REVIEW OF

BASIC
STATISTICS
MEASUREMENT SCALES/LEVELS

The Nominal Scale

•simply represents qualitative difference in the
variable measured
•can only tell us that difference exists without the
possibility of telling the direction or magnitude of
the difference
•e.g. Program in college, race, gender, occupation,
religion, etc.
MEASUREMENT SCALES/LEVELS

The Ordinal Scale

•the categories that make up an ordinal scale
form an ordered sequence
•can tell us the direction of the difference but
not the magnitude
•e.g. coffee cup sizes, socioeconomic class, T-
shirt sizes, food preferences
MEASUREMENT SCALES/LEVELS

The Interval Scale

•categories on an interval scale are organized sequentially, and all categories
are numerically measured
•we can determine the direction and the magnitude of a difference
•may have an arbitrary zero (convenient point of reference) but has no true zero
point
e.g. temperature in Fahrenheit, time in seconds
MEASUREMENT SCALES/LEVELS

The Ratio Scale

•consists of equal, ordered categories anchored by a zero point that is not
arbitrary but meaningful (representing absence of a variable
•allows us to determine the direction, the magnitude, and the ratio of the
difference
•e.g. reaction time, number of errors on a test, scores in a test, speed of
cars, weight loss, etc
CLASSIFICATION OF DATA ANALYTIC METHODS

Dependence Method
The dependence methods test for the presence of or absence of
relationship between two sets of variables – the dependent and
independent variables. Common dependence methods are t-test,
ANOVA, ANCOVA, regression analysis, chi-square test, MANOVA,
discriminant analysis and, logistic regression.
CLASSIFICATION OF DATA ANALYTIC METHODS

Interdependence methods
When data sets do exist for which it is impossible to conceptually
designate one set of variables as dependent and another set of variables
as independent. For these types of data sets the objectives are to
identify how and why the variables are related among themselves.
Common examples are correlation analysis, principal component
analysis, and factor analysis.
RELATIONSHIPS OF VARIABLES
Dependency

Independent
Variables
•Age
Hypertension
•Lifestyle
•BMI
•Family History
RELATIONSHIPS OF VARIABLES

Interdependency

•Age
Systolic Pressure
•Weight
Blood Sugar Level
•Cholesterol level
 INTERPRETING STATISTICAL RESULT
Important Terms

The test statistic is a value computed from the sample data, and it
is used in making the decision about the rejection of the null
hypothesis.
The critical region (or rejection region) is the set of all values of
the test statistic that cause us to reject the null hypothesis. It is
decided by Critical Value.
The significance level (denoted by ) is the probability that the
test statistic will fall in the critical region when the null hypothesis
is actually true. Common choices for  are 0.05, 0.01, and 0.10.
 INTERPRETING STATISTICAL
RESULT
The statement of the problem/hypothesis is the basis for
interpreting results.
The null hypothesis is either rejected or not to be rejected
Significant result is met when the null hypothesis is
rejected. Not significant when the null hypothesis is not
rejected.
 INTERPRETING STATISTICAL
RESULT
Significance can mean any of the following:
• There is a relationship.
• There is an association between or among variables.
• There is an effect.
• The treatment is effective.
• A variable is dependent on the other variable/s.
• There is a difference/different effect.
INTERPRETING STATISTICAL RESULT

Question:
• When and how do you reject or fail to reject the
null hypothesis?
• When do we say that the result is Significant?
TRADITIONAL METHOD

 Reject H0 if the test statistic falls within the critical region.

 Fail to reject H0 if the test statistic does not fall within the critical
region.

Critical Critical
Value Value
 P-VALUE METHOD

Reject H0 if P-value   (where  is the significance level, such

as 0.05).
Fail to reject H0 if P-value > .
DESCRIPTIVE STATISTICS

THE FREQUENCY DISTRIBUTION TABLE

(FDT)
An FDT is a statistical table showing the frequency or
number of observations contained in each of the defined
classes or categories.
TYPES OF FDT
Qualitative or Categorical FDT – an FDT where the data are group
according to some qualitative characteristics; data are grouped into
non numerical categories.
Quantitative FDT – an FDT where data are grouped according to
some numerical or quantitative characteristics.
DESCRIPTIVE STATISTICS
EXAMPLE QUALITATIVE FDT

Table 1. Distribution Respondents by Educational Level

Educational Level Frequency Percentage (%)
Highschool 30 20
College 75 50
MA/Ms 45 30
Total 150 100

Interpretation: Most of the respondents are college graduate which constitute

50% (75 out of 150) of the total respondent.
DESCRIPTIVE STATISTICS
EXAMPLE QUANTITATIVE FDT

Table 1. Frequency Distribution of the Age of the Respondents

Age Frequency Percentage (%)
16-19 134 66
20-25 64 32
26 above 4 2
Total 202 100

Interpretation: Majority of the respondents, about 134 out of 202 (66%), are 16-
19 years of age.
DESCRIPTIVE STATISTICS

Charts and Graphs

• Pie Chart
• Bar Chart
• Line Chart
• Histogram
• Scatter Diagram
DESCRIPTIVE STATISTICS

Distribution of Respondent s by Type of Behavior Distribution of Respondent s by Type of Behavior

120

100

23%
33% 80

60

100
23%
40
20% 70 70
60

20

0
Envious Optimistic Pessimistic Trusting Envious Optimistic Pessimistic Trusting
DESCRIPTIVE STATISTICS

SALES IN MILLION
25

20 19.7 20.1
18.2
17.5

15 14.5
13.8
12.8 12.5
11.3
10.2
10

0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
DESCRIPTIVE STATISTICS

5.5

4.5

3.5

2.5

2
2 2.5 3 3.5 4 4.5 5
DESCRIPTIV
E STATISTICS
1. MEASURES OF
CENTRAL
TENDENCY
 Measures of absolute dispersion are expressed in the units of
DESCRIPTIV the original observations.

