Professional Documents
Culture Documents
Oliver U. Emata
Presenter
A simple reminder:
Do not correct a
fool, or he will
hate you;
correct a wise
man, and he will
appreciate you
Source:www.facebook.com/lessonslearninlife
What is Quantitative Research?
What makes it different with
other type of research?
Considerations:
What statistical tools
to be used in
How to interpret the processing and
result of my study? analyzing my data?
What is the role of Statistics in
Qualitative Research?
Also:
-It will describes the kind of research that
we are working.
-It will tells the method or design of the
study particularly in the processing of data.
-Itwill tells you if there is significant
difference or significant relationship
Statistics
is a branch of mathematics mainly concerned with
describing and interpreting a collection of data,
and with drawing conclusions about population
from a knowledge of the properties.
-Melecio C. Deauna
What is Quantitative Research?
-involve numbers
Includes attributes such as: -results of counting
-sex (male or female) e.g
-attitude (favorable or not) -Number of students
-emotional (happy or sad) -number of teachers with masteral
-amount of parents financial support
Ratio
Classification of data -proportion has meaning
-has an absolute zero
Interval
Ordinal -Arithmetic differences
-No true zero values
Nominal Ordered or ranked data
e.g grades
Categorical data/dummy e.g
Rank, order
Real Nominal
e.g male or female
Artificial Nominal
e.g Pass or failed
Exercises
1. Gender Real Nominal
2. Test Results (Passed/Failed) Artificial Nominal
3. Performance Rating of Teachers Ordinal
4. Scores in Exam Interval
If you want to
determine if 2 or more
variables are related
to associated
3
Goal of the Analysis: Relationship or
Association
Copyright, Wilbur W. Stanton, Ph.D. and Angela DAuria Stanton, Ph.D., 2003.
Assumption/Hypothesis
Dropping of daily temperature would change the
treescolor and start the leaves falling.
Prediction
Reaching a 5 oC average daily temperature would result
to 90% of the trees leaves fall.
Example: Table 1. SCORES OF 10 STUDENTS IN A 20-ITEM TEST FOR
PHYSICS AND MATHEMATICS
How will you describe this results?
Mean (Math)=11+10+12+10+6+3+10+7+17+14
10
= 10
Median (Physics)=n +1 th observation Student Physics Mathematics
2 1 7 3
=10+1 th observation
2 2 8 6
= 11 th observation
2
3 9 7
=5.5 th observation 4 9 10
= 10
5 10 10
6 =
Median 10
10 10
10
7 10 11
Median (Math)=n +1 th observation 8 11 12
2 9 12 14
=10+1 th observation
2 10 14 17
= 11 th observation
2 Mean 10 10
=5.5 th observation Median 10 10
= 10
Mode 10 10
Descriptive Statistics
Frequency Distribution in Tabular Form:
Poor 28.00
Fair 87.00
Good 7.00
Excellent 0.00
Source:Katigaman,pp.12
Descriptive Statistics
Frequency Distribution in Graphical Form:
100.00
90.00
Frequency Distribution
80.00
70.00
60.00
50.00
87.00
40.00
30.00
20.00
28.00
10.00
7.00
0.00 0.00 0.00
Poor Fair Good Very Good Exellent
Description
Source:Katigaman,pp.12
Descriptive Statistics
Description
Poor
Fair
Good
Very Good
Excellent
N NAT MPS Parent's Contributions Parent's Attendance Parent's Attendance Average Parents' Support
1 2 3 4
1 Phi
Coefficient
2 Phi Tetrachoric
Coefficient Correlation
Coefficient
3 Rank biserial Rank biserial Sphearmans
Rank
4 Point biserial Biserial Multiserial Pearson
Product
Legend: 1. Real Nominal
2. Artificial Nominal
3. Ordinal Data
4. Interval/Ratio
Descriptive Statistics
Commonly used statistical tool in describing the
relationship between two or more variables
Sex (x)
Male (1) Female(0)
Against(0) 10 5 15
Opinion(y)
Favor(1) 6 12 18
16 17
Descriptive Statistics
Commonly used statistical tool in describing the
relationship between two or more variables
1. Phi-Coefficient
2. Tetrachoric Correlation Coefficient
3. Rank Biserial
4. Spearmans Rho
5. Point Biserial
6. Biserial
7. Multiserial
8. Pearson r
Descriptive Statistics
A correlation may be
statistically significant (it
didnt happen by chance) but
be weak or low which means
it is nothing to get excited
about. It has no practical
significance.
computed/tabular p-value
6000
strong positive
4000 correlation
3000
2000
1000
0
100 150 200 250 300 350 400
Weight
Scatter plots are often used to depict
correlations
45
Miles from Krispy Creme
1. t- test
1.1 mean of the population
1.2 difference of means for independent samples
1.3 difference of means for dependent (correlated) samples
1.4 inference about correlation coefficients
2. z-test
2.1 inference about phi coefficient
2.2 inference about population proportion
3. ANOVA
4. MANOVA
5. ANCOVA
6. MANCOVA
In interpreting the data
Decide how you will organize or classify your data
(e.g., by individual students, by gender or subgroups,
by classes)
Summarize data in table/graph/chart form
Analyze quantifiable data with statistics or technical
assistance
Analyze qualitative data holistically noting important
elements or themes
Never beg for a
relationship.
Be brave to accept
the one who
really wants to be
you
and reject the one
who just
pretends to be
with you
www.facebook.com/RelationshipPost
Daghang Salamat!!!!!!!!
Oliver U. Emata