Types of scales and Stats tests? - USMLE Forums
 USMLE Forums         Your Reliable USMLE Online Community     Members     Posts
 Home USMLE Articles USMLE News USMLE Polls USMLE Books USMLE Apps
 USMLE Forums Types of scales and Stats tests?
 Register FAQs Members List Search Today's Posts Mark Forums Read

 USMLE Step 1 Forum USMLE Step 1 Discussion Forum: Let's talk about anything related to USMLE Step 1 exam

#1
05-31-2011
 USMLE Forums Veteran Steps History: 1+CK+CS+3 Posts: 264 Threads: 16 Thanked 349 Times in 165 Posts Reputation: 359
Types of scales and Stats tests?

So, there are 4 types of scales in stats.
a. Nominal (categorical)
b. Ordinal
c. Interval
d. Ratio

1. Can somebody explain me what's the difference between them. I mix up ordinal and interval scales as both of them can have different groups in sequence, or not? Any examples which can clear this out.

2. When to use t-test, chi square, one way ANOVA, matched pairs t-test and repeated measures ANOVA?
 The above post was thanked by: benzo (06-03-2011), hassanahk (06-02-2011), struggle (06-02-2011)

#2
05-31-2011
 USMLE Forums Master Steps History: Not yet Posts: 557 Threads: 69 Thanked 1,038 Times in 351 Posts Reputation: 1058

Great questions! I don't know the answer to them all, but here are a couple basic quick tricks that I use (of course it's more complicated but I've found i can answer most biostats questions this way):
1) t-test and ANOVA are basically the same in that they both compare the means. The big difference is that t-test compares 2 means, ANOVA compares the means of of 3 or more groups.
2) If you are not given means, then use a chi square test. This is used to compare categories (eg. number of people who responded yes/no, percentage of people who do xyz etc)

Those are the basics. A matched t-test is again just looking at the means between groups, except now you are matching variables. For instance, you want to look at the mean drug dose required to treat people who smoked for a mean number of years vs the mean dose in folks who smoked for a different mean number of years. So in this case drug dose and smoking is matched.
 The above post was thanked by: drnirajmavani (06-02-2011), excellence (05-22-2014), Gulara (09-09-2011), mle guy (06-01-2011), struggle (06-02-2011)
#3
05-31-2011
 USMLE Forums Master Steps History: 1+CK+CS+3 Posts: 1,357 Threads: 194 Thanked 3,259 Times in 881 Posts Reputation: 3269

Quote:
 Originally Posted by heights Great questions! I don't know the answer to them all, but here are a couple basic quick tricks that I use (of course it's more complicated but I've found i can answer most biostats questions this way): 1) t-test and ANOVA are basically the same in that they both compare the means. The big difference is that t-test compares 2 means, ANOVA compares the means of of 3 or more groups. 2) If you are not given means, then use a chi square test. This is used to compare categories (eg. number of people who responded yes/no, percentage of people who do xyz etc) Those are the basics. A matched t-test is again just looking at the means between groups, except now you are matching variables. For instance, you want to look at the mean drug dose required to treat people who smoked for a mean number of years vs the mean dose in folks who smoked for a different mean number of years. So in this case drug dose and smoking is matched.
Nice rule of thumb!

I would like to add: a matched-pairs t-test is used to test whether there is a significant mean difference between two sets of paired data (correlated data).

for example:
- pair of twins
- where the same people are being measured in before-and-after comparison
- when the group is given two different tests at different times
 The above post was thanked by: drnirajmavani (06-02-2011), excellence (05-22-2014), Gulara (09-09-2011), mle guy (06-01-2011), struggle (06-02-2011)
#4
05-31-2011
 USMLE Forums Guru Steps History: Not yet Posts: 324 Threads: 70 Thanked 236 Times in 143 Posts Reputation: 246

Are we supposed to know how to use Chi square test?
I didnt get any details in the High yield Biostatistics..
#5
06-01-2011
 USMLE Forums Veteran Steps History: 1+CK+CS+3 Posts: 264 Threads: 16 Thanked 349 Times in 165 Posts Reputation: 359

Quote:
 Originally Posted by struggle Are we supposed to know how to use Chi square test? I didnt get any details in the High yield Biostatistics..
No we don't have to use the chi square test or compute any other formula. We just have to know that WHEN we have to use them.
 The above post was thanked by: benzo (06-03-2011), struggle (06-01-2011)
#6
06-01-2011
 USMLE Forums Veteran Steps History: 1+CK+CS+3 Posts: 264 Threads: 16 Thanked 349 Times in 165 Posts Reputation: 359
Thanks

Thanks a lot heights and bebix.
Now if any can explain the differences between the scales: nominal, ordinal, interval and ratio. Thanks

Last edited by mle guy; 06-01-2011 at 12:37 AM.
 The above post was thanked by: benzo (06-03-2011)
#7
06-01-2011
 USMLE Forums Master Steps History: 1+CK+CS+3 Posts: 1,357 Threads: 194 Thanked 3,259 Times in 881 Posts Reputation: 3269
Types of Numerical data

Types of Numerical data

As data we consider the result of an experiment. A rough classification is as follows:

 Nominal data Numbers or text representing unordered categories (e.g., 0=male, 1=female // 0=death, 1=alive // 0=north, 1=south, 2=west, 4=east)

 Ordinal data Numbers or text representing categories where order counts (e.g., 1=fatal injury, 2=severe injury, 3=moderate injury, etc.)

 Discrete data: This is numerical data where both ordering and magnitude are important but only whole number values are possible (e.g., Numbers of deaths caused by heart disease (765,156 in 1988) versus suicide (40,368 in 1988).

 Continuous data Numerical data where any conceivable value is, in theory, attainable (e.g., height, weight, etc.)

Hope this help
Ref: Principles of Biostatistics. Marcello Pagano
 The above post was thanked by: Claus_CU (06-02-2011), drnirajmavani (06-02-2011), hassanahk (06-02-2011), mle guy (06-02-2011), struggle (06-02-2011)
#8
06-02-2011
 USMLE Forums Veteran Steps History: 1+CK+CS+3 Posts: 264 Threads: 16 Thanked 349 Times in 165 Posts Reputation: 359

Thank you so much. It really helped
 The above post was thanked by: benzo (06-03-2011)
#9
06-02-2011
 USMLE Forums Master Steps History: 1+CK+CS+3 Posts: 1,357 Threads: 194 Thanked 3,259 Times in 881 Posts Reputation: 3269

Quote:
 Originally Posted by mle guy Thank you so much. It really helped
Super!

now try to solve the Biostatistic questions!!!

I'm going to try to upload 2-3 per day
 The above post was thanked by: benzo (06-03-2011)

 Tags Biostatistics-Epidemiology

Message:
Options

## Register Now

In order to be able to post messages on the USMLE Forums forums, you must first register.
User Name:
Medical School
Choose "---" if you don't want to tell. AMG for US & Canadian medical schools. IMG for all other medical schools.
 AMG IMG ---
USMLE Steps History
What steps finished! Example: 1+CK+CS+3 = Passed Step 1, Step 2 CK, Step 2 CS, and Step 3.

Choose "---" if you don't want to tell.

 Not yet Step 1 Only CK Only CS Only 1 + CK 1 + CS 1+CK+CS CK+CS 1+CK+CS+3 ---
Favorite USMLE Books
 What USMLE books you really think are useful. Leave blank if you don't want to tell.
Location
 Where you live. Leave blank if you don't want to tell.

## Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.