E STATISTICS There are three main measures of absolute dispersion:

• The range
• Variance

1. MEASURES OF • Standard deviation

DISPERSION
Measures of relative dispersion are unit-less and are used
when one wishes to compare the scatter of one distribution with
another distribution.

Some measures of absolute dispersion:

• Coefficient of Variation
• Standard Score
THE STANDARD NORMAL DISTRIBUTION
The distribution of a normal random variable with mean zero and standard
deviation equal to 1 is called a standard normal distribution.

If X follows a normal distribution, then X can be transformed into a standard

normal random variable through the following transformation.
INFERENTIAL
STATISTICS (BASIC
TOOLS)
T- TEST

• T-test is a parametric test that is commonly used

to test difference between 2 group means. Means
may be from independent or dependent groups
• A dependence method, usually a univariate tests
and is most effective to use when the independent
variable is non-metric.
Example: testing the relationship between level of
job satisfaction and gender.
ONE-SAMPLE T-TEST

Used to test single population mean

Usually compare the mean to existing population mean or to the standard
norm
Example is comparing the performance in the medical board exam of a
certain school to the national result
T-TEST FOR INDEPENDENT SAMPLES

• 
Also called the two sample t-test for independent samples
Assumptions maybe equal or unequal variances
It intends to test whether there is a significant difference
between the means of two unrelated groups
It is use to test the null hypothesis:
T-TEST FOR DEPENDENT SAMPLES

•  
Also called the paired t-test
It intends to test whether there is a significant
difference between the means from the same
group.
Mostly used in comparing pre-test and post-
test results
It is use to test the null hypothesis:
 ANOVA – ANALYSIS OF VARIANCE

It is an appropriate technique for estimating the parameters of a linear model, Y

= α + βx + ε, when the independent variables are nominal or categorical.
In practice, it is used to test significant differences among group means (more
than 2 groups)
Mostly use in experimental research, esp. when design of experiment is applied.
Example: Consider the case where a medical researcher is interested about the
effect of occupation on cholesterol level. The independent variable, occupation,
is nominal.
CORRELATION ANALYSIS

Correlation is a measure of the direction and strength of linear

relationship between two variables.
 Direction means positive or negative.
 Strength can be perfect, strong or high, moderate, low or zero or no
correlation.
Correlation between two variables does not prove X causes Y or Y
causes X.
SCATTER DIAGRAM
PEARSON CORRELATION COEFFICIENT R

Pearson Correlation coefficient is a numerical value that measures strength and

direction of linear relationship
Symbol: r
r can range from -1.0 to +1.0 
Sign (+/-) indicates “direction”
Value indicates “strength”
Measures a “linear” relationship only
Significance of the Pearson r can be tested using t-test
PEARSON CORRELATION COEFFICIENT R

Illustration:
-1 0 1
Perfect Negative No/Zero Perfect
Correlation Correlation Positive
Correlation

 Closer to 0 = weaker
 Closer to 1.0 = stronger
 r close to 1.0 perfect
 r  0 could mean many things:
No correlation at all between X & Y
Non-linear relationship between X & Y
Restricted range on X and/or Y
Outlier may be causing problems
ACTIVITY: INTERPRET THE FOLLOWING R
COEFFICIENT
1) r = 0.85
2) r = -0.69
3) r = -0.37
4) r = -0.11
5) r = 0.09
6) r = 0.32
7) r = -0.92
8) r = 0.75
ACTIVITY: INTERPRET THE FOLLOWING R
COEFFICIENT
1) r = 0.85 Ans.: Very Strong Positive
2) r = -0.69 Ans.: Moderate/Strong Negative
3) r = -0.37 Ans.: Weak Negative
4) r = -0.11 Ans.: No/Very weak
5) r = 0.09 Ans.: No/Very weak
6) r = 0.29 Ans.: Weak Positive
7) r = -0.92 Ans.: Very Strong Negative
8) r = 0.75 Ans.: Strong Positive
INTERPRETING R (Evans, 1996)
r Verbal Interpretation
-1 Perfect Negative Correlation
-0.8 to -0.99 Very Strong Negative Correlation

-0.6 to -0.79 Strong Negative Correlation

-0.4 to -0.59 Moderate Negative Correlation
-0.2 to -0.39 Weak Negative Correlation
-0.01 to -0.19 Very Weak Negative Correlation
0 No Correlation
0.01 to 0.19 Very Weak Positive Correlation
0.2 to 0.39 Weak Positive Correlation
0.4 to 0.59 Moderate Positive Correlation
0.6 to 0.79 Strong Positive Correlation
0.8 to 0.99 Very Strong Positive Correlation
1 Perfect Positive Correlation
CHI-SQUARE TEST

• The Chi-Square test is known as the test of goodness of fit and

Chi-Square test of Independence. In the Chi-Square test of
Independence, the frequency of one nominal variable is
compared with different values of the second nominal variable.
• The Chi-square test of Independence is used when we want to
test associations between two categorical variables.
CHI-SQUARE TEST

Assumptions
Independent random sampling
Nominal/Ordinal level data
No more than 20% of the cells have an expected frequency less than 5
No empty cells