A modest introduction to statistics from a clinician.
With great thanks to Professor Dr. Ahmed Shokeir, Head of Urology and Nephrology center, Mansoura University, Egypt
Arterial health throughout cancer treatment and exercise rehabilitation in wo...
1 variables - Statistics
1. Introduction to Statisticsfrom a clinician
1 - Variables
Ahmed Abdelaziz Ghanem
Assistant lecturer of General & Minimally Invasive surgery
Mansoura University Hospital
Mansoura University, Egypt
2. Introduction
When we decide to do a study, we call it a trial .
In any study there is the population, from which we choose a sample to
conduct the trial on.
Data we want to collect is called the variables each of which has a value .
3. Definitions
Trial : the study we want to conduct.
Population : is the entire pool from which a statistical sample is drawn, e.g. ( if study on
HTN then population will be all people with HTN all over the world )
Sample : is a set of data collected and/or selected from a statistical population by a
defined procedure, e.g. (if study on HTN them the sample will be the selected 100 cases
under trial) .
Case : Every single object or subject.
4. Definitions
Variables : A variable is any characteristics, number, or quantity that can
be measured or counted. A variable may also be called a data item. Age,
sex, country of birth, eye color etc.
Value : each variable has a value which differs according to the variable.
5. Examples
A RCT for efficiency of new drug for treatment of hypertension. The study was done on 200
patients.
?? Trial
?? Population
??sample
??variables
6. Types of variables
Two Classifications :
1- Continuous Vs categorical ( Ordinal / Nominal ).
2- Dependent VS Independent.
7. Continuous variable
If a variable can take on any value between its minimum and its maximum
value, it is called a continuous variable; otherwise, it is called a categorical
variable.
For example : Age can be1 year = 12 m = 52 weakks = 365 d = no limit for
expression between 2 points = continuous variable.
1 is the value, year is the discrimination .
8. Ordinal variable
At least 3 and limited .
Containing hierarchy
Example : stage of disease, Glasgow coma score , number of pregnancies
(no number between 2 and 3 no 2.5 pregnancy).
9. Nominal variable
mostly 2 values defined by their name: data can’t be ordered.
Example : Group a and b / Male vs female / life vs dead / Recovery vs No
recovery .
Or >2 but can’t be ordered e.g. race: Egyptian, Saudi, Libyan, etc…
10. Advice
Biggest amount of information can be collected from continuous variable
> ordinal > nominal .
In your trial try to collect continuous data.
11. Dependent Vs Independent
This classification is according to outcome :
Primary outcome = dependent variable / Secondary outcome =
independent
Example : A trial to study the probability of CHD Vs smoking
* 2 variables : CHD and smoking
** We are testing the hypothesis of the probability of CHD so CHD is the
dependent variable. Furthermore we want to compare the probability of
CHD among smokers Vs the probability among non-smokers so smoking
status is the independent variable
12. Example: -in a RCT subject receives either : drug X or placebo (phantom–
starch-sugar) to facilitate recovery , age and gender are confounding
variables .
* 4 variables ?
If the aim is to study the effect of the drug, What is the dependent
variable??
13. (Age) – (Gender) – (Recovery- No recovery ) – ( drug vs placebo)
Recovery is the dependent variable if the aim is to study the effect of the
drug.
If aim is the distribution of age: then age is dependent
Variable is Dependent according to study question.