Experts are full of valuable knowledge and are ready to help with any question. Credentials confirmed by a Fortune 500 verification firm.

Get a Professional Answer

Via email, text message, or notification as you wait on our site. Ask follow up questions if you need to.

100% Satisfaction Guarantee

Rate the answer you receive.

Ask Martin Your Own Question

Martin, Physicist

Category: Calculus and Above

Satisfied Customers: 781

Experience: 20+ years of research, engineering and teaching

47181708

Type Your Calculus and Above Question Here...

Martin is online now

I have 2 groups (1, test and 2, no test) and 2 race

Customer Question

I have 2 groups (1, test and 2, no test) and 2 race categories (1, one race and 2 other race). I am trying to show that one race is dominant in both groups-I am not trying to compare the two groups so if there a statistical test to run to show this or would I just use percentage? I also want to show that within the one group (1, test group) there are 2 locations (1, East and 2, West) and I want to show the the significance of race in each location-is this another percentage or is there a statistical test to run?

Hello there! I can help you with this. Let me get a little more information though. Can you explain in more specific detail exactly what you are trying to test for, and what your procedure was and your data sets are? Thanks!

In case you're unable to come back with more details, and just want a quick answer, I would say based on the description you provided that you will need to use a "contingency table". In such an analysis, you will arrange a table with column headings "Race 1" and "Race 2", and row headings "Yes" and "No", indicating if they passed the test or not. I won't go into detail yet about how to determine if there is a statistical association evident in the data set, since this may not be exactly what you're trying to to. I will just point out that there *is* a well known and accepted method for dealing with such cases, and the underlying statistical test is called a "Chi-squared test".

I am trying to show percentage of frequency. I am trying to to test for significance of one race to another in the same group 2 groups (1 received testing and 2 did not receive testing) I did a demographic breakout for the 2 groups (1, Caucasian and 2, Not Caucasian) I am not trying to compare the 2 groups demographics, just demonstrate the population In these 2 groups the breakout was: 1st Group Testing: Caucasian 80 (86%) and Non Caucasian 13 (14%) 2nd Group Testing: Caucasian 83 (89%) and Non Caucasian 10 (11%) I then Broke it out further to show see if there was a difference by location and the breakout was: 1st Group Location: Testing: Caucasian 22 (63%) and Non Caucasian 13 (37%) No Testing: Caucasian 25 (71%) and Non Caucasian 10 (29%) 2nd Group Location: Testing: Caucasian 58 (100%) and Non Caucasian 0 (0%) No Testing: Caucasian 58 (100%) and Non Caucasian 0 (0%) As you can see the 1st group was more diverse and I am trying to figure out if is significant? All of the groups were tested together and all of the data is combined, the only difference is that when the two locations were broken out by demographics, the 1st group was smaller but was diverse where the 2nd group was larger and not diverse. When combined, Caucasian dominates but when broken out, the 1st group is 2/3 Caucasian and 1/3 Non. Is this significant? Would I even need to show this? It doesn't impact results

OK, let's focus on the first question first. You say you just want to "demonstrate the population", but are not trying to compare demographics. You list number of white and non-white subjects in each group, and calculate these as percentages of the total. What else do you want to do with these data, if not to "compare demographics"?

No, I am trying to do a write up of findings from a project. Not trying to compare just to show that the majority are Caucasian but within this, there was a more diverse group at one site. Would percentages just be fine or would you recommend a non parametric chi square of the Diverse Group Site? Do I even need to show significance if I am just reporting on race? I had run a non parametric chi square on the total (93 in each group) and got a significance level of p=<0.001 And when I ran a NP chi square on the one site I got a significance level of .111 so not statistically significant (Caucasian over Non). Were these the correct tests? Do they even mean anything?

You want to know if you can conclude, based on your data, that the majority of Group 1 people are White in the general population. The answer is yes. The odds of drawing 80 or more whites out of 93 random draws from a general population that is only 50% white is vanishingly small. So you can say with great confidence that Group 1 must contain more than 50% white people. Same goes for Group 2. This probability can be calculated with Excel or any other statistical package using the binomial distribution. If you need the formula, I can supply it. Note that for 93 samples its fine to use a Normal approximation)