Вы находитесь на странице: 1из 32

STATISTICAL

ANALYSIS
WHAT IS STATISTICS??

Statistics is the science of


collecting, organizing,
analyzing and interpreting data
in order to make decisions
ROLE OF STATISTICS IN RESEARCH
Statistical method help the researcher in making his
research design, particularly in experimental research
Statistical techniques help the researcher in
determining the validity and reliability of his research
instruments
Statistical manipulations organize raw data
systematically to make the latter appropriate for study
Statistics are used to test the hypothesis
Statistical treatments give meaning and interpretation
to data
ERROR BARS
WHAT IS AN ERROR BARS
 An error bar is a line through a point on a graph, parallel to
one of the axes, which represents the uncertainty or
variation of the corresponding coordinate of the point.
 the error bars most often represent the standard deviation
of a data set
WHY INCLUDE ERROR BARS ON GRAPH???

How spread the •Small SD bar = low spread, data are


clumped around the mean
data are around •Larger SD bar = larger spread, data
the mean value are more variable from the mean

The reliability of the


mean value as a
representative
number for the data •small SD bar = more reliable
set. In other words, •larger SD bar = less reliable
how accurately the
mean value
represents the data
The likelihood of there being a significant
difference between between data sets
WHAT DO ERROR BARS INDICATE ABOUT STATISTICAL
SIGNIFICANCE?

When standard deviation errors bars


overlap quite a bit, it's a clue that
the difference is not statistically
significant.
When standard deviation errors bars
overlap even less, it's a clue that
the difference is probably not statistically
significant.
When standard deviation error bars do not
overlap, it's a clue that the difference may be
significant, but you cannot be sure.
Commonly Used Multipliers
Multiplier Number (z*) Level of Confidence

3.0 99.7%

2.58 (2.576) 99%

2.0 (more precisely 1.96) 95%

1.645 90%

1.282
80%
1.15 75%

1.0 68%
YOU MUST ACTUALLY PERFORM A
STATISTICAL TEST TO DRAW A
CONCLUSION.
How many GROUPS
(means) are being
compared?

2 samples 2 Or more
involved samples involved
2 GROUPS 3 or more
In each Variables GROUPS

Independent Dependent
Sample T- Sample T- ANOVA
test test
INDEPENDENT SAMPLE T TEST

TO TEST WHETHER 2 DIFFERENT


GROUPS FROM 1 INDEPENDENT
VARIABLE HAVE SIGNIFICANT EFFECT ON
DEPENDENT VARIABLE

p-value / sig (2-tailed) < 0.05 (α)


There is significant effect
INDEPENDENT SAMPLE T-TEST

Test whether there is significant effect of different


thermoplastic composites on the tensile strength.

Bamboo Kenaf
(MPa) (MPa)
46.7 47.0
40.1 46.3
49.2 37.3
52.1 42.9
40.9 48.7
ONE WAY ANOVA
TO TEST WHETHER 2 OR MORE
DIFFERENT GROUPS FROM 1 (or more
than 2) INDEPENDENT VARIABLE HAVE
SIGNIFICANT EFFECT ON DEPENDENT
VARIABLE

p-value / sig (2-tailed) < 0.05 (α)


There is significant effect
ONE WAY ANOVA
Test whether there is significant effect of different
thermoplastic composites on the tensile strength.

Bamboo Kenaf Oil palm


(MPa) (MPa) trunk
(MPa)
46.7 47.0 40.1
40.1 46.3 35.7
49.2 37.3 34.3
52.1 42.9 45.6
40.9 48.7 40.1
TWO WAY ANOVA
TO TEST WHETHER RELATION BETWEEN AT
LEAST 2 INDEPENDENT VARIABLE HAVE
SIGNIFICANT EFFECT ON DEPENDENT VARIABLE

FIRST INDEPENDENT HAS SIGNIFICANT EFFECT ON


DEPENDENT VARIABLE

SECOND INDEPENDENT HAS SIGNIFICANT EFFECT


ON DEPENDENT VARIABLE

RELATIONSHIP BETWEEN FIRST AND


INDEPENDENT HAVE SIGNIFICANT EFFECT ON
DEPENDENT VARIABLE
TWO WAY ANOVA
An experiment was run to study the effect of pectin
dosage (mg/L) and pH on flocculating activity value.

pH Pectin dosage (mg/L)

10 50

3 92.6 92.4 96.5 96.5 99.4 96.8 96.8 36.5 47 98.4 54.7 97.8

9 93.0 93.7 95.0 95.4 98.0 98.3 98.3 98.0 97 98.5 95.2 96.5
MULTIPLE
REGRESSIO
N ANALYSIS
 PREDICTING UNKNOWN VALUE OF A VARIABLE FROM
THE KNOWN VALUE OF 2 OR MORE VARIABLES

 THE PREDICTED VALUE IS KNOWN AS DEPENDENT


VARIABLE

 THE KNOWN VALUES ARE KNOWN AS INDEPENDENT


VARIABLE

 THIS METHOD USED FOR STUDYING THE


RELATIONSHIP BETWEEN A DEPENDENT VARIABLE
AND TWO OR MORE INDEPENDENT VARIABLE
MULTIPLE REGRESSION FORMULA
EXAMPLE 1
WHAT IS R-SQUARED
 The Coefficient of deteremination, 𝑹𝟐 is a
measure of goodness of fit

 𝑹𝟐 Range from 0 to 1

 𝑹𝟐 =1 is a perfect fit ( all data points fall on the


estimated line or curve

 𝐀 𝐧𝐧𝐧𝐧𝐧𝐧𝐧𝐧 𝑹𝟐 𝐢𝐢𝐢𝐢𝐢𝐢𝐢𝐢𝐢 𝐯𝐯𝐯𝐯 𝐩𝐩𝐩𝐩 𝐦𝐦𝐦𝐦𝐦 𝐟𝐟𝐟


𝑅2 is equal to 0.9653
⇒ the actual values showed
high and good correlation

⇒ The value of the adjusted


determination coefficient
(adjusted R2 = 0.933) was
also highly support for a high
significance of the model
THANK YOU AND
GOOD LUCK !!!!

Вам также может понравиться