Variable and Types of Statistical Variables
Variable and Types of Statistical Variables
There are several operational terms that we shall often use in the field of statistics.
Ex.
Age, Height, Weight, Mortality rate, Income, etc.
Qualitative Data - refers to descriptive attributes that cannot be subjected to
operations of mathematics.
Ex.
Gender, Civil Status, Business Type, Religion, Province, etc.
Discrete Variable - these are variables that can be obtained through counting.
These variables are exact in measure.
Ex.
Number of births, Identification numbers, Pages in a textbook.
Continuous Variable - these are the variables that can be obtain by measuring
objects or attributes. They can never be exact in measure.
Ex.
Height, Temperature, Weight, Area. Time.
Measurements are quantifiers of data in order for these variables to have a specific
magnitude. Note that knowing the type of measurement used in a given data is a must
for a statistician in order for him to avoid wrong choice of statistical test. There are
four types of measurement scales used in statistics.
Ex.
Engineering Courses 5 - Mechanical Engineering
4 - Electrical Engineering
3 - Civil Engineering
2 - Electronics Engineering
1 - Chemical Engineering
Ex.
Volume of Customers 1 - Low 2 - Medium 3 - High 4- Very High
Ranks of Honor Students
Order of grades from highest to lowest in statistics.
Average score classes in Chemistry
Ex.
Test Results , Gravity , Water Level, Temperature
Ex.
Height, Area, Time, etc.
II. RANDOM SAMPLING OF STATISTICAL VARIABLES.
Sampling is the process of selecting and gathering of data. It can minimize the labor
exerted by the researchers and also the cost, time, and effort needed in the study.
There are two types of sampling technique and it is according to the chances of
different elements in a population to be a part of the sample.
a) Simple Random Sampling - also called as the “lottery method” wherein the
elements are randomly selected or with the aid of table of random numbers. This
method limits the number of individual to be chosen as a sample and everyone has an
equal chance to be a part of the sample.
c) Systematic Random Sampling - also called as the “nth technique” wherein the
nth element in the population is being selected until the desired sample size is
selected.
d) Cluster Sampling - is is very applicable when the population is spread out over
a wide geographical area. This is called also as the area sampling technique.
a) Convenience Sampling - this method selects sampling units that come to hand
or are convenient to get information. This technique is used when the accessibility of
the researchers towards the given sample is limited and requires a more easy way to
get that sample or element.
c) Quota Sampling - this is done by drawing samples randomly but if the desired
quota is reached, the drawing of the samples is terminated. This method selects a
specified number of sampling units possessing certain characteristics.
d) Judgement Sampling - in this method, the selection of the sample is based with
accordance of the expert’s judgments towards the given study.
e) Accidental Sampling - only those whom the researcher meets by chance are
included in the sample when using this technique.
To determine the number of samples needed in the study, we will use the Slovin’s
formula.
N
n
1 Ne2
Where
The margin of error is the confidence interval of the formula. Usually, 5% and 1%
are commonly used as the error size which means that 95% and 99% of the sample’s
responses are accurate.
We can also derive a formula for the population and the margin of error.
A. Given: Sample size and Margin of error B. Given: Sample size and Population
Missing: Population Missing: Margin of Error
N N
n n
1 Ne2 1 Ne2
(4) n N( 1 ne 2 ) N n
(4) e 2
Nn
(5) N
n N n
(5) e 100
1 ne 2 Nn
Given: n = 90
e = 5% = 0.05
N=?
Solution
n
(1) N
1 ne 2
90
(2) N
1 90( o.o5 )2
Given: N = 250
95% accuracy = 5% margin of error = 0.05
N=?
Solution
N
(1) n
1 Ne2
250
(2) n
1 250( 0.05 )2
(3) n 154
Example 2.3.3 A researcher wants his sample size for his study is 600 respondents.
Given that the total population is 50,000. What would be his margin of error?
Solution
N n
(1) e 100 (3) e 4.06%
Nn
50000 600
(2) e 100
( 50000 )( 600 )
1. MEAN - use if wamt to have a single value that represents the whole study,.
2. STANDARD DEVIATION -to determine whether the data in a scatter plot
diagram is compact or scattered. When the data points are compact, the more reliable
the data is. In other words, the lower the standard deviation, the accurate the data is.
3. VARIANCE - it is the square of standard deviation and can be used also to check
how the data variates.
4. Z-TEST - use only if the population/sample are assumed to be normal, and also the
number of samples must be greater than 30.
5. T-TEST - use only if the population/sample are assumed to be normal, and also the
number of samples must be lesser than 30.
6. ANOVA - Analysis of Variance is used to determine if there is significant
difference between two or more groups
7. CORRELATION - use only if the researcher wants to determine the strength of
relationship between two or more variables.
8. REGRETION - use only if the researcher wants to forecast a certain data, out of
the given data.
B. NON-PARAMETRIC TESTS- tests that are usable for nominal and ordinal
data.
1. MANN-WHITNEY U TEST - used if want to determine the significant difference
of two variables into one category(ex. Gender to their preference)
2. SPEARMAN RHO - like the correlation analysis but for rank data.