conducting a quasi-experiment therefore you're not going to be able to infer any causal relationship (note that off the top of my head i can think of four other potentially confounding factors that very well might be affecting your data).
it seems what you're actually interested in is observing whether there is any relationship between a person's age and his/her preference. hence you might want to simply conduct some correlational analyses (e.g., a pearson correlation matrix) to see what if any significant relationships exist.
I would consider a non parametric test - likely a chi-square. You will need a larger n than usually necessary for a parametric test, and you may have to combine some categories in your IV if you get a warning from your program that some of your expected frequencies are too low. (You might want to do this anyway after examining your data.) A non-parametric rank based test like the Kristall-Wallis may also be appropriate.
Parametric tests require continuous data - but these tests are 'robust' - ANOVA's (or t-tests) can be sometimes be used on data that is not fully continuous. But they can't be used on categorical measures like the DP you are using.
If you have to do stat often, I would recommend some college level courses - which are not fun, but will help you avoid some of the many pitfalls that can occur when performing data analysis; they can also help keep you employed.
A t test compares the means of two groups. For example, compare whether systolic blood pressure differs between a control and treated group, between men and women, or any other two groups.
Don't confuse t tests with correlation and regression. The t test compares one variable (perhaps blood pressure) between two groups.
Use correlation and regression to see how two variables (perhaps blood pressure and heart rate) vary together.
Also don't confuse t tests with ANOVA. The t tests (and related nonparametric tests) compare exactly two groups. ANOVA (and related nonparametric tests) compare three or more groups.
Finally, don't confuse a t test with analyses of a contingency table (Fishers or chi-square test).
Use a t test to compare a continuous variable (e.g., blood pressure, weight or enzyme activity).
Use a contingency table to compare a categorical variable (e.g., pass vs. fail, viable vs. not viable).
More Questions and Answers:
How would you answer this interview question? it makes no sense...?
In your words what are some treatment for Encopresis?
Talking to oneself , wt do u call it?
3 questions. Do you really care if a suicidal person...?
He had a felt-need to see Europe before he died. When was FELT-NEED first used?
I'm an Atheist but am I alone in that?
What did Jung mean by "complexes"? how did he diagnose these complexes?
What is your greatest fear?why?
What is the use of crying?
Because I am now currently analysis a set of data, and wanna know whether person's age would have effect on their perference of having an ID card.
So the DV is perference on ID card (For and Aganist) and the IV would be the Age Group (which has six levels)
All the books said that the DV has to be continous but in this case both of my DV and IV are catergorial.
So wht should I do?
Thx
Answer:
are you suggesting that age causes preference?
seems to me that neither is the appropriate analysis for the variables you have. given than you can't manipulate either, you're actually conducting a quasi-experiment therefore you're not going to be able to infer any causal relationship (note that off the top of my head i can think of four other potentially confounding factors that very well might be affecting your data).
it seems what you're actually interested in is observing whether there is any relationship between a person's age and his/her preference. hence you might want to simply conduct some correlational analyses (e.g., a pearson correlation matrix) to see what if any significant relationships exist.
I would consider a non parametric test - likely a chi-square. You will need a larger n than usually necessary for a parametric test, and you may have to combine some categories in your IV if you get a warning from your program that some of your expected frequencies are too low. (You might want to do this anyway after examining your data.) A non-parametric rank based test like the Kristall-Wallis may also be appropriate.
Parametric tests require continuous data - but these tests are 'robust' - ANOVA's (or t-tests) can be sometimes be used on data that is not fully continuous. But they can't be used on categorical measures like the DP you are using.
If you have to do stat often, I would recommend some college level courses - which are not fun, but will help you avoid some of the many pitfalls that can occur when performing data analysis; they can also help keep you employed.
A t test compares the means of two groups. For example, compare whether systolic blood pressure differs between a control and treated group, between men and women, or any other two groups.
Don't confuse t tests with correlation and regression. The t test compares one variable (perhaps blood pressure) between two groups.
Use correlation and regression to see how two variables (perhaps blood pressure and heart rate) vary together.
Also don't confuse t tests with ANOVA. The t tests (and related nonparametric tests) compare exactly two groups. ANOVA (and related nonparametric tests) compare three or more groups.
Finally, don't confuse a t test with analyses of a contingency table (Fishers or chi-square test).
Use a t test to compare a continuous variable (e.g., blood pressure, weight or enzyme activity).
Use a contingency table to compare a categorical variable (e.g., pass vs. fail, viable vs. not viable).
The answers post by the user, for information only, FunQA.com does not guarantee the right.
More Questions and Answers: